About the Author

Jeff Waincymer,

Jeff Waincymer is a Professor of Law at Monash University, Melbourne,

Australia. He is also a legal practitioner specialising in international trade,
investment and arbitration as counsel and arbitrator. He is a Fellow of ACICA
and is on the panels of HKIAC, KLIAC, ICDR Asia and SIAC. He is a nominee on
the WTO government panel list.

page "v"
Foreword

Michael Pryles,

It is a great pleasure to be asked to contribute a foreword to Professor

Waincymer's new book. We were colleagues at Monash University in Melbourne
some two and a half decades ago. I left the university to enter practice while
Professor Waincymer moved to Deakin University, although he has since
returned to Monash. At the university we both had an interest in International
Trade Law and Conflict of Laws. The latter subject naturally led me to explore
International Arbitration. When the opportunity came for me to lead an
Australian team to the Vis Moot in Vienna, I passed it on to Professor Waincymer
as I had, by that time, entered practice. Professor Waincymer took up the baton
and has, over many years, coached a succession of successful teams participating
in the Vis Moot. He is an excellent teacher and has acquired a deep
understanding of international arbitration, having taught the subject for many
years. In more recent years he has commenced practice, first as a tribunal
secretary and then as counsel and arbitrator, and this has undoubtedly expanded
his appreciation of arbitration.

Professor Waincymer's treatise, 'Procedure and Evidence in International

Arbitration' suggests a book of limited compass. A scrutiny of its Table of
Contents quickly reveals otherwise. It covers many topics pertinent to
International Arbitration and in considerable depth. There are chapters, as one
would expect, devoted to Evidence, Arbitrators, Hearings, the Award, Remedies
and Costs, amongst others. But the work also deals with important theoretical
underpinnings such as Policy Considerations and difficult subjects such as
Complex Arbitrations. In short it is much more than a practical guide; it is a study
in depth.

The work is firmly predicated on theory but is well supported by reference to

other writers and judicial and arbitral decisions. The frequent reference to other
writers does not mean that the book is a compendium of other people's views.
Far from it. There is much Waincymer throughout the text and in the
arrangement and treatment of subjects.

page "xxxix"

I read with interest the chapter on Choice of Law. It is still not fully recognized
that choice of law in international arbitration is very different from choice of law
in litigation. For a start there are many more choice of laws that can arise in
arbitration. These include the laws applicable to the following matters: the
arbitral regime, the arbitration agreement, the particular reference to arbitration
and the substantive rights of the parties. Often counsel seem to simplify issues by
classifying them as either 'procedural' or 'substantive', following court practice.
In relation to the former I prefer to speak of the law applicable to the arbitral
regime rather than the procedural law. How can matters relating to the
constitution of the tribunal and its powers and responsibilities be regarded as
merely 'procedural'?

Moreover in international arbitration there arises the question of whether the

traditional choice of law methodology is applicable. Some would suggest that the
traditional process of classification and selection of the applicable law is not
appropriate.

It is through works like Professor Waincymer's treatise that the regime of

international arbitration will be better understood and studied. After all
international arbitration is no longer a form of alternative dispute resolution. It
is the primary procedure for the resolution of international commercial and
investment disputes. As such, national litigation has become a form of alternative
dispute resolution for international disputes. Lawyers' familiarity with domestic
litigation should not be allowed to unduly colour approaches to international
arbitration.

I congratulate Professor Waincymer on the publication of his detailed treatise,

which makes a valuable contribution to literature on international arbitration. It
deserves to be consulted by practitioners and students alike.

Michael Pryles

page "xl"
Preface and Acknowledgements

Jeff Waincymer, Professor of Law, Monash University, Melbourne, Australia

The aim of this book has been to combine practical analysis of the procedural
and evidentiary stages of international arbitration, with a theoretical and
comparative perspective, in order to identify optimal solutions to promote
fairness and efficiency. While there is a range of exceptionally fine treatises
dealing with all aspects of international arbitration, they vary from those that
provide an exemplary introduction and overview of key areas, to those that
magisterially and at great length, outline the laws and rules in most key
jurisdictions. By concentrating on procedure and evidence, this book aims to fill
a gap between these two extremes, in particular by devoting more time to
articulating the arguments for and against various practical responses to
particular procedural and evidentiary issues. The aim was to be more
exhaustive as to general issues and practical options, but not as to every
relevant rule or case wherever they might be found.

The book aims to cover each and every procedural and evidentiary stage in
rough chronological order and be informed by variations in approaches
between legal families, different institutions and different lex arbitri. Essentially,
the book seeks to articulate what parties can and should do at each stage of the
arbitral process. It considers how tribunals should behave in order to promote
the fairest and most efficient dispute resolution exercise and how and why
trade-offs should best be made when fairness and efficiency inevitably conflict.
It also seeks to show how important procedural and evidentiary discretions are
to the outcome and quality of arbitral adjudication and how many supposedly
distinct topics such as choice of law, remedies and costs, can be better
understood only when sufficient attention is given to the implications in those
fields of evidentiary and procedural choices.

A project of this magnitude by a sole author would not be possible without the
encouragement and assistance of a range of people who have added
immeasurably to the work but who of course are not responsible for remaining
errors or page "xli" idiosyncratic views. I am particularly indebted to the
following practitioners who commented on individual chapters, namely (in
alphabetical order): Brooks Daly, Hew Dundas, Tony Canham, Justice Clyde
Croft, Graham Easton, Martin Hunter, Mark Kantor, Neil Kaplan, Pierre Karrer,
Julian Lew, Albert Monischino, Tim Nelson, Michael Pryles, Lucy Reed and
Matthew Secomb.

In similar vein, the project would simply have been impossible without the
support of a significant number of exceptionally fine research assistants, the
bulk of whom are former student participants in Willem C Vis Arbitration Moot
teams that I have coached over the last nineteen years. Again I wish to highlight
the assistance of (in alphabetical order): Rosehanna Amin, David Barda, Michael
Beaconsfield, Keren Benjamin, Thomas Dreyfus, Amy Greenberg, Jarred Hofman,
Catherine Miller, James Patto and Nita Rao, with a special mention to those of
my (sensibly non-lawyer) children who were in the country at key times,
namely, Ben and Ilan, who stepped in at short notice to do footnotes
extraordinaire when the regular pool of assistants dried up from time to time. I
wish to pay particular tribute to three former students and research assistants
who from time to time also coordinated with the research team at the same time
as doing the lion's share of the work. Here I wish to thank in chronological
order, Angus Dempster, Alex Fawke and Chris Collie.

Nothing I do or achieve is ever possible without the wholehearted support of all

my family, Sara, Ben, Sophie and Ilan and the tireless and incomparable work of
my friend and secretary Lorna Frick, ably supported by Jeanette Harlock.
Eleanor Taylor at Kluwer was always a supportive and relaxed editor who was a
pleasure to work with along with her broader team. Finally, the research aims to
be current as at the end of 2011.

Jeff Waincymer

Melbourne

page "xlii
Part I: Policy and Principles, Chapter
1: The Nature of Procedure and
Policy Considerations

Jeff Waincymer,

page "1" page "2"

1.1. The Nature of International Commercial Arbitration and The Specific

Role of Procedure

1.1.1. Introduction

It has long been accepted that arbitration is the dominant form of dispute
resolution in international trade and investment. This has come about for a
range of reasons. Initially, international business practitioners were concerned
about parochial attitudes of courts, the expense and delays in litigation and
general lack of commerciality displayed by judges in their processes and
conclusions. In some countries, the quality of the judiciary was thought to be
questionable. At the extreme, corruption may have applied. (1) Resort to a
national court may also raise perceptions of bias when the quality is high. Even
where judges are highly competent, they are unlikely to have significant
experience with international disputes. A rigorous understanding of conflicts of
law principles and an ability to effectively apply foreign law and to
accommodate the wishes of the parties is an important element of international
adjudicatory expertise that may not have been prevalent within national courts,
even in highly developed systems.

There are also many unique procedural uncertainties in international litigation.

In some cases it is not clear what jurisdictional ambit a particular court will have
as it can be a matter of the court applying connection tests whose application
may be uncertain. The tests may differ and more than one court might assert
jurisdiction page "3" over a particular dispute. Different legal systems have
different attitudes to this scenario where more than one court claims
jurisdiction. (2) There may be problems with service of process overseas and
recognition by courts of choice of law or choice of forum clauses. Some systems
will accept party choice. Other legal systems impose limits on enforceability of
forum selection clauses. Litigation in the national court of one party will also
pose differential logistical challenges for the foreigner. Particular problems arise
under international litigation where State trading or investment disputes are
involved. Here there could be issues of sovereign immunity or the application of
the Act of State doctrine.
There are also problems of lack of familiarity with foreign court procedures and
variances in norms. Where parties come from different legal families, there will
always be a problem in that the court selected will not be able to establish
neutral procedures from the parties' perspectives. For example, local courts may
apply evidentiary principles that are far removed from the norms previously
experienced by the foreign party. Examples include pre-trial discovery in the US
or some civilian rules which prevent a party being a direct witness in its own
case. Where the jury trial system applies to commercial matters, this may be
unfamiliar to most international litigants and may lead to surprising damages
results from time to time. Foreign litigation would also be subject to appellate
challenge and there remain significant problems in attempting to enforce
foreign judgments. While it is open for countries to negotiate regional or
bilateral judicial enforcement provisions, this is difficult to achieve in the face of
vastly different domestic systems. For example, US treble damages in a number
of areas such as antitrust litigation, are a disincentive to other countries
engaging in reciprocal negotiations. (3)

These factors have meant that from the outset, the development of international
arbitration was largely reactive to these problems of international litigation. The
establishment of the New York Convention on the Recognition and Enforcement
of Foreign Arbitral Awards (4) provided the single most important advantage
over international litigation through the establishment of a unique and effective
enforcement mechanism. Respect for party autonomy was also seen as crucial
and a natural advantage for arbitration. International arbitration, with its
deference to consent, allows control by the parties, and provides at least an
opportunity, if not a guarantee, that the process will be as neutral as possible,
cost effective, completed in a timely manner, conducted by a person or persons
with appropriate expertise and independence and result in a final determination
that may be enforced internationally in most jurisdictions.

page "4"

While international arbitration remains the preferred form of international

dispute resolution and will continue to do so, it has not always operated
optimally or been seen to do so by critics and users. In recent years, greater
attention has been given to analysing the strengths and weaknesses of this form
of dispute settlement and considering what reforms would best promote the
fairness and efficiency of its procedures. Many of the current debates and
reforms are dealing with the growth in cost and timing of arbitration and the
difficulty that arbitrators have in dealing with some specific problems, such as
multiple parties, multiple claims, interim measures and dilatory tactics. Thus
arbitration has its own challenges.

Furthermore, some of arbitrations greatest challenges are not as problematic in

litigation. For example, challenges to adjudicators on independence and
impartiality grounds will always be less of a concern where, as in litigation,
there is a separate pool of adjudicators appointed by government and kept
separate from commercial transactions. In addition, comparative utility of
arbitration as against litigation is not static in the areas where it has had an
advantage. Most advancing economies soon find that their courts seek to reform
their processes in relation to international transactions, hence the relative
cost/benefit of arbitration over litigation will vary over time. If arbitration
stands still and does not seek to constantly improve both fairness and efficiency
in its processes, its relative advantages may dissipate and it may indeed be
overtaken by litigation as a preferred adjudicatory model. That already may be
so in certain circumstances where arbitration faces particular problems such as
those alluded to above, for example where multiple parties are concerned. Even
if arbitration remains preferred to litigation on a regular basis, if it is not
optimised, it may lose popularity as against truly alternative dispute resolution
(ADR) methods such as mediation and conciliation. (5)

Considerations of ADR can also point to particular factors being seen as values
worthy of promotion within an adjudicatory system and hence being criteria on
which to evaluate systems. One possible criterion for evaluating formal
adjudication processes is to ask whether arbitration or litigation is the more
appropriate forum within which to reach a mutually agreeable solution, as the
latter will always be a preferred outcome to an adjudicated determination. Gary
Born suggests that there are a number of reasons why arbitration facilitates
amicable settlement, including the fact that arbitral proceedings generally
require procedural cooperation and parties are likely to select commercially
oriented arbitrators. He notes, however, that there is no empirical data to
suggest that it provides a better opportunity in this regard. (6) The factors he
points to as arguably supporting arbitration as page "5" a settlement-oriented
process are certainly valid, although there are some countervailing factors. The
first is the problematic position of an arbitrator who seeks to promote
settlement. This is discussed further in section 8.11. Allied to that is the fact that
many domestic litigation systems now demand mediation as a required step in
any litigation process, (7) although that itself is contentious. (8)

While there is now much literature for and against the advantages of arbitration
over litigation, the most important observation is that arbitration, through the
relevant consent of the parties and the broad discretionary powers of the
arbitrator, at least affords the best opportunity to tailor the processes so as to
achieve maximum fairness and efficiency. A related observation is that parties
and their advisers who are most aware of this opportunity, are more likely to be
able to reap the benefits. A further observation is that because of the various
advantages and disadvantages and the ability to shape these factors through
party autonomy, the decision to resort to arbitration should be an informed one.
Transactional lawyers should consider a number of factors when making the
choice. These include the type of transaction involved, where assets are held,
jurisdictional links and the attitude of courts in those jurisdictions, the
particular features of arbitration being contemplated and the kinds of issues
that may arise in disputes that could render variations in the relative
cost/benefit of litigation versus arbitration. (9) While such a considered approach
should be the norm, Gary Born also observes that arbitration is often chosen not
because it is the optimal ideal but rather ‘the least unfavourable forum that the
party can obtain in arm's length negotiations’. (10)
If an informed and efficient choice is to be preferred, it is necessary to further
consider the respective merits of various options. This book aims to consider all
aspects of procedure and evidence from beginning to end in the arbitral process.
The aim is to identify how pervasive issues of procedure and evidence are at
each and every stage, hence showing how important the topic is and how
articulation of rules or the exercise of discretions will have a fundamental
impact on the outcome of arbitration and the respect in which it is held. To
Wetter, ‘arbitration as a subject is procedure’. (11) It has also been said that
‘procedure really does matter in arbitration’. (12)

On one level, these are intensely practical issues, dealing with significant
problems such as whether arbitration can adequately handle the above-
mentioned problem scenarios such as multi-party claims or interim measures.
There is, page "6" however, a need to situate any practical discussion into a
theoretical and policy-oriented framework. As with any contentious issue within
the arbitration paradigm, the views one may have on practical options and
solutions will invariably be coloured by one's theoretical views as to arbitration
per se. For example, if arbitration is fundamentally seen as a consent-based
agreement to bypass State-based litigation systems, then party autonomy would
be seen as the dominant means to resolve procedural questions. Conversely, if
arbitration is ultimately seen as being derived from the jurisdictional
imprimatur of national legal systems, supported by key international
conventions, then procedural solutions consistent with the values of those very
same national systems or consistent with transnational norms may be more
appealing. This book does not aim to resolve those theoretical differences, but
aims to be informed by them and explain where they would affect the ranking of
various procedural options. Because there is greater acceptance of a hybrid or
similar theoretical construct, there is a greater need to identify the best means
to integrate autonomy and jurisdictional issues into procedural and evidentiary
questions.

To the extent that procedure is central, it is important to identify what are the
optimal norms and by what principles procedural and evidentiary discretions
should be employed. Similarly, to the extent that party autonomy defines the
parameters of arbitration, what procedural and evidentiary norms should
parties agree to? Even posing the questions in this way is contentious. Some
might question the utility of a search for principles, arguing that ‘international
arbitration is not a science, it's an art’. (13) Part of the aim of this book is to
determine to what extent some articulation of principles is possible or by what
principles the ‘art’ may best be employed. It will certainly not be argued that
there is some rigid formula that can lead to a fair and efficient procedural
decision in all cases. Much has to be left to the intelligence and integrity of the
tribunal involved. But if that is all we can say and no clearer guidance can be
provided, how do parties arrive at procedural agreements as is their general
right? How can we evaluate the determinations made, either by the parties or by
a tribunal when exercising its discretionary powers? How can counsel make
submissions on disputed procedural discretions? How can institutions set up
optimal models? How can budding arbitrators be educated as to best practice
standards?

The aim of the book is thus to be practically useful within a rigorous theoretical
paradigm. The aim is more about exploring all issues to aid the reader, rather
than presenting one scientific theory. The aim is also more in the form of
discussion than an attempt to collect all relevant rules and cases. Given that
international arbitration occurs in so many places, subject to so many
procedural and substantive norms, identification of all relevant rules is best left
to case-by-case research, with resort being made to the major treatises, internet
resources and national compendia. Thus the book will refer to examples from
leading jurisdictions around the world page "7" but does not aim to be a
reference manual for the rules of all systems. The book is also not aiming to
outline the elements of court procedures that are ancillary to the arbitral
process. Such procedures include stay of proceedings applications, supportive
court-ordered interim relief, court assistance in obtaining evidence abroad and
procedural challenges to the award and enforcement. While these are important
in their own right, they are only dealt with in the context of the way they would
impact upon an arbitrator's powers, rights and duties. For example, because an
arbitrator aims to render an enforceable award, a tribunal might consider the
potential for legitimate challenges in the procedural decision-making matrix. To
that extent, a tribunal might be predicting the likely behaviour of supervisory or
enforcement courts.

Being concerned with procedure and evidence, the book is also not directly
covering substantive issues. Nevertheless, a number of such issues are in fact
dealt with in part, simply because it is too easy to overlook the importance of
procedural and evidentiary issues where resolution of all substantive matters is
concerned. Hence the book looks at the procedural and evidentiary aspects of
choice of law in Chapter 13 and remedies in Chapter 14 These do not aim to be a
comprehensive exposition on those topics, but instead, an outline of all of the
procedural and evidentiary issues that arise when identifying, proving and
applying substantive norms, including determination of remedies.

The book is also divided into three parts. The first part is an introduction as well
as an analysis of policy and principle. Part II is an analysis of the arbitral process
from the beginning to the close of proceedings. Part III deals with the award and
remedies.

Some key themes are addressed at a general level in this and the following
chapter. Many will also be recalled at the beginning of each specific chapter
wherever they constitute a significant cause of divergent views on the topic
being discussed. These would include how can a tribunal be fair, and in
particular be neutral, if parties are so diverse? How do we make trade-offs
between fairness and efficiency? How does a tribunal do what is best if the
parties are choosing a suboptimal process? Is there a consistent paradigm in
which to make such decisions? Because a tribunal's determination of such
questions involves an amalgam of rights, duties and powers, Chapter 2 looks at
these more generally as a precursor to further consideration on a topic-by-topic
basis.

1.1.2. The Nature of Procedure

In a book dealing solely with procedural and evidentiary matters, it is of course

necessary to consider how to distinguish between procedure and substance.
This is not only necessary for determining the ambit of the book, but more
importantly, so that arbitrators and practitioners can consider what rules and
principles apply to any kind of question. The distinction between matters of
procedure and matters of substance has been important in the conflict of laws
because, generally speaking, matters of procedure are governed by the lex fori,
whereas matters of substance are page "8" governed by the lex causae. (14) In
arbitration, we consider the lex arbitri as opposed to a lex fori. For example, in
the UNCITRAL Model Law, procedural matters are determined under Article 19,
subject to Article 18 norms, while the identification of substantive law is dealt
with under Article 28. The processes are different and hence it is important to
categorise an issue to know how to proceed.

While characterisation is necessary, it is problematic as there are many

scenarios in the grey area where the concepts can overlap. There is no universal
consensus on what matters are procedural and what matters are substantive. (15)
Domestically, the exact characterisation of matters as either procedural or
substantive or hybrid depends on the particular jurisdiction. In some domestic
systems, the position has changed over time. (16) This has meant that today, some
common law and civil law jurisdictions classify a greater number of matters the
same way. (17) Nevertheless, because different legal systems still take differing
approaches to what may be characterised as substantive or procedural issues,
an arbitrator must begin with such a characterisation question. Because an
arbitrator is not bound by the habitual characterisation under a domestic legal
system, there must be some other methodology for approaching such an issue.

The first question is to consider whether the lex arbitri resolves the issue. If so,
this should be determinative. Lex arbitri will rarely provide an answer, although
they may do so indirectly, simply by referencing certain matters expressly in
procedural rules. (18) While most lex arbitri do not address the question of
characterisation, there are some exceptions. For example, section 34(2) of the
English Arbitration Act 1996 defines procedural and evidentiary matters to
include when and where the proceedings are to be held; the language of the
proceedings and whether translations are to be supplied; the use, provision and
amendment of statements of claim and defence; disclosure and production of
documents; questions to be put to and answered by the respective parties;
whether to apply strict rules of evidence or other rules as to admissibility,
relevance or weight; time, manner and form of the exchange and presentation of
evidentiary material; whether and to what extent the tribunal should itself take
the initiative in ascertaining facts and law; and whether and to what extent
there should be oral or written evidence or submissions.

page "9"
In the absence of such guidance, various abstract definitions of laws relating to
procedure, as opposed to laws relating to substance, have been proposed – for
example: procedure is ‘concerned with manner, whereas…substance (is)
concerned with matter.’ (19) Substantive issues are those which are elements of
the decision on the case and the reasoning behind it. Procedural issues deal with
the process by which that decision has been reached. Another method of
determining the distinction between substance and procedure is to consider
whether a norm has autonomous substantive content or relates to the
application of another norm. (20) While it is relatively easy to state the distinction
in the above manner, at the margin it is difficult to clearly categorise some
questions as falling in one group or the other. An example of a mixed
procedural/substantive determination occurred in Brasoil. (21) A tribunal had to
consider an allegation of fraud as the basis for a request to review a partial
award. While the ruling called for was procedural, a review court considered
that the application also called for a prima facie consideration of a substantive
claim as to fraud.

Pierre Karrer suggests a functional approach, delineating procedural matters as

merely relating to organisation of the procedure without affecting the inherent
fairness of the process. Substantive issues should be those which influence the
outcome of the case directly. He suggests that the latter would include burden of
proof, interest and currency. (22) While it is certainly appropriate to attempt a
more harmonised delineation not dependent on conflicts methodologies, it is
problematic to try and differentiate by a principle as to which matters influence
the outcome of a case directly, although as noted below, Karrer's suggestion is
particularly useful in determining when a chair can make decisions alone. It is
more problematic as a general definitional test. For example, if an application
was made for production of a document by a claimant that could not succeed in
the absence of such a document, and the target of the application is from a
civilian legal system, the production decision will be wholly determinative of the
outcome. As another example, if a statement made during negotiations was
alleged to be misleading and there is a dispute as to whether third-party
evidence about that statement can be presented, different legal families have
different responses about how to categorise the issue, as well as how it ought to
be dealt with from a policy perspective.

There are a number of reasons why some situations are difficult to categorise as
either substantive or procedural. At times this may be because of their inherent
nature which displays features of each. At other times it is because of differences
of approach between different legal families, although this cannot usually be
traced page "10" back to inherent nature, but instead result from differing
systems selecting different factors as determinative. For example, the limitation
of actions was traditionally considered a matter of procedure in the UK, (23) and a
matter of substance in civil law jurisdictions. (24) Both systems now treat it as a
matter of substance, although some common law jurisdictions still continue to
treat it as a matter of procedure. (25) Another contentious area is in relation to
set-off rights. Other matters, like burden of proof, are also not straightforward.
Briggs states that the law is basically uncertain in the UK. (26) In Germany, the
burden of proof is a matter of substance. (27) Born argues that the issue of burden
of proof has both procedural and substantive elements and that arbitrators
should not directly apply the rules with regard to the burden of proof from any
particular domestic legal system. (28) Burden is considered further in section
10.4.1.

A further problem of categorisation arises if questions relate to the arbitration

agreement itself. On the one hand, an arbitration agreement is the mechanism
that establishes the procedure and hence could be seen as such. On the other
hand, being an agreement, it is generally part of a contract and often embedded
into the substantive contract between the parties. Hence it does not have a
wholly unique status within the substance/procedure categorisation. This can
impact on the question of its applicable law. So too can its autonomous nature.
These issues are discussed further in sections 3.2.1 and 3.2.2.

If different legal systems categorise procedural and substantive issues in

different ways, one challenge facing an arbitrator who needs to characterise an
issue in order to properly deal with it, is to determine how to do so in a fair and
reasonable manner. Where the parties come from similar legal systems, an
arbitrator might choose the characterisation they are familiar with, based on an
implied form of ex ante consent. Conversely, an arbitrator may approach the
matter from a conflicts perspective, looking at the applicable conflicts laws and
how such laws would characterise the issue. In some cases that would be
circular, as an arbitrator might use conflicts methodology to identify procedural
and substantive laws but still not know at the margin, through which of the two
fields a particular matter should be considered.

Regardless of the difficulties, there are other important reasons why procedure
and substance need to be distinguished in addition to the need to identify
relevant rules. In particular, most challenges to awards by way of annulment
claims or enforcement defences need to be based on procedural as opposed to
substantive page "11" error, although the issue is not so much one of
classification but instead, the proper interpretation and application of
annulment and enforcement provisions. The question is also typically one of
degree, in terms of whether the procedural flaws are so significant that a due
process challenge is merited. Perhaps the key area where the distinction is
needed and where there is no optimal guidance is in relation to procedural
directions by a tribunal. This is particularly important with multi-person
tribunals where it is commonly the case that the chairperson or president of the
tribunal may make procedural orders alone. At the extremes, it is relatively easy
to distinguish between simple procedural matters such as a timetable for
written submissions and at the other extreme, a determination on the merits. In
between, however, there are those matters that are classified differently
between legal families as noted above and the cases where procedural and
evidentiary determinations will fundamentally affect substantive
determinations. As argued in Chapter 6, in that context a restrictive notion of
procedure should apply. As a general rule, anything that can truly affect the
substantive outcome such as a limitation determination should be a decision of
the entire tribunal after hearing from the parties. In this context, Karrer's
functional perspective has much to commend it.

1.2. Policy Criteria for Evaluating Procedural Models

1.2.1. Fairness and Efficiency

The previous section looked at how one seeks to define and identify procedural
questions. This section looks at the features by which we may go about
designing or evaluating procedural models. Any manuscript seeking to evaluate
competing procedural models must have some policy criteria by which the
evaluation may legitimately occur. Virtually all regulatory policy questions,
including legal ones, can be analysed from the perspectives of fairness and
efficiency. Fairness connotes the justice of a situation and is inherently a
question of philosophy. Efficiency is an economic concept.

This has not been lost on arbitration scholars and practitioners. Fortier has
referred to arbitration ‘as a never ending battle between the interests of justice
and fairness on the one hand, and finality and efficiency on the other.’ He
suggests that it is the responsibility of the arbitrator to ensure ‘both the fair and
the efficient functioning of the arbitral process.’ (29) Bernardini observes that
‘(r)econciling speed with justice is a basic requirement of any arbitral process,
given also the finality of this process and the limitations for the means of
recourse against the award.’ (30) Park page "12" observes that ‘(a)rbitral case
management implicates the delicate counterpoise between efficiency and
fairness.’ (31) Van den Berg has articulated these policy norms in the context of
the overriding duties and powers of an arbitrator:

The existence, nature and scope of the arbitrator's discretionary powers

constitute the hallmarks of arbitration, particularly in the international context,
but these powers are not unfettered. In particular, they are tempered by the
parties' rights of due process and the arbitrator's duty to ensure that those
rights are respected and enforced. An arbitrator is accordingly exposed to a
potential conflict between fairness, on the one hand, and efficiency, on the other.
This reality bespeaks a profound confidence in the ability of arbitrators to
conduct proceedings in accordance with the needs of the parties and the
demands of justice. (32)

Where there are complex procedural issues with multiple options,

considerations of fairness and efficiency will not always point in the same
direction, hence tradeoffs will often need to be made as alluded to in the above
quotes. This can either be expressed in the procedural rules or left to be dealt
with on a discretionary basis by the tribunal. Perhaps the greatest procedural
challenge facing arbitrators is to determine how trade-offs ought to be made on
a case-by-case basis when fairness arguments conflict with efficiency concerns.
The same issues arise in design of arbitral laws and rules and in the design of
practical guides. Before looking at how this may occur, attention is first given to
a consideration of the individual criteria.
1.2.2. Fairness

The notion of a fair hearing is enshrined in various legal and human rights
models. (33) In addition to actually promoting fairness, an arbitrator should also
ensure that there is a sufficient appearance of fairness. This is important
because the parties' perceptions as to the process are most important. (34)

page "13"

1.2.3. Process versus Outcome Fairness

Fairness has two distinct elements, one dealing with the fairness of the outcome
and the other with the process by which the outcome is reached. Some
practitioners and philosophers would argue that the only just outcome is one
where the correct result was achieved. René David stated that the ‘concern of
arbitrators, in accordance with the parties' wishes, is to arrive at a just solution’.
(35) While noting the importance of efficiency, William Park argues strongly that

‘(e)fficiency without accuracy will prove an empty prize’ and that ‘award
accuracy remains the lode-star.’ (36) On that view, a just outcome is an inherent
element of fairness. In arbitration, however, there is a further nuance. Is the
fairest outcome the one based on the technical application of law, or the most
equitable solution, or the one that seems to conform more closely with the true
intent or reasonable expectations of the parties? (37) Another complex nuance in
the context of fairness and efficiency is to consider an outcome not in the
context of the arbitral award alone, but in terms of the effect on the parties'
business activities. A process that seeks to resolve a dispute in an acceptable
manner and in the least disruptive fashion may not be that which is most
conducive to accurate fact finding. (38) Considering dispute resolution as being
about peace-keeping, consensus building and preserving contractual
relationships can thus provide a different perspective. (39) Because there are
differing views on the nature of arbitration and the methodologies leading to
particular solutions, there can be differences in view as to what makes for a just
outcome.

To others, the essence of justice is procedural. Whatever outcome is reached by

a fair process is in that sense seen to be a fair outcome. (40) In looking at process
issues, considerations of fairness can be further broken down into general
principles of due process, equality of treatment, and providing each party with a
sufficient opportunity to present its case. It might even be argued that decisions
are so process-dependent that there is rarely an objectively ‘right’ decision in
any case which must be searched for by the arbitrator. Instead, the task is to
accurately apply the logic of factual and legal analysis within procedures that
are so well respected, that even the loser will respect what has occurred.

An example of a distinction where one's view of justice matters is where a party

who will otherwise lose a case, wishes to introduce compelling new evidence
page "14" after the close of proceedings. The distinction is particularly stark if
the new evidence has only become available after the close of proceedings.
Hence the party wishing to introduce the material is not in any way at fault for
not having done so earlier. One may approach this intensely practical issue
differently depending on one's theoretical preference between process or
outcome justice.

While it is easy to state the difference between the two notions of fairness or
justice, it is not always easy to distinguish between just procedures and just
outcomes because they are not clearly separated. There will also be differences
in view as to what constitutes a just procedure. The differing methodologies
alluded to above are again relevant. A most illustrative example in international
arbitration is a hypothetical one involving a US claimant and a German
respondent where the claimant can only succeed by obtaining discovery of
internal documents held by the respondent. To the US party, full discovery
rights are an essential part of a just procedure. To the German respondent, it
offends against its notion of justice for a respondent to have to help a claimant
who should otherwise have had a proper case before commencing it. This raises
a more contentious aspect of fairness through incorporation of a libertarian
perspective, in particular as viewed from the position of the respondent. Parties
from diverse legal systems would differ in their views and would not concur on
whether broad discovery powers are a just procedure or not.

Fairness could also include elements of efficiency, as excessive cost and delay
would not connote fairness. In a purely commercial dispute, would a party
prefer to lose within a year with out-of-pocket expenses of USD 150,000, or win
the case after three years, with net losses after award payment of USD 250,000?
(41) This in turn raises the question whether a just procedure is based on truth

alone or involves libertarian perspectives or efficiency perspectives as well. The

notion of fairness itself might thus be based on the same trade-offs that occur in
establishing any procedural model.

1.2.4. Is a Free Choice of Procedure Inherently Fair?

Procedural fairness can be looked at as an absolute concept, or alternatively, or

in addition, in the context of party autonomy. Party autonomy would emphasise
that, unless procedures selected freely by the parties offend against mandatory
norms, they are arguably fair for that reason alone. The parties could just as
easily have made their consent to arbitration conditional upon adoption of the
procedure selected. However, at times autonomy can lead to questionable
outcomes from the perspective of procedural fairness. For example, the freedom
to select an arbitrator may lead to appointment of biased persons. At other
times, the parties may have disparities in bargaining power and hence the
consent may really be to a non-neutral model preferred by the stronger party. At
other times, the parties through ignorance may simply have agreed on a
procedure that the arbitrator page "15" considers is suboptimal from a fairness
and efficiency perspective. Hence it is hard to conclude that any choice is ipso
facto a fair choice. Hence a more objective delineation is necessary.

1.2.5. The Elements of Procedural Fairness

When we speak of procedural fairness, we typically speak of due process or

natural justice or concepts to similar effect. (42) Within such notions there are
two key concepts, one being equality of treatment and the other being an
adequate opportunity to fully present a case. While different observers may
have differing views about the content of each of these sub-elements, virtually
all would agree that they are important. (43)

While these are vital elements in their own right they impact significantly and at
times controversially on a range of other specific arbitral processes. For
example, can a party demand that it has an equal amount of time and an equal
number of witnesses under equality concepts? Must a full or adequate
opportunity to present a case allow for late submission of evidence? Once again,
these are highly practical questions that inevitably call for attention to be given
to the essential nature of these norms. The following sections briefly elaborate
on these concepts.

1.2.6. Equality of Treatment

While everyone would agree that parties should be treated equally in any just
adjudicatory forum, it is more difficult to describe what the treatment should be.
Aristotle made the point that identical treatment of unequal persons is not
equality. (44) It is suggested that equal treatment means ‘relative’ equal
treatment. That entails treating comparable situations equally and dealing
appropriately with differences. (45)

This has numerous practical implications. For example, when one considers
elements of arbitration such as the time to be given to each party to make page
"16" submissions, should each party have equal time regardless of the
substantive issues that they have pleaded or should they be given a fair time in
proportion to those issues? Equal time may not be equality if the issues and facts
each relies upon are different. The presumption will still generally be in favour
of equal time. Even though one party might only be pleading a simple issue, for
example, failure to pay for goods, the other party's defences, such as defects in
the goods or misrepresentations about their quality, need to be responded to.
Furthermore, if time was allocated in proportion to the issues raised, that would
be an inducement for parties to raise a greater number of issues to capture a
greater allocation of time. This would lengthen proceedings and increase the
costs. As always, issues of fairness interrelate with efficiency concerns. In
addition, optimal design of any procedure must think about how it might be
sought to be abused. In some cases, a procedure that would work well with
parties arbitrating in good faith, may be sub-optimal in the hands of parties
seeking to disrupt proceedings.

One situation where equal time may not be fair is where the parties have
significantly different aptitude in presenting their arguments, although there
could be philosophical debate on this as well. In terms of practical examples,
what if one party is presenting in their native tongue while the other must
present in a foreign language because of a stipulated choice of language of the
arbitration? Should allowances be made for the differences? In an investment
dispute, if a third world host country is represented by its non-legally trained
bureaucracy against multinational lawyers, should any deference be allowed?
Another issue is if one party is seeking to increase the costs with the ostensible
aim of accuracy, but in reality, in the hope that the other party, being less
wealthy, will be forced to settle on favourable terms. In domestic legal systems,
allowances might be given to some degree for parties who represent themselves
and financial aid may be given for those who could not otherwise fully present
their arguments. Many procedural models allow the adjudicator to protect the
interests of the weaker party. Arbitrators have to consider such questions
within very broad procedural discretions. All of these questions suggest that we
aspire to substantive equality and not any purely formalistic notion. Yet a
substantive notion involves qualitative assessments about which people may
legitimately disagree, both conceptually and on a case-by-case basis.

1.2.7. The Ability of a Party to Fully Present Its Case

Due process and fairness lead to the principle that each party should have the
ability to properly present its case. A differential ability to present would also
offend against an equal treatment norm. The ability to fully present a case
includes the concept of natural justice, which in part, entitles a party to
understand the challenges it faces and be permitted to make representations
before final determinations are made. Mantilla-Serrano suggests that elements
of this principle include receiving proper notice of each relevant stage, being
given a reasonable time and page "17" opportunity to respond and respecting
the general right to an adversarial proceeding. (46)

There will again be conceptual and practical questions in determining the

content of this right. In addition, it poses greater challenges for drafters in
determining whether to mandate an adequate, reasonable or full opportunity.
(47)

1.2.8. Neutrality:Bridging the Divide between Legal Families

As noted, one aspect of procedural fairness is equal treatment. One difficulty in

truly employing equal treatment in international adjudication arises where
parties come from differing legal traditions. If different legal families can
legitimately adopt quite distinct approaches to procedural problems and to the
design of procedural norms, how can we develop truly neutral procedures that
do not provide an a priori bias in favour of one legal culture over another?

Leading arbitrators and commentators have seen this as both an opportunity as

well as a problem. The whole history of comparative law scholarship has sought
to consider when and how lessons may be learnt about the operation and
improvement of any legal system, by considering how other legal families
approach similar concerns. A sophisticated comparativist would not presume
that solutions in one legal system are readily transplantable into others, in part
because of differing historical experiences, culture, values and aims.
Nevertheless, because so many difficult issues involve a trade-off between
fairness and efficiency, looking at how other legal systems have developed
trade-offs and critically evaluating the cost-benefits of each model can be a very
useful guide to policy development. Differences in approach between legal
families permeate arbitral procedure, beginning with differences in the way
claims are first expressed, the rights and obligations vis-à-vis production of
documents, treatment of witnesses, control over proceedings, involvement of
the adjudicator, oral versus written hearings and remedies.

These problems face any international adjudicatory body and the development
of optimal norms in the arbitration field can seek to draw inspiration from the
long-standing traditions within such bodies as the International Court of Justice,
the World Trade Organisation, the European Court of Justice and the Permanent
Court of Arbitration. There should be no presumption that an optimal model,
best suited to the politics and values of a particular institution, will naturally
translate to private arbitral disputes. But to the extent that those institutions
and their adjudicatory bodies have sought to reconcile the key procedural
differences between legal families, they can only help in identifying the extent to
which there is an emerging common procedural model accepted as an inherent
part of international adjudicatory practice. The same is true where domestic
litigation systems page "18" undertake major reforms, as all are grappling with
fairness and efficiency challenges and are aiming for optimal trade-offs.

This has both theoretical and practical aspects. It is an important academic

question to consider whether emerging harmonised models are discernible.
From a practical perspective, to the extent that this is so, arbitrators would
naturally look to such models to fill gaps in the procedure selected. In seeking
convergence, some urge concern for the identification of the essential character
and spirit of desirable norms rather than technical sophistication. (48) Ahmed El-
Kosheri has noted:

(I)n order to achieve a truly universal culture, it is necessary to transcend the

particularities of any given culture and, going beyond regional domestic
experience, seek common values in juridical and ethical canons that command
universal acceptance. (49)

These issues can operate at the highest level of discourse or operate at the most
basic level within actual arbitrations. As to the latter, neutrality can not only
encompass rules and discretions, but also demeanour. In pursuit of neutrality it
may even be appropriate to seek to use neutral language and not use
phraseology that would be respected differently in various legal families. For
example, it is better to speak of production of documents rather than discovery,
the latter being the process in common law systems which is not favoured in
many civilian jurisdictions.

1.2.9. Fairness and Certainty

It has been suggested that fairness can be broken down into procedural or
outcome fairness, each or both of which can be aspirations of any system.
Whichever form of fairness is being considered, it can also be looked at from the
perspective of consistency and certainty. It is highly desirable that procedural
determinations are consistent and predictable. The same is so for decisions on
the merits, although this raises other questions as to the relevance of past cases
on arbitral analysis. (50) If procedural and evidentiary determinations are
inconsistent, some or all may be problematic for that reason alone. At the
extreme, the determinations may be arbitrary, which itself would seem unjust.
This issue is central to the contentious debate as to whether there should be
detailed procedural rules and guides to promote certainty and consistency. It is
contentious because certainty and predictability can too easily become rigidity,
which itself could be unfair and inefficient. It is important that like cases are
treated alike but that different cases are given appropriately differential
treatment. Flexibility is an important value in ensuring page "19" that proper
attention is given to the nuances of individual cases. Yet finding an appropriate
trade-off between flexibility and certainty is difficult.

In institutional arbitration, finding a balance between flexibility and certainty

also involves finding a balance between institutional constraints on arbitrator
behaviour to ensure consistency and certainty, while at the same time allowing
sufficient discretion so that arbitrators can exercise good commercial judgment
appropriate to the facts before them. (51) Institutions have to determine whether
they wish to be known for a ‘light’ or ‘heavy’ touch, with no obvious optimal
strategy. (52)

1.2.10. Efficiency

As noted at the outset, the two key means to evaluate any legal system or
prospective reform is through philosophy per medium of fairness
considerations and economics per medium of efficiency factors. Efficiency is the
study of how one goes about allocating scarce resources when these are not
sufficient to meet every worthwhile aspiration. It has both a positive and
normative potential. The positive aspect looks at how resources are likely to be
allocated. The normative aspect looks at how they ought to be allocated.

Efficiency in an adjudicatory context has a number of elements. Efficient

proceedings ought to have as low transaction costs as possible. In arbitration,
this includes the direct expenses of counsel, the arbitrator, at times an arbitral
institution and the travel and other logistics of international adjudication. An
arbitrator certainly has control over the conduct of the proceedings, but has
little control over counsel, generally the most significant expense. Adjudicatory
systems also need to remain aware that there are psychological costs of stress to
participants and the opportunity cost (53) of the parties and witnesses being
diverted from other commercial concerns by reason of the dispute process
which can rarely be compensated for. An efficient process might at least try and
minimise these deadweight costs. (54)

Another aspect of efficiency relates to questions of timing and how long it takes
for the resolution of the dispute. Questions of timing will not only impact on the
direct costs, but will also impact upon the overall value of arbitration as a means
to promote efficient performance under the underlying contractual agreements.
page "20" A sub-element of this aspect of efficiency considers the various
elements of an adjudicatory process and how these can be conducted in the
most resource-effective way possible. Here we might deal with questions such
as whether it is preferable to tender arguments by way of written submissions
developed by the parties in their own time as opposed to oral hearings where all
parties are present, which is typically more costly. (55)

Efficiency can also be considered in a broader context in terms of the

commercial relationship between the parties. Given that a very large part of
modern international trade and virtually all of international investment involves
ongoing relationships, an individual arbitral dispute is often simply one aspect
of an ongoing commercial relationship that the parties wish to continue.
Ultimately it is the efficiency of that relationship and the ability of dispute
resolution to foster it that matters, both in a systemic sense and in the context of
their aspirations. Even if the parties have a dispute requiring an independent
adjudicator, they will typically wish the underlying commercial relationship to
remain on foot on a mutually productive basis. There is then the question as to
whether the tribunal has the right or the duty to conduct the proceedings in a
way to best enhance the ongoing commercial relationship, (56) as opposed to
concentrating on a static problem. One practical, albeit contentious, aspect of
this is the role that an arbitrator might take in fostering settlement. (57)

1.2.11. Efficiency and Cost

One of the key inspirations for the development of arbitration has been the
desire to reduce cost as compared to litigation. Yet lower costs and timeliness
are not inherent advantages of arbitration. Instead, party autonomy and tribunal
discretion simply provide opportunities for improvements over litigation.
Without careful thought towards those goals, arbitration will not naturally lead
to these preferred outcomes.

Indeed there are a number of reasons why arbitration will not have a cost
advantage unless this is very carefully orchestrated. Unlike judges, an arbitrator
must be paid fees. If an institution is involved, it will also charge fees. Hearing
rooms are not provided by the State but must be paid for privately. A neutral
venue will mean that all parties must travel to the Seat and pay for
accommodation as well. Witnesses will also have to travel, as may large bundles
of documents. In international litigation, it is likely that only one party must
actually move to a different jurisdiction. As noted above, the key cost is that of
legal representation. Parties are always likely to consider the cost/benefit of
expensive legal representation in any dispute settlement forum, whether
litigation or arbitration. It is page "21" arguable, however, that the costs rulings
of the adjudicator will impact to some degree on these choices. For example,
domestic litigation in a jurisdiction that is only willing to award relatively low-
scale fees to the winner makes each party think very carefully as to how much
over scale they are willing to pay. Conversely, if arbitrators are more willing to
award costs on closer to an indemnity basis, this may ultimately suit the
winner's immediate interests, but may increase the overall transaction costs of
the dispute. In the long term, increased transaction costs are unlikely to be in
anybody's best interest. Indemnity costing in arbitration hence raises both
fairness and efficiency concerns in ways that potentially conflict. Legal costs of
litigation will certainly be higher where there are multiple fora seeking to have
control of a dispute. Arbitration costs will also be lower than in litigation in
circumstances where foreign judgments need to be relitigated as a step in the
process of enforcement. Appeal stages in litigation also have significant legal
costs.

The most contentious question is whether a tribunal should be concerned with

proportionality in establishing arbitral procedures, namely determining that the
amount of time and cost to allocate to a case should be in proportion to the
amount in dispute and perhaps the difficulty of the competing contentions. This
raises natural challenges as to the trade-offs between fairness and efficiency. (58)

1.2.12. Efficiency and Finality

Arbitration is an alternative to litigation and aims to be a final and binding

method of dispute resolution that is readily enforceable in foreign jurisdictions.
In relation to finality, arbitration does not tend to have built-in appeal
mechanisms. Party autonomy is presumed to instead prefer expeditious
resolution and avoid undue interaction with courts. Even so, efficiency and
fairness break down if an award is rendered in ways which are open to
challenge, hence undermining finality. In this sense, evaluation of procedure and
procedural inadequacies is not conducted in a vacuum. It is generally framed in
the context of potential challenges to the award or challenges to enforcement.
Here there is obviously a trade-off between one aspect of efficiency, being
finality and one aspect of fairness, being a just outcome.

Removing an appeal mechanism is also a trade-off, preferring timeliness to

accuracy. It is unlikely that finality itself is highly valued as opposed to
timeliness of resolution of a dispute. Most legal litigation systems accept that
lower level courts may err at times, hence the need for appeal processes, often
leading to a three-tiered court system. Domestic legal systems allow for appeal
rights on the basis that lower courts can deal with a mass of disputes and the
appeal process can filter more complex matters to more expert judges. That is
generally less relevant in arbitration as the parties have the ability to select
arbitrators of appropriate expertise from the outset. Nevertheless, because
party appointment may lead to less than ideal arbitrators in some cases and
because differences in expertise and page "22" culture between arbitrators or at
times busy arbitrator schedules may lead to less than optimal decisions,
empirical studies suggest that users of arbitration are more ambivalent about
whether some greater appeal rights may be desirable or not. Nevertheless,
Bühring-Uhle suggests that finality is still preferred over appeal rights. (59) Some
lex arbitri allow for limited rights of appeal as a compromise. Party autonomy
allows for this to be incorporated if the parties so wish, but this is rare and
parties cannot contract for court jurisdiction that otherwise does not exist.

1.2.13. Efficiency and the Tension between Flexibility versus Certainty

As noted above, both procedural fairness and outcome fairness encompass

tensions between flexibility on the one hand, and certainty on the other. Similar
issues arise in relation to efficiency. Certainty of procedure means that a
tribunal can perform its functions without having to make costly and time-
consuming determinations on individual questions. Where such determinations
are needed, there are added costs, delays and perhaps tensions if the parties
cannot agree. When there is sufficient certainty as to the processes and
methodologies, there can even be increased certainty as to the likely outcome,
which can only assist the parties in attaining a mutually agreeable settlement.

Again, as noted in the context of the fairness analysis, certainty should not be
over-valued. Rules which aim to be too prescriptive in order to promote
certainty will often fail to do justice to unique circumstances that might require
unique solutions. Certainty may not promote efficiency. Certainty is always
opposed to flexibility, the latter, which is also a value, often supported in
isolation. We would also hope that arbitrators would be creative and adaptable.
Most would still agree that minimum standards of predictability and consistency
are desirable in the abstract at least. This itself raises comparativist concerns as
to the means by which such goals can be achieved. One controversial debate is
the extent to which rules and guides should try and carefully elaborate the way
tribunals should deal with procedural and evidentiary matters. Traditionally,
civilian lawyers have been more comfortable with generalist Codes, while the
common law tradition tended to involve very detailed and prescriptive statutory
drafting, although there is now greater convergence between the statutory
models of the two families and an understanding of the trade-offs that each
necessitates.

There is a lively debate amongst leading arbitration practitioners as to the

appropriate trade-off between flexibility and certainty. To some, attempts to
overly circumscribe arbitral behaviour simply denies experienced arbitrators
the ability to tailor solutions to the needs of a particular case. They will often
argue that in many page "23" circumstances it is simply impossible to be precise.
Similarly, some would argue that rules should not aim to ‘micro-manage’
disputes. Overly prescriptive rules can also lead to particular problems where
they are not comprehensively followed. Should a failure of a party to follow
rules in their entirety be a basis for it losing the ultimate case; should
substantial compliance suffice; or conversely, should procedural failings only
lead to costs decisions and not affect substantive outcomes? A contrary view is
that arbitrators extolling flexibility are simply concerned to be unconstrained
and rely unduly on their own experience and expertise to do justice in individual
circumstances. In particular, given that there are so many different arbitrations
and arbitrators, increased certainty ought to promote more uniform
approaches, which should of itself be good for the development of clearly
understood arbitral models. In that sense uniformity is a distinct value.

As with any area of trade-off of values, neither view is inherently correct. A

proper understanding of the reasons for differences in view should at least
guide sensible behaviour. There are two key reasons why high levels of clarity
may not be possible in any rules or guides. The first is that trying to predict how
all future situations should be dealt with will typically fail to consider some
important permutations that might require differing solutions. The second
reason, commonly applicable in arbitral procedure, is that if the parties and
counsel come from legal families with very differing approaches to problems, it
would be hard to agree on a common solution.

One value of articulated codes, rules or guides is that efforts to determine

procedures by way of ad hoc consent can be time consuming and hence costly.
That is exacerbated if there is no agreement between the parties, requiring
discretionary decisions by the tribunal, often after conflicting submissions by
the parties. In some cases, there may need to be face-to-face hearings to
determine procedural matters. When a tribunal has to make a determination on
a contested matter of this nature, there is also a problem in that the party that
loses the point may feel that the tribunal is against it. These issues are taken up
again in sections 3.10 and 6.5.

1.2.14. Reconciling Conflicting Policy Criteria

As noted throughout the above discussion, any area of legal regulation will
inevitably involve some trade-off between fairness and efficiency. If a
philosophically just or fair theoretical solution is economically inefficient or vice
versa, how are we to respond? In the early historical development of any
adjudicatory system, considerations of fairness will tend to dominate. Policy
makers find it hard to openly acknowledge compromise-inspired cost-effective
positions that are not simply aimed at maximising fairness. Over time, practical
problems arising from myopic concerns for fairness lead to efficiency and
certainty problems that are often seen as equally problematic, or at least as
being highly significant to stakeholders. Thus over time, some regulators and
practitioners have sought to elevate concern for efficient solutions. They might
seek to justify this through a consent-based paradigm, arguing that the parties
chose arbitration in the hope of efficiency and page "24" informality gains over
the litigation process. In any event, it is now well accepted that optimal
arbitration should aim to promote both fairness and efficiency.

No rule system could possibly articulate an optimal trade-off between the

various policy criteria. Once it is understood that any general rules and
procedures are compromises, it must be acknowledged that their application
may appear to be biased for or against one or both parties in a particular case.
(60) This is simply because no a priori trade-off can be neutral as to how it will

then apply in all fact permutations. Some of the more perennial debates in
arbitration are thus overly simplistic, failing to acknowledge the importance of
trade-offs between meritorious but conflicting values. Polarised critics observe
their opponent's solution, hone in on the worst aspects as if their identification
is proof alone, without acknowledging the internal inconsistencies with their
own positions. Considering policy perspectives on each procedural issue in
arbitration may at times show that preferences have been made in the rules.
More often, such an exercise will show potential benefits and problems at each
stage that should at least forewarn a tribunal to be both proactive and reactive
in order to minimise problems and maximise benefits. What is important is to
ensure that all criteria are given due consideration. Thus a number of arbitral
laws and rules not only refer to fairness and due process criteria, but also allude
to expeditious and economical proceedings. For example, Article 21 of the
Swedish Arbitration Act states that disputes must be handled in a ‘practical, and
speedy manner’. The English Arbitration Act 1996 indicates that ‘the object of
arbitration is to obtain fair resolution of disputes by an impartial tribunal
without unnecessary delay or expense.’ (61) Importantly these articulations put
both potentially conflicting norms within the same statement of objectives,
hence showing that there is no a priori ranking.

One approach to effecting the trade-off is to identify different institutional rules

and their different approaches to key elements and select according to the
parties preferences. Yet another approach is to attempt a case-by-case solution,
but informed by general principles. Because each arbitration brings together a
different mix of arbitrators, parties and counsel from differing legal families and
experiences, ‘each arbitral tribunal for each international arbitration has to
strike a new balance between the fairness and the efficiency of the
proceedings…’ (62) The challenge for international arbitration is to respect a
range of potentially conflicting goals. These include respect for party autonomy,
deference to the legitimate concerns of governments who provide the legal
infrastructure for international arbitration, fairness to the parties, efficiency,
flexibility, certainty and neutrality in the face of distinct norms and values of
different legal cultures. These challenges are all of central concern for the
development of optimal page "25" procedural models, without which arbitration
cannot meet the objectives set by those using it.

The most controversial issue alluded to above is that of proportionality,

although domestic systems now accept this expressly or impliedly.
Proportionality involves setting time and money limits proportional to the value
of the dispute even where this must have some impact on the quality of the
adjudicatory analysis and hence the outcome. ‘A notion of proportionality lies at
the heart of intelligent truth-seeking in arbitration, accommodating the
interconnected pillars of due process and efficiency.’ (63) This is discussed further
in section 6.2. Based on the foregoing analysis and the developments in
institutional rules and guiding notes from respected bodies, one possible
articulation would be to say that modern arbitral procedure aims to maximise
objective truth within a cost structure that fits within a range that is reasonable
in the context of the commercial value involved. David Rivkin has neatly
identified the goals of international arbitration as: ‘(i) a fair and neutral process,
(ii) conducted by intelligent and experienced arbitrators, (iii) resulting in a
timely and well-reasoned decision, and (iv) benefiting from an effective
enforcement mechanism.’ (64)

A further problem in reconciling conflicting policy criteria even within such

dictates emanates from differences in view as to how to rank the sources of
arbitral power and responsibility. Here the key issue is the debate as to the
respective roles of consent and arbitral laws. This is discussed in the following
sections.

1.3. Consent Versus Jurisdictional Theories of Arbitration and Their

Relevance to Evaluating Procedural and Evidentiary Models

The previous section looked at the values of fairness and efficiency and
considered the problems when they conflict. Arbitration builds on another
problematic interplay of potentially conflicting values in terms of the role of
consent as opposed to the role of the national legal systems that support
arbitration and which support enforceability of arbitral awards and at times
allow for annulment. These two features of international arbitration raise a
perennial debate about its essential nature. How can a private agreement to
avoid national court systems lead to binding and enforceable awards? To what
extent is arbitration essentially consent-based, or alternatively, is it essentially
based on the jurisdictional support of national legal systems and international
treaties? These are not purely theoretical questions that operate in a vacuum.
The question of arbitration's essential nature is the single most important issue
in determining how different arbitrators page "26" approach contentious
questions and approach the exercise of procedural and evidentiary discretions.
The views one holds on this question can have an important impact on the way
one views rights and powers of arbitrators generally, their status, the relevance
and application of mandatory substantive laws, the circumstances in which a
tribunal may override an agreement of the parties and the general approach to
proactivity by the tribunal in pursuit of fairness and efficiency. Hence, some
further explanation of the concepts and their inter-relationship is appropriate.

Arbitral scholars and practitioners have varied in their views about this issue
over the years, but it is important to understand that there is as yet no
consensus on any detailed prescription. Nor is there ever likely to be. The
important corollary of this is that all contentious questions within arbitration
must be considered in the context of conflicting views about its essential nature.
Those who prefer a consent-based paradigm will primarily look for procedural
solutions per medium of consent. Those who see arbitration as essentially
jurisdiction-based, may prefer solutions respected by the jurisdictions that
support a particular arbitral process.

Historically, scholars and commentators advocated quite extreme positions. FA

Mann argued forcefully for a jurisdictional perspective, questioning how any
form of behaviour can be seen as valid without some national legislative
support. (65) A party can only arbitrate in a particular location if the laws of that
State permit this to occur. Those inclined to the jurisdictional view would also
be more inclined to see arbitrators having a quasi-judicial function, an arbitrator
being an alternative dispute resolver empowered as such by the relevant State.
(66) In turn, considering arbitrators to have quasi-judicial status can have

implications for questions of impartiality and challenges for bias and also in
relation to arguably inherent powers, rights and duties. It would also imply
immunity even where this is not contractually bargained for by an arbitrator.
Conversely, scholars such as Réne David argued for delocalised arbitration,
completely separated from national jurisdictions based on consensual and
contractarian theories. In a 1970 arbitral award, Pierre Lalive stated that
‘international commercial arbitration may be entirely “detached” or separated
from the “national” laws of the parties: it shall only be governed by the rules of
arbitration chosen by the parties or referred to by the parties in their
agreement.’ (67) This consent-based or contractual theory builds on the notion
that arbitration only exists because of agreement of the parties. The ultimate
expression of consent is the freedom to be detached from any national
supervisory system. Proponents of that theory do acknowledge that when
enforcement is necessary, resort must be had to a national system, but the
presence of that system is not essential to the very concept of arbitration.
Gabrielle Kaufmann- page "27"Kohler describes the consent-based approach as
a subjective test looking to the intent of the parties. (68) She describes the
jurisdictional approach as being determined via an objective or territorial test.
(69)

Another way to highlight the tension between consent and jurisdictional

paradigms is to question whether arbitration is simply the fulfilment of a
demand for professional business services or is instead a means to provide
equivalent standards of natural justice to that of domestic litigation. (70) Given
the ongoing debate, it is not surprising that various rule systems do not express
a clear preference for consent over jurisdictional theories or vice versa.
Invariably it is not a complete choice between consent or jurisdictional models,
but an understanding of how the two conflicting theories are best reconciled.
Article 1(2) of the UNCITRAL Model Law provides for a territorial link to its
operation within any jurisdiction. Nevertheless, Article 19 provides that subject
to certain exceptions, the parties are free to agree on the procedure to be
followed by the tribunal in conducting the proceedings. Party autonomy as to
procedure is subject to the overriding obligations in Article 18 that the parties
shall be treated with equality and each party shall be given a full opportunity of
presenting its case. Thus we see a jurisdictional link in Article 1(2), mandatory
procedural laws in Article 18 and subject to those caveats, full deference to
party autonomy. (71) Such a framework typically applies in other lex arbitri.

Judicial attitudes in supervisory and enforcement courts would also be

important. For example, in Naviera Amazonica v Compania Internacional (72) the
English Court of Appeal in obiter dicta considered that an arbitration might exist
in a delocalised form but this may render it practically unworkable. This logic
was approved in American Diagnostica v Gradipore. (73) A converse jurisprudence
has been considered in continental systems although there is no consistent view
or even dominant view against some jurisdictional element. (74) Different
governments may also have very different attitudes to delocalisation. It is less
likely to be acceptable page "28" to centrally planned economies that do not
respect party autonomy as an ultimate value.

Enforcement norms can also impact on the analysis. Article V.(1)(d) of the New
York Convention allows for challenges where the composition of the arbitral
authority or the arbitral procedure was not in accordance with the agreement of
the parties or failing such agreement, was not in accordance with the law of the
country where the arbitration took place. This indicates that party autonomy
takes priority over the jurisdictional procedural laws of the Seat on that issue at
least.
Over time, more scholars have accepted that arbitration is inevitably a hybrid,
albeit one where there has been no consensus as to the relative importance of
the two main sources of arbitral validity. The two conflicting perspectives are
both valid. Arbitration cannot exist without consent. However, an arbitrator will
derive rights, powers and duties from the relevant arbitral law in the Seat of
arbitration. Hence the reality is a hybrid of the two perspectives. More recently,
some scholars have tried to move forward without resolving that debate by
considering a functionalist view of arbitration. This is also described as the
autonomous or sui juris theory. (75) Such an approach is concerned to look at
arbitration as it is and to identify ‘what it does, what it aims to do, how and why
it functions in the way it does. It recognises that the relevant laws have
developed to help to facilitate the smooth working of arbitration.’ (76)

While it is easy to state that a hybrid or functionalist approach may be more

accurate or helpful, a significant problem is how to reconcile the differing
elements where they pull in different directions on specific problems. For
example, what should an arbitrator do if the parties agree to a procedure that
the arbitrator believes is unfair or inefficient? Should an arbitrator ignore
substantive rules that would see the parties' transaction as problematic where
the parties direct the tribunal to ignore those provisions or simply do not raise
them in submissions? Must an arbitrator apply national laws of this nature, such
as competition or securities laws, even without any direction to do so by the
parties? Can the parties agree to arbitrate matters that particular national
systems might see as inappropriate for arbitration? A hybrid theory does not
specify which of two valid perspectives should win when they clash in the above
examples. A functionalist perspective, being essentially sociological, can be more
concerned to first articulate what law is before identifying why it ought to do
certain things in such circumstances.

Because there are so many similar situations where the extreme versions of the
consent and jurisdictional paradigms would tend to differing views, some might
seek to delve deeper into theoretical foundations in the hope of developing a
more scientific means of resolving such questions, or at least in order to better
understand why differing viewpoints form in the way they do. Others warn of
the page "29" problems of any scientific analysis. To this end, Emmanuel
Gaillard has spoken of the ‘representations’ of international arbitration. (77) In
Gaillard's view, the ‘visions’ or ‘representations’ of international arbitration we
have inform the views we take on what may appear merely technical issues. In
describing them as visions or representations, he points out that they relate to
controversies over which there cannot be demonstrated scientific resolution,
but rather, involve matters of belief or faith. As just one example, he makes the
important observation that simply speaking of party autonomy or consent is
inadequate as there is a need to consider ‘the more fundamental question of the
origin and binding character of the principle of party autonomy.’ (78) Gaillard
identified the first representation as that which equates the arbitrator with a
local judge. The source of power is the legal order of the Seat. That can also
claim a consent-based justification where the parties select the Seat or delegate
that choice to an institution or the tribunal. The second representation
concentrates on the ultimate outcome and the need for enforcement where an
award is not voluntarily complied with. On this representation, ‘the legal order
of the place of enforcement legitimises a posteriori the whole arbitral process.’
(79) The legal validity stems from whichever legal orders are willing under

appropriate conditions to recognise the effectiveness of the award. The third

representation contemplates the body of States prepared to recognise and
enforce awards collectively as opposed to individually, thus removing the
arbitrator from the notion that they are legitimised by a particular State. (80)
Such an approach situates arbitration within a global context, theoretically as
well as physically. Whichever representation is embraced will determine the
extent to which the arbitrator abides by court decisions in the Seat, applies its
procedural rules, applies mandatory laws or applies national or general
principles.

Again, this book does not suggest its own scientific theory or critique. The point
instead is that if one digs deep enough into any controversial practical question
of arbitral procedure and evidence, one returns to these issues. The following
sections look more deeply into the notion of consent in the terms of its impact
on procedural and evidentiary questions.

1.4. Consent as the Basis of Procedure and Limits on Party Autonomy

The previous section looked at the question of consent vis-à-vis jurisdictional

bases for arbitral procedural models. It also looked at more nuanced and
sophisticated page "30" theories aiming to deal with the way in which different
participants and observers reconcile these concepts. Regardless of one's view on
that issue, all would agree that consent is important and acknowledge that party
autonomy is a fundamental aspect of all aspects of arbitral procedure,
commencing with the arbitration agreement itself. Consent is also important in
determining the trade-off between fairness and efficiency. Anything that is
consistent with the parties' actual choices has some claim to being arguably fair
for that reason alone and consistent with the essential nature of arbitration,
even if it may add to the expense.

Having said that, there are vital questions of evidence and procedure in
ascertaining consent. In a book dealing with these issues, particular attention is
given to the way consent is first identified and then how it is integrated with
other sources of rights, duties and powers in order to determine how some of
the contentious issues outlined above might best be dealt with. The following
discussion looks at some of the sub-elements of consent in the context of
fairness and efficiency as guiding criteria by which to evaluate arbitral models
and by which to make procedural determinations where discretions exist.
Particular aspects of reconciling consent, fairness and efficiency on an issue-by-
issue basis are dealt with in individual chapters.

1.4.1. Express and Implied Consent

Where consent is seen as relevant, this raises contentious questions about the
evidence on which it is to be determined. As with any form of agreement, it is
natural for legal systems to accept both express or implied forms of consent.
Even express consent may be ambiguous, calling for some form of
interpretation. This may itself be problematic. Implied consent is likely to be
even more uncertain. In arbitration, parties may have little understanding or
experience with such forms of dispute settlement. Hence it is difficult to
determine what would be an appropriate evidentiary basis to identify implied
consent.

These questions are particularly difficult given that they will only arise when the
parties are in dispute on a procedural matter. This is because if the parties have
a newly agreed position during the arbitral process, they can ask for this to be
respected even if there had been previous disagreement on the issue. Hence
express and implied consent will be evaluated in contested situations. Dealing
with disputes as to express consent will typically relate to contests about the
meaning of the arbitration agreement itself. Here the first question will be the
approach to interpretation, whether it be based on selection of applicable law or
some presumptions of intent. There may also be overriding interpretational
predispositions such as promoting effectiveness of the arbitration agreement.
This is discussed further in relation to potentially pathological arbitration
agreements in section 3.2.7.

Some of the issues in that regard also impact upon identification of implied
consent, given that pro-arbitral interpretative presumptions build on notions as
to what parties impliedly hope their arbitration agreement would lead to. At one
extreme, tribunals might be reluctant to readily imply consent and could look
page "31" instead for cogent evidence in the face of such disputes. At the other
extreme, some might argue that parties could regularly be presumed to have
impliedly consented to efficient procedures. That could even be so for reluctant
disputants who are clearly engaging in dilatory tactics, as the relevant consent is
that which was in existence at the time of the original agreement to arbitrate.
Furthermore, parties can be presumed to have intended to arbitrate in good
faith, hence later tactical behaviour is even less valid as an indication of prior
intent. Nevertheless, such approaches are contentious, not only in the context of
consent versus jurisdictional paradigms, but also in terms of whether such
arbitrators are truly justifying efficient solutions through a consent analysis, or
are instead paternalistically imposing their own desired processes in the guise
of this source of authority. The point is not to argue for one particular view over
the other, but instead, to point out the central role of evidentiary determinations
of intent in identifying one of the foundations of the entire arbitral process.

1.4.2. When Can or Must an Arbitrator Ignore Procedural Choices Made by

the Parties?

Once consent has been determined, there are then questions as to whether the
parties are entitled to agree on that particular matter, whether the tribunal is
bound by that agreement or whether it can or must ignore it in certain
circumstances. In the latter event, it is then necessary to determine what those
circumstances are. Inevitably, one's answers to these questions depends on
one's overriding view as to the nature of arbitration. A purely delocalised
consent-based view would strongly defer to party autonomy. Conversely, a
strictly jurisdictional perspective would look to see what the law of the Seat
accepts as permitted behaviour. A view which concentrates on enforceability,
might look at party agreements that would impact on enforceability. Under the
hybrid view, most would agree that the parties cannot agree to override a
mandatory procedural law pertaining to the arbitration, typically derived from
the law of the Seat. Examples of mandatory procedural laws are the duty to treat
parties equally and give each an adequate opportunity to present their case. A
contractual view could come to the same conclusion by implying due process
and good faith as inherent terms in any agreement to arbitrate.

In addition to mandatory procedural laws, an important question is what

attitude an arbitrator should take to procedural decisions arrived at by consent
of the parties which are not in alignment with the arbitrator's own fundamental
values about optimal procedural fairness. For example, how should an arbitrator
react if the parties do not wish to present witness testimony in a case that could
not be confidently resolved without it? How should an arbitrator proceed if the
parties wish to take a lengthy amount of time for witness cross-examination and
submissions, where the arbitrator would know that the total transaction costs
will far exceed the amount in dispute? Some would not wish to interfere with
the parties' choice. Others might seek to give a caution to the parties, invite them
to reconsider, page "32" but accept their decision if they do not resile from it.
Other arbitrators would say that they have an obligation to the institution of
arbitration itself to ensure that it is conducted in a fair and efficient manner and
would not accept directions or continue in the case if the directions led them to
feel that fair and efficient outcomes would not be possible.

What an arbitrator can or must do in the face of party consent is also impacted
upon by their own rights and powers, in part emanating from the arbitration's
legislative basis, in part from party consent, express and implied, in part from
arguably inherent powers and in part from arbitrator's contractual relationship
with the parties. This complex mixture of powers and duties needs to be
separately analysed and integrated, both generally and in relation to specific
scenarios. These issues are discussed further in Chapter 2.

1.5. Insights from Empirical Studies

Previous sections have looked at fairness and efficiency as the criteria by which
to evaluate arbitration and the consent versus jurisdiction paradigm. This was
done so as to understand its nature and determine how trade-offs ought to be
made when the key goals conflict. Such an approach is deductive, seeking to
identify an optimal model from theoretical insights. An alternative approach is
inductive, looking to past practice to discern problems, ideal responses and the
preferences of users.

In considering current practice, practice trends and reform options from an

inductive perspective, we would ideally wish to be able to evaluate
developments to date. Hence, there is seen to be a value in surveying and
analysing participants and arbitral data. Indeed there have been a range of
studies in recent times, the results of which have been broadly published and
widely debated. Before considering the insights of such studies, it is important
to be aware of the problems of statistical analysis in the arbitral field. Even
broad-ranging surveys may be subject to significant methodological flaws.
Bühring-Uhle et al., (81) note that empirical studies are rare because it is difficult
to generate appropriate data on the ‘inner workings’ of international arbitration.
Arbitral awards are generally confidential. An individual arbitrator, even
unconstrained by confidentiality obligations can only make impressionistic
comments on a relatively small percentage. There are also a relatively small
number of cases within each discernable category that might be tested. For
example, it is difficult to compare the performance of various institutions in
relation to jurisdictional or impartiality challenges as the data sets on each
factual matrix are small and it would be contentious as to the way factual
distinctions are warranted. The empirical work may also be problematic if it
mixes investment and commercial arbitration. For example, arbitrating with
States will have inevitable delays, problems of enforcement and greater
uncertainty as to the nature of applicable laws as well as problems in obtaining
evidence from the opposing party.

page "33"

It is also difficult to rely solely on the impressionistic views of participants.

Arbitrators have a vested interest in perceiving the system and their behaviour
in particular ways. Certainly those centrally involved have unique insights and
expertise, but the parties themselves may be unduly influenced by the outcome
if the surveys are left until after the award is published as is the case with
various institutions' satisfaction surveys. There may be a major difference in
outcome if the survey is conducted prior to the submission of the award. At this
stage the parties would simply be evaluating the procedures from a due process
and efficiency perspective. However, once the award has been rendered, the
losing party may find its views coloured by concerns with the outcome and
perhaps the reasoning. For example, if an arbitrator made a fair and reasonable
decision to limit the number of witnesses, once an adverse finding is known, the
losing party might wrongly think that a different procedural decision might have
led to a different outcome, hence giving a negative review of the arbitrator
concerned. Furthermore, any regulatory measure establishes certain incentives.
This is desirable when these are positive but there is also a need for concern as
to negative incentives. For example, an arbitrator that knows that she or he is
being surveyed and that this will impact on an institution's future
recommendations, may be discouraged from taking a strong stance when one or
both parties are being tardy or are inappropriately trying to run the matter
identically to domestic litigation models, simply through inexperience.

When looking at statistics, we must also always remember that correlation does
not prove causation. At times, a survey will simply highlight a problem without
giving an accurate identification of its cause or causes. For example, are
excessive costs and delay the fault of arbitration per se or the behaviour of
external or in-house counsel or both? Does it arise from undue deference to
autonomy or ready application of indemnity costs awards? Is there a mixture of
causes and if so, in what proportion? In some cases we may be able to advance
our response to such questions through analytical reasoning. For example, a
priori, one would presume that there would be little difference in counsel
behaviour between arbitration and litigation. Hence the key distinguishing
feature is the discretionary behaviour of arbitrators and institutions and the
direct costs that they impose.

With these caveats in mind, it is still appropriate to consider and critically

evaluate the insights from various empirical studies. Various surveys have
sought to rank the benefits that stakeholders aspire to through the use of
international commercial arbitration. This can help guide procedural
determinations and regulatory reforms when trade-offs must be made. A study
by Naimark and Keer in 2000 considered that a fair and just result was far and
away the most important objective, significantly outweighing speed, arbitral
expertise, cost and receipt of a monetary award. (82) The results are to be
contrasted with a more recent study which did not rank a fair and just result as
a key criterion. In 2008 a survey was conducted page "34" by Price Waterhouse
Coopers and the School of International Arbitration at Queen Mary College,
University of London. Responses from in-house counsel showed that 73%
preferred to use arbitration either distinctly, or in conjunction with mediation.
Ninety five per cent expected to continue using arbitration. In order of
importance, advantages cited were flexibility, enforceability of awards, privacy
and the ability to select arbitrators. (83) Obviously the disparity in aims is itself
important. Differences may arise because of different participants, different
times and different formulation of the questions. For example, multiple criteria
forced ranking of values may be problematic for a range of reasons. Anyone
asked to consider whether they value a fair and just result will obviously be
strongly disposed to say yes. Yet as noted above, a fair and just result to some, is
simply the outcome of a just procedure.

As noted above, another methodological issue is whether participants are

surveyed about their general life experience with arbitration, or about their
experiences in a current arbitration. In the latter case, we can differentiate
between those surveyed during the currency of the proceedings or those who
are surveyed after the conclusion for the reasons noted above. The Naimark and
Keer study surveyed participants both before the first hearing and after the
award. Some survey results after the award still show that losing parties are
able to rank arbitrators highly, which may be used to argue for the validity of
this approach. Michael Hoellering noted some time ago that surveys of AAA
arbitrators showed that responding parties ranked arbitrators as excellent to
good in all cases. Ninety per cent indicated that they would choose the same
arbitrator again. (84)

Key findings that have led to the greatest debate have been in the context of
costs. While the Queen Mary–Price Waterhouse survey reiterated the traditional
advantages, the survey participants highlighted significant concerns with
expense and length of time for resolution of disputes. (85) Alarmingly, 65%
believed it to be more expensive than international litigation, while a further
23% believed it to be equally costly. When evaluating the views of those
preferring arbitration, the key aspect was a clear willingness to try and use
mediation in conjunction with and hopefully as an alternative to arbitration, the
latter to be used in due course and only if necessary. Thus the respondents were
not articulating a preference for arbitration as the primary dispute resolution
method, but instead, a preference for it as a fallback, as against international
litigation.

While an express preference for mediation or indeed negotiation as an

alternative is only to be expected, the concern for delay and cost is more
problematic. A recent study by the Corporate Counsel of International
Arbitration Group (CCIAG) considered that international arbitration ‘takes too
long’ and page "35" ‘costs too much’. (86) Sophisticated in-house counsel are
aware that adjudicatory proceedings will have significant costs. For these
experienced professionals to articulate particular concerns about the costs and
delay in arbitration is a suggestion that the costs and delay are higher than their
expectations. Excessive costs may even be a barrier to cases being brought in
the first place. External counsel will often give estimates on being approached
and may demand advance fees, guarantees or monies in trust. If the expectations
of the corporate counsel surveyed were reasonable, arbitral practice contrary to
those expectations is hence problematic. If their expectations are not
reasonable, there is at least an educative problem with a mismatch between
expectations and reality that itself will require a solution. For arbitration to be
optimised. In this context, it would appear that impressionistic views, survey
results and empirical analysis all point to some problem with timeliness in
international arbitration. A recent study showed that the average time in which
an award is rendered is some fifteen to eighteen months after the close of
proceedings.

Here there is a need to temper empirical findings with analytical reasoning,

particularly if appropriate solutions are to be identified. One reason why
arbitration may systemically take longer than litigation is the separate need to
constitute the tribunal. Court procedures apply immediately on service of
proceedings and judges as public servants are already in place. Furthermore,
they are required to be available as and when a coordinating authority sets them
down to hear particular matters. Arbitrators, however, are entitled to try and fit
key stages into their schedule of other commitments. A related problem is that
court coordinators will always seek to balance the workload of judges, while it is
up to the integrity of arbitrators whether they take on too much work or not. If
they do, excessive delay will be unavoidable, given their inability to schedule the
particular arbitration in a timely fashion. Another reason why arbitration can
take more time than litigation is that it lacks a mechanism for summary
judgment. A party without any legitimate defence who chooses not to
participate, can still expect that the tribunal will force the claimant to present its
case in a thorough manner. Because of the importance of such issues, individual
problems and responses will be dealt with in individual chapters below.

1.6. Insights from Economic Theory: How Arbitral Persons Respond to Price
Signals

Another approach to regulatory reform and analysis is to consider likely

behaviour from an analytical perspective. In considering trade-offs between
fairness and efficiency and procedural regulations in general, it is important to
give adequate consideration to the way rule changes will affect behaviour.
Ideally, the effects will page "36" be positive and will support the very objectives
sought by the rule change. Yet complex law reforms often have costs as well as
benefits. Often the costs are unintended. For example, it is intuitively reasonable
to oblige tribunals to give parties a full opportunity to prepare their case.
Nevertheless, such a norm allows strategic threatening behaviour by some
parties who aim to abuse that right to delay and obfuscate the proceedings.

Importantly, this example shows not only that rule changes will affect party
behaviour but also may affect arbitrator behaviour. The more an arbitrator is
concerned to pre-empt challenges to the award, the more the arbitrator might
accede to such requests during the arbitral process. Unduly intrusive
supervisory attitudes by enforcement courts would also tend to limit the
creativity of tribunals. Arbitrators determined to keep matters efficient and
party focused might become too concerned by threats of challenges to
enforcement because of denial of due process. The fact that arbitrators might
respond to economic price signals has been noted by Dezalay and Garth and also
by Bruno Oppetit. (87) Similar comments have been made by Pierre Lalive (88) and
Lord Mustill. (89)

An economical analysis of the arbitral process could also suggest a number of

undesirable incentives in terms of arbitral bias in favour of the appointing party,
undue deference in the hope of having repeat work and mutual support in
gaining appointments. This is not to suggest that this occurs to any significant
degree or even at all. At the very least, however, if participants are aware of the
likely economic incentives, they can ensure they maintain the highest ethical
standards and fight against any such urges or pressures if and when they arise.

1.7. Methods of Harmonisation or Convergence

If there is to be any practical utility in a theoretical analysis of the aspirations

and problems in arbitral procedure and evidence, it will come in the form of
some principles and criteria by which reforms should be made and discretions
exercised. One option is to consider the value of harmonisation or convergence
in laws, rules and practices. A key reason why harmonisation or convergence is
a valid aspiration is the sheer diversity between legal families on questions of
procedure and evidence. Such divergence is far more significant than differences
in substantive law. Arbitration can try and resolve such differences afresh or can
in addition, look to other models for inspiration. In addition to international
adjudicatory processes page "37" which require some harmonisation or at least
consideration of alternative perspectives, so too has the development of the
European Union forced a resolution as to the way to bring together common law
and civilian perspectives. Inspiration might also be sought from domestic
developments, although it is important to again understand the insights of
comparative law that transplants cannot be expected to operate optimally
without due consideration for differences in culture and procedural norms.
Nevertheless, efforts have been made to harmonise procedure in domestic
litigation. UNIDROIT has sponsored a study which led to draft Principles and
Rules of Transnational Civil Procedure. (90)

There are a number of ways in which convergence of arbitral procedural values

and processes may occur. These are not mutually exclusive. At one extreme is
the attempt to promote uniform procedural laws and arbitral rules. Here the
work of UNCITRAL is the most significant, through the establishment of the
UNCI-TRAL Model Law on International Commercial Arbitration and the
UNCITRAL Arbitration Rules. As the name suggests, the Model Law has become
a recommended framework for any country seeking to have a statutory regime
supporting arbitrations conducted within it. The UNCITRAL Arbitration Rules
are not only a discrete set of ad hoc rules in their own right, but also constitute a
permitted model for individual institutions to use and/or modify as they see fit.
Many institutions have built their own rules in this way. While harmonisation
initiatives exist, the bulk of the work occurs within such individual institutions.
Even here we can see a significant convergence amongst national procedural
frameworks for international arbitration and various institutional rules on some
key issues. Institutional developments together with ad hoc practices and
scholarly analysis have already done much to promote desirable convergence of
procedural elements. Such a process will inevitably continue, aided by
international and comparative examination such as those promoted by
conferences.

One contentious issue relates to the degree of detail or specificity in rules. Park
suggested that it is only sensible that institutional rules provide detailed
procedural norms, subject to the parties' rights to opt out. This will help
arbitrators avoid making an undue amount of procedural determinations that
inevitably bias one legal culture over another. (91) It is certainly true that such
models can only aid certainty and consistency, but the problem is in part
circular. If truly neutral, fair and efficient procedural criteria could be
encapsulated in procedural rules, page "38" one would expect that international
arbitrators, even from different legal cultures, would gravitate to such norms in
any event through practices, experiences, scholarly writing and seminar
discussions. However, Park rightly observes that if a norm is widely accepted for
these reasons, it ought to be set forth in a clear rule. If it is not widely accepted,
then it may be desirable to discourage an arbitrator from adopting such a
practice. (92)

Yet there are still significant differences between arbitral laws and rules. These
are not necessarily a bad thing as they give potential arbitration participants an
option of selecting different models. For example, the English Arbitration Act
1996 gives much broader rights of appeal on questions of law to English courts
than is generally found within other arbitration laws. (93) Swiss arbitration rules
allow much broader rights to bring set-off claims from separate activities that
are not otherwise directly covered within the jurisdiction of an arbitration
agreement. (94) Some observers could see these unique features as desirable
additions to arbitral norms, or at least useful alternatives to be considered on a
case-by-case basis. Others might see them as being more contentious,
representing undesirable variations from a happy medium. For example, English
rights of appeal might be valued as a safety valve or criticised as having less than
ideal respect for arbitral determinations. The Swiss model could be argued to
allow for efficient and comprehensive dispute resolution or be asserted to go
too far in trying to deem consent beyond the four walls of the arbitration
agreement in order to more efficiently resolve all disputes between parties. The
aim of this discussion is not to express a view on these issues, but merely point
out that various models will have advantages and disadvantages about which
observers can legitimately disagree. The other observation is that when
institutions attempt to objectivise procedural norms, they have to broadly
choose between alternatives where the above factors may play out. Some might
try and establish unique features in the hope of attracting business. After all,
institutions are competing in the marketplace for custom. History has shown
that such experiments can be dangerous. An example was Belgium's short-lived
attempt to promote autonomy of arbitration by virtually removing all aspects of
a court's supervisory jurisdiction. (95) Most institutions would tend to gravitate
to a common norm for fear of losing business. From a risk management
perspective, it is better to emulate the features of the most respected
institutions than to seek to go it alone. Where parties have equal bargaining
power, both sides would need to be confident in the value of unique provisions
before actively selecting an atypical institution.

Another approach to convergence is the development of recommended

practices. Examples include the UNCITRAL Notes on Organising Arbitral page
"39" Proceedings, (96) the IBA Rules on the Taking of Evidence in International
Commercial Arbitration, (97) and the IBA Guidelines on Conflicts of Interest in
International Arbitration. (98) A third approach is simply to rely on the reflections
of leading practitioners and scholars in trying to find the best features of all legal
systems in advancing arbitral procedural norms. A number of works have
provided important guidance in that regard. (99) In that sense, harmonisation and
convergence can come from express rules or through a common approach to the
exercise of discretions, that is, convergence via rules or convergence via
methodology. Where harmonisation or convergence is sought to be developed
by rules, there is still a question of the way to go about drafting. Here there is a
contrast between the expansive versus general styles of drafting commonly
underlying historical differences between civil and common law systems. There
is then the way in which rules are interpreted. Even a common articulation of
rules will not necessarily lead to convergence if arbitral interpretation or
discretionary practice leads to a replication of domestic biases.

It is important to note, however, that not all harmonisation is necessarily

advantageous. Convergence may simply flow from economic actors following
inefficient incentives. Harmonisation initiatives are interesting from an
economic perspective as there are conflicting pressures. On the one hand, areas
with the greatest divergence between legal systems are those with the greatest
need for clarification, hence there will be a greater demand for such initiatives.
On the other hand, the significance of divergence makes it less likely that there
will a consensus and hence harmonised norms that may be articulated. This can
then lead to suboptimal compromises. For example, counsel from different
jurisdictions with differing views about the respective value of oral and written
submissions may end up agreeing to do both, with the disadvantage of excess
costs to the parties concerned. In some cases, matters that are controversial but
which are permitted in some legal systems, may end up becoming the norm
simply because of the belief that the duty of equality cannot allow one counsel
an unfair advantage over the other, regardless of the merits of the procedure
employed. An example is preparation of witnesses. In Park's view this can lead
to ‘a race to the lowest common denominator’. (100)

page "40"

One question recently debated is whether there has been undue

‘Americanisation’ of arbitration, in the sense of adoption of domestic norms
simply because of the economic power of certain parties in drafting, or because
of the role of American firms and arbitrators. (101) Arbitral experience to date has
shown that directly or indirectly, many of the more contentious common law
elements are incorporated to some degree at least. This is not necessarily
because common law perspectives dominate over civilian perspectives. It might
instead simply be a result of allowing all permitted processes from each legal
system. Thus we often see a number of rounds of written submissions, oral
testimony, cross-examination, applications for production of documents, valid
standing for all witnesses and the ability to prepare witnesses prior to a hearing.
Such an approach can easily be inefficient, leading to increased time and
expense and unnecessary duplication without careful control by the tribunal.
Furthermore, to select everybody's positive procedural elements is a rejection of
their negative ones, so overall it seems fair to conclude that common law norms
tend to dominate. To some law and economics scholars that is in fact a desirable
outcome. (102)

1.8. Comparative Law Perspective

A central reason why theory and practice inevitably combine in the field of
international arbitration is because different legal families have different
approaches to dispute settlement procedures and questions of evidence. Anyone
searching for an optimal model has to be alert to these differences and draw on
comparative law methodology with a view to discerning world's best practice,
or at least with a view to finding optimal compromises between different
systems' approaches. Legal systems rarely disagree because they have
fundamentally different values. Instead they disagree because they rank
irreconcilable values differently while aspiring to the best of all worlds.

While a comparative approach is thus important, it can be dangerous to discuss

procedure in the context of supposed differences between civil and common law
jurisdictions. This is because there are significant differences in approach within
legal families as well as between them. (103) For example, one division amongst
civil law countries is between those in the north where Germanic influence was
most page "41" significant and those in the south where the Roman tradition has
been more influential. Paulsson has also cautioned that merely considering
issues from common law, civil law or adversarial versus inquisitorial
perspectives is ‘more conducive to lazy thinking and unhelpful approximations
than to a reliable conceptual framework that would help fashion a transnational
modus vivendi out of the inconsistent patterns of national practice.’ (104) To some,
international commercial disputes ‘need not show so much consideration for
national legal traditions and the requirement of a certain regional and social
environment…’ but can ‘be formed on a worldwide common understanding of
values and standards which are the necessary bases of the international
exchange of goods, services and capital.’ (105)

More specific comparative analysis is left for individual chapters where

individual topics are discussed, such as document production and expert
witnesses. In this section, only general observations are made. It is important to
understand that differences in one area typically have logical implications for
other areas, hence leading to further differences between systems. For example,
if an adversarial system depends on the parties to generate the relevant
evidentiary record, it makes sense to allow for broad document request
processes to keep the parties honest. In terms of the general approach of legal
families and how this may affect procedural decisions, the common law trial
model involved a written pleading phase, a pre-trial phase dealing with
collection of facts and evidence and an oral trial phase. (106) The adversarial
model saw the adjudicator as ‘a passive judge who is merely the umpire of the
justice game between the litigating parties.’ (107) The Roman tradition also
typically adopted a three-stage process, the first being the written introductory
phase, the second fact finding by an instructing judge and the third, a final
hearing without further evidence-taking before a judicial panel. (108)

It has been noted that while common law systems were traditionally defined as
adversarial, ‘they are presently placing more and more emphasis on the role and
powers of the judge especially as far as the management of procedure is
concerned.’ Similarly, civilian systems traditionally labelled as inquisitorial,
‘have made clear their adherence to the principles of parties' initiative, and
adversary procedure’ and have at times adopted typically common law
institutions such as cross-examination, limited discovery and bifurcation of
proceedings into page "42" pre-trial/trial preparation. (109) Stürner describes
recent developments in England, Germany and Spain that could be described as
a ‘main hearing model’ with preparatory clarification after the written
introductory stage that aimed to clarify as much as possible before the main
hearing and also aimed to warn the parties of the matters that need final
determination. (110)

While developments in domestic litigation procedure may be influential in

arbitral reform and practices, the converse is also true, with modern arbitration
arguably being a key inspiration for moves to case management in common law
reforms and civilian jurisdictions such as France and Spain. (111) Litigation
studies have also been influential. In 2004, the Principles of Transnational Civil
Procedure were approved by the Governing Council of UNIDROIT and the
members of the American Law Institute. The ALI/UNIDROIT Principles opted for
the main hearing model, which has become the norm in international
arbitration.
It is important to acknowledge that a comparison between civil and common
law traditions is also incomplete in failing to give appropriate deference to the
values and practices of Asian and Middle Eastern systems. Colonial histories also
do much to shape individual developing economies, often leading to hybrid
systems with a highly dynamic nature as influences have changed. Islamic law
and jurisprudence is more complex, given that there is no uniform position as to
the role of Shari'ah law in the context of commercial matters. Islamic law varies
from countries where it has little impact on commercial matters to those where
it dominates. (112) Furthermore, many Islamic jurisdictions are mixed, some with
civilian backgrounds such as Egypt and Indonesia, while others such as Pakistan
and Malaysia have strong common law elements. From a procedural
perspective, there may also be a difference as to whether there is an ad hoc
arbitration in an Islamic country allowing resort to Shari'ah principles, or
whether there is an exclusive arbitral code that itself seeks to resolve all issues.
El-Kosheri suggests that ‘Islam recognises arbitration in its broadest sense as
the normal method of restoring peaceful harmony at all levels of human
relations.’ (113) This broad notion encompasses conciliation on the choice of
arbitrator known to and trusted by the persons involved. The latter led to
historical concerns with prospective arbitration agreements.

While some speak of Asian culture and Asian attitudes, this is even more
problematic than erroneous presumptions of homogeneity within civilian and
common law legal families. There are fundamentally different cultures, religions,
page "43" languages and political systems applying in differing Asian countries.
With that caveat in mind, it is at least worth considering observations that have
been made about distinct attitudes to dispute resolution methodologies and
processes in the region. Taniguchi speaks of a preference for a conciliation
culture in Asia as a whole, especially East Asia. (114) The Confucian tradition
considered that ethics and persuasion were preferable to compulsion and
legalism. (115) These differences in view can impact on questions such as whether
an arbitrator can or should assist the parties in reaching a mutually agreeable
solution and even whether institutional rules should bifurcate conciliation and
arbitration processes. (116) Differing attitudes may also lead to unexpected
implications at subsequent stages, with the suggestion that notwithstanding
many Asian parties' reluctance to enter arbitration, many will nevertheless find
it harder to settle disputes, being entrenched in their positions and being
annoyed at being forced to have an involvement in the process. (117) Conversely,
another suggestion is that in certain Asian jurisdictions, there is a ‘tendency
toward equity-based, compromise decisions…’. (118) Another aspect of Asian
culture is ‘face, or respect of self and others demonstrating respect for us.’ (119)
This has obvious implications in cross-examination scenarios.

There are also North/South perspectives that can be brought to bear. For
example, Sornarajah considers lex mercatoria to be ‘a creation of a coterie of
Western scholars and arbitrators who have loaded it with norms entirely
favourable to international business.’ (120) Whatever ones views on these
philosophical and political perspectives, a comparison between systems at least
sets an agenda as to the key areas where legal procedural models may diverge. A
comparative methodology will rarely provide a simple solution for the very
reason that systems tend to diverge on more problematic issues where there is
no simple solution to complex problems. A comparative methodology is useful,
however, because each system will tend to concentrate on one policy reason for
the choices it has made, which page "44" then has to be properly evaluated
against the conflicting policy choices of other systems.

1.8.1. Culture and Comparative Perspectives

Comparative perspectives can also be strongly influenced by cultural values,

which in turn can have much to say about the attitude to adjudicatory processes
per se. For example, alternative dispute resolution and informal processes have
long been preferred in China. (121) In considering culture, there is the potential
impact of cultural differences between the parties, culture of counsel and
culture of the tribunal. With multi-person tribunals the latter can be mixed.
When we speak of culture in the context of arbitrators and counsel, we are more
often concerned with their domestic legal traditions and ethical norms. The
term is used in differing senses when we consider parties on the one hand and
counsel and arbitrators on the other. From a party's perspective, we might
speak of the culture of dispute settlement and whether arbitration is in fact
valued as an appropriate methodology. Even though arbitration is based on
consent, hence showing support for that process by the individuals concerned,
because many transactions are offered on a take it or leave it basis, those
cultural perspectives can remain important. The arbitrators themselves,
however, freely choose to engage in the profession and would not have the same
a priori cultural concerns.

An important aspect of different cultural perspectives is that even where parties

consent to arbitration, they may have a fundamentally different view as to the
nature of the process and the way each participant should behave. Examples
might be proactivity by arbitrators and aggressive cross-examination from
common law counsel. Cultural attitudes may also impact on notions of truth and
how testimony should be given. (122) There are aspects of corporate culture in
terms of the way a party might behave and the way corporate officials may
conduct themselves. (123) Another aspect of culture involves the norms of the
particular industry or profession that is a party to the arbitration. (124) Differing
cultures may also have different attitudes to time and deadlines. That can have
implications for proactive arbitration and even timely attendance for procedural
conferences or hearings. (125) Legal cultural differences have also been discerned
in the page "45" methodology of writing and reasoning in awards. (126)
Prejudices in terms of intelligence and sophistication between first world and
third world parties can be particularly problematic. Neutrality is challenged
when arbitrators are aligned to one party only from a socio-economic
perspective. Western adjudicators treating a third world legal system as
primitive is a particularly sobering aspect of such cultural perspectives. (127) An
important aspect of culture is also the modern diversity in office-bearers,
arbitrators, writers, conference presenters and Secretariat officials.

1.9. Concluding Remarks

This chapter has sought to present a set of theoretical criteria within which to
pursue a practical analysis of all aspects of arbitral procedure. It has sought to
identify the crucial roles that philosophy plays through notions of fairness, that
economics plays through notions of efficiency and that sociology plays, both
generally and through notions of culture. These impact strongly on challenges
and solutions. The chapter has also sought to highlight the fact that optimal
goals from each theoretical perspective are likely to clash in the most
contentious areas of arbitral procedure. This is simply because we aspire to a
range of incompatible values and require trade-offs within regulatory design.
Both critical and comparative analysis needs to rise above simple criticism of
the defects of opposing models and instead seek to articulate why one second-
best trade-off is properly preferred to another.

In many instances, designing and implementing effective arbitral procedures

simply must get on with the job and cannot wait for any resolution of such
theoretical debates. In other cases, without at least some attention to the theory,
the chances of a uniformly respected solution are diminished. Where that is
thought to be the case, these broader issues will be revisited in individual
sections throughout this book.

page "46"

1 See Djibo Abdoulaye et al., ‘Global Corruption Report 2007’, available at,
<www.transparency.org/publications/gcr/gcr_2007>, 2007.
2 Many courts in the common law tradition will apply a forum non conveniens

standard to decide which is the most appropriate court, although different

common law countries vary as to the standards they apply. Other methods of
resolving such tensions include the lis pendens doctrine giving preference to the
first in time. This is discussed further in sections 8.8 and 8.9.
3 An example of an important regional arrangement for enforcement of
judgments is contained in Council Regulation No. 44/2001 of the European
Union, OJ L 012, 16/01/2001, 1–23 which replaced the Brussels Convention.
4 Convention on the Recognition and Enforcement of Foreign Arbitral Awards,

done at New York 10 June 1958 (entered into force 7 June 1959) 330 UNTS 38;
TIAS No. 6697.
5 Jean-Claude Najar, Chairman of the Corporate Council of International
Arbitration Group, has suggested that ‘(a)rbitration is no longer fulfilling the
basic need of business customers for early and efficient resolution of disputes.
We are increasingly turning elsewhere, to mediation and other forms of ADR’.
Jean-Claude Najar, ‘Users View on International Arbitration’, (Speech delivered
at Clayton Utz and the University of Sydney International Commercial
Arbitration Lecture, Sydney, 6 November 2008, quoted in Doug Jones,
‘International Dispute Resolution in the Global financial Crisis’, The Arbitrator
and Mediator 28, no. 2 (2009): 45).
6 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 88–89.

7 Examples include Australia and Hong Kong.
8 To many ADR theorists, mandatory mediation is an oxymoron.
9 Doug Jones, ‘International Dispute Resolution in the Global Financial Crisis’

The Arbitrator and Mediator 28, no. 2 (2009): 45.

10 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 73.

11 J Gillis Wetter, ‘The International Arbitral Process, Public and Private’ (1979),

vol. II, 288 cited in Sigvard Jarvin, ‘The Sources and Limits of the Arbitrator's
Powers’, Arbitration International 2, no. 2 (1986): 140.
12 V.V. Veeder, ‘Evidence: The Practitioner in International Commercial
Arbitration’, International Law FORUM Du Droit International 1, no. 4 (1999):
228–231.
13 Gerald Aksen, ‘Reflections of an International Arbitrator’, Arbitration
International 23, no. 2 (2007): 255–260.
14 Gernot Biehler, Procedures in International Law (Berlin: Springer, 2008), 7;

Adrian Briggs, Conflict of Laws (Oxford: Oxford University Press, 2008), 37;
Lawrence Collins (ed.), Dicey, Morris & Collins, The Conflict of Laws (London:
Sweet & Maxwell, 2006), para. 7-002; George Panagopoulos, ‘Substance and
Procedure in Private International Law’, Journal of Private International Law 1,
no. 1 (2005): 69.
15 Born states that a uniform definition is ‘elusive’. Gary B. Born, International

Commercial Arbitration (The Hague: Kluwer Law International, 2009), 2148.

16 See, e.g., Adrian Briggs, Conflict of Laws (Oxford: Oxford University Press,

2008), 42–43.
17 There has been a certain amount of harmonisation in Europe since the 1980

Convention on the Law applicable to Contractual Obligations (the Rome

Convention), but this does not apply to arbitration. The Rome Convention has
now been replaced by EC Regulation No. 593/2008. See EC Regulation No.
593/2008, Art. 2(e) which states that it shall not apply to arbitration
agreements.
18 An example would be the provisions on waiver.
19 George Panagopoulos, ‘Substance and Procedure in Private International

Law’, Journal of Private International Law 1, no. 1 (2005): 71.

20 Andrew D. Mitchell & David Heaton, ‘The Inherent Jurisdiction of WTO

Tribunals: The Select Application of Public International Law Required by the

Judicial Function’, Michigan Journal of International Law 31 (2010): 568, 574.
21 Judgment of 1 July 1999, Brasoil v. GMRA, Cour d'appel de Paris, Revue de

l'Arbitrage (1999), no. 4: 834.

22 Pierre A. Karrer, ‘Freedom of an Arbitral Tribunal to Conduct Proceedings’,

ICC International Court of Arbitration Bulletin 10, no. 1 (1999): 14–26.

23 Adrian Briggs, Conflict of Laws (Oxford: Oxford University Press, 2008), 43.
24 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2149.

25 Ibid., 2149–2150.
26 Adrian Briggs, Conflict of Laws (Oxford: Oxford University Press, 2008), 38.
27 Burkhard Bastuck & Burkhard Göpfert, ‘Admission and Presentation of
Evidence in Germany’, Loyola of Los Angeles International and Comparative Law
Journal 16, no. 3 (1994): 619; Gernot Biehler, Procedures in International Law
(Berlin: Springer, 2008), 15.
28 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law
International, 2009), 1311.
29 Yves Fortier, ‘International Arbitration and National Courts: Who Has The

Last Word?’, in International Arbitration and National Courts: The Never Ending
Story, ICCA Congress Series No. 10, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 2001), 69.
30 Piero Bernardini, ‘The Role of the International Arbitrator’, Arbitration
International 20, no. 2 (2004): 121.
31 William W. Park, Arbitration of International Business Disputes (Oxford:
Oxford University Press, 2006), 48. The tension between justice and fairness on
the one hand and finality and efficiency on the other has also been noted by
E.D.D. Tavender, ‘Considerations of Fairness in the Context of International
Commercial Arbitrations’, Alberta Law Review 34, no. 3 (1996): 509.
32 L. Yves Fortier, ‘The Minimum Requirement of Due Process in Taking
Measures against Dilatory Tactics: Arbitral Discretion in International
Commercial Arbitration – A Few Plain Rules and a Few Strong Instincts’, in
Improving the Efficiency of Arbitration Agreements and Awards: 40 Years of
Application of the New York Convention, ICCA Congress Series No. 9, ed. Albert
Jan van den Berg (The Hague: Kluwer Law International, 1999), 396.
33 See, e.g., European Convention on Human Rights Art. 6.
34 See, e.g., IBA Rules of Ethics for International Arbitrators, 2; Richard Garnett

et al., A Practical Guide to International Commercial Arbitration (New York:

Oceana Publications, 2000), 83.
35 René David, L'Arbitrage Dans Le Commerce International (Paris: Economica,

1982), 453.
36 William W. Park, ‘Arbitrators and Accuracy’, Journal of International Dispute

Settlement 1, no. 1 (2010): 53. As a result, he has described accuracy, fairness

and efficiency as the ‘three musketeers of arbitral duty’. William W. Park,
‘Arbitration in Autumn’, Journal of International Dispute Settlement 1, no. 2
(2010): 6.
37 See generally, Pierre Mayer, ‘Reflections on the International Arbitrator’s

Duty to Apply the Law: The 2000 Freshfields Lecture’, Arbitration International
17, no. 3 (2001): 235–247.
38 Mirjan R. Damaska, The Face of Justice and State Authority: A Comparative

Approach to the Legal Process (New Haven: Yale University Press, 1986), 122–
123.
39 Carrie Menkel-Meadow, ‘Practising “in the Interests of Justice” in the Twenty-

First Century: Pursuing Truth as Justice’, Fordham Law Review 70 (2002): 1763–
1764.
40 See, e.g., Dennis James Galligan, Due Process and Fair Procedures: A Study of

Administrative Procedures (Oxford: Clarendon Press, 1996).

41 The scenario contemplates less than full indemnity costs to the winner.
42 William Park speaks of le principe du contradictoire or le droit d'être entendu

in Francophone legal systems and rechtsstaatliches verfahren (fair-trial

principles) and Anspruch aus rechtliches Gehör (hearing in accordance with law).
William W. Park, ‘Arbitration in Autumn’, Journal of International Dispute
Settlement 1, no. 2 (2010): 5, n. 21.
43 This is notwithstanding the fact that different systems may describe them in

different ways or even use differing concepts with similar intent. An example of
the latter is the new ICC Arbitration Rules 2012 that have preferred a reference
to fairness over equality, on the basis that the former includes the latter in any
event and is less open to casuistic abuse.
44 See Aristotle, Nichomachean Ethics, trans. Martin Ostwald (Upper Saddle

River: Prentice Hall, 1999), 1131a22–1131a25.

45 Tina Wüstemann & Cesare Jermini, ‘Minimal Procedural Requirements (Part

1)’, in Swiss Rules of International Arbitration Commentary, ed. Tobias

Zuberbühler, Christoph Müller & Phillip Habegger (The Hague: Kluwer Law
International, 2007), 64.
46 Fernando Mantilla-Serrano, ‘Towards a Transnational Procedural Public
Policy’, Arbitration International 20, no. 4 (2004): 342. Adversarial in this
context means the ability of each to present opposing arguments, and not
common law style procedures.
47 This is discussed in section 2.7.7.
48 Ahmed El-Kosheri, ‘Universalism versus Regionalism in Today's Arbitration

Culture’, in Global Reflections on International Law, Commerce and Dispute

Resolution: Liber Amicorum in Honour of Robert Briner, ed. Aksen et al. (Paris:
ICC Publishing, 2005), 249.
49 Ibid., 248.
50 This is discussed in section 13.18.
51 See Toby Landau, ‘The Day before Tomorrow: Future Developments in
International Arbitration’, <www.claytonutz.com/ialecture/2009/>, 21 October
2009. Aksen also argues against undue restraints by institutions. Aksen et al.
(eds), Global Reflections on International Law, Commerce and Dispute Resolution:
Liber Amicorum in Honour of Robert Briner (Paris: ICC Publishing, 2005), 27–28.
This is discussed further in sections 3.14 and 3.16.
52 See sections 3.14 and 3.16.
53 Opportunity cost is a means of measuring value of any selected option, by

looking at the value of the next most desirable option that would otherwise have
been selected. The selected option must have been seen as more valuable by
some on a range of measures to have been selected.
54 They are seen as dead-weight costs in the sense that they have no
commensurate benefit and arise from the process alone.
55 Such evidentiary issues are discussed further in Chapters 6, 10, 11 and 12.
56 Carrie Menkel-Meadow, ‘Practising “in the Interests of Justice” in the Twenty-

First Century: Pursuing Peace as Justice’, Fordham Law Review 70 (2002): 1761,
1763–1764.
57 This is discussed in section 8.11.
58 This is discussed in sections 6.2.6 and 6.2.8.
59 Christian Bühring-Uhle, ‘A Survey on Arbitration and Settlement in
International Business Disputes’, in Towards the Science of International
Arbitration: Collected Empirical Research, ed. C. Drahozal & R. Naimark (The
Hague: Kluwer Law International, 2005), 25.
60 V.V. Veeder, ‘Evidence: The Practitioner in International Commercial
Arbitration’, International Law FORUM Du Droit International 1, no. 4 (1999):
228–231.
61 English Arbitration Act 1996 s. 1(a).
62 Hilmar Raeschke-Kessler, ‘Witness Conferencing’, in The Leading Arbitrators’

Guide to International Arbitration, 2nd edn, ed. L. Newman & R. Hill (Huntington:
Juris Publishing, 2008), 415. See also William W. Park, Arbitration of
International Business Disputes: Studies in Law and Practice (Oxford: Oxford
University Press, 2006), 48.
63 William W. Park, ‘Arbitrators and Accuracy’, Journal of International Dispute

Settlement 1, no. 1 (2010): 54.

64 David W. Rivkin, ‘Towards a New Paradigm in International Arbitration: The

Town Elder Model Revisited’, Arbitration International 24, no. 3 (2008): 377.
65 F.A. Mann, ‘Lex facit arbitrum’, in Liber Amicorum Martin Domke, ed. Pieter

Sanders (Boston: Martinus Nijhoff, 1968), 157.

66 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 75.
67 Award of 14 January 1970, ICC Case 1512, in Yearbook of Commercial
Arbitration V, ed. Pieter Sanders (The Hague: Kluwer Law International, 1980),
176. See also Jan Paulsson, ‘Delocalisation of International Commercial
Arbitration: When and Why It Matters’, International and Comparative Law
Quarterly 32, no. 1 (1983): 53.
68 G. Kaufmann-Kohler, ‘Globalization of Arbitral Procedure’, Vanderbilt Journal

of Transnational Law 36, no. 4 (2003): 1313–1334.

69 Ibid.
70 J.G. Wetter, ‘The Conduct of the Arbitration’, Journal of International
Arbitration 2, no. 2 (1985): 33.
71 Laurence Craig notes that the Model Law, whose drafters had opined on the

unimportance of the place of arbitration, nevertheless exclusively adopted the

criteria of territoriality. W.L. Craig, ‘Some Trends and Developments in the Laws
and Practice of International Commercial Arbitration’, Texas International Law
Journal 30, no. 1 (1995): 36. While that is true, it would also be something of an
oxymoron as the sole criterion to enshrine autonomy in a nationally adopted
version of the Model Law. A legislator still has to decide whose autonomy is to
be respected.
72 [1988] 1 Lloyd’s Rep 116 (CA).
73 (1998) 44 NSWLR 312, 328.
74 Poudret and Besson discuss the difference between geographical and legal

delocalisation. Geographical delocalisation connotes submitting an arbitration

to a lex arbitri other than that of the Seat. This is possible but not desirable and
is discussed further in section 3.8. Legal delocalisaiton connotes the removal of
arbitration from all national legal systems. They argue strongly against the
concept and note a failure of any consistent jurisprudence to the contrary. See
Jean-François Poudret & Sébastien Besson, Comparative Law of International
Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 95–99.
75 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 81.
76 Ibid. This connotes a sociological perspective. For a sociological perspective

see Yves Dezalay & Bryant G. Garth, Dealing in Virtue: International Commercial
Arbitration and the Construction of a Transnational Legal Order (Chicago: The
University of Chicago Press, 1996).
77 Emmanuel Gaillard, Legal Theory of International Arbitration (The Hague:

Martinus Nijhoff Publishers, 2010); Emmanuel Gaillard, ‘The Representations of

International Arbitration’, Journal of International Dispute Settlement 1, no. 2
(2010): 1.
78 Emmanuel Gaillard, ‘The Representations of International Arbitration’,
Journal of International Dispute Settlement 1, no. 2 (2010): 4.
79 Ibid., 7.
80 Gaillard notes that this third view has both a naturalist and positivist strand

in terms of the source of underlying values.

81 C. Bühring-Uhle, G. Scherer & L. Kirchhoff, ‘The Arbitrator as Mediator: Some

Recent Empirical Insights’, Journal of International Arbitration 20, no. 1 (2003):

81.
82 Richard W. Naimark & Stephanie E. Keer, ‘What Do Parties Really Want from

International Commercial Arbitration?’, Dispute Resolution Journal 57, no. 4

(2002–3): 80.
83 Queen Mary School of International Arbitration, ‘International Arbitration:

Corporate Attitudes and Practices’,

<www.arbitrationonline.org/docs/IAstudy_2008.pdf>, 2008.
84 See Michael F. Hoellering, ‘The Role of the International Arbitrator’, Dispute

Resolution Journal 51, no. 2 (1996): 100–107.

85 Queen Mary School of International Arbitration, ‘International Arbitration:

Corporate Attitudes and Practices’,

<www.arbitrationonline.org/docs/IAstudy_2008.pdf>, 2008.
86 See Lucy Reed, ‘More on Corporate Criticism of International Arbitration’,

<http://kluwerarbitrationblog.com/blog/2010/07/15/m...>, 16 July 2010.

87 See Yves Dezalay & Bryant G. Garth, Dealing in Virtue: International
Commercial Arbitration and the Construction of a Transnational Legal Order
(Chicago: The University of Chicago Press, 1996), 31–63. See also Bruno Oppetit,
Théories de l'Arbitrage (Paris: Presses universitaires de France, 1998), 9–11,
127.
88 Pierre Lalive, ‘Sur une “commercialisation” de l’arbitrage international’, in

Liber Amicorum Claude Raymond: Autour de l'Arbitrage, ed. Piero Bernardini et

al. (Paris: Litec, 2004), 167.
89 Lord Mustill, ‘The History of International Commercial Arbitration – A
Sketch’, in The Leading Arbitrators' Guide to International Arbitration, ed.
Lawrence W. Newman & Richard D. Hill, 2nd edn (New York: Juris Publishing,
2008).
90 UNIDROIT, ‘ALI/UNIDROIT Principles of Transnational Civil Procedure’
<www.unidroit.org/english/principles/civilprocedure...>, 2004. See also Axel H.
Baum, ‘International Arbitration: The Path towards Uniform Procedures’, in
Global Reflections on International Law, Commerce and Dispute Resolution: Liber
Amicorum in Honour of Robert Briner, ed. Aksen et al. (Paris: ICC Publishing,
2005). More effort has gone into developing harmonised norms of substantive
law through such mechanisms as the Vienna Convention on Contracts in
International Sale of Goods (1980), the UNIDROIT Principles of International
Commercial Contracts and the Energy Charter Treaty, than is the case with
international dispute settlement procedure.
91 William W. Park, ‘Arbitration's Protean Nature: The Value of Rules and the

Risks of Discretion’, Mealey's International Arbitration Report 19, no. 5 (2004): 1.

92 Ibid., 8.
93 English Arbitration Act s. 69.
94 Swiss Rules 2012 Art. 21.5.
95 In relation to the Swiss Rules, it is unlikely that its unique rule allowing for a
broad range of set-off rights has had any significant impact on its overall arbitral
business.
96 UNCITRAL, ‘UNCITRAL Notes on Organising Arbitral Proceedings’
<www.uncitral.org/uncitral/en/uncitral_texts/arbitr...> (1996).
97 IBA, ‘IBA Rules on the Taking of Evidence in International Commercial

Arbitration’ <www.ibanet.org/Publications/publications_IBA_guide...> (1999).

98 IBA, ‘IBA Guidelines on Conflicts of Interest in International Arbitration’

<www.ibanet.org/Publications/publications_IBA_guide...> (2004).
99 See, e.g., Lawrence W. Newman & Richard D. Hill (eds), The Leading
Arbitrators Guide to International Arbitration 2nd edn (Huntington: Juris
Publishing, 2008); Michael Pryles & Michael J. Moser (eds), The Asian Leading
Arbitrator's Guide to International Arbitration (New York: JurisNet, 2007); the
ICCA Congress Series; and the ICC Dossiers Series. Liber Amicora for leading
practitioners will also typically perform such a function. See, e.g., Aksen et al.
(eds), Global Reflections on International Law, Commerce and Dispute Resolution:
Liber Amicorum in Honour of Robert Briner (Paris: ICC Publishing, 2005).
100 William W. Park, ‘Arbitration's Protean Nature: The Value of Rules and the

Risks of Discretion’, Mealey's International Arbitration Report 19, no. 5 (2004):

1–21.
101 See Roger Alford, ‘The American Influence on International Arbitration’, Ohio

State Journal of Dispute Resolution 19 (2003): 69–88; Bernard Audit,

‘L'Américanisation du droit’, Archives de philosophie du droit 45 (2001): 7–11.
102 While highly contentious, some assert that common law systems best
support economic development and success. The claim was famously made by
Hayek. See Friedrich A. Hayek, Law, Legislation and Liberty: A New Statement of
the Liberal Principles of Justice and Political Economy (Chicago: The University of
Chicago Press, 1973), 94. For a more recent analysis, see Paul G. Mahoney, ‘The
Common Law and Economic Growth: Hayek Might Be Right’, Journal of Legal
Studies 30 (2001): 503–525.
103 Claude Reymond, ‘Civil Law and Common Law Procedures: Which Is the

More Inquisitorial? A Civil Lawyer's Response’, Arbitration International 5, no. 4

(1989): 357–368.
104 Jan Paulsson, ‘Overview of Methods of Presenting Evidence in Different Legal

Systems’, in Planning Efficient Arbitration Proceedings/The Law Applicable in

International Arbitration, ICCA Congress Series No. 7, ed. Albert Jan van den
Berg (The Hague: Kluwer Law International, 1996), 112.
105 Rolf Stürner, ‘The Principles of Transnational Civil Procedure: An
Introduction to the Basic Conceptions’ RabelsZ 69 (2005): 210. See also Tibor
Várady et al., International Commercial Arbitration: A Transnational Perspective,
3rd edn (Eagan: West, 2006).
106 Rolf Stürner, ‘The Principles of Transnational Civil Procedure: An
Introduction to the Basic Conceptions’, RabelsZ 69 (2005): 224.
107 Ibid., 226.
108 Ibid., 223.
109 Nicolò Trocker & Vincenzo Varano (eds), ‘Concluding Remarks’, in The

Reforms of Civil Procedure in Comparative Perspective (Torino: G. Giappichelli

Editore, 2005): 244.
110 Rolf Stürner, ‘The Principles of Transnational Civil Procedure: An
Introduction to the Basic Conceptions’, RabelsZ 69 (2005): 224–225.
111 Ibid., 227.
112 Egypt is an example of the former while Saudi Arabia is an example of the

latter.
113 Ahmed El-Kosheri, ‘Universalism versus Regionalism in Today’s Arbitration

Culture’, in Global Reflections on International Law, Commerce and Dispute

Resolution: Liber Amicorum in Honour of Robert Briner, ed. Aksen et al. (Paris:
ICC Publishing, 2005), 247.
114 Yasuhei Taniguchi, ‘Is There a Growing International Arbitration Culture? –

An Observation from Asia’, in International Dispute Resolution: Towards an

International Arbitration Culture, ICCA Congress Series No. 8, ed. Albert Jan van
den Berg (The Hague: Kluwer Law International, 1998), 38.
115 Grant Kim, ‘East Asian Cultural Influences’, in The Asian Leading Arbitrator's

Guide to International Arbitration, ed. M. Pryles & M. Moser (New York: Juris Net,
2007), 27–28.
116 S.F. Ali, ‘Approaching the Global Arbitration Table: Comparing the
Advantages of Arbitration as seen by Practitioners in East Asia and the West’,
Review of Litigation 28, no. 4 (2009): 827–828.
117 Simon Greenberg, Christopher Kee & J. Romesh Weeramantry, International

Commercial Arbitration: An Asia-Pacific Perspective (New York, Cambridge

University Press, 2011), 47.
118 Michael Pryles & Michael J. Moser, ‘Introduction’, in The Asian Leading

Arbitrator’s Guide to International Arbitration, ed. Michael Pryles & Michael J.

Moser (New York: JurisNet, 2007), 14.
119 Karen Mills, ‘The Importance of Recognising Cultural Differences in
International Dispute Resolution’, in The Asian Leading Arbitrator's Guide to
International Arbitration, ed. Michael Pryles & Michael J. Moser (New York:
JurisNet, 2007), 59.
120 M. Sornarajah, ‘The UNCITRAL Model Law: A Third World Viewpoint’,
Journal of International Arbitration 6, no. 4 (1989): 7. See also Catherine A.
Rodgers, ‘The Arrival of the “Have-Nots” in International Arbitration’, Nevada
Law Journal 8 (2007): 341.
121 William C. James, ‘Trying to Understand the Current Chinese Legal System’,

in Understanding China's Legal System, ed. C. Stephen Hsu (New York: NYU
Press, 2003), 7, 18. See generally Tom Ginsburg, ‘The Culture of Arbitration’,
Vanderbilt Journal of Transnational Law 36, no. 4 (2003): 1335–1345.
122 Karen Mills, ‘The Importance of Recognising Cultural Differences in
International Dispute Resolution’, in The Asian Leading Arbitrator's Guide to
International Arbitration, ed. Michael Pryles & Michael J. Moser (New York:
JurisNet, 2007), 66.
123 Ibid., 72.
124 Ibid., 54.
125 Ibid., 63.
126 Christoph A. Hafner, ‘Professional Reasoning: Legal Cultures and Arbitral

Awards’, World Englishes 30, no. 1 (2011): 117–128.

127 See Ibrahim Fadlallah, ‘Arbitration Facing Conflicts of Culture’, Arbitration

International 25, no. 3 (2009): 303–317. For other examples of prejudice in

culture see Benjamin Davis, ‘The Colour Line in International Commercial
Arbitration: An American Perspective?’, American Revue of International
Arbitration 14 (2004): 461.
Part I: Policy and Principles, Chapter
2: Powers, Rights and Duties of
Arbitrators

Jeff Waincymer,

2.1. Introduction

This chapter considers the rights, duties and powers of arbitrators as they relate
to procedural matters. While arbitrators usually have a range of express
procedural powers together with a broad residual discretion, it will often be the
case that the parties have very different views as to how these should be
exercised. For example, how does a tribunal deal with an application for
production of documents when one party comes from a legal culture that
believes this is a fundamental right, but the other comes from a culture that
believes it is an invasion of personal liberty to be asked to produce documents
that aid the opponent in the dispute? In other circumstances there may be no
consensus as to whether a particular power or duty even exists. An example is
whether there is a power to exclude counsel if it is felt that counsel for one of the
parties is deliberately seeking to interfere with the smooth running of the
process. Another example of a contentious power is whether a tribunal may
decide on legal principles not submitted by either party. In other cases parties
may believe that an acknowledged power should nevertheless be constrained by
a separate duty. For example, how does a tribunal ensure that each party has an
opportunity to adequately present its case but still deal with the matter
expeditiously? In the document production example above, how does an
arbitrator make a choice if both parties argue that a choice adverse to their
interests is contrary to fair and reasonable due process norms?

A discussion of rights, duties and powers in the context of such examples

involves consideration of three aspects of arbitral authority: what an arbitrator
can do in a particular circumstance, meaning whether they have requisite
authority to do that; what they ought to do, meaning what best practice would
suggest in the circumstances; and what they must do to comply with any legal or
moral obligations on them. (1)

This chapter looks at these issues at the general level although it will use a range
of examples in support of the analysis. More specific discussion of the rights,
duties and powers over particular aspects of the arbitral procedure are
elaborated upon in specific chapters. As Lalive has noted, ‘international
arbitration is a domain where generalisations are particularly apt to be
misleading.’ (2) It is important to understand, however, that what an arbitrator
can or ought to do in relation to many specific procedural questions, such as
control over evidence or control over counsel's ethical behaviour, depends on
the view which is taken as to the general nature and sources of rights and
obligations discussed in this chapter.

The following discussion first looks at sources of powers, rights and duties. It
then gives attention to their suggested content. The discussion of content is
separated. The chapter first discusses the content of duties and then gives
separate attention to the content of powers. Separate attention is then given to
rights. This is simply done for ease of discussion. For reasons alluded to above,
there is inevitable cross-referencing between duties, powers and rights.
Furthermore, in many contentious areas it is difficult to identify whether they
best fit in one category or the other. An example would be the question whether
a tribunal can decide on issues of law not raised by either party. Some might
argue this to be a duty, some a mere power, while some argue that it should not
occur at all. At the very least, an attempt at categorisation may help with the
essential analysis.

Some treatises on international arbitration deal separately with questions of

jurisdiction and power of arbitrators on the one hand and an arbitrator's rights
and duties on the other. (3) Others integrate the analysis. (4) Some commentators
who discreetly discuss rights and duties do so in the context of identifying the
obligations flowing between the arbitrator and the parties to determine what
claims can be made between them outside of the arbitral process itself.
Traditionally, a discussion of an arbitrator's rights was limited to entitlement to
fees or entitlement to immunity against liability. A broader notion of rights
should also be understood to relate page "48" to the procedural rights necessary
to promote fairness and efficiency of the arbitral process.

This chapter ultimately deals with powers, rights and duties concurrently
because it is not easy to separate the concepts in determining how to deal with
contentious procedural and evidentiary questions. For example, if one can
clearly articulate a particular arbitral duty, that will often impose a constraint
on an otherwise broad power. Conversely, if one can clearly articulate a
particular right held by an arbitrator, that might imply certain consequential
powers, or at least confirm that generally expressed powers should be
interpreted broadly in order to support that right. (5) A final decision on a
procedural matter might thus involve the tribunal in considering how to
reconcile various rights, duties and powers which appear to otherwise be in
conflict. To this end, the aim of this chapter is to analyse the nature of the
various rights, duties and powers in some detail and consider how they
interrelate. An understanding of how this ought to occur may then assist in
providing a logical basis for the resolution of the various practical questions
raised in the balance of the book.

2.2. Sources of an Arbitrator's Powers, Rights and Duties

While the range of sources of the rights, duties and powers of arbitrators is easy
to identify, more contentious is the question of their relative importance. This is
to be contrasted with the situation of a domestic judge. A judge's powers, rights
and duties are generally more carefully articulated. In a modern parliamentary
democracy at least, these will first emanate from a constitution establishing
separation of powers and a judicial branch of governance. There will then
typically be a court-related statute outlining rights, duties and hierarchies.
Subsidiary court rules may then establish a procedural framework. Where
international matters are concerned, a domestic judge is also subject to that
country's private international law principles, which will cover issues of choice
of forum and choice of substantive law. The same simple framework is not
prevalent in international arbitration. As with any area of arbitral law and
practice, not only are there a range of potentially conflicting sources but there is
also the overriding tension between a consent-oriented paradigm and a
jurisdictional one. Thus, in identifying the exact ambit of the rights, duties and
powers of arbitrators, one could concentrate on the rights and obligations
emanating from the consent of the parties or concentrate upon the rules
emanating from the governing procedural law. The two are of course page "49"
connected given the usual role of party autonomy in selecting a Seat and hence
the lex arbitri, but there interaction is contentious.

As noted in Chapter 1, most would now see arbitration as a hybrid of consent

and jurisdictional paradigms. The hybrid theory recognises that an arbitrator's
jurisdiction is in part a function of the agreement of the parties but also a
function of the willingness of national legal systems to recognise the validity of
such agreements and to provide practical utility to the process through
guaranteeing enforceability of awards. Because it is still difficult to understand
the exact interplay between the elements of a hybrid theory, some concentrate
simply on a functional approach, without seeking to resolve the consent versus
jurisdiction debate. From a practical perspective, however, recognising
arbitration as a hybrid or arguing for a functional approach does not provide a
methodology to determine how to resolve any tension between rights, duties
and powers where they emanate from differing sources and where there may be
a need to rank the sources themselves or at least consider how they interrelate
on a specific procedural or evidentiary issue. There is also a need to consider the
extent to which permitted variations in the nature of rights emanating from one
source may impact upon the rights and obligations under others. This is
particularly so when one of the sources involves contractual rights. Here the
question is whether arbitrators and/or the parties can use contractual
agreements to increase their rights and/or reduce their obligations. Once again,
the view one takes on questions of this nature can reflect one's view on the
consent versus jurisdiction debate.

Such a debate can appear in other guises. For example, some commentators
speak of the status of arbitrators even though this concept is not expressly
addressed in arbitral rules. (6) A consideration of the status of an arbitrator may
provide strong indirect guidance as to the rights and obligations. Born even
considers the question of status to be central in defining the rights, powers and
obligations of arbitrators. (7) A discussion of status typically seeks to draw
analogies between the role of an arbitrator and that of a domestic judge. The
more the status is seen to be similar, the more the rights and obligations of
judicial officers might be seen to inherently or impliedly flow to international
arbitrators. The more one aligns with the jurisdictional end of the arbitral
spectrum, the more one might have sympathy with these views. At times, such
views may be affected by comparativist perspectives. Where some contentious
procedural issues are concerned, differences in view can be tied to
fundamentally different positions historically taken by legal families. For this
reason, an arbitrator concerned to appear neutral and impartial has some
difficulty in selecting between legitimate but differing page "50" responses to
questions such as production of documents, duty to know the law regardless of
submissions of the parties, or treatment of set-off rights.

As to the sources themselves, at the very least there is a need to consider:

(i) consent of the parties, in particular the arbitration agreement;

(ii) the contractual relationship between the parties and the arbitrator;
(iii) the relationship between the parties, arbitrators and an institution if
utilised;
(iv) the lex arbitri;
(v) any agreements or modifications made by the parties as to the procedural
norms applicable, including selection of a set of arbitral rules;
(vi) any mandatory laws of procedure applicable to the arbitration (whether
emanating from the lex arbitri or otherwise) or the possibility that there
are some general principles of arbitral procedure that apply in any event,
i.e., a lex mercatoria of procedure;
(vii) potentially applicable international treaties, conventions and
international law norms;
(viii) customs and usages;
(ix) powers, duties and rights that might be argued to be inherent, flowing
from the essential nature of the adjudicatory process itself and/or the
status of an arbitrator; and
(x) the impact if any on arbitrators' powers, rights and duties by the law of
the place or places where the arbitration agreement is sought to be
recognised or the award enforced.

There is also a need to consider powers, rights and obligations that might be
implied into express provisions under the above categories.

The following sections elaborate on these sources. It is important to understand

a number of things about these sources before they are individually discussed
further. First, the above list is not presented in order of importance. The list
begins with three aspects of consent, being that of the parties, any agreement
between the parties and an arbitrator and additional agreements where an
institution is concerned. The list then turns to the key jurisdictional bases, in
particular the lex arbitri, although there is also a reference to a choice by parties
of arbitral rules. Some of the sources listed are contentious but at least need to
be considered in some circumstances at least. Contentious categories would
include powers that might be implied as flowing from express provisions;
powers which are argued to be inherent in the nature of the adjudicatory
process where, as noted, some would try to analogise the position of an
arbitrator with that of a judge; the potential application of mandatory
procedural norms of a jurisdiction other than the Seat; and the impact if any of
the laws or practices of potential enforcement jurisdictions. Because there can
be legitimate debate about the nature and ambit of any of these sources, it is
simply not possible to define with precision the exact ambit of rights, duties and
powers of international arbitrators. Nevertheless, a proper consideration of the
sources and the way they are to be integrated will aid in understanding the
proper ambit of any procedural discretions given to the arbitrator. Any
arbitrator page "51" must at least be able to present a reasoned justification for
any contentious procedural determination that is made. This would include a
reasoned justification for ranking different sources.

2.2.1. Consent and the Arbitration Agreement

If consent is one of the keys to arbitrator powers, and almost certainly the most
significant, the arbitration agreement itself is the primary evidence of that
consent, subject to any subsequent agreement between the parties. Approaches
to drafting and interpreting arbitration agreements are discussed in sections 3.2
and 3.3. This section looks at limits on party autonomy.

2.2.2. The Limits of Consent Emanating from Party Autonomy: Consent and
the Lex Arbitri and Arbitral Contract

While party autonomy is central, the overriding question is whether the

jurisdictional source of an arbitrator's power, commonly the lex arbitri,
recognises a particular form of party consent. Party autonomy is ranked above
broad arbitrator discretions in virtually all leading lex arbitri. An example is
Article 19(1) of the Model Law which states ‘(s)ubject to the provisions of this
Law, the parties are free to agree on the procedure to be followed by the arbitral
tribunal in conducting the proceedings’. The tribunal's broad discretion only
applies failing such agreement. Party autonomy is also enshrined indirectly
through Article V(1)(d) of the New York Convention which sets up conformity
with agreed procedures as one of the criteria by which to test enforceability of
an award.

Nevertheless, the parties cannot consent to the exclusion of mandatory norms of

procedure enshrined in the relevant lex arbitri. (8) An example would be a norm
such as Article 18 of the Model Law which requires each party to be treated
equally and to be given an opportunity to present its case. An example of a lex
arbitri which might constrain party autonomy over and above the more general
mandatory norms such as equality of treatment is the English Arbitration Act
1996 which also calls for an arbitrator to be concerned with efficiency and
timeliness. (9) Mandatory procedural norms are discussed further in section 3.7.

It is also important to understand that virtually all lex arbitri give the parties the
power to agree on new or modified procedures from time to time, save for their
obligation to comply with mandatory procedural norms. In some cases, such
page "52" agreements could constitute changes to an earlier arbitration
agreement. In other cases they are merely supplementary to it. Being
agreements supported by the lex arbitri, they do not need to meet form
requirements of the arbitration agreement itself as per Article II of the New
York Convention. Nor would they need to satisfy contractual principles such as a
requirement of consideration under the common law where it applies.

It is also necessary to consider how party autonomy impacts upon the

arbitrator's rights under any contract with the parties and/or institution.
Because individuals cannot bind third parties without some legal basis for doing
so, the role of party autonomy in constraining arbitrators' powers must consider
both the lex arbitri and the arbitrator's own agreement to perform. The nature
of that agreement is discussed in section 2.3 below. An example of such a
potential limitation on party autonomy would arise where parties seek to
fundamentally change the procedural parameters after the tribunal has already
accepted appointment. For example, if an arbitrator has agreed to undertake a
matter for a fixed fee on the basis that it will be a documents-only arbitration, if
the parties subsequently want a full oral hearing they cannot compel the
arbitrator to do so against the arbitrator's wishes as this would conflict with the
contract between them and the arbitrator. The parties still retain the right to
agree on modified procedures but would then need to renegotiate with the
arbitrator if necessary. In the extreme, an arbitrator might be entitled to resign
from the position if no modified agreement is achievable between the arbitrator
and the parties. If the parties change their agreement after appointment, an
arbitrator cannot normally reject the proposed change and conduct the
arbitration as per the terms of their original mandate unless the parties'
variation can be ignored, for example, on the basis that it offends against a
mandatory procedural norm that must apply in any event or because it is illegal
or unethical. In these circumstances, the tribunal must still consider why it may
continue with the original mandate and whether it should instead resign. Once
again, the conclusion one comes to on this issue can in part be dependent on
one's view on the consent/jurisdiction debate. The more one adopts a
jurisdictional perspective, the more one sees a general adjudicatory status that
allows an arbitrator to continue after ignoring an improper proposal.
Conversely, the more one adopts a consent paradigm, the more one might see an
arbitrator being fundamentally compromised in the face of even unreasonable
directions from the parties. Even a consent orientation can still lead to the same
outcome if the view is taken that party autonomy is not unlimited, but that it is
always subject to implied limitations as to legality or ethics or implied terms to
respect tribunal determinations in such circumstances. This is discussed further
in the next section and also in section 2.7.12.4 and dealing with the duty not to
exceed the mandate and section 2.10.3 dealing with the power to override a
procedural determination of the parties. It is also discussed in section 6.2.4
dealing with proactivity.

Limitations on consent can also arise even where the tribunal and parties are in
agreement. The parties and the arbitrator have no right to externalise such
powers over non-consenting third parties. Thus the arbitration agreement
cannot by itself grant an arbitrator powers to make coercive determinations in
relation to third- page "53" party property or witnesses. Such powers can only
emanate from some national legal system. Similarly, two parties cannot grant an
arbitrator powers to make coercive determinations as to joinder of third parties.
Some jurisdictional support will be needed for this to occur.

Finally, it needs to be borne in mind that many key issues are not usually
resolved by party agreement and are instead left to broad arbitrator discretion.
Often party autonomy will simply not address some of the more crucial issues
facing an arbitrator such as choices between strict and purposive interpretation,
whether proactive and practical solutions should be fashioned and whether an
attempt should be made to temper strict legal principles by a consideration of
the reasonable anticipation of the parties. (10)

2.3. The Contractual Relationship between the Arbitrator and the Parties

Arbitral consent will typically emanate from, or at least result in, two separate
agreements. The first being the arbitration agreement, is accepted as a
necessary precondition to arbitration. The second, being any contractual
relationship between the arbitrator and the parties, is more controversial as
there are differences in view as to whether such a contract exists as a matter of
course and if so, what its content is. Where a contractual relationship with the
arbitrator applies, this may also seek to extend rights and duties as would
otherwise arise from lex arbitri, arbitral rules and even the arbitration
agreement or may seek to limit them. It is thus necessary to consider the
contractual relationship in more detail. Because it may impact upon procedural
rights, duties and powers, its ambit is in that sense appropriate for this book.

2.3.1. Is the Relationship Truly Contractual?

The common law had at first been reluctant to accept the contractual basis of
the relationship, that has now been recognised by the courts, although with
somewhat differing articulations, (11) while civilian systems were always
prepared to page "54" acknowledge a contractual relationship. A contractual
basis is also accepted in a number of Islamic countries. (12)

One reason for the original difference in view emanates from differences in view
as to the essential status of an arbitrator. In turn, this reintroduces differences in
view as to the consent versus jurisdiction paradigms in arbitration per se.
Despite the reasons for the historical differences, the contract/status debate as
to whether an arbitrator is in a contractual relationship with the parties can be
too easily overstated. The leading proponents of the status view have been said
to be Mustill and Boyd. (13) The authors question whether it makes much sense to
try and identify an implied contract with terms and conditions in all
circumstances, or whether this was instead a strained exercise. The scholarly
debate seems to be more about whether there is always some implied contract,
a view that Mustill and Boyd question, rather than an assertion by contract
proponents that it is solely dependent on contractual agreements between the
arbitrator and the parties. Even those legal systems that readily recognise a
contract between the arbitrators and the parties also recognise the
jurisdictional link and the way this may impact on contract rights and
obligations. The hybrid nature of the sources of rights means that the contract
with the arbitrator simply needs to be looked at in the context of the regulatory
framework that otherwise applies. (14) The contractual versus status relationship
between arbitrator and parties may also be impacted upon by the degree of
specificity of the lex arbitri. For example, Mustill and Boyd have observed that
the more detailed provisions of the English Arbitration Act 1996 have moved
arbitration in that jurisdiction towards a statutory conception as opposed to a
paradigm of implied powers. (15)

The ICC Final Report on the Status of the Arbitrator considered that in all cases
the arbitrator and the parties are bound by a specific contract. (16) There are
good reasons for this emergent position. Prospective arbitrators are not bound
to take a case against their wishes. Hence the very existence of an arbitration
depends on some arrangement between the arbitrator and the parties. It is hard
to see how some form of contract cannot eventuate to the extent that an
arbitrator can demand a particular level of fees.

page "55"

Karrer also speaks of the arbitrators forming a partnership in the sense of

putting common effort into achieving a common goal. (17) It is certainly true to
describe them as having a common goal but under most applicable partnership
law systems they would not be seen as forming such a relationship, not having a
common view to profits and not intending unlimited liability as to the behaviour
of each.

2.3.2. When and how is the Contract Formed?

In the context of its impact on procedure, the timing of the arbitrator's contract
may be important as a party may not unilaterally vary contract terms. The
contract should not be seen as consummated until the particular arbitrator
accepts the mandate. That might be by way of an express acceptance or by
conduct demonstrating a willingness to perform. Considering the contract to be
completed on acceptance by the tribunal would be consistent with a range of
statutes indicating that arbitrators are not bound to accept an appointment or
that the tribunal is only constituted upon acceptance. (18)

In looking for the time when an offer has been accepted, in many cases this may
be complicated as there could be a range of stages and various forms of
communication which both directly address potential terms and also
incorporate terms by reference. This can again be important when a prospective
arbitrator imposes conditions or suggests qualifications or recommendations.
The complex chain of communications leading to an ultimate contract can be
even more complex where there is a multi-member tribunal, given that in most
cases, the party-appointed arbitrators then select the Chair. Where there is a
multi-member tribunal, even the appointment by a party of its own selected
arbitrator leads to contractual rights with all parties as each is seen as agent for
the other in this process. The same would be so where the two party-appointed
arbitrators select a Chair or President. (19) This is thought to arise because each
party-appointed arbitrator is in a contractual relationship with all parties and
then acts on their behalf in nominating the Chair. These contracts are also seen
to arise where an institution is involved in appointment, although Onyema
queries whether this is so. (20) This is discussed further in section 2.4 below.

While it is reasonable to accept that each party impliedly accepts that the other
may act as its agent in entering into a contract with a nominee, less clear is the
page "56" agreement as to fees if a discussion is not held with both parties. One
could conclude that there was implied agreement to submit to reasonable terms
and conditions. Under common law, the non-nominating party might be
estopped from objecting to reasonable terms. (21) Civilian systems might rely on
broader good faith notions. This would still not obviate an argument as to
whether the terms as proffered were reasonable or not. There is also a question
as to whether there are two separate contracts between the arbitrators and each
party or whether ultimately one contract is formed. If there is more than one
contract, the question of governing law might pose problems where they differ.
Fee entitlement will only give rise to procedural issues where a tribunal would
halt proceedings on the basis of inadequate payment. If the other party cannot
be bound to all terms and conditions proposed by a nominee arbitrator, another
question is whether they can then frustrate the proceedings by refusing to
accept such terms or through seeking to negotiate modifications. (22)

While a contract might only be formed upon acceptance by a nominee

arbitrator, there are nonetheless pre-contractual obligations imposed on
prospective arbitrators through national laws and institutional rules. Key
obligations are disclosure and the requirements for accepting appointment.
Duties of inquiry and disclosure and ethical considerations on acceptance of
appointment are discussed in sections 5.10 and 5.9 respectively. From the
perspective of potential contracts with the parties, depending on the applicable
law, there may be broader or lesser liability as to an arbitrator's pre-contractual
behaviour. For example, in systems applying culpa in contrahendo, an arbitrator
who expresses availability and then pulls out unjustifiably, could find that there
are claims for any disruption thereby caused, subject to overriding immunity
provisions, if any. (23)

2.3.3. What are the Elements of the Contract?

If there is a contractual relationship, key questions include what terms and

conditions might be required by an arbitrator, whether an arbitrator can insist
upon certain conditions and whether there must be an agreement of both
parties for these terms to be binding. (24) While these issues go well beyond
procedure, they impact upon it as the contract elements would be one factor in
determining when and why a tribunal may act contrary to the wishes of the
parties.

One obvious approach to determining the rights and obligations of an arbitrator

would be to look to the express and implied terms in such a contract. Another
approach is to first or concurrently consider the essential nature of such a
contract page "57" and what this implies as to interpretation of express terms
and content of implied terms. From the latter perspective, the contract between
the arbitrator and the parties is unusual for two reasons. The first is that it
combines the normal elements of contractual agreement with the arguable
status of a judicial style appointee. The second and related issue is that in one
sense, an arbitrator is subject to the wishes of the parties, given that jurisdiction
emanates from original consent to arbitrate and is further subject to the parties'
rights to agree on procedural matters. However, the relationship is by no means
entirely subordinate to the parties because of the quasi-judicial nature of the
function. That means that the arbitrator must always give effect to certain
fundamental principles of due process and justice. In turn, this implies that
there will always be certain discretions and prerogatives that may be implied.
(25)

Where civilian jurisdictions are concerned, there may be a further need to

characterise the contract in terms of traditional sub-categories of contract law, if
possible, in order to determine what legal principles apply to it. This is because
civilian systems are more inclined to look for a unifying scientific theory
underlying a legal relationship. For example, the respective rights and
obligations may vary depending on whether civilian law characterises the
contract as a contract for services or akin to a modified agency agreement, or as
a contract with a unique or mixed status or finally, as a contract of
empowerment. (26) The hybrid nature of the contract has made it difficult to
classify. Fouchard, Gaillard, Goldman describe the relationship as ‘contractual in
source, judicial in object’. (27)

Poudret and Besson suggest that there may be less differences in reality
notwithstanding the different typologies. (28) It would be undesirable to allow
such questions of characterisation to dominate. In particular, they ought only to
help in providing certain starting presumptions in defining rights and
obligations. Furthermore, whatever presumptions arise from civilian contract
law vis-à-vis contracts of agency and professional services, these cannot indicate
how statutory arbitral norms should limit rights and promote obligations
irrespective of the wishes of the parties themselves. Furthermore, as Born
points out, the various options typically applied in civilian legal systems simply
do not fit the arbitrator/party relationship. An agency arrangement is one
where an agent must respond to the directions of the principal. A professional
service contract is concerned with the relationship vis-à-vis a client and does
not deal with what is essentially an page "58" adjudicative function. (29) Born
thus argues persuasively for a distinct treatment on the basis of a sui generis
adjudicatory function of arbitrators. (30)

Similar issues arise from a common law perspective. When arguing against a
purely contract-based approach, Mustill and Boyd were also making the point
that the rights and obligations of an arbitrator could not be determined solely by
trying to take ordinary principles of contract law and seek to imply appropriate
terms. The authors invited the courts to consider ‘what rights and duties ought,
in the public interest, to be regarded as attaching to the status of arbitrators'.
While most disagree with Mustill and Boyd's suggestion that rights and
obligations are not determined by contract, their caution is a sound one that the
proper balance will not necessarily arise through a blinkered consideration of
applicable contract law norms alone but must instead consider the nature and
status of arbitrators per se. That ought to be required under a thoughtful
contract law analysis in any event as a contract for any activity will invariably
have implied terms supporting that activity itself. Thus a contract for arbitration
services must impliedly contain the entitlement to do those things that are
inherent in the arbitral function, so the debate is returned to that central and
controversial question. Utilising a contract approach, a court would inevitably
need to look for implied terms based on the necessary expectations of the
parties in an arbitrator/party relationship, terms implied by law and the
implications of public policy, such as in relation to immunity and validity of
exemption clauses. (31) Thus for common law commentators to say that a
contract theory in and of itself is unhelpful is simply acknowledging that the
outcome will differ depending on the nature of the contract involved. This is the
same as civilian systems that are concerned to characterise different types of
contractual relationships.

Even if contract terms are not determinative, they must be relevant.

Nevertheless, a simple consideration of express terms would be problematic for
a range of reasons. First, in many cases there is no written contractual
agreement between arbitrators and parties, although an ad hoc arbitration will
typically have one. Even where there is no written agreement whatever, some
oral discussion must have occurred for an arbitrator to accept appointment.
Where arbitrators require detailed written agreements, these will easily provide
evidence of the mutual rights and obligations, save for the impact of legislative
and other rule-based norms that limit the freedom of contract. More typically,
however, where an arbitrator is approached with an offer of a position and
accepts it in simple form, the terms page "59" of the contract may largely flow as
a result of implied terms or terms incorporated by reference.

In some cases, the final contract between an arbitrator and parties may also
arise from a number of stages where earlier agreements were modified. For
example, an arbitrator might agree on a particular fee based on a party's
estimation of the likely time it will take. If the procedure then proves to be more
complex, there might be a revision of the arbitrator's entitlement. That may be
impacted upon by ethical principles as to whether an arbitrator may seek to
revise a fee structure after appointment. (32) An arbitrator might also indicate to
the parties that after consideration of early submissions, he or she is unwilling
to conduct the proceedings in the manner originally envisaged by the parties. In
some circumstances this may constitute an offer to modify the arrangement. If
the parties accept, the situation is easy. If the parties refuse, the arbitrator's
rights may depend on the view one takes as to the contract that previously
existed. If the parties are themselves seeking to fundamentally change the
parameters of the agreement with the arbitrator, the latter is entitled to reject
that. If they are determined that the procedure be modified in any event, this
may give rise to an entitlement to resign without liability under the contract. If
instead, the arbitrator is seen as attempting to change a previous agreement
freely made, then it would be well within the parties' entitlement to refuse and
demand that the arbitrator completes the mandate.

If one looks to the wording of any express agreement between the arbitrator
and the parties to determine rights and obligations, one should also be careful as
to the method of interpretation and the attitude to be taken to express
provisions that seek to prefer the rights of one party over the other. As with any
negotiated contract, one party might try and increase its own rights at the same
time as increasing obligations on the other. One would not wish to start an
analysis of an arbitrator's contract from the perspective of caveat emptor,
allowing an experienced arbitrator to write voluminous exclusion clauses
limiting many onerous procedural obligations that might otherwise apply.
However, when the parties are sophisticated commercial actors, there may be
no problem in accepting the implications of such rights freely ceded to the
arbitrator under an open contract negotiation. A jurisdictional perspective may
be less disposed to this view, although it could still accept that exclusion clauses
simply seek to promote the very immunity that quasi-judicial officers are
entitled to.

Even if one started from a presumption of freedom of contract, there may be

limitations on what may be agreed upon from the lex arbitri. For example,
certain mandatory norms, such as giving each party an opportunity to present
its case and to be treated equally, plus the proscription of bias, imply certain
obligations regardless of what might be written into a contract to perform
arbitral duties; otherwise, it could cease to be properly seen as a contract for
arbitration. (33) All of the key elements in the arbitration agreement, including
selection of the Seat and page "60" the rules of procedure would also have to be
incorporated by reference into the contract with an arbitrator. That is not only
the natural implication but in addition, the arbitrator could not act under a
separate contract in ways inconsistent with the arbitration agreement. (34)

To the extent that rights and obligations need to be determined from the
contract between the parties and the arbitrator, this will be an international
contract and raises conflict of laws issues. Determining the applicable law of
such a contract is outside of the choice of law rules of the lex arbitri applicable to
the arbitration itself. Nevertheless, under most domestic conflict systems,
absent an express choice by the parties, one would either look at the law of the
domicile of the arbitrator as the provider of the key services or the place with
the closest connection to the contract. Where the parties have made an express
choice, that would normally be accepted by any adjudicatory body. (35) Even if
the arbitration agreement had a choice of law clause this would not necessarily
apply to a contract with an arbitrator. Lew, Mistelis and Kroll point to the
obvious difficulty of applying the domicile of the arbitrator in multi-member
tribunals. (36) Where the closest connection test is concerned, that is likely to be
the Seat of arbitration. The Seat of the arbitration is not only where the work is
typically done, that is, the place of characteristic performance, but it is also
where the lex arbitri governs the behaviour of the arbitrator, even if hearings
and other functions occur elsewhere. (37) One problem might be where the Seat is
not known when the contract with the arbitrator is concluded. (38) Other
possibilities might be the Seat of the arbitral institution where one is involved.
Fouchard, Gaillard, Goldman suggest that an international arbitration's contract
should be governed by international law and not domestic laws as determined
under conflicts methodology. (39) This is not the dominant view, however.

page "61"

If the contract is subject to common law, implied terms would need to be

particularly considered as this is a key means by which the common law of
contract develops standards in certain types of contractual relationships. It is
necessary to differentiate between terms implied as a matter of fact and those
implied by law. As to terms implied by fact, this is based on a necessity test to
give business efficacy to the transaction. It flows from implied intent that the
parties would have expected all terms necessary to give the arrangement
adequate business efficacy. (40) Terms implied by law would include the
necessary elements from lex arbitri, such as the duty to remain impartial. (41) In
all systems there would need to be a consideration of the impact of protective
legislation seeking to exclude the operability of unreasonable contract
provisions. There are also questions as to the degree to which exemption
clauses should be restrictively interpreted. Identifying the nature of the contract
would also have implications for resignation rights or rights to withdraw
nominations. Because the arbitrator is linked to all of the parties, not just the
one that directed appointment, termination permission or acquiescence in
resignation must also come from all of the parties.

Poudret and Besson make the interesting observation that the contract with the
arbitrator rarely contains an arbitration clause, hence disputes must be brought
before the courts. (42) Their view seems correct as a contract between the parties
and an arbitrator is not an arbitration proceeding as such.

2.3.4. Form and Content Requirements and Recommendations for Arbitrator

Contracts

Few legal systems provide form requirements for any arbitral contract. One
exception is the Dutch Code of Civil Procedure, requiring a written acceptance of
an arbitral mandate. (43) Nevertheless, it would be rare for acceptance to occur
other than in writing in any appointment process.

Because of the uncertainties, prudent arbitrators should only accept

appointments after terms have been agreed with all parties. In drawing terms of
appointment, an arbitrator might consider covering the following matters: (44)

2.3.5. Relationship to the Arbitration Agreement

Some common law courts have considered that the arbitrator becomes a party
to the original arbitration agreement, (45) ‘which becomes a trilateral page "63"
contract.’ (46) Some Islamic countries also take the view that the arbitrator
becomes party to the arbitration agreement. (47) French courts see the
contractual arrangement being distinct from the original arbitration agreement.
(48)
The better view is that the contracts are separate but terms of the arbitration
agreement are incorporated in the separate contract between the arbitrator and
the parties or at least define its scope, given that the separate contract is an
agreement to perform services under the initial arbitration agreement. (49) For
example, if the arbitration agreement designates that an award must be
rendered within a particular period of time, if the arbitrator accepts
appointment, then the timeframe is impliedly agreed to. The same is so for
selection of the Seat, lex arbitri and arbitral rules. Thus while independent, the
arbitration agreement is controlling unless modified by consent, although many
examples of apparent conflict would be explained on the basis that the later
agreement is a modification of the former. In other cases the arbitration
agreement can constrain the contract between the parties and arbitrator. That
still would not undermine the contract between the parties and the arbitrator
but would indicate which contracts give proper effect to the arbitration
agreement and which do not. Because of this dominant view, the validity of one
does not depend on the validity of the other.

One important difference is the ongoing implication for all of the parties if the
arbitrator was truly seen as an equal party to the original arbitration agreement.
One problem in such a circumstance would be that the arbitrator has an equal
ability to lock in the status quo by simply refusing to agree to any amendment to
it. That would go against the essential autonomy of the parties and would even
mean that once an arbitration had been commenced, no changes can be made
without the arbitrator's permission unless such power was said to impliedly
flow from the lex arbitri selected. Such a scenario would even make it difficult to
determine how the parties could agree on a settlement of their dispute without
an arbitrator being able to claim an implied cancellation fee as damages.

page "64"

Another problem in seeing the arbitrator being bound in a tripartite relationship

under the original arbitration agreement would be where the latter is invalid.
An arbitrator would expect to have an independent valid agreement entitling
him or her to fees when making such a determination even under express
competence-competence powers. (50) Problems would also arise in seeing an
arbitrator as party to the arbitration agreement given that in most cases, its
form is that of a general agreement covering all future disputes between the
parties. A further problem in seeing it as a tripartite contract is the fact that
many laws and rules grant the parties the power to dismiss the arbitrator. (51) If
the arbitrator was instead seen as being an essential party to the original
agreement, it would be hard to see how termination was permissible without
the arbitrator's consent, although one could certainly imply such a right via
incorporation by reference of the lex arbitri or institutional rules.

2.3.6. The Role of the Arbitrator's Contract after the Award is Rendered

While some institutional rules expressly indicate that the mandate of the
arbitrator is terminated upon the making of a final award, care should be taken
to ensure that this is not presumed to automatically terminate inter-party rights
under a contract between an arbitrator and the parties. It is certainly true that
the service element of such a contract is at an end. Virtually all rules also
indicate that an arbitrator's power extends to allow for interpretation and
correction of an award, and for additional awards often within a limited
timeframe. (52)

2.4. The Relationship between the Parties, Arbitrators and an Institution if

Utilised

A contract still arises between the arbitrator and the parties even where an
institution makes the appointment. In such circumstances the parties have
conferred power on the institution to bind them in this way, hence making it an
agent for such page "65" purposes. (53) The previous section considered whether
an entitlement to fees helped show that there must be some form of contractual
arrangement between the parties and an arbitrator. This will be so even where
an institution is involved that regulates the fees, as it is the appointment of the
arbitrator by the parties that starts the legal relationship empowering the
institution to do so.

Some commentaries suggest different conclusions. Berger considers that by

acceptance of the mandate, an arbitrator enters into a contractual relationship
with the parties or the arbitral institution. (54) Hausmaninger argues that in
institutional arbitration the arbitrator and the parties are not in a direct legal
relationship. Instead he considers the relationship to be through two contracts,
one between the parties and the institution, the other being a contract between
the institution and the arbitrators. (55) Such a view seems undesirable. If there
are to be two contracts, there may be problems of jurisdiction and
interpretation between them. They might also have different applicable laws.
For example, a contract between the parties and the ICC in Paris that is seen as
separate to the contract with the arbitrators, might see a judge conclude that
Paris is the place most closely connected with the institutional contract while
the Seat as selected might be seen as most closely connected to the arbitrators'
contract. The better view is to see there being a direct contract with the parties
in institutional arbitrations as well as in ad hoc arbitrations.

Acceptance of an appointment under a particular institution's rules without

further specification is still sufficiently definite to constitute a binding service
agreement as broad acceptance would impliedly accept the fee structure and
methodology as elaborated upon in the institutional rules. There may well be a
separate contract between the arbitrator and the institution, but the terms of
this would be imported into the contract with the parties in any event. An
institution that simply acts as an appointing authority on an ad hoc basis would
not be seen to be in a contract with an arbitrator.

There may also be a range of materials to consider. Contracts through

institutions can arise from an amalgam of correspondence and oral
communications. In many instances, parties may begin by recommending
certain names to each other. These may be discussed amongst themselves
and/or communicated to an institution. A written or oral invitation will be made
to an arbitrator. An arbitrator may be asked to sign an institutional acceptance
form. The institution may then communicate this to the parties. The institution
might provide the parties with page "66" details of fee schedules and perhaps
practice notes setting out various rights and obligations. The institutional
communication may make it clear that it has ultimate control over certain
issues, including determining an applicable fee within parameters based on the
amount in dispute. Acceptance of a mandate under a particular institution's
rules would also typically cover timing obligations, obligations as to the form
and nature of the award and certainly mandatory procedural norms. From a
rights perspective, they would provide for rights to remuneration and a broad
range of discretions subject to mandatory norms.

2.5. Lex Arbitri, Rules and other Legal Provisions

It is widely accepted that the key jurisdictional basis of an arbitrator's rights,

duties and powers is to be found in the arbitration law that is applicable. This
will usually be the arbitration law in the Seat of the arbitration. (56) When parties
select the place of arbitration within a particular geographical location, this
ought to mean that their intention is ‘that the arbitration is conducted within the
framework of the law of arbitration of (that location)’. (57) This is generally
described as the lex arbitri. (58) A lex arbitri might either expand on the powers
conferred on a tribunal by consent or may seek to limit them in some way.
Typically it will do both.

2.5.1. Choosing a Procedural Law other than that of the Seat

While the widely accepted norm is that the lex arbitri is the arbitral law of the
Seat of arbitration, it is not inherently so. It is potentially possible to select the
Seat but also agree to a different procedural law, although this is certainly not
recommended. This is discussed in section 3.8.

2.5.2. General Features of Lex Arbitri

For many countries, concern to identify the lex arbitri in fact leads to application
of the procedural norms in the UNCITRAL Model Law on International
Commercial Arbitration, either because the Model Law has been incorporated
verbatim into a page "67" domestic statute applicable to international
arbitrations, or because the Model Law has been the inspiration for a similarly
worded national law. The key features of the Model Law are discussed in section
3.6.1. The nature and extent of mandatory norms and their general applicability
is discussed further in section 3.7. The role of the lex arbitri and key differences
amongst leading jurisdictions are discussed in section 3.6. In this section, only
general comments are made about the impact of the lex arbitri on the rights,
powers and duties of an arbitrator.

While jurisdictions dealing with international arbitrations may have different

lex arbitri, they will generally display common features. The lex arbitri frames
the arbitrator's rights, duties and powers by determining what is arbitrable
under its rules and by setting out a procedural framework. (59) All will enshrine
key mandatory norms. Fortier suggests that mandatory rules of the forum are
usually limited to the basic requirement to conduct the matter consistent with
due process. (60) Key examples are those giving each party an opportunity to be
heard (61) and requiring that tribunals treat parties equally or fairly.

In most instances, because arbitral statutes will allow the parties to vary all but
mandatory procedural norms and because they provide a broad discretion to
arbitrators, they do not present a complete picture as to the final powers that
are held. For example, when parties select a set of arbitral rules, these would
take precedence over any conflicting norms in the arbitral statute other than
those of a mandatory nature, simply because the arbitral statute has expressly
allowed this variation to be made. The lex arbitri of each key arbitral jurisdiction
will usually expressly indicate that, save for such mandatory procedural norms,
a tribunal is bound by procedural choices agreed to by the parties. In cases
where there is no such agreement, each key lex arbitri gives the tribunal the
broadest discretion to select appropriate procedures.

There is some variation between those that simply offer a broad and undefined
discretion and others which seek to provide guidance and constraints as to how
such discretions ought to be exercised. Article 19(2) of the Model Law simply
provides a broad discretion. An example of an alternative approach is the
English Arbitration Act 1996 which directs the tribunal to concern itself with
costs, efficiency and timeliness as well as key due process norms. (62) Mayer also
makes the point that lex arbitri will vary in focus depending upon whether the
government concerned sees arbitration as a system of justice with different
aims to State courts page "68" or whether it is seen as a substitute. The more
they take the latter view, the more the ordained procedures may follow
domestic litigation models, although over time, most have gravitated to the
former view. (63)

Another variation affecting procedural matters may relate to the range of

powers that a lex arbitri grants to a national court. These may be supportive of
an arbitrator's powers or may seek to limit them. In the latter case, distinct
supervisory powers might, for example, allow a court to remove an arbitrator
for breaches of duty in extreme circumstances. (64) Where broader assistance
powers are concerned, an example is the English Arbitration Act 1996 which
allows for court support of orders made by the tribunal which, if not complied
with within a reasonable time, can constitute a separate contempt of the court.
(65)

Section 3.5.1 looks at the criteria by which parties might wish to select the Seat.
One important aspect is the nature of the lex arbitri that would thus apply. In the
context of this chapter dealing with the rights, powers and duties of an
arbitrator, one particular circumstance is where the arbitrator has the power to
select the Seat. (66) If the choice would have a significant impact on the way the
dispute was to be resolved, an arbitrator may at times be faced with a conflict
between alternatives, none of which are ideal. An example might be a choice
between a place which seems to be the most efficient and a place which is
neutral. A good faith exercise of such a discretion should be seen as valid,
regardless of the choice made. (67)

2.5.3. Agreement or Modifications by the Parties as to the Procedural Norms

Applicable, Including Selection of a Set of Arbitral Rules

As noted in the previous sub-section, most lex arbitri give the parties a broad
right to agree on procedures, subject to the obligation to comply with
mandatory procedural norms. It was also noted that a permitted party choice
takes precedence over what would otherwise appear to be conflicting norms in
the lex arbitri. In that sense there is no real conflict as the lex arbitri itself
permitted such a choice. Agreements may involve ad hoc determinations of
particular procedural matters. More commonly, parties select a largely
comprehensive model via selection of existing institutional or ad hoc arbitral
rules.

page "69"

2.5.4. Other National Laws and Mandatory Procedural Rules

When considering the role of national procedural laws, the first question is
which laws need to be considered. As noted above, it is clearly the case that the
procedural laws of arbitration of the Seat are of paramount importance. They
typically provide the lex arbitri of the arbitration. They are expressly referred to
in Article V(1)(e) of the New York Convention. The public policy in that place
would also be relevant under New York Convention Article V(2)(b).

It has been noted, however, that both party autonomy and arbitrator's
discretions are subject to certain mandatory norms of procedure. There is then a
need to consider what are such mandatory norms. By definition, a mandatory
norm is one that must be applied regardless of the wishes of the parties and/or
the arbitrator. A key example of a mandatory norm within a lex arbitri is Article
18 of the UNCITRAL Model Law requiring the parties to be treated equally and
to be afforded an appropriate opportunity of presenting their respective cases.
Considerations of the nature of mandatory procedural laws are dealt with in
section 3.7. That leads on to the more general proposition that a range of other
legal provisions in an anticipated enforcement country may also need to be
considered. These do not directly control behaviour of an arbitrator but may do
so indirectly given the desirability of conducting an arbitration in a way which
promotes enforceability as a fundamental goal. The general duty to render an
enforceable award and the impact of enforcement country norms on arbitral
powers and duties is discussed further in section 2.7.14.

2.5.5. Procedural Support from Foreign National Laws

National laws may also indirectly impact upon an arbitrator's powers

beneficially by granting support through the court system. A relevant court may
be approached to grant interim measures in aid of the arbitral process. A court
might also be used to assist in gaining evidence. (68)
2.5.6. International Treaties, Conventions and International Law Norms

International treaties, conventions and customs may also be relevant sources of

arbitral rights and obligations. International conventions may directly or
indirectly influence arbitral procedure. Direct impact will naturally be so where
investment arbitration is concerned. The agreement to arbitrate may itself
partly emanate from a treaty such as the ICSID Convention whereby the host
State makes a general offer page "70" to arbitrate with future investors. Such an
offer is commonly accepted simply via the tendering of a notice of arbitration.

There is also the 1994 Energy Charter Treaty. Investment disputes may also be
subject to bilateral investment treaties. Regional arrangements such as NAFTA
or the Lomé Convention may also be applicable in some disputes. These may
contain discrete jurisdictional bases for dispute resolution, express procedural
norms and general provisions that may indirectly impact upon such norms.
International norms may also impact upon private commercial arbitration if due
process aspects of human rights treaties can be said to apply to arbitration.

One of the procedural challenges where investment arbitration is concerned is

that there may then be a multiplicity of treaty, statutory and contractual sources
of jurisdiction and procedural control. Furthermore, there is a need to consider
public international law concepts quite distinct in nature from those typically
underlying private commercial disputes. Some of these international law
principles may impact upon procedural matters. An example of a general norm
that may impact upon express procedural norms is a most favoured nation
obligation typically contained in such intergovernmental agreements. (69) The
essence of a most favoured nation clause (MFN) is to protect the bargain in a
particular international treaty from being undermined subsequently by a more
favourable bargain made with another country. Treaties typically contain
express norms, plus an MFN promise that ongoing treatment will continue to be
no less favourable than the most favourably treated other country. In a range of
investment arbitration cases a question then arose as to whether this promise
should be limited to substantive obligations or whether this should also apply to
procedural norms. This is discussed further in section 13.11.

Where indirect influence is concerned, it should be remembered that the New

York Convention is a binding international treaty. While it does not directly
address arbitral powers, it contains crucial criteria by which enforcement may
be opposed. These criteria are predominantly procedural in nature. Hence it sets
benchmarks for arbitral tribunals that ought to help guide the processes. New
York Convention provisions that allow for non-enforcement when awards
violate norms of public policy entitle courts to consider question of public policy
sua sponte. This is discussed further in section 2.7.14 dealing with the duty to
render an enforceable award.

2.5.7. Customs and Usages

Other sources of arbitral procedure, either directly or indirectly, may include

international customs and usages. In some specialty areas, there are habitual
page "71" ways of conducting certain kinds of arbitration. (70) There might then
be an argument that such habitual processes are implied terms of the
arbitration agreement between the parties.

2.6. Inherent Powers and the Status of Arbitrators

It is common for judges to identify inherent powers flowing from the essential
nature of the judicial office. A question then arises as to whether arbitrators can
apply the same logic. This is a highly contentious question. One reason why it
will remain contentious is that it is impacted upon by the difference in
perspective between those preferring a consent-based theory of arbitration and
those preferring a jurisdiction-based approach. Those arguing in favour of
consent would be less inclined to see such inherent powers, preferring instead
to look at the powers expressly or impliedly granted by the parties. Conversely,
those more concerned with the jurisdictional perspective might at least argue
that a statutory mandate of arbitrator jurisdiction has sufficient similarities to
statutorily mandated judicial powers that some inherent powers may be
considered to apply. (71) Even those supporting a more anational approach could
still argue for a judicial style status. For example, one leading arbitrator has
argued that while arbitrators may not be organs of a particular State, they can
best be seen as organs of a community of States or of the international
community of business persons. (72) That is not the only distinction that
commentators would draw. Hausmaninger argues that inherent powers or
obligations are a common law concept and should thus not be valid sources of
law applicable to arbitral conduct in any event. (73) He suggests that inherent
powers are incompatible with the principle of legality. (74)

Some consider these issues in the context of discussing the status of arbitrators.
Norjarl describes arbitrators as being in ‘a quasi-judicial position.’ (75) Scrutton
page "72" J described an arbitrator as becoming ‘a judicial officer’ on
appointment. (76) Rau describes arbitrators as ‘private judges’. (77) Regardless of
one's view as to status, an arbitrator takes the place of a judge and engages in an
adjudicatory function as a neutral, third party binding decision-maker. (78)
Because the status of an arbitrator could have implications for a range of
procedural, contractual and liability issues, it is not surprising that conventions,
statutes and rules have not sought to directly define it in an overriding manner.
Drafters may well have considered that the rights of arbitrators vis-à-vis the
parties are simply to be negotiated between them, (79) subject to mandatory
procedural norms.

One reason why analogies with judges may be inappropriate is that cases
discussing the inherent powers of judges will often draw such powers from the
national constitution and considerations of the appropriate separation of
powers between legislative, executive and judicial branches of government.
Because a private arbitral function is not part of such a political compact, one
cannot find a similar intellectual basis to justify importation of inherent judicial
powers that are constitutionally inspired. There are other differences as well.
Judges are natural custodians of societal interests in fields such as criminal law,
family law, competition law, environmental law, refugee law and securities
regulation. Court actions are also generally conducted in public. Common law
judges are also an essential element in the process of development of the law.

In many arbitral cases, little will turn on these questions. Many of the inherent
powers discussed in the context of domestic litigation relate to matters that are
required of arbitrators in any event. These include inherent powers to ensure
the right to a fair trial, the right to equality and the duty to respect and promote
due process. Domestic courts typically consider such questions when forced to
determine what inherent powers are needed to effectively support such rights
in the litigants before them. Such debates inevitably arise in the context of what
might appear to be undesirable gaps in the express procedural framework.

While there are unlikely to be many areas where arbitrators would need to
consider inherent powers and status in resolving procedural questions, this will
be more likely to at least be arguable in an area where lex arbitri and
institutional rules simply make no attempt at coverage. An example is an
arbitrator's control over counsel. While the right to counsel of choice might be
seen as a mandatory norm of procedure, can an arbitrator with only a broad
procedural discretion interfere with unethical behaviour of counsel and/or a
subsequent appointment of counsel with an unacceptable relationship to an
existing tribunal member? A theory of inherent page "73" rights might support
such a position. This particular example is discussed in sections 2.9.5 and
6.2.8.7. Another example would be the right to adopt amicus curiae briefs. This
is dealt with in section 7.16 below.

An additional corollary is that if powers are inherent, it may also be that they
cannot be excluded by the parties. (80) Some guidance as to the ambit of such
rights, if accepted, might be gathered from commentary and cases on this issue
in the field of public international law. The ICJ has concluded that it has inherent
jurisdiction in a number of circumstances. For example, in the Nuclear Test case
it observed that it:

possesses an inherent jurisdiction enabling it to take such action as may be

required…to provide for the orderly settlement of all matters in dispute, to
ensure the observance of the “inherent limitations on the exercise of the judicial
function” of the Court, and to “maintain its judicial character”. Such inherent
jurisdiction, on the basis of which the Court is fully empowered to make
whatever findings may be necessary for the purposes just indicated, derives
from the mere existence of the Court as a judicial organ established by the
consent of States, and is conferred upon it in order that its basic judicial
functions may be safeguarded. (81)

In a further case, and in a separate opinion, Judge Higgins considered that:

the court's inherent jurisdiction derives from its judicial character and the need
for powers to regulate matters connected with the administration of justice, not
every aspect of which may have been foreseen….

She further noted that the court has ‘inherent power to protect the integrity of
the judicial process’. (82) Importantly, however, Brown notes that an
international court ‘cannot claim to possess an inherent power if that power is
not necessary for the performance of its particular functions’. (83)

Carlston has argued that in addition to observing provisions in governing

instruments, a ‘tribunal is also expected to conform its operations to the basic
procedural norms. Accordingly, the fundamental procedural norms, whether or
not expressly provided for, comprise (1) “certain fundamental rules of
procedure” (2) which are “inherent in the judicial process” and (3) generally
recognised in all procedure.’ (84) Pauwelyn also suggests that inherent
jurisdiction of any international tribunal includes the power to interpret the
parties' submissions to isolate the real page "74" issues and the object of the
claim, determine its competence to hear a matter, determine whether to refrain
from exercising jurisdiction once established and deciding all issues pertaining
to substantive jurisdiction including burden of proof, due process and other
general international rules on the judicial settlement of disputes. (85) Other
examples would include estoppel and abuse of rights. The ICJ considers that
acquiescence and estoppel follow from fundamental principles of good faith and
equity. (86)

2.7. Duties of Arbitrators

The previous section looked at the broad range of sources of arbitral procedural
powers and obligations. The section did not seek to comprehensively identify
the particular types of powers and duties that an arbitrator derives from such
sources. This section seeks to identify all of the key general duties. The following
section deals with all key powers granted directly or indirectly to arbitrators. A
further section looks at rights where these can be distinct from powers. Once
again, there will ultimately be a need to understand and integrate potentially
conflicting norms. For example, duties may restrict otherwise broadly stated
powers. It is even possible and perhaps likely that there will be potential
conflicts within a range of duties. Thus this section seeks to identify individual
duties and consider what principles should be utilised for reconciling any
conflicts.

A true consideration of the nature of duties must also contemplate the

responses available to alleged breaches of such duties. If there is no meaningful
recourse, it is harder to argue that a meaningful duty exists in the first place.
This question involves two elements. In what forum should such an allegation
be aired? Secondly, if the allegation is made out, what implications and remedies
flow as a result? Liability and immunity of arbitrators are discussed briefly in
section 5.18. They are only relevant to this book in the context that they can or
should circumscribe a tribunal's procedural behaviour. At this stage it is
important to at least be aware of the various potential responses in
understanding how significant a particular duty is. There are five major avenues
of redress against an arbitrator for alleged breach of duties. The first is removal.
The second is annulment of an award. The third is blocking enforcement. The
fourth is a direct cause of action against the arbitrator, either for breach of
contract or via some tortious liability. The fifth relates to the impact on the
arbitrator's ongoing reputation and impact on future appointments. The first
three will directly impact upon procedures, while the latter two can be expected
to have indirect influence.

page "75"

Returning to the question of identifying principles by which conflicts between

duties can be reconciled, there is a need to consider how we might identify a
hierarchy. The two broad approaches are to look for the source of the duties as
articulated in the previous section and look for a hierarchy in that context, or
instead concentrate on the content of the conflicting duties and try and identify
a policy basis for choosing one over the other or otherwise reconciling them.
Where source is concerned, Redfern and Hunter differentiate between duties
imposed by law, those imposed by parties and ethical duties. (87) The source of a
particular duty may also impact on whether it in fact applies. For example,
duties imposed by the parties at the outset and agreed upon by the tribunal
would be contractually binding. Duties sought to be imposed on the arbitrator
during the currency of proceedings require a separate agreement if they fall
outside any express or implied powers to do so under the arbitrator's contract.

Duties might also be categorised in relation to their subject matter as well as

their source and timing. Different typologies have been suggested. General
duties and obligations include a duty of competence and impartiality, a duty of
disclosure, a duty to communicate, a duty to act professionally and with due
care, a duty to act in a fiduciary manner, a duty to uphold the integrity and
fairness of the proceeding and a duty to render a decision. (88) It is possible to
categorise key norms in a range of other ways. For example, Kessedjian has
suggested five procedural rules to promote fairness and enforceability. The first
is the duty to instruct the parties about matters of procedure and seek their
agreement. The second is a duty of transparency so that the parties are not
surprised. The third is to work diligently and sensibly to allow the parties time
to satisfy their due process entitlements. The fourth is to maintain open-
mindedness. The fifth is to prohibit any unilateral treatment. (89)

Another approach is to identify an overriding duty from which others might be

derived. For example, Böckstiegel suggests that ‘(t)he fundamental duty of the
arbitrators is to come to a reasoned decision on the claims put before them after
giving the parties an equal and full opportunity to present their case’. (90) Wetter
describes the primary duty of an arbitral tribunal as being ‘to strive to achieve
justice under law, consistent with the legitimate demands of the international
commercial community’. (91) Born suggests that the arbitrator's task is in hearing
page "76" ‘the parties' submissions and evidence in fair, objective proceedings
and rendering an impartial, reasoned decision that finally decides their rights’.
(92) Philip describes the overriding duty as follows:

It is the duty of the arbitrators in international arbitration to be independent of

the parties and in an unbiased way and in accordance with due process and the
applicable lex arbitri and arbitration rules to make themselves acquainted with
the facts of the case and the claims, allegations and defences of the parties and,
within a reasonably short period of time, to make a reasoned award, based upon
the applicable law, which fulfils the requirements for the award to be
enforceable. (93)

IBA Rules of Ethics, Article 1 indicates that:

Arbitrators shall proceed diligently and efficiently to provide the parties with
the just and effective resolution of their disputes.

Such general articulations are particularly useful in directing attention to the

essential nature of the arbitral process but they can be problematic in the
context of this section in that there would invariably be a need to identify
subsidiary duties that flow from the broader proposition. Because of these
uncertainties, the following sub-sections seek to individually explore all of the
express or implied duties that might thus apply. Some of the categories to be
discussed below are contentious. The categories are sought to be ordered more
in the context of the chronology as to when they are likely to apply in an
arbitration and not in terms of their relative importance. Thus the discussion
begins with the duties, if any, prior to appointment and proceeds through to
completion of the award and post-award behaviour. Only preliminary and
general observations are made at this stage. More detailed analysis of the duties
and the way they should apply in practice is taken up in individual chapters that
follow.

2.7.1. Duties during Pre-appointment Discussions and Interviews and Chair

Appointment Procedures

Section 2.7.8 below outlines the general communication duties of an arbitrator

vis-à-vis the parties. While it is generally accepted that an arbitrator should not
have ex parte communications with a party, it is equally accepted that ex parte
pre-appointment communications are permissible, subject to important
guidelines. These are discussed in section 5.6.

page "77"

2.7.2. Duty of Competence

As noted in the previous section, prospective appointees can be interviewed to

determine their availability and suitability. There ought to be commensurate
duties on prospective nominees to ensure that they are both available and
competent to deal with the case. (94) Duties as to availability are discussed in
sections 2.7.9, 5.5.4 and 5.7 below.

In terms of a duty of competence, while it is easy to state such a principle, it is

not as easy to indicate what criteria ought to apply. This is because there is no
single set of attributes that are necessary to be engaged as an arbitrator. Section
5.5 discusses the various views as to the attributes of an ideal arbitrator. That
looks to identification by a party or institution of a suitable nominee, but notes
that there are differing views on the subject and few constraints. Where a duty
of competence is concerned, such a duty would imply that in extreme
circumstances, a nominee should refuse to accept appointment. Nevertheless,
because parties have the freedom to choose and are the ones who will generally
invite an appointee to take on a position, it is difficult to articulate exactly what a
duty of competence entails in the context of the factors to consider when
contemplating rejecting a party's preference. The essence of the obligation
should be for the nominee to consider their own experience and attributes and
make an objective and dispassionate conclusion as to whether they are truly
suitable and/or whether there are any concerns they have in that regard that
they may wish to communicate to the appointor.

2.7.3. Duty to Remain Independent and Impartial and Duties of Disclosure

Chapter 5 looks at the selection and challenge of arbitrators. It outlines the

requirements for a tribunal to remain impartial and/or independent throughout
the process. Because of this requirement, an arbitrator has a duty to that effect,
with a commensurate duty of disclosure in relation to any circumstance that
may give rise to justifiable doubts as to impartiality or independence. The duty
of independence and impartiality is an ongoing one. Hence, arbitrators should
refrain from any behaviour that could lead to justifiable doubts as to their
ongoing suitability. Born also suggests that the arbitrator's duty of
independence and impartiality includes an obligation to conduct the arbitration
in an impartial manner, treating parties equally and fairly. This seems to be a
logical corollary of it being seen as an ongoing obligation. This is discussed as a
separate obligation in section 2.7.6. (95)

page "78"

Potential tensions can arise between various rights and duties. For example,
does this duty go so far as to suggest that an arbitrator should not accept a
prestigious new appointment until a potentially conflicting existing arbitral case
is completed? Does this duty also imply that such an appointment should be
delayed or rejected if delay is not possible? Section 2.7.12 below looks at the
duty to complete the mandate. These questions are considered further in the
context of that section.

2.7.4. Duties of Disclosure

In addition to the duty to be and remain independent and impartial, there is a

commensurate duty to disclose facts that may give rise to justifiable doubts in
that regard to the parties so that they can form their own views. In the context
of this duty, it is appropriate to consider what should be disclosed, what
investigations an arbitrator should undertake in order to identify potentially
problematic circumstances and finally, what are the implications of inadequate
disclosure. Policy considerations may go much of the way to identifying the
appropriate answers to these questions.

In particular, it is essential to see the duty of disclosure as pertaining to the

rights of the parties to consider whether to appoint a particular arbitrator or
challenge an appointment once made. This is because an arbitrator has a
separate duty to consider whether they have sufficient independence and
impartiality to accept the nomination. If they feel that this is not the case, they
will simply refuse the appointment and there is no discretion left to the parties.
Hence the significant circumstance is where the arbitrator does not believe that
circumstances warrant refusal of a nomination but ought to be aware that at
least one of the parties might take a different view. In this case the parties need
to be made aware of the circumstances before they can come to some conclusion
on the issue. The ambit of what ought to be disclosed is considered further in
section 5.10.

A related question is what investigations must an arbitrator undertake in

support of the duty of disclosure. It is difficult to marry the duty to give parties
an adequate opportunity to consider their own position, with the complexity of
the modern world and the fact that in some circumstances, it may be
unreasonable or impossible for arbitrators to make adequate inquiries. Section
5.10.5 looks at this question further and argues that the obligation should be to
make reasonable inquiries, within the context of the strong duty to give parties
an adequate opportunity to consider their positions. There is also a question as
to the implications of a failure to properly disclose. This is discussed in section
5.10.3.

2.7.5. Due Process Obligations

Regardless of the debate as to whether an arbitrator's essential function is a

creature of contract or results from a quasi-judicial status, all would agree that
there is page "79" an obligation to respect due process principles. A leading
common law scholar has suggested that due process ‘is probably the greatest
contribution ever made to modern civilisation by lawyers or perhaps any other
professional group’. (96) Redfern and Hunter describe this as the duty to act
judicially. (97) Born describes the obligation ‘to resolve the parties' dispute in an
adjudicatory manner’. (98) Civilian texts tend not to speak in such generalities.

Principles of due process will generally be mandatory norms. Thus they are
binding on the tribunal at all stages and cannot normally be waived in advance
by any agreement of the parties. Nevertheless, due process rights may not be
unlimited to the extent that the lex arbitri allows for waiver in the case of
express decisions in that regard or in respect of late objections. Examples in the
Model Law include Article 4 requiring a prompt objection, and Articles 13 and
16(2) which set time limits for certain objections.

There are challenges in understanding the exact ambit of due process in the
arbitral context and the extent to which this can be impacted upon by party
consent or unilateral waiver by one of the disputants. These questions also
impact upon the question of whether an arbitrator can depart from the wishes
of the parties. As to the nature of due process obligations, the ambit of due
process can still be open to debate. Best practice in litigation presumes that
open hearings and rights of appeal are key elements of a just procedure. Yet
these are not elements of arbitration at all. Furthermore, how due process
norms apply in practice may be dependent on the type of arbitration involved as
they are generally expressed in broad terms. For example, in certain specific
industries where the parties seek expedited dispute resolution, that may limit
what would otherwise be fundamental norms or at least provide for variations
in the details of a norm's application. (99)

There are a range of potential due process obligations. Some would be inherent
in all legal systems. Others may flow from the essential nature of a particular
legal family. An ILA Report considered that ‘(d)ue process is an elastic and
imprecise term, with no fixed meaning…’ although it referred to ‘bedrock rules
of procedural fairness’ including the right to be heard. (100) Bernardini describes
due process requirements as having three fundamental principles, the right to
be heard; audi alteram partem, that is the right to be made aware of an
opponent's case and be allowed to rebut it; and the right to be treated alike. (101)
The obligation to treat page "80" parties equally and afford them an adequate
opportunity to present their case is enshrined in virtually all key rules. (102)
Other key due process elements in the Model Law have been said to include
Article 24(1) being the right to require an oral hearing; Article 24(2) which
affords the right to adequate notice of a hearing; Article 24(3) requiring
communication of all documents and Article 26(2) establishing procedures
regarding expert reports and rights to respond. Individual aspects of broad due
process obligations are discussed in sub-sections below.

Even with such fundamental norms there is still a need to consider how other
duties are affected. Concern for key due process norms such as equality and
opportunity to present a case, particularly when there is a concern to err on the
side of removing potential challenges to enforcement, can come at the expense
of efficiency and flexibility. Due process obligations remain fundamental, but the
way they are applied in an instant case can depend upon these other
considerations. The key challenge for a tribunal is to promote efficiency and
expediency without allowing for any legitimate due process complaints.
Furthermore, under some lex arbitri duties of efficiency and expediency may
themselves be mandatory norms binding on a tribunal. In some cases at least,
due process needs to be considered in the context of the trade-off between
fairness and efficiency. Perhaps the most contentious question is whether
proportionality should apply in the procedural design of the process. In this
sense proportionality means that the timing and costs of proceedings should be
in proportion to the amounts in dispute so that the total transaction costs do not
significantly outweigh the sums in dispute.

2.7.6. Treating Parties Equally or Fairly

Lex arbitri will typically enshrine a duty to treat parties equally as a mandatory
due process norm. The ICC Rules have now opted to replace this duty with the
duty to act fairly, the latter being considered to include the former. (103) The
assumption behind the ICC's shift is presumably that this concept better meets
due process concerns and is less capable of abuse by those wishing to disrupt
proceedings who at times argue that exact equality on all matters is required.
Situations where that is a potential problem include the number of witnesses
allowed for each party and time periods for submissions and entitlement to
extensions.

Most laws and rules retain the notion of equality of treatment. Substantive
equality must also be a part of the fairness notion in the ICC rules. Even so, it is
important to consider equality alongside other norms and to understand that its
implications will often be contentious. Equality and fairness are not synonyms.
page "81" Two parties can be treated equally but both be treated equally badly.
Equality must also be integrated with other due process norms. Equality
without an appropriate opportunity to present a case is wholly insufficient to
meet due process norms. Treatment must always be applied based on principles
of fairness and efficiency, perhaps another reason for the ICC shift. It is only
where it is demonstrable that the treatment was unequal based on these
considerations that a challenge ought to be accepted.

Mantilla-Serrano suggests a number of sub-elements of the right to equal

treatment. These include the right to an impartial tribunal, equality in the
appointment of the tribunal, and the right to equal communications with the
tribunal. (104) On this basis, the duty to treat parties equally can impact upon the
decision whether to accept a nomination. An arbitrator who lacks sufficient
impartiality is for that reason alone, unable to treat them equally both in
substance and in appearance. There may also be issues as to unequal rights in
selection where there are multiple parties. (105)

While the ICC reform has much to be said for it, equality is a concept that can
easily be applied fairly by sensible arbitrators. The duty does not require rigidly
identical treatment. For example, it is not necessarily unequal treatment if a
tribunal asks more questions of one party's witnesses than the other's. Similarly,
it is not necessarily unequal treatment if the tribunal awards one party
extensions but not the other, although tensions may arise if one party constantly
seeks adjournments opposed by the other side. Each situation needs to be
looked at in relation to its specific circumstances and in the context of a
substance over form analysis. An example of a substance over form approach to
equal treatment during the currency of an arbitral process relates to time limits.
A useful example is the obligation to provide written submissions. A rigid
application of equality of treatment would suggest that each party should have
the same amount of time to prepare written submissions and hence they should
be simultaneously filed. That would not always be either fair or efficient. For
example, if a claimant's case is based on misrepresentations in contractual
negotiations, fairness would suggest that the complainant particularise the
nature of the representations and by whom they were made. Only then can the
respondent analyse the allegations, take appropriate instructions, identify
relevant witnesses in response and prepare its arguments. To require the
respondent to make a simultaneous submission on the grounds of equal
treatment would require it to guess at the nature of the claimant's particulars
that have not as yet been provided. A substance over form analysis should also
remind us that a claimant can typically take as long as it likes within statutory
limitation periods to prepare and bring a claim. It is only once a claim is brought
that much tighter timeframes then apply to the respondent. Equal time will not
necessarily be true equality in such circumstances.

page "82"

Another contentious aspect in the context of equal treatment is the number of

submissions and the number of times witnesses are questioned by counsel.
Different legal systems have fundamentally different views about equal
treatment where witnesses are concerned. Common law legal systems generally
involve evidence in chief, cross-examination and re-examination. The same is
true for pleadings and submissions, with the norm being primary claims,
responses and replies. Conversely, in some civilian legal systems each side
engages in two presentations with rebuttals and surrebuttals. In the civilian
system, numerical equality equates with essential notions of fairness. In the
common law system, equality is looked at in the context that each side needs
only one opportunity to respond to new arguments of the other.

Importantly, the duty does not apply in a vacuum and must be employed
alongside other due process norms. One challenging question is whether duties
of equal treatment can be tempered by equivalent duties to give each party an
opportunity to present its case in an appropriately effective manner. This may
even mean that true equality may require nuances in dealings with differing
levels of ability between the parties and even counsel. The following sub-
sections include consideration of the right to present one's case and this
subsidiary question as to whether a tribunal should assist an inexperienced
party as part of a duty to provide it with an adequate opportunity to present its
case.

2.7.7. Providing Each Party with an Opportunity to Present its Case

All lex arbitri and rule systems expressly indicate that each party must be given
an opportunity to present its case. This is a widely accepted mandatory norm.
This duty is supported by Article V(1)(b) of the New York Convention which
allows enforcement to be refused if the party against whom the award is
invoked ‘was otherwise unable to present his case’.

Some lex arbitri express this mandatory norm as a requirement for adversarial
proceedings, described in French law as le principe de la contradiction (106) while
Swiss law refers to the parties droit d'être entendues en procédure contradictoire.
(107) Obviously such stipulations in civilian systems are not opting for common

law style proceedings. Instead, they enshrine elements of fundamental due

process, being an entitlement to hear the case as put by one's opponent and the
entitlement to respond. The right to be aware of one's opponent's case and
attempt to rebut it is described as the principle of audi alteram partem. Another
corollary of this notion of an adversarial proceeding is that tribunals ought not
to decide cases on issues that have not been addressed by the parties as it would
effectively take away the page "83" parties' rights to address the issues of
central concern. The practical impact of the principle can still be controversial,
especially when it impacts upon the debate as to whether there is a duty to
know the law and apply the principle iura novit curia and if so, whether
tribunals can decide on principles not addressed by the parties. This is
discussed in section 13.19.

Thus an English Court of Appeal concluded that a Swiss arbitrator who refused
to hear oral testimony did not lead to a breach of natural justice. (108) The
essential question is what obligations the tribunal has in regard to a duty to
afford a reasonable opportunity to present each party's case. Because of
differences in legal families, it is not necessarily the case that the principle of
audi alteram partem comes with a right to be heard orally. Written submissions
and evidence may suffice. The implications of such a norm will depend upon
other aspects of the procedure. For example, if the procedure includes oral
witness testimony, the principle suggests a right to cross-examination. The right
to be heard also does not apply to each and every stage of proceedings. For
example, parties ought not to be able to demand a preparatory conference on
this basis alone. (109)

2.7.7.1. Guiding Counsel and the Parties as to What to Address

The obligation to afford the parties an adequate opportunity to present their

respective cases can include a duty to give sufficient guidance to the parties as to
what the challenge facing them actually is at each stage of the process. If a party
is misguided as to the nature of the case it faces, it may argue that it has not had
a meaningful opportunity to present the case required of it. This would certainly
be so if the tribunal itself caused the confusion. For example, in Brasoil, a review
court considered that the tribunal confused the parties about the proper
approach to a challenge of a partial award based on an allegation of fraud. The
court considered that the tribunal had failed to clarify whether such a challenge
was to be considered based merely on an allegation of fraud or required some
proof of fraud. That confusion meant that the right to proper adversarial
proceedings was violated, as this included the right to know the case of each
party's adversary. (110)

A related question is whether there is a general duty to indicate which issues are
more important and hence a duty to direct the parties as to where they should
devote their energies. While it would naturally seem beneficial to provide such
insight to the parties even without such a duty, the risk is that some parties
might assert that the tribunal has thus prejudged the relative importance of
issues without hearing fully from the parties themselves. Similar issues arise
where a tribunal indicates that it has sufficient arguments and evidence on a
particular issue but page "84" invites the parties to fill the gaps in other aspects
of the case. In these circumstances, a tribunal might be accused of lacking
impartiality, given that interventions of this nature are likely to benefit one
party over the other, although that should not be a valid accusation. The extent
to which an arbitrator may choose to give guidance of this nature will be
impacted upon by the general view the arbitrator takes as to a duty to be
proactive. While that is itself a contentious question, it should at least be
recognised that there is no safe view. Whether arbitrators are alleged to be
either too active or too passive, each could constitute a ground for challenge,
whether successfully or not. These issues are discussed further in section 6.2.
2.7.7.2. Guiding Counsel as to Quality of Performance

A different situation arises where the tribunal may form the view that counsel is
performing sub-optimally and is not taking points that might well be put or has
a severe lack of understanding of international arbitration. A tribunal has no
duty or right to actively run a case for either or both parties even in the face of
lack of optimal behaviour by counsel. In particular, where the obligation is to
provide an ‘opportunity’ to present a case, this relates to the procedural
framework and the entitlement to counsel of choice. Providing a party with an
opportunity is not the same as ensuring that the party presents its case in an
optimal fashion. Having said that, there is a debate as to the educative role of an
arbitrator and what it means in substance to truly give a party an adequate
opportunity when in most cases, this is done through the agency of counsel.
Because an expert arbitrator will in most cases know more about arbitral
processes and options than the parties and even their counsel, an arbitrator can
aid the efficiency of the process and the opportunity of a party to fully present
its case by explaining the procedures and options.

Furthermore, to argue that a tribunal does not have a duty to optimise each
party's case simply as a result of due process obligations does not mean that a
tribunal may not exercise a discretion to assist the parties. In considering
whether to do so, however, a tribunal needs to consider other due process
norms such as the duty to treat parties equally and the ongoing duty to ensure
that justifiable doubt as to impartiality should not arise. If the parties were
equally able to select counsel of choice and one is more able than the other, a
tribunal should be careful in seeking to interfere to try and redress that
imbalance except in extreme circumstances.

2.7.7.3. Duty to Assist someone appearing on their Own behalf or being

represented by Non-lawyer Counsel

The previous sub-section looked at whether a tribunal has a duty to assist

inexperienced or incompetent legal counsel. Conceptually, the situation should
not be different if a party appears in person or uses a non-lawyer as counsel.
From a practical perspective, however, there may be a greater inclination to
provide support in such circumstances. For example, inexperienced public
servants often represent States in ICSID cases. It has been suggested that an
arbitrator page "85" must be sensitive in such circumstances. (111) Once again,
the key advice and warnings from the tribunal should be about what needs to be
done and should not go so far as to performing counsel duties on behalf of the
parties.

2.7.7.4. Duty to Assist Non-appearing Parties

The first aspect of the duty is one of communication, ensuring that the
nonrepresented party is always informed of progress and given an opportunity
to take part at any stage. If the party remains unrepresented, the tribunal will
then seek to test the claims as presented to ensure that they have been made out
to the appropriate level of proof. This is discussed further in section 6.15.
2.7.7.5. Timing and Opportunity to Present the Case

The next issue relates to the timing of procedures so as to afford each party an
adequate opportunity to present its case. The obligation to present one's own
case also includes an entitlement to be aware of the opponent's case in a timely
manner. At times this will lead to allegations that a shift in an opponent's case
should not be allowed or should come with an appropriate extension for the
responding party. This will depend on whether the shift is a fundamental change
in the nature of the case or is merely an improved argument.

There may be decisions as to the length of hearings, the number of witnesses,

the time permitted for cross-examination and re-examination, the balance
between oral and written submissions, the length of written submissions and
the possibility of post-hearing briefs. (112) Furthermore, a tribunal must make
these determinations in the light of other obligations on it such as the duty to
avoid delay and unnecessary expense. The latter duties do not narrow the duty
to provide an opportunity to present a case as this is a mandatory norm, but
they do add relevant factors into the discretionary analysis where efficiency
duties also apply. Giving parties unlimited time to both present and challenge all
possible evidence may aid the pursuit of truth but would lead to horrendous
expense that no one would support. While the duty on the arbitrator should be
seen as mandatory, a party does not have an unlimited right to take whatever
time it needs to present as much evidence as it thinks appropriate. It is for the
tribunal to determine what a reasonable opportunity is. Some rules now invite
considerations of proportionality, i.e., the timing and extent of the process
should be proportional to the amounts in issue.

page "86"

2.7.8. General Communication Duties with the Parties

At the outset, parties will communicate with prospective arbitrators, seeking to

sound them out as to their qualifications and availability. There may also be
further communications as to selection of an appropriate presiding member in a
multimember tribunal. Where an institution is involved, parties will
communicate via the institution in commencing the arbitration and formally
selecting the tribunal. A prospective arbitrator's duties as to appointment were
discussed in section 2.7.4. Section 5.10 deals with the related duty of
investigation and disclosure as to whether there is sufficient independence or
impartiality to accept the appointment. Section 4.8 is concerned with the
subsequent communication duties of an arbitrator. They key issue is to treat
parties equally and avoid ex parte communications.

2.7.9. The Duty of Efficiency and Expediency

Historically, procedural obligations concentrated on due process, fairness,

independence and impartiality factors. While these remain fundamental, over
time, concerns have been raised with the expense and delay in many
arbitrations. Some arbitral statutes seek to impose express obligations on
arbitrators to avoid unreasonable delay and expense.

Aksen even suggests that there is a systemic problem with the time taken to
complete arbitrations. He juxtaposes the natural time to resolve a dispute under
a particular deal and the time that was typically taken to cement the deal in the
first place. (113) Even factoring in the time taken to negotiate and draft complex
contractual documents, such a disparity is likely to be the norm although it does
not presume that arbitration is hence being conducted sub-optimally. Obviously
they are different processes and it may be better to consider what each is trying
to do to best identify whether inordinate delays are being occasioned. Having
said that, empirical work has suggested legitimate concerns as to delays and
costs. (114) Concern with delays has tended to be a key rationale in institutional
revision of rules, (115) and is the basis of arguments in favour of proactivity.

While duties of expediency and efficiency have been seen as more important as
costs and timing of arbitrations escalate, these objectives need to be integrated
with the duty to give all parties an appropriate opportunity to present their case
and the tribunal's duty of due care in consideration of factual and legal matters.
Thus the time taken will in part be proportional to the complexity of the case.
Whether it should be proportional to the amounts in dispute is more
contentious. In addition, page "87" where international arbitration is concerned
and where there may be multiple witnesses in different jurisdictions, problems
of language and translation and the physical need to inspect sites in some cases
and at times undertake scientific testing, the duty of expediency can only be
sensibly considered in the context of the mandates of the adjudicatory
requirements in such circumstances.

2.7.9.1. Efficiency and Expediency Issues when Accepting Appointment

The duty to act expeditiously also imposes ethical obligations on arbitrators at

the appointment stage. An important aspect of expediency is availability. Unlike
judges who can naturally devote their attention to a case once appointed,
arbitrators conduct multifaceted practices, at times combining arbitration work
with other legal or non-legal work. Even those wholly engaged in arbitration
will have a number of cases whose timeframes are not centrally coordinated, all
imposing pressures on the arbitrator. If an arbitrator knows that he or she is too
busy to resolve a particular dispute expeditiously, the appointment should not
be accepted. Bedjaoui speaks of a duty to refuse appointment if a nominee
cannot fulfil the appointment honourably. (116)

This obligation as to reasonable availability is sometimes described as an ethical

duty as it predates the arbitral function and is not in and of itself a breach of that
function. (117) For example, an arbitrator who takes on a case when there does
not appear to be time to do so, might then find that other cases settle and they
can perform adequately. Hence the initial acceptance, when it was unlikely that
they could perform adequately, is better described as a failure to act ethically at
that moment than any failure to comply with an actual obligation of expediency
and efficiency.
2.7.9.2. Exercising the Duty of Timeliness

While it sounds easy to speak of a duty of timeliness applying to arbitrators, this

can readily clash with other norms such as the duty to afford the parties a
reasonable opportunity to present their respective cases. An example of a
potential clash arises where one party seeks simultaneous written submissions
while the other argues for sequential ones on the basis that it needs to have the
claims properly particularised before it can adequately research a response. As
noted previously, timeliness alone would support simultaneous submissions. In
some cases, however, fairness issues will naturally take priority and suggest the
contrary.

page "88"

One question is what an arbitrator should do if the parties seem to be taking too
long or seem to wish to address matters of lesser importance to an inordinate
degree. These issues are discussed further in Chapter 6 in relation to
establishing a procedural framework.

2.7.9.3. Express Time Limits and Arbitrator Duties and Powers

In addition to a general duty of due diligence and expediency, if there are

express time lines in the arbitrator's mandate these must of course be met or
jurisdictional power may be lost. An example would be an arbitration agreement
that requires an award to be rendered within a stipulated timeframe. The
impact on arbitrator power is discussed in section 6.3.

2.7.10. Duty of Commerciality

Julian Lew has argued that because of the special status of arbitrators in
promoting international commerce, there are natural duties of commerciality
that flow as a result:

They are the guardians of the international commercial order:… imply the
presence of commercial bona fides in every transaction; respect the customs
followed in international trade practice and the rules developed in relevant
international treaties; uphold the commonly accepted views of the international
commercial community and the policies expressed and adopted by appropriate
international organisations; and enforce the fundamental moral and ethical
values which underlie every level of commercial activity. (118)

These factors also suggest that arbitrators must remember the ultimate aim of
the arbitral process. This is to provide a resolution of a dispute in a manner
which is consistent with legal norms and is respected in terms of the
commercial aspirations that led the parties to select arbitration in the first place.
Ideally, the losing party should believe that both the methodology and logic of
the award are valid and commercially sound, notwithstanding that the tribunal
may have taken a view on conflicting evidence or legal contentions that the
losing party wished it had not taken. A proper concern for commerciality should
draw attention both to procedural efficiency and fairness and also to the way
substantive law and evidence is viewed. Some lex arbitri or arbitral rules will
support some of these considerations in any event, for example, where an
arbitrator is directed to include consideration of commercial practices and
usages in the determination. (119)

One important aspect of a concern for commerciality is that much of modern

international commerce involves long-term relationships between parties.
Hence page "89" in many international arbitrations, parties that otherwise wish
to have ongoing relationships are engaged in a short-term dispute. A concern for
commerciality in such circumstances should not only be a concern to resolve the
particular dispute but should also be about helping the parties resume mutually
beneficial commercial relationships. (120) In some cases of this nature, a tribunal
is simply helping the parties establish a ruling to guide future behaviour. In
other cases a tribunal might wish to encourage a settlement so that such long-
term interests can rightly dominate over the historical dispute particularly
where key commercial problems are not being addressed in the instant dispute.
Duties and powers in that regard are discussed in section 8.11. More
contentious aspects of such a duty would be in the context of selecting the
applicable law including the possible application of lex mercatoria, adapting the
contract and/or deciding as amiable compositeur. These matters are discussed
in sections 13.6 and 13.4 respectively.

2.7.11. A Duty to Cooperate, Act in Good Faith and Act with Integrity

The tribunal, the parties and any institution would all have mutual duties of
cooperation. One aspect of that is a duty to communicate fully. This has been
dealt with in section 2.7.8. A duty to cooperate also involves cooperation with
other members of multi-member tribunals. Born describes such obligation as
being ‘to make oneself available for deliberations in a timely fashion, not to
delay the arbitral process by unduly protracting or obstructing the deliberations
and to participate collegially in drafting and reviewing the arbitral award’. (121)
In multi-member tribunals, the co-arbitrators must respond promptly to
requests for input from the Chair. The Chair must give co-arbitrators prompt
and sufficient opportunity for such input.

A duty to act with integrity includes an obligation to give appropriate respect to

relevant ethical rules or general rules as to professional behaviour, although it
should be noted that most sets of ethical guides developed for arbitration are
not binding norms emanating from the lex arbitri. Furthermore, professional
ethical rules may be territorially limited and may not expressly or even
impliedly apply to arbitral functions. Nevertheless, arbitral ethical standards
and professional codes will generally deal with matters of ethics and
professional integrity that one would readily wish to apply to arbitrations. Born
has also noted sensibly that even when non-binding, to the extent that leading
codes express ‘the reasonable expectations of commercial parties and most
arbitrators’, then to that extent they could be seen as part of the contractual
obligations owed by arbitrators, at least where directed towards the protection
of the parties and where otherwise consistent with the page "90" arbitration
agreement. (122) The various codes as to ethical conduct are discussed further in
section 2.8. (123) Wetter also suggests that ‘(t)here is such a thing as observing
good manners in international arbitral proceedings, just as in other contexts.’
(124) This is not only a proper way to behave but should aid in the parties' respect

for the process.

2.7.12. Completing the Mandate

Because arbitration is a final, binding and enforceable independent

determination of a dispute, the primary obligation on an arbitrator is to make
such a determination. By agreeing to adjudicate the dispute, an arbitrator is thus
promising to complete the work. This is described as ne ultra petita partium.
The obligation to complete the mandate is expressed in some arbitral rules and
statutes. The duty both identifies what must be done and provides limits on the
power of an arbitrator to resign without just cause. This duty requires diligence
in that task and thus also relates to the duty of expediency. The duty to complete
the mandate also implies a duty to be prepared to conduct proceedings at the
Seat of arbitration unless the parties agree to another place for hearings (125) and
the duty to use the particular language of the arbitration. (126) A tribunal must
apply the relevant law after it is found, otherwise there is a failure to complete
the mandate with appropriate due care. A duty to complete the mandate also
means that a tribunal may not render a finding of non liquet based on gaps or
ambiguities in the law. (127)

The duty to complete the mandate also requires a proper determination

according to law. A tribunal should not accept the role of amiable compositeur
without authorisation. The duty to complete the mandate also involves by
implication a duty not to delegate duties to others. This is seen as implicit in the
arbitral function page "91" and in the fact that the mandate is a personal one.
(128) The duty not to delegate has two elements. The first is that a member of a

multi-member tribunal should not improperly delegate functions to co-

arbitrators. This does not mean that each arbitrator must engage in each
function, however. First, it is perfectly permissible, subject to contrary
agreement of the parties, for the presiding member to make procedural orders
from time to time. In addition, a tribunal member, expert in a particular field or
better able to take evidence in particular physical locations, may be delegated to
do so. Some statutes expressly authorise this, (129) although Born suggests that
such authority should be implied in any event. (130) Care should be taken to
ensure that the process of doing so cannot be said to undermine the co-
arbitrators' mandate to properly consider evidence and make an independent
assessment of it.

The second aspect of an obligation not to delegate duties relates to the use of a
tribunal secretary or other support staff in performing arbitral functions. Here it
is important to delineate between acceptable and non-acceptable tasks. It is
acceptable to delegate a range of tasks as long as this does not involve the non-
member in exercising an adjudicative function or otherwise unduly influence
the tribunal. This means that all routine organisational tasks may readily be
delegated. It is also accepted that research work and even drafting of parts of
the award may be permissible, as long as these are under the direction of the
tribunal and flow from its own independent determinations. The proper ambit
of the role of a secretary to the tribunal is considered further in section 6.8.
Finally, while an arbitrator must complete the mandate, an arbitrator is not
required to comply with any unlawful agreements, procedures or rules. (131)

2.7.12.1. Is there a Duty to Propose Settlement?

It was suggested at the outset that this chapter needed to deal with rights, duties
and powers concurrently to better understand the true status and functions of
an arbitrator. In some cases this even means that it is difficult to identify
whether a particular topic should be discussed under the heading of duties or
powers. This is the case where settlement of disputes prior to an adjudicatory
outcome is concerned. The duty to complete the mandate is, of course, subject to
the parties page "92" rights to settle the dispute. If they settle, they do not
require an adjudicatory outcome.

To some commentators, there is an actual obligation on arbitrators to propose

settlement or at least support the possibility. While some jurisdictions expressly
empower arbitrators to consider advising the parties to consider settlement,
(132) most statutes pertaining to arbitration do not go so far as to make this

obligatory. Nevertheless, even statutes that are merely permissive imply that an
arbitrator has a duty to at least consider when such a discretion ought to be
exercised. The very idea of a statutory discretion implies that the person given
such a power must not refrain from considering when and why it should be
exercised. Nevertheless, because the sources of arbitral duties do not expressly
mandate such an obligation, it is preferable to consider an arbitrator's role in
relation to settlement in the context of a consideration of the powers to do so.
Thus the policy and power issue is discussed below in section 2.10.7, while
section 8.11 looks at protocols which might be employed when this occurs.

2.7.12.2. Completing the Mandate in the Absence of a Party or in the Absence

of Another Arbitrator

Where there is a multi-member tribunal, the duty to complete remains on foot

notwithstanding the possible non-cooperation or incapacity of a particular
arbitrator. The powers of truncated tribunals are considered further in sections
5.16 and 16.4.

The same is true when a party fails to take part. Arbitral rules and statutes
indicate that claims can be dismissed when their proponent fails to pursue the
case. Even where a claimant waives or abandons certain claims, the mandate
may need to be proceeded with in order to render an award in favour of a
respondent, including costs. (133) Where a respondent to a claim fails to
participate, there is no opportunity for summary or default judgment in
international arbitration. Instead, the tribunal must proceed to evaluate the
claims. Furthermore, adverse factual inferences should not be drawn merely as
a result of non-participation. This is particularly so as many instances of non-
participation arise from assertions that the tribunal does not have jurisdiction.
These principles are articulated in most lex arbitri or arbitral rules. Various
courts have also upheld the validity of proceedings conducted in the absence of
an unwilling party. (134) If a tribunal is continuing in the absence of a party, it
must not only consider its duty to complete the mandate but also ensure that it
follows due process in properly evaluating the claims. In the absence of a party
or its counsel being present to test the claims, the tribunal must inevitably take a
more active role in that regard. Sections 2.7.7.4 and 6.15 deal with the duty to
assist non-represented parties. Section 2.7.8 above also noted the page "93"
communication duty in such circumstances, ensuring that the absent party is
fully informed of all steps and that it is given an opportunity to have an input at
any stage.

2.7.12.3. Resignation of an Arbitrator

The duty to complete the mandate also means that an arbitrator should not be
entitled to resign without just cause. Circumstances may of course arise where
an arbitrator might have valid personal reasons to withdraw, such as personal
health or family problems requiring attention. National laws and institutional
rules vary as to the preconditions for an arbitrator to resign. They range from
those which express no conditions to those which provide for a requirement for
permission from the parties or an independent authority. This is discussed
further in section 5.14.

An arbitrator may also resign where some fraudulent or otherwise improper

behaviour of the parties becomes known after acceptance of the appointment. A
more contentious scenario is where the parties agree to conduct the procedure
in a way that the tribunal believes to be suboptimal. Sections 2.10.3 and 6.2.5
deal with this and asserts that the right to resign applies in some circumstances
at least.

2.7.12.4. Exceeding the Mandate

A corollary of the duty to complete the mandate is that an arbitrator should not
exceed the mandate as granted by the parties. This has a number of elements.
First, an arbitrator should not assume jurisdiction where none was provided.
The New York Convention Article V(1)(c) allows for enforcement to be refused
where:

the award deals with a difference not contemplated by or not falling within the
terms of the submission to arbitration, or it contains decisions on matters
beyond the scope of the submission to arbitration. (135)

A related question is whether an arbitrator is entitled to decide a case on an

issue not raised by either of the parties but in relation to a matter that is within
the ambit of the arbitration agreement. A similar question is whether an
arbitrator may override a procedural agreement of the parties. Some
conclusions to these questions would be readily accepted while others are
contentious. It would readily be accepted that the duty to complete the mandate
means that an arbitrator should not decide any issues that were not submitted
for adjudication. If a party has limited its claims, a tribunal should not find for
that party under other potential claims not brought, simply because it believes
the party to be deserving of some redress through that mechanism.

A more contentious question is whether an arbitrator is under a duty to know

the law and apply legal principles not argued by the parties. While this is a
matter of page "94" some unresolved debate, a number of distinct situations
need to be considered. These include less controversial areas such as the
entitlement of the tribunal to adopt an interpretation of law not postulated by
either party, to the other extreme where most would agree that it is not the
tribunal's responsibility to raise a claim not presented. For the purposes of this
section, even that may require qualification. In some cases, the tribunal may feel
that a particular issue would need to be determined, otherwise enforcement
might be considered to be contrary to public policy. This was considered by the
European Court of Justice in the Eco Suisse case in the context of EU competition
law. (136) This can also relate to questions of proactivity, the duty to educate and
the duty to render an enforceable award. As to procedural agreements, a
similarly contentious question is whether in the absence of a mandatory norm,
an arbitrator can ignore a procedural agreement of the parties and still be said
to be completing the mandate as provided by their consent. The latter issue is
discussed in section 6.2.5. Once again, differences in view make it impossible to
naturally classify the question as either one of identifying a duty or a power.

Some institutional rules seek to clarify the limits of the mandate and hopefully
avoid some of these conceptual debates. This is one of the functions of the
Terms of Reference as used under the ICC system. (137) However, the conceptual
debate can simply shift to a question as to whether terms of reference may be
revised and/or when new matters may be raised after terms of reference have
been completed. This is discussed in section 6.13.3.

2.7.12.5. The Status of an Arbitrator after Completion of the Mandate

The obligation to complete the mandate also invites consideration of the status
of the arbitrator once the mandate is complete. Most legal systems would
consider that an arbitrator become functus officio in such circumstances. Some
rules expressly indicate this, albeit with different phraseology. (138) Most lex
arbitri will give an arbitrator the opportunity to rectify errors, interpret awards
or fill gaps with an additional award. In some cases an arbitrator may have
expressly given the parties an entitlement to apply for variations in the award if
certain events occur. (139)

page "95"

2.7.13. Resolving the Dispute in an Adjudicatory Manner

The duties elucidated above relate to the procedure by which an arbitral hearing
will be conducted. It is then necessary to consider what the arbitrator must
ultimately do. As discussed in the previous section, commentators speak of the
duty to complete the mandate, which involves an obligation to proceed to the
end and not resign without just cause. Even that does not indicate what ultimate
task the arbitrator is to perform.

This has led some commentators to speak of the duty to resolve the dispute in
an adjudicatory manner. (140) Commentators often revert back to procedural
norms when discussing the adjudicatory function of an arbitrator but it is
important to consider both the process and also the features and methodology
of the ultimate determination to be made. As previously noted, Böckstiegel
suggests that the ‘fundamental duty of the arbitrators is to come to a reasoned
decision on the claims put before them after giving the parties an equal and full
opportunity to present their case’. (141) Another way to describe deciding in an
adjudicatory or judicial manner is to say it involves deciding the issues between
the parties, after hearing arguments of the parties, considering the evidence
(written and oral), taking account of the contract terms and trade usages and
applying the applicable law or relevant rules. (142) This is in contrast to simply
determining what is ‘fair’ or applying discretion.

Perhaps the most significant benefit in speaking of an obligation to resolve the

dispute in an adjudicatory manner is that this serves to distinguish an
arbitrator's function from that of a mediator or conciliator. Mediators and
conciliators attempt to promote any solution that is mutually agreeable as
between the parties. Such solutions may have nothing to do with the correctness
or otherwise of either or both parties' historical behaviour. An adjudicatory
function is different. It is described as quasi-judicial in nature and involves
making determinations of fact under applicable principles of evidence, including
burdens and standards of proof and determining the implications of the facts so
found under applicable principles of law. An adjudicatory function is essentially
a rights-based rather than an interest-based function, and would be described
by economists as a zero sum game whereby there is ultimately a winner and
loser on each defended claim and cross-claim. (143)

page "96"

2.7.14. Rendering an Enforceable Award

All agree that an arbitrator has a duty to render an enforceable award.

Nevertheless, simply stating this is largely unhelpful. It is important to identify
the source of the duty, determine how it is to be fulfilled, consider what, if any,
limits there are on such a duty and ascertain what options are available to
parties if an arbitrator breaches the duty. (144) The following sub-sections
address each of these issues.

2.7.14.1. Source and Nature of the Duty

Some commentators see the obligation to render an enforceable award as a

central element of the arbitral function. Lew suggests that if an award is not
enforceable the arbitrator has failed the responsibility as vested. (145) Derains
and Schwartz suggest that enforceability is ‘the raison d'etre of the arbitration
process’. (146) Philip describes this as the primary duty of arbitrators. (147)
Böckstiegel suggests that ‘the procedure shaped by the arbitrators must from
the very beginning aim at producing an enforceable award’. (148) Some
institutional rules thus express an obligation to render an enforceable award.
For example, LCIA Rules Article 32.2 states that ‘the LCIA Court, the Arbitral
Tribunal and the parties…shall make every reasonable effort to ensure that an
award is legally enforceable.’ ICC Rules 2012 Article 41 indicates that a tribunal
‘shall make every effort to make sure that the award is enforceable at law’.
Karrer rightly observes that a proper reading should not ignore the opening
words, ‘(i)n all matters not expressly provided for in the Rules…’ which comes
before the obligation to make every effort to make sure that the award is
enforceable at law. (149) On one view that is a gap-filling role. On another view it
is even more restrictive, warning an arbitrator not to fill a gap in a way which
would undermine enforceability.

While all would give some deference to the duty, more problematic is to
determine exactly what it entails and what, if any, additional obligations it
imposes on the tribunal. Given the fact that the grounds for challenging
enforceability cover the key elements of procedural due process, an arbitrator
rightfully concerned page "97" about the latter will automatically be dealing
with those aspects of the duty to render an enforceable award in any event. For
example, Article V(1)(b) of the New York Convention deals with central
mandatory norms of due process, being appropriate notice and opportunity to
present a case. Article V(1)(d) covers situations where the tribunal was
improperly composed or the procedure was not according to the parties’
agreement or applicable law. (150) Jarvin also suggests that the tribunal should
take care to ensure the award does not offend the public policy of the place of
enforcement. (151) This is because Article V(2)(b) dealing with the right to refuse
enforcement on the grounds of public policy, encompasses procedural as well as
substantive matters. Because the better view is that public policy referred to
should be international or transnational public policy, this means widely
recognised norms of procedure. In this sense it may again add little to the duty if
the enforcement country takes a reasonable approach to interpretation. It would
be hard to identify norms from such transnational public policy that would not
also fit within other categories of Article V(1). (152) Furthermore, where public
policy is concerned, while a court may rely on this provision sua sponte, to the
extent that procedural matters overlap with Article V(1), a court might consider
that except in the most extreme cases, if a party has not sought to invoke Article
V(1), then the court should not rely on public policy norms of its own volition.
(153)

The duty to render an enforceable award may also not help to identify the ambit
of other duties simply because of the discretionary nature of enforceability.
Thus a court might still enforce an award in the face of procedural irregularities.
Thus a mere consideration of the grounds upon which enforceability of an
award may be refused adds little if anything to an understanding of an
arbitrator's duties. That is not to say that the duty should not be prominent in an
arbitrator's consideration or that an indication of the duty might not affect
procedural determinations from time to time. This is discussed further in
section 2.7.14.3 below.
2.7.14.2. At What Stages does the Duty to Render an Enforceable Award
Apply?

Most commentators agree that a duty to render an enforceable award should

not impact upon the thought processes underlying a decision on jurisdiction.
(154) There page "98" are a number of reasons for this. Because an award may be

enforced in many jurisdictions and because the tribunal and even the parties
may at that stage be unaware of where assets may be held and hence where
enforcement may be sought, this should not be a consideration when utilising
the direct legal principles in determining jurisdiction.

Furthermore, if the award might be enforced in some jurisdictions but not

others, it is still enforceable. The duty does not extend to making an award
enforceable in any possible jurisdiction. Any opposing suggestion has a number
of fundamental flaws. First is the obvious difficulty of a tribunal determining
where enforcement may be sought given that there is no obligation to notify a
tribunal of each party's assets, both present and potentially future ones.
Secondly, it is conceivable that enforcement norms of differing jurisdictions may
conflict so that the same set of procedures cannot justify enforceability in all
jurisdictions at once. Thirdly, there may be differences in view even within a
particular jurisdiction or between jurisdictions as to whether enforceability
should occur. (155)

One further problem in trying to colour arbitral behaviour by considering the

likely attitude of an enforcement court is that the latter may not be limited to the
factual findings of the arbitrator, although that may be very dependent on the
jurisdiction. (156) Finally, circumstances may well arise where there is a clear
clash between a question of jurisdiction and a question of enforcement. If the
proper legal conclusion is that a tribunal has jurisdiction, but it is also clear that
a particular legal system will block enforcement, a tribunal cannot obviate its
duty to make the right decision under competence-competence simply because
of its knowledge of the latter. (157) However, a tribunal should be alert to the fact
that rejection of enforcement in one jurisdiction may be used as a basis to
support similar applications in other enforcement jurisdictions.

2.7.14.3. What are the Implications of the Duty when it Applies?

While many rules express a duty to render an enforceable award and while such
a duty could readily be implied in any event, it is less easy to articulate exactly
what a tribunal should do in light of the duty. As noted above, this is primarily
because it is a negative duty which in some senses merely reinforces other
obligations that a tribunal may have in any event. The duty is essentially about
not doing anything of page "99" a procedural nature which would leave the
award open to legitimate challenge. Because those procedural duties are all
positive in nature and because tribunals must seek to follow both mandatory
and stipulated procedural norms to comply with their mandate, by simply doing
that to the best of their ability, tribunals are effectively meeting their duty to
render an enforceable award.
To understand the practical ambit of the obligation, it may thus be necessary to
distinguish between binding procedural norms and procedural discretions
which may be affected by the duty. It is more where the latter are concerned
that the duty to render an enforceable award may have practical implications. A
binding norm must be followed in any event. It is where discretions are involved
that parties might use threats against enforceability to urge a particular
discretionary determination, yet even this category is hard to circumscribe.
Typical examples might include challenges to strict timetables, limitations on
length of submissions, limitations on the number of witnesses and the use of
tight Chess Clock arbitration. Parties who do not wish to be bound by such tight
timeframes might argue that the discretionary determinations are interfering
with their right to fully present their case. Yet this is also not a mere
discretionary question if the assertion of denial of rights is valid. Another area
which is not really discretionary is the duty of disclosure on arbitrators as to
matters affecting their independence and impartiality and the decision whether
to accept an appointment or remain in that position when new factors arise
which might lead to a reasonable apprehension of bias. A challenge to an
arbitrator might be made concurrently with a threat to block enforcement if the
challenge is not responded to by voluntary resignation.

Another possible grey area as to any duty to render an enforceable award

involves concern for mandatory rules of the likely place of enforcement. The ICC
Court's internal rules indicate that where awards are scrutinised, the Court shall
consider ‘to the extent practicable, the requirements of mandatory law at the
place of the arbitration’. (158) A former Secretary-General also suggested that
consideration may be given to the mandatory law of the place of enforcement of
the award. (159) If the law should apply, it should do so under the duty to apply
law correctly and not because a duty to render an enforceable award should
change one's theory as to mandatory substantive laws. This issue is discussed
further in section 13.8.

A distinction needs to be drawn between proper challenges to enforceability

that should be upheld anywhere enforcement is sought and conversely,
challenges in jurisdictions that do not give sufficient deference to arbitration,
but which instead look for parochial reasons to block enforcement against their
citizens. If such jurisdictions apply the law improperly, tribunals should not
pander to those predilections. In those circumstances, informed parties may
well have been aware page "100" of the potential problems with enforcement
when deciding to do business. Having selected arbitration in any event, they are
entitled to expect the tribunal to approach questions fairly and reasonably. The
situation is quite clear where there is a clear clash between the arbitration
agreement itself and the provisions of a particular enforcement country. In such
circumstances an arbitrator must follow the arbitration agreement. An example
might be where the arbitration clause establishes a Seat outside of the
enforcement country, but the latter takes the view that certain forms of disputes
must be resolved within its borders. Another area where there may be a
legitimate difference in view between the Seat and enforcement country is as to
arbitrability. If the Seat recognises arbitrability of disputes in relation to such
matters as competition law, an arbitrator cannot come to a different view simply
because a potential enforcement country may have a different attitude. Given
that parties knowingly consented to an arbitration that is unlikely to be
enforceable in that country and empower the arbitrators accordingly, their
wishes must be followed. In any event there are many reasons why persons
might commercially wish to do so. They might expect voluntary compliance.
They might simply want a ruling to guide practice in the future. They might
believe that there are assets in another jurisdiction. They might believe that a
more accurate outcome would be achievable in a neutral Seat. An award might
simply be used as a basis for a claimant's insurance claims. In some countries an
award might be self-enforcing because of an attachment of assets before the
award was rendered. (160) In the above examples of clashes between the
arbitration agreement and enforcement attitudes there is not even a discretion
facing the arbitrator once appointment is accepted. There is simply a duty to
complete the mandate.

2.7.14.4. Is there a Duty to Investigate Enforceability Requirements?

An arbitrator would not normally be expected to investigate enforceability

requirements. Obviously an arbitrator is not expected to consider every possible
enforcement country and its form requirements. It is the primary obligation of
the successful party to alert the tribunal to any such requirements. Nevertheless,
a prudent tribunal might make inquiries in such circumstances or at least invite
the parties to make submissions in that regard. In some circumstances, an
arbitrator might already know that assets are only held in a particular country.
This might be because there has been a security for costs application with each
party providing evidence as to asset holdings. Some procedural obligations
emanating from national systems will be so easy to comply with that there is no
reason not to expect an arbitrator to do so where the facts are known. An
example would be a national law that says that each page of an award must be
signed by the tribunal.

page "101"

2.7.14.5. Is there a Duty to Aid in Enforcement?

The normal position is that an arbitrator is functus officio once an award has
been rendered. Thus an arbitrator has no ongoing legal role in promoting
enforcement. Nevertheless, some commentators have argued for some ongoing
role in appropriate circumstances. Bedjaoui has even suggested that an
arbitrator may, in select circumstances, aid in the enforcement of an award by
convincing the relative authorities of the importance of enforceability norms.
This may be particularly so in investment arbitration. (161) This would certainly
be contentious and would be very dangerous in a practical sense unless the
arbitrator had such a stature that it could be expected that governmental
respect would be almost automatic.

2.7.14.6. Is there a Duty to Produce an Award Protected from Annulment

Challenges?
While arbitral rules only express a duty to render an enforceable award, the
same policy arguments would suggest that a tribunal should take reasonable
steps to ensure that annulment challenges will not be made. Jarvin speaks of ‘a
duty…to avoid a petition by the party to a national court seeking to declare the
procedure invalid’. (162) As with the duty with regard to enforceability, in most
circumstances it does not oblige a tribunal to do anything distinct from its other
duties in terms of following applicable procedures. This should be particularly
so with annulment challenges as a tribunal must be concerned with the duty to
follow procedural requirements emanating from the Seat. From a practical
perspective, a tribunal should also be alert to the fact that a successful
annulment may jeopardise enforcement in other jurisdictions. (163)

2.7.14.7. What are the Implications of a Failure to Comply with the Duty to
Render an Enforceable Award When It Applies?

While there may be a duty, there is no natural recourse against an arbitrator

who fails to render an enforceable award, absent any alternative remedy under
some other more direct procedural obligation. The parties can only know that
the award is unenforceable when an unsuccessful attempt has been made to
enforce it. Even then that may only relate to one of a number of potential
enforcement jurisdictions. Hence, there are no sanctions within the currency of
the arbitration. (164)

page "102"

2.7.15. Duty of Confidentiality

This section only deals with the duty of confidentiality in relation to the
arbitrator. Section 10.16.7 deals with the general duty of confidentiality in
relation to arbitration and in particular, the rights of the parties or third parties
in relation to the use of information obtained through the proceedings. Where
the parties are concerned, key questions are whether confidentiality can be a
basis to withhold otherwise relevant documents where there is a duty to
disclose information to certain regulatory authorities that demand it and what,
if any, liability there is between the parties in circumstances of breach of
confidentiality duties. While these are separate questions to the situation of an
arbitrator's own duty of confidentiality, the answer to each depends in part on
the view one takes as to the essential nature of arbitration. Furthermore,
confidentiality vis-à-vis the arbitrator can have significant procedural
implications where there are parallel proceedings and an attempt is made to
transplant evidence and determinations from one to another.

Confidentiality in arbitration has not been viewed uniformly by all domestic

courts. Nevertheless, the predominant view is that arbitration is inherently a
confidential process and hence there are duties of confidentiality imposed on
arbitrators as a matter of course, whether expressed in the rules or not.
Nevertheless, such principles do not indicate what attitude to take to parallel
proceedings. Some rules do indeed require arbitrator confidentiality. (165) Lew,
Mistelis and Kröll point out that the confidentiality obligations on an arbitrator
are less controversial than in relation to the parties, presumably because the
latter are the beneficiaries of the duty and hence the ones who can waive its
entitlements. (166)

It is generally accepted that deliberations of the arbitrators are confidential.

This is enshrined in Rule 9 of the IBA Ethics Rules unless the parties provide to
the contrary. Article 9 states:

The deliberations of the arbitral tribunal, and the contents of the award itself,
remain confidential in perpetuity unless the parties release the arbitrator from
this obligation. An arbitrator should not participate in, or give any information
for the purpose of assistance in, any proceedings to consider the award unless,
exceptionally, he considers it his duty to disclose any material misconduct or
fraud on the part of his fellow arbitrators.

page "103"

Confidentiality obligations also ensure that procedurally, the tribunal should not
improperly allow third parties to be present. (167) Rogers argues that while
arbitral rules will often not directly deal with arbitral confidentiality, party
expectation would lead to similar effect so that ‘confidentiality may be one area
where the personal integrity and ethical discretion of individual arbitrators
provides the most important protection’. (168) However, Carter argues that the
duty of confidentiality must be seen ‘as coexisting with the right of the arbitrator
to contribute to knowledge’ in the field through education and publication. (169)
That is particularly important given the role of leaders in the field in many
reform exercises, such as through working groups for the IBA, ILA and
UNCITRAL. An arbitrator's obligation of confidentiality may also vary depending
upon whether issues in the dispute become matters of public record. Investment
disputes are semi-public where challenges are made to governmental actions
otherwise subject to public scrutiny. (170)

There may also be complications where certain activities ought to be disclosed

to public authorities. ICDR Arbitration Rules Article 34 states:

Confidential information disclosed during the proceedings by the parties or by

witnesses shall not be divulged by an arbitrator or by the administrator. Unless
otherwise agreed by the parties, or required by applicable law, the members of
the Tribunal and the administrator shall keep confidential all matters relating to
the arbitration or the award.

This suggests that there may be a clash between the agreement inter-parties in
an arbitration and general duties of disclosure under some other statutory
regime. The latter typically apply in relation to criminal actions, or disclosure
necessary for securities and competition law regulation. This is discussed
further in the following sub-section.

One concern with a broad duty of confidentiality relates to arbitrator control

over counsel. In cases of the most extreme unethical behaviour by counsel, a
domestic judge would typically report counsel to the relevant Bar Authority.
Where arbitration is concerned, any desire to do so must be considered in the
context of the duty of confidentiality. General control over counsel is considered
further in sections 2.9.5 and 6.2.8.7.

page "104"

2.7.15.1. Duty to Investigate and/or Disclose Fraud or Corruption

The first question will be whether an arbitrator is compellable in a personal

capacity to answer questions from certain regulatory authorities. That would
depend on those laws themselves and cannot be impacted upon by any general
principles of arbitration law, although in most cases an express protection in an
arbitration statute would protect against the more general operation of a
regulatory statute, if found within the same legal system. (171)

A separate question is whether an arbitrator has a duty to inform authorities of

known or suspected criminal behaviour even when not asked. The first question
is whether the duty of confidentiality acts as a bar in all circumstances to any
such notification. In the event that this is considered not to be the case, the
second question is what evidentiary basis there should be of such a suspicion
before any disclosure is made. There are differences in view between those who
would suggest that disclosure is contrary to the nature of arbitration and those
who would say that extreme forms of improper behaviour undermine the
arbitration itself.

2.7.16. Settlement and the Question of an Arbitrator Exercising a Mediation

Function

All would agree that a mutually agreeable solution is preferable to an

adjudicated outcome. In the context of this chapter there is a question as to
whether there is any duty to promote a settlement and if so, how this relates to
the duty to complete the mandate. A second question is whether an arbitrator
can directly engage in a settlement function or is limited to a duty to advise
parties when an adjudicated outcome would be far less desirable that a mutually
agreeable solution. The key question that is oft debated is whether an arbitrator
once appointed, may adopt a mediation function. Different legal systems have
greater or lesser concern as to whether there would be unacceptable problems
for an arbitrator if he or she had previously unsuccessfully sought to facilitate
settlement during arbitral proceedings. As a result, there is a very significant
difference between legal systems as to the extent to which they permit or even
encourage arbitrators to facilitate settlement by engaging in a mediator or
conciliator function.

While the matter remains contentious, the debate is more about whether an
arbitrator has a legitimate power to engage in mediation or conciliation rather
than whether they have a broad duty to do so, although some speak in the
language of duty. (172) For this reason, the power in relation to assisting
settlement and suggested protocols when doing so are discussed in section 8.11.
page "105"

2.7.17. Duty to Educate

Some of the duties discussed above imply some educative function on the part of
an arbitrator, although the ambit of powers and obligations in that regard and
whether this can be described as an independent duty would be controversial.
Problems may arise where there are inexperienced counsel and the duty to
educate could be argued to offend against the duty to treat the parties equally.
Conversely, this could be asserted to be promoting substantive equality and be
supported by duties to cooperate and act in good faith and by duties of
efficiency.

An educative function also relates to the drafting of Awards. Arbitral awards are
required to contain sufficient reasoning in support of the decisions reached. The
most important value of such reasoning, particularly in systems without a
doctrine of precedent, is to explain to the losing party the basis for its loss and
ideally earn that party's respect for the process, if not the final result. Where
ongoing relationships are involved, the educational function of the award may
also give guidance as to expected behaviour, hence preventing future disputes.
Finally, any educative role can come into conflict with the duty of confidentiality
insofar as an arbitrator seeks to publicise the content and reasoning of an
award, whether through scholarly writing, seminar presentations or reform
exercises.

2.7.18. Duty of Due Care

Redfern and Hunter consider the question as to whether there is a general duty
to act with due care on the part of arbitrators, just as would be the case under
most domestic systems where professionals are concerned. The authors outline
two schools of thought, being a contractual approach preferred in civilian legal
systems (173) and the status approach more prevalent in the common law. (174)
There are two aspects to this question. The first is whether such a duty exists
and if so, what its content would be. Another way to approach the question is to
consider the standard of care required if there is such a duty. Civil liability
systems such as tort law will typically be asked to consider such questions of
standard of care.

page "106"

Contract versus status issues have previously been discussed in sections 1.3,2.1
and 2.2. As has been suggested, polarised positions emanate in large part from
different views as to the consent/jurisdiction paradigm in arbitration and also
as a result of historically divergent thinking between different legal families. The
current preference for a hybrid or functional approach at least makes the more
extreme perspectives somewhat problematic. Furthermore, it has been
observed that some of the extreme comments within each legal family are hard
to reconcile with other aspects of their own jurisprudence. Finally, because the
common law has ultimately accepted that there is a contractual relationship
between the arbitrator and the parties, the better view is that there could be
said to be a general duty of care.

To the extent that there is an obligation of due care, it arises when considering a
nomination for an arbitral position. IBA Rules of Ethics, Article 2(2) indicates
that an arbitrator should only accept an appointment if fully satisfied of
competence and adequate knowledge of the language of the arbitration. Once a
position is accepted, the obligation of due care would relate to the need to
devote appropriate time and attention to the task, concern for the proper
exercise of a procedural mandate and concern for the proper processes of
evaluating evidence and law in coming to a determination.

In most cases, it does not appear to be suggested that there are separate
obligations under a general duty of care than those which would apply under
the other duties of due process and efficiency. For example, an arbitrator who
breaches a presumed duty of care by simply not reading the materials, has failed
to allow each party to fully present their case in a meaningful way. A mandatory
norm such as Article 18 of the UNCITRAL Model Law should not be interpreted
to merely allow for submission of materials, but should also encompass proper
consideration of them by the tribunal. Even here there may be differences in
view depending on the view one takes as to the degree of preparation a tribunal
should engage in prior to the hearing. At one extreme is the view that it is for the
parties to do all necessary research and prepare the case and that a tribunal can
expect to be taken through all necessary elements at the hearing and in the
context of final written submissions. At the other extreme, some would suggest
that an arbitrator has a duty to become fully informed at the earliest possible
time as to the key issues in the case. This would include considering matters not
appropriately addressed by the parties so that they can be forewarned of the
tribunal's views and adequately present their case as best as it can be put. At the
very least, the better view is that an arbitrator should be sufficiently on top of
the issues and materials as they are presented so that sensible timeframes can
be prepared, adequate terms of reference can be drafted where they are to be
used, requests for production of material documents can be dealt with sensibly
and appropriate decisions can be made as to the appropriate number of
witnesses and the length of the hearing. Appropriate preparation and planning
should also allow the hearings to be tailored more effectively towards the key
issues. (175)

page "107"

2.7.19. The Duty to Know the Law – Iura Novit Curia

There is a difference in view as to the degree to which a tribunal must apply the
appropriate law regardless of the contentions of the parties under the principle
iura novit curia. (176) Regardless of the view one takes as to whether a tribunal
should consider the appropriate law notwithstanding the failure of the parties
to do so, the second issue is a procedural one, and potentially of a fundamental
due process nature. If the answer to the first question is yes, the issue is whether
the tribunal should draw the parties' attention to the relevant law at an early
enough stage to allow them to make submissions and engage in an adversarial
contest or instead, whether the arbitrator is entitled to or indeed bound to apply
a law after the conclusion of the hearing if it is considered the appropriate one
notwithstanding the failure of the parties to address it.

The issue is also impacted upon by the view one takes as to the duty or
entitlement of a tribunal to be proactive. The more one supports that approach,
the more one accepts that a tribunal should ensure that appropriate legal
principles are brought to bear on the dispute, regardless of the adequacy or
otherwise of the parties submissions. The issue is also complicated by
differences in view between civilian and common law jurisdictions which in part
are impacted upon by the fundamental approach that each takes to adjudicatory
dispute resolution. It is far more natural to find civilian jurisdictions supporting
the principle of iura, given an inquisitorial model that makes the judge the
primary actor in seeking a just solution. Conversely, common law jurisdictions
that historically saw judges as passive recipients of adversarial arguments by
opposing litigants, were far less supportive of any notion that a judge could
decide on principles not presented by the parties. Different attitudes between
civil and common law may also impact upon notions as to what an adjudicator is
seeking to do. An adjudicator seeking to get the correct decision may follow the
principle. Conversely, an adjudicator who simply believes that their job is to
choose which version is more likely to be true, may refrain. (177)

These issues will be explored further in Chapter 13 dealing with identification

and interpretation of applicable law. While the issues are subject to significant
scholarly debate and differences in view, the vast majority believe that there is
no actual duty to know the law under the iura principle, but in appropriate
circumstances, tribunals are entitled to do so. Furthermore, the vast majority of
jurisdictions and commentators are of the view that an arbitrator who believes
there is a power to consider other principles of law than those cited in
submissions, should at least warn the parties that he or she is inclined to do so
and allow them to make submissions accordingly. This view is to be
commended. Any contrary view page "108" offends against the duty to give
parties an opportunity to best present their case, given that even legal systems
that do not demand legal submissions by parties still allow them to make them.
It would also offend against party autonomy to deny them the entitlement to
direct the tribunal as to the attitude it should take to this question.

2.7.20. The Duty to Apply Mandatory Laws

A subset of the question as to whether there is a duty to apply the law is

whether that varies depending on the type of law involved. The most obvious
example would be a law that purports to apply regardless of the will of the
parties or a tribunal. This relates to mandatory laws. Born suggests that the duty
to apply the law includes a duty to apply relevant mandatory law provisions. (178)

All would agree that mandatory procedural norms of the Seat must be applied
regardless of the wishes of the parties. Furthermore, an arbitrator following
such norms will of necessity give sufficient warning to the parties of their
content. Without appropriate warning, due process mandatory procedural
norms are simply not satisfied. For example, allowing for an oral hearing with
equal time and sufficient time periods, allows for equality of treatment and an
adequate opportunity to present the case. The more contentious question is
whether arbitrators must apply mandatory substantive laws in all
circumstances. Here there is again a need to differentiate between a number of
scenarios. There is nothing controversial where both parties want such laws to
apply. Hence the three problematic scenarios are where the parties simply do
not raise the question, where they do but are not in agreement as to
applicability and finally where they expressly direct the tribunal not to apply
such mandatory substantive laws. Responses of either resigning or ignoring an
express prohibition are canvassed in sections 2.10.3, 2.12 and 6.2. Further
analysis of the potential applicability of mandatory substantive norms is
undertaken in section 13.8.

2.7.21. Is There a Duty to Respect Court Orders?

Another way to consider the ambit of rights and obligations of arbitrators is to

consider the interaction between the tribunal and supervisory and enforcement
courts. The more the latter may circumscribe arbitral behaviour, the more we
should understand that there are thus limits on arbitral powers. Conversely, the
more that an arbitrator can resort to courts for assistance purposes, the more
effective their direct powers are likely to be.

The first question is which orders from which court an arbitrator ought to
consider during the currency of a hearing. This should only arise with a
supervisory page "109" court as articulated in the lex arbitri. There are a range
of reasons why an arbitrator ought to respect a decision of a court in a
supervisory jurisdiction, although the contrary view could be propounded by
those who adopt a wholly anational approach to arbitration. In terms of the
arguments in favour, given that a hybrid or functional approach is now the
predominant view, there is the issue of respect for sovereign rights of the State
affording an arbitrator jurisdictional power; the concern that an arbitrator
might be in contempt of court failing compliance and the duty to render an
enforceable award and the related duty to not conduct an arbitration so as to
encourage annulment. (179)

In many cases an arbitrator is expressly permitted to continue while court

challenges are proceeding. For example, Article 8 of the Model Law provides for
situations where a tribunal is empowered to proceed notwithstanding seizure of
the matter or an important element of it before a relevant court. A similar
situation is covered in Article II(3) of the New York Convention. A similar
approach is taken under Article 16(3) of the Model Law in relation to challenges
to jurisdiction. A tribunal might also proceed in the face of a challenge to an
arbitrator where the rules permit. (180)

In cases where a court declares that an arbitration should not proceed,

circumstances may arise where the behaviour of that court is so unreasonable
that a tribunal might wish to ignore it. Obviously that is problematic in cases
where the lex arbitri makes the supervisory court the higher authority.
Nevertheless, because enforceability is possible in relation to annulled awards, a
tribunal with a duty to complete a mandate and operate with due care should at
least be prepared to countenance a circumstance where it concludes that no
reasonable supervisory court could have come to the view that was taken. (181)
An extreme example would be if the tribunal knows that the supervisory court
made an adverse ruling following a bribe. The distinction ought to be between
good faith and abusive determinations by a supervisory court. If the decision is
in good faith and the supervisory court is given a higher authority under the lex
arbitri, then its view should be respected simply because of the hierarchy of
adjudicatory power. The difficult question is then when and how an inferior
adjudicator is entitled to rule that the behaviour of a superior body is abusive
and can be ignored.

page "110"

2.1.22. Is There a Duty of Loyalty to Arbitral Institutions?

Hunter and Paulsson have queried whether there is a distinct duty of loyalty
towards an arbitral institution that is involved in proceedings. (182) At the very
least, they note that an arbitrator should not undermine the authority of an
institution that is involved. That could flow from the need to reconcile
conflicting contractual obligations to the institution on the one hand and the
parties on the other. Having said that, there may be conflicts between the way
an institution wishes a matter to proceed and the way the arbitrator believes
would be optimal.

2.8. Ethical Rules and Principles

Previous sections have tried to identify a range of applicable or arguable duties

that arbitrators are subject to. Another way to clarify the nature and range of
duties is to consider what, if any, broad ethical duties they may have. Related to
that is the possibility of clarifying certain duties by articulating ethical
standards. Some ethical principles would go beyond procedural questions but
many would circumscribe the way procedural decision are to be taken.

There is a reasonably vigorous debate as to whether there ought to be ethical

rules imposed on arbitrators and if so what they should contain. (183) Some
academic commentators will typically call for such standards. Some institutions
will attempt drafts or at least establish working parties aiming to do so.
Conversely, some leading practitioners will question the need, arguing that the
system ultimately depends on the personal integrity of leading individuals. In
the context of such polarised positions, some policy observations are
appropriate in the context of identifying the impact of ethical considerations on
arbitral behaviour. First, most legal systems are unwilling to rely on complete
personal self-regulation unconstrained by any express standards. Secondly, lex
arbitri already contain a number of significant direct or indirect ethical
standards such as the duties of impartiality and independence. Arbitrators
would all agree on proscriptions of some of the more significant violations of
ethical behaviour such as an ex parte discussion about likely predilections on
the merits when nominations for appointment are being considered. Some
ethical standards might be implied into contracts between arbitrators and the
parties. Whether standards are expressed in lex arbitri, institutional rules,
professional associations or are left to be determined on an ad hoc basis by
individual arbitrators or courts, reasonable standards must exist and should be
applied in a sensible and consistent manner. The practitioner who says page
"111" that the optimal solution is to rely on the individual integrity of an
arbitrator, is simply saying that he or she will know when an ethical conflict
arises and know how best to deal with it. They are implying that they know key
principles without having to be told. If that is true, it might be desirable to try
and articulate the nature of those situations and the recommended methodology
when they arise for the benefit of others. Rogers also makes the important
observation that ethical standards not only constrain arbitrator behaviour, but
also gives them protection against spurious challenges by setting standards that
parties must respect. (184) While the foregoing argues that ethical standards
could well be articulated one way or another, the caution of sceptical
practitioners is at least useful insofar as it raises doubts as to whether very
detailed prescriptions could be agreed upon that do not involve complex
nuances on a case-by-case basis.

With these policy concerns in mind, it is appropriate to consider what rules,

cases and commentary have to say on the way that ethical issues impact upon
the rights, powers and duties of arbitrators. Various attempts have been made to
incorporate ethical obligations into express codes of conduct applicable to
arbitrators. These include the IBA Rules of Ethics for International Arbitrators
1987 (IBA Ethics Rules) and the IBA Guidelines on Conflict of Interest in
International Commercial Arbitration (2004) (IBA Guidelines). The IBA Ethics
Rules overlap to some degree with the IBA Guidelines. Paragraph 4.57 of the
latter indicates that the Guidelines replace the IBA Ethics Rules to the extent of
overlap. At the domestic level there is the American Arbitration
Association/American Bar Association Code of Ethics for Arbitrators in
Commercial Disputes (2004). (185) A number of US states have enacted ethical
rules but these largely apply to domestic arbitrations and are outside the scope
of this book. (186)

While these are non-binding guides, they could be made binding if parties
expressly incorporate them into their arbitration agreement or otherwise
directed arbitrators to comply with the guidelines. Another question is whether
a supervisory court evaluating an arbitrator's behaviour is constrained by
ethical rules adopted by the parties. Such a commitment by the parties should
not be ignored but there may well be other and perhaps overriding ethical
considerations that a court might apply. (187) An institution that promulgates a
Code will typically make it a condition of appointment that an arbitrator abides
by it. An example of an international institutional code is that of the Singapore
International Arbitration page "112" Centre, Code of Practice: Code of Ethics for
an Arbitrator. (188) Even if not expressly included, this might be implied. At the
very least, these ought to give strong guidance to arbitrators as to emerging
norms of behaviour. Schwartz also argues that ‘ethical’ obligations are inherent
in the arbitral function itself. (189)

Regardless of the reasons why they may apply, there is then a need to consider
their content. The individual elements of the key codes are not articulated here,
as relevant provisions are discussed in relation to particular duties and powers
throughout this chapter and other more specific chapters. As a general
observation, however, it should be noted that while there have been a range of
admirable initiatives, it is only to be expected that broad rules of ethics,
negotiated between experts from diverse legal systems that are not aimed to be
binding in and of themselves, are expressed in general language which may at
times fail to give adequate direction in relation to a particular concern.

In addition to ethical rules directly applicable to arbitrators as such, a number of

other rules might potentially be applicable as a result of an arbitrator's
particular professional qualifications. Commonly an arbitrator will be a
professional of some description, perhaps lawyer, engineer, architect or
accountant. There is then a question as to whether the general professional
ethical rules have implications for the arbitral function. Normally this would not
be so. Such rules rarely expressly refer to an arbitral function. (190) Absent any
express direction, there would be a need to consider whether they apply only
when a person is exercising that direct professional function or whether they
also purport to apply when the professional acts as arbitrator, given that the
latter role does not require the particular professional qualification except
where expressly sought by the parties. Born makes the point that professional
rules are concerned with a professional/client relationship, which is quite
distinct to the relationship between an arbitrator and the parties. (191)

They may still indirectly have some implications, for example in relation to
potential conflicts of interest. However, it would be unlikely that they would
impose standards over and above those expressed or implied in relation to
general arbitrator duties. Court decisions reviewing arbitrator behaviour also
help to identify standards of conduct that are required, although such views may
not be consistent or indeed accurate. (192)

page "113"

2.9. Powers of Arbitrators

While this chapter has dealt separately with the question of duties of
arbitrators, it was noted at the outset that there is a natural relationship
between duties, powers and rights. The more there is a duty to do something,
the more one would expect that there are implied powers to that end.
Conversely, where there is a duty to refrain from acting in a particular way, that
might circumscribe otherwise broad powers or discretions. Indeed the previous
sections inevitably dealt with the discussion of powers alongside a discussion of
various duties. Furthermore, it was observed at the outset that in some cases it
is debateable whether a particular practice should be seen as emanating from a
power or duty. While any division is to some degree artificial and contentious it
can still be a useful way to list and analyse various matters that ought to be
considered by arbitrators. With that caveat again in mind, the following sub-
sections deal with individual powers as may be applicable to arbitrators. Where
appropriate, these sub-sections cross-reference back to the duties which may
impact upon their ambit. In most cases, they also cross-reference to other parts
of the book that deal with the specific application of the rights. This chapter is
only concerned with identifying their general nature and how they may
integrate with other rights and duties.

2.9.1. Competence-Competence, Separability and Jurisdictional Challenges

Chapter 8 looks in detail at the way an arbitrator may wish to deal with
jurisdictional challenges. For the purposes of this chapter, it is appropriate to
consider the conceptual basis for the powers in that regard. All recognised
arbitration statutes and rules provide a power for arbitrators to resolve
questions as to their own jurisdiction, described as competence-competence. Key
examples of the power include Article 16 of the Model Law. Without such a
power, arbitration would be dysfunctional in the face of such challenges. If an
arbitrator only has valid authorisation where there is a valid agreement to
arbitrate, how can a person whose authority is in dispute make such a
determination? The issue is circular without an express grant of such
competence.

Allied to the concept of competence-competence is the doctrine of separability

or autonomy of an arbitration agreement. Just as arbitration would fail if there
was no way out of a challenge to the arbitrator's jurisdiction to decide matters,
so too would there be problems if allegations as to the invalidity of the
underlying contract might naturally taint an arbitration clause in that contract,
hence also undermining arbitral jurisdiction, at least until a final determination
was made on the merits. The solution is that of the doctrine of separability or
autonomy of the arbitration agreement. This treats the agreement to arbitrate as
separate from the balance of the contract even if it is contained as a clause
within it. Even if the contract as a whole is invalid because of a failure to follow
form requirements or because it is void or voidable under principles such as
misrepresentation or duress page "114" or if the contract is at an end because of
a fundamental breach accepted as repudiation by the innocent party, the
doctrine of separability holds that the arbitration agreement remains on foot as
a separate agreement to arbitrate, not tainted by any of the defects in the
balance of the contract.

The power to effectively consider questions of jurisdiction may also come with a
duty to do so, although that is contentious. On one view, either a duty to
complete the mandate and not exceed it and/or the duty to know the law might
imply an independent assessment of the basis for jurisdiction even when this
has not been challenged by the respondent. A contrary argument would be that
challenges to jurisdiction are waivable matters and should not be raised by the
tribunal in the absence of such a claim by a party. That is discussed further in
section 6.2.8.10 and also in section 8.1. These powers may not even need to be
expressed. An accepted example of an inherent power is competence-
competence. (193) Another is the right to hear a preliminary objection on
jurisdiction. (194)

2.9.2. Power of General Control, Procedural Discretions and Case

Management by Arbitrators

It has been noted that subject to party autonomy and the application of
mandatory procedural norms, an arbitrator has the broadest discretion as to
procedural matters. While that is expressed in virtually all lex arbitri and
arbitral rules, the discretions are not unbounded and are subject to the various
duties outlined above.

There will commonly be a number of questions as to the optimal way to

proceed. Practical considerations are dealt with throughout this book, in
particular in Chapter 6. For the purposes of this chapter, some of the key issues
in terms of exercising general discretionary power include the degree to which a
tribunal may choose to be proactive in the context of due process norms,
including helping a less able party outline its case; and when a tribunal may
refuse to follow a procedural direction from the parties. The question of the
limits on the entitlement to be proactive were considered in section 2.7.7 in the
context of providing each party with an opportunity to fully present its case. The
right to override a procedural agreement of the parties is considered in section
2.10.3.

2.9.3. Choice of Law Discretions

Chapter 13 deals broadly with issues of interpretation and choice of law by

arbitrators. In this section it is simply appropriate to note that as with
procedural matters, the tribunal is generally given a broad discretion as to
choice of law, page "115" subject again to party autonomy and subject to the
more controversial questions as to the applicability or otherwise of mandatory
substantive laws and the relevance of the principle iura novit curia. They are
addressed further in Chapter 13.

In addition to party autonomy, which circumscribes both procedural and choice

of law discretions, where the latter is concerned, the tribunal's residual
discretion is also more circumscribed simply because the rules will typically
choose between affording a discretion to pick the appropriate conflicts of law
rule or empowering a direct choice of substantive law by the arbitrator. Many
commentators point out, however, that the distinction is more formal than real.
A tribunal choosing amongst a range of possible conflicts methodologies may
consciously or subconsciously consider the outcome under each as a relevant
factor in making the determination. An arbitrator concerned to pick a
substantive law directly, may consider the kinds of factors considered
appropriate under leading conflicts methodologies in coming to that conclusion.

2.9.4. Powers in Response to a Failure to Attend or Respond by a Party

and/or Waiver of Jurisdictional Challenges

The duty to complete the mandate, as noted above, comes with a commensurate
right to continue with the proceedings in the absence of a party, or following
waiver of a potential right to assert lack of jurisdiction. A distinction is drawn
between a person whose obligation it is to pursue a claim and a person
responding to that claim. The case must proceed in the face of the latter's
inaction. Where the proponent of the claim is concerned, appropriate
circumstances of inaction the tribunal may conclude that the claim has been
abandoned. As noted above, however, there may be a need to proceed to render
an award in favour of the respondent, including as to costs.

There are important evidentiary and interpretation issues that arise when
proceeding in the absence of a party that have been outlined in earlier sections.
The first is that arbitration rules sometimes ensure that a failure to appear is not
deemed to be an admission of the other party's assertions. (195) That confirms
that a claimant cannot succeed outright in the absence of a respondent's
attendance. Other important duties in these circumstances besides the duty to
complete the mandate are the duty to communicate and the duty to render an
enforceable award. As suggested in section 6.15, a tribunal facing an unwilling
party needs to ensure that there is a strong evidentiary record showing that
they were always invited to participate and were invited to have input into the
determination of an appropriate timeframe and procedural framework. More
controversial is the question of the degree to which the tribunal can test
claimant's submissions and/or engage in its own research. This is discussed in
section 6.15.3.

page "116"

2.9.5. Power over Counsel

There are rarely any express rights in relation to counsel given in lex arbitri or
arbitral rules. A tribunal's powers and rights in that regard are controversial
given that counsel is not a direct party to the mandate via consent as given to
the arbitrator and is not a party to the contract between the parties and
arbitrators. It may be arguable that control over the parties can be used to force
them to control the behaviour of their counsel. Two added complications are the
duty to afford due process and allow a party an adequate opportunity to present
a case. This is normally seen as giving a right to counsel of choice. Finally, where
counsel are thought to be behaving inappropriately, duties of confidentiality
may make it difficult for tribunals to determine whether notification to
professional authorities can be made. Hence the issue remains controversial.
Section 6.2.8.7 deals further with a tribunal's control over counsel. (196)

2.9.6. Deciding as Amiable Compositeur

In the normal course of events, an arbitrator exercises an independent

adjudicatory function. In doing so, the arbitrator determines questions of fact
based on the evidence provided and subject to the applicable law. An arbitrator
is expected to rigorously apply such principles of law and undertake an
appropriate fact finding procedure.
While that is the norm, most institutional rules allow the parties to expressly
empower the tribunal to act as amiable compositeur or ex aequo et bono. Where
this occurs, the tribunal is not required to strictly apply legal rules. (197) While it
is easy to say that a tribunal empowered to act as amiable compositeur is
entitled to depart from the strict application of legal rules it is less easy to
determine the proper ambit of that departure. This is discussed further in
section 13.14.

2.9.7. Interim, Anti-suit and Sanction Powers

Specific aspects of a tribunal's interim and sanction powers are considered in

Chapter 8 Only some broad observations are appropriate in the context of this
chapter. Regardless of the view one takes as to the status of an arbitrator, in a
practical sense, tribunals have little of the effective power of judges. Some
arbitral statutes and rules give tribunals a power to award security for costs and
other page "117" interim measures. (198) The UNCITRAL Model Law and Rules
were revised to broaden such powers. (199)

Some legal systems allow arbitrators to provide punitive sanctions in support of

their orders or awards. Astreinte and judicial penalty powers are discussed in
section 14.21. As a general rule, the tribunal has no direct coercive powers
although in some circumstances it may provide effective sanctions in any event.
For example, if there is a failure to provide security for costs, the claims may
simply not be proceeded with. In some circumstances, adverse evidentiary
inferences might be drawn. Poudret and Besson make the point that regardless
of the debate about the existence of such powers, an important issue is whether
such a power is directly enforceable or not. The authors doubt the enforceability
of such determinations. (200) Also important is the question as to whether a
tribunal may seek the assistance from courts in enforcing such interim
measures. Some lex arbitri provide for applications to a court by an arbitrator or
at least provide for court support of arbitral determinations.

2.10. The Arbitrators' Rights

As noted at the outset, while there is a significant interrelationship between

duties, powers and rights, nevertheless it was thought appropriate to separate
discussion to both, analyse each individually, and consider which other elements
conflict or at least constrain each of them individually. The following sub-
sections simply deal with those rights of a tribunal that could discretely be
described as such. The aim is not to repeat reference to the various powers
where it naturally flows that a tribunal with a particular power has a right to
exercise it.

2.10.1. Right to Remuneration

The essential nature of the contract between the arbitrator and the parties is to
provide arbitral services in return for remuneration. Such a right is hence the
most central from the perspective of the arbitrator. Such a right is supported by
various national laws and arbitral rules and would also be implied into any valid
contract, although a failure to identify a remuneration mechanism may in some
circumstances render a contract ineffective. This would rarely be so. Where an
institution page "118" is involved, it would be implied that the remuneration is
as per the institutional rules. In an ad hoc arbitration, many legal systems would
imply an agreement to reasonable fees.

The right to remuneration is constrained by a range of factors. In many cases the

rights are subject to supervisory control and even determination by an
independent body such as an institution. It may also be impacted upon by the
duty to complete the mandate where this has not occurred or where duties of
diligence and efficiency have arguably not been met. Hence a range of factors
need to be considered in relation to who determines fees, by what methods,
what ethical obligations are there on arbitrators in negotiating fees,
entitlements to cancellation or commitment fees and what may be done where it
is alleged that excessive fees have been charged. These questions are discussed
further in section 5.17. In the context of the scope of this book, disputes as to
fees may impact on procedure where a tribunal refuses to proceed or claims a
lien over an award. A tribunal might also be challenged where behaviour as to
fees is considered improper. Undue fee pressure may be asserted to be an
interference with due process.

2.10.2. Right to Cooperation from the Parties

In section 2.7.11 above, it was suggested that an arbitrator has a duty to

cooperate and act in good faith. Because the duties are mutual, they also entail
rights for the tribunal. Born considers that there is a right to cooperation from
the parties emanating from the arbitrator's contract and applicable law. (201) This
can be supported analytically. Arbitration is consent based. It is appropriate to
presume that such consent was given in good faith and was intended to be
maintained in good faith. That should be so regardless of the actual behaviour of
individual parties from time to time. If that is so, an arbitrator can extrapolate
certain duties applicable to the parties and rights for the tribunal from this
notion alone, even in the absence of further express direction. These principles
should be seen as implied terms of the contract, being fundamental aspects of it.
This means there should also be interparty duties in this regard. Other
commentators discern such a right otherwise than via implied contractual
terms. Fouchard, Gaillard, Goldman describe it as a ‘moral right’. (202) Fouchard
previously suggested that ‘the arbitrator has the right to meet with faithful and
co-operative behaviour from the parties throughout the whole of the arbitration
procedure’ seeing this as a principle which ‘goes without saying’. (203)

page "119"

Parties are obliged to act in good faith, provide appropriate information and act
expeditiously. They must also provide the fees and expenses as required. (204)
Jarvin suggests that the parties must also refrain from acting in a way that
makes enforcement more difficult or aggravates an ongoing arbitration. (205) The
obligation to cooperate in good faith is also said to extend to the presentation of
evidence. For example, Rule 34 of the ICSID Rules indicates that the parties are
to cooperate with the tribunal in the production of evidence. (206) More
controversial would be whether good faith obligations of cooperation would
apply to presenting legal arguments contrary to a party's position.

In the extreme, if a party seriously breached the obligation of cooperation, that

would not only give rise to claims by opposing parties but may also entitle the
arbitrator to redress. If the non-cooperation caused financial loss, damages
equating to a higher remuneration might be applicable. In an extreme case it
might provide justification for resignation. The latter would depend on what
constitutes a fundamental breach of the contract with the parties, where that is
the source of the right.

2.10.3. The Right to Override a Procedural Agreement of the Parties

As argued throughout, an arbitrator's duty is to meet certain standards of

procedural fairness and efficiency, all considered in the context of party
autonomy and the consent which underpins jurisdiction and which frames
procedural decisions made from time to time. Section 2.12 below deals more
generally with how these different factors may be reconciled where they
conflict. This section is concerned with a specific question that is a subset of that
analysis, namely when can a tribunal ignore a procedural agreement of the
parties and when ought it to do so? The question can arise in a range of ways
and follow a conflict between a range of differing duties. The need to follow the
agreement of the parties flows naturally from party autonomy and the duty to
complete the mandate. The mandate emanates from the parties' agreement. Any
well developed lex arbitri will also provide that procedural discretions of an
arbitrator are subject to the parties' agreement. Nevertheless, it has also been
observed that parties cannot override a mandatory norm of procedure. Thus the
first situation where a tribunal may override a procedural direction of the
parties is where they have made an agreement contrary to a mandatory
procedural norm. For example, the parties are page "120" not allowed to agree
that only one of them is entitled to present its case or be represented by counsel,
or have witnesses give evidence.

This only indicates what a tribunal should not do. It is a separate question as to
how a tribunal should proceed in the face of such a purported agreement by the
parties. As noted throughout, this can involve a complicated interaction between
the arbitrator's contract with the parties, the role of party autonomy, and the lex
arbitri, all in the context of the timing of the purported improper agreement. If
the parties' original arbitration agreement purports to override a mandatory
procedural norm or if such an agreement predates the arbitral nomination, the
arbitrator should simply refuse to accept a nomination unless the parties change
their preferences. If the purported agreement arises after the nomination has
been accepted, the arbitrator should again counsel the parties against the
change. If they refuse, the arbitrator ought to have legitimate grounds for
resignation under any respected lex arbitri, including those requiring just cause
or the imprimatur of a supervising authority. In appropriate circumstances an
arbitrator might continue with the original mandate, arguing that the attempt to
override a mandatory procedural norm can simply be ignored as invalid.
An even more difficult situation is where the mandatory norm involves
questions of degree and the arbitrator believes that the parties agreement is
unreasonable in the circumstances. An example might be unduly tight
timeframes. This would relate to the mandatory procedural norm to give each
party an adequate opportunity to present its case. The converse timing situation
is where an arbitrator imposes strict Chess Clock timelines that the parties are
unhappy with. At times ethical conflicts might arise, for example, where an
arbitrator in an ad hoc arbitration has agreed to conduct the arbitration for a flat
fee and sets tight timeframes based on a typical daily rate. Another difficult
situation is where the parties have agreed on certain facts or a certain
methodology of determining the facts or interpreting law which the tribunal
believes to be inadequate. An example would be an agreement to have a
documents-only hearing where oral evidence seems the most pertinent. It is
important in these circumstances to distinguish on the one hand between issues
that were not placed before the tribunal and on the other hand determining how
best to resolve an issue within jurisdiction. In some cases it will be easy for the
tribunal to simply say that its mandate is to make certain determinations and it
will do so in the optimal manner regardless of defects in the submissions of the
parties. This can also be complicated by the question as to whether there is a
duty to know the law. The argument presented in this book is that the dominant
view is correct that there is no such duty but at most a power, with that power
being clearly subject to the right of the parties to be heard.

Another category of cases is where there is a potential conflict between party

autonomy and the duty to render an enforceable award. (207) This might arise if
the parties' agreement could be seen to be a violation of some norm of
international page "121" public policy or would otherwise offer grounds for
blocking enforcement under the New York Convention. This is unlikely to be a
sufficient independent ground to ignore party autonomy. In any respected
enforcement country, international public policy and the approach to Article V
generally would mirror key due process norms in any event. If they are
separately of concern, then enforceability need not be the determining factor.
Thus again the duty to render an enforceable award would not require an
arbitrator to do anything other than he or she is already obliged to do. If the
jurisdiction has a particularly unique way of viewing international public policy
in the context of procedural matters, the duty to render an enforceable award
does not require it to be enforceable in all jurisdictions. A party who entered
into business transactions with a national of that country could well be said to
have taken the risk. The majority of arbitral awards are voluntarily honoured in
any event.

More contentious again is the degree of control that an arbitrator chooses to

take, often in conflict with the preferences espoused by counsel. This raises the
whole question as to how proactive a tribunal should be or how deferential to
party preferences it should be. This was briefly discussed above and is dealt
with more comprehensively in section 6.2 dealing with proactivity generally.

2.10.4. The Right to Determine Issues Not Raised by the Parties

Section 2.7.19 above raised the question whether there is a duty to know the
law and whether arbitrators are required to apply laws not raised by the parties
under the principle iura novit curia. It was suggested in that section that even
where a tribunal has an in principle power to do so, it would be rare indeed for a
tribunal to be able to justify failure to give the parties the right to address those
issues and perhaps even determine whether they wish to revoke the mandate of
the tribunal, subject to their entitlement to do so.

This is discussed further in section 13.19. At that stage, policy questions are
considered, including the need to balance any rights with duties of impartiality,
giving persons full opportunities to present their case, treating them equally and
not exceeding the mandate. Duties of enforceability are also considered but
these are not thought to be the most determinative.

2.10.5. Right to Confidentiality

Section 2.7.15 discussed the duty of confidentiality from the perspective of the
arbitrator. It is also possible to consider whether an arbitrator has a right to
confidentiality. This might include questions as to disclosure of the award and
general comments by the parties as to the way the arbitral proceedings were
conducted. To the extent that one sees arbitration as inherently confidential, one
could argue that such a right is implied into the arbitrator's agreement with the
parties. It is more likely that the essential confidentiality of arbitration emanates
from the page "122" parties' agreement to use this form of adjudication and
does not automatically inure for the benefit of the arbitrator. (208)

2.10.6. The Right to Dissent

Notwithstanding the fact that there is a general duty of confidentiality as to

deliberations, most systems accept that there is an entitlement for a minority
arbitrator to present a dissenting opinion. Nevertheless, such an opinion is not
part of the award as such. This is discussed further in section 16.7.1.

2.10.7. The Right to Act as Mediator or Conciliator in Facilitating Settlement

Section 2.7.12.1 looked at the question of whether there was a duty to promote
settlement. While there is no absolute consensus, the better view is to consider
whether there is a power to do so as there is no overriding duty in that regard.
Section 8.11 looks at directions in lex arbitri and arbitral rules and considers the
policy issues with a view to making practical suggestions about how an
arbitrator might approach such a function were it to be considered appropriate.
The question of settlement agreements as awards and whether arbitrators have
a power to refuse to render a settlement agreement as an award are considered
in sections 16.3.7 and 16.3.8.

2.11. Are There Special Rights and Obligations for Party-Appointed

Arbitrators or The Tribunal Chair?

The foregoing discussion of the duties and powers of an arbitrator apply to all
arbitrators whether sole arbitrator, party-appointed, neutral chair or
institutional or supervisory court appointee. Nevertheless, it is appropriate to
give specific attention to two particular types of arbitrators, party-appointed
arbitrators and the chair or presiding member. These issues are dealt with in
sections 6.7 and 6.6 respectively.

2.12. Reconciling Conflicting Powers, Duties and Rights

This chapter has tried to identify and analyse the various powers, rights and
duties of arbitrators insofar as they impact upon the procedural elements of an
arbitration. From the outset it was noted that the ambit of duties may impact
upon rights and page "123" vice versa. Furthermore, rights and duties may at
times be in tension. Hence there is a need to determine how to reconcile this in a
fair and efficient manner. While complex permutations of fact will need to be
considered on a case-by-case basis, relying in the main on the common sense of
experienced and able arbitrators, some general principles ought to be accepted.

While previous sections have suggested that there are a range of potentially
conflicting sources of norms, the reality is that many procedural questions in
practice are simply resolved by looking at the applicable procedural law or
party agreement and identifying an express directive. In some cases it may even
be that an entire arbitration can be conducted without ever having to concern
oneself with identifying a lex arbitri or procedural norm. While the latter might
appear to be the case for experienced arbitrators, the reality is that they must
impliedly be accepting the validity of the procedures they are employing under
some such applicable norms. Such experienced arbitrators could readily
articulate the legal basis for their behaviour if forced to do so.

In many other cases, the rules might be ambiguous or even silent as to an

important issue. At other times arbitrators are given a broad discretion but no
specific guidance as to how that discretion is to apply to a particular procedural
question about which the parties might be in dispute. At other times there may
be a debate as to how a broad obligation, such as the duty to treat parties
equally, operates at specific stages of the process. There might be a potential
clash between various rights and obligations. It has also been noted that the
obligations on a tribunal can also be determined indirectly by looking at the
challenge powers as to tribunal determinations. At the very least, arbitrators
need to be prepared to consider how to deal with gaps in the procedural norms
or conflicts between various sources. An example of a gap is the lack of express
power of an arbitrator over the conduct of counsel. The example is used at this
stage merely to point out how a methodology of resolution might be identified.
Areas where there may be potential clashes in relation to this gap include a right
to counsel of choice and a right to an independent and impartial tribunal. A
problem may arise where a party seeks to select counsel and an arbitrator in a
conflicted relationship. In circumstances such as this, an answer to a procedural
question cannot be obtained simply by looking at the procedural rules. It is
necessary to consider the rights, duties and powers of arbitrators more
generally in the hope that a fair, efficient and consistent solution will be
identifiable.
The next principle is that certain notions of due process and fairness are
fundamental to arbitration and must always be promoted. These include the
right to equal or at least fair treatment and an appropriate opportunity to
present a case and the right to have a tribunal that is unbiased or does not lead
to reasonable apprehensions of bias at any stage through the process. Lew,
Mistelis and Kröll in fact describe these norms as the Magna Carta of arbitration.
(209) Over time, however, it has become better understood that fairness concerns

without efficiency concerns page "124" can lead to injustices in substance.

Excessive costs and delays in arbitration may go hand in hand with a correct and
thorough decision on the merits, but will not meet the legitimate expectations of
the parties and will not necessarily support the viability and respect for
arbitration as a process. Hence, it has become more common for arbitral rules
and statutes to also express the arbitrator's obligation to meet norms of
efficiency and timeliness without detracting from the need to uphold due
process standards. The more significant and challenging observation is that all
dispute resolution systems must involve some compromise between seeking
truth and minimising expense and delay in their processes. The aim of each
system is to find a reasonable balance. That should be the role of an arbitrator
when applying procedural norms and when utilising procedural discretions.
Another way to describe the need to reconcile competing rights and obligations
is to see the arbitrator's task as requiring ‘careful application of discretion and a
deep understanding of the full spectrum of competing interests and diverse legal
cultures’. (210)

The added complication where an arbitrator is concerned is the issue of party

autonomy and consent. Consent is both a jurisdictional basis for the arbitrator's
powers and is also expressly determined to be the primary way of identifying
applicable procedures, except where mandatory procedural norms are involved.
Even these are consent based either via selection of a lex arbitri containing such
norms or mere selection of arbitration involving these inherent elements.
Fairness in the arbitral context must always consider fairness as perceived by
the parties through their consent. However, the arbitrator is the protector of
minimum standards, hence parties cannot be allowed to agree to unjust
processes. Such agreements generally arise when one party has undue power
over the other or when both parties seek to abuse an arbitral process for an
unmeritorious purpose.

One of the most troubling situations is where the parties have agreed to a
particular procedure but the arbitrator is strongly convinced that the agreement
is suboptimal. Examples might include unduly tight timeframes which prevent a
sufficiently thorough analysis of the evidence. A similar situation is where
parties might agree on a documents-only hearing where the arbitrator believes
the only cogent evidence in the face of wholly ambiguous and inadequate
documentation would be that of the parties themselves. No clear-cut answers
can be given as to the way to proceed in such circumstances as there will be
many variables that must be taken into account. How strong is the evidence of
consent and how clearly advised were the parties of the implications of their
choices? How much money is involved when compared to the likely cost of the
arbitral processes? How important was expediency of outcome given their
financial status? Will any innocent third parties be adversely affected by a hasty
determination? Are there enforcement concerns that justify a speedy
resolution? At the very least, an arbitrator alert to the principles and tensions
and concerned to best reconcile due process, efficiency and consent ought to be
better able to identify the most reasonable response on a case-by-case page
"125" basis. The arbitrator should be concerned to raise these issues at the
earliest possible opportunity and give each party an adequate opportunity to
respond. An arbitrator's educative function and the respect in which they are
held will often lead to solutions by agreement or at least agreement to be bound
by the tribunal's determination.

This will only leave a smaller number of extreme cases where the parties try and
bind the tribunal to a solution considered to be inappropriate. If that occurs at
the outset, the nomination should not be accepted without an agreement by the
parties to vary the procedural agreement. If the parties seek to impose such a
position on an arbitrator after commencement of the process, then in any
circumstance where a tribunal may legitimately seek to ignore the parties'
directions, the tribunal ought to be entitled to resign without penalty. In an even
smaller number of cases the tribunal might be permitted to complete the
mandate, ignoring the parties' directions if they clearly offend against a
mandatory norm of procedure and perhaps even if they could otherwise be seen
to be inconsistent with a good faith original consent to arbitration and good
faith grant of tribunal discretion.

The balance of the book will attempt to identify some of the more troubling
scenarios while further analysing the particular steps in the arbitral process.

page "126"

1 The International Law Association made this tripartite distinction in the

context of discussing the content of applicable law, but the observation can
apply more broadly. See International Law Association (International
Arbitration Committee), ‘International Law Association Recommendations on
Ascertaining the Contents of the Applicable Law in International Commercial
Arbitration’, <www.ila-hq.org/en/committees/index.cfm/cid/19> (2008), 8.
2 Pierre Lalive, ‘The Role of Arbitrators as Settlement Facilitators: A Swiss View’,

in New Horizons in International Commercial Arbitration and Beyond, ICCA

Congress Series No. 12, ed. Albert Jan van den Berg (The Hague: Kluwer Law
International, 2005), 561.
3 For example, Born discusses rights and duties of international arbitrators

(Chapter 12) separately from matters of jurisdiction and competence which are
considered under the topic of international arbitration agreements and
competence-competence (Chapter 6), separate also to the tribunal's general
powers over procedure (Chapter 14). See Gary B. Born, International
Commercial Arbitration (The Hague: Kluwer Law International, 2009).
4 Redfern and Hunter deal with them all in a single chapter. (Chapter 5 ‘Powers,
Duties, and Jurisdiction of an Arbitral Tribunal’). See Nigel Blackaby et al.,
Redfern and Hunter on International Arbitration, 5th edn (Oxford: Oxford
University Press, 2009), 313–362.
5 The insight that the proper understanding of legal relationships can be aided

through considering the rights, duties, privileges and powers of one person
alongside the corresponding elements as held by others arose through the work
of Hohfeld. See W.N. Hohfeld, ‘Some Fundamental Legal Conceptions as Applied
in Judicial Reasoning’, Yale Law Journal 23, no. 1 (1913): 16–59; W.N. Hohfeld,
‘Fundamental Legal Conceptions as Applied in Judicial Reasoning’, Yale Law
Journal 26, no. 8 (1917): 710–770.
6 The notion of status was considered sufficiently important for the ICC
International Court of Arbitration to devote an entire special supplement to its
Bulletin on the issue. ICC Commission on International Arbitration Working
Party on the Status of the Arbitrator, ‘Final Report on the Status of the
Arbitrator’, ICC International Court of Arbitration Bulletin 7, no. 1 (1996): 29.
7 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1595.

8 See, e.g., UNCITRAL Model Law Art. 18; Swiss Private International Law Art.

182(1); English Arbitration Act 1996 ss 4(3) & 34(1); French New Code of Civil
Procedure Art. 1510. Section 1B of the English Arbitration Act 1996 also states
as a general principle that ‘a party should be free to agree how the disputes are
resolved, subject only to such safeguards as are necessary in the public interest’.
9 English Arbitration Act 1996 ss 1(a) & 33.
10 Pierre Mayer, ‘Reflections on the International Arbitrator's Duty to Apply the

Law: The 2000 Freshfields Lecture’, Arbitration International 17, no. 3 (2001):
241.
11 See, e.g., Companie Europeénne de Céréales SA v. Tradax Export S [1986] 2

Lloyd's Rep 301, 306; Cohen & Ors v.Baram [1944] 2 Lloyd's Rep 138; K/S
Norjarl A/S v. Hyundai Heavy Industries Co. Ltd [1991] 1 Lloyd's Rep 524, 536.
These cases also held that the arbitrator becomes a party to the arbitration
agreement itself, but this is not seen as the better view. This is discussed further
in section 2.3.5. Other English cases have referred to derivation of an
arbitrator's powers as arising from statute and the consent of the parties. See,
e.g., Bremer Vulkan Schiffbau und Maschinenfabrik v. South India Shipping Corp.
Ltd [1981] AC 909, 921 (HL).
12 Abdel Hamid El-Ahdab, Arbitration with the Arab Countries, 2nd edn (The

Hague: Kluwer Law International, 1999), 348–349, 430, 457, 520 and 755. For a
comparative discussion of the contractual basis of arbitral rights see Susan D.
Franck, ‘The Liability of International Arbitrators: A Comparative Analysis and
Proposal for Qualified Immunity’, New York Law School Journal of International
and Comparative Law 20, no. 1 (2000): 1. See also Klaus Lionnet, ‘The
Arbitrator's Contract’, Arbitration International 15, no. 2 (1999): 161–170.
13 Michael J. Mustill & Stewart C. Boyd, Commercial Arbitration: 2001 Companion

to the Second Edition (London: Butterworths, 2001), 61.

14 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1605.

15 Michael J. Mustill & Stewart C. Boyd, Commercial Arbitration: 2001 Companion

to the Second Edition (London: Butterworths, 2001), 57–61.

16 ICC Commission on International Arbitration Working Party on the Status of
the Arbitrator, ‘Final Report on the Status of the Arbitrator’, ICC International
Court of Arbitration Bulletin 7, no. 1 (1996): 29.
17 Pierre Karrer, ‘Responsibility of Arbitrators and Arbitral Institutions - Let

Sleeping Dogs Lie', in The Leading Arbitrators' Guide to International Arbitration,

2nd edn, ed. Lawrence W. Newman & Richard D. Hill (New York: Juris
Publishing, Inc., 2008), 614.
18 See, e.g., French New Code of Civil Procedure Art. 1456; Austrian Code of Civil

Procedure Art. 579.

19 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 277.
20 Emilia Onyema, International Commercial Arbitration and the Arbitrator's

Contract (London and New York: Routledge and Taylor and Francis, 2010), 115–
117.
21 Murray L. Smith, ‘Contractual Obligations Owed by and to Arbitrators: Model

Terms of Appointment’, Arbitration International 8, no. 1 (1992): 34.

22 Ibid., 24–25.
23 Questions of liability and immunity are discussed in section 5.18.
24 Murray L. Smith, ‘Contractual Obligations Owed by and to Arbitrators: Model

Terms of Appointment’, Arbitration International 8, no. 1 (1992): 17.

25 Philippe Fouchard, ‘Relationships between the Arbitrator and the Parties and

the Arbitral Institution’, in The Status of the Arbitrator, ICC International Court of
Arbitration Bulletin: 1995 Special Supplement, ed. Jean-François Bourque (Paris:
ICC Publishing, 1995), 20.
26 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 368.

27 Philippe Fouchard, ‘Relationships between the Arbitrator and the Parties and

the Arbitral Institution’, in The Status of the Arbitrator, ICC International Court of
Arbitration Bulletin: 1995 Special Supplement, ed. Jean-François Bourque (Paris:
ICC Publishing, 1995), 17.
28 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 369.

29 Gary B. Born, International Commercial Arbitration (The Hague, Kluwer Law

International, 2009), 1608.

30 Ibid., 1608.
31 In Norjarl, one of the judgments in the English Court of Appeal considered

that the arbitrator's rights and duties flow from a conjunction of contractual and
status considerations. The natural implication of accepting that both apply
would be that this would influence implied terms and the way in which public
policy would ameliorate contractual principles. K/S Norjarl AS/S v. Hyundai
Heavy Industries Co. Ltd [1991] 1 Lloyd's Rep 524, 536 (CA). See also Murray L.
Smith, ‘Contractual Obligations Owed by and to Arbitrators: Model Terms of
Appointment’ Arbitration International 8, no. 1 (1992): 21.
32 This is discussed further in section 5.17.7.
33 Philippe Fouchard, ‘Relationships between the Arbitrator and the Parties and

the Arbitral Institution’, in The Status of the Arbitrator, ICC International Court of
Arbitration Bulletin: 1995 Special Supplement, ed. Jean-François Bourque (Paris:
ICC Publishing, 1995), 14.
34 Murray L. Smith, ‘Contractual Obligations Owed by and to Arbitrators: Model
Terms of Appointment’, Arbitration International 8, no. 1 (1992): 33. There may
also be ethical issues that could apply regardless of the express terms. An
example might include excessive cancellation fees demanded by an arbitrator. It
may even be the case that the contract law applicable to the agreement between
the arbitrator and the parties would find such a provision offensive. Civilian
approaches might reduce the amount. The common law might consider it to be
an unenforceable penalty provision. Pressurising the parties to agree to such a
provision during the currency of a hearing might be argued to be duress. Under
civilian principles, such behaviour might be contrary to norms of good faith.
35 In some circumstances there may be a clash between a law as selected and a

mandatory norm that should apply in any event. Gary B. Born, International
Commercial Arbitration (The Hague: Kluwer Law International, 2009), 1670.
36 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 279.
37 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 369; Klaus Lionnet, ‘The
Arbitrator's Contract', Arbitration International 15, no. 2 (1999): 168; Klaus
Peter Berger, International Economic Arbitration (Deventer: Kluwer Law and
Taxation Publishers, 1993), 233.
38 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 278.
39 Philippe Fouchard, ‘Relationships between the Arbitrator and the Parties and

the Arbitral Institution', in The Status of the Arbitrator, ICC International Court of
Arbitration Bulletin: 1995 Special Supplement, ed. Jean-François Bourque (Paris:
ICC Publishing, 1995), 14.
40 Luxor (Eastbourne) Ltd v. Cooper [1941] AC 108.
41 Murray L. Smith, ‘Contractual Obligations Owed by and to Arbitrators: Model

Terms of Appointment’, Arbitration International 8, no. 1 (1992): 21.

42 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 370. Poudret and
Besson also challenge the assertion of some authors that EU Council regulation
of 22 December 2000 does not determine jurisdiction over disputes between
the parties and an arbitrator because of the exclusion of arbitration from the
Convention.
43 Dutch Code of Civil Procedure Art. 1029.
44 The list is in part inspired by a draft contract suggested in Murray L. Smith,

‘Contractual Obligations Owed by and to Arbitrators: Model Terms of

Appointment’, Arbitration International 8, no. 1 (1992): 21.
1. Establishment of the fees or fee formula. Questions of fees should be raised
with both parties at the same time, to ensure utmost propriety. page "62"
2. An indication as to pre-payment or staged payment of fees.
3. To the extent that fees are on a daily basis, an indication of how that is
defined.
4. Agreement that each party is jointly and severally liable for fees and
disbursements.
5. The nature of the disbursements to be paid by the parties. An indication
 whether the disbursements are to be on an itemised basis or based on a per
diem agreed figure.
6. An indication of who is obliged to pay any value added or similar taxes.
7. Entitlement to cancellation or commitment fees. Cancellation fees should
indicate the percentage and the time between cancellation and the hearing
date which triggers entitlement, (notification sufficiently in advance should
not entitle an arbitrator to cancellation fees as there is ample time to get
other work). A cancellation fee clause might also need to consider what
happens if the arbitration continues after the fee has been paid, for example
if the hearing was cancelled based on a presumed settlement which then
falls through. An indication should be given whether all or part of the
cancellation fee needs to be returned.
8. Entitlement to withhold any functions when fees are not paid.
9. Indicate that due diligence obligations are to be subject to other reasonable
personal and professional commitments.
10. Outline the circumstances where resignation is legitimate (e.g. acceptance
of a judicial appointment).
11. Indicate if anyone is to insure against death, incapacity or other inability to
continue.
12. Exemptions of liability and general immunity.
13. Interest on outstanding moneys.
14. Applicable law and jurisdiction, whether a court's or whether this
agreement will have its own arbitration clause.

45 Cie Européenne de Céréales SA v. Tradax Export SA [1986] Lloyd's Rep 301. It

is also worth noting that English courts took the view that the arbitrator was
party to the arbitration agreement when considering questions of a supervisory
jurisdiction. Tradax Export considered whether an injunction could be sought
under contract law against an arbitrator acting outside the terms of the
arbitration contract. The court considered that by becoming party to the
arbitration agreement, an injunction could be imposed, for example, if the
arbitrator sought to hold hearings in a place other than that specified in the
arbitration agreement. While it seems intellectually strained to suggest that the
arbitrator becomes a party to the original arbitration agreement, it at least
obviates a problem as to which supervisory court would be invited to grant such
an injunction. If the contract is seen as separate and if a view is taken that an
injunction could be successfully sought to compel certain arbitral behaviour, the
difficulty would be in deciding what is the proper court that would have
jurisdiction in relation to this distinct contract. The contract could have a
different law to that of the arbitration agreement and the jurisdiction of courts
would vary between normal attitudes to jurisdiction in contract disputes
between civilian and common law countries. It would be wholly inappropriate
to have a different court seek to guide the arbitrator's behaviour via contract
considerations than the supervisory court under arbitral theory and practice.
46 K/S Norjarl AS/S v. Hyundai Heavy Industries Co. Ltd [1991] 1 Lloyd's Rep 524,

535.
47 Susan D. Franck, ‘The Liability of International Arbitrators: A Comparative
Analysis and Proposal for Qualified Immunity’, New York Law School Journal of
International and Comparative Law 20, no. 1 (2000): 6.
48 Judgment of 19 December 1996, Société Qualiconsult v. Groupe Lincoln, Cour

d' Appel de Paris, Revue de l'Arbitrage (1998), no. 1: 121.

49 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1606. Michael Pryles, ‘Limits to Party Autonomy in Arbitral

Procedure’, Journal of International Arbitration 24, no. 3 (2007): 333. The
suggestion that the arbitrator becomes a party to the arbitration agreement as
held in Tradax is also criticised by Jean-François Poudret & Sébastien Besson,
Comparative Law of International Arbitration, 2nd edn (London: Sweet &
Maxwell, 2007), 371.
50 See, e.g., Judgment of 28 April 1998, Oberster Gerichtschof (Supreme Court of

Austria), in Albert Jan van den Berg (ed.), Yearbook of Commercial Arbitration
Volume XXVI (The Hague: Kluwer Law International, 2001), 221.
51 See, e.g., UNCITRAL Model Law Art. 14(1); Swiss Private International Law

Art. 179(1); French New Code of Civil Procedure Art. 1458; English Arbitration
Act 1996 s. 23(3) and (4); ICDR Rules Art. 9; SCC Rules Art. 15(4). See also ICC
Rules 2012 Art. 15, which enables the Court to dismiss the arbitrator upon a
request by all parties.
52 This is discussed in Chapter 16. While the service element would cease once

the award is concluded, certain key elements should remain on foot. These
would include the entitlement to remuneration if that had not as yet been paid,
provisions as to liability and immunity and duties of confidentiality which may
relate to general discussions about the content of the arbitration or the
arbitrators' rights to publish commentary as to the content of the proceedings.
53 See Klaus Lionnet, ‘The Arbitrator's Contract’, Arbitration International 15,

no. 2 (1999): 165; Philippe Fouchard, ‘Relationships between the Arbitrator and
the Parties and the Arbitral Institution’, in The Status of the Arbitrator, ICC
International Court of Arbitration Bulletin: 1995 Special Supplement, ed. Jean-
François Bourque (Paris: ICC Publishing, 1995), 15.
54 Klaus Peter Berger, International Economic Arbitration (Deventer: Kluwer

Law and Taxation Publishers, 1993), 232.

55 Christian Hausmaninger, ‘Rights and Obligations of the Arbitrator with
Regard to the Parties and the Arbitral Institution – A Civil Law Viewpoint’, in The
Status of the Arbitrator, ICC International Court of Arbitration Bulletin: 1995
Special Supplement, ed. Jean-François Bourque (Paris: ICC Publishing, 1995), 37.
56 This is often also described as the Place of Arbitration, including in lex arbitri.

While both phrases are commonly used, place of arbitration is somewhat more
confusing given that it is also accepted that hearings may be held in places other
than the Seat, without any impact on governing law. Because the word ‘Seat’ is a
more technical term than ‘place’, it is thus less confusing in reminding
practitioners of the distinction, albeit being somewhat arcane as an English
word at least.
57 Claude Reymond, ‘Where Is an Arbitral Award Made?’, Law Quarterly Review

108, no. 1 (1992): 3.

58 If an international arbitration has subjected itself to national laws pertaining

to domestic arbitration these may contain quite distinct norms. This is discussed
in section 3.8.1.
59 Sigvard Jarvin, ‘The Sources and Limits of the Arbitrator's Powers’,
Arbitration International 2, no. 2 (1986): 149–150.
60 L. Yves Fortier, ‘The Minimum Requirements of Due Process’, in Taking

Measures against Dilatory Tactics: Arbitral Discretion in International

Commercial Arbitration: “A Few Plain Rules and a Few Strong Instincts’”, in
Improving the Efficiency of Arbitration Agreements and Awards: 40 Years of
Application of the New York Convention, ICCA Congress Series No. 9, ed. Albert
Jan van den Berg (The Hague: Kluwer Law International, 1999), 399.
61 They may vary in terms of whether they afford each party a ‘full’ or
‘reasonable’ opportunity to be heard. Sections 2.7.7 and 3.7 argue that nothing
really turns on this difference in terminology, although the converse is of course
arguable simply because of the difference in wording.
62 English Arbitration Act 1996 s. 33.
63 Pierre Mayer, ‘Comparative Analysis of Power of Arbitrators to Determine

Procedures in Civil and Common Law Systems’, in Planning Efficient Arbitration

Proceedings/The Law Applicable in International Arbitration, ICCA Congress
Series No. 7, ed. Albert Jan van den Berg (The Hague: Kluwer Law International,
1996), 26.
64 See, e.g., English Arbitration Act 1996 s. 24.
65 English Arbitration Act 1996 s. 42; Dutch Code of Civil Procedure Art. 1035.
66 See UNCITRAL Rules (2010) Art. 18; UNCITRAL Model Law, Art. 20; Swedish

Arbitration Act s. 22; ICDR Rules Art. 13.

67 Section 3.5.1 deals with the factors that ought to be considered in selecting a

Seat under such a discretion.

68 See UNCITRAL Model Law Art. 27; English Arbitration Act 1996 s. 43; French

New Code of Civil Procedure Arts 1449 and 1469; Swedish Arbitration Act s. 26;
Swiss Private International Law Art. 184(2).
69 Emilio Agustin Maffezini v. The Kingdom of Spain ICSID No. Arb/97/7 decision

of the tribunal on objections to jurisdiction, 25 January 2000,

<http://ita.law.uvic.ca/documents/Maffezini-Jurisdi...>.
70 Michael J. Mustill & Stewart C. Boyd, Commercial Arbitration: 2001 Companion

to the Second Edition (London: Butterworths, 2001), 249.

71 In one sense arbitrators have less power than judges as they depend on the

parameters of consent. However, the lack of appeals on questions of law makes

them more powerful in that sense at least.
72 Pierre Lalive, ‘Transnational (or Truly International) Public Policy and
International Arbitration’, in Comparative Arbitration Practice and Public Policy
in International Arbitration, ICCA Congress Series No. 3, ed. Pieter Sanders
(Deventer: Kluwer Law and Taxation Publishers, 1987), 271–272.
73 Christian Hausmaninger, ‘Rights and Obligations of the Arbitrator with
Regard to the Parties and the Arbitral Institution – A Civil Law Viewpoint’, in The
Status of the Arbitrator, ICC International Court of Arbitration Bulletin: 1995
Special Supplement, ed. Jean-François Bourque (Paris: ICC Publishing, 1995), 38.
74 Described in continental systems as ‘legälitatsprinzip’ or ‘le principe de

légalité’.
75 K/S Norjarl AS/S v. Hyundai Heavy Industries Co. Ltd [1992] QB 863, 885.
76 In Re an Arbitration between Crighton and Law Car and General Insurance

Corporation, Limited [1910] 2 KB 738, 745.

77 Allan Scott Rau, ‘Integrity and Private Judging’, South Texas Law Rev 38

(1997): 485.
78 Catherine Rogers, ‘The Ethics of International Arbitrators’, in The Leading
Arbitrators' Guide to International Arbitration, ed. Lawrence W. Newman &
Richard D. Hill (Huntington: Juris Pub, 2008), 630.
79 Howard M. Holtzmann & Joseph E. Neuhaus, A Guide to the UNCITRAL Model

Law on International Commercial Arbitration: Legislative History and

Commentary (The Hague: Kluwer Law and Taxation Publishers, 1989), 464–465.
80 S. Besson, ‘Arbitrage International et Mesures Provisoires’, Etudes Suisses de

droit International (Société Suisse de droit International) 105 (1998): 131.

81 Nuclear Test (Australia v. France) (Judgment) [1974] ICJ Rep 253, 259–260

citing in part Northern Cameroon (Cameroon v.United Kingdom) (Preliminary

Objections) [1963] ICJ Rep 15, 29.
82 Separate Opinion of Judge Higgins in Legality of Use of Force (Serbia and

Montenegro v. Belgium) (Preliminary Objections) [1999] ICJ Rep 279, 338–339.

83 Chester Brown, A Common Law of International Adjudication (Oxford: Oxford

University Press, 2007), 79.

84 Kenneth Carlston, The Process of International Arbitration (New York:
Columbia University Press, 1946).
85 Joost Pauwelyn, Conflict of Norms in Public International Law: How WTO Law

Relates to Other International Law (Cambridge: Cambridge University Press,

2003), 447–448; Joost Pauwelyn, ‘The Role of Public International Law in the
WTO: How Far Can We Go?’, American Journal of International Law 95 (2001):
555.
86 Gulf of Maine (Canada v. United States) (Judgment) [1984] ICJ Rep 246, 305.
87 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 326.

88 Richard Garnett et al., A Practical Guide to International Commercial
Arbitration (New York: Oceana Publications, 2000), 81–89.
89 Catherine Kessedjian, ‘Principe de la contradiction et arbitrage’, Revue de

l'Arbitrage (1995): 406–410.

90 Karl-Heinz Böckstiegel, ‘The Role of the Arbitrators in Investment Treaty

Arbitration’, in International Commercial Arbitration: Important Contemporary

Questions, ICCA Congress Series No. 11, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 2003), 371.
91 J. Gillis Wetter, ‘Procedures for Avoiding Unexpected Legal Issues’, in
Planning Efficient Arbitration Proceedings/The Law Applicable in International
Arbitration, ICCA Congress Series No. 7, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 1996), 97.
92 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1626.

93 Allan Philip, ‘The Duties of an Arbitrator’, in The Leading Arbitrators' Guide to

International Arbitration, ed. Lawrence W. Newman & Richard D. Hill (New

York: Juris Publishing, Inc., 2008), 67.
94 The Introductory Note to the IB A Rules of Ethics indicates that arbitrators

should be competent. See also Catherine Rogers, ‘The Ethics of International

Arbitrators’, in The Leading Arbitrators' Guide to International Arbitration, ed.
Lawrence W. Newman & Richard D. Hill (Huntington: Juris Pub., 2008), 646.
95 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1619.

96 P.S. Atiyah, Law and Modern Society (Oxford: Oxford Paperbacks, 1983), 42.
97 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn
(Oxford: Oxford University Press, 2009), 335.
98 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1594–1595.

99 See Dutch Code of Civil Procedure Art. 1047 exempting expedited commodity

arbitrations from some general procedural norms.

100 International Law Association (International Arbitration Committee),’
International Law Association Recommendations on Ascertaining the Contents
of the Applicable Law in International Commercial Arbitration’, <www.ila-
hq.org/en/committees/index.cfm/cid/19>(2008), 20.
101 Piero Bernardini, ‘International Arbitration and A-National Rules of Law’, ICC

International Court of Arbitration Bulletin 15, no. 2 (2004): 58, 117.

102 See UNCITRAL Rules 2010 Art. 17; LCIA Rules Art. 14; SCC Rules Art. 19(2);

Swiss Rules 2012 Art. 15.1; ICDR Rules Art. 16.1; ACICA Rules Art. 17.1. The
ICSID Arbitration Rules adopted differing structure to similar effect, giving
greater detail as to the content of such norms.
103 ICC Rules 2012 Art. 22(4).
104 Fernando Mantilla-Serrano, ‘Towards a Transnational Procedural Public

Policy’, Arbitration International 20, no. 4 (2004): 342.

105 See further section 5.3.
106 French New Code of Civil Procedure Art. 1510.
107 Article 182(3) Swiss PILA provides: ‘Regardless of the procedure chosen, the

arbitral tribunal shall guarantee equal treatment of the parties and their right to
be heard in contradictory proceedings.’
108 Dalmia Dairy Industries Ltd v. National Bank of Pakistan [1978] 2 Lloyd's Rep

223.
109 Gerold Herrmann, ‘Power of Arbitrators to Determine Procedures under the

UNCITRAL Model Laws’, in Planning Efficient Arbitration Proceedings/The Law

Applicable in International Arbitration, ICCA Congress Series No. 7, ed. Albert Jan
van den Berg (The Hague: Kluwer Law International, 1996), 45.
110 Judgment of 1 July 1999, Société Braspetro Oil Services (Brasoil) v. GMRA,

Cour d'appel de Paris, Revue de l'Arbitrage (1999), no. 4: 834.

111 Karl-Heinz Böckstiegel, ‘The Role of the Arbitrators in Investment Treaty

Arbitration’, in International Commercial Arbitration: Important Contemporary

Questions, ICCA Congress Series No. 11, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 2003), 373.
112 Goldsmith refers to a Supreme Court of the Federal Republic of Germany

case where an award was overturned based on the violation of the rights of
defence when arbitrators decided not to hear some of the witnesses produced
by the defendant. No case is actually cited. See Jean-Claude Goldsmith, ‘How to
Draft Terms of Reference’, Arbitration International 3, no. 4 (1987): 299.
113 Gerald Aksen, ‘Reflections of an International Arbitrator’, Arbitration
International 23, no. 2 (2007): 257.
114 The issue was considered to be sufficiently important for the ICC
Commission on Arbitration to publish a report entitled ‘Techniques for
Controlling Time and Costs in Arbitration’ in 2007. See section 6.2.8.4.
115 A. A. de Fina, ‘The Party Appointed Arbitrator in International Arbitrations –

Role and Selection’, Arbitration International 15, no. 4 (1999): 386, referring to
the ICC 1998 amendments.
116 Mohammed Bedjaoui, ‘The Arbitrator: One Man-Three Roles: Some
Independent Comments on the Ethical and Legal Obligations of an Arbitrator’,
Journal of International Arbitration 5, no. 1 (1988): 16.
117 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th

edn (Oxford: Oxford University Press, 2009), 337. Philip describes this as a
moral duty. See Allan Philip, ‘The Duties of an Arbitrator’, in The Leading
Arbitrators' Guide to International Arbitration, ed. Lawrence W. Newman &
Richard D. Hill (New York: Juris Publishing, Inc., 2008), 71.
118 Julian Lew, Contemporary Problems in International Arbitration (Oceana:

New York, 1978), 540.

119 For example, ICC Rules 2012 Art. 11.
120 Piero Bernardini, ‘International Arbitration and A-National Rules of Law',

ICC International Court of Arbitration Bulletin 15, no. 2 (2004): 116.

121 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1622.

122 Ibid., 1622–1623.
123 While beyond the scope of this book, a duty to act with integrity may also

impact upon the way fees are determined. While an arbitrator has a right to be
concerned for their own remuneration, bargaining or haggling would be
improper. Henry Gabriel & Anjanette H. Raymond, ‘Ethics for Commercial
Arbitrators: Basic Principles and Emerging Standards’, Wyoming Law Review 5,
no. 2 (2005): 453, 466. A duty to act in good faith and with integrity also
involves not using confidential information for personal gain. An example would
be insider share trading. Henry Gabriel & Anjanette H. Raymond, ‘Ethics for
Commercial Arbitrators: Basic Principles and Emerging Standards’, Wyoming
Law Review 5, no. 2 (2005): 453, 465.
124 J. Gillis Wetter, ‘Procedures for Avoiding Unexpected Legal Issues’, in
Planning Efficient Arbitration Proceedings/The Law Applicable in International
Arbitration, ICCA Congress Series No. 7, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 1996), 98.
125 Allan Philip, ‘The Duties of an Arbitrator’, in The Leading Arbitrators' Guide to

International Arbitration, ed. Lawrence W. Newman & Richard D. Hill (New

York: Juris Publishing, Inc., 2008), 68.
126 Ibid.
127 This is expressly articulated in Art. 42(2) ICSID Convention, but applies as a

matter of course in any adjudicatory proceedings.

128 Eric A. Schwartz, ‘The Rights and Duties of ICC Arbitrators’, in The Status of

the Arbitrator, ICC International Court of Arbitration Bulletin: 1995 Special

Supplement, ed. Jean-François Bourque (Paris: ICC Publishing, 1995), 86; Gary B.
Born, International Commercial Arbitration (The Hague: Kluwer Law
International, 2009), 1627; Constantine Partasides, ‘The Fourth Arbitrator? The
Role of Secretaries to Tribunals in International Arbitration’, Arbitration
International 18, no. 2 (2002): 147.
129 See, e.g., Dutch Code of Civil Procedure Art. 1039(3).
130 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1629.

131 This is articulated in AAA/ABA Code of Ethics, Canon 1(E). See also Alexis

Mourre, ‘Application of the Vienna International Sales Convention in

Arbitration’, ICC International Court of Arbitration Bulletin 17, no. 1 (2006): 1.
132 For example, Austrian Code of Civil Procedure Art. 204.
133 See, e.g., UNCITRAL Model Law Art. 32(a).
134 See G.R. Delaume, Transnational Contracts: Applicable Law and Settlement of

Disputes, vol. 2 (New York: Oceana, 1978), [13.11].

135 Similarly the French New Code of Civil Procedure Art. 1520(3) allows for

awards to be set aside if the tribunal made its decision without complying with
the mandate conferred on it.
136 Eco Suisse China Time Ltd v. Benetton International NV Case C-126/97, ECR

1991, 1-03055; See also Yves Brulard & Yves Quintin, ‘European Community
Law and Arbitration -National versus Community Public Policy’, Journal of
International Arbitration 18, no. 5 (2001): 533–548.
137 See ICC Rules 2012 Art. 23.
138 For example, UNCITRAL Model Law Art. 32(3) indicates that termination of

the arbitral proceedings terminates the tribunal's mandate. See also French New
Code of Civil Procedure Art. 1485; German Code of Civil Procedure Art. 1056(3);
Swedish Arbitration Act s. 27.
139 Even where an arbitrator is functus officio, this only relates to arbitral

powers. There are separate and ongoing rights and obligations in terms of
unpaid remuneration and liability for breaches of arbitral duties where
immunity provisions do not intervene.
140 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1594–1595.

141 Karl-Heinz Böckstiegel, ‘The Role of the Arbitrators in Investment Treaty

Arbitration’, in International Commercial Arbitration: Important Contemporary

Questions, ICCA Congress Series No. 11, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 2003), 371.
142 I am indebted to Julian Lew for this articulation.
143 The situation may or may not be conceptually different when an arbitrator is

empowered to act as amiable compositeur as there is some debate as to the

degree to which an arbitrator can depart from general principles of law. See
further section 13.14.
144 Günther J. Horvath, ‘The Duty of the Tribunal to Render an Enforceable

Award’, Journal of International Arbitration 18, no. 2 (2001): 135.

145 Julian Lew, Contemporary Problems in International Arbitration (Oceana:

New York, 1978), 537.

146 Yves Derains & Eric A. Schwartz, A Guide to the New ICC Rules of Arbitration

(The Hague: Kluwer Law International, 1998), 353.

147 Allan Philip, ‘The Duties of an Arbitrator’, in The Leading Arbitrators' Guide to

International Arbitration, ed. Lawrence W. Newman & Richard D. Hill (New

York: Juris Publishing, Inc., 2008), 67.
148 Karl-Heinz Böckstiegel, ‘Major Criteria for International Arbitrators in
Shaping an Efficient Procedures', in Arbitration in the Next Decade, ICC
International Court of Arbitration Bulletin: 1999 Special Supplement, ed. ICC
(Paris: ICC Publishing 1999), 50.
149 Pierre A. Karrer, ‘Must an Arbitral Tribunal Really Ensure that Its Award is

Enforceable?’, in Global Reflections on International Law, Commerce and Dispute

Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al.
(Paris: ICC Publishing, 2005), 429.
150 The UNCITRAL Model Law provisions on Recognition and Enforcement of
Foreign Arbitral Awards mirror the New York Convention provisions.
151 Sigvard Jarvin, ‘The Sources and Limits of the Arbitrator's Powers’,
Arbitration International 2, no. 2 (1986): 155.
152 Fernando Mantilla-Serrano, ‘Towards a Transnational Procedural Public

Policy’, Arbitration International 20, no. 4 (2004): 338–339.

153 Ibid., 339.
154 Sigvard Jarvin, ‘The Sources and Limits of the Arbitrator's Power’,
Arbitration International 2, no. 2 (1986): 153; Martin Platte, ‘Arbitrator's Duty to
Render Enforceable Awards’, Journal of International Arbitration 20, no. 3
(2003): 309–310; Marc Blessing, ‘Arbitrability of Intellectual Property Disputes’
Arbitration International 12, no. 2 (1996): 206.
155 An example of different views on enforceability occurred in the Egyptian

Pyramids case. Compare Judgment of 12 July 1984, Cour d'appel de Paris, in

Yearbook of Commercial Arbitration Volume X, ed. Pieter Sanders (The Hague:
Kluwer Law International, 1985), 113 with Judgment of 12 July 1984, District
Court of Amsterdam, Yearbook Commercial Arbitration Volume X, ed. Pieter
Sanders (The Hague: Kluwer Law International), 487.
156 Philippe Fouchard, ‘Relationships between the Arbitrator and the Parties

and the Arbitral Institution’, in The Status of the Arbitrator, ICC International
Court of Arbitration Bulletin: 1995 Special Supplement, ed. Jean-François
Bourque (Paris: ICC Publishing, 1995), 254.
157 For a case confirming that a tribunal must uphold jurisdiction even in

situations where the award might not be enforced see ICC Case 4695 (interim
award) reported in Sigvard Jarvin (ed.), ICC Arbitral Awards 1986–1990 (The
Hague: Kluwer Law International, 1994), 33.
158 ICC Arbitration Rules 2012 Appendix II Internal Rules of the International

Court of Arbitration of the ICA, Art. 6.

159 Grigero Naon, ‘The Powers of the ICC International Court of Arbitration Vis-

à-Vis Parties and Arbitrators in Arbitration in the Next Decade’, Special

Supplement of the ICC International Court of Arbitration Bulletin 55 (1999): 70.
160 Pierre A. Karrer, ‘Must an Arbitral Tribunal Really Ensure that Its Award Is

Enforceable?’, in Global Reflections on International Law, Commerce and Dispute

Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al.
(Paris: ICC Publishing, 2005), 433.
161 Mohammed Bedjaoui, ‘The Arbitrator: One Man-Three Roles: Some
Independent Comments on the Ethical and Legal Obligations of an Arbitrator’,
Journal of International Arbitration 5, no. 1 (1988): 19.
162 Sigvard Jarvin, ‘The Sources and Limits of the Arbitrator's Powers’,
Arbitration International 2, no. 2 (1986): 154.
163 Günther J. Horvath, ‘The Duty of the Tribunal to Render an Enforceable

Award’, Journal of International Arbitration 18, no. 2 (2001): 142.

164 If the tribunal has not breached any other express or implied procedural

obligations, the tribunal cannot be said to be negligent. In circumstances where

liability is possible, it would be hard to envisage situations where there might be
liability because of nonenforceability when there was no separate cause of
action for a direct breach of a procedural obligation. An example might be a
failure to comply with a form requirement for enforceability but only if the view
was taken that an arbitrator has a duty of inquiry, absent any advice from the
parties. It is unlikely that many courts would conclude that there was such an
obligation. In any event, most rule systems would provide for immunity for
arbitrators in such circumstances.
165 See ICDR Rules Art. 34; HKIAC Rules Art. 39; LCIA Rules Art. 30.2; SCC Rules

Art. 46; SIAC Rules Art. 21.4; WIPO Rules Art. 76. See also IBA Rules of Ethics,
rule 9.
166 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 283.
167 ICC Rules 2012 Art. 26(3); LCIA Rules Art. 19.4; ICDR Rules Arts 20(4) and

34; UNCITRAL Rules (2010) Art. 28.3; Swiss Rules 2012 Art. 25.6.
168 Catherine Rogers, ‘The Ethics of International Arbitrators’, in The Leading

Arbitrators' Guide to International Arbitration, ed. Lawrence W. Newman &

Richard D. Hill (Huntington, Juris Pub, 2008), 646–647. Express confidentiality
obligations vis-à-vis arbitrators are contained in ICDR Rules Art. 34 of and SCC
Rules Art. 46.
169 James H. Carter, ‘The Rights and Duties of the Arbitrator: Six Aspects of the

Rule of Reasonableness’, in The Status of the Arbitrator, ICC International Court

of Arbitration Bulletin: 1995 Special Supplement, ed. Jean-François Bourque
(Paris: ICC Publishing, 1995), 35.
170 WTO and NAFTA rules also allow for third-party rights.
171 This would depend on the drafting and interpretation of the statutes as the

legislature could give either priority. When in doubt, an express arbitral

immunity or confidentiality mandate would most likely be seen as lex specialis.
172 Bedjaoui sees it as a duty to promote settlement flowing from the
fundamental difference between arbitrators and judges, but that is based on a
particular notion as to the status of a judge. Mohammed Bedjaoui, ‘The
Arbitrator: One Man–Three Roles: Some Independent Comments on the Ethical
and Legal Obligations of an Arbitrator’, Journal of International Arbitration 5, no.
1 (1988): 17. Marriott suggests ‘that it is the primary duty of arbitrators to
encourage parties to settle’. See Arthur Marriott, ‘Arbitrators and Settlement’, in
New Horizons in International Commercial Arbitration and Beyond, ICCA
Congress Series No. 12, ed. Albert Jan van den Berg (The Hague: Kluwer Law
International, 2005), 533. In his view, this supports both the public interest to
promote dispute settlement and the private interests of the parties concerned.
173 Pierre Karrer, ‘Responsibility of Arbitrators and Arbitral Institutions – Let

Sleeping Dogs Lie’, in The Leading Arbitrators' Guide to International Arbitration,

2nd edn, ed. Lawrence W. Newman & Richard D. Hill (New York: Juris
Publishing, Inc., 2008), 615.
174 English cases supporting a professional duty of care include the Myron

[1970] 1 QB 527; Succula Ltd v. Harland & Wolff [1980] 2 Lloyd's Rep 281;
Palacath Ltd v.Flanigan [1985] 2 All ER 161.
175 Allan Philip, ‘The Duties of an Arbitrator’, in The Leading Arbitrators' Guide to

International Arbitration, ed. Lawrence W. Newman & Richard D. Hill (New

York: Juris Publishing, Inc., 2008), 79. The more significant debate in relation to
a supposed duty of care is what flows from an alleged breach and hence what, if
any, repercussions there should be for an arbitrator. These issues are touched
on in section 2.3 dealing with the contractual relationship between the
arbitrator and the parties and section 5.18 dealing with liability and immunity of
arbitrators.
176 A number of authors describe it as jura novit curia.
177 Michael J. Mustill & Steward C. Boyd, The Law and Practice of Commercial
Arbitration in England (London: Butterwoths, 1989), 299.
178 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1626.

179 Eric A. Schwartz, ‘Do International Arbitrators Have a Duty to Obey the

Orders of Courts at the Place of the Arbitration? Reflections on the role of the
Lex Loci Arbitri in the Light of a Recent ICC Award’, in Global Reflections on
International Law, Commerce and Dispute Resolution: Liber Amicorum in Honour
of Robert Briner, ed. Gerald Aksen et al. (Paris: ICC Publishing, 2005), 795–797.
180 See, e.g., UNCITRAL Model Law Art. 13(3).
181 For example, Himpurna California Energy Ltd v. PT (Persero) Perusahaan

Listruik Negara (Republic of Indonesia) in Albert Jan van den Berg (ed.),
Yearbook of Commercial Arbitration Volume XXV (The Hague: Kluwer Law
International, 2000), 11.
182 J. Martin Hunter & Jan Paulsson, ‘A Code of Ethics for Arbitrators in
International Commercial Arbitration?’, International Business Lawyer 13
(1985): 158.
183 Catherine A. Rogers, Tit and Function in Legal Ethics: Developing a Code of

Conduct for International Arbitration’, Michigan Journal of International Law 23

(2002): 321 and Catherine A. Rogers, ‘Regulating International Arbitrators: A
Functional Approach to Developing Standards of Conduct’, Stanford Journal of
International Law 41 (2005): 53 contain the most sophisticated treatments of
this issue to date.
184 Catherine Rogers, ‘The Ethics of International Arbitrators’, in The Leading

Arbitrators' Guide to International Arbitration, ed. Lawrence W. Newman &

Richard D. Hill (Huntington: Juris Pub, 2008), 648.
185 The AAA/AB A, Code of Ethics for Arbitrators in Commercial Disputes was

first developed in 1977 and then revised in 2004.

186 See, e.g., California Rules of Court, Ethics Standards for Neutral Arbitrators in

Contractual Arbitration (2007).

187 For conflicting American cases on this issue see Catherine Rogers, ‘The

Ethics of International Arbitrators’, in The Leading Arbitrators' Guide to

International Arbitration, ed. Lawrence W. Newman & Richard D. Hill
(Huntington: Juris Pub, 2008), 626, n. 15.
188 Singapore International Arbitration Centre, ‘Code of Ethics for an Arbitrator’,

<www.siac.org.sg/images/stories/documents/code-of-p...>, 2009.
189 Eric A. Schwartz, ‘The Rights and Duties of ICC Arbitrators’, in The Status of

the Arbitrator, ICC International Court of Arbitration Bulletin: 1995 Special

Supplement, ed. Jean-François Bourque (Paris: ICC Publishing, 1995), 67.
190 An example that does is the Code of Professional Ethics of the Italian Bar

Association which requires arbitrators to be independent, preserve the trust of

the parties and make appropriate disclosure. Art. 55.
191 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1601.

192 Catherine Rogers, ‘The Ethics of International Arbitrators’, in The Leading

Arbitrators' Guide to International Arbitration, ed. Lawrence W. Newman &

Richard D. Hill (Huntington: Juris Pub, 2008), 622.
193 Chester Brown, A Common Law of International Adjudication (Oxford: Oxford
University Press, 2007), 63; Interpretation of Greco-Turkish Agreement [1928]
PCIJ (ser B) No. 16, 20.
194 Mavrommatis Palestine Concessions (Greece v. United Kingdom) (Judgment

No. 2) [1924] PCIJ (ser A) No. 2, 16.

195 See UNCITRAL Rules 2010 Art. 10.1(b); ICSID Rules Art. 45(1).
196 See generally J. Waincymer, ‘Reconciling Conflicting Rights in International

Arbitration: The Right to Choice of Counsel and the Right to an Independent and
Impartial Tribunal’, Arbitration International 26 no. 4 (2010): 597.
197 If tribunals are empowered to act in this way, then subject to any contrary

applicable law they are also under a duty to do so. See Nigel Blackaby et al.,
Redfern and Hunter on International Arbitration, 5th edn (Oxford: Oxford
University Press, 2009), 327.
198 See ICC Rules 2012 Art. 28; UNCITRAL Rules 2010 Art. 26; ICDR Rules Art.

21; HKIAC Rules Art. 24; SCC Rules Art. 32; SIAC Rules Art. 26; Swiss Rules 2012
Art. 26; ACICA Rules Art. 28; French New Code of Civil Procedure Art. 1468;
UNCITRAL Model Law Art. 17 & 17E; Swedish Arbitration Act s. 25; English
Arbitration Act 1996 s. 38(3). The English Act gives a consequential power to
dismiss the claim if the award for security is not complied with: s. 41(6).
199 See UNCITRAL Model Law Chapter IVA; UNCITRAL Rules 2010 Art. 26.
200 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 468.

201 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1595.

202 Emmanuel Gaillard & John Savage (eds), Fouchard Gaillard Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 627.
203 Philippe Fouchard, ‘Relationships between the Arbitrator and the Parties

and the Arbitral Institution’, in Jean-François Bourque (ed.) The Status of the
Arbitrator, ICC International Court of Arbitration Bulletin: 1995 Special
Supplement, (Paris: ICC Publishing, 1995), 20.
204 Christian Hausmaninger, ‘Rights and Obligations of the Arbitrator with

Regard to the Parties and the Arbitral Institution – A Civil Law Viewpoint’, in
Jean-François Bourque (ed.) The Status of the Arbitrator, ICC International Court
of Arbitration Bulletin: 1995 Special Supplement, (Paris: ICC Publishing, 1995),
46.
205 Sigvard Jarvin, ‘The Sources and Limits of the Arbitrator's Powers’,
Arbitration International 2, no. 2 (1986): 152.
206 See also English Arbitration Act 1996 s. 40.
207 Catherine Rogers, ‘The Ethics of International Arbitrators’, in The Leading

Arbitrators' Guide to International Arbitration, ed. Lawrence W. Newman and &

Richard D. Hill (Huntington, Juris Pub, 2008), 645.
208 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1663.

209 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 95.
210 Ruth M. Teitelbaum, ‘Book Review: Newman and Hill (eds), The Leading

Arbitrator's Guide to International Arbitration’, Arbitration International 25, no.

1 (2009): 143.
Part II: The Process of an Arbitration,
Chapter 3: The Procedural
Framework for International
Arbitration
Jeff Waincymer,

3.1. Introduction

page "127" page "128"

Part I looked at the principles and policy issues underlying the process of
international arbitration. This Part separately analyses the various stages of the
arbitral process, in part, informed by those general observations. This chapter
looks at the procedural framework for international arbitration. It considers
issues of consent through an analysis of the arbitration agreement and the laws
and rules applicable to the agreement.

Because much of the procedural framework is established or circumscribed by

party autonomy and hence by the arbitration agreement, attention is given to
issues of drafting and interpretation both generally and in relation to some
typical specialty clauses. In many instances, interpretation of the agreement to
arbitrate will be necessary to determine what was agreed or intended. This may
arise where consent to certain procedural aspects is disputed or when it is
alleged that there is no valid and binding agreement at all. In the extreme,
inadequate drafting of an arbitration agreement may lead to the whole process
being inoperative or ineffective. Because interpretation can be guided by
applicable law, the chapter considers what law applies to arbitration agreements
to determine their validity and to interpret their terms where valid. As a related
exercise it considers the various general principles by which arbitration
agreements are interpreted and the evidence that may be utilised in aid of that
exercise.

The chapter then considers the role of the Seat, the lex arbitri, mandatory
procedural laws and arbitral rules if selected. Consideration is also given to other
sources of procedural law. Finally, the chapter looks at the role of institutions
and the advantages and disadvantages of different models of arbitration, both
institutional and ad hoc.

page "129"

3.2. The Arbitration Agreement

Arbitration by its essential nature is based on an agreement between the parties
to submit their dispute to binding and final adjudication by an identifiable
tribunal. One way or another, one must be able to identify an arbitration
agreement to that effect. (1) That need not take the form of a contract and does
not require contemporaneous offer and acceptance. Nevertheless, the norm is to
include an arbitration clause in commercial contracts where that is the preferred
dispute resolution model. A separate agreement would also typically constitute a
contract. The situation is generally different where investment arbitration is
concerned. In that field there is typically a two-step agreement. A host State first
provides a standing offer to arbitrate with future investors under a bilateral or
regional investment treaty. At the stage the offer is made, the identity of the
investors or relevant investments is usually not known. The ambit of the offer is
instead simply outlined via definitions as to coverage of ‘investors’ and
‘investments’ in the treaty. A person coming within the definition of an investor
may accept the offer to arbitrate with respect to investments as defined, simply
by commencing arbitration within the terms specified. In some cases there can
be an overlap or conflict between such an agreement and an express investment
contract that contains its own arbitration clause. (2)

In addition to establishing the required consent, an arbitration agreement must

provide for certain key elements for the process to be viable. These elements
might be expressed in the agreement itself or incorporated or empowered
through choices otherwise made. Where the agreement stipulates the Seat or
place of arbitration, it in turn invariably leads to the arbitration law of that Seat
forming the lex arbitri. The agreement may also select particular arbitral rules
and specify a range of other powers, rights and duties of various parties and the
tribunal. As noted, because it forms an agreement between the parties, it will
typically be contractual in nature under some substantive law of contract. At the
same time, it creates the jurisdictional basis for the arbitration. Hence an
arbitration agreement has a dual character. This raises challenges when it comes
to interpreting the agreement where there is a need to identify the applicable
law. Should it be a procedural or a substantive law that is utilised or should more
general international principles apply? Because of the dual nature, there can be
no unassailable conclusion on this question, which makes tribunal discretion
more problematic in the absence of direction from the parties. This is discussed
further in section 3.2.2. There is also a need to consider which elements of party
agreement are acceptable and how party autonomy interrelates with legal
provisions framing the arbitration. Mandatory procedural laws of the Seat will
take priority. More contentious is whether other purportedly page "130"
mandatory norms must or may apply. These questions are considered in sections
3.7.2 and 13.8.

Historically, arbitration agreements fell into two broad groups, those dealing
with a defined category of future disputes and those dealing with an existing
dispute. The latter were also described as compromis. These were necessary in
those jurisdictions that refused to enforce agreements in relation to future
disputes, on the basis of their being thought to be too uncertain. (3) Each type of
agreement is still a consent-based express agreement to arbitrate. Thus each
form of agreement may offer powers and authority to arbitrators. Each may also
directly or indirectly provide restrictions on those powers, identify particular
duties (4) or indicate the degree to which discretions and powers may be
circumscribed. While there are many similarities, the main difficulty with
submission agreements in terms of setting a procedural framework is that at the
time a dispute is known, positions are likely to be hardened and it is much less
likely that the parties will agree to fill any gaps in the procedural framework in a
neutral and efficient manner. Hence, particular care needs to be taken to ensure
that there are no gaps or that there is a mechanism that will adequately fill any
such gaps if found.

3.2.1. Autonomy and Separability of the Arbitration Agreement and

Competence of the Tribunal

While the existence of an arbitration agreement is vital, arbitration theory and

practice still had to establish additional distinct notions for it to be a viable
dispute resolution mechanism. One of the most important is the doctrine of
autonomy or separability of the arbitration agreement. (5) While a submission
agreement over an existing dispute is indeed separate, the more common
situation is a dispute resolution clause embedded within a commercial contract.
In that circumstance, there is a natural problem for a tribunal if there is a dispute
as to whether the contract has in fact been concluded or whether the contract is
void or voidable, perhaps because of some misrepresentation or fraud. While
parties in dispute are entitled to have those allegations determined, if the very
jurisdictional source of the arbitrator's power is a clause in the contract itself,
there is a problem in presuming jurisdiction to decide validity of the very
contract from which jurisdiction purports to emanate. The problem becomes a
circular one, with no obvious solution absent some guidance in the applicable
law governing the arbitration or from some broadly accepted general power. The
solution adopted in lex arbitri, arbitral rules and general theory page "131" is to
treat the arbitration agreement as autonomous and separate. Hence the
agreement to arbitrate survives and is untainted by any illegality or by any
alleged void or voidable nature in the balance of the contract. (6) While the
doctrine of separability indicates that the agreement is not per se tainted by
fraud or illegality as found under the balance of the contract, it is possible that
there may be such an allegation in relation to the clause itself. In such
circumstances a determination may be made that the agreement, as well as the
balance of the contract, is invalid.

The doctrine of separability goes hand-in-hand with the doctrine of competence-

competence, (7) which allows a tribunal to decide on its own jurisdiction,
including in circumstances where the contract may be defective. This goes so far
as to empower a tribunal to conclude that there is in fact no jurisdiction at all,
which would otherwise seem illogical. (8) Separability and competence together
also provide for a res judicata effect of a negative jurisdictional determination,
subject to any express appeal or challenge rights in the lex arbitri. Fouchard,
Gaillard, Goldman make the observation that autonomy extends the tribunal's
power beyond competence alone, as it allows for valid arbitral jurisdiction under
the separable clause in circumstances where the balance of the contract is
inoperative. (9)
Separability is enshrined in all leading arbitral laws and institutional or ad hoc
rules. (10) National courts have also identified separability as a general principle
even absent any statutory foundation. (11) This has led scholars to consider that it
is an element of transnational arbitral law or the lex mercatoria of arbitration. (12)
Where the latter view is concerned, one issue is how one might justify
separability under a pure consent paradigm. It would presumably be a form of
implied intent logic; anyone who puts an arbitration clause into a document
wants it to be effective and knows that without separability, there can be no
adequate resolution of disputes about the status of the underlying contract. The
consent-based argument is enhanced if the drafting of the agreement expressly
indicates that disputes as to page "132" the validity or existence of the contract
are all matters for arbitration, but the converse should not hold.

Some authors distinguish situations where the issue is whether there is a valid
contract in the first place, considering such circumstances to be more
problematic in terms of a tribunal's competence. The paradigm example, where
it is difficult to imply consent to arbitral jurisdiction, is where a respondent
alleges that the contract is a forgery and it never entered into the transaction,
including the arbitration clause itself. This is different to a situation where a
party argues that the contract was not consummated because a purported
acceptance in fact constituted a revocation and counter offer. In the latter case
one can still use the implied consent logic that if the parties wanted arbitration
over a consummated contract, they have impliedly agreed to a tribunal deciding
whether it was in fact consummated. The offer and acceptance cases are more
problematic if the counter offer relates to a different dispute resolution method,
as there is then no evidence of consent to a particular process. (13) This challenge
could also arise where there is an argument in relation to the agreement's
coverage as to a group of companies or as to an entity that controls a signatory.
This is considered separately in Chapter 7.

Fouchard, Gaillard, Goldman disagree that there is a distinction between cases

where there is an allegation of non-existence of a contract as opposed to whether
it is void, (14) arguing that there is competence in each case. (15) This certainly
seems the better view both as to intent and logic where both parties at least
contemplated arbitration but one contends the agreement was not
consummated. Even where it is alleged that a contract has not been formed, in
some cases it may be arguable on the facts that the parties agreed on an
arbitration clause as part of the negotiations and intended this to provide the
mechanism to decide whether the negotiations were themselves validly
concluded. (16) In other cases, because the parties have selected an arbitration
clause in the documentation, it seems preferable to conclude that they intended
all questions to be determined by a tribunal, including whether the page "133"
contract itself was consummated. Stated differently, if the parties clearly
intended arbitral jurisdiction over a consummated contract, they can be taken to
have intended such jurisdiction over the very question of formation itself. Yet
another way to look at allegedly unconsummated agreements is to consider two
conflicting hypothetical scenarios, namely, that the allegation of non-completion
is correct or it is not. If the non-completion argument is not correct, the only
natural adjudicatory forum is arbitration if the contract contains an arbitration
clause. Conversely, if there is no consummated contract, then the claim should be
denied with res judicata effect. The only legal fiction in the latter event is the
logic by which this determination comes from an arbitrator and not a court. If the
issue is left to be determined by courts, there would be a problem as to which
court to choose. If it was a court other than a court of the Seat, such other court
cannot make a decision as to arbitral jurisdiction that is binding on a tribunal. It
also cannot validly injunct a tribunal from finding jurisdiction on the facts. Yet it
may be the court that would have jurisdiction but for an arbitration agreement.
Even a court of the Seat may be problematic as courts in a neutral Seat are only
relevant because of the allegation of arbitration itself. Furthermore, some
supervisory courts can only overturn positive findings. In most cases there is an
indication in lex arbitri that a tribunal can decide on jurisdiction prior to a court.
Hence the better view is that lex arbitri providing for competence should not be
limited in any way, save where the parties own arbitration agreement uses
language that might limit an arbitrator's powers in such circumstances. Some
arbitral laws obviate the need for such analysis and expressly indicate that there
will be competence to decide allegations that the agreement never came into
existence. (17)

The situation is more complicated if one party argues that it never had any
interest in the arbitration and has never evidenced any intent to arbitrate, for
example under the forgery, group of company or corporate veil scenarios alluded
to above. If there is no a priori evidence of some hope to arbitrate by that party, it
is harder to identify an optimal solution. If the forgery argument is valid, a court
should not easily be bypassed by that forgery. The converse is equally true. Even
if there is some conceptual sympathy with a more circumspect view in these
cases, the problem of misuse of the argument by way of guerrilla tactics cannot
be ignored. (18) As Chapter 7 demonstrates, while multi-party scenarios are
contentious, the better view is to allow a tribunal to ascertain the coverage of a
clause as to persons. This should apply to the case where a person at least
appears to have signed in the forgery scenario.

page "134"

3.2.2. Applicable Law as to the Arbitration Agreement

The first issue facing a tribunal may be determining whether a valid arbitration
agreement exists and if so, what it requires and allows. This may include
questions such as whether an agreement to arbitrate has been consummated and
has not been overturned by subsequent conduct; whether it covers the parties
sought to be brought within the dispute; whether it covers the subject matter of
the dispute; whether it is sufficiently certain to support the arbitration; and
whether it meets the form requirements of the particular lex arbitri. The natural
starting point in seeking to determine the ambit of the parties' consent in the
arbitration agreement is to first look to the terms of the arbitration agreement
itself. In some cases the terms of the agreement will be clear. In other cases they
may be ambiguous, requiring interpretation. In turn this raises a conflict of laws
question as to the law applicable to an arbitration clause. While identification of
applicable law is primarily for a tribunal when considering the validity and ambit
of an arbitration agreement, courts may also have to consider this question, for
example, when determining whether an arbitration agreement should be
recognised and court proceedings otherwise stayed on a challenge to a positive
jurisdictional decision, or under an enforcement defence.

Because of the dual nature of an arbitration agreement, combining procedural

and substantive elements, there is an important conceptual and practical
question as to the proper law governing the arbitration agreement itself. Because
of its separability, it can, in theory, have a different law to the balance of the
contract. Party autonomy means that the parties can certainly select the law
applicable to their arbitration agreement but many arbitrators would consider a
general choice of substantive law in a commercial contract to only be dealing
with the balance of the contract and not the arbitration agreement itself. Hence
in many cases, there may be no designated law by the parties for the arbitration
agreement even though it would ultimately be within their power to make a
choice. Even if a choice expressly pertained to the arbitration clause, it might at
times be unclear as to whether this would encompass the substantive
contractual law of that chosen law or its procedural arbitration laws or both.

If the parties have not made a choice or are in dispute as to whether this has
occurred or what an ambiguous choice means, there are problems for a tribunal.
The first issue is one of classification. As noted, is interpretation of the
arbitration agreement a substantive issue, typically being part of a substantive
contract or is it a procedural issue, dealing as it does with the jurisdictional
framework? Given that it has elements of both, classification becomes difficult. A
further problem is that arbitral laws rarely provide an answer to this question.
Because of the hybrid nature, some are uncomfortable with a conflicts
methodology or a methodology based on provisions dealing with selection of
substantive law, and suggest an anational approach, looking for the intent of the
parties without resort to national law. On this approach, interpretation is based
on the wording and circumstances without resort to an applicable law. This is
discussed further in section 3.2.4.

page "135"

Absent any other stipulation, if a tribunal aims to apply a conflicts approach and
considers it a quasi-substantive matter, then it would look at the discretions it is
given in that regard, either via a conflicts or a direct choice methodology as
directed in the lex arbitri and/or arbitral rules. If the rules differ from the lex
arbitri, they would normally prevail as an express agreement of the parties as to
methodology at least. (19) Where a conflicts methodology is selected, there are
problems whichever conflicts rule is chosen. The two most prevalent tests are
the closest connection test and the domicile of the person exercising
characteristic performance test. If a closest connection test applies, will it be
analysed with respect to facts distinct from the balance of the contract, given that
the agreement to arbitrate is treated under notions of separability as a separate
agreement to arbitrate? If so, would that lead inexorably to the law of the Seat,
being the place where the arbitration is being conducted? A simplistic
application of the closest connection test will always be problematic. Separability
is required to make arbitration sustainable, but does not invite considerations to
occur in a vacuum. Hence, if the overriding issue is that a contract containing an
arbitration clause was never consummated, or that the signature was forged, it
makes little sense to have a different applicable law to the arbitration agreement
than to the balance of the provisions. One can readily conceive of cases where the
parties would want the arbitration clause to remain on foot even if the balance of
the contract is void, but it would be hard to envisage a contrary scenario. (20)
Similarly, to make the obvious statement that the characteristic performance of
an arbitration agreement is arbitration itself and hence the closest place is that of
the Seat, may be tenable, but the laws of most Seats do not carefully articulate
whether validity of an arbitration clause is purely a question of procedure or a
question of substance. The broad discretion to pick an applicable direct law or an
applicable conflicts rule also means that the Seat contemplates a tribunal finding
an appropriate law for the particular circumstances. Also, concentrating on the
law of the Seat concentrates on the mere activity of arbitration, but not the
content of the matters to be considered. It is even more problematic if the parties
have not selected the Seat and where instead this is left to a discretionary
determination by an institution or the tribunal. Where the domicile of the person
exercising characteristic performance is concerned, there are other problems. Is
that a reference to the parties, in which case there is no single venue, or to the
arbitrator, in which case it is problematic where there is a multi-person tribunal?
The discretionary problems are exacerbated if validity of the arbitration
agreement may be impacted upon by the choice of applicable law.

There is no consistent view on these issues. In the English Court of Appeal

decision of C v D it was suggested in an obiter dictum that under a closest
connection test, it would either be the law of the underlying contract or the law
of the page "136" Seat, with the answer more likely to be the latter. (21) Berger
argues for the application of the law of the Seat. (22) Conversely, Julian Lew argues
for a strong presumption in favour of the law governing the substantive
agreement. (23) Fouchard, Gaillard, Goldman similarly argue strongly that an
arbitration agreement is simply a contract. Hence the applicable law should not
be that of the Seat. (24) The authors argue for a substantive rules method which
would fall back on general principles of public policy. (25) Their suggested
approach is discussed further in section 3.2.4.

From both a jurisdictional and functional perspective, choosing the law of the
Seat supports enforcement indirectly, given that the New York Convention
allows enforcement to be barred if the arbitration agreement is not valid under
the law where the award was made. This only goes to validity, however, and does
not truly address the question of the optimal approach to determining intent via
interpretation of all aspects of otherwise valid agreements. Furthermore, as was
discussed in section 2.7.14, the duty to render an enforceable award, while well
recognised, should not go so far as to demand that this duty should automatically
predominate over other considerations when exercising a clear discretion as to
choice of law. A slavish concern to pre-empt spurious challenges to enforcement
may not otherwise be optimal from fairness and efficiency perspectives.

However, it is not simply a choice between the law of the Seat or the contract.
Marc Blessing has in fact identified nine possible laws that could apply to the
interpretation and validity of an arbitration clause. (26) The options he identifies
are the law of the place where the arbitration agreement has been concluded; the
law of the Seat of the arbitral tribunal, i.e., the lex arbitri; the proper law of the
arbitration agreement to the extent it has been chosen by the parties or can be
established under the circumstances; the proper law of the substantive contract
in which the arbitration clause is found (lex causae); the law of the parties or one
of them; the law of the country whose courts would have jurisdiction but for the
arbitration agreement; page "137" the law of the country where the award is
most likely to be enforced; a combination of previously articulated laws; or an
anational or denationalised approach by which the clause should be governed by
common and fundamental principles of law. While he accepts the central
importance of the lex arbitri, he argues strongly in favour of an analysis of the
real intention and objectives of the parties through the widest analysis of all
surrounding circumstances. The aim of such approach is to ultimately determine
what is the ‘subjectively fair and objectively reasonable expectations of each one
of the parties … ’ (27) His direction to consider intent in the context of what is
objectively fair and subjectively reasonable is in line with some decisions of
French courts that did not require reference to any State law. (28) The anational
approach was also adopted in Dalico, (29) although Poudret and Besson point out
that an anational approach could contradict Article V(1)(a) of the New York
Convention in terms of its stipulated connecting factors. (30)

The lex arbitri might in fact provide a direction. Some civilian countries opt for a
presumption in favour of the law of the contract. (31) Others opt for the law of the
seat of arbitration. (32) A similar approach is taken under Article 59(c) of the
WIPO Rules. Swiss law has a very broad approach to validity, expressly
indicating that ‘an arbitration agreement is valid if it conforms either to the law
chosen by the parties, or to the law governing the subject matter of the dispute,
in particular the main contract, or to Swiss law’. (33) Whatever one's view as to
applicable law, because there are different approaches and theories, a tribunal
should make the parties aware that this is a matter for determination and allow
submissions to be made. In appropriate cases a preliminary award or direction
as to applicable law could even be necessary prior to a determination of a
jurisdictional challenge.

page "138"

3.2.3. Interpreting the Arbitration Agreement

The previous section looked at the law applicable to an arbitration agreement in

order to determine how its validity and ambit might be assessed. This is a step in
the process of determining that there has in fact been an agreement and if so, the
nature of that agreement. As was pointed out, arbitrators might either apply any
of a number of different national laws or use an international approach and may
vary as to the evidence they would rely on, some restricting themselves to the
words of the agreement, some looking at surrounding circumstances. If an
arbitrator follows the approach of utilising a national contract law, such law will
also have principles of contract interpretation which would apply. This will
include preferences or discretions in relation to either or both plain meaning or
purposive approaches and either reliance on express wording or additional
concern for evaluation of surrounding circumstances. Inevitably, domestic
contract systems leave room for differences in approach on such questions,
either via broad discretions or via differing jurisprudential schools of thought
about the nature of the law concerned.

These challenges all face arbitrators but arbitrators face an additional conceptual
challenge when approaching the interpretation of arbitration agreements as
opposed to the interpretation of the balance of contractual terms. This is because
the arbitration agreement is the source of jurisdiction and the prime source and
evidence of consent. In this context, should an arbitrator take an expansive
approach to interpretation seeking to promote jurisdiction wherever possible,
perhaps on the basis that parties who use an arbitration clause want a valid
process? Is that approach best able to support the presumed intent of the parties
or does it involve a potential conflict in promoting one's own jurisdictional
validity? How should arbitrators approach the interpretation of potentially
pathological clauses where some key elements have been left out, presumably
because of ignorance or poor drafting? Should an arbitrator strive to make sense
of such clauses? Conversely, can a party challenging such a clause argue validly
that an arbitrator should not fill gaps to complete consent to arbitrate that was
not there at the outset? Is plain meaning instead the best approach? These are
questions about which there can be legitimate disagreement. They underlie the
methodological options discussed in the following sections.

3.2.4. Methodology of Interpretation

In addition to considering the broad approach to interpretation, an arbitrator

must also consider an appropriate methodology, particularly in relation to
evidence and inferences as to intent. Here one can again expect differences. First
there would be those favouring a consent paradigm who may simply look to all
of the negotiations, perhaps utilising a presumption in favour of validity as this
would be consistent with the original intent of the business persons involved.
Those favouring a more jurisdictional or scientific approach might follow a
conflicts methodology, looking page "139" to the proper law of the arbitration
agreement and the interpretative processes thereby identified.

As noted, one of Blessing's references was to an anational approach. This would

look for general principles of interpretation of arbitration agreements. Fouchard,
Gaillard, Goldman provide one such suggested methodology when discussing
interpretation in the context of identifying the parties' consent. The authors
recommend a number of principles of interpretation under an anational
approach which appear to have much merit. The first principle they contend for
is that of interpretation in good faith. The authors use the notion of good faith in
a limited sense to assert that a party's true intention should always prevail over
its declared intention where the two differ. Thus a party who freely entered into
an arbitration clause but challenges its form in order to avoid liability would fall
foul of this principle. (34) This principle of interpretation invites the tribunal to
look behind the declared intention for what could have been the true intention at
the time of entering the arbitration agreement. At the time of negotiation of the
arbitration agreement, parties hope that there will not be disputes, would
usually not be aware whether they would be claimant or respondent in the event
of any dispute and could in good faith be presumed to want their disputes
resolved efficiently and amicably. Once a dispute has arisen, however, parties
tend to think of their own best interests, which could include looking for the
most tactically advantageous dispute settlement process. It is the initial intent
and not the latter, which should be determinative. Nevertheless, the authors note
that merely challenging the validity or ambit of an arbitration clause is not
presumptively an act of bad faith.

Fouchard, Gaillard and Goldman go on to derive more specific rules of

interpretation flowing from the obligation to interpret contracts in good faith.
The tribunal must look for intention in context, taking into account the
consequences which the parties reasonably and legitimately envisage. (35) The
authors also suggest that the attitude of the parties after signature and up until
the time of the dispute arising should be taken into account as subsequent
conduct confirming their original perceptions. They note that this is sometimes
referred to as ‘practical and quasi-authentic interpretation’ or ‘contemporary
practical interpretation’, commonly applied in arbitral case law. (36) The authors
also note that this leads to a requirement that the agreement must be interpreted
as a whole. While certainly an appropriate suggestion, it will still pose challenges
of circularity. For example, in group of contract situations, what is the ‘whole’ to
be interpreted? The authors do see this as appropriate to such circumstances. (37)

The second principle they refer to is that of effective interpretation (effete utile).
Where words can be interpreted in more than one way, the interpretation should
be preferred which gives practical effect to the words rather than one which
page "140" would undermine the agreement. This is a natural consent-based
approach, doing no more than relying on the assumption that the parties
intended their agreements to be effective. They note that this is a recognised
element of most legal systems, sometimes described under the maxim ut res
magis valeat quam pereat. It has also commonly been used in international law
interpretation. (38) While Fouchard, Gaillard and Goldman treat it separately to
the principle of interpretation in good faith, both the ICJ and the ILC have in fact
seen it as a subset of that concept. (39) The authors cite approvingly from the
preliminary award in ICC Case No 2321 (40) when in considering a potentially
pathological clause, the tribunal said:

When inserting an arbitration clause in their contract the intention of the parties
must be presumed to have been willing to establish an effective machinery for
the settlement of disputes covered by the arbitration clause.

Where pathological clauses are concerned the principle of effective

interpretation salvages the true intent which was ‘distorted by the parties'
ignorance of the mechanics of arbitration’. (41) No persons can be presumed in
good faith to have wanted to expressly provide for arbitration in an ineffective
manner. Hence it is natural to add meaning to the inadequate express clause,
filling the gaps where possible or treating it as a nullity where the faults are
irreparable. A consent-based approach would also presume that the parties
would rarely wish to have separate proceedings. (42) While the authors strongly
support this principle, they correctly distinguish it from a supposed principle of
interpretation in favorem validitatis, which asserts some general presumption in
favour of arbitration as, for example, has been applied by an American court in
Moses H Cone. (43) They criticise any such presumption and also argue
persuasively that there similarly should not be a presumption against
arbitration. Some have asserted that an arbitration agreement should be
interpreted restrictively on the basis that parties should not be seen to have
given away their rights to national court jurisdiction lightly. Examples of the
latter approach include ICC Award No 2138, (44) and page "141" ICC Award No
2321. (45) Instead, effective interpretation should look to the true intent of the
parties without presumptions either way. The difference between the two
concepts is that effete utile works within the words used and resolves ambiguity
in favour of validity. In favorem validitatis goes further and provides a
presumption in all circumstances. Importantly, the effete utile principle works
best in terms of validity and scope of claims but not where the issue is scope as
to parties. As to the first category, this will at times support an expansive
interpretation of an agreement so that all issues between the parties can be aired
at one time. As to the latter, it is not readily applicable when the assertion is that
a non-signatory should be included as there typically is no starting point where
that person can be presumed to want a valid arbitration.

The third principle they refer to is interpretation contra preferentem. The

authors again use this in a restrictive fashion, akin to the duty to interpret in
good faith. The authors contend for the principle in circumstances where a party
who drafted an ambiguous or obscure text then seeks to rely on their own
ambiguity to read down the clause. (46) They do not propose the more expansive
notion where ambiguities are uniformly resolved contrary to the position of the
drafter.

The above principles are also strongly intent based, albeit building on good faith
intent assumptions alone. There is much to be said for the views of Blessing and
Fouchard, Gaillard, Goldman calling for a comprehensive analysis of the parties'
intent wherever that may resolve the issue. This is at least a legitimate
methodology and cannot be ignored. It should be understood, however, that
some of these principles will vary in their utility depending on the factual
circumstances. Where plain meaning is concerned, views can still differ
depending on whether broad or narrow approaches are taken to ambiguous
scenarios in terms of the ambit of an otherwise valid agreement to arbitrate.
Principles such as effete utile may not be as relevant in this context. Problems
thus arise under this methodology where there is clearly an agreement to
arbitrate but uncertainty as to ambit of the mutual promises. As noted, there is
also less logical justification to use an effete utile approach in some complex
arbitration scenarios, for example, where an attempt is made to apply an
arbitration agreement to a non-signatory-related company or controlling
shareholder. This is simply because that person had not evidenced a clear wish
to arbitrate. At the very least that is the question in issue and cannot be a starting
presumption for the tribunal's analysis. Interpreting arbitration agreements in
the context of complex scenarios is further considered in Chapter 7.

Another approach is to adopt a conflicts methodology, select the law applicable

to the arbitration agreement from the range identified above and then use that
law's interpretative maxims. In some cases this will lead to similar
considerations. Whichever law applies, most legal systems would support
concern for plain meaning in context, subject to considerations of the purpose
behind the agreement, at page "142" least where sufficient ambiguity exists.
Legal systems may vary as to the extent to which extraneous evidence may be
used to help interpret an arbitration agreement. Common law systems are less
inclined to do so, but are more disposed to see implied terms in arbitration
agreements. (47) Where extraneous evidence is concerned, given the writing
requirement for arbitration agreements, for example, in Article 7(2) of the
UNCITRAL Model Law 2006 Option I and Article II of the New York Convention,
some arbitrators may refrain from considering such evidence for that reason
alone. This should not be the preferred view. The writing requirement should
not constrain arbitrators who feel that extraneous evidence would assist in
resolving ambiguities on the face of an agreement that itself satisfies the writing
requirement. The writing requirement is about ensuring adequate proof of
waiver of litigation rights. If the choice in favour of arbitration is clear,
extraneous evidence may be the best way to understand the parameters of such
a choice. (48)

3.2.5. The Scope of the Agreement

Until recently English courts adopted a pedantic approach to the construction of

an arbitration agreement. They focused on the words employed and drew fine
distinctions between the terms used. Thus some judges took the view that a
clause referring disputes ‘arising under’ a contract was narrower than a clause
referring disputes ‘arising out of a contract (49) and it was commonly accepted
that a clause referring disputes ‘in relation to’ or ‘in connection with’ a contract
was broader and therefore conferred a more extensive jurisdiction on the
arbitral tribunal. These fine distinctions were particularly relevant in relation to
the question of whether an arbitral tribunal had jurisdiction to hear disputes
concerning pre-contractual representations, (50) rectification (51) and claims
founded on statute for misleading and deceptive conduct or for breach of
competition law. (52)

page "143"

In contrast to the strict and pedantic construction of arbitration clauses adopted

in many common law jurisdictions, civil law jurisdictions are often more liberal
and tend to construe arbitration agreements in a broader way. However it seems
that the difference between the classic common law approach and the more
liberal civil law approaches may now have disappeared. The decision of the
House of Lords in Premium Nafta Products (53) marked a change in the position
adopted in the UK. In that case the House was concerned with arbitration clauses
in eight charterparties. The relevant clauses gave each party an election to refer
‘any dispute arising under this charter…’ to arbitration in London. One of the
questions for determination was whether, as a matter of construction, the clause
was broad enough to cover the question of whether the contract was procured
by bribery. Lord Hoffmann observed:

Both of these defences raise the same fundamental question about the attitude of
the courts to arbitration. Arbitration is consensual. It depends upon the intention
of the parties as expressed in their agreement. Only the agreement can tell you
what kind of disputes they intended to submit to arbitration. But the meaning
which parties intended to express by the words which they used will be affected
by the commercial background and the reader's understanding of the purpose
for which the agreement was made. Businessmen in particular are assumed to
have entered into agreements to achieve some rational commercial purpose and
an understanding of this purpose will influence the way in which one interprets
their language….

If one accepts that this is the purpose of an arbitration clause, its construction
must be influenced by whether the parties, as rational businessmen, were likely
to have intended that only some of the questions arising out of their relationship
were to be submitted to arbitration and others were to be decided by national
courts. Could they have intended that the question of whether the contract was
repudiated should be decided by arbitration but the question of whether it was
induced by misrepresentation should be decided by a court? If, as appears to be
generally accepted, there is no rational basis upon which businessmen would be
likely to wish to have questions of the validity or enforceability of the contract
decided by one tribunal and questions about its performance decided by another,
one would need to find very clear language before deciding that they must have
had such an intention.

Having regard to these considerations, Lord Hoffmann said that the court was
required to give effect to the commercial purpose of the arbitration clause so far
as the language used by the parties would permit. He then referred to earlier
English decisions and further observed, with considerable frankness, that the
fine distinctions drawn in these cases ‘reflect no credit upon English commercial
law’. He then stated the new approach which should hence forth be adopted:

In my opinion the construction of an arbitration clause should start from the

assumption that the parties, as rational businessmen, are likely to have intended
page "144" any dispute arising out of the relationship into which they have
entered or purported to enter to be decided by the same tribunal. The clause
should be construed in accordance with this presumption unless the language
makes it clear that certain questions were intended to be excluded from the
arbitrator's jurisdiction. (54)

3.2.6. Interpretation and Procedural Directions

Interpretation may also be needed to determine the nature and ambit of

procedural directions included in the agreement as well as its substantive
coverage. Directions might even be implied in some factual circumstances. There
is then a question as to the source of such implied terms. There is also a question
as to whether any may be implied from the inherent nature of arbitration and
from international public policy norms. In the context of procedural norms,
international public policy would be unlikely to lead to any considerations
besides those directly applicable in any event under the lex arbitri, although the
possibility that a particular enforcement court may take a different view cannot
be overlooked.

Another suggested source of implied terms might be customs and usage. If this is
asserted to be the basis for an implied procedural term, it is not really a separate
source of arbitral procedure, but instead a mechanism of determining implied
consent. Common usage can even be justified on the basis of the legitimate
expectations of the parties, hence again being an element of implied consent. No
problem arises where the parties and arbitrator all wish a particular usage to be
followed. Where the parties disagree as to whether a common usage is part of an
implied term in the arbitration agreement, the arbitrator will need to make a
determination. The argument would be that parties in a particular trade that
expect arbitrations to be conducted in a particular way have impliedly qualified
their consent in this way. Whether the latter argument would succeed would
depend on the facts of the particular industry and its practices. There ought to be
a significant evidentiary burden on anyone asserting this to be so. One would
also wish to look to see whether they might have wanted an atypical approach or
waived their rights to complain, based on acquiescence in the original
arbitration. The same problem should not arise where the common usage is only
pressed upon the arbitrator as part of the arbitrator's accepted broad discretion.
Because an arbitrator has a broad discretion to do justice in the instant case, he
or she may decide that this is best achieved by following or not following a
habitual practice.

page "145"

3.2.7. Pathological Clauses and Tribunal Interpretation

The above-mentioned approaches to identifying applicable law and interpreting

arbitration agreements are most commonly employed when there is a challenge
to arbitral jurisdiction alleging that the clause is fatally pathological. This will
typically arise because of inadequate or ambiguous drafting. If the better view is
that the approaches of Fouchard, Gaillard, Goldman and Blessing are to be
preferred, then many problematic scenarios will still lead to valid jurisdiction. An
example of the expansive approach to interpretation was enunciated by an ICSID
tribunal in Amco v Indonesia. The tribunal stated:

Conventions to arbitrate, should be construed in good faith, that is to say by

taking into account the consequences of their commitments the parties may be
considered as having reasonably and legitimately envisaged. (55)

This is similar to the logic in Premium Nafta discussed in the previous section.
While this is the preferred and emerging view, tribunals may still vary depending
on whether they are willing to consider extraneous evidence to identify intent.
Some would instead simply presume an intent to valid and effective arbitration.
That could be so where the parties have clearly identified arbitration as the
preferred model but have proved less than competent in identifying the Seat, an
institution and/or applicable rules. A paradigm example of a pro-arbitration
approach to interpretation involved the High Court of Hong Kong in Lucky-
Goldstar v Ng Moo Kee Engineering. (56) The clause referred to arbitration in a
‘third country under the rule of the third country and in accordance with the
rules of procedure of the International Commercial Arbitration Association’. The
judge relied on an intent to arbitrate and ignored the reference to the non-
existent arbitration institution.

Another scenario is where a contract inadvertently contains conflicting

provisions. The primary terms might incorporate a choice of forum clause but
also refer to general conditions which contain an arbitration agreement. Here it
would be inappropriate to use effete utile as there are two possible fora that
either party may wish to see as effective. In such circumstances, and absent other
evidence, adjudicators will typically use presumptions of intent such as that an
express direction prevails over a general stipulation or a later designation
prevails over an earlier one. These are only presumptions of intent and an
arbitrator may look at all the circumstances to determine if there was truly an
agreement within the contemplation of the parties.

page "146"

3.2.8. Arbitrability

Previous sections have looked at validity of arbitration agreements and the

related questions of applicable law and approaches to interpretation. In addition
to questions as to the ambit and status of the parties' agreement to arbitrate,
there is a distinct question as to whether the matters they have sought to bring
to arbitration are capable of being decided in that format. In this context, it is
necessary to consider whether legal systems aim to limit certain kinds of
disputes to court jurisdiction alone, hence rendering them non-arbitrable
notwithstanding party intent to the contrary. One difficulty in fully
understanding the concept of arbitrability is that it is used in different ways by
different jurisdictions. US courts and authors often use it in a broader sense
covering a number of questions, including jurisdiction generally, matters that are
capable of being dealt with by arbitration and at times, covering the relevance or
otherwise of mandatory substantive laws. It is more appropriate to limit the
term to the question as to which matters parties may be permitted to submit to
arbitration. In this sense it is separate to the question of the existence and scope
of the arbitration agreement. The concept of arbitrability is also different to the
question of the ambit of a tribunal's powers, particularly in relation to remedies.
Poudret and Besson note that the two can be related, for example, when a
tribunal has asserted jurisdiction over a punitive law with commensurate
damages. (57)

Matters that are typically held to be non-arbitrable are criminal liability and
family law. Other more debateable categories about which there are differences
in view include insolvency, registration of intellectual property rights, labour
disputes, grants of statutory licences (58) and illegal or corrupt contracts. In some
countries there may be greater restrictions on arbitrability of claims related to
administrative law, (59) property leases, (60) consumer protection and general
protection of parties seen as being in weaker positions. (61) Where there is a
legislative provision seeking to bar arbitrability, this may at times raise
questions of interpretation as to extraterritorial intent, and if so, whether that is
effective where the legislation purports to apply in a foreign Seat. (62)

page "147"

Some important questions relating to arbitrability include what law applies to

determine the question and whether a tribunal may or should consider this of its
own volition or should wait for a challenge by a respondent. As noted by Lew,
Mistelis and Kröll, the analysis is more complex when it is a tribunal and not a
court that considers questions of arbitrability. They identify eight different
approaches in both practice and scholarly writing as to the law to be used to
determine arbitrability. (63) While the dominant view is in favour of the law of the
Seat, others argue for an international approach, (64) or for the law of the
enforcement country. The latter view relies, in part, on the fact that the New York
Convention refers to arbitrability in the context of the law of the enforcement
country. (65) At the earlier recognition stage, Article II(3) of the New York
Convention allows an exception to recognition where the dispute is not capable
of settlement by arbitration, but gives no guidance as to the law to be used to
determine whether this is so or not. The predominant and better view is that
arbitrability should be determined by the law of the Seat, as this is the country
that is determining which forms of arbitration it seeks to encourage. The laws of
any enforcement country ought not to be determinative as enforcement could
occur in a range of jurisdictions, only some of which may have restrictive notions
of arbitrability. Such restrictive notions ought not to apply to assets in foreign
jurisdictions. (66) There would also be a problem if there were assets in two
jurisdictions where these had opposing views on arbitrability. At the very least,
the potential application of a foreign country's arbitrability notions should
simply be seen as a subset of the mandatory law question, dealing with specific
cases where arbitral jurisdiction is purported to be barred. The application of
mandatory laws is discussed further in section 13.8.

Conversely, a consent-based paradigm might be said to invite attention to the

law of the arbitration agreement to determine what the parties have consented
to. This should not be accepted as the means to deal with arbitrability. Under a
pure consent paradigm, arbitrability would not need to be separately considered.
The only question would be to determine what the parties have agreed to.
Arbitrability notions are instead a potential constraint on autonomy. Looking for
the constraint within an express law selected by the parties would also raise
interpretational problems when the arbitration clause is broad but the
applicable law provides arbitrability qualifications. Even in the absence of
inconsistencies, such a law, commonly not selected by the parties expressly,
ought not to be seen as the source of any limitation on their autonomy itself.
Hence it is better to see it as a question of mandatory procedural rules of the Seat
that bar page "148" arbitrability. (67) Poudret and Besson raise the additional
concern that an approach concentrating on the parties' choice of law would allow
parties to try and bypass a mandatory procedural law by selection of a foreign
law pertaining to their arbitration agreement.

Returning to the question of the categories of arbitrability, in view of the above

commentary, this may vary depending on the circumstances, including the lex
arbitri involved. Where lex arbitri are concerned, many will simply use a general
criterion such as referencing ‘claims in respect of which the parties may reach a
settlement’ . (68) This will then be a question of interpretation, with domestic case
law being a useful guide. Some lex arbitri attempt a more specific definition of
arbitrability. Swiss PILA refers to the fact that ‘any dispute of financial interest
may be the subject of an arbitration’. (69) French jurisprudence deals with the
question under notions of international public policy. (70) Where a dispute does
not involve a financial interest, German law adds an entitlement of the parties ‘to
conclude a settlement on the issue in dispute’ to the pecuniary criterion as an
alternative factor leading to arbitrability. (71) Only some general observations can
otherwise be made. Over time, it has been accepted that competition law issues
are arbitrable. Indeed a failure to consider competition law even sua sponte may
be a ground for annulment, in Europe at least. (72) There may be tangential issues
as to the ability of an arbitrator with jurisdiction to consult with competition
authorities. Where intellectual property is concerned, there are differences in
approach between countries, some with very liberal arbitrability rules, while
others reserve jurisdiction for courts. It is also important to distinguish between
a mere contractual dispute about the exploitation of such property and
conversely, disputes directly concerning the property rights themselves. (73) It is
natural to accept that the pure contract claim is arbitrable (74) while it may be
more problematic when the challenge is as to who has the priority rights to
registration. Even then that should not be a concern if the tribunal is not seeking
to bind the registration authorities. A decision that one contracting party has
rights over the other simply means that the latter should be barred from seeking
registration. The party granted priority through the arbitration would still need
to satisfy the relevant authority as to entitlement.

Where insolvency and bankruptcy laws are concerned, there are particular
difficulties in identifying the proper ambit of arbitrability for two key reasons.
page "149" First, efficient insolvency law looks to collect all disputes in one
forum so that a liquidator or administrator can make rational decisions on which
to pursue from a cost-benefit basis so as to maximise the returns to creditors.
The controlling forum would thus wish to have as much control over foreign
assets as possible. There can then be foreign mandatory procedural laws to be
considered and limits on enforceability. A separate problem where insolvency
and bankruptcy are concerned is that the ongoing capacity of one of the parties
to direct the arbitration may be adversely affected. As a matter of insolvency law,
the insolvent certainly cannot continue to exercise independent decision-making
functions, but less clear is whether the arbitral rights and duties are taken over
by the liquidator. (75) The impact of insolvency on arbitral procedures is
considered further in section 7.15.

In addition to considering whether a tribunal may arbitrate certain matters, it is

also appropriate to consider what are the implications of any positive
determination to do so in these contentious areas. A tribunal can only determine
rights between the parties and cannot make binding decisions in relation to
other relevant persons. For example, a tribunal dealing with a competition law or
intellectual property matter cannot expect its decision to be res judicata vis-à-vis
government officials and cannot make rulings that determine the rights,
liabilities and licences of third parties. For similar reasons, arbitral
determinations may not lead to issue preclusion or issue estoppel when such
matters are subsequently raised by public authorities.

Where States are concerned there can also be limits on their liability to be
subject to suit. The key principle is that of sovereign immunity. Where it applies
it has a similar effect to arbitrability notions. However, this does not generally
apply in arbitration. A number of international treaties and arbitral statutes
indicate that a State agreeing to arbitrate has waived its right to claim sovereign
immunity. (76) Enforceability can still be a problem in some countries. The
Washington Convention does provide for enforcement of awards but also
indicates in Article 55 that nothing is construed as derogating from the law in
force in any contracting State relating to execution immunity of that State or of
any foreign State. Even where sovereign immunity has not been waived, modern
articulations of the concept would exclude the principle from applying to most
matters that are subject to arbitration because they typically exclude immunity
in relation to commercial activities.

Finally, there are different views as to whether a tribunal should of its own
volition consider arbitrability even if not raised by the parties. The better view is
that this should be so. This is particularly the case where limits on arbitrability
are found within the law of the Seat, the place from which the tribunal derives its
very jurisdiction. A prohibition against arbitration of certain disputes within that
law should be seen as mandatory and hence should be considered sua sponte by
the tribunal. (77) However, due process and the right to be heard will demand that
a page "150" tribunal warn the parties of its intention to consider such issues
and allow them to make submissions. If the parties then sought to bar the
tribunal from considering the question, this would raise the issues considered in
sections 2.10.3, 6.2.4 and 6.2.5.

3.2.9. Capacity

As with any form of agreement, arbitration clauses will only be recognised in law
if the relevant parties have the capacity to enter into such agreements. Lack of
capacity is a ground for challenging enforcement under Article V(1)(a) of the
New York Convention. In determining capacity, there is a need to consider which
law applies. The New York Convention does not identify the relevant law, simply
referring to the law ‘applicable to them’. Typically, that will be the law of the
domicile or residence of the relevant legal person. The capacity of a corporation
would normally be dependent on its country of incorporation. Some tribunals
would take the view that an express choice of law should apply to capacity as
well. This cannot lead to a binding conclusion as to corporate validity in the
country of incorporation but it may be a valid view as to intent for inter-party
rights, for example, who may accept an assignment of arbitral rights.

An important area, particularly in relation to investment arbitration, is where a

State is a party and the question arises as to whether various statutory bodies,
State departments or semi-government authorities are either entitled to enter
into arbitration agreements themselves or can bring the government into such
processes through administrative law principles or concepts of agency. (78) On the
one hand, the laws of that country naturally determine what is permissible, but
on the other, it is a general principle of international law that a State cannot rely
on its own laws to escape legal obligations. That principle is itself a form of good
faith or equitable reliance. (79)

3.2.10. Form and Proof of Arbitration Agreements

Traditionally, the major issue with form and proof of arbitration agreements was
the writing requirement, contained in the New York Convention and in the
original version of the UNCITRAL Model Law and Option 1 of the 2006 revision.
Similar provisions are contained in other lex arbitri. Article II(2) of the New York
page "151" Convention refers to arbitration clauses contained in a contract or an
arbitration agreement signed by the parties and also covers an arbitration
agreement contained in an exchange of letters or telegrams. Courts and
commentators have extended the meaning of the latter to telexes, telefaxes and
emails in line with more modern forms of communication. There is also a
reservation in relation to States whose laws do not require that an arbitration
agreement be made in writing. There is a debate as to whether this refers to the
State of the Seat, that of recognition (80) or to the States of the parties involved. (81)
As originally drafted, Article 7(2) of the UNCITRAL Model Law requires an
arbitration agreement to be in writing but provides a slightly more expansive
interpretation than that contained in the New York Convention. It was effectively
updated for more modern forms of communication and also contemplated the
exchange of statements of claim and defence, which could have been seen as an
implied agreement in writing in any event. Article 7(2) also allows for a
reference in a contract to another document containing an arbitration clause
‘provided that the contract is in writing and the reference is such as to make that
clause part of the contract’.

While it may have been natural to impose a writing requirement for confidence
of proof of an arbitration agreement, there are many cases where an adjudicator
would naturally conclude that there was in fact an agreement but where the
strict definition of writing may not have been satisfied. Examples included
corroborated oral acceptances of written offers and acceptance by conduct.
Various tribunals and courts took differing views in these scenarios. (82) Leading
practitioners called for modifications (83) and the 2006 revision of the UNCITRAL
Model Law gave enacting States two options, one a far more liberal writing
requirement and the other, removal of the writing requirement entirely. Under
the revised version of the Model Law, Option I Article 7(2) still requires the
agreement to be in writing but Article 7(3) defines that to occur if the
agreement's ‘content is recorded in any form, whether or not the arbitration
agreement or contract has been concluded orally, by conduct, or by other means’.
Article 7(4) elaborates on the meaning of electronic communications and Article
7(5) and (6) repeat reference to exchange of statement of claim and defence, and
reference to documents containing clauses respectively. Option II leaves out the
writing requirement entirely.

page "152"

It needs to be remembered that most countries have not accepted the 2006
revisions to the UNCITRAL Model Law, hence in most cases, the original writing
requirement will remain. Conversely, some laws did not follow the original
UNCITRAL Model. Some lex arbitri simply call for proof of the arbitration
agreement in writing. (84) Some allow for passive acceptance. (85) Some expressly
allow for oral acceptance. (86) It also needs to be remembered that national courts
do not always approach these issues consistently and there is a separate
question as to whether the writing requirement is satisfied for enforceability
purposes. Because the New York Convention calls for recognition as well as
enforcement, a country which has adopted the Convention should not impose
more rigorous form requirements than those contained within the convention
itself.

3.3. Drafting Arbitration Agreements

There is no single model arbitration agreement that is ideal. A system based on

consent needs to be tailored to the individual needs of particular parties in the
circumstances of their commercial transactions. Nevertheless, there are certain
key features that should be directly or indirectly contained in any agreement to
ensure that it will be effective, binding and enforceable. Other important drafting
objectives are to ensure that the procedures cannot be abused by a party seeking
to avoid payment of money that is due and payable. Ambiguities and
uncertainties within the drafting of the arbitration agreement can lead to
unnecessary delays while jurisdictional questions are challenged before the
tribunal and/or the courts. In the extreme, inadequate drafting could render the
clause fatally pathological. For any arbitration, there needs to be at least a
mechanism for selection of the Seat, empanelling of arbitrators and
establishment of the procedures. A choice of the Seat or a set of respected
arbitral rules will commonly lead to the other elements being appropriately
established. For example, selection of a Seat will invigorate the lex arbitri of that
Seat, which in most cases will provide an appropriate framework for appointing
arbitrators and establishing procedure.

While it is easy to state the ideals where dispute resolution drafting is concerned,
it is important to situate this into the context of overall contract negotiations to
understand why problems continually arise. Transactional lawyers often know
little about dispute settlement, arbitration or the problems of pathological
clauses. Many law firms have artificial divisions between ‘front end’ transactional
lawyers and ‘back end’ dispute resolvers, which can detract from the firm's
ability to appropriately prepare for fair and efficient dispute resolution at the
drafting stage. Furthermore, the whole notion of drafting is often undertaken in
the context page "153" of compromise. Yet compromises in drafting dispute
resolution clauses can often render them ambiguous, internally inconsistent and
ultimately raise the possibility of invalidity. Finally, at the time that contracts are
being negotiated, the commercial parties are not likely to want to spend time on
dispute resolution clauses as they are hopefully trying to encapsulate a win-win
arrangement in documented form, albeit one where they may have contrary
views about the proper balance between risk and reward for each party. In an
environment where everyone is hoping for mutual gain, to raise complex issues
of dispute resolution can undermine the ambience. The more that one party
raises concerns about dispute settlement, the more the other may wonder
whether there is a distinct possibility that it expects some breach or dispute in
the future. Some contract negotiations have broken down simply because of
concerns about dispute resolution clauses. This can be particularly problematic
in multi-party scenarios where the needs of complex clauses are great, but so too
are the costs and difficulties. An example would be a construction contract where
an owner may naturally wish to have one dispute resolution forum and demands
that the builder includes matching arbitration clauses in any subcontract
arrangement. Yet the builder may feel that this unduly constrains it from being
able to get the best subcontractors at the best price.

Practitioners are not required to address these issues from a blank slate. Various
institutions provide guides to drafting arbitration agreements and other dispute
resolution clauses. (87) The most important observation about such guides is that
they are tried and true means of ensuring the viability of arbitral processes.
Practitioners without experience in arbitration should be wary of attempting to
draft their own arbitration agreements. ‘Specially drafted arbitration clauses are
more likely to give rise to uncertainty, ambiguity or formal invalidity than a
standard form.’ (88) This can remain a problem even if an institutional or other
guide is used as a basis, as any variations can still lead to potential problems of
interpretation. Nevertheless, in some cases, standard forms do not cover all
features that may be thought desirable. Where this is the case, it is advisable for
practitioners to add to a standard form clause in order to incorporate specific
requirements. This is in order to ensure that the clause continues to address the
core aspects considered necessary by the institution responsible for the standard
form clause. (89) In any event, it is appropriate to consider what are the key
elements required so that any model, whether institution or ad hoc, can be
properly page "154" evaluated. With these caveats in mind, the following is a
checklist of matters to consider both when drafting and when reviewing a draft
proposed by another contracting party:

1. The agreement should cover both existing and future disputes.

2. The clause should indicate its coverage and should make it clear that it is
broader than purely contractual disputes if the essential arrangement is one
of contract. This is so that related claims can also be included, such as those
in relation to misleading conduct or fraud. A broad formulation also allows
for claims in relation to pre-contractual negotiations. To achieve this, it is
typical to speak of disputes ‘arising out of or in relation to’ or ‘in connection
with’ the contract. (90)
3. It is preferable for the parties to expressly refer to ‘arbitration’ and not use
 words that could allow an interpreter to confuse this concept with distinct
concepts such as expert determination. (91) If a clause simply indicated that
disputes are to be resolved by a designated person without using the term
‘arbitration’ an adjudicatory body would most likely conclude that
arbitration was implied, but it would be wasteful to force this determination
to occur.
4. It is important to use mandatory language to the effect that the parties are
required to arbitrate, e.g., all disputes ‘shall’ be resolved by arbitration.
While courts will generally support clauses drafted permissively, for
example, those which indicate that disputes ‘may’ be referred to arbitration,
it is preferable that the drafting is clearly mandatory in language.
5. Ideally the parties should choose the Seat, with proper consideration as to
which Seat would be best. There are a number of factors that are relevant to
selecting an ideal Seat discussed further in section 3.5.1 below. For example,
it should have an appropriate lex arbitri. The country should be party to the
New York Convention to maximise enforceability. Geographically and
logistically, the cost should be appropriate and it should be in a convenient
location. Finally, its courts should be supportive of arbitration, helping
where necessary and not being unduly interfering. In some cases,
inadequate drafting would lead to a dispute as to whether a reference to a
particular place is a reference to the Seat or merely to the place of hearing.
The wording of the drafting should make this clear. (92)
6. Parties may wish to make it clear that hearings can be held in places other
than the Seat. This is made clear in most laws and rules.
7. Parties need to decide between institutional and ad hoc arbitration and also
decide whether they will utilise any internal procedural rules. There page
"155" are advantages and disadvantages to each approach. This is discussed
in sections 3.14, 3.15 and 3.16. Where an institution is selected, a choice
must be made as to which is most appropriate. That would include
questions of stature and competence, cost and in some cases, geographical
location. While it makes sense to specify an institution if that is the
preferred methodology, if for some reason the institution no longer exists or
changes its nature or status, there might be debate as to whether the
arbitration agreement was conditional and no longer operative. Different
courts and tribunals have taken different views in such circumstances. (93)
Where a less well known institution is to play a part in this mechanism, the
parties should confirm that the institution will be ready and willing to play
the relevant part in the future. (94)
8. The parties must be careful to accurately state the name of any institution
that will be involved in the arbitration if that is to be the form selected. (95)
For an institutional arbitration, this will be the relevant institution. For an
ad hoc arbitration, this may include, for example, an institution tasked with
selecting the arbitrators or otherwise acting as an appointing authority.
Simply stating the site of that institution is also ambiguous as it will often be
unclear whether that is a reference to a selected Seat or simply further
descriptive details as to the institution selected.
9. Parties might consider whether they wish to select the number of
 arbitrators and/or designate any special qualifications. There are
advantages and disadvantages to doing so. One concern, particularly in
relation to future disputes the nature and size of which is unknown, is that
the specification of particular qualifications or other attributes may turn out
to be less relevant to the dispute when it arises. Also, parties should
consider the effect that specification of particular qualifications may have on
the size of the pool of potential arbitrators. (96) For similar and other reasons,
an arbitration agreement will not select the arbitrators by name. (97) Instead,
an arbitration agreement will directly or indirectly provide a mechanism by
which arbitrators can be selected. An indirect method of selection of
arbitrators would be through the selection of a Seat and its lex arbitri and/or
specific procedural rules, each of which would be expected to contain a
selection mechanism.
10. Parties may wish to select the applicable substantive law. This can also
include consideration of broad non-national laws such as lex mercatoria,
which are accepted principles common to business persons around the page
"156" world. (98) A typical example is the United Nations Convention on the
International Sale of Goods, which many countries have adopted as
applicable in any event. If a law is selected, the drafter may expressly
indicate that the conflicts rules of that law are not to apply, to avoid renvoi
problems, although this is articulated in many lex arbitri in any event. There
will also be a need to consider whether the choice of law is expressed to
apply to interpretation of the arbitration agreement itself, given that there is
no presumption to this effect owing to the doctrine of autonomy or
separability. Furthermore, it may be advisable to stipulate the substantive
law within the contract but do so in a clause separate from the arbitration
agreement. This is in order to make it clear that the selection of substantive
law governs the contract generally, and is not relevant only in the context of
arbitration. (99) However, stating its intent clearly is more important than
deciding where it is situated.
11. Arbitral rules also allow an arbitrator to decide ex aequo et bono (that is
without rigidly following legal principles) but only when the parties agree to
this. Hence the parties should at least consider whether they wish to so
empower the tribunal. (100)
12. The parties might need to consider whether they each have capacity to enter
into the commercial arrangement including any arbitration agreement.
Proof of capacity may be particularly important when State entities are
concerned.
13. Similar concerns arise where there is multi-party arbitration, for example, a
group of companies engaged in a building transaction. There is a benefit in
ensuring that all parties can come within the arbitration even though only
one might have signed the arbitration clause. This may require clear side-
agreements from all relevant parties.
14. In specific circumstances, the parties may wish to exclude certain kinds of
disputes from arbitration. This would arise where certain disputes might be
handled more cheaply and expertly in other ways, for example, ‘pricing and
technical disputes’. (101) Such exclusions may be drafted into the arbitration
agreement, but should be approached with caution and with the guidance of
 a specialist in order to ensure that they do not inadvertently reduce the
scope of arbitrable disputes beyond the level intended or raise
interpretation problems at the margin.
15. In trying to ensure that arbitration is conducted expeditiously, some suggest
including deadlines in the arbitration clause itself. This might be more
dangerous than beneficial. Time limits might be considered, page "157"
although care should be taken to ensure that when breached, these do not
undermine the entire arbitration. It would be important to ensure that a
failure to comply with a specified deadline does not render the entire
arbitration agreement nugatory. Another approach is to simply direct the
tribunal to exercise its procedural discretions in order to ensure a speedy
conclusion. (102) In cases where time is crucial, it can be helpful to include
time limits on the selection of arbitrators by the parties, and to provide that
upon expiry of that time limit the arbitrators will be selected by a relevant
institution. (103)
16. Parties may wish to designate the language of the arbitration when each
party has a different native language.
17. In some cases parties might wish to designate the currency in which the
award will be rendered where currencies fluctuate widely.
18. Parties sometimes consider specifying notice requirements (e.g. is email
acceptable?) although this is typically left to a procedural conference after a
dispute has arisen.
19. Parties might wish to consider express confidentiality requirements as not
all courts take the same view as to the inherently confidential nature of
arbitration. Alternatively, if parties do not wish to be bound by
confidentiality requirements, they should indicate this expressly. (104)
20. Parties may wish to consider what evidentiary methods will be utilised, e.g.
for document production or for appointment of expert witnesses. The
parties might consider whether they are happy to agree to adopt the IBA
Rules of Evidence 2010.
21. Parties might also wish to consider stepped or escalation clauses, i.e., where
they begin with negotiation, then perhaps move to mediation and only move
to arbitration if ADR techniques have been unsuccessful. Care is needed in
drafting to avoid uncertainty and avoid an opportunity to delay the
arbitration unduly by debating the adequacy of the ADR processes.
22. Drafting of arbitration clauses may also need to give thought to the impact of
a dispute on the ongoing performance of the transaction. If a party is not
claiming termination of the underlying contract, in some commercial
scenarios it makes sense to indicate whether each party is otherwise
required to continue its contractual obligations while the dispute is
occurring. In a construction dispute it may also be appropriate to indicate
that the bringing of the arbitration is not itself the basis for a claim for
extension of time or additional costs. (105)
23. page "158"The parties may also wish to consider whether to exclude
recourse against the award where the relevant law entitles them to do so.
Some European systems allow waiver of such rights. If there are features of
the lex arbitri that are not required and which are not mandatory, these
 might be expressly excluded. For example, under the English Arbitration Act
1996, parties can agree to waive their right to appeal on points of law. They
cannot do so in relation to claims of substantive lack of jurisdiction or
serious irregularity. (106)
24. Parties sometimes wish to consider what kind of damages may be awarded
(in some systems punitive damages are not allowed), what attitude there
should be to interest (e.g. at what rate, and whether simple or compound),
costs (e.g. should the loser pay costs to the winner), taxation (e.g. should
damages be net of tax liability or a gross amount), offsets via counterclaims
or set-off rights (e.g. can matters not naturally within the arbitration clause
but due between the parties be offset in relation to matters that are within
the agreement?). Care should be taken when the parties seek to limit a
tribunal's powers, particularly in relation to remedies. They should indicate
whether they are seeking to remove such remedies entirely on an inter-
party basis or whether they are instead allowing for the excluded remedies
to be claimed through court proceedings.
25. There is a need to make sure that form requirements are satisfied, for
example, satisfying Option 1 of the UNCITRAL Model Law if it applies and
Article II of the New York Convention.
26. Other possible elements include provision for interim measures, and a ‘last
best offer procedure’. (107) Some of these are discussed discretely in the
following sections.

3.3.1. Drafting Specialty Clauses

The previous section warned against the problems that may arise when clauses
are drafted afresh, rather than relying on well tested models from international
institutions. Nevertheless, there are a number of specialty scenarios were
specific drafting may be required. The key variations are discussed below.

3.3.1.1. Unilateral or Optional Arbitration Clauses

Sometimes agreements will wish to give the claimant an option as to whether to

arbitrate or litigate. To an individual disputant, the preferred dispute settlement
avenue may only become clear once the details of the dispute are known. This
has encouraged some drafters to use unilateral arbitration clauses, allowing the
page "159" claimant to determine whether the matter will be brought to
arbitration or to the courts. On the one hand, party autonomy would suggest that
this is acceptable, although on the other hand, it might be seen to offend against
procedural equality. These clauses will generally be enforced, although some
courts have intervened on the basis of lack of mutual obligations to arbitrate. (108)
That should not be so with a broad optional clause as a priori, neither party may
know who will be claimant in any dispute so they each benefit equally from the
option. Even where that is not the case, procedural equality must be considered
within the confines of the arbitration as agreed to. If one party wishes to
effectively waive its entitlement to commence arbitration actions by agreeing to
a unilateral clause, this should prima facie be acceptable. When reviewing
attitudes of courts to such clauses, a distinction also needs to be made between
clauses that seek to give only one party any form of dispute resolution rights and
other clauses that simply limit the arbitration option to one party. (109) A clause
that purports to take away one party's rights entirely would more naturally be
vulnerable.

When drafting unilateral clauses, it is useful to express this clearly as a

discretion. To simply say that disputes are resolved either by arbitration or
litigation could allow for arguments that it is void for uncertainty, although
respected courts have not adopted this position. (110)

3.2.1.2. Escalation Clauses

Escalation clauses involve staged dispute settlement procedures. Typically they

might ask the parties to first attempt to negotiate and/or conciliate or mediate a
mutually agreeable solution. In construction disputes, resort must often be first
made to a referee or engineer. In many cases, one of the parties is only entitled to
commence arbitration if such a designated process fails to achieve a successful
outcome. Another variant of an escalation clause is a situation where there is an
obligation to exhaust local remedies before an international dispute can be
progressed. Escalation clauses are sometimes described as integrated dispute
resolution systems. (111) Other descriptions are ‘multi-tiered’ or ‘step clauses’. (112)

page "160"

An advantage of an integrated approach is that it ensures that these steps will

occur without requiring an ad hoc agreement once a dispute is known, which
may be less likely. However, if the parties are unwilling to even agree to
negotiate or mediate at the stage of the dispute, it is less likely that a prior
contractual requirement to do so will bring them to the negotiating table in the
appropriate mindset. One advantage of escalation clauses is that they can help
promote proportionality, particularly in long-term commercial relationships. In
situations such as construction contracts, when there are disputes about small
items, as will inevitably occur, these can be resolved through negotiation or
resort to an adjudicator such as an engineer. When a dispute is large enough or
an accumulation of smaller disputes renders the relationship problematic, it is
appropriate to move to a more formal adjudicatory stage. (113) When considering
the use of escalation clauses, the drafter should also consider the suitability of
the alternative steps on a cost/benefit basis and the impact on the arbitral
process if they are not successful. (114)

While escalation clauses may often be desirable on policy grounds, it needs to be

understood that they are a particularly problematic type of clause from a
drafting point of view as they can too easily be abused or become pathological.
One of the problems of drafting is that they must inevitably use general phrases
and terms about which there can be legitimate disagreement. Transactional
lawyers seeking to draft such clauses have particular challenges as a result.
Because of the inherent difficulty of such clauses, institutions have either
refrained from proposing models or where they do, they have not been fully
successful in avoiding ambiguities. There will also be interpretation issues in
evaluating such clauses, which in turn raise conflict of laws questions as to the
law applicable to the arbitration agreement and/or general approaches to
determining arbitral intent. This has been discussed in sections 3.2.2–3.2.4.

The drafting should ensure that it is clear whether the pre-arbitral stages are
optional or whether they are pre-conditions to the right to go to arbitration.
Tribunals often conclude that the right to arbitration is unconstrained,
notwithstanding a direction to first attempt more conciliatory processes. An
arbitrator considering jurisdiction in such circumstances is evaluating the
importance of the procedural methodology to the determination of arbitral
consent. In the event that the prearbitral steps are seen as pre-conditions, there
needs to be a clearly delineated methodology of determining when one can move
from one form to the next level. It is particularly important that there is minimal
chance to have a separate dispute as to whether such pre-conditions have been
met. The important drafting issue is to consider whether to utilise either
‘mechanical or judgment-based’ page "161" standards to determine when each
stage has appropriately come to an end. (115) If a negotiation stage is to be
included, it is appropriate to carefully draft the obligation to indicate who may
end the negotiations and at what point in time. (116) It would be problematic to
indicate that one stage comes to an end only on agreement of the parties. One
simple method is to include time limits for each stage, after which the dispute
automatically progresses to the next stage. In any case, care must be taken to
ensure that progression from one stage to the next must, rather than may, occur
once specified criteria (such as the expiry of a time limit) are satisfied. (117)
Furthermore, wherever time limits are involved, they should be triggered by
clearly defined and unambiguous events which all parties will understand as
triggering the relevant time limit. For example, the time limits should not be
triggered simply by notice of an issue leading to a dispute, but rather, by a
request for negotiation in relation to that issue. (118) The simplest trigger is to
allow either party to terminate the ADR processes at will and move to arbitration
wherever they wish. This might be permissible in any event if the parties have
agreed to a particular set of mediation rules that include such a right. (119)

Where time periods are utilised to delineate stages, it would be standard to allow
the time to be extended by mutual agreement. More difficult is whether the time
period can be reduced at the request of one party if the other party refuses to act
in good faith, as this again requires qualitative judgment about which there can
be disagreement. A multi-tiered clause also needs to consider whether one party
should be able to speed up the process or jump a level if it is necessary to
prevent serious damage, (120) although this could be handled by interim measures
powers in some cases. Problems may also arise if particular persons or types of
persons are designated in the escalation clause as the ones to engage in
negotiations or mediation. An expansive approach was taken in ICC Case No
9977 where ‘senior management representatives’ were required to attempt to
reach a settlement. While they did not negotiate directly, the arbitrator
concluded that they were in contact with and had control over the relevant
officers. (121) There may also be a need page "162" to consider the impact upon
limitation periods. Under some legal systems the limitation period might be
suspended during ADR processes or it may be in bad faith to invoke such a
provision in such circumstances. (122)

From a procedural perspective, a number of other questions arise in such

circumstances. To the extent that the promise to negotiate, mediate or conciliate
is seen as a condition precedent, how would an arbitrator make factual findings
as to the way these steps were conducted if there is no agreed statement of facts
in that regard? Will those obligations be treated as subject to a duty of good faith,
perhaps requiring consideration of the spirit in which parties engaged in such
proceedings? Complex factual scenarios may include cases where a party simply
refuses to negotiate, where a party offers only narrow timing opportunities for
the required processes that are not convenient to the other party, where they
merely restate their position and refuse to show any willingness to compromise,
when their proposals would appear grossly unreasonable ambit claims to an
independent observer and where they withhold or misrepresent key
information. (123) Supervising legal systems take very different views as to
whether they are willing to second-guess the quality and bona fides of
negotiations. Most will be naturally reluctant to do so. Generally speaking, except
in extreme circumstances, arbitrators would be unlikely to engage in a
qualitative assessment of the approach each party took to a negotiation
requirement. Nevertheless, a number of ICC cases have looked at adequacy on
the facts where judgment-based criteria have been utilised. (124) If a tribunal is to
look at whether good faith negotiations have legitimately come to an end, ICC
Case No 6276 looked to see whether the parties ‘arrive in good faith at the
conviction that they have reached a persistent deadlock’. (125) In some cases they
will merely look to see whether negotiations occurred and not attempt to
consider qualitative aspects. (126) Where the ICC is involved, the Court itself may
be asked to consider whether there is a prima facie case of jurisdiction if there is
an allegation that one of the pre-conditions has not been met. The Court is more
likely to leave it to a tribunal to make such a determination.

Another uncertainty arises if there is a tentative agreement through the ADR

process but there is a dispute as to whether that has been complied with.
Another concern is that some legal systems see negotiation provisions as
unenforceable. (127) page "163" If a negotiation provision could be unenforceable,
thought should be given to separating it from the arbitration agreement in case it
could taint the latter. Mediation agreements have been supported in France (128)
and in Switzerland. (129) Agreements to negotiate have been questioned in
England in Courtney & Fairbairn Ltd v Tolaini Brothers (Hotels) Ltd. (130) US courts
have enforced negotiation and mediation clauses if they are drafted with
sufficient certainty. (131) That can depend solely on the terms of the clause itself
or in some circumstances, may involve disputes about ambiguities within a
selected mediation centre's own processes, leading to arguments that the
agreement is void for uncertainty for that reason alone. This led to
unenforceability in the Australian case of Elizabeth Bay Development Pty Ltd v
Boral Building Services Pty Ltd. (132) In some cases an adjudicator may hold that a
poorly drafted preliminary step may be ignored simply because it is unworkable.
(133) There are also timing questions as to the proper notice for the ADR stages.

Where there is a mandated negotiation or consultation phase, one issue is how

much notice the claimant gives the other party of the dispute. In extreme
circumstances the recipient may argue that insufficient notice by the claimant
has taken away the respondent's right to truly seek an alternative solution. If the
consultation notice is too late, the respondent may argue that it has not been
given its full substantive right to the designated consultation period.

The next issue is what remedy would apply if the preliminary step was
considered to be enforceable but did not take place. Common law courts would
rarely allow for specific performance. (134) The tribunal would be left with a
question as to whether a condition precedent to arbitration that has not been
met prevents arbitral jurisdiction. Another drafting option is therefore to
indicate in the multi-tiered clause what the consequences are of non-compliance
with a particular stage. It would be far preferable to allow for arbitration
residually in all circumstances, perhaps giving the tribunal a discretion as to
costs and damages for breach of the escalation clause. A tribunal might also be
entitled to stay the arbitration and even order specific performance of the ADR
stage in appropriate circumstances. Questions of stay are problematic, however.
In a jurisdictional challenge relating to alleged non-compliance with an
escalation clause, for example, where there is a page "164" preliminary
stipulation to place the dispute before an engineer or mediation, a question is
whether a tribunal can stay the arbitration while the parties redress the failure
to use such a preliminary process. It has been suggested that this may be
possible. While that seems intuitively efficient, the contrary view would be that if
there is jurisdiction, the tribunal should just go ahead. That should flow from the
arbitrator's duty to complete the mandate. The party who justifiably asserts
jurisdiction is also entitled to an expeditious proceeding. Conversely, if the
tribunal took the view that a necessary pre-condition was not performed, then
once the tribunal makes such a determination, it might more properly be seen as
functus officio. Certainly an arbitrator can invite the parties to reappoint and
adopt all pleadings, but it seems inappropriate for an arbitrator to attempt to
effectively bind them to the existing arbitration through a questionably sourced
discretion to stay proceedings.

Martin Hunter has also raised the question of whether it might be better to draft
multi-tier clauses with arbitration as the first option. (135) While ADR theory
would naturally suggest that the process start with the least adversarial and
move up the scale, the Hunter proposal has the advantage of having a clear
agreement to arbitrate without debates about pre-conditions, leaving it to the
tribunal to utilise alternative processes on a discretionary basis. This will also
have an advantage of crystallising time limits and setting dates for calculations of
remedies, interest and costs. It will also crystallise legal privilege where that
applies. It also overcomes the conceptual challenge as to general stay of
proceedings powers. Having early rounds of arbitral submissions, particularly in
the context of full disclosure of arguments and documents, can also help in a
proper evaluation of strengths and weaknesses of each party's case, an
important step on the way to promotion of successful negotiations. Conversely, if
the mediation is left too late in the arbitral process much of its time saving
benefits will be lost. (136) Care would still need to be taken in drafting an
arbitration first clause to the extent that it purports to bind the parties to
alternative processes once the arbitration has commenced. The more the parties
seek to stipulate such requirements, the more that arguments can still ensue to
the effect that the tribunal's jurisdiction is constrained and that the arbitration
cannot proceed while an alternative mandated process must be completed.
Hence it may be preferable to draft a clause giving a broad discretion to the
tribunal and select an arbitrator who is adept at such processes or at least in
their management. (137) The problem is that if the clauses do not make any
commitment, page "165" then they may in practice have little value if the parties
are unwilling to engage in ADR processes. Conversely, the more they stipulate
requirements, the more they can be misused to disrupt the arbitral process.
Because of this, giving a tribunal discretion in terms of timing and then as to
costs may be the best compromise.

Attention also needs to be given to confidentiality over the pre-arbitral processes

and what, if anything, can be used at a later stage within arbitral proceedings if
they ensue. The broad drafting choice is between relying on broad categories of
privilege or instead listing specific items. (138) Article 10 of the UNCITRAL Model
Law on International Commercial Conciliation adopts a list approach.

Courts may also look at compulsory mediation obligations in the context of

interpreting their responses to arbitration agreements where stay applications
have been made. For example, Section 9(2) of the English Arbitration Act 1996
allows for a stay of proceedings in the face of an arbitration agreement even if
the latter may only apply after the exhaustion of other dispute resolution
procedures. Similarly, the Australian case of Aiton v Transfield (139) considered
that the authority to enforce arbitration agreements allows a court to enforce a
mediation pre-condition to arbitration. The case also supported the idea of an
agreement to negotiate.

A similar set of issues arises in investment arbitrations where there is an

obligation to notify the other party as to the terms of a dispute so as to afford an
opportunity for negotiations. There is a need to consider what is required to
satisfy the consultation norm and when challenges about the consultation
process can be used to prevent the arbitration proceeding. For example, a debate
may arise as to whether the case, as brought to arbitration, is sufficiently similar
to the dispute as originally notified. (140) If the notice of arbitration contradicts or
varies any of the claims at the consultation stage, the respondent might seek to
undermine one by reference to the other and in extreme cases may even argue
that the arbitration is sufficiently different in focus to require a distinct
consultation period. (141) A tribunal in Generation Ukraine, Inc v Ukraine (142) held
that there need not be ‘a precise congruity in the investor's articulation of its
grievances …’. (143) A distinction may be made between situations where new
facts are alleged and situations where new claims or new legal characterisations
occur in relation to facts which had been well notified. (144) There may also be
issues as to the appropriate service of such a request, page "166" particularly
when it is unclear which part of the bureaucracy ought to benefit from the
consultation entitlement. In investment cases, there may also be questions of
interpretation if an investment treaty contains a most favoured nation clause
raising the question as to whether this applies to procedural as well as
substantive issues.
3.3.1.3. Multi-party and multiple claims

As noted above, in some multi-party and multiple claim scenarios it would be

most efficient to ensure that all relevant persons and issues can be brought
within one arbitral proceeding. Where multiple parties are concerned, it is
desirable to direct that subcontracts or related contracts include identical
arbitration clauses and cross reference the primary agreement. The primary
contract might also indicate that it is a condition of the right to subcontract that
such an arbitration clause will be utilised. There is no one boilerplate model
clause that would cover all situations. Nevertheless, there are some useful
examples that can be a guide to optimal practice. (145) Where multi-parties are
concerned, there is also a need to ensure that there are no challenges to
composition of the tribunal on the basis that the joined party has not had equal
rights to appointment. This is discussed further in section 7.5.

Where multiple contracts are concerned, it is again important to have identical

dispute resolution mechanisms in various related contracts. A failure to do so
will make it difficult if not impossible to employ consolidation powers to bring all
the disputes under one umbrella. Another approach is to have a separate general
dispute resolution agreement that refers to each individual contract. The optimal
terms will depend in part on the consolidation powers in the relevant lex arbitri
or arbitral rules and should at least ensure that gateway requirements are
satisfied. Where matters may be consolidated, it is appropriate to consider the
same substantive law and language to ensure smooth operation of any
consolidated proceedings. (146) Particular care needs to be taken when there are
both multi-party and multi-contract scenarios where the ultimate aim is to have
single proceedings covering all relevant matters. These issues are canvassed
more fully in Chapter 7.

3.4. Sources of Procedural Laws

The following sections aim to outline the various sources of procedural law as
applicable to international arbitration. It is important to understand that there
are a range of sources and that these need to be identified and integrated. Where
they appear to conflict, there is a need to identify a hierarchy. That is not
straightforward as identification and analysis of the relevant sources can be
affected by theoretical page "167" perspectives about arbitration itself.
Hierarchy is not simply a technical matter but may also be a guide to best
practice. For example, while lex arbitri indicate that party autonomy takes
precedence over arbitral discretion, Herrmann makes the apt observation that in
practice this can best result from a dialectic relationship where a consultation
occurs and the views of the parties are solicited at all appropriate stages. (147)

The following sections look at arbitral laws and rules. In terms of sources of
procedural mandates, it is equally important to consider the parties' arbitration
agreement itself, which will typically provide prescriptions as to procedural
requirements and at times proscriptions as to procedures that the parties have
rejected. Arbitration agreements and their interpretation were considered in
section 3.2. Once again there is a need to consider how express or implied
procedural agreements of the parties are integrated with the dictates of
applicable procedural laws. This essentially involves a consideration of the
nature and ambit of the lex arbitri and any mandatory procedural laws, discussed
in sections 3.6 and 3.7 below. Attention is also given to other general principles
that may impact upon the conduct of an arbitration, for example, international
law and human rights norms that purport to articulate procedural justice.

3.5. The Seat or Place of Arbitration

The law of the Seat or place of arbitration generally plays a central role in
arbitral proceedings. Statutes, rules, cases and commentaries use the phrase
‘place of arbitration’ synonymously with the notion of the ‘Seat’. The arbitration
law of that place will, in most cases, form the lex arbitri. An award that does not
comply with those norms can be annulled or enforcement can be refused on that
basis. (148) Because lex arbitri typically identify the supervisory court that has
jurisdiction, tribunals ought in principle to abide by such courts' determinations
unless their injunctions are clearly abusive, although there can at times be timing
questions as to which forum should deliberate first on a particular issue. (149)
Aside from that timing issue, courts in the Seat will not only have supervisory
functions in relation to jurisdictional determinations and annulment, but may
also be asked to make tribunal appointments, may at times hear challenges as to
independence or impartiality and may support the process in relation to interim
measures or the taking of evidence. (150) Because an award is generally deemed to
be made in the Seat, this will page "168" not only identify the relevant law for
enforcement purposes but also determine whether it is truly domestic or foreign
in the context of the New York Convention's requirements.

The concept of the Seat thus identifies the ‘external’ controlling law (151) and also
indicates the key supervisory court. The Seat of arbitration can be seen as the
legal or juridical framework of the arbitration. (152) The Seat ‘is less a matter of
real geography than a link to the legal order of the place whose curial law will
govern many aspects of the proceedings’. (153) In that sense, selection of the Seat
is simply selection of the jurisdictional link for the legal imprimatur to arbitrate
and ultimately to best shape an enforceable award. Because of this, Gabrielle
Kaufmann-Kohler describes the Seat of arbitration as a legal fiction. (154) The lex
arbitri must also be considered in relation to other choices made by the parties.
Most countries' arbitration laws allow significant scope for party autonomy.
Many countries' lex arbitri also allow for opt-out provisions.

Fouchard, Gaillard and Goldman suggest that there is a trend against necessarily
applying the law of the Seat to govern arbitral procedure. (155) While there has
been a long debate as to the possibility for truly delocalised arbitration, (156) it is
now accepted that arbitration is in practice at least something of a hybrid,
requiring an appropriate procedural framework in the Seat and appropriate
provisions to recognise arbitration agreements and support enforcement of
awards.

The role of the Seat is not as relevant in ICSID-based investment arbitration.

While Article 2 of the ICSID Convention nominates Washington DC as the Seat,
arbitrations are governed solely by the ICSID Convention and Arbitration Rules.
(157) Annulment proceedings do not use national courts but are instead conducted

before specially constituted ICSID annulment committees. (158) Enforcement does

not proceed under the New York Convention but instead pursuant to page "169"
Article 54(1) of the ICSID Convention which demands that the contracting State
treats such an award as if it were a final judgment of a court in that State.

3.5.1. Seat Selection

As noted, the norm is for the law of arbitration in the Seat to be the lex arbitri.
Hence in selecting a Seat, parties are normally selecting the framework
procedural model as well as the supervisory court. Section 3.8 below deals with
the possible but undesirable scenario where a procedural law other than the Seat
is selected. If the parties have not selected a Seat but have selected arbitral rules,
these will provide a mechanism for establishing the Seat. (159) Where ad hoc rules
are concerned, this decision will typically be delegated to the tribunal. Where
institutional rules are concerned, the institution itself may designate the Seat,
although some such rules still delegate the power to the tribunal. The UNCITRAL
Model Law and Rules also leave selection of the Seat to the tribunal. ICC and
Stockholm Chamber of Commerce Rules provide for the institutional
administrator to make the determination. Another possibility is for the rules to
make a provisional determination and the administrator to confirm this after
hearing from the parties (160) or a provisional determination is made by the
institution which is then to be confirmed or revised by the tribunal. (161)

In some cases it will be unclear as to whether the parties have in fact designated
a Seat. Where parties expressly refer to a particular arbitration statute, most
would see this as designating a Seat, although this is less valid for jurisdictions
that readily accept arbitration in the Seat under a foreign procedural law. (162)
This question will also typically arise in relation to pathological clauses, such as
an agreement to arbitrate under the ‘Chamber of Commerce Rules, Geneva’. Is
this to be taken to mean an agreement to ICC Rules with Geneva as the Seat, or a
reference to the Swiss Arbitration Rules or is it void for uncertainty? Selecting
‘ICC Paris’ is also ambiguous as the reference to Paris may be to the proposed
Seat or may simply be identification of the location of the ICC itself. In such
circumstances, principles of interpretation, including designation of applicable
law will also be relevant. These were discussed in section 3.2.4. It was argued
that an effete utile approach is to be preferred when the parties to the dispute
have adopted the clause in issue.

In selecting a Seat, a number of factors are worthy of consideration. Some of the

factors that should be considered include respect for arbitration at that place,
elements of the lex arbitri, i.e., does the law provide for all appropriate powers,
adoption of the New York Convention to maximise enforceability, geographical
location, ease of transport and suitable facilities, the role and attitude of the local
page "170" courts in a supervisory and assistance sense, and perhaps the
availability of high quality arbitrators of that nationality. A neutral venue is
invariably selected but this need not be so. The party in the stronger negotiating
position might seek to select its own country. Parties may even wish to
strategically select the opponent's home country as the Seat if that would help in
gaining access to evidence and witnesses through subpoena powers in
jurisdictions where such powers are statutorily provided to tribunals. Similarly,
this may aid in obtaining anticipated interim measures over assets.

The ASA has recommended:

– positive attitude of the Seat towards international arbitration

– political stability of the Seat country
– political and cultural neutrality
– reliability and neutrality of the legal system
– legal considerations: laws likely to apply to various questions
– tax considerations: no VAT or withholding tax on arbitrators' fees
– quality of local arbitrators and local Bar (out of whom the presiding arbitrator,
and local counsel to the extent required, may be selected)
– non-interventionist courts
– limited remedies against awards
– language considerations
– convenient geographical location
– telecommunication, airports, hotels, restaurants, banks that work well. (163)

William Park suggests that the best place for arbitration would normally be a
country where the judiciary will safeguard the integrity of the procedure, but not
unduly interfere with honest mistakes of law or fact. (164) Considerations of the
respect for arbitration and the role of the courts requires actual analysis of the
practice in the particular jurisdiction as well as the way its lex arbitri is framed.
Most horror stories arise where courts unduly interfere with the arbitral process
in ways that would not be evident from their arbitral statutes.

Holtzmann and Neuhaus noted that a suggestion at the Model Law drafting stage
to have a requirement of a genuine link between the Seat and the actual
proceedings was rejected. There is no abuse in selecting a technical Seat to best
support the arbitral process. (165) Some scholars suggest that it may be prudent to
page "171" perform at least some functions within the Seat to prevent arguments
that the choice is fictitious and otherwise inoperative. (166) Few supervisory
courts would take this view however. This is quite different to selecting a
registered office of a corporate shell as a conduit to perhaps gain the substantive
benefits in a particular bilateral investment treaty. In the latter case there is a
separate question as the arbitral rights only come from the treaty and there may
be a jurisdictional question as to whether it does indeed apply. Where selection
of the Seat is concerned, a court is only likely to be concerned if it thought that
selection aimed to avoid some otherwise applicable mandatory norm. In such a
scenario, it should not impact on the analysis if some of the hearings were
conducted in the selected Seat.

3.5.2. Changing the Seat

The accepted view is that parties can change the Seat during the currency of the
arbitration proceedings, although this would be rare and could raise problems as
there may be questions as to the impact on previous events. Some legal systems
would take the view that a transfer has retroactive effect. (167) There may be
problems as to which court would have supervisory jurisdiction and what time
limits may apply to challenges. In addition to the power of the parties to transfer
the Seat, a separate question is whether an appointing body, an institution or the
tribunal itself may transfer the Seat where problems arise with the first selected
jurisdiction. Where there are significant changes in circumstances, it is at least
arguable that any party agreement as to the Seat was impliedly conditional on
there being no major changes. That would be a matter of interpretation and
would be a difficult argument to pursue except in extreme circumstances.

An example of a motivating circumstance may be a court in the first designated

Seat rendering an unreasonable injunction against the continuance of the
arbitration. There is no fundamental need to change the Seat in such
circumstances given the ability to hold hearings elsewhere. If that jurisdiction
will also be the natural enforcement country, there is little practical value in
avoiding a potential annulment decision by transferring the Seat given that such
a decision is unlikely to have a positive or a negative impact on enforcement in
due course.

3.5.3. Seat versus Place of Hearings

It is clear that while parties ought to select the Seat and in turn an applicable lex
arbitri, they are not required to hold most or indeed any of the hearings at that
Seat. While it is well accepted that hearings may be conducted at more
convenient locations than the Seat, a tribunal concerned to truly deal equally
with the parties page "172" must be very careful when making such a selection
absent party consent. Any alternative place to the Seat is unlikely to have the
same comparative cost or logistical utility for both parties. The tribunal should
ensure that such a selection at least decreases total transaction costs. The
situation is different where there is a physical barrier to conducting hearings in
the Seat. (168)

3.6. The Lex Arbitri

Countries aiming to support arbitration will enact a statute seeking to ensure an

appropriate framework. The UNCITRAL Model Law on International Commercial
Arbitration (169) is important as at the time of publication, the Model Law has
been adopted in over sixty-five jurisdictions either in toto or through legislation
largely based on the Model Law. (170) Key common law jurisdictions that have not
adopted the UNCITRAL Model Law include England and the US, while in civilian
jurisdictions key countries include Belgium, France, The Netherlands, Sweden
and Switzerland. Some countries maintain a distinct statute, although more often
than not, this will be drafted with an eye to the Model Law. (171) Some civilian
jurisdictions include their arbitration laws as part of a general Code of Civil
Procedure as opposed to housing them in a separate arbitral statute. (172)
Particular elements of the Model Law are dealt with under the relevant topics
throughout this book. In terms of its structure and general thrust, the Model Law
contains Articles dealing with recognition and enforcement of arbitration
agreements, appointment of arbitrators and challenges to arbitrator
appointment, separability and competence of arbitrators, provisional measures,
procedure and evidence, determining applicable law, arbitral awards,
challenging awards and recognition and enforcement of the awards themselves.
The UNCITRAL Secretariat commenced a study for review and possible
modification of the Model Law in 1999. (173) The Model Law was revised in 2006.
Because UNCITRAL ultimately considers reform options through
intergovernmental discussions and negotiations, it will tend to take time and be
conservative in terms of its coverage although that is by no means uniformly so.
Amendments in 2006 included the addition of interpretative principles, (174)
options for reform of the writing requirement, (175) controversial provisions on
interim page "173" measures, (176) and mechanisms for the recognition of awards.
(177) It has also been observed that the style of drafting in the 2006 amendments

is somewhat different to the concise drafting style of the Model Law itself. (178) A
number of other matters that had been recommended by some participants were
not adopted, including a choice of law provision governing the arbitration
agreement, clarifications of the competence-competence principle and
elaboration of the grounds for holding arbitration agreements and awards to be
invalid.

National statutes dealing with arbitration have tended to separate out

international and domestic arbitrations for differing treatment. More recently,
however, some jurisdictions have combined the two, albeit with some special
provisions relating to each. This is because some jurisdictions have sought to
bring their domestic arbitration more into line with the efficiency and consent
oriented norms of international arbitration. (179) Historically, the key difference
was that there was thought to be more rationale for judicial control of domestic
arbitrations both in a supervisory sense and also in providing assistance, given
that the parties would be naturally within the jurisdiction of the courts.

3.6.1. Elements of the Lex Arbitri

Practitioners will need to give careful consideration to the features of various

arbitral statutes in a range of circumstances. The first is where the parties are
considering selection of a Seat. It is important to ensure at the very least that the
Seat has an appropriate arbitral model. A second situation is to consider whether
a particular statute is sufficiently comprehensive where it has already been
selected in the agreement or is otherwise selected for the benefit of parties. Here
practitioners not only need to be aware of the nature and ambit of the law but
also be able to consider what additions, if any, may be needed to ensure fair and
efficient proceedings.

Lex arbitri may deal with issues directly or provide a conflict of laws rule to deal
with certain scenarios. In some cases, lex arbitri may simply fail to deal with a
particular issue. In other cases, they may have a unique feature that is
contentious. A well constructed lex arbitri needs to do the following things:
(1) Ensure that the courts in that jurisdiction will recognise and enforce
arbitration agreements.
(2) Provide a framework for the appointment of arbitrators and stipulate the
grounds and processes by which arbitrators may be challenged. page "174"
(3) Confirm that arbitrators have jurisdiction to decide on their own
jurisdiction (competence-competence).
(4) Enshrine the separability doctrine to ensure that an arbitration agreement
remains workable even in the face of a void or voidable contract.
(5) Articulate mandatory principles of procedure such as fair treatment and an
equal opportunity to be heard and otherwise give strong support to party
autonomy.
(6) Provide a broad residual discretion to arbitrators on procedural matters.
(7) Give some guidance as to matters not otherwise determined, including
selection of a Seat if that has not occurred, language of the arbitration,
modes of evidence and methodology of determining applicable law.
(8) Indicate the requirements of an arbitral award.
(9) Provide a mechanism for rectification and perhaps interpretation of
awards.
(10) Provide a mechanism and limited grounds for setting aside or challenging
awards.
(11) Provide for recognition and enforcement of foreign arbitral awards and
also indicate the grounds upon which enforcement may be blocked.

A well-drafted statute would also clearly indicate its jurisdictional coverage and
the minimum requirements for a valid arbitration agreement. A well-drafted lex
arbitri might also make it clear that courts have no jurisdiction in arbitration
save where expressly granted through the statute. More contentiously, a well-
developed lex arbitri may also provide for court support in the way of interim or
provisional measures and court support for an arbitral tribunal seeking to
exercise such powers.

The following sections elaborate on some of the key gateway provisions in lex
arbitri. Before looking to the key elements, some general comments are first
made about drafting and regulatory style. While it is relatively easy to state the
key requirements, much can depend on the way these are drafted and certainly
on the way these are interpreted by tribunals and domestic courts where
disputes arise before either. In terms of drafting, there is a difference between a
model which presumes the validity of arbitration agreements and places the
burden on anyone challenging and the converse. An appropriately drafted law
would provide for such presumption as to validity. Even then, much can depend
on both the criteria and standard of proof to be applied by a tribunal in hearing
any challenges. At times, a proper interpretation of provisions in lex arbitri will
need to consider implied terms or at least purposive or contextual
interpretation. Because there is a very significant difference in specificity
between arbitral laws such as the English Arbitration Act on the one hand and
the French statute on the other, merely selecting London or Paris as the Seat
without any other procedural agreements will have significant differences in
terms of the detail of the procedural model. (180)

page "175"

Most countries have sought to reform their arbitration laws to better facilitate
the process of international arbitration. Historically, the most significant trend
has been the acknowledgment of party autonomy as governments become more
comfortable with respect for arbitration and the quality of its processes. Some
developments are more contentious, at times being motivated by a desire to
make a particular Seat more attractive to potential users. An example of a failed
policy in that regard was the Belgian statute that for a time, limited court
supervision. That proved to be unpopular and Belgium's current statutory
regime is more in line with that of other countries. The example highlighted the
fact that where governments try and reform their laws so as to attract arbitral
business, there is a question as to whether the best strategy would be to design a
unique model that might arguably be world's best practice or might capture a
niche market or instead adopt a familiar model such as the Model Law so that
prospective users do not feel they need to learn about a novel system. In this
context, policymakers should be very careful in how they seek to predict the
particular desires of commercial entities, and their legal advisers. As noted
above, because there is a need to balance fairness and efficiency and rely to a
significant degree on discretionary behaviour of hopefully expert arbitrators, any
attempt to identify discrete areas that might attract significant business to a
venue are unlikely to be accurate as it will involve some a priori setting of
contentious trade-offs. Even a successful reform of this nature is likely to be of
short-term benefit in terms of attracting business, as if successful, it would soon
be copied by other jurisdictions, although that is of course no reason not to adopt
a meritorious reform.

3.6.1.1. ‘International’ Dispute

Where lex arbitri distinguish between international and domestic matters,

gateway criteria are needed to distinguish between the two. These can look at
the features of the transaction, the nationality or domicile of the parties or the
separate location of the Seat. Article 1(3) of the UNCITRAL Model Law outlines
four alternative criteria of internationality: the parties have their places of
business in different States at the conclusion of their agreement; the place of
arbitration is outside the State in which the parties have their places of business;
any place where a substantial part of the obligation is to be performed or in
which the subject matter of the dispute is most closely connected is outside the
State in which the parties have their places of business; or the parties have
expressly agreed that the subject matter of their arbitration agreement relates to
more than one country. The article further indicates that if a party does not have
a place of business, reference is to the party's habitual residence. If a party has
more than one place of business, the relevant place is that which has the closest
relationship to the arbitration agreement. (181) A number of national systems that
have not adopted the Model Law are more restrictive. (182)
page "176"

As noted above, while a distinction is rightly drawn between domestic and

international arbitration, a number of jurisdictions are moving towards a
uniform lex arbitri for both. (183) There will then be a question as to the extent to
which cases discussing domestic matters will inform international commercial
arbitration practice.

3.6.1.2. ‘Commercial’ or ‘Investment’ Subject Matter

The UNCITRAL Model Law also requires that the dispute be commercial. This is
defined broadly in footnote 2 to Article 1(1) to cover those transactions that
would fit within the plain meaning of that term in any event.

Where investment arbitration is concerned, an important issue is the definition

of ‘investment’. This may vary from treaty to treaty. Historically, such treaties
were drafted broadly, typically at the inspiration of Western States seeking to
provide the maximum protection to their citizen investors. Over time, as such
norms have operated both ways through such developments as NAFTA,
participants and observers have noted the potential for very broad and
contentious arguments on jurisdiction where the widest definitions have been
utilised. Most importantly and often overlooked, is the fact that just as with a
private arbitration agreement, a key aspect is the connecting words used to
describe the link between the dispute and the investment. Thus a stipulation in a
treaty that disputes may be brought arising ‘out of or in relation to’ an
investment, means that the disputes are not necessarily investment disputes per
se. This allows for a very broad potential ambit, perhaps even going beyond that
subjectively intended by the treaty negotiators. For example, a contentious
question arises if an import barrier or discriminatory trade norm might be
alleged to interfere with the ‘asset’ of market share or ‘relate’ to a particular
investment being a manufacturing process underlying the trade transaction. (184)
The scope of the arbitration agreement may then depend on the approach to
interpretation that a tribunal takes to the particular treaty instrument.

3.6.1.3. Agreement to Arbitrate

Because arbitration is consent based, there must be an agreement to arbitrate in

order to found a tribunal's jurisdiction. Article 7 of the UNCITRAL Model Law
defines an ‘arbitration agreement’ as ‘an agreement by the parties to submit to
arbitration all or certain disputes which have arisen or which may arise between
page "177" them in respect of a defined legal relationship, whether contractual
or not’. Other lex arbitri would be worded with similar generality.

There are a range of scenarios where there is a dispute as to whether there has
in fact been an agreement. One party may allege that no agreement was ever
consummated. The party may allege that the agreement was induced by fraud.
Arbitration may be subject to a battle of the forms, where parties expressly
agreed to certain factors but the purported acceptance referred to general
conditions containing matters inconsistent with an arbitration clause. Depending
on the applicable law, that purported acceptance could in fact constitute a
counter offer. In other circumstances there could be disputes as to whether the
party that agreed did so on their own account or as agent for another. A non-
signatory may be alleged to have agreed by implication, simply through its status
in a group of companies or as an alter ego. There may be a question as to
whether a range of contracts, only one of which contains an express arbitration
clause, nevertheless expressly or impliedly were all agreed to be subject to that
clause. There may be a question as to whether a subsequent agreement
overrides an earlier consent to arbitration, whether expressly or as an implied
waiver. The latter raises a contentious aspect of the notion of agreement, namely,
whether agreement can be positively presumed on the basis of waiver or
estoppel. Civilian legal systems come to similar conclusions through concepts
such as good faith and venire contra factum proprium. (185)

As discussed in sections 3.2.2–3.2.4, questions as above will entail consideration

of the law applicable to the arbitration agreement and the proper approach to
interpretation. As noted in those sections, there is much debate and potential
discretion underlying these questions. In the context of this book, there are also
important evidentiary and procedural issues that arise. The key procedural
questions include when decisions are to be made as to whether there is an
agreement to arbitrate, who has the burden of proof on the issue and by what
evidence will the tribunal make a determination? Where the latter is concerned,
this may be caught up with questions of applicable law, for example, whether
common law restrictions on the use of external evidence to interpret written
agreements may be applicable or instead whether all circumstances will be
analysed to determined the true intent of the parties.

As to timing, the complexities of these legal and evidentiary questions and the
distinct possibility that in some cases this will be bound up in a general review
on the merits, may mean that the determination may not easily be bifurcated
even though it would invariably be desirable to resolve any jurisdictional
challenges from the outset. Nevertheless, when factual disputes about the very
presence of an arbitration agreement are caught up in related factual disputes
about the contract as a whole, it will often be the case that a complete hearing
will be needed before the tribunal will be in a position to determine whether
there was in fact an agreement to arbitrate.

page "178"

3.6.1.4. Defined Legal Relationship

Article 7 of the Model Law also requires that the agreement be in respect of a
‘defined legal relationship’. Little turns on this in practice, as it would be natural
to take an expansive view and consider any circumstance where a remedy could
apply between one person and another as itself constituting the requisite
relationship. (186)

3.6.1.5. Requirement for a ‘Dispute’

There has been some debate as to whether there must be a dispute. Obviously
there must be a matter that the parties disagree upon otherwise whatever is
claimed will be subject to a consent agreement. While some older cases call for a
distinct consideration as to whether there is in fact a ‘dispute’, the better view is
that this is now presumed whenever there is not a complete acceptance of all
claimant's contentions.

3.6.1.6. Writing

Section 3.2.10 above dealt with questions of form and proof of arbitration
agreements. As noted, some lex arbitri contain express writing requirements.
These vary from quite restrictive norms calling for the agreement itself to be in
writing, to broader provisions simply calling for some written evidence of what
might otherwise be an oral agreement. The 1985 version of the UNCITRAL Model
Law requires arbitration agreements to be in writing. Article 7(2) defines that as:

An agreement is in writing if it is contained in documents signed by the parties

or in an exchange of letters, telex, telegrams or other means of
telecommunication which provide a record of the agreement, or in an exchange
of statements of claim in defence in which the existence of an agreement is
alleged by one party and not denied by another. The reference in a contract to a
document containing an arbitration clause constitutes an arbitration agreement
provided that the contract is in writing and the reference is such as to make that
clause part of the contract.

This is a more expansive version of the writing requirement than contained in

the earlier New York Convention which defines an ‘agreement in writing’ as a
term that ‘shall include an arbitral clause in a contract or an arbitration
agreement, signed by the parties or contained in an exchange of letters or
telegrams’. (187) The 2006 version of the Model Law relaxed the writing
requirement and indeed provided those adopting it with the choice between
retaining a more expansive version in Option I or dispensing with it entirely in
Option II. Nevertheless, being simply a model law, it is important to identify
which countries still operate under page "179" the 1985 version and which have
adopted the new provisions and in what form. The bulk of the countries
operating under the Model Law are still doing so under the 1985 version.

Where the updated version is concerned, Option I further defines an arbitration

agreement by indicating that it ‘may be in the form of an arbitration clause in a
contract or in the form of a separate agreement’. Article 7 goes on to stipulate as
follows:

(2) The arbitration agreement shall be in writing.

(3) An arbitration agreement is in writing if its content is recorded in any form,
whether or not the arbitration agreement or contract has been concluded
orally, by conduct, or by other means.
(4) The requirement that an arbitration agreement be in writing is met by an
electronic communication if the information contained therein is accessible
 so as to be useable for subsequent reference; ‘electronic communication’
means any communication that the parties make by means of data
messages; ‘data message’ means information generated, sent, received or
stored by electronic, magnetic, optical or similar means, including, but not
limited to, electronic data interchange (EDI), electronic mail, telegram, telex
or telecopy.
(5) Furthermore, an arbitration agreement is in writing if it is contained in an
exchange of statements of claim and defence in which the existence of an
agreement is alleged by one party and not denied by another.
(6) The reference in a contract to any document containing an arbitration clause
constitutes an arbitration agreement in writing, provided that the reference
is such as to make that clause part of the contract.

Where sub-paragraph (6) refers to incorporation by reference, the qualification

that this occurs where the reference makes the clause part of the contract draws
attention to the applicable contract law and whether that would be the valid
conclusion under that law. (188) Option II simply does not call for any writing
requirement and hence does not need to stipulate when this is satisfied.

In addition to their express terms, these stipulations raise conceptual questions

as to whether they are jurisdictional gateways or are merely evidentiary norms.
There are differences in view as to the mandatory nature of the writing
requirement and how strictly it should be applied in practice. One option in
answering this question is to look to the purpose of the provisions. The key aim
of such stipulations is simply to provide a sufficient evidentiary record of an
agreement to waive litigation rights. Hence to some, the better view is that there
is only a requirement that an agreement be evidenced and that an oral
agreement appropriately evidenced will suffice. (189) The contrary view is
presented by Poudret and page "180" Besson. (190) One can naturally take a
jurisdictional perspective and argue that the Model Law and similar provisions
only purport to apply when pre-conditions are met, including the writing
requirement. Because there is no clearly correct position, there is also significant
variance between tribunals and supervisory courts on this question and also as
to whether these provisions are applied strictly or whether substantial
compliance is accepted even where required for jurisdictional purposes.
Differences in view are to be expected given the number of circumstances that
have arisen over the years where the facts do not strictly comply with the norms
in the lex arbitri but are sufficiently clear to show on balance that the parties
truly preferred arbitration.

Where enforceability is concerned, there may still be a need to consider the more
restrictive writing requirements under the New York Convention, although
UNCITRAL adopted a recommendation on its interpretation to the effect that the
circumstances referred to in Article II (2) should not be seen as exhaustive. (191)
Such a recommendation is of course not binding on an enforcement court,
although it is certainly the preferred view from a policy perspective.

3.6.1.7. Finality and Appeal Mechanisms

One key aspect of arbitration is that it leads to a binding and final award. (192)
Save for the ability to seek annulment or block enforcement, the decision of the
tribunal is generally the last word on the merits. Nevertheless, some laws
provide for limited appeal rights. An example is Section 69 of the English
Arbitration Act 1996 which allows for limited rights of appeal on points of law to
English courts. This is not a mandatory provision and can be excluded.

In addition to lex arbitri that establish some appellate mechanism, the parties
themselves could establish one if they wish. Hence they could agree to a
preliminary determination by one adjudicator, and expressly allow appeal rights
to a separate body on terms as determined in the arbitration agreement. While
parties could conceivably do so, there could be clear challenges with the drafting
and unintended and undesirable consequences. If parties did attempt to
establish an appellate mechanism through their own drafting as opposed to
selecting a Seat with limited appeal rights, there are a number of pitfalls. One
approach would be to page "181" agree to a preliminary determination by one
adjudicator and expressly allow appeal rights to a separate private body or
person on terms determined in the arbitration agreement. The first question
would be whether the first stage could even give rise to an award and what, if
any, enforceability rights would arise as a result. If it is capable of appeal, it may
not be seen as final and binding, at least until the time for appeal has expired. A
second possibility is that a party might seek to bypass the first step and move
directly to the ultimate body, seeking a final and binding award at the earliest
opportunity. In such an event, its opponent could argue that the first stage was a
pre-condition to the latter. That would depend upon the drafting. Setting up a
private mechanism in this way may also lead to various other gaps, such as the
appropriate standard of review by the appellate body or whether it undertakes a
de novo hearing and the ambit of the matters that could be subject to appeal.

Some practitioners have also tried to establish rights of appeal to domestic

courts under their arbitration agreements. Some would support such clauses as
being simply a question of supporting party autonomy. Others would raise
concerns as to the inability of individuals to create court jurisdiction that
otherwise does not exist. The better view is that while the parties can adopt their
own appeal mechanisms, they have no rights to bind national courts to such a
function, absent government agreement to allow them to do so. Recent attempts
to establish appellate jurisdiction in US courts within arbitration agreements has
failed in a number of instances. (193) At the extreme, an arbitration agreement
invalidly attempting to establish enhanced appeal rights may be asserted to be
pathological on the basis that it was conditional on an impossible element. (194)

3.7. Mandatory Procedural Laws

While the starting presumption in arbitration is that party autonomy prevails

and hence the parties can set up whatever procedural model they wish, this is
not without constraint. The recognition of a hybrid theory of arbitration must
contemplate that the parties' agreements must be within the range permitted by
whatever page "182" laws support arbitral jurisdiction per se. One potential limit
on party autonomy or tribunal power is where either seeks to ignore a
mandatory procedural law. It is then necessary to consider exactly what is meant
by a mandatory procedural law and which jurisdiction's laws are relevant in that
context. To some commentators, the question of the applicability of mandatory
substantive laws is highly contentious. (195) The same is not true with mandatory
procedural laws, at least those emanating from the Seat, although there are still
circumstances where this may also be debated as discussed below.

3.7.1. Mandatory Procedural Laws of the Seat

It would generally be accepted that mandatory procedural laws of the Seat apply.
Because it is accepted that the lex arbitri emanates from the Seat and is
controlling, any mandatory provisions of that law which would impose limits on
party autonomy and on the tribunal ought to be accepted as this is the very
source of the tribunal's competence. This may be open to debate from those
proposing a delocalised view of arbitration, as absent party selection of such
norms, there is less justification for their application. A further uncertainty arises
in the rare and undesirable case that parties select a Seat but direct that other
procedural laws are to apply as lex arbitri. (196)

While most would accept that there are always some mandatory procedural
norms that constrain party autonomy and arbitral discretion, in some cases it is
hard to identify whether norms are by nature mandatory. In resolving the
question as to potential applicability, it is important to remember the nature of a
mandatory norm. It is simply one where the relevant legislature wishes the norm
to apply regardless of the will of the parties or regardless of any discretion
otherwise given to an arbitrator. If the governing arbitral law is clear, that should
be the end of the matter, either by indicating that a provision is mandatory or
conversely indicating that it may be varied. However, in many cases arbitration
laws are silent on this issue, leaving it to a question of interpretation. Some
procedural norms would readily be seen as mandatory regardless of whether
they are described as such or not. Hence they must be applied by tribunals
regardless of any contrary wishes of the parties. Where the UNCITRAL Model
Law is concerned, the key mandatory norm is Article 18, which requires equality
of treatment between the parties and an opportunity for each to present its case.
Article 19, the broad procedural provision, makes it clear that it is subject to
Article 18. Hence, neither the parties nor the tribunal can derogate from the
principles of Article 18. The principles contained in Model Law Article 18 are
found in virtually all arbitral statutes and rules. (197)

page "183"

Some lex arbitri, such as the English Arbitration Act 1996 expressly indicate
which provisions are mandatory. By selecting such a Seat, the parties have bound
themselves to those norms, hence the provisions should apply from both a
jurisdictional and consent paradigm. The French New Code of Civil Procedure
Article 1464(2) provides that the fundamental principles of court proceedings
set forth in Articles 4 to 10, 11(1) and 13–21 of the Code shall always apply to
arbitral proceedings. Similar concepts are contained in most arbitral rules. For
example, Article 1(2) of the UNCITRAL Arbitration Rules 2010 indicates that the
rules are subject to any provision of the law applicable to the arbitration from
which the parties cannot derogate. In considering the earlier version of the
UNCITRAL Arbitration Rules, Redfern and Hunter suggest that provisions
entitling either party to call for a hearing, provisions dealing with one
consecutive exchange of written submissions and the provision entitling the
parties to question a tribunal-appointed expert and to present their own expert
witnesses are mandatory. In some cases mandatory norms do not apply
automatically but depend upon a request by one of the parties. An example
would be the right to oral hearings.

The next important question is to understand what such mandatory norms

require. Mandatory procedural norms will rarely be mechanical in nature but
will instead involve questions of degree in relation to matters such as due
process. Hence it will often be debatable as to how they should apply. In this
sense there are potentially important nuances between the expressions used in
different laws and rules, with some such as the UNCITRAL Model Law speaking
of parties having a ‘full opportunity’ of presenting a case, while others such as
the ICC Arbitration Rules 2012 speak of ‘a reasonable opportunity’ to present the
case. Previously, UNCITRAL Rules 1976 Article 15.1 afforded each party a ‘full’
opportunity, but the 2010 version has changed that to a ‘reasonable’
opportunity. (198) Section 33(1)(a) of the English Arbitration Act 1996 speaks of
giving each party ‘a reasonable opportunity of putting his case and in dealing
with that of his opponent…’. Article 182(3) Swiss PILA simply refers to the ‘right
to be heard…’.

These due process norms are considered throughout the book as they typically
underlie many challenges on questions such as timing of hearings and
presentation of evidence, but some general observations are made at this stage.
On plain page "184" meaning, a ‘full’ opportunity could be asserted to go further
than a reasonable opportunity but that is not a preferable view. While different
rules express the test differently, the better view is that each is concerned with
fairness and due process and the differing expressions have little, if any, impact
on such overriding values. A reasonable opportunity to present a case ought to
be seen as a full opportunity in any event as long as a full opportunity is merely
understood as a form of opportunity. Anything less than a full opportunity means
that part of the case has not been adequately allowed to be presented.
Legislators who use the term ‘reasonable’ or ‘adequate’ rather than ‘full’, are
almost certainly not seeking to limit the due process obligation. Instead it is
more likely that they accept the reality that the way this is to apply on a case-by-
case basis requires some rule of reason by the tribunal. The concern is that a
reference to a ‘full’ opportunity might be abused and might invite inappropriate
assertions by a party that it should be the sole determinant of the ambit of its
entitlement, a contention that should be rejected in any event. An example of
misuse would be an assertion that a party can present as many witnesses as it
likes as anything less is not a full opportunity. The proper response is that the
opportunity is with respect to the case and superfluous witnesses cannot be
justified on due process grounds. The entitlement to fully present a case also
does not entitle a party to ignore reasonable deadlines even where material
evidence is concerned. A tribunal is entitled to close proceedings after each party
has had a reasonable opportunity to present oral and written submissions and
not wait indefinitely while either tardily seeks to avail itself of that opportunity.

Courts may also elaborate on these norms as they infer them from the lex arbitri
and perhaps principles of transnational public policy. (199) For example, the Swiss
Federal Supreme Court has elaborated on the fundamental requirements of due
process in the context of Article 182(3) and Article 190(2)(d) PILA:

… each party has the right to express itself on the facts that are essential for the
decision, to present its legal arguments, to propose its evidence with respect to
relevant facts and to participate in the hearings of the arbitral tribunal. On the
other hand, the right to be heard does not encompass the right to express oneself
orally. As to the right to propose evidence, it is necessary that it be exercised in a
timely way and respects the applicable formal rules. The arbitral tribunal can
refuse to admit a piece of evidence if the latter is not apt to prove the fact that the
party proposing it purports to prove, if the relevant fact has already been proved,
if it is not relevant or if the arbitral tribunal – on the basis of a so-called
‘anticipated weighing of the evidence’ -concludes that it has already formed its
conviction and that the new piece of evidence cannot modify it….

The equal treatment of the parties, also guaranteed by Art 182(3) and Art
190(2)(d) PILS, implies that the proceedings be regulated and conducted in such
a way that each party has the same opportunities to present its case.

page "185"

The principle of adversarial proceedings, guaranteed by the same provisions,

requires that each party can comment on the other party's presentation of its
case, examine and discuss the evidence produced by the other party and refute it
with its own proof. (200)

While these broad principles would apply in nearly all jurisdictions, care needs
to be taken with jurisdictions that may have unique mandatory features. For
example, a Spanish Supreme Court vacated an award because the award had not
been registered before a notary as was required under an earlier form of the
Spanish Arbitration Act. (201) In due course, countries concerned to promote
themselves as neutral arbitration centres will remove such provisions, but in
each case, care needs to be taken with unfamiliar provisions. In this sense it will
sometimes be difficult to evaluate the laws without seeking local advice. For
example, the arbitration law of the Republic of China contains numerous
references to the application of the Code of Civil Procedure and other domestic
ROC laws. (202) One commentator has been critical. Li states that such references
‘… are probably above the comprehension of the foreign parties and lawyers’.
(203)

3.7.2. Mandatory Procedural Laws from Other Than the Seat

In some circumstances there is a need to consider whether mandatory

procedural laws of other jurisdictions purport to apply and whether they validly
do so. In an international arbitration, the fact that one government expresses
such a will does not mean that it should necessary apply extraterritorially. There
are two questions. First, can it be concluded that the foreign government intends
the norm to apply to a foreign arbitration? If so, should it do so?

The various potentially relevant countries could include the countries of

residence or domicile of the parties, the place of performance or the likely places
of enforcement. Where residence or domicile of the parties is concerned, some
countries might seek to apply norms extraterritorially on the basis that they see
their role as protecting their citizens’ due process rights regardless of where
they do business or seek to resolve disputes. Alternatively, the government may
simply wish to apply such norms domestically, but accept that where citizens do
business overseas and freely consent to other norms, the latter should apply. The
country of performance may wish certain regulatory standards to apply to
commerce within page "186" its borders regardless of the wishes of the parties,
although this is more conceivable with substantive laws and not procedural
ones. Regardless of the country concerned, in many cases it will not be clear
whether the government actually intends for extraterritorial operation of such
norms. Statutes are often unclear in that regard.

One would usually not see procedural norms of a foreign country purporting to
apply as mandatory norms in arbitration in a different Seat. Nevertheless, the
fact that some governments might wish to achieve this outcome ought to be
noted, particularly when that country might also be the most likely enforcement
country and may utilise public policy exceptions as a ground to block
enforcement when their norms have not been complied with. Fouchard, Gaillard
and Goldman suggest that regardless of the law of procedure, the mandatory
provisions of law in the jurisdictions where the award is liable to be reviewed by
the courts cannot be entirely ignored, as these are the only laws that limit the
autonomy of the parties and the tribunal in the conduct of the proceedings. (204)
That view is certainly consistent with the authors' general theory of arbitration.
A number of scenarios should be considered before a conclusion is postulated on
this issue. First, even if a foreign provision purports to apply as a mandatory
norm, in most cases this should have no practical effect. As noted below, an
enforcement court applying truly transnational standards of public policy would
not require procedural steps that were not already required by the mandatory
norms of the lex arbitri. Conversely, where such foreign norms are unreasonable
in attempting to give unfair rights to one of the adversaries, they would also
conflict with the mandatory norms of the Seat itself and hence should naturally
be ignored. (205) More challenging scenarios are where the alleged mandatory
rule is not widely accepted but would make a fundamental change to the agreed
arbitration. Examples might include form requirements in arbitration
agreements; limitations on the rights of certain parties to arbitrate; limitations as
to the qualifications for arbitrators and provisions as to non-arbitrability. (206)
These should generally not be controlling as they would still conflict with more
permissive lex arbitri. In extreme cases, foreign mandatory norms may purport
to remove jurisdiction. However, the doctrine of separability means that certain
purportedly mandatory requirements may not invalidate the arbitration. For
example, in Sojuznefteexport v. Joc Oil Ltd (207) the Bermuda Court of Appeal
allowed enforcement when the respondent asserted that the arbitration
agreement was invalid when contained in a contract that did not conform to
mandatory signature requirements under applicable Russian law pertaining to
foreign trade transactions.

page "187"

If the foreign norms are not inherently unreasonable and the two jurisdictions
impose fundamentally different procedural models, an arbitrator seeking to
promote enforceability may have particular difficulties in determining how to
proceed. An example of a challenge would be a foreign law that demands oral
hearings in all cases where the lex arbitri leaves this to tribunal discretion absent
agreement of the parties. There would also be confusion as to which law should
be relevant for the purposes of Article V(1)(d) of the New York Convention
where enforcement challenges are concerned. In other cases, if such a
purportedly mandatory norm does not conflict with mandatory norms of the lex
arbitri and is at least arguably reasonable but not binding, an arbitrator might
still accommodate it through procedural discretions for that reason alone and
also in order to promote enforceability. For example, if arbitration is conducted
in a Seat where witnesses need not testify under oath, but one of the parties
comes from a jurisdiction purporting to require this extraterritorially, an
arbitrator with the power to do so might employ a discretion to apply oaths to
avoid any potential conflict. In some cases that might not be possible, however,
as the arbitrator might not be entitled under the law of the Seat to administer an
oath. The point is simply that some potential conflicts might be avoided through
a judicious use of arbitral discretion.

A number of concluding remarks may be made. There seems no theoretical

justification for taking the duty to render an enforceable award into account over
and above the normal considerations as to the applicability of mandatory laws.
Either the tribunal considers the foreign norms binding or reasonable by way of
proper analysis or not. If not, the fact that an enforcement jurisdiction may take a
different view should not be persuasive. A tribunal may feel comfortable in
adopting a different view where there is a mandatory procedural norm that
would be easy to comply with. The situation should be different if mandatory
procedural norms of an enforcement country would ask an arbitrator to conduct
proceedings contrary to their strong preferences based on fairness and
efficiency. This is even more so where a supposedly mandatory procedural norm
could alter the outcome on the merits, perhaps by changing the relative bodies of
evidence of each party. It is hardly an exercise of a duty to render an enforceable
award where the award itself is altered in ways contrary to an arbitrator's
preference. Similar principles would apply if the award might be considered to
be contrary to the public policy of a potential enforcement country. Furthermore,
while it is certainly important to consider the role of controlling courts, they will
never be given an opportunity to limit the autonomy of the parties unless one of
the parties brings a challenge to a previous determination. If parties agree on a
particular procedure and choose never to challenge the outcome, then autonomy
remains inviolate. Conversely, where a successful challenge occurs even in
respect of a matter agreed to between the parties, it is only because that court is
allowing a party to resile from its earlier agreement. That should only arise
where that earlier agreement offended against some fundamental principle of
procedural justice.

This book seeks to point out the many instances where procedure, evidence and
substance overlap. In this context, it at times addresses substantive issues where
important procedural and evidentiary questions are involved. In the context
page "188" of the current question as to foreign mandatory procedural laws,
there is even some relevance of the related question whether a foreign
substantive law must apply. If so, this can be said to limit the procedural
discretion under the lex arbitri as to choice of law. The most troubling issue is
where a tribunal is determining applicable law and knows that the key
enforcement court demands application of allegedly mandatory substantive
norms such as competition law. The key example is the Eco Suisse decision. (208)
This topic is discussed in detail in section 13.8.6. For present purposes it may be
said that even here, if the tribunal is properly of the view that the law does not
apply, this is the proper decision under the duty to complete the mandate.

There will also be evidentiary questions about the purported application of

foreign mandatory procedural norms. Where a party asserts that a mandatory
law external to the lex arbitri applies, it will need to establish this to the
tribunal's satisfaction. While it is not accurate to speak of burden of proof in
relation to applicable law, the party seeking to make the point will need to show
why the law is mandatory, why it applies to the arbitration and why it is not
otherwise dealt with or overruled by the lex arbitri or procedural rule selected.
(209)

3.7.3. Mandatory Institutional Norms

The above discussion contemplated mandatory procedural laws of a foreign

State. Another question arises in relation to mandatory requirements of an
arbitral institution. There is a difference between rules that are mandatory per
se because they enshrine fundamental principles of due process and rules that a
particular institution requires to be mandatory for their agreed involvement. The
first scenario need not be separately considered. Provisions in rules simply
replicate key norms in lex arbitri which cannot be varied in any event. As to an
institution's own mandatory requirements, an example would be the ICC
refusing to administer an arbitration where the parties seek to remove the
supervisory jurisdiction of the ICC Court. Such norms must be complied with.
That simply arises as a matter of contractual agreement by a private provider
where its offer of services denies the right to vary certain rules. Not all rules are
mandatory and institutional attitudes to modifications vary. This is discussed
further in sections 3.9.2 and 3.14.

3.8. Selecting Procedural Laws Other Than Those of The Seat

Much has been written as to whether parties can effectively select a procedural
law other than that of the Seat. Consent-based paradigms suggest that it is
theoretically page "189" possible for parties to select the Seat but designate a
different procedural law to apply. While that is so, commentators uniformly
criticise this option as it is fraught with uncertainty and potentially raises a range
of disputes and challenges that would not otherwise occur in the usual case
where the lex arbitri of the Seat is controlling. (210) Practical problems would arise
in determining which court would have supervisory jurisdiction over procedural
challenges. Would it be the court of the Seat or would it be the courts in the
country of the procedural laws selected? What if the law of the Seat and that
other procedural model took different views about the distinction between
procedural and substantive matters or about party autonomy in such
circumstances? What if they had different mandatory requirements before an
award would be in enforceable form? It is conceivable that in such
circumstances, two different courts might feel seized of supervisory jurisdiction,
with all of the problems of unnecessary duplication and potentially inconsistent
findings. (211)

Some rule systems make it clear that even if parties select a different procedural
set of norms than those of the Seat, nevertheless, the courts of the Seat retain
supervisory jurisdiction. This is the position in Switzerland. (212) French law also
ensures that even if a foreign procedural law is chosen to govern arbitration,
mandatory French rules of procedure must be complied with. (213) It has been
argued that selection of a different procedural law is evidence of consent to the
jurisdiction of that country's judicial system. (214) The better view is to the
contrary, however, both in terms of the weight of case law and express
references in arbitration statutes. Because of these problems, a tribunal or court
is unlikely to conclude that parties have selected a different procedural law to
that of the Seat without sufficiently clear evidence to that effect. (215) Some
individual legal systems take a different view, which is simply another example
of contentious extraterritorial judicial fora. For example, Indian courts have at
times accepted supervisory jurisdiction over arbitrations simply where the
transactions are governed by Indian substantive law. (216)

page "190"

3.8.1. Selecting Domestic Arbitral Statutes as a Procedural Framework

Party autonomy would technically allow for selection of domestic arbitration

statutes to apply in respect of international matters. (217) Most lex arbitri such as
the Model Law allow for their provisions to be expressly excluded by the parties.
Various international arbitral rules make clear that domestic procedural laws
will not normally impact upon an arbitration as a matter of course. Exceptions at
times arise where mandatory laws of a national legal system are applicable,
where interim measures are sought from a national judicial body and where
national registration requirements for the award must be complied with. (218) As
noted, national procedural laws can also apply where the parties choose them to
be applicable or where an arbitrator does so in exercise of a broad discretion
absent party choice. As Laurence Craig has pointed out, if care is taken in drafting
and in selection of an appropriate Seat, most arbitrations will be conducted
under rules agreed to by contract. It is only in a small number of less than
optimally prepared cases where the intersection between domestic procedural
norms and arbitration rules remain problematic. (219) Some practitioners might
select domestic arbitration rules simply because of familiarity. It is never
desirable for practitioners to operate in this way. This is for both ethical and
practical reasons. As to ethical considerations, the duty of counsel is to give the
best possible advice to the client. Advice as to an appropriate procedural model
can only be optimal if counsel has properly considered a range of alternatives.
Merely choosing familiar rules would be better seen as a conflict of interest
between the entitlement of the client and the self-interest of counsel.

There may be a number of significant practical problems when selecting

domestic rules. First, because such rules are not concerned with the
requirements of international arbitration, they may fail to deal with an essential
element and might even render the arbitration agreement pathological.
Secondly, it is typically the case that most legislatures opt for a greater degree of
court scrutiny of domestic arbitration over that of international arbitration.
There are natural reasons for this. Where domestic arbitration is concerned, it is
simply seen as a permitted alternative to domestic litigation for local persons
and is not aiming to be a neutral venue for foreign parties. The view is often
taken that this alternative needs as much if not more court scrutiny than first
instance litigation. Conversely, where international arbitration is concerned,
lesser scrutiny flows in part from an page "191" intergovernmental agreement to
allow as much as possible for party autonomy and support for arbitration
through broad enforceability commitments, (220) and also from the desirability of
establishing neutral venues with suitable laws. This does not necessarily mean
that international rules will always be preferable to domestic rules. In some
disputes, one or more of the parties might prefer the more interventionist
approach of the courts typically allowed for under domestic arbitration statutes,
particularly where intervention is supportive of the process. An example might
be where urgent interim measures are required to ensure any meaningful value
in an ultimate award and where the domestic laws have stronger provisions that
have adequate territorial reach. Nevertheless, while there may be instances
where parties could legitimately prefer to choose domestic procedures, this does
not detract from the general obligation to consider the alternatives and seek
appropriate instructions. It also remains true that, a priori, international
arbitration rules are likely to be more appropriate for international disputes.

3.8.2. Selecting Domestic Litigation Procedures as a Framework

The situation is even more problematic where parties do not choose domestic
arbitration rules but instead seek to run arbitration via domestic rules of court-
oriented civil procedure. In such a circumstance, the rules may be inadequate
from a policy perspective and there may be complex interpretational questions
as to which parts of the rules apply when some elements might seem only
applicable to court procedures. (221) Such an approach would never seem
desirable. (222)

3.9. Arbitral Rules

Because the lex arbitri will generally enshrine party autonomy, the right to select
particular rules is recognised. For example, Article 19(1) of the UNCITRAL Model
Law indicates that the tribunal is bound by procedural choices made by the
parties save where these conflict with any mandatory provisions of the Model
Law. Thus if the parties select rules under that or similar mandate, to the extent
of any conflict between the lex arbitri and the rules selected, the latter will
prevail unless they purport to override a mandatory procedural norm. In other
page "192" circumstances, there is indeed no conflict as the allowance of
variation was mandated by the Model Law itself.

Some institutional and ad hoc rules add important detail to the lex arbitri. In
some cases, however, they primarily replicate the general norms as found in
advanced arbitral statutes. In some cases arbitral statutes are now even more
detailed than institutional or ad hoc rules. This is partly a feature of the ongoing
reform process that any government, intergovernmental body or arbitral
institution engages in to ensure that their framework is state of the art.
Sometimes it is simply a matter of timing as to which is more elaborate and up to
date at any point in time. The most influential set of rules are the UNCITRAL
Arbitration Rules. The current version is the 2010 Rules. These are not only
influential in terms of their direct use, but over the years, various versions of the
UNCITRAL Rules have formed a model when individual institutions are
considering their own set of procedural rules. Various aspects of the UNCITRAL
Rules and other key models are discussed on a topic-by-topic basis. Where
institutional rules are concerned, a brief introduction to the key institutions,
their rules and processes is contained in section 3.14 below. Here the revised ICC
Rules 2012 are particularly important.

3.9.1. Modifications to Arbitral Rules and the Impact on Arbitration

Many of the key arbitral institutions make regular amendments to their rules as
an ongoing reform and improvement exercise. In addition to dealing with
specific problems that have arisen from time to time, rule amendments will often
introduce or strengthen key policy norms such as delocalised aspects, party
autonomy, fair and effective institutional control and effective time lines. An
important question is whether revised institutional rules apply retrospectively
to arbitration agreements entered into beforehand and which refer generally to
such rules. In all cases it is a question of interpretation of party intent. If the
parties wish to have the older rules apply, they could designate them expressly
or refer to the rules ‘in force at the time of the arbitration agreement’.
Conversely, if they wish to be subjected to the updated rules, they could stipulate
a reference to the rules ‘as amended from time to time’ or ‘those in force at the
time of the dispute’. Where the parties simply refer to named rules without
indicating which of the above they prefer, the drafting is suboptimal and an
adjudicator will need to form a view as to the parties' intent.

Modern rules will themselves seek to indicate that a reference to the rules is
presumed to be a reference to rules in force at the time of the dispute. That by
itself cannot be determinative as a statement by an institution or by UNCITRAL
at a particular point in time cannot be determinative of the true intent of the
parties at an earlier time. Nevertheless, as such stipulations become more
common, it is more reasonable to presume that parties selecting such rules
would be aware of the stipulation and hence have impliedly agreed to it.

page "193"

3.9.2. Modification of Published Rules by the Parties

One of the important reasons why parties would choose a particular institution
or set of ad hoc rules published by a body such as UNCITRAL is respect for the
fairness and efficiency of the rules they have established. At times, however,
parties might wish to utilise the services of a particular institution more for its
expertise and stature and might wish to modify the rules for their own
circumstances. In those cases the question is whether the institution itself will
allow this to occur. Party autonomy means that they can designate anything they
like, including modification of rule systems. But party autonomy does not extend
to compelling an institution to act in a way that it does not wish to do. An
institution is entitled to offer its services on terms that it wishes, including that
rules not be modified without the institution's consent. The presumption is that
the institution has made an offer to provide services on condition that the rules
are properly adhered to in full. (223) However, one cannot presume that an
institution and/or the tribunal established through it would necessarily refuse to
accept a particular modification agreed to by the parties. It is obviously prudent
for parties who are interested in this to carefully consider why they wish to do so
and seek advice in advance from the institution as to its response if the rules are
not clear on this issue. Where they fail to do so, parties might find that their
variation agreement is ineffective. In an extreme case, if the variation is
contained within the arbitration agreement, this may be pathological as it may be
conditional on an institution agreeing to do something that it simply will not do.
Another scenario is where the parties seek to add elements not articulated in the
rules rather than detract from them. If they do not conflict with the rules, this
should not be problematic.

Most variations would be non-contentious and many modern rule systems are
strongly supportive of the parties' rights to modify the rules, including going so
far as to allow the institution to administer an arbitration conducted under
wholly distinct rules. (224) The CIETAC Rules allow non-CIETAC Rules to be
utilised provided that the use of such rules is not ‘inoperative or in conflict with a
mandatory provision of the law as it applies to the arbitration proceedings’. (225)
Similarly, where ad hoc arbitration is concerned, Article 1(1) of the UNCITRAL
Rules 2010 is to similar effect. In Insigna v Alston, (226) the Singapore Court of
Appeal upheld the validity of an arbitration agreement agreeing to SIAC
arbitration but under ICC Rules. Importantly, that was only possible because
SIAC was prepared to accept such an appointment, a view which was apparently
not appealing to the ICC at the time.

page "194"

ICC practice is the most illustrative, simply because it has a tradition of strong
control over the process and a more elaborate set of prescriptions through the
role of the Court, terms of reference and scrutiny of awards. While the ICC would
not regularly agree to modified rules, one cannot categorically state which
modifications the Court would accept from the outset. The court would look at
purported modifications on a case-by-case basis. (227) Craig, Park and Paulsson
suggest that the Court is only likely to refuse to administer an arbitration ‘when a
fundamental characteristic of ICC arbitration (such as Court scrutiny of the
award) is omitted’. (228) Schwartz notes that while there is no obligation to do so,
the ICC has accepted rule changes from time to time. (229) He suggests that the ICC
has refused to administer a case where parties provided for non-binding
arbitration and where they designated that an award could be made by the
chairman alone in a multi-person tribunal. (230) He also suggests that the ICC
would not accept a case if the parties sought to avoid court scrutiny or utilisation
of terms of reference. He notes the case of Sumitomo Heavy Industries Ltd v Oil
and Natural Gas Commission (231) where the ICC refused to accept the
appointment of two arbitrators and an umpire where the umpire was not
prepared to act as a normal chair as per the ICC Rules. (232)

3.9.3. The Language of Institutional Rules

Many arbitral institutions will tend to publish their rules in foreign languages,
particularly English as well as in the national language. It is then important to
determine whether the English or other foreign version is official and/or can be
an aid to interpretation of the original draft. Importantly, there are many
instances where meaning may differ depending on which version of the rules is
being considered. (233) In other cases, one version may be resorted to in order to
clarify ambiguity in the official version.

page "195"

3.9.4. Specialised Rules

One of the features of international arbitration is that it allows the parties to

tailor their dispute to their particular commercial circumstances. In some cases,
trade and industry bodies have developed their own specialised rules after
reflecting on the optimal needs of their trade or industry. Specialty rules have
been developed in the fields of maritime, transport, insurance, commodities and
employment disputes. (234)

3.9.5. Interpretation of Lex Arbitri and Arbitral Rules

Once arbitral laws and rules are identified, they will at times need to be
interpreted. There are a number of broad questions. Where laws are concerned,
should they be interpreted in the same manner as other laws emanating from
that jurisdiction? If so, is there a separate law on statutory interpretation in that
jurisdiction? Does the law allow for both plain meaning and purposive
interpretation? As to the latter, when may it apply and what extraneous evidence
of purpose may be resorted to? Conversely, would a more international approach
to interpretation be permitted? If so, is that desirable and how would it differ? It
has been argued that if there are truly transnational procedural norms they will
not only be applicable as such but they are more likely for that reason alone to be
interpreted from an international rather than a parochial perspective. (235)

Where rules are concerned, does the promulgating body have available travaux
preparatoire which can be resorted to? If not, can journal articles written by
members of drafting committees be accepted as reliable guides to intent? For
example, Holtzmann and Neuhaus have provided a detailed analysis on the
travaux preparatoire of the Model Law. (236) While some legal systems would not
accept secondary sources as being valid sources of this nature, both the
involvement and stature of the authors has ensured that it has had appropriate
influence in the interpretation of Model Law provisions. Most of these questions
would be answered positively although views may legitimately differ.

3.10. Arbitral Guides to Procedure and Evidence

Other potentially relevant sets of norms are the recommended practices of

institutional or professional bodies. If expressly or impliedly agreed to by the
parties, they are binding in any event unless they offend against mandatory
norms. Even if page "196" not agreed to by the parties, if viewed as
recommended best practice they may be used as guides by tribunals when
exercising broad discretions. Such documents might also constitute common
usages or customs, although this would not normally be so. This is because such
guides are more commonly developed when there are clear differences in
approach that require some harmonisation or direction. Their utility is
contentious, however, some arguing that they add greater transparency and
certainty, while others argue that they add unnecessary complexity and expense
without being able to accurately outline key norms. These issues are discussed
further in section 6.5.

3.11. International Law and Procedure

3.11.1. Investment Arbitration and Procedural Aspects of International Law

Where investment arbitration is concerned, elements of public international law

and the terms of relevant investment treaties will be central to the identification
of jurisdiction and applicable procedures. A number of regional or content-
specific conventions may be relevant. These include the European Convention on
International Commercial Arbitration, (237) the Inter-American Convention on
International Commercial Arbitration (238) and the ICSID Convention. (239) There
may also be regional trade or bilateral trade and investment agreements and
other forms of agreement that provide for investor protection. (240)

In some cases it is unclear as to the impact that such norms may have on
procedural matters. One of the most contentious debates has been whether most
favoured nation clauses (MFN) apply to procedural as well as substantive
matters. The debate has permeated a range of investment disputes beginning
with Maffezinni. The case involved a bilateral investment treaty that barred
arbitration of disputes until a particular time period had elapsed and also until
there had been exhaustion of domestic remedies. Arbitration was commenced
before the time period had expired and before exhaustion had occurred. The
investor claimant argued that the particular country had not imposed these
limitations on arbitration in other treaties. Hence the investor argued that he
was entitled to the better procedural entitlements found under other treaties
through the operation of the MFN clause within the treaty under which he
brought his claim. The tribunal in Maffezinni accepted this argument. Since then,
a number of tribunals have divided, some following Maffezinni while others have
asserted that MFN obligations are not intended to deal with procedural matters.
(241)

page "197"

The issue should properly be seen as one of identification of intent. What did the
countries concerned truly intend in adopting an MFN clause? It is
understandable that without clear guidance, various tribunals might differ. Some
argue that without any limitation on an MFN obligation it extends broadly on its
plain meaning to encompass procedural matters. They might support that view
through resort to ordinary meaning interpretation as required by the Vienna
Convention on the Law of Treaties. (242) Others would argue to the contrary,
reasoning that expressly including time limits in an arbitration clause shows a
clear intent in that regard and would be a specific agreement overriding a more
general most favoured nation norm. The division between the various cases has
quite rightly led treaty negotiators to clarify this issue of intent in subsequent
negotiations. The norm is now to exclude procedural matters from the ambit of
the most favoured nation clause. Whether clarified or not, a tribunal at least
needs to consider the impact of this source of law on its procedural mandate.

3.11.2. Human Rights and Procedure

A general question is whether human rights based procedural justice guarantees

apply to voluntary arbitration. This would not be limited to investment
arbitration. This is a contentious question that has largely been debated within
the context of the potential applicability of Article 6(1) of the European
Convention on Human Rights. (243) More recently, consideration has concentrated
on the elements of the protections as well as their more general application.

Article 6(1) of the European Convention on Human Rights indicates that ‘(i)n the
determination of his civil rights and obligations … everyone is entitled to a fair
and public hearing within a reasonable time by an independent and impartial
tribunal established by law’. Article 6(3) may even be relevant in relation to the
right to cross-examine and/or suggest questions to the tribunal. The first issue is
whether this Convention applies to arbitration. Some would take the view that
the Convention is entitled to be all-encompassing as this is the essence of human
rights. While it has been noted that Article 6(1) was drafted with criminal cases
in mind, some suggest that the norm could readily be applicable in civil matters
as well. (244) Others might argue that an arbitral tribunal is not ‘established by
law…’. page "198" The better view should be that in establishing a lex arbitri as a
framework for arbitral tribunals, sovereigns subject to this Convention's norms
should meet its stipulations. Other potential international norms include Article
14 of the UN Convention on Civil and Political Rights, 1966. Article 14(1) states
that ‘(a)ll persons shall be equal before the courts and tribunals’. Article 16
stipulates that everyone shall have the right to recognition everywhere as a
person before the law. Article 17 may also be relevant which aims to protect
privacy. This could impact upon the role of witnesses and the attitude to
arguably privileged communications. To similar effect is Article 8 of the
European Convention on Human Rights.

More important than the overall question of applicability is the question of what,
if anything, might be added or varied by such norms. In most cases there should
be consistency between the norms emanating from human rights treaties and
those that are an essential part of arbitral due process in any event. Human
rights conventions are difficult to establish and will typically concentrate on the
least contentious norms, being those that must apply in any event to arbitration.
It is even less likely that any customary norms of procedure would apply via a
human rights theory separate to any treaty norms or principles contained in lex
arbitri. That might more possibly arise in relation to certain issues argued to be
norms of international public policy.

The compatibility between Article 6 and arbitration in England was considered

in Stretford v Football Association Ltd. (245) Mr Stretford was a football agent
licensed by the Football Association under rules that included an arbitration
agreement. He commenced proceedings before the courts in relation to
disciplinary proceedings brought against him by the Association in arbitral form.
The Association sought to stay the proceedings pursuant to s 9 of the Arbitration
Act 1996. Mr Stretford's application was rejected. The Court of Appeal held that
an arbitration agreement freely entered into is deemed to be a waiver of the
right to a public hearing and a waiver of a right to have a case dealt with in an
ordinary court. Such a waiver would not apply in circumstances such as duress,
undue influence, mistake or where the waiver is contrary to an important public
interest. The principles in Stretford were endorsed by the House of Lords in
Premium Nafta Products Ltd and Others v Fili Shipping Company Ltd and Others.
(246) Similarly, applicants have been unsuccessful in seeking to argue that Article

6 of the Convention entitles them to appeal procedures, although where leave to

appeal provisions are included as in the English Arbitration Act 1996, the Court
of Appeal maintains a residual jurisdiction to determine whether the exercise of
the discretion was fair. (247) To the extent that they could be described as
inalienable guarantees, waiver would not be possible. Another way to describe
such a guarantee would be to see it as a mandatory norm of procedure fully
applicable in the case of arbitration. The European Court of Human Rights has
addressed an argument challenging tribunals' partiality on the merits. While it
rejected the application, it did not do page "199" so on the basis that human
rights guarantees had been waived. (248) On this approach it would be necessary
to consider the particular rights in issue and determine whether they were
waivable or not. (249)

3.12. Transnational Public Policy and Procedure

Public policy can be relevant to the procedural conduct of arbitration in a

number of ways. The parties might expressly indicate that public policy is to be a
guiding criterion in the arbitral procedure, although that would not be typical. In
some jurisdictions public policy is seen as the key means by which validity and
arbitrability may be considered. The duty to render an enforceable award may
invite a tribunal to consider the way the public policy exception to enforceability
may apply.

It is accepted that the public policy exception in the New York Convention should
be limited to transnational or truly international public policy. (250) Hence it
should only deal with norms having the broadest international recognition.
Hence this would add little, if anything, to the range of norms that should already
apply through due process mandates within the lex arbitri. It is of course possible
for domestic courts to have a different view although that would be problematic
behaviour on their part and not a reason why a tribunal should accept such
dictates contrary to the parties' consent. The Report of the UNCITRAL
Commission on the Model Law noted that it was understood that public policy as
used in the New York Convention and other treaties ‘covered fundamental
principles of law and justice in substantive as well as procedural respects’. (251)
An ILA Report referred to both procedural international public policy and
substantive international public policy. (252) Issues of substantive international
public policy are considered in section 13.8.10.2. There may also be classification
questions as between procedural and substantive public policy. The ILA Interim
Report states ‘substantive public policy (ordre public au fond) goes to the
recognition of rights and obligations by a tribunal or enforcement court in
connection with the subject matter of the award (as opposed to procedural
public policy, which goes to the process by which the dispute was adjudicated)’.
The civilian legal tradition also generally sees notions of public policy
encompassing breaches of procedural justice. (253)

page "200"

Even where public policy may be thought to apply, there are problems in
understanding its true meaning and essential content. The Swiss Federal
Tribunal stated:

It should be underlined that procedural public policy will constitute only a

simple exclusion provision, namely that it will merely have a protective function
and will not generate any positive rules. This is because the legislature did not
desire that procedural public policy should be extensively interpreted and that
there should arise a code of arbitral procedure to which the procedure, as freely
selected by the parties should be subjected. (254)

While it is sensible to not attempt an exhaustive definition of procedural public

policy, the distinction postulated above between protective and positive rules is
difficult to maintain. If an act or omission is of such a problematic nature to allow
enforcement to be validly blocked on international public policy grounds, it must
ipso facto be a matter that should be addressed by any sensible tribunal.
However, it has been argued throughout that truly transnational public policy
will not add anything to mandatory norms of a well-drafted lex arbitri.
3.13. Lex Mercatoria And Procedure

Section 13.6 considers the possibility that a tribunal may apply lex mercatoria as
the substantive law of the contract. This section is instead concerned with
whether there may be said to be a lex mercatoria as to procedure. There needs to
be consideration of the means by which this could be applicable and the content
of any suggested norms. The issue could arise in a number of ways. While it
would be extremely rare for parties to do so, they might stipulate that the
arbitration is to be conducted utilising procedural norms of the lex mercatoria, In
such a case, the tribunal would be bound to do so and would then need to make a
determination as to the content of such requirements. Another scenario is to
consider whether there is some accepted content of procedural lex mercatoria
when trying to flesh out broad principles of due process or international public
policy. A further possibility is an allegation that procedural lex mercatoria might
apply as a matter of custom or usage. This would be a variant of an implied intent
argument.

While such assertions are possible, procedural lex mercatoria is highly unlikely
to add any meaningful content over and above principles that are either
expressly referred to in lex arbitri and rules, or which flow from good faith
approaches to interpretation of such norms. The nature and content of lex
mercatoria is thus highly contentious, but practitioners should at least be aware
of the potential scope of such arguments. For example, Professor Klaus Peter
Berger in page "201" his ongoing project seeking to analyse the nature and scope
of lex mercatoria, (255) has suggested the following principles that would pertain
to arbitral procedure, although not all would be truly accepted by all parts of the
arbitral community as consensus based or common usage norms:

The parties are free to enter into contracts and to determine their contents
(principle of party autonomy)2026

Where there is doubt about the meaning of a contract term, an interpretation

should be preferred that makes the contract lawful or effective …

If a contract has contacts to more than one jurisdiction and the parties have not
agreed on the applicable law, it is in the presumed interest of the parties to apply
the law, both as to form and to substance, that validates the contract…

Interpretation: contra proferentem,… take into account whole contract,…

common intention prevails where incorrect term used…

The invalidity of the main contract does not automatically extend to the
arbitration clause contained therein (principle of separability)…

Arbitration proceedings are not suspended if one of the parties goes bankrupt…

Absent a choice of law by the parties, the contract is governed by the law with
which the contract has the closest connection …
(As to Set-off)… If parties have mature and liquidated claims of an identical
nature vis-a-vis each other, each party may declare the set-off of these claims.
The parties may also agree on the set-off of these claims by contractual consent…

No one may set himself in contradiction to his previous conduct…

A right that has been forfeited may not be raised …

The burden of proof rests on the claimant…

A written contract may be proved through any means of modern

telecommunication… if it provides a record of the information contained therein
and can be reproduced in written form…

Circumstantial evidence is permissible. (256)

3.14. Nature of Institutional Arbitration

Because of the importance of institutional arbitration, the following sections seek

to outline the general nature of arbitral institutions, their impact on procedural
norms and discretions, the relationship between the institution, arbitrator and
the parties and levels of institutional control. It also seeks to broadly outline the
similarities and differences between institutional models.

page "202"

Before doing so, some brief general comments are provided as a precursor.
Arbitral institutions are not all identical even if they have similar rules.
Institutions range from non-profit to profit-making, government controlled to
independent, parochial or truly international. In some countries there is more
than one institution vying for the work. (257) Institutions may vary as to whether
they use lawyers to carry out administrative functions or non-lawyers. They may
also vary as to how intrusive administrative officers prove to be. To properly
evaluate institutions one must of course consider these issues, their published
rules and procedures together with their practices. This book does not attempt
to provide any comparative analysis of this nature. Such an attempt would be
fraught with numerous methodological problems. For these reasons, the
following sections concentrate more on the published rules and procedures.
Even this is problematic as there is significant convergence between different
institutional rules, hence practices become even more important as a
distinguishing feature. It is natural to expect such convergence in express rules.
All will engage in widespread comparative analysis in considering reforms.
Leading practitioners are on numerous boards and panels. Cross-pollination of
ideas is to be expected. In addition, the UNCITRAL Rules will always be a guide in
considering what issues to cover. Mandatory procedural norms also find their
way into all rule systems. Subject to those rules, the primacy of party autonomy
and broad residual discretionary powers for arbitrators will both be enshrined
in any system. Hence the broad structure should be close to identical. Not only is
there convergence in what the rules do, but also in what they do not seek to
articulate in detail. Institutional rules would generally not seek to clarify complex
evidentiary questions such as when production of documents may be required.
Because the institutions wish to be popular with parties from all legal cultures,
they will tend to refrain from too specific directions in areas such as this.

While there is thus significant convergence amongst arbitration rules, some have
distinct features. These include the terms of reference process of the ICC Rules,
summary proceedings in the NAI Rules, express rights to bring set-off
entitlements in the Swiss Rules 2012 and the close scrutiny of awards by the
International Court of Arbitration of the ICC and by SIAC. In certain areas the
rules might diverge simply because those charged with reform exercises take
legitimately different views about the cost-benefit ratio of certain options.
Taking just one of the differences as an example, providing for terms of reference
may help efficiency by directing the parties’ attention to the key issues from the
earliest stage. However, the process comes with its own cost and may lead to
separate arguments about whether new lines of reasoning are merely extensions
of matters raised within the terms or are in fact new claims that may not be
brought. This and other examples are considered throughout this book. The
point is simply that there may be anticipated differences that then impose
obligations on practitioners to page "203" consider which model may be best for
particular contracts and relationships. Institutional reformers considering these
options are forced to think of implications for their own business as well as the
prospective parties. Thus they will also concern themselves with the potential
impact of differences on the market for institutional services. Some may try and
build a unique model in the hope of attracting business. Others will be concerned
that differences are more likely to deter users. The latter view tends to be the
dominant one. While different institutions are effectively competing for arbitral
business, there is also a trend towards greater collaboration between them. For
example, ICDR has cooperative links with a range of other institutions. This can
help grow global market share without undue infrastructure costs.

Finally, care needs to be taken with some national institutions that may impose
restrictions as to nationality of arbitrators, language of proceedings and venue
for hearings, although in most cases rules have been broadened to meet
international best practice. For example, over the years CIETAC Rules have been
amended to allow foreign nationals as arbitrators, to use languages other than
Chinese and to allow hearings to be conducted outside of key Chinese cities. (258)

3.14.1. The Relationship between Institution, Arbitrator and Parties

This book concentrates on procedure and evidentiary issues within the

arbitration itself. Hence it is only concerned with the relationship between the
institution, arbitrator and parties where that relationship might impact upon
those elements. While it will generally be outside the scope of this book, it is
appropriate to note that an institution is in a contractual relationship with the
parties who employ its services. The key implication for procedure as noted
above is that the institutional offer of services is generally conditional on its rules
applying. Where key rules are concerned, these then cannot be varied by the
parties without institutional consent. Other procedural concerns might arise if
the parties take issue with the procedural conduct of an institution. This would
be a contractual issue under their contract.

As an international contract, the first question in identifying its status and

interpreting it would be to decide what is the proper law of that contract if that
has not been specified. In many legal systems, the institution's marketing would
constitute an offer to provide services on the terms and conditions as publicised.
The offer would be accepted when a notice of arbitration is sent pursuant to an
arbitration agreement that has selected that institution. On this view, the
institution would already be in breach if it refused to provide services at that
stage. An alternative view in some systems is that a contract is only
consummated when the institution accepts its role following the notice, either by
express statement or by necessary implication from its actions. On this view, it is
the notice that constitutes the offer. As to the terms, it can be implied that the
institution promises the page "204" parties that it will have appropriate
oversight of the procedures and provide the necessary technical and
administrative support to the tribunal. For some institutions, that will also
include scrutiny of a draft award. An institution's rules will naturally be
incorporated in any resulting contract between it and the parties. The
communications may also incorporate other provisions by reference, such as
procedural directions and fee schedules.

Where the relationship between the arbitrator and the institution is concerned,
this is less clear cut. There has been some debate as to whether there is a
separate contract between the institution and the arbitrator. Fouchard, Gaillard,
Goldman support the view that if the institution merely acts as an appointing
agency, it is only an agent for that task and is not in a direct contractual
relationship with the arbitrator. (259) Where an institution is fully involved, they
argue that a contract arises. The tribunal consents to that contract by providing
the arbitrator with its rules and through its advice as to the arbitrator's
functions. The arbitrator accepts the arrangement by agreeing to take on the
appointment. (260) This is discussed further in section 2.4.

3.14.2. Arbitral Fees and Institutional Costs

Where fees are concerned, institutional rules will set standard schedules of costs
and fees. In addition to the arbitrator's fees, institutional arbitration requires a
payment to be made to the institution itself. Some institutions set fees based on a
percentage of the amount in dispute. Others set fee entitlements on an hourly
basis. An advantage of a percentage based schedule is that the parties know the
position at the outset and can factor this into their decision to proceed and/or
seek settlement from time to time. Nevertheless, certainty can come at the
expense of flexibility and a fee schedule must provide some leeway depending on
the complexity of a matter, regardless of the amount in dispute. Hence, most
schedules will provide a range and at least indicate maximum and minimum
figures. The parties can presume that the more complex the issues and/or the
lengthier the processes they choose to adopt, the higher in the range the ultimate
fee will be.
A well-established schedule will try and ensure that the costs do not escalate to
an unreasonable percentage of the amount in dispute, although the institution
cannot control the parties' own legal costs. This is a key reason why the total
transaction costs of any arbitration can reach unacceptable proportions. (261) In
this environment, some debate whether the added costs of an institution are
warranted. Merely pointing to the added institutional cost is overly simplistic as
it page "205" depends on whether those same functions would be performed in
any event. In most cases those functions would be performed by an ad hoc
arbitrator, often at significantly added expense. (262) It is also possible that the
tighter fee schedules for arbitrators imposed by institutions more than counters
any direct institutional administrative fees. A clearly defined institutional
structure and rule system also means that time will not be spent on dealing with
as many procedural questions, in turn saving overall expense. Finally,
institutional involvement helps shelter the arbitrators from the process of
negotiating their fees with the parties. Any such negotiation, while proper,
provides some degree of conflict of interests from the outset given that a
prospective arbitrator would wish to reduce total transaction costs of an
arbitration but still must be entitled to try and achieve the level of professional
fees that they desire. (263)

A comparison of the different arbitral fee schedules of the key institutions shows
that some are far more generous in their fee structure than others. There will be
positive and negative benefits of this, in particular through the way that
institutions indirectly compete for arbitral work. Those with tighter fee
schedules might see themselves as more attractive to prospective parties
through more rigorous management of costs. However, more eminent
arbitrators who are the most in demand will tend to prefer more lucrative cases
to those where fees are more tightly controlled. There is also a difference
between institutions that will pay arbitrators on a pro rata basis and those who
hold fees until an award is rendered. These examples suggest that it is important
to understand that any fee system provides various incentives and disincentives
and practitioners and ethical arbitrators must be wary of these. A maximum fee
level regardless of the work undertaken provides some incentive to work
efficiently but also provides a disincentive to at times do extra work that might
otherwise be desirable. An ethical arbitrator would understand that in taking the
appointment they are still undertaking to do whatever is necessary to promote a
fair and efficient outcome and should not vary from that ideal simply because
they have agreed to accept a modest fee structure. Parties who are aware that
they have a complex dispute involving a relatively small amount of money,
should at least ensure that they select arbitrators who they could be confident
would adopt such an ethical stance.

3.14.3. Institutional Control

There is quite a degree of difference between institutions as to the extent to

which they choose to get involved in, and control the process. From the very
beginning, some such as the ICC and CIETAC directly serve the notice of
arbitration on the page "206" respondent and serve the answer to the notice on
the claimant. (264) In other rule systems, the parties are asked to serve key
pleadings directly. (265)

Some institutions take an active role in determining the validity of the

arbitration agreement. The CIETAC Rules indicate that it is CIETAC itself rather
than the tribunal that determines whether there is a valid arbitration agreement
in existence. (266) The prima facie position under the ICC rules is that such
questions are determined by the tribunal, unless the Secretary-General refers the
question to the ICC Court of Arbitration. In that case, the Court makes a
preliminary determination whether there is a prima facie agreement that may
exist, after which any decisions on validity are made by the tribunal. (267) Most
other institutions leave it for the tribunal to determine all validity questions. (268)
After tribunal constitution, most systems then have parties directly serve
pleadings and submissions and copy in the institutional secretariat. A most
significant aspect of control is as to scrutiny of the award, discussed in the
following section.

3.14.4. Scrutiny of the Award

Some institutions take an active role in scrutinising the award. The institutions
are not concerned with evaluating the decision on the merits and are not seeking
to engage in an appellate function. Instead, the institutions are concerned to
review draft awards to see if the terms of reference or memorandum of issues
have been properly addressed, that the language and reasoning is unambiguous,
clear and sufficient and there are no other inconsistencies between the award
structure and the institutional requirements.

Under ICC Rules 2012, a draft award is submitted to the ICC Court. Under Article
33 of the ICC Rules, the Court may ‘lay down modifications as to the form of the
award and, without affecting the arbitral tribunal's liberty of decision, may also
draw its attention to points of substance’. While the Court is not a review body,
any critical observations it makes will allow revisions in order to enhance the
quality of the award. It is important to understand, however, that the Court is not
in any way concerned with directing an arbitrator as to the particular outcome.
The CIETAC Commission also reviews awards (269) as does SIAC. (270) LCIA and
ICDR Rules simply provide that the award is delivered to the institution and then
subsequently communicated to the parties. (271) It is important to understand,
however, that the practice of a particular institution may go beyond the express
page "207" stipulations in the rules. For example, the former General Counsel of
the AAA has indicated that AAA staff will review the form of an award. (272)

3.14.5. Time Limits

Some institutions impose time limits on the rendering of an award. For example,
the ICC Rules require the tribunal to render an award within six months of
signing the Terms of Reference or such later period as may be allowed by the ICC
Court of Arbitration. (273) Time limits may be extended and this is the norm.
Hence they will not typically provide constraints on arbitral jurisdiction. In
extreme cases, an arbitrator failing to meet numerous deadlines may be removed
under powers contained in the rules. This is discussed further in Chapter 5.
3.14.6. Selection of Arbitrators

Some institutions have public lists of national and international arbitrators that
meet specific criteria of expertise. In other cases, institutions might informally
make recommendations from non-publicised lists. The ICC uses national
committees to make recommendations. A number of institutions are prepared to
act as appointing authorities for a fee even where ad hoc arbitration is
concerned.

There are advantages and disadvantages of published lists. If the institution

scrutinises appropriately, the advantage is that a person is only appointed to a
list if they meet the rigorous standards. Parties with less experience of
arbitration can rely on the quality control exercised by the body that developed
the list. Disadvantages include at times unduly restrictive or circular criteria for
inclusion, such as the need to do a certain number of arbitrations in a jurisdiction
before the imprimatur of that jurisdiction can be granted. Another problem is
that a list shields those on the list from competition from outside. There are also
availability problems if the list is too restricted. Institutions around the world
will typically include the same leading arbitrators. While that is natural, a
situation soon arises where those persons have more work than they can handle
and each individual institution cannot truly and directly offer their services.

3.14.7. Liability of Arbitrators

Institutional rules generally provide for immunity to arbitrators as do arbitral

statutes. Even absent such rules, courts have at times implied that as part of the
page "208" inherent functions of an arbitrator. (274) Arbitrator liability is
discussed further in section 5.18.

3.14.8. Immunity for Institutions

Arbitral statutes do not generally provide indemnities to institutions, although

that may occur for some functions. For example, section 74 of the English
Arbitration Act 1996 provides an indemnity for institutions as appointing
authorities. (275) Institutional rules will typically express an indemnity for
arbitrators and the institution itself. The ICC Rules expressly exclude liability for
the institution and its employees. (276)

Given the immunity of judges and arbitrators, it has been suggested that this
would be meaningless if liability simply shifted to the institution. (277) That does
not necessarily follow in all factual circumstances. It should certainly be the case
that the institution should not be argued to be vicariously liable for the decision-
making process of the arbitrator. However, there seems no reason why a
commercial enterprise that negligently fails to perform its own functions in a
timely manner might not, in some circumstances at least, be asked to take
responsibility. Consider for example, an arbitration clause that directs an
institution to be the appointing authority and indicates that the tribunal must be
appointed within a specified time period. If the institution negligently fails to
make the appointment and hence undermines the agreement to arbitrate, there
seems no policy reason to bar redress for the complainant. Other circumstances
are more problematic and there may be differences in view about the cost-
benefit ratio of immunity or liability. Examples would be situations where claims
were made for lack of general timely coordination of the procedure.

In any event, being private commercial entities, institutions are entitled to offer
their services on terms and conditions as they see fit. As noted, these now
invariably include provisions as to immunity. Importantly, however, the validity
and scope of such immunity provisions would need to be looked at as per any
contractual terms under the applicable law. Section 2.4 looks at the contractual
relationship between the parties and the institution. One would need to
determine the applicable law and whether there are any mandatory rules that
might seek to protect the consumer against unreasonable limitations of liability
by a service provider.

page "209"

3.15. Ad Hoc Arbitration

Ad hoc arbitration occurs where no institution is used to coordinate and

supervise the process. Those organising an ad hoc arbitration need to be more
careful in seeking to ensure that an appropriate procedural framework exists.
Any established institution will inevitably ensure that all key elements are
contained within their rules and general powers.

The most significant set of ad hoc rules available for use by any interested
parties are the UNCITRAL Arbitration Rules. These Rules were adopted in 1976
(278) and recently modified in 2010. The UNCITRAL Rules cover a similar

procedural framework to that of the UNCITRAL Model Law. The UNCITRAL Rules
are also suitable for investor/State disputes and disputes between States and
were utilised by the Iran-US Claims Tribunal. Other ad hoc rules available for
general use include the Rules for Non-Administered Arbitration of the
International Institution for Conflict Prevention and Resolution. (279) The
Permanent Court of Arbitration has also developed rules for use in arbitrations
between States, between States and private parties and between international
organisations and private parties. (280)

3.16. Pros and Cons of Ad Hoc Versus Institutional Arbitration

Institutional and ad hoc arbitration are commonly considered to each have

certain advantages and disadvantages. It might be more accurate to state that
institutional and ad hoc arbitration each have certain qualities which may be
advantageous or disadvantageous depending on the interests of the parties and
the circumstances of their dispute. Advantages and disadvantages of each are
addressed separately below, although it is inappropriate to only consider the
differences between ad hoc and institutional arbitration in the extreme without
considering hybrid models. Some essentially ad hoc arbitrations still provide for
resort to be had to an institution for certain key matters such as appointment of
a tribunal.
page "210"

3.16.1. The Advantages of Institutional Arbitration

The two main differences between institutional and ad hoc arbitration involve
control over the process and control of costs. While the attributes of institutional
arbitration obviously depend on the particular institution, one of the main
advantages of institutional arbitration in general is in the use of institutional
arbitration rules. (281) The rules of well-established institutions offer a ‘tested
procedural framework’ , (282) and will usually be effective and up to date. (283)
Institutional arbitration is said to offer ‘convenience, security and administrative
effectiveness’. (284) Also important are the administrative support and the
assistance in the management of the proceedings offered by the institution. (285)
‘They will ensure that the arbitral tribunal is appointed, that advance payments
are made in respect of the fees and expenses of the arbitrators, that time limits
are kept in mind and, generally, that the arbitration is run as smoothly as
possible.’ (286) The oversight of the institution ‘… reduces the risks of procedural
breakdowns, particularly at the beginning of the arbitral process, and of
technical defects in the arbitration proceedings and arbitral award’. (287) In this
way, the institution can ensure procedural efficiency. (288) Institutions can make
arbitration ‘more reliable and expeditious’. (289) Where an arbitrator is less
experienced, technical assistance will also involve an educative function, helping
the arbitrator to understand how various steps ought to be approached.

page "211"

Some institutional rules also impose time limits on the process. Some would
suggest that institutions are more cost effective, although their general
involvement adds costs that do not inherently arise in ad hoc arbitration. As
noted in section 3.16 above, there will be both positive and negative cost
implications of institutional involvement, hence it is difficult to make a blanket
statement as to relative merit in that regard. Institutional arbitration may
enhance the likelihood of voluntary compliance with an arbitral award, (290) as a
result of the ‘cachet’ of the institution, (291) and the particular ‘credibility’ of
awards issued under an institution. (292) In some cases, a specific aspect of
institutional oversight is the ‘quality control’ maintained through the review of
awards for completeness, errors and other drafting issues. (293) Some further
advantages of institutional arbitration are: that the institution can act as an
intermediary in certain decisions (e.g., as to the remuneration of the arbitrators);
(294) that the institution can organise the physical resources for the arbitration

(e.g., the venue for the hearings); and that some institutions are tailored to
specific industries. (295) Some institutions have also developed a particular
expertise with certain geographically based parties. For example, the Arbitration
Institute of the Stockholm Chamber of Commerce deals with a significant number
of cases involving Russia and other members of the former Soviet Union. The
International Arbitral Centre of the Federal Economic Chamber in Vienna also
deals with a significant amount of East-West arbitration.

Another potential advantage of institutional over ad hoc arbitration relates to

reforms of the rules. Because the institutions are commercial entities and are not
concerned with politics and diplomacy, any reforms they make can simply be
directed to bringing their rules into line with world best practice. Many of the
key arbitral institutions make regular amendments to their rules as an ongoing
reform and improvement exercise. In addition to dealing with specific problems
that have arisen from time to time, rule amendments will often introduce or
strengthen key policy norms such as delocalised aspects, party autonomy, fair
and effective institutional control and effective time lines. (296) Conversely, page
"212" UNCITRAL being an intergovernmental process, is slower in its
deliberations and will inevitably tend to more conservative outcomes as
consensus amongst differing legal families and countries of differing levels of
economic development is a precondition to reform. However, that can provide
some benefits in ensuring that any reforms are suitable to a broad range of
potential users rather than those with similar views to the dominant legal
culture in the particular institution's home country.

3.16.2. Disadvantages of Institutional Arbitration

The principal disadvantage of institutional arbitration is said to be its cost, (297)

although as noted above, a proper evaluation would look at which functions
would be performed by ad hoc arbitrators in any event, often at higher cost. This
is not only an issue of the overall cost of institutional arbitration but also of
timing in cases where payment of the fee is required up-front. (298) However, even
added cost must be balanced against the advantages of institutional arbitration.
(299) Other potential disadvantages are the time taken to go through the

administrative process of the institution and the inconvenience of time limits

that might be imposed by the institution. (300) Finally, there may be less flexibility
where institutions are unwilling to allow variations to their rules.

3.16.3. The Advantages of Ad Hoc Arbitration

In comparing the two forms of arbitration, it is important to understand exactly

what is contemplated by ad hoc arbitration. Here there is a difference between
ad hoc arbitration under well respected and recently updated rules such as the
UNCITRAL Rules and instead, a fully ad hoc process where counsel have
purported to design all elements beginning with drafting the arbitration
agreement. In the latter page "213" case, there is the risk that the drafting will be
inadequate. A well constructed ad hoc arbitration cannot by definition be more
or less advantageous than an arbitration supervised by a respected institution.
The parties are free to choose a respected Seat, high quality arbitrators and an
established ad hoc set of procedural rules.

The main advantages of ad hoc arbitration is the flexibility and control of the
procedure by the parties. (301) Two further potential, but not always realised,
advantages are time (where the administrative delays of an institution are
avoided), (302) and cost (where the fees charged by institutions are avoided). (303)
Ad hoc arbitration is also said to be ‘arguably more confidential’, (304) although
this is not inherently so. It is also important to understand that the revised
UNCITRAL Rules 2010 provide for various appointing authority and related
controls as to various matters including fees, costs and expenses. While ad hoc
published rules do not contain schedules of fees as is the case with institutional
arbitration, some rules still give some directions in that regard. For example,
Article 41 of the UNCITRAL Rules 2010 requires the fees to be ‘reasonable in
amount’ taking into account the amount in dispute, the complexity of the subject
matter, the time spent and any other relevant circumstances of the case. If an
appointing authority has been agreed or designated and the authority has issued
a schedule of fees, the tribunal shall ‘take that schedule of fees into account’ in
fixing its fees ‘to the extent that it considers appropriate in the circumstances of
the case’. A party can request the appointing authority to review the statement
setting forth the basis of establishing fees. Where a party so page "214" requests
and the appointing authority, or by default the Secretary-General of the PCA,
determines that the tribunal's determination is inconsistent with its proposal or
is otherwise manifestly excessive, it can make binding adjustments. Peter
Sanders, commenting on an earlier version of these provisions, rightly pointed
out that a party may be very reluctant to utilise these powers at the final stage of
the arbitral proceedings when the tribunal is working on the draft award. (305)

3.16.4. The Disadvantages of Ad Hoc Arbitration

The main risk associated with ad hoc arbitration is the lack of cooperation of one
or more of the parties in the arbitration proceedings. (306) ‘The principal
disadvantage of ad hoc arbitration is that it depends for its full effectiveness on
co-operation between the parties and their lawyers, backed up by an adequate
legal system in the place of arbitration.’ (307) In the event of an uncooperative or
disruptive party, the potential advantages of ad hoc arbitration can be reversed.
The time and cost taken to resolve problems could be significant. (308) The
problems are multiplied where there are shortcomings in the arbitration clause
itself, (309) so careful drafting of the arbitration clause is acutely important in ad
hoc arbitration. (310) The time and effort necessary to establish the procedure for
the ad hoc arbitration should be taken into account, (311) and the arbitration may
take longer without the time limits imposed by an institution, (312) although the
parties are free to impose whatever limits they wish.

3.17. Pre-Arbitral Applications

In many instances a question is of such urgency that directions would wish to be

sought prior to the establishment of a tribunal. Parties may make application to
page "215" appropriate courts in such circumstances. Well-drafted lex arbitri
make it clear that such an application is not a waiver of a right to arbitrate.
Nevertheless, there are problems arising from differing attitudes of courts and
the fact that courts with jurisdiction to order such measures may not be in places
where orders need to be effective. Some institutions have sought to deal with
this through the establishment of pre-arbitral referee procedures. This is
discussed in Chapter 8 alongside interim measures as the principles overlap.

page "216"
1 An exception where a separate agreement is not required is a domestic statute
that allows parties to be compelled to arbitrate. This is outside the scope of this
book.
2 See generally Noah D. Rubins, ‘Betting the Farm on International Arbitration: Is

It Time for an Appeal Option?’, American Revue of International Arbitration 11,

no. 4 (2000): 531.
3 Historically, some jurisdictions only recognised submission agreements,
arguing that consent in advance of knowledge of the actual dispute was not a
true agreement to be bound. Redfern and Hunter suggest that this is still the case
in Argentina and Uruguay. Nigel Blackaby et al., Redfern and Hunter on
International Arbitration, 5th edn (Oxford: Oxford University Press, 2009), 18, n.
69.
4 Bernard G. Poznanski, ‘The Nature and Extent of an Arbitrator's Powers in

International Commercial Arbitration’, Journal of International Arbitration 4, no.

3 (1987): 83.
5 It is sometimes also described as severability.
6 The doctrine of separability was recently upheld by the House of Lords in Fiona

Trust & Holding Corporation & Ors v. Privalov & Ors [2007] UKHL 40 (17 October
2007).
7 Kompetenz-kompetenz in German.
8 For example, Gossett v. Carapelli, Cour de Cassation, 7 May 1953 [1953] Dalloz

545 accepted competence in France.

9 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on
International Commercial Arbitration (The Hague: Kluwer Law International,
1999), 214.
10 See, e.g., UNCITRAL Model Law Art. 16(1), ICC Rules 2012 Art. 6(4), and

UNCITRAL Rules 2010 Art. 23.1. For the uncertain historical position in China
see Gu Weixia, ‘China's Search for Complete Separability of the Arbitral
Agreement’, Asian International Arbitration Journal 3, no. 2 (2007): 163.
11 Prima Paint Corp v. Flood and Conklin Mfg Co. 388 US 395 (1967); Emmanuel

Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on International

Commercial Arbitration (The Hague: Kluwer Law International, 1999), 197, para.
391.
12 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 201; K.P. Berger, International Economic Arbitration (Deventer: Kluwer
Law and Taxation Publishers, 1993), 121; Julian Lew, Loukas Mistelis & Stefan
Kröll, Comparative International Commercial Arbitration (The Hague: Kluwer
Law International, 2003), 106.
13 For an argument in favour of a distinct approach to contract existence
questions see George A. Bermann, ‘The “Gateway” Problem in International
Commercial Arbitration’, in Liber Amicorum Eric Bergsten: International
Arbitration and International Commercial Law: Synergy, Convergence and
Evolution, ed. Kröll et al. (Alphen aan den Rijn: Kluwer Law International, 2011),
65.
14 Commenting in relation to Art. 21, para. 2 of the UNCITRAL Rules adopted in

1976, Fouchard, Gaillard and Goldman state that ‘[c]ontrary to what has
sometimes been suggested, this text does not distinguish between the various
kinds of jurisdictional challenges which can be put forward as a consequence of
the flaws affecting the main contract. In particular, no distinction is to be made
between situations where the action is based on the purported absence or
nonexistence of the main contract, and those where it is based on the allegation
that such contract is void.’ Emmanuel Gaillard & John Savage (eds), Fouchard,
Gaillard, Goldman on International Commercial Arbitration (The Hague: Kluwer
Law International, 1999), 201.
15 Their view is supported by Simon Greenberg, Christopher Kee & J. Romesh

Weeramantry, International Commercial Arbitration: An Asia-Pacific Perspective

(New York: Oxford University Press, 2011), 156–157.
16 Chartered Institute of Arbitrators, Practice Guideline 6: ‘Guidelines for
Arbitrators Dealing with Jurisdictional Problems in International Cases’, para
5.5.2.
17 See, e.g., s. 7 English Arbitration Act 1996.
18 It is interesting that the debate is at times confined to the notion of
separability, as it ultimately pertains to competence and whether a tribunal can
even begin to deal with a case where the contract is disputed.
19 The situation would be different if the lex arbitri provision is mandatory, but

this would rarely be so.

20 If all relevant connecting factors were considered, that may also make it more

difficult to decide on jurisdiction as a preliminary matter, before all evidence is at

hand.
21 C v. D 2007 EWCA Civ 1282 at [22]; Bernard Hanotiau, ‘What Law Governs the

Issue of Arbitrability?’, Arbitration International 12, no. 4 (1996): 391; Julian

D.M. Lew, ‘The Law Applicable to the Form and Substance of the Arbitration
Clause’, in Improving the Efficiency of Arbitration Agreements and Awards: 40
Years of Application of the New York Convention, ICCA Congress Series No. 9, ed.
Albert Jan van den Berg (The Hague: Kluwer Law International, 1999), 114.
22 Klaus Peter Berger, ‘Re-examining the Arbitration Agreement: Applicable Law

– Consensus or Confusion?’, in International Arbitration 2006: Back to Basics?,

ICCA Series No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer Law
International, 2007), 301.
23 Julian D.M. Lew, ‘The Law Applicable to the Form and Substance of the

Arbitration Clause’, in Improving the Efficiency of Arbitration Agreements and

Awards: 40 Years of Application of the New York Convention, ICCA Congress Series
No. 9, ed. Albert Jan van den Berg (The Hague: Kluwer Law International, 1999),
114.
24 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 221.
25 Ibid., 231.
26 Marc Blessing, ‘The Law Applicable to the Arbitration Clause and
Arbitrability’, in Improving the Efficiency of Arbitration and Awards: Forty Years of
Application of the New York Convention, ICCA Congress Series 9, ed. Albert van
den Berg (The Hague: Kluwer Law International, 1999), 169.
27 Ibid., 170.
28 See Menicucci: Paris, 13 decembre 1975, Revue de l'arbitrage (1977): 147 and

note by P.H. Fouchard; Jean-François Poudret & Sébastien Besson, Comparative

Law of International Arbitration 2nd edn (London: Sweet & Maxwell, 2007), 145
criticise the decision although they note that it is supported by the majority of
French authors. See also Judgment of 8 July 2009, Soerni v. ASB, Cour de
Cassation.
29 Judgment of 20 December 1993, Comité populaire de la municipalité de Khoms

El Mergeb v. Dalico Contractors, Cour de Cassation, Journal du Droit International

(1994): 432.
30 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 147. The authors also
note the scholarly debate among French authors on this issue and the historical
reasons why such an approach may have been taken to overcome constraints of
French law.
31 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 259.

32 See, e.g., Swedish Arbitration Act s. 45. Gary B. Born, International Commercial

Arbitration (The Hague: Kluwer Law International, 2009), 470–475. China

applies the law of the seat. Simon Greenberg, Christopher Kee & J. Romesh
Weeramantry, International Commercial Arbitration: An Asia-Pacific Perspective
(New York: Oxford University Press, 2011), 161.
33 Swiss Private International Law Art. 178(2) PIL.
34 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 257.
35 Ibid., 256–258.
36 Ibid., 258, n. 79.
37 These scenarios are considered in Chapter 7.
38 See, e.g., Corfu Channel Case 1949 ICJ Reports 24.
39 Ambatielos [1953] ICJ Reports 10; Case Concerning Rights of United States

Nationals in Morocco [1952] ICJ Reports 196; Yearbook of the International Law
Commission (1966-II) 219 as cited in Ian Sinclair, The Vienna Convention on the
Law of Treaties (1984) 124.
40 ICC Case No. 2321 (1974), as extracted in Pieter Sanders (ed.), Yearbook of

Commercial Arbitration Vol I (The Hague: Kluwer Law International, 1976), 133.
41 Emmanuel Gaillard & John Savage (ed.), Fouchard, Gaillard, Goldman on
International Commercial Arbitration (The Hague: Kluwer Law International,
1999), 266–268.
42 For an example of selecting a law to render the arbitration clause valid see

Interim Award in ICC Case No. 4145, in Albert Jan van den Berg (ed.),Yearbook of
Commercial Arbitration Vol XII (The Hague: Kluwer Law International, 1987), 97;
Preliminary Award in ICC Case No. 5505, in Albert Jan van den Berg (ed.),
Yearbook of Commercial Arbitration Vol XIII (The Hague: Kluwer Law
International, 1988), 110; Partial Award in ICC Case No. 7920, in Albert Jan van
den Berg (ed.), Yearbook of Commercial Arbitration Vol XXIII (The Hague: Kluwer
Law International, 1998), 80.
43 Moses H Cone Memorial Hospital v Mercury Construction Corp 460 US 1, 24–25,

(1983).
44 Journal du Droit International (1975): 934.
45 Ibid., 938.
46 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 259–260.
47 Michael J. Mustill & Stewart C. Boyd, Commercial Arbitration, 2nd edn
(London: Butter-worths, 1990), 66.
48 This section draws on and extracts from Michael Pryles & Jeff Waincymer,

‘Multiple Claims in Arbitrations Between the Same Parties’, Fifty Years of the New
York Convention: ICCA International Arbitration Conference, Congress Series No.
14, ed. Albert Jan van den Berg (Alphen aan den Rijn: Wolters Kluwer, 2009), and
is included with the permission of the coauthor.
49 Heyman v. Darwins Ltd [1942] AC 359, 399.
50 See the decision of the Supreme Court of South Australian in Main Electrical

Pty Ltd v. Civil & Civic Pty Ltd [1978] 19 SASR 34.
51 See Printing Machinery Co. Ltd v. Linotype and Machinery Ltd [1912] 1 Ch 566.
52 There are many Australian cases examining whether claims under the former

Trade Practices Act 1974 (Cth) are arbitrable. See, e.g., the early decisions in
White Industries v. Trammel (1983) 51 ALR 779 and Allergan Pharmaceuticals
Inc. v. Bausch andLombe Inc. (1985) ATPR 40–636. Compare the more recent
approach in Comandate Marine Corp. v. Pan Australia Shipping Pty Ltd (2006)
238 ALR 457. As to claims involving breach of competition laws see the liberal
decisions of the United States Supreme Court in Mitsubishi Motors Corp. v. Soler
Chrysler-Plymouth Inc. 473 US 614 (1984) and the decision of the High Court of
New Zealand in Attorney-General v. Mobil Oil NZ Ltd [1989] 2 NZLR 649.
53 Premium Nafta Products Limited v Fili Shipping Company Limited [2007] UKHL

40.
54 Lord Hoffmann concluded at paragraph 12 that the language of the arbitration

clause contained nothing to exclude disputes about the validity of the contract,
whether on the grounds that it was procured by fraud, bribery,
misrepresentation or anything else.
55 Decision on jurisdiction, Amco Asia Corp and Others v. Republic of Indonesia, 23

ILM 351 (1984) 359.

56 Lucky-Goldstar International (HK) Ltd v. Ng Moo Kee Engineering Ltd, in Albert

Jan van den Berg (ed.) Yearbook of Commercial Arbitration XX (The Hague:
Kluwer Law International, 1995), 280.
57 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 282–283.

58 For different arbitrability categories see Julian Lew, Loukas Mistelis & Stefan

Kröll, Comparative International Commercial Arbitration (The Hague:Kluwer Law

International, 2003), 201–219.
59 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 312, referring to French
law.
60 German Code of Civil Procedure Art. 1030(2).
61 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 313–314.

62 A specific example of legislation barring certain forms of arbitrability is s. 11

of Australia's Carriage of Goods by Sea Act 1991 if contracts call for arbitration in
respect of a bill of lading or similar document outside of Australia. The policy
reason is that in certain areas countries have very differing views on matters
such as liability as between shippers and carriers and are concerned with
potential biases in foreign jurisdictions.
63 Julian Lew, Loukas Mistelis & Stefan Kröll, Comparative International
Commercial Arbitration (The Hague: Kluwer Law International, 2003), 196.
64 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 363.
65 New York Convention Art. V(2)(a).
66 One problematic scenario where this logic may not pertain is in insolvency,

where the controlling jurisdiction does want to have power over all assets so as
to provide an orderly collection and distribution for the benefit of creditors. This
is discussed in section 7.15.
67 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 284–285.

68 See, e.g., Sweden Arbitration Act Art. 1(1).
69 Swiss PILA Art. 177(1).
70 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 289–290.

71 German Code of Civil Procedure Art. 1030(1).
72 Eco Swiss China Time Ltd v. Benetton International NV, Case C-126/97, ECR

1991, I-03055.
73 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 303–305.

74 Chartered Institute of Arbitrators, ‘Guidelines for Arbitrators Dealing with

Jurisdictional Problems in International Cases’, 121.

75 Insolvency issues can also impact significantly on interim measures scenarios.
76 For example, the European Convention on State Immunity Art. 12(1).
77 This was the view taken by the ICJ in Eco Swiss China Time Ltd v. Benetton

International NV, Case C-126/97, ECR 1991, I-03055.

78 Julian D.M. Lew, ‘The Law Applicable to the Form and Substance of the

Arbitration Clause’, Improving the Efficiency of Arbitration Agreements and

Awards: 40 Years of Application of the New York Convention, ICCA Congress Series
No. 9, ed. Albert Jan van den Berg (The Hague: Kluwer Law International, 1998),
116.
79 This is expressed in Swiss Private International Law Art. 177(2). See further P.

Dunham & S. Greenberg, ‘Balancing Sovereignty and the Contractors Rights in

International Construction Arbitrations Involving State Entities’, International
Construction Law Review 23, no. 2 (2006): 130.
80 As per the view of Jean-François Poudret & Sébastien Besson, Comparative

Law of International Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 155.
81 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 377.
82 An UNCITRAL Working Party identified numerous circumstances where
jurisdiction had been accepted, notwithstanding the lack of strict compliance; see
United Nations Commission on International Trade Law, Working Group II
(Arbitration) Forty-fourth session, New York, 23– 27 January 2006 Settlement of
commercial disputes - Preparation of uniform provisions on written form for
arbitration agreements - Article II(2) of the Convention on the Recognition and
Enforcement of Foreign Arbitral Awards (New York, 1958), 18 UN Doc.
A/CN.9/WGII/ WP.139.
83 See, e.g., Neil Kaplan, ‘Is the Need for Writing as Expressed in the New York
Convention and the Model Law Out of Step with Commercial Practice?’,
Arbitration International 12, no. 1 (1996): 27.
84 See, e.g., Netherlands Code of Civil Procedure Art. 1021.
85 See, e.g., German Code of Civil Procedure Art. 1031(2).
86 For example s. 5(3) English Arbitration Act, 1996, which provides that ‘where

parties agree otherwise than in writing by reference to terms which are in

writing, they make an agreement in writing.’
87 See, e.g., Drafting Dispute Resolution Clauses – A Practical Guide (2004) AAA

International Centre for Dispute Resolution available at <www.adr.org>; ICC

Suggested Clauses are available at <www.iccwbo.org>; CEDR Clauses and Rules
available at <www.cedr.co.uk>; LCIA Recommendations available at <www.lcia-
arbitration.com>; CPR Institute for Dispute Resolution available at
<www.cpradr.org>.
88 Julian D.M. Lew, ‘The Law Applicable to the Form and Substance of the

Arbitration Clause’, in Improving the Efficiency of Arbitration Agreements and

Awards: 40 Years of Application of the New York Convention, ICCA Congress Series
No. 9, ed. Albert Jan van den Berg (The Hague: Kluwer Law International, 1999),
128.
89 IBA Guidelines for Drafting International Arbitration Clauses, International

Bar Association, 2010, 8.

90 The IBA Drafting Guidelines para. 14 seem to suggest that the latter
expression is broader.
91 Simon Greenberg, Christopher Kee & J. Romesh Weeramantry, International

Commercial Arbitration: An Asia-Pacific Perspective (New York: Oxford University

Press, 2011), 191.
92 See, e.g., Naviera Amazonica Peruana, SA v.Compania Internationale de Seguros

del Peru [1988] 1 Lloyd's Rep 116 (CA).

93 Julian Lew, Loukas Mistelis & Stefan Kröil, Comparative International
Commercial Arbitration (The Hague: Kluwer Law International, 2003), 158.
94 IBA Guidelines for Drafting International Arbitration Clauses, International

Bar Association, 2010, 16.

95 Ibid., 15–16.
96 Ibid., 18.
97 This would be particularly problematic if those arbitrators were not willing or

able to undertake tasks, in which case the arbitration clause may fail in its
entirety.
98 This is discussed further in section 13.6.
99 IBA Guidelines for Drafting International Arbitration Clauses, International

Bar Association, 2010, 19.

100 This is discussed further in section 13.14.
101 IBA Guidelines for Drafting International Arbitration Clauses, International

Bar Association, 2010, 11.

102 Jack J. Coe, ‘Pre-hearing Techniques to Promote Speed and Cost-
Effectiveness: Some Thoughts Concerning Arbitral Process Design’, Pepperdine
Dispute Resolution Law Journal 2, no. 1 (2002): 61.
103 IBA Guidelines for Drafting International Arbitration Clauses, International

Bar Association, 2010, 16.

104 Ibid., 24.
105 Jane Jenkins & Simon Stebbings, International Construction Arbitration Law
(The Hague: Kluwer Law International, 2006), 84.
106 English Arbitration Act 1996 ss 67–69.
107 William W. Park, ‘When and Why Arbitration Matters’, Mealey's International

Arbitration Report 13, no. 1 (1998): 25.

108 See cases cited in William W. Park, ‘When and Why Arbitration Matters’,

Mealey's International Arbitration Report 13, no. 1 (1998): n66.

109 Unilateral clauses have been accepted by the English High Court in NB Three

Shipping Ltd v. Herebell Shipping Ltd [2004] All ER (D) 152. For a survey of the
attitude of other jurisdictions to such clauses see Simon Nesbitt & Henry Quinlan,
‘The Status and Operation of Unilateral or Optional Arbitration Clauses’,
Arbitration International 22, no. 1 (2006): 133.
110 See, e.g., Liverpool City Council v. Casbee Pty Ltd 2005 [NSWSC 590].
111 See Arthur W. Rovine, ‘Introduction to Session on Issues in Integrated
Dispute Resolution Systems’, in New Horizons in International Commercial
Arbitration and Beyond, ICCA Congress Series No. 12, ed. Albert Jan van den Berg
(The Hague: Kluwer Law International, 2005), 439.
112 Arthur W. Rovine, ‘Introduction to Session on Issues in Integrated Dispute

Resolution Systems’, in New Horizons in International Commercial Arbitration

and Beyond, ICCA International Arbitration Congress Series No. 12, ed. Albert Jan
van den Berg (The Hague:K-luwer Law International, 2005), 439.
113 The desirability of escalation clauses may also be considered in the context of

the Price Waterhouse/Queen Mary Study where those favouring arbitration over
litigation nevertheless wished to employ arbitration in combination with other
ADR mechanisms. Chapman notes that this was identified in the 2006 Study but
apparently not repeated in the 2008 Study. Simon Chapman, ‘Multi-Tiered
Dispute Resolution Clauses: Enforcing Obligations to Negotiate in Good Faith’,
Journal of International Arbitration 27, no. 1 (2010): 90.
114 Klaus Peter Berger, ‘Law and Practice of Escalation Clauses’, Arbitration

International 22, no. 1 (2006): 4.

115 James H. Carter, ‘Part I – Issues Arising from Integrated Dispute Resolution

Clauses’, in New Horizons in International Commercial Arbitration and Beyond,

ICCA International Arbitration Congress Series No. 12, ed. Albert Jan van den
Berg (2004), 446,462.
116 Arthur W. Rovine, ‘Introduction to Session on Issues in Integrated Dispute

Resolution Systems’ in New Horizons in International Commercial Arbitration and

Beyond, ICCA International Arbitration Congress Series No. 12, ed. Albert Jan van
den Berg (The Hague: Kluwer Law International, 2005), 443.
117 IBA Guidelines for Drafting International Arbitration Clauses, International

Bar Association, 2010, 31.

118 Ibid.
119 See, e.g., UNCITRAL Model Law on International Conciliation Art. 11(d).
120 The English case of Cable & Wireless v IBM United Kingdom Ltd [2002] EWHC

2059 considered a structured clause preventing litigation unless a party ‘has

reasonable cause to do so to avoid damage to its business or to protect or
preserve any right of action it may have’.
121 See James H. Carter, ‘Part I – Issues Arising from Integrated Dispute
Resolution Clauses’, in New Horizons in International Commercial Arbitration and
Beyond, ICCA International Arbitration Congress Series No. 12, ed. Albert Jan van
den Berg (The Hague: Kluwer Law International, 2004), 463.
122 Klaus Peter Berger, ‘Law and Practice of Escalation Clauses’, Arbitration
International 22, no. 1 (2006): 6.
123 For an analysis of a range of scenarios in the context of enforcement of

contractual obligations to negotiate see E. Allan Farnsworth, Farnsworth on

Contracts, 2nd edn (New York: Aspen Publishers, 1998).
124 ICC Case No 6276, ICC International Court of Arbitration Bulletin 14, no. 2

(2003): 76.
125 Klaus Peter Berger, ‘Law and Practice of Escalation Clauses’, Arbitration

International 22, no. 1 (2006): 12.

126 See, e.g., ICC Case No. 7422 of 1996 cited in James H. Carter, ‘Part I - Issues

Arising from Integrated Dispute Resolution Clauses’, in New Horizons in

International Commercial Arbitration and Beyond, ICCA International Arbitration
Congress Series No. 12, ed. Albert Jan van den Berg (The Hague: Kluwer Law
International, 2004), 462–63.
127 Eric Van Ginkel, ‘The UNCITRAL Model Law on International Commercial

Conciliation: A Critical Appraisal’, Journal of International Arbitration 21, no. 1

(2004): 4. The New South Wales Court of Appeal upheld an obligation to
undertake genuine and good faith negotiations in United Group Rail Services Ltd
v. Rail Corp. New South Wales [2009] NSW CA 177.
128 Charles Jarrosson, ‘Observations on Poiré v. Tripier’, Arbitration International

19, no. 3 (2003): 368.

129 Judgment of 23 April 2001, Obergericht des Kantons Thurgau, ASA Bulletin

21, no. 2 (2003), 418–420.

130 Courtney & Fairbairn Ltd v. Tolaini Brothers (Hotels) Ltd [1975] 1 All ER 716

at 720 (per Lord Denning).

131 Howtek, Inc. v. Relisys 958 F. Supp 46 (D.N.H. 1997).
132 Elizabeth Bay Development Pty Ltd v. Boral Building Services Pty Ltd (1995) 36

NSWLR 709.
133 Julian Lew, Loukas Mistelis & Stefan Kröll, Comparative International
Commercial Arbitration (The Hague: Kluwer Law International, 2003), 158
referring to Stockholm Institute, Interim Award of 17 July 1992.
134 See, e.g., POS Media Online Ltd v. Queensland Investment Corporation [2000]

FCA 1154, para. 29.

135 Martin Hunter, ‘Commentary on Integrated Dispute Resolution Clauses’, in

New Horizons in International Commercial Arbitration and Beyond, ICCA

International Arbitration Congress Series No. 12, ed. Albert Jan van den Berg
(The Hague: Kluwer Law International, 2004), 470. See also Michael Pryles,
‘Multi-Tiered Dispute Resolution Clauses’, Journal of International Arbitration 18,
no. 1 (2001): 159.
136 Martin Hunter, ‘Commentary on Integrated Dispute Resolution Clauses’, in

New Horizons in International Commercial Arbitration and Beyond, ICCA

International Arbitration Congress Series No 12, ed. Albert Jan van den Berg (The
Hague: Kluwer Law International, 2004), 476.
137 An example of a clause that retains discretion in the arbitrator is the Baltic

and International Maritime Council's Proposal available at

<www.bimco.org/Chartering/BIMCO%20Clauses/Dispute_Resolution_Clauses.as
px>.
138 Arthur W. Rovine, ‘Introduction to Session on Issues in Integrated Dispute
Resolution Systems’, in New Horizons in International Commercial Arbitration
and Beyond, ICCA International Arbitration Congress Series No. 12, ed. Albert Jan
van den Berg (The Hague: Kluwer Law International, 2004), 443.
139 [1999] NSW SC 996.
140 See, e.g., CMS Gas Transmission Company v.The Republic of Argentina, Decision

on Objections to Jurisdiction, ICSID Case No. ARB/01/8, IIC 64 (2003).

141 See, e.g., Goetz v.Burundi, ICSID ARB/05/3, Award 10 February 1999, paras

92–93.
142 ICSID Case No. ARB/00/9, Award 13 September 2003.
143 Ibid., para 14.5.
144 Sarah François-Poncet & Caline Nouawad, ‘So You Want to Start an
Investment Treaty Arbitration? Getting the Notice of Dispute Right’, in 50 Years
of the New York Convention: ICCA International Arbitration Conference, ICCA
Series No. 14, ed. Albert Jan van den Berg (Alphen aan den Rijn: Wolters Kluwer,
2009), 167, 178, 184.
145 A number of such clauses have been extracted by Hanotiau. Bernard
Hanotiau, Complex Arbitrations, Multi Party, Multi Contract, Multi-Issues and Class
Actions (The Hague: Kluwer Law International, 2006), 101–162.
146 IBA Guidelines for Drafting International Arbitration Clauses 2010, para. 108.
147 Gerold Herrmann, ‘Power of Arbitrators to Determine Procedures under the

UNCITRAL Model Laws’, in Planning Efficient Arbitration Proceedings/The Law

Applicable in International Arbitration, ICCA Congress Series No. 7, ed. Albert Jan
van den Berg (The Hague: Kluwer Law International, 1996), 42.
148 Article V(1)(d) New York Convention.
149 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 89.

150 Where the latter issues are concerned, this may often be impractical as a

neutral seat will have little extraterritorial power over persons and information
in other jurisdictions.
151 An expression used in Union of India v.McDonnell Douglas Inc. [1993] 2

Lloyd's Rep 48. (QB).

152 Treating a legal Seat as a conceptual rather than a practical question also aids

in support of Internet-based arbitration where it would otherwise be difficult to

identify a physical centre for the process. Gabrielle Kaufmann-Kohler,
‘Globalization of Arbitral Procedure’, Vanderbilt Journal of Transnational Law 36,
no. 4 (2003): 1318.
153 William W. Park, ‘Procedural Evolution in Business Arbitration: Three
Studies in Change’, in William W. Park, Arbitration of International Business
Disputes: Studies in Law and Practice, (Oxford: Oxford University Press, 2006).
Section 3 of the English Arbitration Act 1996 expressly indicates that the Seat is a
juridical concept.
154 Gabrielle Kaufmann-Kohler, ‘Globalization of Arbitral Procedure’, Vanderbilt

Journal of Transnational Law 36, no. 4 (2003): 1318.

155 See Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 643, citing several relevant cases.
156 For an analysis of delocalised arbitration see Jean-François Poudret &
Sébastien Besson, Comparative Law of International Arbitration, 2nd edn
(London: Sweet & Maxwell, 2007), 91; Emmanuel Gaillard & John Savage (eds),
Fouchard, Gaillard, Goldman on International Commercial Arbitration (The Hague:
Kluwer Law International, 1999), 643.
157 ICSID Convention Art. 44.
158 ICSID Convention Art. 54.
159 Section 4.6 below looks more generally at commencement of arbitration

without a designated Seat.

160 This is the case under the LCIA Rules. See LCIA Rules Art. 16.1.
161 ICDR Rules Art. 13.
162 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 112.

163 ASA Special Series No. 31, June 2009, International Arbitration ‘Do's and

Don'ts’, 6th edn. Poudret and Besson make the valid observation that logistical
factors are less relevant given that hearings can be conducted elsewhere than at
the Seat. Jean-François Poudret & Sébastien Besson, Comparative Law of
International Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 119.
164 William W. Park, ‘When and Why Arbitration Matters’, Mealey's International

Arbitration Report 13, no. 1 (1998): 25.

165 The only argument to the contrary would be a financial one from the
perspective of the government of the Seat. If the parties are not truly engaging in
arbitration in the jurisdiction and hence not adding directly to the economy,
would the government want the court system used when problems arise,
particularly where these are not funded on a user pays basis?
166 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 103.

167 Ibid., 112 referring to the situation in Switzerland.
168 An example occurred with arbitrations in Hong Kong while SARS was at its

height.
169 The UNCITRAL Model Law was approved by the UN General Assembly in

Resolution No. 40/72, 1985 and was subsequently amended in 2006.

170 The UNCITRAL website lists the jurisdictions which have based their
legislation on the Model Law; the list can be accessed at
<www.uncitral.org/uncitral/en/uncitral_texts/arbitr...>.
171 The most notable example is the English Arbitration Act 1996.
172 See, e.g., Belgian Judicial Code; French New Code of Civil Procedure; German

Code of Civil Procedure; Dutch Code of Civil Procedure.

173 Note of the Secretariat on the Possible Future Work in the Area of
International Commercial Arbitration, UN Doc. A/CN.9/460 XXX UNCITRAL YB
395 (1999).
174 UNCITRAL Model Law 2006 Revision, Art. 2 A.
175 Article 7.
176 Article 17 and 17A-J.
177 Article 35.
178 See APRAG Report on 43rd and 44th Sessions (9 January 2006).
179 For example, recent Australian developments include the advent of the

Commercial Arbitration Bills, which are intended to create a uniform arbitration

regime in Australia which is aligned more with the international model.
180 Eric A. Schwartz, ‘Choosing between Broad Clauses and Detailed Blueprints’,

in Improving the Efficiency of Arbitration and Awards: Forty Years of Application

of the New York Convention, ICCA Congress Series 1998, Paris 9, ed. Albert Jan
van den Berg (The Hague: Kluwer Law International, 1999), 105, 106.
181 UNCITRAL Model Law Art. 1(4).
182 See generally Jean-François Poudret & Sébastien Besson, Comparative Law of

International Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 30–35.
183 See, e.g., Hong Kong (with the passage of the new Arbitration Ordinance in

2010); Australia (with the Commercial Arbitration Bill 2010 (Cth)).

184 Mexico – Anti-Dumping Investigation of High Fructose Corn Syrup (HFCS)

from the United States WT/DS132/R. For further discussion on these issues see
Martin Hunter & Alexei Barbuk, ‘Reflections on the Definition of an
“Investment”’, in Global Reflections on International Law, Commerce and Dispute
Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al.
(Paris: ICC Publishing, 2005), 381.
185 Simon Greenberg, Christopher Kee & J. Romesh Weeramantry, International

Commercial Arbitration: An Asia-Pacific Perspective (New York: Oxford University

Press, 2011), 167.
186 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 256.

187 New York Convention Art. II (2).
188 Note accompanying the 2006 UNCITRAL Model Law United Nations
Publications, Sales No. E08.V.4 (2008), 28.
189 See, e.g., Simon Greenberg, Christopher Kee & J. Romesh Weeramantry,

International Commercial Arbitration: An Asia-Pacific Perspective (New York:

Oxford University Press, 2011), 147.
190 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 145.

191 Recommendation regarding the interpretation of Art. II, para. 2, and Art. VII,

para. 1, on the Convention on the Recognition and Enforcement of Foreign

Arbitral Awards adopted by UNCITRAL, 7 July 2006, 39th Session.
192 A number of scholars have questioned whether an appeal mechanism would

be appropriate, particularly in the context of investment disputes, where

complex interpretation of principles of customary and treaty law do not find
consistent treatment from a range of tribunals. The Appellate Body model of the
WTO may have been an inspiration to some commentators. William H. Knull III &
Noah D. Rubins, ‘Betting the Farm on International Arbitration: Is it Time for an
Appeal Option?’, American Revue of International Arbitration 11, no. 4 (2000):
531.
193 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2660–2670, 2734–2736. US appeal rights were denied in

Hall Street Associates LLC v. Mattel Inc. (2008) 552 US 576 (S. Ct.). Such
approaches have been supported in New Zealand (see Arbitration Act 1996 s. 6
and cl. 5 of Schedule 2) and Switzerland (Private International Law Art. 191). A
Paris Court of Appeals decision held that an arbitration clause was invalid when
it purported to grant appeal rights to a court. See Emmanuel Gaillard & John
Savage (eds), Fouchard, Gaillard, Goldman on International Commercial
Arbitration (The Hague: Kluwer Law International, 1999), 269. See also William
H. Knull III & Noah D. Rubins, ‘Betting the Farm on International Arbitration: Is It
Time for an Appeal Option?’, American Revue of International Arbitration 11, no.
4 (2000): 531.
194 See, e.g., the Paris Court of Appeal approach to that effect in Société de Diseno
v. Société Mendes, Paris Cour d'Appel 27 October 1994, Revue de l'Arbitrage
(1995): 263. An appeal provision was accepted on appeal in LaPine Technology
Corporation v Kyocera Corporation 97 C.D.O.S. 9183.
195 The application of mandatory substantive laws is discussed in section 13.8.
196 See section 3.8.
197 For example, see ICC Rules 2012 Art. 22(4): ‘(t)he Arbitral Tribunal shall act

fairly and impartially and ensure that each party has a reasonable opportunity to
present its case’; UNCITRAL Arbitration Rules 2010 Art. 17(1) ‘the arbitral
tribunal may conduct the arbitration in such manner as it considers appropriate,
provided that the parties are treated with equality and that at an appropriate
stage of the proceedings each party is given a reasonable opportunity to present
its case’; LCIA Rules Art. 14 the tribunal must ‘act fairly and impartially as
between all parties, giving each a reasonable opportunity of putting its case and
dealing with that of its opponent’; ICDR Rules Art. 16 ‘subject to these rules, the
tribunal may conduct the arbitration in whatever manner it considers
appropriate, provided that the parties are treated with equality and that each
party has the right to be heard and is given a fair opportunity to present its case’.
The CAA Rules speak of a ‘full’ opportunity in the English version at least. Art. 23
of the Arbitration Rules of the Chinese Arbitration Association, Taipei, available
at <www.arbitration.org.tw/english/image/Arbitration/C...>.
198 UNCITRAL Rules 2010 Art. 17.1.
199 Transnational public policy is discussed in section 3.12 below.
200 Quoted in Tobias Zuberbühler, Christoph Müller & Philipp Habegger (eds),

Swiss Rules of International Arbitration Commentary (The Hague: Kluwer Law

International, 2005), 246– 247.
201 Fernando Mantilla-Serrano, ‘Towards a Transnational Procedural Public
Policy’, Arbitration International 20, no. 1 (2004): 348, n. 81.
202 See, e.g., Arts 15 (independence and impartiality), 19 (procedural rules), 27

(service of procedure), 36, 39 (conservation provisions) and 52 (Non-

contentious Matters Law).
203 Catherine Li, ‘The New Arbitration Law of Taiwan: Up to an International

Level?’, Journal of International Arbitration 16, no. 3 (1999): 132.

204 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 644.
205 Michael J. Mustill & Stewart C. Boyd, Commercial Arbitration, 2nd edn
(London: Butter-worths, 1990), 64.
206 Karl-Heinz Böckstiegel, ‘The Relevance of National Arbitration Law for
Arbitrations under the UNCITRAL Rules’, Journal of International Arbitration 1,
no. 3 (1984): 234.
207 Albert Jan van den Berg (ed.), Yearbook of Commercial Arbitration Vol XV (The

Hague: Kluwer Law International, 1990), 407.

208 Eco Suisse China Time Ltd v.Benetton International NV Case C-126/97, ECR

1991, I-03055.
209 Karl-Heinz Böckstiegel, ‘The Relevance of National Arbitration Law for
Arbitrations under the UNCITRAL Rules’, Journal of International Arbitration 1,
no. 3 (1984): 229.
210 German procedural law previously expressly allowed this but has now
amended the situation through s. 1025 of the German Civil Code of Procedure.
See Gabrielle Kaufmann-Kohler, ‘Globalization of Arbitral Procedure’, Vanderbilt
Journal of Transnational Law 36, no. 4 (2003): 1316.
211 In any case where there is a potential conflict of jurisdiction the converse is

also a potential problem where both courts might feel that they are not the
appropriate forum.
212 See, e.g., Art. 176 Swiss PILA.
213 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 902.
214 See, e.g., Sergio Le Pera, ‘Where to Vacate and How to Resist Enforcement of

Foreign Arbitral Awards: International Standard Electric Corporation v Bridas

Sociedad Anónima Petrolera, Industrial y Comercial’, American Review of
International Arbitration 2 (1991): 48, 55–56.
215 See, e.g., Naviera Amazonica Peruana, SA v. Compania Internationale de
Seguros del Peru [1988] 1 Lloyd's Rep 116 (CA).
216 W. Laurence Craig, ‘Some Trends and Developments in the Laws and Practice

of International Commercial Arbitration’, Texas International Law Journal 30, no.

1 (1995): 45; and see, e.g., National Thermal Power Corp v. Singacorp, in Albert
Jan van den Berg (ed.), Yearbook of Commercial Arbitration Vol 18 (The Hague:
Kluwer Law International, 1993), 403. For criticisms of this approach see Jan
Paulsson, ‘The New York Convention's Misadventures in India’, Mealey's
International Arbitration Report 7, no. 6 (1992): 18.
217 Earlier versions of some rules provided a presumption in favour of national

procedures absent a choice to the contrary.

218 Karl-Heinz Böckstiegel, ‘The Relevance of National Arbitration Law for
Arbitrations under the UNCITRAL Rules’, Journal of International Arbitration 1,
no. 3 (1984): 228.
219 W. Laurence Craig, ‘Some Trends and Developments in the Laws and Practice

of International Commercial Arbitration’, Texas International Law Journal 30, no.

1 (1995): 57.
220 Domestic procedures are less likely to be optimal given that they often

emanate from historical libertarian distrust of sovereigns.

221 Pierre A. Karrer, ‘Freedom of an Arbitral Tribunal to Conduct Proceedings’,

ICC International Court of Arbitration Bulletin 10, no. 1 (1999): 18.

222 Previously, many civilian national legal systems had imposed a duty on an

arbitrator to respect rules of procedure of domestic courts unless the parties

agreed otherwise. Pierre Mayer, ‘Comparative Analysis of Power of Arbitrators to
Determine Procedures in Civil and Common Law Systems’, in Planning Efficient
Arbitration Proceedings/The Law Applicable in International Arbitration, ICCA
Congress Series No. 7, ed. Albert Jan van den Berg (The Hague: Kluwer Law
International, 1996), 24–25.
223 Yves Derains & Eric A. Schwartz, A Guide to the ICC Rules of Arbitration, 2nd

edn (The Hague: Kluwer Law International, 2005), 7–8.

224 See, e.g., LCIA Rules Art. 14.1; AAA Rules Art. 1(a). Dana H. Freyer, ‘Practical

Considerations in Drafting Dispute Resolution Provisions in International

Commercial Contracts’, Journal of International Arbitration 15, no. 4 (1998): 15.
225 CIETAC Rules 2012 Art. 4(3).
226 [2009] SGCA 24.
227 Yves Derains & Eric A. Schwartz, A Guide to the ICC Rules of Arbitration, 2nd
edn (The Hague: Kluwer Law International, 2005), 7–8, 225.
228 W. Laurence Craig, William W. Park & Jan Paulsson, Annotated Guide to the
1998 ICC Arbitration Rules with Commentary, 3rd edn (Oxford: Oxford University
Press, 2000), 295.
229 Eric A. Schwartz, ‘Choosing between Broad Clauses and Detailed Blueprints’,

in Improving the Efficiency of Arbitration and Awards: Forty Years of Application

of the New York Convention, ICCA Congress Series 1998, Paris 9, ed. Albert Jan
van den Berg (The Hague: Kluwer Law International 1999), 105, 111.
230 Ibid.
231 [1994] 1 Lloyd's Rep 45.
232 Eric A. Schwartz, ‘Choosing between Broad Clauses and Detailed Blueprints’,

in Improving the Efficiency of Arbitration and Awards: Forty Years of Application

of the New York Convention, ICCA Congress Series 1998, Paris 9, ed. Albert Jan
van den Berg (The Hague: Kluwer Law International, 1999), 105, 112.
233 See Isabelle Liger, ‘100 Translation Errors in Institutional Arbitration Rules’,

<http://ssrn.com/abstract=1781078>, 8 March 2011.

234 See, e.g., London Maritime Arbitration Association Terms (2006); Society of

Maritime Arbitration Rules (2003); National Grain and Feed Association

Arbitration Rules (2009).
235 See Hilmar Raeschke-Kessler, ‘The Arbitrator as Settlement Facilitator’,
Arbitration International 21, no. 4 (2005): 652.
236 H.M. Holtzmann & J.E. Neuhaus, A Guide to the UNCITRAL Model Law on

International Commercial Arbitration: Legislative History and Commentary (The

Hague: Kluwer Law and Taxation Publishers, 1989).
237 21 April 1951, 484 UNTS 349.
238 30 January 1975, O.A.S.T.S. No. 42.
239 18 March 1965, 575 UNTS 159.
240 See North American Free Trade Agreement, 32 ILM 289, 605 (1993).
241 See, e.g., Salini Costruttori SpA and Italstrade SpA v. The Hashemite Kingdom of

Jordan, decision on jurisdiction, 9 November 2004, ICSID Case No. ARB/02/13

available at <http://ita.law.uvic.ca/chronological_list.htm>; Siemens AG v.The
Argentine Republic, decision on jurisdiction, 3 August 2004, ICSID Case No.
ARB/02/8 available at <http://ita.law.uvic.ca/chronological_list.htm>; Tza Yap
Shum v. Peru, ICSID Case No. ARB/07/6; Renta No. 4 SVSA v. Russian Federation,
SCC Case No. 24/2007, Award on Preliminary Objections, 20 March 2009, paras
80–101.
242 Article 31 Vienna Convention on the Law of Treaties, Signed at Vienna 23

May 1969, entered into force 27 January 1980.

243 Council of Europe, European Convention for the Protection of Human Rights

and Fundamental Freedoms, 4 November 1950, ETS 5.

244 Pierre A. Karrer, ‘Freedom of an Arbitral Tribunal to Conduct Proceedings’,

ICC International Court of Arbitration Bulletin 10, no. 1 (1999): 26.

245 [2007] EWCA (Civ) 238.
246 [2007] UKHL 40.
247 Astra Zeneca Insurance Co. Ltd v.SGU International Insurance Plc and Ors

[2006] EWCA (Civ) 1340.

248 Osmo Suovaniemi & Ors v. Finland, EHR Court, 23 February 1999 (No.
31737/96).
249 Ibid.
250 International Law Association (International Arbitration Committee), Final
Report: Ascertaining the Contents of the Applicable Law in International
Commercial Arbitration, <www.ila-hq.org/eri/committees/index.cfm/cid/19>
(2008), 5.
251 UN Doc. A/40/17, [297].
252 International Law Association (International Arbitration Committee), Final

Report: Ascertaining the Contents of the Applicable Law in International

Commercial Arbitration, <www.ila-hq.org/eri/committees/index.cfm/cid/19>
(2008), 5.
253 International Law Association Committee on International Commercial
Arbitration, Interim Report on Public Policy as a Bar to Enforcement of
International Arbitral Awards, Report of the 69th Conference, London 2000, 10.
254 Judgment of 30 December 1994, R and U v. W, Swiss Federal Supreme Court,

ASA Bulletin 13, no. 2 (1995): 221.

255 See, e.g., Klaus Peter Berger, The Creeping Codification of the Lex Mercatoria

(The Hague: Kluwer Law International, 1999).

256 Klaus Peter Berger, The Creeping Codification of the Lex Mercatoria (The

Hague: Kluwer Law International, 1999), 278–302.

257 See, e.g., AAA, CPR (International Institute for Conflict Prevention and
Resolution) and JAMS (formerly Judicial Arbitration and Mediation Services) in
the United States, NAM (National Arbitration and Mediation) and NASD (National
Association of Securities Dealers).
258 CIETAC Rules 2012 Arts 7 and 71.
259 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 22.
260 Ibid., 23.
261 That can be indirectly managed through proactive tribunals throughout, and

in the end through cost rulings as part of the award. This is discussed in Chapters
6 and 15.
262 Jonathan L. Greenblatt & Peter Griffin, ‘Towards the Harmonization of
International Arbitration Rules: Comparative Analysis of the Rules of the ICC,
AAA, LCIA and CIET”, Arbitration International 17, no. 1 (2001): 104.
263 Ibid.
264 ICC Rules 2012 Art. 4(5); CIETAC Rules 2012, Art. 13(2).
265 See, e.g., AAA Commercial Arbitration Rules Art. 4(a)(i); LCIA Rules Art.

1.1(g).
266 CIETAC Rules 2012 Art. 6(1).
267 ICC Rules 2012 Art. 6(3)-(5).
268 See, e.g., LCIA Rules Art. 23.1; AAA Art. 7(b).
269 CIETAC Rules 2012 Art. 49.
270 SIAC Rules Art. 28.2.
271 LCIA Rules Art. 26.5; ICDR Rules Art. 27(5).
272 Michael F. Hoellering, ‘Administering International Arbitration Proceedings’,

Dispute Resolution Journal 53, no. 1 (1998): 64, 69.

273 ICC Rules 2012 Art. 30(1).
274 See Michael F. Hoellering, ‘The Role of the International Arbitrator’, Dispute

Resolution Journal 51, no. 2 (1996): 100.

275 The US court has also extended immunity on policy grounds where a tribunal
was composed contrary to the rules of a particular institution. Austein v. Chicago
Board Options Exchange, Rnc 898 F. 2d 882 (2d Cir) cert denied, 498 US 850
(1990).
276 ICC Rules 2012 Art. 40.
277 Michael F. Hoellering, ‘The Role of the International Arbitrator’, Dispute

Resolution Journal 51, no. 2 (1996): 106.

278 Resolution 31/98 adopted by the General Assembly of the United Nations, 15

December 1976, Report of the UNCITRAL on the Work of its Ninth Session, UN
Doc. A/31/17, VII YB UNCITRAL 9, 20–27, 66–82 (1976). See generally D.D.
Caron, M. Pellonpaa & L.M. Caplan, The UNCITRAL Arbitration Rules: A
Commentary (Oxford: Oxford University Press, 2006) as to the 1976 Rules;
Thomas H. Webster, Handbook of UNCITRAL Arbitration: Commentary, Precedents
and Materials for UNCITRAL Based Arbitration Rules, (Thomson Reuters (Legal)
Limited 2010, London); and C. Croft, C. Kee and J. Waincymer, The UNCITRAL
Rules 2010 (forthcoming, Cambridge University Press).
279 Effective 1 November 2007 and available at
<www.cpradr.org/Resources/ALLCPRArticles/tabid/265/...>.
280 The various sets of rules are available at <www.pca-
cpa.org/showpage.asp?pag_id=1188>.
281 Christian Bühring-Uhle, Lars Kirchhoff & Gabriele Scherer, Arbitration and

Mediation in International Business (The Hague: Kluwer Law International,

2006), 39; Nigel Blackaby et al., Redfern and Hunter on International Arbitration,
5th edn (Oxford: Oxford University Press, 2009), 34–35; Anne Veronique
Schlaepfer & Cristian Girod, ‘Institutional vs. Ad Hoc Arbitration’, in International
Arbitration in Switzerland: A Handbook for Practitioners, ed. Gabrielle Kaufmann-
Kohler & Blaise Stucki (The Hague: Kluwer Law International, 2004), 12; William
K. Slate II, ‘International Arbitration: Do Institutions Make a Difference?’, Wake
Forest Law Review 1 (1996): 53.
282 Gordon Blanke, ‘Institutional versus Ad Hoc Arbitration: A European
Perspective’, ERA-Forum 9, no. 2 (2008): 276.
283 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 35.

284 Christian Bühring-Uhle, Lars Kirchhoff & Gabriele Scherer, Arbitration and

Mediation in International Business (The Hague: Kluwer Law International,

2006), 36.
285 Gordon Blanke, ‘Institutional versus Ad Hoc Arbitration: A European
Perspective’, ERA-Forum 9, no. 2 (2008): 276–278; Nigel Blackaby et al., Redfern
and Hunter on International Arbitration, 5th edn (Oxford: Oxford University
Press, 2009), 35; William K. Slate II, ‘International Arbitration: Do Institutions
Make a Difference?’, Wake Forest Law Review 1 (1996): 54–55.
286 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 35.

287 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 149.

288 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 35.
289 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 149.

290 Ibid.
291 Julian Lew, Loukas Mistelis & Stefan Kröll, Comparative International
Commercial Arbitration (The Hague: Kluwer Law International, 2003), 35.
292 Christian Bühring-Uhle, Lars Kirchhoff & Gabriele Scherer, Arbitration and

Mediation in International Business (The Hague: Kluwer Law International,

2006), 36.
293 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 35; William K. Slate II, ‘International
Arbitration: Do Institutions Make a Difference?’, Wake Forest Law Review 1
(1996): 54.
294 Gordon Blanke, ‘Institutional versus Ad Hoc Arbitration: A European
Perspective’, ERA-Forum 9, no. 2 (2008): 279–280; Julian Lew, Loukas Mistelis &
Stefan Kröll, Comparative International Commercial Arbitration (The Hague:
Kluwer Law International, 2003), 35.
295 Gordon Blanke, ‘Institutional versus Ad Hoc Arbitration: A European
Perspective’, ERA-Forum 9, no. 2 (2008): 281–282.
296 The ICC Rules were most recently amended to take effect from 1 January

2012. LCIA Rules were most recently amended on 1 January 1998. CIETAC Rules
were most recently amended on 3 February 2012. The UNCITRAL Rules were
most recently revised on 25 June 2010. ACICA Rules were most recently
amended on 1 August 2011.
297 Christian Bühring-Uhle, Lars Kirchhoff & Gabriele Scherer, Arbitration and

Mediation in International Business (The Hague: Kluwer Law International,

2006), 36; Nigel Blackaby et al., Redfern and Hunter on International Arbitration,
5th edn (Oxford: Oxford University Press, 2009), 35; Anne Véronique Schlaepfer
& Cristian Girod, ‘Institutional vs. Ad Hoc Arbitration’, in International Arbitration
in Switzerland: A Handbook for Practitioners, ed. Gabrielle Kaufmann-Kohler &
Blaise Stucki (The Hague: Kluwer Law International, 2004), 12.
298 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 35.

299 Anne Véronique Schlaepfer & Cristian Girod, ‘Institutional vs. Ad Hoc
Arbitration’, in International Arbitration in Switzerland: A Handbook for
Practitioners, ed. Gabrielle Kaufmann-Kohler & Blaise Stucki (The Hague: Kluwer
Law International, 2004), 12.
300 Christian Bühring-Uhle, Lars Kirchhoff & Gabriele Scherer, Arbitration and

Mediation in International Business (The Hague: Kluwer Law International,

2006), 36; Nigel Blackaby et al., Redfern and Hunter on International Arbitration,
5th edn (Oxford: Oxford University Press, 2009), 35.
301 Christian Bühring-Uhle, Lars Kirchhoff & Gabriele Scherer, Arbitration and

Mediation in International Business (The Hague: Kluwer Law International,

2006), 38; Emmanuel Gail-lard & John Savage (eds), Fouchard, Gaillard, Goldman
on International Commercial Arbitration (The Hague: Kluwer Law International,
1999), 33; Julian Lew, Loukas Mistelis & Stefan Kröll, Comparative International
Commercial Arbitration (The Hague: Kluwer Law International, 2003), 34; Nigel
Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn (Oxford:
Oxford University Press, 2009), 35; Anne Véronique Schlaepfer & Cristian Girod,
‘Institutional vs. Ad Hoc Arbitration’, in International Arbitration in Switzerland: A
Handbook for Practitioners, ed. Gabrielle Kaufmann-Kohler & Blaise Stucki (The
Hague: Kluwer Law International, 2004), 5–6.
302 Christian Bühring-Uhle, Lars Kirchhoff & Gabriele Scherer, Arbitration and
Mediation in International Business (The Hague: Kluwer Law International,
2006), 38; Anne Veronique Schlaepfer & Cristian Girod, ‘Institutional vs. Ad Hoc
Arbitration’, in International Arbitration in Switzerland: A Handbook for
Practitioners, ed. Gabrielle Kaufmann-Kohler & Blaise Stucki (The Hague: Kluwer
Law International, 2004), 6.
303 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 149; Christian Bühring-Uhle, Lars Kirchhoff & Gabriele

Scherer, Arbitration and Mediation in International Business (The Hague: Kluwer
Law International, 2006), 38; Julian Lew, Loukas Mistelis & Stefan Kröll,
Comparative International Commercial Arbitration (The Hague: Kluwer Law
International, 2003), 34; Anne Véronique Schlaepfer & Cristian Girod,
‘Institutional vs. Ad Hoc Arbitration’, in International Arbitration in Switzerland: A
Handbook for Practitioners, ed. Gabrielle Kaufmann-Kohler & Blaise Stucki (The
Hague: Kluwer Law International, 2004), 7.
304 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 149. See also Anne Véronique Schlaepfer & Cristian Girod,
‘Institutional vs. Ad Hoc Arbitration’, in International Arbitration in Switzerland: A
Handbook for Practitioners, ed. Gabrielle Kaufmann-Kohler & Blaise Stucki (The
Hague: Kluwer Law International, 2004), 6.
305 Pieter Sanders, ‘Has the Moment Come to Revise the Arbitration Rules of

UNCITRAL?’, Arbitration International 20, no. 3 (2004): 258.

306 Gordon Blanke, ‘Institutional versus Ad Hoc Arbitration: A European
Perspective’, ERA-Forum 9, no. 2 (2008): 275–282, 282; Nigel Blackaby et al.,
Redfern and Hunter on International Arbitration, 5th edn (Oxford: Oxford
University Press, 2009), 37.
307 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 37; See also Christian Bühring-Uhle,
Lars Kirchhoff & Gabriele Scherer, Arbitration and Mediation in International
Business (The Hague: Kluwer Law International, 2006), 38–39.
308 Anne Véronique Schlaepfer & Cristian Girod, ‘Institutional vs. Ad Hoc
Arbitration’, in International Arbitration in Switzerland: A Handbook for
Practitioners, ed. Gabrielle Kauf-mann-Kohler & Blaise Stucki (The Hague:
Kluwer Law International, 2004), 6.
309 Ibid., 7–10.
310 Ibid., 7–10, 14.
311 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 35.

312 Anne Véronique Schlaepfer & Cristian Girod, ‘Institutional vs. Ad Hoc
Arbitration’, in International Arbitration in Switzerland: A Handbook for
Practitioners, ed. Gabrielle Kauf-mann-Kohler & Blaise Stucki (The Hague:
Kluwer Law International, 2004), 6.

Part II: The Process of an Arbitration,

Chapter 4: Written Notices,
Submissions and the Articulation of
Claims and Defences

Jeff Waincymer,

4.1. Introduction

This chapter examines the various opening stages including the request or
notice of arbitration, answers to the request and defences and counterclaims.
Some matters that could otherwise be dealt with in this chapter for
chronological reasons, being matters to be dealt with at the outset, are instead
left for other chapters for conceptual reasons. The key stage in that regard is
that of any pre-arbitral referee process which is discussed in Chapter 8 when
the whole question of interim measures is considered.

4.1.1. Policy Issues and the Articulation of Contentions

Any formal legal dispute should commence as a result of a written notice on

behalf of the claimant articulating the identity of the respondent, the basis of
claim and the relief sought. Where international arbitration is concerned,
requirements for such notices are contained in lex arbitri and any applicable
arbitral rules. The principle that ‘the parties' claims determine the
commencement, scope and end of the proceedings’ is seen as fundamental to all
procedural systems. (1) The outline of claims and cross-claims can be given
differing names by practitioners, tribunals or arbitral rules. Generally speaking,
the claimant will outline the claim in a Request or Notice of Arbitration, but a
more fulsome outline would be described as a Statement of Claim, Statement of
Case or phrase to similar effect. In some cases these page "217" may be
combined. After the notice of claim is served, the respondent may then be called
on to provide an Answer to the Notice or Request and a more detailed document
would be described as a Defence or Rebuttal. There may also be a counterclaim
and/or set-off by the respondent, which in turn would call for claimant's
defences to those documents and contentions.

There are important policy questions in setting out standards for such
documents. The nature and use of written submissions outlining each party's
contentions must be analysed in the context of mandatory norms of due process,
general principles of efficiency and specific provisions in lex arbitri and selected
rules. Where mandatory norms are concerned, an important issue is the right to
an adequate opportunity to present one's case. That inherently involves the
need to understand the opposing party's position on each issue and then have a
right to respond. That cannot mean that there is an unlimited right to reply to
the last observation of the opposing party. It would simply be impossible to give
each party that right concurrently. A claimant will typically be allowed a reply to
respondent's defence as will the respondent to claimant's defence to any
counterclaims. Tribunals vary as to whether they allow a further response by
the respondent to claimant's reply. This is because legal systems vary as to
whether they allow the claimant alone to make a reply to the defence or
whether each party has one right to rebuttal or surrebuttal respectively.
Arguments in favour of an equal number of submissions would be based on a
simple notion of equality. The contrary position, where only the claimant may
reply, would flow from the argument that effective equality means each party
must have one opportunity to answer what it hears from the other side. The
defence answers the statement of claim and the reply answers the defence.
Tribunals differ in their approach but strict equality is probably the
predominant approach.

Due process principles can have significant impact in relation to notice and
detail of claims and defences. Nevertheless it is hard to state general principles.
There is no uniform view amongst legal families as to the appropriate detail in
the originating document in any adjudicatory process. Some comparative
analysis is appropriate before considering rules and practices in arbitration, as
these tend to provide little guidance. Even where guidance is provided, a
tribunal will have to apply rule of reason tests where adequacy of compliance is
contested. In domestic litigation, there is a significant difference between
common law and civil law jurisdictions as to the minimum requirements for
notices commencing proceedings. There is said to be a distinction between what
is described as notice pleading in the common law tradition and fact pleading in
the civilian tradition, although a historical analysis shows that the divisions are
not so marked. (2) The key differences are the degree of detail, whether
principles of law as well as facts are articulated and whether reasoning is to be
included. The common law tradition is to have a more minimal document that
identifies the parties, the basic facts and the relief sought. Common law legal
systems commence proceedings with the serving and page "218" filing of a
statement of claim that accompanies the formal writ or summons. (3) While
statements of claim will often directly or indirectly articulate the basis of the
claim, they will often not do so formalistically or comprehensively. For example,
it is perfectly acceptable in common law systems to state that a contract was
entered into calling for delivery of goods of a sufficient quality and then simply
allege that goods of inadequate quality were delivered. The statement of claim is
not defective by reason of failing to expressly articulate that there was a breach
of contract or that the breach was fundamental. The argument in favour of this
minimalist approach would be that the defendant merely needs to be given a
sufficient indication of the nature and the amount of the claim to determine
whether to defend, comply with the demand or seek a settlement. Common law
systems also allow the respondent to call for further particulars of the claim. At
the other extreme is the idea that the initial request should outline the claim,
evidence and arguments as comprehensively as possible. Civilian legal systems
have adopted a much more libertarian approach vis-à-vis the defendant, which
in turn inspires more comprehensive requirements. If a party's liberty is to be
interfered with by the compulsion of defending a claim, then the view is that the
claimant should have a fully considered and articulated case from the outset and
should be able to articulate it immediately. However, this is complicated by the
fact that civilian systems typically apply the iura novit curia principle whereby
the judge must know the law, hence parties need not necessarily include legal
argument in their documentation. This principle is discussed further in section
13.19.

Whatever the differences between legal families, these are in part explained by
differences in approach to document production and differences between staged
or single hearings. A legal system that does not oblige a party to produce
information to its opponent upon demand during the currency of the
proceedings by reason of a fundamental right of the latter to that effect, can
naturally call for each to present a comprehensive outline of their case from the
outset. Conversely, common law systems that invite all relevant documents to
be produced, more naturally allow submissions to be deferred until such
material is fully available. This dovetails with the common law reliance on a
single final hearing. In common law systems, more detailed aspects of the claim
and responses and relevant material are developed over the interlocutory
stages which would include discovery, interrogatories (being questions posed to
the other party), and in the US, depositions of witnesses. Because it is seen as
perfectly reasonable to seek information from the opposing party in developing
one's case, the corollary is that the claimant is not expected to have a complete
articulation at the outset.

Emerging arbitral practice has tended to find a compromise position between

these extremes but looks more closely akin to civilian models, although that is
more a case of developing practice and arbitrator discretions than clear
exposition in the rules. On the one hand, timeliness and the wish to afford each
party the fullest page "219" opportunity to respond to its opponent's case would
suggest that the more complete approach is preferable. On the other hand, if
there are to be multiple submissions, calling for too much detail at a preliminary
stage may lead to unnecessary duplication when subsequent submissions are
made after review of produced documents and witness statements.

4.2. Request or Notice of Arbitration

As noted above, while there are differences in approach between legal families,
the role and requirements of a request or notice of arbitration should be looked
at in the context of fundamental norms of due process. The person served with
such a request or notice should understand the nature of the claims made and
relief sought and be in a position as soon as possible to decide whether to
defend, concede or try and settle the dispute. While the parties could themselves
designate what is required as an element of their arbitration agreement, more
typically it is a question of provisions in the lex arbitri and/or arbitral rules,
although as noted below, these will often be general in nature. Some important
questions are: what are the minimal standards for a valid request or notice; can
challenges be made as to adequacy; and what implications flow in terms of
timing and validity if a request or notice has not met required standards.
4.2.1. What Should a Request or Notice of Arbitration Contain?

There is no consistent model in arbitral rules or lex arbitri as to the required

contents of requests or notices of arbitration. Nevertheless, it would be
widespread practice to provide a detailed outline in the notice of arbitration
together with the key documents upon which the claimant relies. Requiring
parties to plead facts and include relevant documentation is an ideal means to
ensure adequate notice to the respondent and also to limit the ambit and costs
of the document production stage.

4.2.2. Content

It is for the claimant to formulate the claims that it seeks to put forward in
arbitration. The claimant will be limited by the terms of the arbitration
agreement which will define matters which the parties have agreed can be
referred to arbitration. This usually can be done by reference to a particular
contract and by specifying the relationship that the claims must possess to that
legal relationship. If the respondent does not appear on the face of the document
to be a party to the arbitration agreement, the claimant would be expected to
articulate the basis for inclusion, for example via an agency or group of
companies contention.

An appropriately drafted arbitration clause will enable a claimant to put

forward claims concerning the contract itself and in respect of pre-contractual
page "220" representations, associated claims in tort and statutory claims
(including competition law matters) concerning the conduct attendant upon the
negotiations and conclusion of the contract and its effect.

Various institutional rules give direction as to the level of content. The ICC Rules
2012 call for the request for arbitration to contain a statement of the claimant's
case and be accompanied by relevant documentary evidence. The Answer is to
conform to similar standards. Article 4 of the ICC Rules 2012 indicates that the
request for arbitration shall, inter alia, contain ‘a description of the nature and
circumstances of the dispute giving rise to the claims and of the basis upon
which the claims are made …’ together with ‘a statement of the relief sought
together with the amounts of any quantifiable claims, and, to the extent possible,
an estimate of the monetary value of any other claims’. These stipulations do not
indicate the extent to which the reasoning and arguments must be articulated.
Article 3 of the UNCITRAL Rules 2010 provides that the notice shall include ‘a
brief description of the claim and an indication of the amount involved, if any’
and ‘the relief or remedy sought…’. Article 36 of the ICSID Convention stipulates
that the request shall contain information concerning the issues in dispute, the
identity of the parties and their consent to arbitration in accordance with the
Rules of Procedure for the Institution of Conciliation and Arbitration
Proceedings. (4)

Because these laws and rules only speak in generalities, there needs to be
greater thought to what ought to be included and how a tribunal may react if a
challenge is made as to the adequacy of the notice. In some cases an institution
must itself consider compliance. (5) There are some important policy
considerations as outlined above. From the respondent's perspective, the notice
of arbitration needs to be sufficiently clear to enable it to determine whether to
defend the claim. If the notice is too brief the respondent will not be adequately
informed of the nature of the claim. In turn, this will mean that it is difficult for
the respondent to decide on the best composition of the tribunal. Insufficient
detail is unlikely to encourage any settlement overtures from the respondent.
Where the arbitration agreement has not already selected a Seat, if the request
is too brief this will also interfere with the optimal selection of a Seat.

If the notice is sufficiently detailed, there may not need to be a separate

statement of claim. If institutional rules call for Terms of Reference, a
sufficiently detailed request will also aid in the drafting. It will also allow the
respondent to adequately consider whether any jurisdictional challenges are
merited. A request that is sufficiently detailed will also allow the tribunal to
move quickly if the respondent does not choose to engage in the arbitration. If
the notice is sufficiently detailed, it will allow the tribunal or institution to
identify an appropriate advance on costs.

Conversely, if the request is too specific there is a problem if something has been
left out. In such circumstances the claimant may need to rely on the discretion of
the tribunal to add new claims at a later stage if the respondent will not
otherwise page "221" consent. From the claimant's perspective, as long as there
is enough detail to satisfy the respondent's legitimate needs, further details can
be provided later after constitution of the tribunal, after which full written
submissions will at some stage be the norm. The more that the request for
arbitration needs to provide details of applicable law and reasoning, the more it
may help the respondent, but the more it will also slow down a claimant in
bringing the action. That may be particularly problematic where a limitation
period might soon expire. An appropriate balance needs to be struck. Elsing and
Townsend suggest that one test of whether a statement of claim provides
sufficient detail is whether the statement and supporting documents would
allow an arbitrator to draw up Terms of Reference without any other
information. The authors suggest that such a practice allows the claimant to
begin to tell the story convincingly from the outset, makes it difficult for the
other party to claim insufficient notice of the claim and helps the arbitrator have
confidence that the key issues are all on the table. (6)

Various laws and rules will also indicate how to treat additional claims or
amendments to claims as made. This is discussed in section 6.13.3.

4.2.3. Documents

The various rules do not demand that the claimant provide the documents it
seeks to rely on contemporaneously with the Notice or Request, although this is
often encouraged and as noted, is emerging practice. For example, Article 4 of
the ICC Rules 2012 indicates that a claimant ‘may submit such other documents
or information with the Request as it considers appropriate or as may
contribute to the efficient resolution of the dispute’. From a tactical perspective,
the advantages of doing so are to force the respondent to think of the key
evidence and to respond accordingly. The disadvantage is that it shows the
claimant's hand and obviates the advantages of surprise. Because arbitration
seeks to promote transparency and minimise gamesmanship, the better view
would encourage the provision of relevant documents at an earlier stage rather
than later. (7) Arbitral rules may also call for production of the arbitration
agreement and an indication as to the number of arbitrators proposed. A
reference to the arbitration agreement would also deal with the writing
requirement under Article 7(2) of the UNCITRAL Model Law or equivalent
provisions where such a requirement applies.

page "222"

4.2.4. Identifying Arbitrators, Language and Applicable Law

Most rules do not provide for proposals as to the identity of arbitrators at this
stage. Article 3.4 of the UNCITRAL Rules 2010 allows for such a designation. If
the arbitration agreement is silent as to the Seat, this might also be proposed at
this stage and is required under Article 3(3)(g) of the ICC Rules 2012.

A request might also comment on the applicable law but this is not always
required. It would be particularly desirable to raise any claimed mandatory
laws. A notice may require an indication of the language of arbitration. (8) Even if
the arbitration agreement has selected the language of the arbitration itself, one
tribunal has considered that the Notice or Request need not be in that language.
(9) A failure to tender a request in a specified language may also not be fatal,

particularly if a translation is also provided. (10) If there is neither any agreement

nor a default language in the rules, then the parties may make their request in
whatever language they believe appropriate.

4.2.5. Timing and When the Notice or Request Must Be Lodged

It is important to meet the minimal standards for a notice or request both in

terms of the due process rights of the respondent and also to preserve rights of
the claimant. The importance of timing questions may depend on whether the
tribunal is interpreting a provision in the parties' agreement subject to
whatever law is considered to be applicable, or conversely, considering the
application of arbitral rules. A key concern is where there may be an applicable
limitation period. Where the claimant's rights are concerned, a valid request
needs to be brought within any applicable limitation period. There is a need to
identify what, if any, statute of limitations may apply, whether such laws apply
to arbitration and when the arbitration can be said to have commenced. There is
a difference between a statutory limitation on rights and an internal limitation.
An example of the latter may be a contractual warranty as to profits on the sale
of a business for a stipulated period of time. The arbitration agreement itself
may impose time limits either by indicating that claims cannot be brought after
a certain period has elapsed or conversely, that there must first be a cooling off
or negotiating period before arbitration can be commenced. Commencement of
an arbitration must satisfy any stipulated preconditions such as an obligation to
attempt mediation or a moratorium period on claims or a stipulation that there
be exhaustion of other remedies before resort to arbitration. A failure to comply
could invalidate the award.

There may also be provisions in the lex arbitri and arbitral rules that purport to
deem the date of receipt of notices. Waiver rules might also be relevant. In some
cases a party may be unclear as to whether there is in fact a valid arbitration
clause page "223" or whether its rights are instead under litigation statutes. If a
limitation period is fast expiring, the party will need to take a view, being aware
that there is a risk that whichever step is adopted might be argued to constitute
a waiver of the alternative. (11)

In addition, there is an important timing question in the context of equality of

the parties and the expectations of a tribunal as to required content. Other than
those arising by agreement or under any applicable statutes of limitations, there
are no time limits on claimants as to when they should present any request for
arbitration. Hence they can take a significant amount of time in preparing an
elaborate claim document. The more elaborate the documents called for, the
more unreasonable it is to expect the respondent to produce an equivalent
standard document in a restrictive time period, normally thirty days.

4.2.6. When Is the Notice or Request Deemed Served

There are a number of reasons why it is important to know the time of service of
a notice or request. The most significant aspect of the timing of a notice is where
limitation periods may apply. The commencement date that flows from a valid
request may also be relevant for calculation of interest on any damages. The
timing of service also triggers obligations as to answers to the request or notice
and in some cases, nomination of an arbitrator.

Sections 4.7 and 4.8 look generally at questions of service and communication.
At this stage it is appropriate to note that provisions in arbitral rules and lex
arbitri combine actual and deemed criteria about which there can at times be
factual debate. There may, for example, be debates about delivery via the
Internet where there is some breakdown between servers or where the alleged
recipient denies receipt.

4.2.7. Challenges to Adequacy

As noted, various rules indicate the minimum required content of a Notice or

Request, such as indicating that a request should refer to the arbitration
agreement, the nature of the claim, the relief sought, the amount of damages and
preferred number of arbitrators. Consideration must be given to circumstances
where the stipulated requirements are not fully met. These circumstances can
range from relatively technical failings such as a failure to provide a designated
number of copies and at the other extreme, defects that go to the heart of due
process such as a failure to articulate claims, relief sought or to properly identify
the respondents. For example, where the respondent is part of a group of
companies, the exact company might be mis-described. A question may then
arise whether the Notice page "224" is defective or whether substantial
compliance with the minimal requirements would suffice. Absent any express
direction as to factual exactitude, the preferred approach is to adopt a substance
over form analysis and consider whether the Notice as drawn is sufficient to
appraise the respondent of the claim. This question may be particularly
important where a limitation period expires after the original Notice was served
and before any correction is possible. Arbitral laws and rules rarely give any
indication as to whether substantial compliance is permitted and if not, what if
any remedial action is possible. In the absence of specific directions, other
relevant provisions would include timelines on when objections to jurisdiction
can be made. (12) If the failure is noticed prior to the expiration of a limitation
period it can obviously be remedied, but there still may be uncertainties as to
the proper dates for response documents, notification of arbitrators and
commencement of damages and interest entitlements. For example, in an ICC
scenario, it would be difficult to know exactly what a respondent should do
where there is a thirty day period to nominate an arbitrator in circumstances
where there is a debate between the parties as to whether the initial request is
adequate or not.

If remedial action is not possible prior to the expiration of a limitation period,

the more serious question is whether the arbitration has been validly
commenced and if not, whether it is now barred. While that also may be
impacted upon by whether the respondent has made objection within a
stipulated time period, even if that is not the case, it is conceptually difficult to
identify who should win when neither party meets stipulated deadlines, either
under a consent or jurisdictional paradigm. Statute of limitations issues are
particularly troubling as there is no consensus as to whether they should be
seen as substantive or procedural. (13) It is problematic to try and resolve this
simply through questions of characterisation. Because a party can always waive
a statute of limitations defence, it is always an issue of consent as to the time
limits that should apply. Hence parties could expressly deal with this in their
arbitration agreement in any event. While that is unusual, it points to the
difficulty in identifying a uniform solution absent clear guidance from the
parties themselves. Because of the ongoing uncertainty, great care needs to be
taken with any statute of limitation period that could arguably apply. Once
again, the better view is that arbitration should not be caught up in technical
debates about sufficiency, except in more extreme cases where the respondent
is truly unable to identify the nature of the claims. (14) In considering the effect of
partial non-compliance the policy test should be whether the respondent is
prejudiced other than through losing the right to succeed on technical
arguments. Other situations are more readily remedied even if the notice is
clearly defective. If the Claimant fails to nominate an Arbitrator in the request,
this will page "225" typically lead to the institution inviting them to make a
subsequent nomination, (15) or allow for appointment by other means.

Some rules specify how to deal with non-compliance with requirements. Swiss
and German rules expressly indicate that if defects are fixed in time, it does not
affect the commencement date. (16) Article 3.5 of the Swiss Rules 2012 indicates
that the Secretariat may request that the Claimant remedy the defect within an
appropriate period and further that if the Claimant complies, the Notice of
Arbitration shall be deemed to have been validly filed on the date initially
served. The ICC Rules 2012 only expressly deal with failure to provide the
advance payment or the required number of copies, allowing extra time to be
fixed by the Secretariat. (17) Where there is no express guidance in the rules, each
individual fact scenario would need to be looked at on its merits, ideally in the
context of the above suggestion to take a substance over form approach. There
is again a two-step process, first determining whether the stipulations are
merely guides to best practice or are instead gateway requirements before
proceedings can be validly established. In ICC Case 6784 of 1990 (18) the Tribunal
considered that the stipulated elements were not preconditions to admissibility
of the claims. (19)

Issues as to adequacy of a request or notice should not be confused with

challenges as to preconditions to a promise to arbitrate. Section 3.2.1.2 above
looked at the question of escalation clauses and when arbitration might be
argued to be premature. Similarly, an arbitration agreement will often refer to
‘disputes’ arising out of or in relation to a particular contract being brought to
arbitration. In some cases, there may be a question as to whether there needs to
be an existing dispute or merely a claim to form the basis of a Request or Notice.
The notice itself should be free from challenge if it identifies a claim, although it
will be a question of interpretation of the arbitration agreement as to whether
the preconditions to an adjudication have arisen. There should in fact be an
existing claim. A mere indication of a wish to claim at a later stage is unlikely to
suffice.

4.3. Answer to Request or Notice

As a general rule, lex arbitri do not require a specific response to a request or

notice for arbitration. The parties might themselves agree on a process or utilise
arbitral rules that designate what response is required. More recently, amended
rules tend to call for such a response. From a policy perspective, a response to
the claim simply needs to provide a brief response and begin any required
processes as to page "226" selection of arbitrators, language and/or the Seat.
The situation is different where counterclaims are concerned. Here the claimant
needs sufficient detail of the nature and basis of any counterclaims in order to
determine what response to take.

Article 5 of the ICC Rules 2012 indicates that within thirty days from the receipt
of the Request from the Secretariat, the respondent shall submit an Answer
which shall contain, inter alia, ‘its comments as to the nature and circumstances
of the dispute giving rise to the claims and the basis upon which the claims are
made’, ‘its response to the relief sought’ and observations as to number and
choice of arbitrators, place of arbitration, applicable rules of law and language of
arbitration. (20) Just as the claimant is given the option to submit relevant
documents and information, the same is so with the respondent's Answer. The
Secretariat may grant the respondent an extension provided that the required
observations or proposal concerning the number or choice of arbitrators is
provided. This is to ensure that tribunal composition is not held up while
respondent turns its mind to its responses on the merits. Article 5(5) indicates
that any counterclaims made shall be submitted with the Answer and shall
provide ‘a description of the nature and circumstances of the dispute giving rise
to the counterclaims and of the basis upon which the counterclaims are made’,
and ‘a statement of the relief sought together with the amounts of any quantified
counterclaims and, to the extent possible, an estimate of the monetary value of
any other counterclaims …’. The rule also requires provision of any relevant
agreements and where counterclaims are made under more than one
arbitration agreement, an indication of the arbitration agreement under which
each is made. Counterclaims and set-off are considered in section 4.4 and again
in section 7.8.

Article 4.1 of the UNCITRAL Rules 2010 introduces a new obligation in the
UNCITRAL regime to provide a response to the notice of arbitration within
thirty days of receipt. Article 21 of the UNCITRAL Rules provides for a separate
statement of defence which responds to the statement of claim. Where the
response to the notice of arbitration is concerned, the response is to give an
indication of the respondent's views as to the matters contained in Article
3.3(c)-(g) outlining elements of the notice of arbitration, being identification of
the agreement, the contract or other relationship, the claim and amount, the
relief or remedy sought and proposals for the place and language of arbitration
and the number of arbitrators (if these details have not been previously agreed).
Article 4.2 indicates that the response to the notice may also include any plea of
lack of jurisdiction, a proposal for the designation of an appointing authority, a
proposal for the appointment of a sole arbitrator, notification of the
appointment of an arbitrator, a brief description of counterclaims or claims for
the purpose of set-off where relevant, including amounts involved and the relief
or remedy sought and a notice of arbitration where a respondent seeks to claim
against a party to the arbitration agreement other than the claimant.

While timeframes are typically articulated, less clear are the ramifications if a
respondent simply fails to meet the deadline. There are no express ramifications
page "227" articulated in the ICC or UNCITRAL Rules. Obviously due process
must still allow it to present its defences in due course. The more contentious
question is whether the right to counterclaim is lost if deadlines are not met.
Even if that was so, a respondent might bring the counterclaim as a new claim
and seek to consolidate the proceedings, although consolidation is not a matter
of entitlement but simply an ability to apply in appropriate circumstances.
Furthermore, parties may amend their claims and counterclaims in appropriate
circumstances and this would normally be interpreted to allow late
counterclaims. In any event the UNCITRAL Rules are only permissive in terms of
counterclaims at that stage. This is not the case with the ICC Rules which
requires notice of the counterclaims at that stage. (21)

4.4. Counterclaims and Set-Off Rights

When considering respondent's claims it is appropriate to first separate

counterclaims and set-off defences. A counterclaim is usually seen as a claim
brought by a respondent in a civil suit against the claimant that is independent
of the primary claim although it may be linked to the same facts. The term is
used in contradistinction to a set-off that is seen as a defence to the primary
claim, albeit one invariably related to different facts. Because it is not simply a
defence, a counterclaim leads to a separate judgment that may be in excess of
the judgment under the primary claim. Furthermore, the counterclaim remains
alive even if the initial claim is withdrawn. Thus it is truly a reverse claim and
not a defence as such. Because a counterclaim remains alive even if the primary
claim is withdrawn or invalid, it must be based on its own independent evidence
of arbitral consent. As always, such consent should be found to emanate from
the arbitration agreement underlying the Notice of Arbitration, either directly or
through a lex arbitri that expressly allows for counterclaims. Even then the
counterclaim should be linked to the original arbitration agreement. Numerous
scholars support this view. (22) While a counterclaim is normally a distinct action,
at times it might be raised on a conditional basis, that is to say, the tribunal
might only be asked to consider it if it should find prima facie liability under the
primary claim. (23) This would not alter its distinct nature, however.
Counterclaims rarely arise in investment disputes as treaties will typically not
provide for this.

page "228"

Civilian systems describe set-off claims in differing ways, the essential meaning
of which is compensation. (24) The case of set-off in international economic
arbitration was comprehensively analysed by Klaus Peter Berger in 1999. (25) As
his contribution shows, the proper treatment of set-off claims is far more
complex and controversial than that of counterclaims. At the domestic level, a
set-off can be in respect of a wholly unrelated dispute or transaction. It then
becomes difficult to integrate this notion into an arbitral paradigm. Berger notes
the particular problem facing international commercial arbitration. On the one
hand, given that there is no harmonised view as to the nature and ambit of set-
off within domestic legal systems, arbitral rules would be reluctant to be too
prescriptive. On the other hand, these very complexities together with the lack
of prescription forces arbitrators to try to find a justifiable methodology for the
treatment of such claims.

The distinction between set-off and counterclaim is important in terms of

finding evidence of consent to admissibility in each case. The consent-based
logic that some commentators employ is to the effect that parties must have
intended that all relevant defences can be raised against claims. It is suggested
that a contrary position would offend against the most fundamental principles
of justice, due process and fairness. Thus there is a presumption in allowing set-
off defences to be admissible. Others argue that there needs to be some express
agreement. This is discussed further in section 4.4.3.3.

4.4.1. Admissibility of Counterclaims

As noted, a counterclaim must find a jurisdictional basis within the arbitration

agreement that supported the primary claim, although that could emanate from
a finding that multiple contracts, only some of which contain arbitration clauses,
are sufficiently connected to allow this to occur. Even though arbitral rules
generally allow for counterclaims, circumstances may arise which show that the
parties have not consented to this occurring. An example would be a
counterclaim brought under a different contract that has a different dispute
resolution clause incompatible with the arbitration agreement under which the
counterclaim has been brought. This will obviously be so where the second
contract has a choice of a domestic litigation forum. It can even be so where the
separate contract has selected arbitration in a different Seat and even where the
arbitral rules might be identical. In such a case the tribunal must make a
determination as to intent. If a tribunal, or indeed an annulment court,
considered that the intent behind the separate contract with a separate Seat was
to treat the disputes independently and not page "229" allow counterclaims,
then accepting a counterclaim would undermine the validity of the award. (26) A
contrary conclusion is also possible. A tribunal might consider that the decision
to have a separate Seat was only if the claim was made as a primary claim and
not if it was taken by way of counterclaim. Because an annulment or an
enforcement court might take a different view to the tribunal, a tribunal should
certainly be circumspect in that regard. The case of multiple contracts is
discussed in detail in section 7.8.

There does not appear to be controversy with respect to the need for distinct
consent to underlie a counterclaim. The key difference in view is based on the
evidentiary standards and methods of determining consent. This will either
involve interpretation of the arbitration agreement or the lex arbitri and
procedural rules. The more broadly the terms are drafted, the easier it is to
allow a counterclaim. The biggest problem is with rules that simply allude to the
procedural steps required, without attempting to define an admissibility
standard. Even rules that have attempted such a definition have rarely dealt
with the question with enough precision to guarantee certainty and consistency.

A properly drawn agreement would allow for both claims and counterclaims
under the contract that contains the arbitration clause. It is the connection to
the contract and not who makes the claim that matters, although there can still
be consent issues as to the constitution of the tribunal which should hear the
case. Tribunals are not standing dispute settlement bodies. Thus a counterclaim
that only arises on the facts after a tribunal is constituted would not fit within
the particular reference to arbitration, unless permitted under the arbitration
agreement or the institutional rules. (27)

Even where the events giving rise to the counterclaim arose before the
constitution of the tribunal, there may still be fairness considerations,
particularly as to tribunal composition and particularly as counterclaims under
some procedural systems do not need to be notified prior to tribunal
composition. In such circumstances, if one party's concern is raised as a claim,
the other party sees the contention in the Notice of Arbitration and can select an
arbitrator with the issues in dispute in mind. Conversely, where a counterclaim
is notified after the constitution of the tribunal, there can be a legitimate
question as to whether the parties have consented to that tribunal hearing a
reverse claim that was unknown to at least one of the parties when the initial
arbitration was commenced. This issue might not be a problem where the clause
expressly refers to counterclaims, as parties may consent to that imbalanced
situation.

page "230"

4.4.2. Counterclaims and Arbitral Rules

In the absence of express agreement by the parties in their arbitration clause to

the allowance of counterclaims, the next possibility is that they have indicated a
similar choice through their selection of a seat and its attendant lex arbitri
and/or their selection of arbitral rules. The various lex arbitri and procedural
models fall into two broad categories. One group simply states that
counterclaims may be brought or identifies the time limits within which they
can be brought. This category does not seek to identify the degree of connection
required for the counterclaim to be admissible. An example of this arises in
Article 3.10 of the Swiss Rules of International Arbitration 2012. (28) Within this
category a distinction should also be drawn between rules expressed in this
form (29) and those that, while not defining the degree of connection required for
a counterclaim to be admissible, refer to ‘any counterclaim’ (emphasis added)
brought by a respondent. The contention that the use of the word ‘any’ instead
of ‘a’ to qualify the class of counterclaim has the significance of broadening the
class of admissible counterclaims is unlikely to have been the clear intention of
the drafters of those rules. A similar general reference is provided by Article 2(f)
of the UNCITRAL Model Law that simply states that the provisions of the law
apply to a counterclaim. (30) An indirect reference is found in page "231" Article
30(5) of the ICC Rules which clarifies the inclusion of counterclaim and set-off
amounts in the advance on costs. (31)

Other rules seek to define the linkage required for admissibility. One group
takes a conservative approach, limiting admissibility to counterclaims from the
same contract. For example, UNICTRAL Arbitration Rules 2010 Article 19(2)
and (3) speak of counterclaims ‘arising out of the same contract’. If it is from the
same contract, the accepted view is that it does not need to be limited to
contractual claims. The test might be broader if it referred to counterclaims
‘relating to’ the same contract, a view alluded to by Kee. (32) There is no
equivalent in the UNCITRAL Model Law although the working group suggested
the Model Law should be interpreted with a similar restrictiveness to the
UNCITRAL Rules. (33) Article 22(3) of the ACICA Rules is worded to give effect to
a broader formulation: ‘…the Respondent may in its Statement of Defence,…,
make a counterclaim or claim for the purpose of a set-off, arising out of, relating
to or in connection with the contract’ (emphasis added).

Another group takes a different approach to the linkage test and draws attention
to the same arbitration agreement or the same relationship rather than the
contract per se. (34)

4.4.3. Introduction to Set-off

As noted above, the term counterclaim is used to describe an independent claim
that is not in the nature of a defence. Set-off, however, is seen as a defence. If the
set-off is allowed and established, it has a number of implications. Because the
setoff is limited in amount to the totality of the original claim, there can be no
monetary award in favour of a person raising it. Furthermore there is no need
page "232" for separate awards on claim and set-off defence. The other corollary
of this is that if the claim is not made out or is withdrawn, there is no need to
adjudicate upon the set-off. A set-off merely provides a defence to the claim.
Hence it operates ‘as a shield, not as a sword’. (35) While a set-off is normally
raised by respondent, it could also be raised by a claimant against a
counterclaim brought by a respondent. (36) However, one set-off cannot be
brought as against another set-off.

From the time that an adjudicator finds in favour of the set-off right, it will
either operate retrospectively or prospectively depending upon whether the
applicable law provides for automatic application or application from the time
of the notice or perhaps even the time of judgment. This can affect rights to
interest and measurement of damages. It might also often have different costs
implications to counterclaims, although costs will of course be a discretionary
matter. By way of example, if a set-off is a full defence, a tribunal might order
costs in favour of a respondent who has succeeded in entirely blocking the
claim. However, a successful set-off that only partially reduces the primary
claim, might still see the claimant successful on costs, to the extent of the net
amount. Conversely, where both a claim and counterclaim are successful, each
successful party might anticipate costs on their successful element.

While these propositions are clear, much of the law of set-off is contentious and
varies greatly between different legal families. This makes it difficult to
determine what the general treatment should be under international
arbitration. For this reason, there is a need to consider the differences in views
between legal families as to the nature of set-off in the context of considering
admissibility factors within international arbitration.

4.4.3.1. The Nature of Set-Off

For an arbitrator, there are both procedural questions as to admissibility, and

substantive questions as to the inherent nature of set-off. One of the
complexities in dealing with set-off is that some legal systems treat it as
substantive while others treat it as procedural. Even within some legal systems
different types of set-off fall into each category. (37) Berger contends that the
long-standing dispute as to whether set-off was procedural or substantive in
nature under civil law, was decided in favour of the latter view. (38) For the
purposes of this section the more it is seen as substantive, the more it might be
argued to inherently undermine the claim and hence be admissible as any
defence should be. The more it is procedural, the more it is seen as an efficiency
measure that simply aims to page "233" reduce total transaction costs in
resolving multiple claims and thus require independent evidence of consent. (39)

As is discussed below, even this distinction does not necessarily lead to differing
outcomes as many see efficiency factors as key guides to implied consent. As
suggested below, efficiency arguments are more justifiable when linked to
consent. Efficiency factors might then apply indiscriminately to both procedural
and substantive reverse claims. This concern to discern the intent of the parties
on a case-by-case basis cautions against automatic transplanting of domestic
litigation notions of set-off into the field of international commercial arbitration.
The aim of avoiding multiplicity of litigation underlying common law allowance
of set-off rights is in the main about efficient use of the courts and allocation of
taxpayers' money, not about the original intent of the litigants. (40) The point is
simply that the consent logic differs greatly depending on which view of set-off
we begin with.

As noted above, Klaus Peter Berger has made a major study of set-off in
international arbitration. (41) As to substantive issues, Berger has also noted and
analysed a recent attempt to articulate harmonised principles of set-off through
the 2004 edition of the UNIDROIT Principles of International Commercial
Contracts (UPICC). (42) Article 8.1 UPICC states:

(1) Where two parties owe each other money or other performances of the
same kind, either of them (‘the first party’) may set off this obligation
against that of its obliger (‘the other party’) if at the time of set-off, (a) the
first party is entitled to perform its obligation; (b) the other party's
obligation is ascertained as to its existence and amount and performance is
due.
(2) If the obligations of both parties arise from the same contract, the first
party may also set off its obligation against an obligation of the other
party which is not ascertained as to its existence or to its amount. (43)
page "234"

There are also similar rules on set-off in Part III of the Principles of European
Contract Law (PECL) published in 2003. (44) Case law has articulated similar
principles. For example, ICC Award No 3540 suggested that:

according to the general principles of law, non-contractual set-off is subject to

four cumulative conditions: similarity and reciprocity of the subjects,
performances of an identical nature, the claims should be certain and liquid, and
finally maturity of the claims (i.e., not subject to a time limit). (45)

Like most harmonisation exercises, the UNIDROIT draft or such arbitral

articulations of principle must either express a preference for the views of one
legal family over others, or find some compromise point between each.
Compromises in harmonisation exercises are often sub-optimal, often papering
over remaining differences in view between negotiators. Such negotiating fora
might seek a compromise simply in order to further the cause of harmonisation,
leaving it to later jurisprudence to make some key refinements. In such
instances, arbitrators stating a preference for one theory of set-off might do so
because they see a preferred uniform policy position or because they adopt a
conflicts methodology and follow this to the legal family from which particular
principles are identified. A comparative analysis of differences between legal
systems and the principles articulated by UNIDROIT should be looked at
primarily in helping us decide whether set-off should be allowed as of right and
if not, by what other principle should admissibility be determined on a case-by-
case basis.

Berger suggests rightly that the historical perspective may help us understand
the different concepts of set-off. In most contentious issues of legal policy, we
are faced with potential conflicts between fairness and efficiency and the
subsidiary elements of certainty versus flexibility. Hence, Roman law saw the
establishment of the right of set-off as an exception to the historical procedural
formalism of classical Roman law which did not initially even allow reverse
claims arising from the same contract. (46) From an early equitable basis, the
right expanded to claims arising out of the same contract to then encompass
claims outside of that contract. That progression in Roman law led civilian legal
systems to develop the right using notions contemplating either set-off or the
alternative description, compensation. Where the common law is concerned, it
also began with the notion of equitable setoff in relation to claims that were
sufficiently connected.

Perhaps the fundamental difference in the development of the laws of set-off in

different legal families is between those that see set-off not simply as a reverse
claim but instead, as a means of performance. On this view, where a claim is
made and a respondent declares a set-off, the respondent is saying that even if
the primary page "235" claim is made out, it has fully performed its obligations
as a result of the declared set-off. When articulated in that manner it can be
argued to be a direct response to the claim and hence an integral part of an
assessment of the continuing validity of that claim by the tribunal. One of the
difficulties with this in the field of arbitration analysis is that it puts all set-off
defences within the one category, regardless of the degree to which the facts
truly relate to the primary claim. A contrary argument might be to the effect that
the only important commercial issue is whether the respondent truly owes
money to the claimant. While that may be so in a purely economic sense, it
cannot be the only way to deal with arbitration where jurisdiction must depend
on some logical consent. Hence a perspective that treats all set-off claims as
performance may not be an ideal way to resolve arbitration jurisdiction
questions as it does not concern itself with the degree to which those facts might
come within the original arbitration clause. An extreme example would be
where the arbitration clause expressly bars set-off rights. The second conceptual
problem is that this doctrinal perspective of set-off as performance fails to
distinguish adequately between the establishment by a tribunal of contested
legal rights and the determination of the remedies that flow from breach of those
rights. For example, if a contested claim is fully made out in law at the same time
as a set-off is made out on unrelated facts, one could argue conceptually that the
set-off merely explains why there will be no remedy by way of an order of
payment of money. One could still say that the primary claim is discretely made
out in full.

Different legal cultures have certainly taken different views on this issue. The
concern is then with the different articulations and theoretical opinions on set-
off and their impact upon a consent-based approach to admissibility questions.
(47) There are a number of different factual permutations that deserve separate

analysis. Where debts are connected, a cross entitlement is often a pure defence
and does not even need to be treated as a set-off. Consider for example, a case of
a buyer and seller who have an ongoing two way commercial relationship with
regular two way payment obligations. The supplier sues the buyer for
outstanding payment but the buyer says the claim fails to take into account
agreed allowances for faulty goods. This need not be separately pleaded as a
setoff if the claimant is only entitled to a net amount under their agreement. It is
simply an allegation that the net position as claimed is wrong. This is at times
described as contractual set-off. If it is expressly or impliedly agreed to in this
way it would generally fall within any arbitration agreement covering the
primary claim.

page "236"

4.4.3.2. Liquidated and Unliquidated Claims and Rights to Set-Off

In most legal systems, more general set-off is not available for unliquidated
damages. It essentially deals with mutual debts. That already raises a challenge
as arbitration agreements are not so limited. The Principles of European
Contract Law (PECL) have moved away from other legal systems on this issue by
not requiring that the set-off must be ascertained as to existence and amount.
(48) Article 13:102(1) PECL provides adjudicators with discretion to allow a set-

off of an unascertained amount where it ‘will not prejudice the interests of the
other party’. Such a test begs the question as to arbitral jurisdiction, as
admissibility in the face of objections would at least be argued to be such
prejudice. However, if admissibility is justifiable via consent, then such an
argument should not succeed as it is initial consent that matters, not consent at
the time of the dispute. The common law has also seen a tempering of this rule
as to unliquidated damages. Hanak v Green (49) saw the English Court of Appeal
allow an unliquidated claim by way of equitable set-off against a damages claim
for defective workmanship. There might also be situations where the reverse
claim, while unascertained, is sufficiently high so that one can be certain that it
exceeds the principal claim. In these circumstances it would be sufficiently
ascertained to support set-off of the entire claim. (50)

Berger suggests that the requirement of an ascertained and existing cross-

obligation can be explained by the function of set-off as a ‘means of private
enforcement of the cross-claim of the party declaring set-off’. (51) Unfortunately,
this policy justification would also not resolve the issue of arbitral treatment. If
that is the essence of the right then it is more in the nature of a distinct remedy
rather than a true defence to the original claim. Furthermore, to describe it as a
means of private enforcement says nothing about the appropriate dispute
resolution forum for determining whether such private enforcement was valid
or not.
4.4.3.3. Automatic Application versus Claims as to Set-Off

There is also the distinct question as to whether a set-off operates automatically

(ipso iure) or whether it requires a declaration by one of the parties. The
distinction should have little relevance to arbitration. If the entitlement is not
within the reference and is not pleaded, it cannot be dealt with by a tribunal. (52)
Berger indeed criticises the development of the ipso iure effect of set-off. Article
8.3 and 8.4 UPICC now indicate that set-off is effected by notice and does not
operate page "237" automatically. This solution better accords with the legal
certainty that is of particular importance in international business. (53) Yet this
arguably removes some of the strength in the argument in favour of automatic
admissibility of set-off. The more it applies automatically to undermine a claim,
the more one could argue that it is an inherent element of a just determination
of the ambit and validity of such a claim. Conversely, if it is a unilateral right of
self-execution, the ability to legally evaluate the validity of that purported self-
execution still must be based on mutual consent if the forum is to be an arbitral
one.

4.4.3.4. Independent and Equitable Set-Off

In different legal systems, it is further classified into independent or equitable

setoff. This distinction is particularly important as it alludes to the degree of
factual connection between the claim and set-off. If the first and perhaps
essential question as to admissibility is consent, then the degree of connection
between primary claim and set-off defence may be relevant to that
determination. An independent set-off at common law is allowed for where it is
capable of being ascertained with suitable precision, that is, being liquidated.
This would also include some damages claims, for example, where they arise out
of an express contractual provision setting up a damages formula, such as in the
case of late performance in construction contracts. Such an independent set-off
need not arise out of related transactions and is seen as purely procedural,
requiring the imprimatur of legal proceedings. As such it cannot be invoked
unilaterally. It is sometimes described as statutory set-off. (54) Aeberli provides a
strong argument that statutory set-off never acquired the characteristics of a
substantive defence and criticises the contrary assertion by Mustill and Boyd. (55)

Conversely, Berger notes that where equitable or transaction set-off is

concerned it operates as ‘a true, substantive defence against the respondent's
liability to pay a debt otherwise due. It may be invoked independently of the
order of a court or arbitral tribunal’. (56) To be characterised as an equitable set-
off the cross-claims must be ‘inseparably connected with the transaction giving
rise to the claim page "238" so that the title of the plaintiff at law to prosecute
his demand is impeached’. (57) The cases suggest that the set-off need not
necessarily arise out of the same contract. Aeberli also argues that at common
law, an equitable set-off has the characteristics of a substantive rather than a
procedural defence. He takes issue with the suggestion by Mustill and Boyd that
equitable set-off is a procedural defence and must therefore come within an
express submission to arbitration. (58) Aeberli notes a number of older common
law cases which dealt with unliquidated damages and which appear to have held
that the only matter to establish for equitable setoff is the simple fact whether
the reverse claim was inseparably connected with the dealings and transactions
which gave rise to the primary claim. After criticising those cases, Aeberli
suggests that equitable set-off in common law should be seen as applicable
‘where the cross-claim alleges matters which can be identified as depriving the
defendant of the benefit for which the plaintiff was demanding payment, or
hinder or prejudice the defendant in enjoyment of that benefit’. (59)

Even if this is an accurate reflection of the position at common law, it does not
give a clear indication of how a tribunal should proceed. If the test is that it
needs to be so inseparably connected as to impeach the title of the plaintiff, how
else could this be so if the cross-claim does not otherwise come within the
arbitration agreement or is otherwise founded on the consent of the parties?
Even if it arises out of the same contract, if it is not within the initial reference to
arbitration, by what principles of consent ought it to be nevertheless included?
This is the nub of the question where set-off admissibility is concerned. Because
it is so complex and uncertain it is unfortunately the case that whichever way a
tribunal goes on these issues, there might be challenges to its determination
and/or enforcement. Even under the common law, an equitable set-off does not
apply automatically. Its equitable basis simply means that it is unconscionable
for the creditor to consider the debtor being in default where an equitable set-
off is sufficient to counter the primary claim. (60) Thus in Aires Tanker
Corporation v Total Transport Ltd the House of Lords considered that a set-off
which had been previously notified, but which was not pleaded in any suit
within a statutory time period, was lost and did not negate the primary claim.

An alternative view of the nature of substantive set-off under the common law is
provided by Wood. (61) He describes equitable or transaction set-off as a ‘self-
help’ remedy. A debtor might unilaterally rely on the remedy or alternatively,
may exercise the right by relying on it as a defence in judicial proceedings. In the
latter case the judgment has a retroactive effect from the point in time at which
it accrued.

page "239"

Derham takes issue with this formulation (62) although it is not necessary for
arbitral purposes to resolve that doctrinal debate. As suggested throughout,
differences between legal families and doctrinal differences within those
families are unlikely to be a sensible gateway for arbitrators to resolve these
questions. Derham also notes situations where a truly substantive set-off
defence may have other significant contractual consequences for the claimant's
rights. He cites examples where one party is entitled to take a particular course
of action only where the other party fails to make a payment as and when due. If,
however, the latter has an equitable basis for refusing to make the payment,
then the express options provided for the benefit of the creditor should not be
seen as coming into play. (63)

In his article on Set-off under UPICC, Berger suggests that ‘(s)et-off is based on
the idea that the performance of mutual claims between two parties must be
simplified and that therefore, whenever equity … requires, they should be set off
against each other’. (64) It is suggested that it would be against good faith to ask
each party to perform its obligations separately. One could readily envisage
circumstances where that would be so, but much is dependent on the facts of
each case and whether the set-off has merit or not, or is instead used as a
delaying tactic. There is also no inherent logical link between saying that it is not
in good faith to claim money when you know you have an equal payment
obligation in reverse and in then saying that such logic must always impose
itself upon the claimant no matter what permutation of dispute resolution
clauses apply in relation to each. Most importantly, good faith notions ought to
apply differently depending on the degree to which the claimant is either aware
of the set-off or accepts its validity. If the claimant does not believe that the set-
off allegation has merit, then the good faith based initial premise simply does
not hold. Thus again it can be concluded that determining admissibility based on
some inherent notion of set-off as a distinct legal concept, particularly when this
varies so much between legal families, is not an ideal way for the arbitral world
to move forward on this issue.

Berger also notes a number of authors who find great similarity between setoff
and counterclaim where as is usual in international commercial arbitration,
money claims are at stake. (65) For example, Bühler and Webster find the
distinction between counterclaims and set-off to be ‘difficult to see’. (66) They
argue that in most systems, the adjudicator must find a claim in a certain
amount and that it arose in a context where a party is entitled to a set-off. They
suggest that if this is an accurate assessment of the law, the first step is really
demonstrating that there is a counterclaim of some nature while the second step
shows that it has a sufficient connection to the primary claim. However, Berger
also suggests that in spite of page "240" similarities, set-off and counterclaims
‘have to be distinguished sharply from each other’. (67)

Even if a tribunal considers that a set-off cannot be brought before it, the mere
presence of such a claim may be relevant to the tribunal's decisions on the
timing of the proceedings and directions as to implementation of the award and
other procedural matters. Parties could be presumed to intend that arbitrators
make such discretionary decisions with an eye to all relevant surrounding
circumstances and not treat the instant dispute as occurring in a vacuum.
However, arbitrators should not seek to resolve the implications of those
external circumstances unless they are directly within jurisdiction.

4.4.3.5. Procedural Rules Dealing with Set-Off

Similar to the variations between procedural arbitration rules with respect to

counterclaims, rules allowing for set-off fall into a number of different
permutations in terms of wording, and by implication, in terms of variations in
the broadness/narrowness of the conditions for admissibility of claims of set-
off. Article 21(3) of UNCITRAL Arbitration Rules 2010 provides that ‘the
respondent may make a counterclaim or rely on a claim for the purposes of a
set-off provided that the arbitral tribunal has jurisdiction over it.’ The ICC Rules
of Arbitration 2012 provision on set-off is contained in Article 30(5), but only in
the context of calculating an advance on costs. That is, as with its provision for
counterclaim, the admissibility of set-off per se is acknowledged, however,
conditions of admissibility are not articulated any further. Article 42(c) of the
WIPO Arbitration Rules (68) falls into the category of provisions which provide
for set-off by the use of the word ‘any’, and as discussed with reference to
counterclaim provisions, the attendant implications of this wording might
distinguish such rules from a rule which provides for ‘a’ claim of set-off, which
could designate a narrower category for the purposes of admissibility, although
it can again be questioned whether this is the clear intent behind the use of the
word ‘any’.

The Swiss Rules 2012 constitute yet another formulation for the admissibility of
claims of set-off. Article 21.5 provides that ‘[t]he arbitral tribunal shall have
jurisdiction to hear a set-off defence even when the relationship out of which
this defence is said to arise is not within the scope of the arbitration clause or is
the object of another arbitration agreement or forum-selection clause’. As such,
this clause unreservedly provides for the widest ambit of admissible set-off
claims. Schedule 2.1 of these rules, dealing with the calculation of the value of
the dispute's claims further provides that ‘[t]he value in dispute is further
increased by the page "241" amount of set-off defences of non-connected claims
to be evaluated by the Arbitral Tribunal’. (69) (emphasis added).

One can readily note various advantages and disadvantages with an approach
such as Article 21.5 of the Swiss Rules 2012. An express rule of that nature
removes most of the uncertainty and potential for costly debate about the
admissibility or otherwise of set-off defences. That alone should reduce
transaction costs significantly. In most cases it will further reduce transaction
costs by removing a multiplicity of actions. A concern to find the appropriate net
figure that a claimant is entitled to will remove problems arising from mutual
payment obligations with attendant timing, cash-flow and in extreme cases
insolvency problems. For those who would see certain types of set-off at least as
inherent elements in identifying the true net amount payable between the
parties, broad rules providing for set-off defences might thus be seen as an
inherent part of a just legal system. Conversely disadvantages of such a rule
include its blanket nature, encompassing both set-offs that are described above
as true defences, primarily equitable set-off, and those that are simply relating
to cash-flow issues. Such a clear-cut rule does not help to distinguish between
these categories to the extent that one believes there are conceptual and policy
reasons for doing so. It also does not help distinguish between valid and bona
fide set-off claims on the one hand and those which are instead aimed at
delaying payment and/or pressuring a claimant into settling for a reduced
amount.

The articulation of the law in the Swiss Rules 2012 also does not make it clear
whether a tribunal has a discretion to consider these matters when a set-off
defence is raised. For example, while the rule stipulates that the tribunal ‘shall’
have jurisdiction over such set-off defences, where these are not part of the
original arbitration agreement it has been suggested that it is not clear whether
the tribunal must consider the defence or whether it instead has a discretion
whether or not to do so. (70) To the extent that it is unclear whether the
discretion exists, this will add transaction costs in individual disputes where this
has to be debated. Because it is a fundamental procedural and jurisdictional
question, uncertainty as to the tribunal's powers may be grounds for challenge
of the award or enforcement proceedings. (71) If there is no discretion, then the
problems alluded to above from a blanket rule point to costs as well as benefits
from such an initiative. If the rules do contain a discretion, there is a question as
to whether different tribunals are likely to be able to apply such a discretion in a
fair and consistent manner.

Even though the inclusionary power seems clear, there still may be uncertainty
where there are clashes between contracts, particularly as this may give rise to
interpretative challenges for the tribunal and complex questions of the interplay
between courts and tribunals in some circumstances at least. As to the first, if
two page "242" separate arbitrations were commenced, each with an Article
21(5) equivalent, could set-off from one be raised under the other? Could the
provisions be used as a basis for seeking consolidation? Could a tribunal say that
the contract later in time is intended to take precedence over the former one as
the most recent indication of the intent of the parties? What if instead, an earlier
contract referred to arbitration subject to Article 21 (5) and a later related
contract has a choice of forum clause? Would it be held to be a variation of the
Article 21(5) entitlements because it is later in time? The tribunals might have
to unravel issues of lis pendens, good faith and abuse of rights. What would
happen if one contract referred to the Swiss Rules including Article 21(5) but
either this or a related contract expressly indicated that there are no rights to
bring set-off defences? What if the second contract was earlier in time,
concurrent or subsequent to the first contract? What if two institutions had a
21(5) equivalent and separate cases were brought to each? Hopefully in most
circumstances, the order of the procedural steps taken by the parties should
indicate what would be a fair procedural determination in the circumstances. At
the very least, in a world where commercial entities can choose between
different arbitral centres, it is a valuable option and experiment to have a highly
respected arbitral venue offering such a model.

At the extreme, a rule might also allow for all cross-claims including unrelated
counterclaims on the grounds of an efficient resolution of all inter-party claims.
Thus Pierre Karrer has even suggested that Article 21(5) could be applied to
counterclaims notwithstanding that it only expressly refers to set-off. (72) The
concerns above would also apply to such an initiative.

4.4.4. Conflict of Laws Issues and Set-Off

The differences between legal families are significant and hence a conflicts
methodology might be one means of making choices between alternatives when
the rules are not explicit. Given that different legal families treat set-off either as
procedural or substantive and also have differing views about the extent to
which a set-off defence is inextricably linked to a claim and hence more likely to
be within the claim's jurisdictional ambit, the question is whether
determinations as to admissibility might simply flow from the answers given by
the law determined to be applicable to the set-off itself.

Here different legal systems diverge as to the proper law that should apply, in
part as a result of differences in view about its essential nature as a defence. For
example, should the law of the set-off follow the law of the primary claim, on the
basis that it inherently undermines it, or should it be found to have its own
governing law based upon the connecting factors to its own essential factual
page "243" elements? Should these questions depend upon the timing of the
claim for set-off? For example, where the set-off predates the claim, it might be
thought to have some level of intrinsic status and value based on the likely
applicable law as and when it arose, although the level is not determined until
adjudication occurs. Yet even that is complicated. If the question is whether set-
off is permissible or not in terms of whether the set-off entitlement is already
due, that would normally be a question to be determined under the law
applicable to that claim, which is normally the law applicable to the contract. (73)
This is to be compared with the contrary view that the law of the set-off as a
defence to the primary claim should follow the law of that claim.

These questions also raise issues of procedural justice. If the applicable law of a
set-off follows the law of the primary claim and if the respondent could choose
instead to raise its entitlement by way of a separate claim rather than as a set-off
defence, it gives the respondent an effective choice over governing law,
depending on where it seeks to raise the issue. Similar strategic questions might
apply to the claimant if the law of set-off follows the law of the claim. Thus if a
claimant could either sue in tort or contract in relation to a commercial joint
venture where a respondent has a separate contractual set-off and if the proper
law of the tortious and contractual primary actions would be different, should
the claimant's choice of how it frames its case affect the law applying to the set-
off?

Berger notes that in France and Belgium a different conflicts principle is applied.
He refers to a cumulative theory under which the set-off is only justified if both
the personal laws of the debtor and creditor would declare it admissible. (74)
Applying this rule to arbitration would make successful applications for
admissibility stronger and more clearly anticipated by both parties. However, it
would deny admissibility to potentially meritorious claims. Berger suggests that
this cumulative approach is also followed by arbitrators to add persuasiveness
to their decisions where they would be justified under a range of applicable
laws. That is quite different to the more restrictive domestic approach that he
identified. On the latter view, unless the set-off satisfies both laws it cannot be
considered. Under the more traditional approach to cumulation by arbitrators,
they are simply looking to be able to say that the more the same outcome would
be mandated under alternative applicable laws, the more they are confident that
their suggested outcome is a just one. More often than not, a cumulative
approach is utilised to identify a false conflict and indicate that the same
substantive principles will apply no matter what choice of law is made. In ICC
Case No 5971, after considering that set-off under related contracts was
admissible, the tribunal turned to the governing law. The tribunal adopted a
comparative approach with a view to seeing that if the page "244" various laws
were essentially harmonised it might not be necessary to discuss the conflicts of
laws principles which might apply. (75) The tribunal did find comparable
principles under each of the potential legal systems that would apply under a
conflicts methodology. On balance, the dual compliance cumulative approach to
set-off admissibility referred to by Berger ought not to be preferred in
international arbitration as such a restrictive approach is unlikely to be
consistent with presumed intent. Berger notes that the approach emanates from
different domestic considerations, namely as a corollary of the ex lege effect of
the compensation légale which historically did not call for a declaration by
either party. (76)

Another question is whether an express selection by the parties of a broad

substantive law which expressly allows for set-off, resolves the admissibility
question in arbitration. Thus if parties agree to arbitrate claims under
procedural rules which do not provide any indication of which set-offs may be
allowed, but have their contractual arrangement subject to UPICC or PECL, could
the express right to set-off within that contract law regime be the basis of a
finding of consent to jurisdiction? Could such an argument be supported by
Article 13:107(a) PECL which provides that the parties can agree to exclude a
right to set-off, thus suggesting that it is an opt-out substantive law rather than
an opt-in model? This might readily apply in simple cases but other problems
arise when UPICC tries to deal with a set-off in the face of multiple claims or
conversely, multiple reverse claims against an individual claim. How should the
set-off be allocated in each case? Article 8.4 UPICC provides discretion to the
party declaring the set-off where it has more than one reverse claim. UPICC does
not directly address the converse situation, although Article 13:105(2) PECL
indicates that a set-off in response to multiple claims also allows for some
choice. Any articulation of substantive law rights as between parties in their
business dealings does not necessarily reflect consent to provide similar choices
in allocating set-off rights to different dispute settlement fora. If the set-off facts
come within the arbitration clause, this would be easy in any event. If they do
not, at most this selection of substantive law is valuable evidence pointing in the
opposite direction to the clause itself, but it is once again a decision forced on a
tribunal based on inadequate drafting.

If a conflicts methodology is to be used under a consent paradigm, a number of

domestic principles might appear to be ill-suited to arbitration. For example, in
some countries it is not possible to set-off claims made in different currencies.
This does not seem to be a sensible presumption in the field of international
commercial arbitration where in most cases, the two parties to the dispute
naturally deal in different currencies. UPICC has also sought to temper this rule.
Article 8.2 UPICC indicates that set-off may be exercised if both currencies are
freely convertible and there is no stipulation that the payment by the party
claiming set-off must be in a specified currency. It might even be the case that
the mutuality of claims called for under Article 8.1 of UPICC which encapsulate
the established principle that page "245" reciprocal claims owing by separate
corporate personalities are not generally eligible for set-off, should not
automatically apply in arbitration. The consent-based paradigm at least allows
for a group of companies theory to provide for jurisdiction in some
circumstances at least. (77)

If a conflicts approach is utilised, the following list of questions might be a guide:

1. There are two distinct questions. First, is the set-off allowed to be raised as a
reverse claim? Secondly, if so, has it been made out?
2. In considering admissibility there are four conflicts options to consider:
(a) the lex causae of the primary claim on the theory that it is a substantive
defence.
(b) the lex arbitri on the basis that admissibility is a procedural issue.
(c) the personal laws of either or both of the parties.
(d) the law of closest connection, however that is to be determined.

This section does not seek to resolve these conflicts issues as to set-off. Instead
it has sought to show problems with undue reliance on conflicts approaches.
There is no consensus as to which conflicts rule to apply and no guarantee that
this would lead either to consistent results or results that would meet the
parties' legitimate expectations. Those advocating a conflicts approach might
wish to consider how they would deal with the following example. Let us
assume that two parties have selected a substantive law that sees a set-off as an
inherent defence that need not be separately declared. At the same time, they
have selected a set of arbitral rules and/or an arbitration agreement that
expressly denies a set-off right. Which element of the parties' consent should
prevail, the clear denial of the set-off right or the indirect inclusion via the
choice of substantive law? The example is not presented in order to advocate for
one view over another. Instead, this example shows how there can be
potentially conflicting but important aspects of consent that a tribunal should
seek to reconcile on a case-by-case basis.

For the purpose of this section it can merely be noted that a conflicts
methodology under current domestic approaches will not solve problems in a
way that would be beyond debate. Given that tribunals are given broad
discretions as to which conflicts rules to apply and given the extreme differences
in approaches of domestic regimes, the outcome would not be clear and
consistent and would often lead to substantive principles applying which are ill-
suited to international arbitration.

4.4.5. Implied Consent to the Raising of any and All Defences as a

Justification for Set-Off

As noted above, many commentators and tribunals have tended to differentiate

between counterclaims and set-off on the basis that the latter is considered to
be a page "246" defence. It is suggested that it flows as a fundamental principle
of justice that a respondent should be able to raise any available defence. This
approach can too easily fail to distinguish between different types of defences. It
also fails to consider the way the concept and justification fit within a consent-
based arbitration paradigm. The previous analysis suggested that there is a
significant difference between different forms of set-off. The mere description of
all as ‘defences’ fails to address the degree of connection between the set-off
claim and primary claim, which should be the key to deciding on the treatment
of set-off within a consent paradigm.

There is also the problem of the interplay between the arbitration agreement,
procedural rules and the nature of the set-off in an individual case. In looking at
setoff, Berger invites separate consideration of the lex arbitri and the scope and
interpretation of the arbitration agreement in deciding on procedural
admissibility of a set-off defence. While these are certainly two distinct sources
of relevant principles, these are both evidentiary aspects of consent. If the
parties have expressly selected a Seat that either expands or contracts the
ability to plead such defences, they have provided express evidence in that
regard. The same is true if they have selected arbitral rules that address the
issue, or if they expressly cover the matter in their contractual arbitral
agreement.

The mere nature of set-off claims as defences does not resolve the issue without
broader consideration of all circumstances. Nevertheless, for the purposes of
analysis it is appropriate to consider when and why a defence should be
admitted as of right, absent any other factors that add to or detract from a
finding of consent between the parties. In this context, it is important to
consider different categories of defences. The first category involves defences
that are inherent in the particular dispute and which any sophisticated legal
system would say are part of the process by which the ultimate rights and
obligations of the parties must be determined. Examples of such defences in the
sphere of commercial contracts include the duty to mitigate loss and the
obviation of damages in the face of force majeure or frustration. Other examples
of the first category that permeate other areas of law include contributory
negligence and abuse of rights under civilian principles. All such defences go to
tempering the ramifications of the claim within the four walls of its own fact
situation. They directly relate to the claim as such and should therefore be
included under any view of justice. In civil cases the proper measure of damages
can only be identified after the claimant has set up its gross damages
entitlement under the applicable law which will usually look at issues of
foresee-ability and causation. That figure then needs to be tempered in light of
the above-mentioned defences if and when they are made out. (78) Because they
are inextricably linked to the facts of the claim, they will also have identical links
to the relevant court or tribunal's jurisdictional mandate. There could not even
be questions of page "247" statute of limitations or out-of-date counterclaims of
this nature. It would also be highly unusual if the tribunal which had been
selected as appropriately expert to deal with the primary claim, was not
similarly expert to deal with the defence. (79) The same logic should apply to
other similar defences which may vary in title and content from jurisdiction to
jurisdiction. (80)

At the other extreme are differences that do not affect the legal entitlement of
the primary claim but deal instead with issues of cash-flow and duplication.
They are premised on the understanding that even if the claimant can make out
its primary claim, the respondent has an equal or greater entitlement under a
distinct right, which means that the claimant has no net right to any legal
remedy. A key policy reason underlying this aspect of the domestic law of set-off
is in cases of potential insolvency. Treating the set-off as a means of identifying
the true net figure, if any, to be passed between the parties, shields the
respondent from the normal position of an unsecured creditor. Otherwise the
respondent might be required to pay the full amount of the claim to a liquidator
and then find that its set-off amount places it in a long list of unsecured
creditors. Yet most legal systems provide separate rules for insolvency set-off
and this category is not included in UPICC or PECL, so this cannot be a key factor
for arbitrators deciding a priori intent.

The distinction between inherent defences and cash-flow situations does not
mean that the second category should never be seen as admissible in arbitral
page "248" proceedings. It is simply that where consent is used as a key to
determining admissibility, the consent logic in the context of cash-flow and
duplication concerns, differs significantly to the consent logic when one is
discussing a defence integral to identifying the validity of the claim. Even that
statement is contentious as in some legal systems and conflicts theories, there is
no inherent bright line distinction between these two categories. At the very
least, however, the fact that a particular domestic legal system describes such an
entitlement as a defence should not be determinative in an arbitral situation. (81)
Such a domestic legislative policy often aims to minimise duplicate proceedings
and save taxpayers costs. Again these are factors which might help a tribunal
identify presumed intent, but a mandatory rule of a government seeking to save
its own court expenses is based on a quite distinct logic. (82)

It is thus suggested that it is important to distinguish between different types of

defences; on the one hand those that are truly inherent as responses to the
primary claim and those which are more in the nature of defences to the cash-
flow implications of a final order in favour of the claimant. Because some set-off
claims fit the first category but most do not, it does not advance the appropriate
consideration of their place in arbitral proceedings by simply acknowledging
that they are treated as defences in domestic litigation systems. The ultimate
question in all cases is whether the parties intended at the outset for there to be
cross-performance of these obligations and whether there is a single dispute
resolution body intended to analyse both elements of that cross-performance.
Inherent defences would fall within the arbitration agreement in any event.
Cash-flow defences call for additional presumptions as to intent to be justifiable.

4.5. Defences to Counterclaims

Where rules call for an answer by respondent to claims made, they will typically
also call for a reply or answer by claimant to any counterclaims. For example,
Article 5(6) of the ICC Rules 2012 requires a claimant to submit a reply to any
counterclaim within thirty days from the date of the receipt of the counterclaim
communicated by the Secretariat. An extension may be granted prior to the
transmission of a file to the tribunal.
page "249"

4.6. Commencement of Arbitration Without a Seat

Section 3.5 above pointed to the importance of the Seat of arbitration in framing
the procedural law. Section 3.5.1 above outlined the various features that the
parties should look for in selecting a Seat, ideally in their arbitration agreement.
While the recommended approach is to make such a selection either at the
outset or when the dispute is known, attention needs to be given to situations
where this has not occurred. Some rules will typically call for comments about a
preferred Seat in the Request and Answer. (83) If that leads to agreement there is
no further problem. In many cases, however, agreement is not possible between
the parties. ICC figures suggest that between 10% to 20% of cases are
commenced without a Seat having been selected. (84) In these circumstances, it is
important to know how to proceed. It is important to determine who is to make
the selection, by what process and at what stage and what criteria they ought to
apply. In the extreme, attention will sometimes need to be given as to whether
the arbitration clause is fatally pathological due to a failure to establish a
minimum framework by such a selection.

It is necessary to distinguish between a number of scenarios in answering these

questions. The first scenario is where the parties have simply not selected a
Seat. In those circumstances, if they have selected a set of arbitral or
institutional rules, one would look to those rules to determine the methodology
for Seat selection. In most cases the residual selection of the Seat would be
delegated to the tribunal, the institution or a domestic court. Where the ICC is
concerned, the Seat is selected by the Court of Arbitration. If the parties have not
selected a Seat, another possibility is that they may be subject to some
international treaty which itself contains direct or indirect methodologies for
determining the place of arbitration. Examples include the 1975 Inter American
Convention on International Commercial Arbitration and the 1961 European
Convention on International Commercial Arbitration. (85) Other rules are more
prescriptive. Where the parties have not selected a Seat, Article 16.1 of the LCIA
Rules provides for London as the Seat. This is different to other institutions that
employ a broad discretion looking at all relevant factors. Nevertheless, the LCIA
will happily supervise an arbitration where the parties have selected a different
Seat. In some cases, it may not even be necessary to identify a Seat, where the
selected rules are sufficiently comprehensive.

The second scenario is where it is not clear from the parties' agreement whether
they have in fact selected the Seat or not. An example would be an arbitration
clause calling for arbitration ‘under the arbitral rules of the ICC, Paris’. This is
ambiguous because it is not clear whether the reference to Paris is simply a
reference to the geographical location of the ICC whose rules have been selected,
or instead, whether the parties are separately stipulating Paris as the Seat page
"250" alongside selection of the ICC Rules. As with any question of ambiguous
drafting, there is a need for interpretation. It would naturally be for the tribunal
to make such a determination. Section 2.3.7 above looked at the approaches to
interpretation in such circumstances. In the above example, the arbitration
should be valid in any event. If it was a separate selection of Paris as the Seat
with the ICC Rules applying, there is more than enough to effectively complete
the arbitration. Similarly, if it was only seen to be a selection of the ICC Rules,
there is a clear mechanism for Seat selection and in turn, identification of the lex
arbitri.

A more troubling scenario is where a clause is drafted in such a way that it is

ambiguous whether the parties have selected the Seat, and if a determination
was made that this did not occur, the agreement would be fatally flawed by
reason of the inadequacy of other aspects of the arbitral agreement. In such
circumstances, the better view is that an effete utile approach to interpretation
might incline to the view that the parties indeed made such a selection as it is
closer to their implied intent to find a valid arbitration as opposed to an invalid
one. Of course the party opposing this conclusion might be able to argue in the
circumstances that they would not have agreed to the particular Seat contended
for, which argument may be stronger if it is not a neutral Seat and there is some
contemporaneous evidence suggesting that they had such concerns at the time
of negotiation.

Where there is ambiguity, another approach is to see if the parties accept that
the tribunal selects a Seat even if they cannot agree on the location. (86) In such
circumstances, such an agreement might obviate a challenge to jurisdiction and
simply leave it for the tribunal to make the particular determination.

4.7. Service

There are three aspects of service and notice requirements. The first need is to
comply with the requirements in the lex arbitri and any stipulated rules and any
agreement of the parties. While parties are entitled to agree on service, in
extreme cases even an agreed method may be improper if it interferes with
fundamental principles of due process. An example would be a stipulation that a
party cannot claim lack of receipt of documents. The second, where some
arbitral discretion remains, is for the tribunal to establish appropriate methods
to ensure due process, minimise guerrilla tactics and promote timeliness. The
third and related aspect is to consider when irregularities in notice and service
elements may give rise to grounds for annulment or refusals to enforce arbitral
awards or otherwise undermine the validity or effectiveness of any steps that
have been taken in the proceedings.

Where lex arbitri and arbitral rules are concerned, these should be fully
complied with. It would be dangerous to consider substantial compliance as
sufficing as sometimes operates in domestic legal systems, given that arbitral
rules and most commonly lex arbitri will follow party consent, hence
constituting a requirement page "251" under their arbitration agreement.
Nevertheless, in some cases, a tribunal will need to rule on whether this suffices
as it may have a crucial impact on whether deadlines have been met.

Even full compliance with lex arbitri and arbitral rules will not necessarily
obviate any potential problems. This is because the rules typically are limited to
requiring proof that relevant documents have been sent in stipulated form and
not proof of actual receipt. The UNCITRAL Rules 2010 are illustrative. They
indicate that notices should be delivered to certain addresses, allow for
transmission by means providing a record and articulate deemed dates of
receipt. Yet deeming provisions cannot presume actual receipt. Nor can they
guarantee that due process is satisfied in all permutations of circumstances.
Hence circumstances may arise where one party has done all that is required
but the other still argues that key documents were not brought to its attention.
The latter scenario may be divided into situations where the fault was outside
its control, such as a fault with an external server where email service is
permitted; to cases where the party claiming lack of notice was at fault but not
intentionally so, for example, inadvertently failing to notify a change of address
for service; and situations where there is a suspicion that a party is seeking to
avoid service of documents or is lying about lack of receipt. A duty to promote
an adequate opportunity to present a case and produce an enforceable award
requires a tribunal to consider how best to deal with each such scenario if and
when it arises. A relevant factor may be that enforcement should not normally
occur against a party who does not know of the initiation of the arbitration. For
example, the Swedish Supreme Court has refused to allow enforcement of an
arbitral award against a respondent who had failed to notify a change of address
as per a requirement under the contract. The court considered that no proper
notice had been received as per Article V(1)(b) of the New York Convention. (87)
It is therefore the case that even where the rules specify a particular form of
communication, a tribunal may need to require additional methodologies to
ensure proper notice. Special care needs to be taken where a respondent is
unrepresented. This scenario is discussed in section 6.15.

Where corporations are concerned there is also a need to consider who may be
authorised to receive notices and communications. If the lex arbitri and arbitral
rules are silent it may be left to domestic law to consider how one properly
communicates with particular legal persons. Where receipt is denied, service by
courier or registered mail will at least have an independent transport entity able
to attest to delivery and usually with a recipient's signature, although no
attempt is made by such couriers to determine whether any recipient has due
authority or whether an unauthorised person will hand the document to the
relevant company officer. In some cases the request for arbitration may indicate
particular counsel known to represent respondent. In such circumstances there
is a need to determine if that counsel actually has the requisite authority as it is
only an assertion to that effect from the opposing party. In many cases that
assertion will be accurate as the page "252" commencement of arbitration may
follow legal letters of demand and discussions between counsel, often with a
warning that proceedings are about to be brought.

Some rules provide for the institution itself to serve the request or notice while
others call on the claimant to do so. While an institution will no doubt deal with
situations where it is obvious that service has not been effected, such as return
of a document by a courier, it is essentially for each party to ensure proper
notice and not rely on an institution to make searches or inquiries.
4.8. Communication

In addition to formal notices as between the parties, there is also a need to

articulate the way that the tribunal will generally communicate with the parties
and vice versa. (88) The two key approaches to communications are for each
party to communicate with the tribunal which then distributes the material to
all of the other parties, or the more typical approach, to have parties submit
simultaneously to the tribunal and other parties. The latter is far more efficient.
In some cases a tribunal might feel it is preferable to be the gateway so that one
party does not gain an unfair advantage. An example would be where
simultaneous submission has been called for. If one party is late but receives the
other party's submission first, it effectively gets a right of reply which was never
intended in the first place.

The tribunal should identify a communication protocol at the earliest possible

opportunity. While email has proved a great boon in relation to international
transactions, care should be taken when it is used as a means for serving
important material in an arbitration. Care needs to be taken to ensure that there
is adequate proof that certain key communications have in fact been received. A
sender is not always appropriately advised when an email has not been
received. Similarly, a party might wrongly claim lack of receipt but the contrary
is difficult to prove. Hence it is often desirable to follow email with hard copy
service of the more important documents within the arbitral process, although it
would be important to know which version triggers time periods. If soft copy
documents are alone utilised, there is even the potential problem as to their
wrongful modification by the receiving party and the need to at least consider
technological means to render documents as read only. In some cases registered
mail or hand delivery may be preferable. The tribunal may need to draw a
distinction between merely administrative and technical communications on the
one hand and more central communications such as Requests and Answers and
applications for joinder or consolidation on the other hand.

It is important that an arbitrator communicates equally with the parties

otherwise this would offend against the right to equal treatment. Arbitrators
ought not to engage in ex parte communications with parties or their counsel
except in unavoidable and otherwise justifiable circumstances. Enforcement has
been refused where an arbitrator received a letter from a claimant but did not
page "253" communicate this to a respondent. (89) It would not be appropriate to
ignore this rule even in situations where there is thought to be no real harm to
the party not involved in the communication. For example, the LCIA Court
considered it to be a failure to act fairly between the parties when an arbitrator
conducted a meeting with one party alone but provided a verbatim transcript to
the other side. (90) Nevertheless, the parties could give an arbitrator the right to
communicate separately, perhaps on minor administrative matters where it is
not necessary to have all parties present. Similarly, Rule 5.5 of IBA Ethics Rules
indicates that substantial hospitality should not be accepted directly or
indirectly from any party. This might arise, for example, where the Seat of
arbitration is in the home city of one of the parties who then seeks to host the
tribunal at social functions. On the one hand, one might wish to have a friendly
and informal environment to an arbitration but on the other, no preference
should be given to one side.

The duty of communication is both with the parties and with any institution that
is involved. Hence even in the case where an arbitrator wishes to make a written
direction relevant to one party alone, copies should be sent to the other parties
and to the institution. For example, this may arise where one party has failed to
meet a time limit for submission of written arguments. The other party needs to
know this in any event as it may wish to apply to extend its own procedural
timelines or seek costs sanctions in due course. An institution also needs to be
kept fully aware of procedural orders so it can plan its own administrative
support accordingly. It would allow the institution to take organisational steps
efficiently, such as organising staff, hearing rooms, telephone conferences or
transcription or translator or interpreter services where these may be required.

page "254"

1 Rolf Stürner, ‘The Principles of Transnational Civil Procedure: An Into to the

Basic Conceptions’, RabelsZeitschrift 69, no. 2 (2005): 221 referring for example
to Art. 1 French NCCP.
2 Ibid., 233.
3 Common lawyers refer to statements of claim and responses as pleadings. As

with many terms, this can confuse civilian lawyers who often use the term to
denote representations made by counsel to an adjudicator. Nevertheless, this
term is now used in Art. 3 of the ICC Rules 2012.
4 Article 36(2).
5 See, e.g., ICC Rules 2012 Art. 4(4).
6 Siegfried H. Elsing & John M. Townsend, ‘Bridging the Common Law-Civil Law

Divide in Arbitration’, Arbitration International 18, no. 1 (2002): 60.

7 In some cases it could even be desirable to have factual witness statements

produced concurrently so that all concerned can evaluate the evidentiary record
in the context of the claims each party is making. If documents are to be
included and they need translation, then it is appropriate for the party
submitting the document to organise the translation at the relevant time. It is
preferable that the parties agree on an independent and expert translator to
avoid disputes as to the proper terminology.
8 See, e.g., HKIAC Rules Art. 4.5.
9 ICC Case No. 6228 of 1990, ICC International Court of Arbitration Bulletin, 8,

no.1 (1997): 53.

10 Ibid.
11 Julian Lew, Loukas Mistelis & Stefan Kröll, Comparative International
Commercial Arbitration (The Hague: Kluwer Law International, 2003), 520.
12 See, e.g., UNCITRAL Model Law Art. 16(2).
13 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2214.

14 Yves Derains & Eric A. Schwartz, A Guide to the ICC Rules of Arbitration, 2nd
edn (The Hague: Kluwer Law International, 2005), 57; Emmanuel Gaillard &
John Savage (eds), Fouchard, Gaillard, Goldman on International Commercial
Arbitration (The Hague: Kluwer Law International, 1999), 656.
15 Françoise Lefebvre, ‘The Scope and Contents of the Request for Arbitration in

the Comparative Perspective’, in Arbitral Procedure at the Dawn of the New

Millennium, Report of the International Colloquium of CEPANI, 15 October 2004,
ed. Stephen Bond (Brussels: Bruylant, 2005), 26.
16 Swiss Rules 2012 Art. 3.5; DIS Rules Art. 6.4.
17 Article 4(4).
18 ICC Case No. 6784 of 1990, ICC International Court of Arbitration Bulletin 8,

no. 1 (1997): 53.

19 See also ICC Case No. 6039 of 1992, ICC International Court of Arbitration

Bulletin, 8, no. 1 (1997), 54.

20 ICC Rules 2012 Art. 5(1).
21 The balance of this chapter draws on and extracts material from Michael

Pryles & Jeff Waincymer, ‘Multiple Claims in Arbitrations Between the Same
Parties’, Fifty Years of the New York Convention: ICCA International Arbitration
Conference, Congress Series No. 14, ed. Albert Jan van den Berg (Alphen aan den
Rijn: Wolters Kluwer, 2009), and is utilised with permission of the co-author.
22 Vladimir Pavić, ‘Counterclaim and Set-Off in International Commercial
Arbitration’, Belgrade Law Review (2006): 104; Gary Born, International
Commercial Arbitration, 3rd edn (2009), 1801 ; Emmanuel Gaillard & John
Savage (eds), Fouchard, Gaillard, Goldman on International Commercial
Arbitration (The Hague: Kluwer Law International, 1999), 660.
23 Final Award in ICC Case No. 7314 of 1995, in Albert Jan van den Berg (ed.),

Yearbook of Commercial Arbitration Vol XXIII (The Hague: Kluwer Law

International, 1998), 49.
24 Descriptions in other languages include compensación, compensazione,
Aufrechnung, Verrech-nung, and Verrekening.
25 Klaus Peter Berger, ‘Set-Off in International Economic Arbitration’,
Arbitration International 15 no. 1 (1999): 53.
26 See, e.g., Société Empresa de Telecomunicaciones de Cuba SA v. Telefonica

Antillana SA and SNC Banco Nacional de Comercio Exterior, Cour d'Appel de Paris
lère Chambre, Decision of 16 November 2006.
27 See Art. 19 ICC Rules.
28 Article 3.10 reads, in part: ‘Any counterclaim or set-off defence shall in

principle be raised with the Answer to the Notice of Arbitration.’

29 Rules which provide for counterclaim in this way include the following:

Article 4.1(b) of the SIAC Rules (‘The Respondent shall send to the Claimant a
Response within fourteen days of receipt of the Notice of Arbitration. The
Response shall contain: […] (b) a brief statement of the nature and
circumstances of any counterclaim […]’), Art. 5.3(d) of the ACICA Rules (‘The
Answer to Notice of Arbitration may also include: […] (d) any counterclaim
[…]’), Art. 15 of the CIETAC Rules 2012 (The Respondent shall file a
counterclaim, if any, in writing within forty-five (45) days from the date of
receipt of the Notice of Arbitration), Art. 2.1(b) of the LCIA Rules (‘Within 30
days of service of the Request on the Respondent, (or such lesser period fixed by
the LCIA Court), the Respondent shall send to the Registrar a written response
to the Request (‘the Response’), containing or accompanied by: […] (b) a brief
statement describing the nature and circumstances of any counterclaims
advanced by the Respondent against the Claimant’.), Art. 36(b) of the WIPO
Expedited Arbitration Rules (‘Any counter-claim or set-off by the Respondent
shall be made or asserted in the Statement of Defense…’), and Art. 5(b) of the
HKIAC Short Form Arbitration Rules (‘Within 14 days of receipt of the
Claimant's Statement of Claim case file, the Respondent shall submit to the
Arbitrator and to the Claimant a Statement of Defence containing: […] (b) any
counterclaim, together with a brief statement of the remedies sought […]’).
However, the purview of the WIPO Rules generally may be understood to
envisage a wider scope for the arbitrability of counterclaims when provisions
allowing for counterclaim are read alongside Art. 63(d) of the Expedited
Arbitration Rules: ‘Where the amount of the counter-claim greatly exceeds the
amount of the claim or involves the examination of significantly different matters,
[…] the Center in its discretion may establish two separate deposits on account
of claim and counterclaim.’ (emphasis added)).
30 ‘For the purposes of this Law […] where a provision of this Law, other than in

articles 25(a) and 32(2) (a), refers to a claim, it also applies to a counter-claim,
and where it refers to a defence, it also applies to a defence to such counter-
claim.’
31 ‘If one of the parties claims a right to a set-off with regard to either claims or

counterclaims, such set-off shall be taken into account in determining the

advance to cover the costs of arbitration in the same way as a separate claim
insofar as it may require the Arbitral Tribunal to consider additional matters.’
32 Christopher Kee, ‘Setoff in International Arbitration - What Can the Asian

Region Learn?’, Asian International Arbitration Journal 1, no. 2 (2005): 147-148.

33 Klaus Peter Berger, ‘Setoff in International Economic Arbitration’, Arbitration

International 15, no. 1 (1999): 63.

34 See, e.g., Art. 11(1) of the Rules of Arbitration and Conciliation of the Vienna

Arbitration Centre. Similarly the ICDR Rules provide that ‘a respondent may
make counterclaims or assert set-offs as to any claim covered by the agreement
to arbitrate’ (AAA International Arbitration Rules Art. 3(2)) and that ‘[a] party
may not amend or supplement a claim or counterclaim if the amendment or
supplement would fall outside the scope of the agreement to arbitrate’ (AAA
International Arbitration Rules Art. 4) (emphasis added).This is shared to a
degree by the ICSID Rules of Arbitration Art. 40(1) of which provides that ‘a
party may present an incidental or additional claim or counter-claim arising
directly out of the subject-matter of the dispute, provided that such ancillary
claim is within the scope of the consent of the parties and is otherwise within
the jurisdiction of [ICSID]’ (ICSID Rules of Procedure for Arbitration Disputes
Art. 40(1)).
35 Stooke v. Taylor [1880] 5 QB 569, 575 as cited in Klaus Peter Berger, ‘Setoff in

International Economic Arbitration’, Arbitration International 15, no. 1 (1999):

59.
36 See, e.g., ICC Award No. 3540 of 1980, reprinted in Michael Jarvin & Yves

Derains (eds), Collection of ICC Arbitral Awards (1974-1985) (Paris: ICC

Publishing, 1990) 105, 112, 399, 402.
37 Klaus Peter Berger, ‘Setoff in International Economic Arbitration’, Arbitration

International 15, no. 1 (1999): 54.

38 Ibid.
39 In some domestic legal systems efficiency arguments even mean that some
counterclaims, where sufficiently linked to the primary claim, must be pleaded
or they will be lost under principles of res judicata, e.g., Rule 13 of the United
States Federal Rules of Civil Procedure (2007) which distinguishes between
compulsory and permissive counterclaims. Aeberli notes that the distinction in
arbitration has become blurred because it emanated from domestic legal
systems but has been reduced in significance in some jurisdictions because of
procedural reforms: Peter Aeberli, ‘Abatements, Setoff and Counterclaims in
Arbitration Proceedings’, Arbitration and Dispute Resolution Law journal 3, no. 1
(1992): 2.
40 The history of the establishment of statutory set-off shows other reasons why

it has little direct relevance for arbitral matters. Aeberli notes that the
legislation sought to remove the potential injustice if a defendant might be
imprisoned for non-payment of debts when in fact money was owed in the other
direction: Peter Aeberli, ‘Abatements, Setoff and Counterclaims in Arbitration
Proceedings’, Arbitration and Dispute Resolution Law journal 3, no. 1 (1992): 4
(citing Stoke v. Taylor (1880) 5 QBD 569; Green v. Farmer (1786) 4 BURR 2214).
A statutory remedy was required to deal with this as equity had only established
set-off rights for connected transactions.
41 Klaus Peter Berger, ‘Setoff in International Economic Arbitration’, Arbitration

International 15, no. 1 (1999): 53.

42 Klaus Peter Berger, ‘Setoff’, ICC International Court of Arbitration Bulletin

2005 Special Supplement, Issue no. 662 (ICC Publication, 2005): 17.
43 UNIDROIT Principles of International Commercial Contracts 2010 Art. 8.
44 Klaus Peter Berger, ‘Setoff’, ICC International Court of Arbitration Bulletin

2005 Special Supplement, Issue no. 662 (ICC Publication, 2005): 17.
45 ICC Case No. 3540 made 3 October 1980, (Journal du Droit International

(Clunet) 4 (1981, no. 4)): 924.

46 Klaus Peter Berger, ‘Setoff in International Economic Arbitration’, Arbitration

International 15, no. 1 (1999): 54.

47 For a more detailed analysis of the comparative position see Klaus Peter

Berger, ‘Setoff in International Economic Arbitration’, Arbitration International

15, no. 1 (1999): 54-56; Philip Wood, English and International Setoff (London:
Sweet & Maxwell, 1989); S.R. Derham, Setoff, 2nd edn (Oxford: Clarendon Press,
1996); Vladimir Pavić, ‘Counterclaim and Set-Off in International Commercial
Arbitration’, Belgrade Law Review (2006): 101.
48 Klaus Peter Berger, ‘Setoff’, ICC International Court of Arbitration Bulletin

2005 Special Supplement, Issue no. 662 (ICC Publication, 2005): 20.
49 [1958] 2 QB 9.
50 Klaus Peter Berger, ‘Setoff’, ICC International Court of Arbitration Bulletin

2005 Special Supplement, Issue no. 662 (ICC Publication, 2005): 22.
51 Ibid., 17.
52 Berger also notes that in litigation the same procedural issue applies. Klaus

Peter Berger, ‘Setoff in International Economic Arbitration’, Arbitration

International 15, no. 1 (1999): 56.
53 Klaus Peter Berger, ‘Setoff’, ICC International Court of Arbitration Bulletin

2005 Special Supplement, Issue no. 662 (ICC Publication, 2005): 23.
54 S.R. Derham, Setoff, 2nd edn (Oxford: Clarendon Press, 1996), 56. Aeberli

notes and criticises an English Court of Appeal decision that a statutory set-off
can be raised against a claim for specific performance. He supports the
dissenting judgment of Kerr LJ that at most the cross-debt is relevant to the
equity of granting specific performance. Peter Aeberli, ‘Abatements, Setoff and
Counterclaims in Arbitration Proceedings’, Arbitration and Dispute Resolution
Law Journal 3, no. 1 (1992): 5 citing Bicc v. Burndy [1985] 1 All ER 417.
55 Peter Aeberli, ‘Abatements, Setoff and Counterclaims in Arbitration
Proceedings’, Arbitration and Dispute Resolution Law Journal 3, no. 1 (1992): 6,
citing Michael Mustill & Stewart Boyd, The Law and Practice of Commercial
Arbitration in England, 2nd edn (London: Butterworths, 1989), 130.
56 Klaus Peter Berger, ‘Setoff in International Economic Arbitration’, Arbitration

International 15, no. 1 (1999): 57 citing AWA Ltd v. Exicom Australia Pty Ltd
[1990] NSWLR 705 and ‘The Kostas Melas’ [1981] 1 Lloyd's Rep 18.
57 The Angelic Grace [1981] 1 Lloyd's Rep 288. See also National Westminster

Bank plc v. Skelton [1993] 1 WLR 72.

58 Michael Mustill & Stewart Boyd, The Law and Practice of Commercial
Arbitration in England, 2nd edn (London: Butterworths, 1989). 130.
59 Peter Aeberli, ‘Abatements, Setoff and Counterclaims in Arbitration
Proceedings’, Arbitration and Dispute Resolution Law journal 3, no. 1 (1992): 10.
60 S.R. Derham, Setoff, 2nd edn (Oxford: Clarendon Press, 1996): 57–58.
61 Philip Wood, English and International Setoff (London: Sweet & Maxwell,

1989): 111-112.
62 S.R. Derham, Setoff, 2nd edn (Oxford: Clarendon Press, 1996): 57–58.
63 Ibid., 62-63.
64 Klaus Peter Berger, ‘Setoff’, ICC International Court of Arbitration Bulletin

2005 Special Supplement, Issue no. 662 (ICC Publication, 2005): 18.
65 Klaus Peter Berger, ‘Setoff in International Economic Arbitration’, Arbitration

International 15, no. 1 (1999): 57–58.

66 Michael Bühler & Thomas Webster, Handbook of ICC Arbitration:
Commentary, Precedents, Materials, 1st edn (London: Sweet & Maxwell, 2005),
76.
67 Klaus Peter Berger, ‘Setoff in International Economic Arbitration’, Arbitration

International 15, no. 1 (1999): 58.

68 WIPO Arbitration Rules Art. 42(c) provides as follows: ‘Any counter-claim or

set-off by the Respondent shall be made or asserted in the Statement of Defense

or, in exceptional circumstances, at a later stage in the arbitral proceedings if so
determined by the Tribunal. Any such counter-claim or set-off shall contain the
same particulars as those specified in Article 41(b) and (c).’
69 The SIAC Arbitration Rules fall into a category which provides no explicit

acknowledgement of set-off claims.

70 Vladimir Pavić, ‘Counterclaim and Set-Off in International Commercial
Arbitration’, Belgrade Law Review (2006): 108.
71 Ibid., 109.
72 Pierre Karrer, ‘Arbitration under the Swiss Rules of Arbitration in
Switzerland and Elsewhere’ (Paper presented at the 12th Croatian Arbitration
Day, Zagreb, 2 December 2004), 6, cited in Vladimir Pavić, ‘Counterclaim and
Set-Off in International Commercial Arbitration’, Belgrade Law Review (2006):
111.
73 Berger notes Art. 10(1)(b) Rome Convention which indicates that the law

applicable to the contract governs all issues relating to the performance of the
contract: Klaus Peter Berger, ‘Setoff’, ICC International Court of Arbitration
Bulletin 2005 Special Supplement, Issue no. 662 (ICC Publication, 2005): 23.
74 Klaus Peter Berger, ‘Setoff in International Economic Arbitration’, Arbitration

International 15, no. 1 (1999): 62.

75 ICC Case No. 5971, para. 138.
76 Klaus Peter Berger, ‘Setoff in International Economic Arbitration’, Arbitration

International 15, no. 1 (1999): 62-63.

77 See section 7.6.12.
78 Another simple situation that does not call for complex admissibility
determinations is where a breach causes both benefit and detriment to the
claimant and hence the damages caused are only the net amount: Klaus Peter
Berger, ‘Setoff in International Economic Arbitration’, Arbitration International
15, no. 1 (1999): 53.
79 In the rare cases where this was not so, the respondent can at least consider

these issues when recommending its choice of arbitrator, although it is

undesirable to allow for notification of defences after tribunal composition
where this would disadvantage the claimant at the selection stage.
80 Craig Park Paulsson note the related doctrine of exception non adempleti

contractus. Under this doctrine, one party's performance is excusable because of

the failure of the other. The authors suggest that there is evidence that it is an
autonomous rule of international arbitration (see ICC Case No. 2583 of 1976 1
ICC Awards 304; Case No. 3540 of 1980 1 ICC Awards 105, 399). While Craig
Park Paulsson define exceptio broadly, Black's Law Dictionary, for example,
defines it in narrower terms: ‘An exception in a contract action involving mutual
duties or obligations to the effect that the plaintiff may not sue if the plaintiff s
own obligations have not been performed.’ Bryan Gardner (ed.), Black's Law
Dictionary, 8th edn (St Paul, MN: West Group, 2004), 603. ICC Case No 3540
Award made 3 October 1980 (Journal du Droit International (Clunet) 1981, No
4, 914) described the exceptio doctrine as meaning that the plaintiff is not
entitled to relief ‘because he has not performed his own part of the agreement’.
It also considered that this principle was part of the general principles of law
which form the lex mercatoria, which was the law they had chosen to apply
when expressly empowered as amiables compositeurs. Another example is that a
right to claim abatement was initially an exception to the common law rule that
counterclaims could not be raised as defences to an action but instead needed to
be brought in separate proceedings (Peter Aeberli, ‘Abatements, Setoff and
Counterclaims in Arbitration Proceedings’, Arbitration and Dispute Resolution
Law Journal 3, no. 1 (1992): 3). Abatement at common law, akin to Art. 50 of the
CISG, flows from a doctrine of partial failure of consideration based on an
allegation that fault of the plaintiff has led the to value of goods or services
supplied being less than was contracted for (ibid, citing Allen v. Cameron (1833)
1 Cr & M 832; Mondel v. Steel (1841) 8 M & W 858). If the primary claim is for
failure to pay for the goods, then a claim as to abatement is obviously a direct
substantive defence that would inevitably come within the four walls of the
initial arbitration agreement.
81 US domestic civil procedure at times even demands that a set-off be raised

lest this be lost for good.

82 Whether something constitutes a defence may also be relevant under the

applicable law of assignment. Under common law, for example, an assignee of a

debt generally takes subject to defences available to the debtor as against the
assignor where they arose prior to the debtor receiving a notice of the
assignment (S.R. Derham, Setoff, 2nd edn (Oxford: Clarendon Press, 1996), 3).
83 See, e.g., DIS Rules Art. 6.3(3).
84 Anibal Sabater, ‘When Arbitration Begins without a Seat’, Journal of
International Arbitration 27, no. 5 (2010): 444, n. 9.
85 Ibid., 450-451.
86 Ibid., 447.
87 See the cases referred to in Hans Dahlberg & Marie Öhrström, ‘Proper

Notification: A Crucial Element of Arbitral Proceedings’, Journal of International

Arbitration 27, no. 5 (2010): 539.
88 See ICC Rules 2012 Art. 3; UNCITRAL Rules 2010 Art. 2; UNCITRAL Model

Law Art. 3.
89 See the Decision of the Court of Appeal of Hamburg, in Pieter Sanders (ed.),

Yearbook of Commercial Arbitration Vol II (The Hague: Kluwer Law

International, 1977), 241. The laxer rules about ex parte communications of
non-neutral party-appointed arbitrators under American domestic ethical
principles are outside the scope of this book as they do not pertain to
international arbitration. See further AAA/ABA Code of Ethics.
90 Decision 10 (dated 13 February 2002) LCIA Court, reported in G. Nicholas & C

Partasides, ‘LCIA Court Decisions on Challenges to Arbitrators: A Proposal to

Publish’, Arbitration International 23, no. 1 (2007): 16.

Part II: The Process of an Arbitration,

Chapter 5: Selection, Challenge and
Change of Arbitrators
Jeff Waincymer,

5.1. Introduction

It has become a cliché to say that an arbitration is only as good as its arbitrators,
but it is nonetheless largely true. The arbitrators make key decisions not only in
the final award, but also in relation to numerous procedural elements of the
arbitration. For that reason, getting the composition of the tribunal right is of
utmost importance.

This chapter looks at the composition of the tribunal from beginning to end. It
starts with an outline of how to select arbitrators, which is a crucial aspect of all
arbitrations. It then considers an issue which arises only rarely: how those
arbitrators can be removed. From there, it considers how arbitrators can resign
and be replaced. Finally, it looks at the related issues of the arbitrators' fees,
liability and immunity to the extent that these may impact upon procedural
conduct.

5.2. General Procedures for Appointment

Almost all modern legal systems allow for parties to choose the arbitrators, or
choose a method for their selection. (1) As with so many areas of international
page "255" arbitration, consent and party autonomy have primacy. The main
ways in which arbitrators can be selected are: (2)

– Direct agreement of the parties.

– Modified inter-party processes, including lists systems.
– Delegated selection processes, including appointment by an arbitral
institution, a professional institution or trade association or a person such as
the Secretary-General of the PCA.
– Appointment of the chair by co-arbitrators.
– Appointment by a national court.

This is not an exhaustive list and the various methods can, of course, be varied
and combined under principles of party autonomy.

5.2.1. Agreement of the Parties

Direct choice of arbitrators by the parties is often said to be the preferable

method of selecting arbitrators, as it maximises party control over the running of
the case. (3) Party autonomy in selecting arbitrators is widely recognised in
international conventions and national legislation, (4) although it rarely occurs in
a pure form where each party agrees to each arbitrator. Where there is to be a
sole arbitrator, the parties would have to agree on a specific person or rely on a
default provision absent agreement. Where the panel is to have three arbitrators,
the most common method is for each party to nominate one arbitrator, with
those two arbitrators to then choose the chair. Hence this is a hybrid aspect of
party choice in that it is partly direct and partly indirect.

From a policy perspective, there are also fairness and efficiency issues with this
aspect of party autonomy. While it would almost be sacrilegious to question the
practice of party-appointed arbitrators, (5) economics would suggest that there
page "256" can be undesirable implications as the parties will to some degree
look for predisposition as opposed to mere expertise. Thus, in an investment
arbitration attacking a government's environmental measures as indirect
expropriation, there would be a strong incentive to examine past awards and
scholarly writings to determine the general attitude to this contentious question
in the field of investment protection. The same would be so if a claim was based
on good faith norms contrary to express contractual wording. This form of
strategic selection behaviour is a zero sum game if played ‘well’, as if each party
selects arbitrators predisposed to contrary positions, they will possibly cancel
each other out, in which case the essential decision will be that of the chair. The
justification of the expense of a three-person tribunal is then dubious. To raise a
problem is not to say that the option should not be followed. One must always
compare the alternatives. Requiring agreement on all three would be
problematic in many cases. Deferring to third-party appointors comes with its
own predispositions even if it is not as easy for a party to take strategic steps to
benefit unfairly from this.

It should also be noted that, although we tend to speak of agreement of ‘the

parties', in practice it is parties' legal counsel that tend to make the choice. (6) In
this context there is a particular duty on counsel as there would be a world of
difference between an informed choice in the best interests of the parties and
instead, a choice based on ignorance by counsel or based on counsel's own
interests. This is considered more generally in section 5.5 dealing with the
attributes to be looked for in a tribunal.

If parties have designated a procedure it must obviously be respected and

attention must also be given to cases where procedures are not fully followed. If
the tribunal is constituted in a manner contrary to the parties' agreement, and
there has been no waiver, the award might not be enforced. (7) At times one party
may be able to argue that arbitral consent was dependent on the proper
application of the relevant procedures, for example, if the method has been
selected in the arbitration clause. (8) It is important to have default processes,
otherwise arbitration agreements could too easily be frustrated. Some
arbitration clauses, and most institutional rules, (9) allow for an institution to
make the nomination as a fallback. In an ad hoc arbitration, it may be necessary
to rely on the lex arbitri, which normally allows appointment by a court or a
national arbitral institution (see section 5.2.3 below), although the UNCITRAL
Rules 2010 provides its own mechanism as noted below. Born notes that the
agreement on the arbitrators should be put in page "257" writing in order to
meet form requirements of international conventions and national legislation. (10)

Policy issues are more complex if a failure to appoint under a designated

procedure arises from mere inadvertence. One issue is whether a party, having
missed a deadline to nominate, may later be permitted to do so. In some cases, it
has been held that failure to comply with time limits results in losing the right to
nominate at a later date. Other courts have been more lenient, finding that
additional time periods should be granted. (11) This may depend on the
circumstances and the wording of arbitral rules. For example, the ICC Rules 2012
allow for a time extension for the Answer to the Request if the respondent's
arbitrator is nominated. (12) If the Respondent is generally challenging
jurisdiction in circumstances where it hopes the ICA may accept the argument
without the need for a tribunal, a failure to nominate an arbitrator under protest
may lead to an inability to do so later if the ICA defers to a tribunal under a strict
interpretation of the ICC Rules. English law allows for the practice that, if one
party fails to nominate its arbitrator, then the other party's candidate can serve
as a sole arbitrator. (13) This can, however, lead to problems in enforcing the
award in some countries. (14)

The scope of the parties' choice may be limited by their arbitration clause. For
example, the arbitral rules named in the arbitration clause may limit the choice
of arbitrators to those who are members of that institution's panel (15) or of a
particular profession. The choice of these rules will bind the parties even if this
list later seems inappropriately restrictive. (16) They can of course come to a
contrary agreement.

5.2.2. Limits on Party Agreement

In most systems, there are no limitations on who parties can choose as an

arbitrator. There are, however, a few exceptions to be noted.

page "258"

5.2.2.1. Nationality

In the past, many systems prohibited foreign nationals from acting as arbitrators,
but this has mostly been abolished. (17) Article 11 of the Model Law in fact
prohibits statutory discrimination based on the arbitrator's nationality. (18) Many
parties select a national from their country as their appointed arbitrator,
although in many instances there would be no inherent benefit (or detriment) in
doing so. (19) However, there remains a trend towards choosing ‘neutral’
nationalities to chair arbitration. Most rules suggest that either a sole arbitrator
or presiding arbitrator should not be from the same nationality as either of the
parties. (20) Article 9(5) of the ICC Rules 2012, for example, provide that a sole
arbitrator or chair should not be from the same country as one of the parties,
although it is permitted if appropriate and neither party objects. The ICC also
typically prefers to appoint a chair who is not of the same nationality as one of
the other arbitrators to ensure a ‘balance of nationalities’ on the panel. (21)
Likewise, Article 6(7) of the UNCITRAL Rules 2010 requires the appointing
authority to have regard to the ‘advisability’ of appointing an arbitrator who
does not have the same nationality as one of the parties.

Arbitral rules may also have to be looked at in the context of other legislative
norms such as human rights and non-discrimination provisions which might be
asserted to impact on the ambit of choice in selection. However, there can be no
presumption that they must apply to arbitration. In the case of Jivraj v Haswani,
the UK Supreme Court quashed a decision of the Court of Appeal which held that
arbitrators were ‘employees’ and thus subject to anti-discrimination laws about
religion. (22)

Born notes that, in practice, even if the arbitral rules do not require appointing a
sole arbitrator/chair from a neutral country, it is typically done anyway to avoid
any impression of bias. It is worth briefly noting that there is some debate on the
precise meaning of ‘neutral’ nationality. In most rules, it means that the
arbitrator does not share the nationality of one of the parties. But Pierre Lalive
has suggested that shared religion, ideology and social background should also
be considered. (23) Moreover, he argues that an arbitrator must have
‘international mindedness’, i.e., openness to other legal cultures and a
comparative approach, in order to be page "259" considered neutral. (24) If there
is any logic in a concern with common nationality, these other criteria must also
be worthy of consideration, although the sheer number of possible items of
concern highlights how illusory notions of neutrality can be.

5.2.2.2. Qualifications

Most systems do not specify any formal qualifications required to be an

arbitrator. There are, however, a number of exceptions. Indonesia is one
example: Article 22 of Law No. 30 of 1999 requires an arbitrator to be at least 35
years of age and have at least fifteen years experience in the field. Likewise in
China, while having no specific qualification requirements, arbitrators must be
‘fair and honest persons’ and have some form of experience or senior title in law,
trade or economics. (25) In a similar vein, Article 14(1) of the ICSID Convention
requires arbitrators to be ‘persons of high moral character and recognized
competence in the fields of law, commerce, industry or finance, who may be
relied upon to exercise independent judgment.’ Saudi Arabia is another
exception, where the arbitrator must be a Saudi national or ‘Muslim expatriate’.
(26) Born argues that, if applied in an arbitration, this amounts to a violation of

Articles II and V(1)(d) of the New York Convention. (27) In many States, a person
may not be an arbitrator if he or she has committed a serious crime or has been
declared bankrupt. (28) Likewise, it is not uncommon for sitting judges to be
prohibited from acting as arbitrators. (29) Some States also require arbitrators to
be natural persons; for example, France in domestic arbitrations. (30) It is unusual
for legal persons to be appointed as arbitrators, though this has occurred when
accounting firms have been made arbitrators in specialised accounting disputes.
(31)

5.2.2.3. Party Equality

A further limitation on the scope of party autonomy in selecting arbitrators is

that a manifestly inequitable agreement may be invalidated. In Belgium, for
example, an arbitration agreement is invalid if ‘it gives one of the parties thereto
a privileged page "260" position with regard to the appointment of the
arbitrators.’ (32) Likewise, common law courts may strike down arbitration
agreements that allow one party to appoint both co-arbitrators or to have its
arbitrator as the presiding arbitrator. (33) That would be the logical corollary of
mandatory due process norms such as Article 18 of the UNCITRAL Model Law. It
is easy to state the norm but more difficult to decide how it should apply in
practice, for example, if one party is given sole default selection rights from a
specified high quality independent pool.

5.2.2.4. Party-Agreed Limitations

The parties themselves may impose limitations on arbitrator selection. This may
be done directly in the contract or by virtue of the rules selected. (34) This often
happens in contracts for highly technical work such as energy, oil and gas,
construction and maritime contracts. (35) Thus a contract may provide that the
arbitrator is to be a ‘shipping man’. (36) Parties may even agree to name a specific
arbitrator in the event of a dispute. (37) It is widely accepted that it is too risky if a
specific person is selected. If they are unwilling or unable to take the mandate,
the entire agreement could then be undermined, again on the basis that it was
conditional on the named appointment.

A contract could conversely prohibit certain types of arbitrators. In some cases,

arbitration agreements have expressly prohibited the arbitrators being lawyers.
(38) Redfern and Hunter suggest that it is unwise to include strict qualification

requirements in the contract, as this reduces flexibility once the dispute arises.
The dispute may be very different from that contemplated in the arbitration
clause. (39) For similar reasons, it would be unwise for the lex arbitri to require
page "261" certain professional qualifications for being an arbitrator. (40) Most
other commentators seem to agree. (41) If those qualifications are truly pertinent
to the dispute when known, the parties can impose those requirements at that
stage.

Conversely, an agreement in the arbitration clause about qualifications can of

course be waived by parties' later conduct. (42) Thus in a German case, the fact
that both parties appointed arbitrators who were not members of a particular
chamber of commerce (which had been required by their contract) was held to
mean that the qualification requirement had been waived and the appointment
could not later be challenged. (43) Care needs to be taken in such circumstances
that the relevant supervisory court would accept the waiver. This may be
impacted on by writing requirements in the lex arbitri, whether the waiver does
not fall foul of such requirements and what law of waiver would apply.

5.2.3. Arbitral Statutes and Tribunal Appointments

Arbitral statutes allow for party autonomy but provide fallback methods as to
number of arbitrators and appointing authority. Different lex arbitri take
different approaches. UNCITRAL Model Law Article 10(2), for example, says that
the number of arbitrators shall be three if parties cannot agree. In the absence of
any procedural rules, many national laws also provide a default number of
arbitrators, typically one (44) or three. (45) Other national laws give discretion for a
national court to determine the number. (46)

5.2.4. Arbitral Institutions and Tribunal Appointments

Institutional rules typically allow for appointment of arbitrators by the

institution if the parties fail to reach agreement. (47) The ICC Rules, for example,
provide for appointment by the Court if the parties have failed to agree within
thirty days of the request for arbitration. (48) Even in an ad hoc arbitration,
institutions may act as page "262" an appointing authority. (49) Although this is an
added cost, the fee is typically small.

Different institutions adopt different methods for appointing arbitrators. Born

has analysed the advantages and disadvantages of the various approaches
adopted by institutions. The AAA first gives the parties an opportunity to agree
and, if this does not succeed after fifteen days, a formal list procedure is used.
That is, names are taken from a ‘National Roster’ and then supplied to the
parties. The parties can then cross off and rank the names accordingly. (50) The
list procedure has the advantage of maximising party involvement, but the
disadvantage of delay. The ICC tends to decide on a nationality and then request
proposals from one of its ‘National Committees’. The final authority for the
choice lies with the ICC Court of Arbitration and Secretariat. (51) To Born, the ICC's
system of National Committees is ‘cumbersome’ and may be more ‘political’ than
‘rational’ (although he asserts that it normally ‘works’). The LCIA is the most
time efficient, but has minimal party involvement and an apparent bias towards
English barristers and retired judges. (52) Despite these criticisms, Born opines
that the major institutions typically make good selections. (53) He notes that
empirical research suggests parties prefer list systems, (54) and concludes that a
list system is preferable. (55) The logic would be that the more the parties feel in
control, the happier they are.

As noted above, an institution may also have to step in if one party fails to
nominate its arbitrator (sometimes as a delaying tactic). (56) Where this occurs,
the ICC endeavours to nominate the type of arbitrator that the defaulting party
would have nominated. In practice, this tends to mean appointing an arbitrator
of the same nationality as the party. This can be problematic if the party comes
from a country with no available arbitrator. (57) In some cases this could even be
contrary to wishes, for example, if a foreign multinational has established a local
investment page "263" subsidiary in a host country and would prefer its own
national to that of the host State.

Institutional rules also have default provisions on the number of arbitrators if

the parties cannot agree. A respondent, for example, may unreasonably request
three arbitrators, knowing that this can delay the process. Likewise, a claimant
may request three arbitrators in order to put economic pressure on the
respondent. (58) Most institutional rules give the institution discretion. Many,
including the ICC, LCIA, SIAC and Swiss Rules 2012 provide for the appointment
of a sole arbitrator, unless the circumstances justify a three-member panel. (59)
The size of the dispute will obviously be a major factor in using this discretion.
(60) Of course, there is no necessary correlation between value and complexity.
The logic would instead be that all cases might benefit from a greater number of
high quality arbitrators but only high value cases can justify the transaction
costs. The ICC, for example, usually appoints a sole arbitrator where the amount
in dispute is less than USD 5 million and typically appoints three arbitrators
where the amount in dispute exceeds USD 30 million. (61) A range of other factors
are of course relevant. Counsel with a preference for one or three arbitrators
should draw the institution's attention to the specific reasons which justify their
preference, as it can be difficult for the institution to appreciate all the details of
the case. (62)

Newmark advocates a greater role for institutions in selecting arbitrators. The

key advantages suggested are greater allegiance by the tribunal to the
institution, including deference to its control and influence over the conduct of
the case and, secondly, to ensure that no member of the tribunal owes allegiance
to a party, thus removing any fear of bias. (63) Deference could be a problem if the
institution itself is behaving in a problematic way and there is a question of how
much analysis the institution should undertake to minimise predisposition.

5.2.5. A List System

Parties seeking agreement on who to select as an arbitrator or institutions

seeking party-input in making an appointment may wish to use a ‘list system’.
Party autonomy can still be maximised via the institution providing a list for the
parties to then rank and veto any unacceptable candidates. As noted above, while
many would be reluctant to give up the right to appointing their own arbitrator,
policy page "264" considerations might suggest that those of this view forget the
corollary of giving their opponent an equal right.

Under this system, a version of which is adopted by various institutions as above,

each party could, for example, compile a list of three to four arbitrators. These
are exchanged and the parties seek to reach agreement. Redfern and Hunter note
that this has several advantages. First, the parties may name the same people on
their respective lists, making agreement easy. Secondly, even if this does not
happen, it focuses the parties’ thinking. (64) However, a number of criticisms are
made of the list system. First, the likelihood of common names happening is
usually remote. (65) Secondly, it can be a slow process, (66) with parties waiting for
each other's responses. Thirdly, parties' peremptory rejections, often based on
nothing more than vague suspicion of the other side's suggestions, can effectively
disqualify suitable candidates. (67)

A variation on this, used by some institutions, is for the institution to send each
party the same list. The parties may then delete any arbitrator to which they
object and list their order of preference for the others. To make this workable,
there would need to be a limit on challenge rights. However, the ingenuity of
procedural challenges is virtually limitless and an institution might still face an
allegation on the basis that due process requires more than the express number
of vetoes because of the nature of the list provided. This is not to say that such
challenges should be accepted, but instead, to make the point that procedural
rules cannot easily be drafted without fear of challenge when mandatory due
process norms sit over any model.

5.2.6. Appointment of Chair by Co-arbitrators

Most rule systems allow the two party-appointed arbitrators to select a chair
under a three-person tribunal. In 55% of ICC arbitrations, the co-arbitrators are
able to reach an agreement on the chair. (68) If the co-arbitrators are unable to
agree on a chair, it will fall to the institution or national court to do so. (69)

Aksen advises co-arbitrators to contact each other immediately after their

appointment to begin the process of selecting a chair (assuming the parties or
rules give them that role). (70) The advantages of beginning the process early are
page "265" that the deadline for tribunal constitution will more likely be met and
the co-arbitrators can begin to develop a rapport. (71) Aksen firmly believes that,
although there is usually a default provision allowing an institution to appoint a
chair, the arbitration tends to run more smoothly if the chair has been selected
by his or her co-arbitrators. He notes, with regret, that less experienced and
State-appointed arbitrators will often think deferring to the institution is ‘the
safer course of action’. (72) Aksen recommends the following process when
selecting a chair with a co-arbitrator:

1. Co-arbitrators should try to agree first on the nationality of the chairperson.

Often, it will be useful if the chair is from the seat of the arbitration.
2. They should then exchange a list of three to five names with attached
résumés.
3. These can then be narrowed down to a list of around three.
4. These three prospective chairs should then be contacted simultaneously via
email with a list of questions about conflicts and time availability.
5. Finally, a chair can be chosen from those who meet the conflict and time
requirements. (73)

When the ICC is called upon to select a chair, it most typically looks to the seat of
the arbitration and tends to appoint a chair from that seat. The theory behind
this is that a party who has agreed to that venue as a seat cannot reasonably
object to the appointment of a chair from that seat. Of course, this must be
balanced against concerns about political neutrality, expertise and the need to
have a balance on the tribunal. (74) Thus, the ICC will typically not appoint a chair
who is of the same nationality as one of the parties. (75) In many cases, it may not
be appropriate to try to have a chair from the seat, particularly if other expertise
is more significant.

It was noted above that there will be a natural tendency for each party to try and
select predisposed arbitrators. If only one does so, this is unfair. If both do so, it
cancels out any advantage. A related question arises as to what a party-
appointed arbitrator should do if the appointing party directs them as to
selection of the chairperson. This relates both to ethics of parties and the
appointed arbitrators. On the one hand, a consent-based paradigm might suggest
that the party has as much right to indicate its preferences about a suitable chair
as it does to select its own arbitral representative. A contrary view is that once
the party-appointed arbitrators are selected, they should simply confer
independently to identify the most able and independent chairperson
appropriate for the nature and size of the dispute.

page "266"

Most institutional rules are silent on this matter. Aksen believes that, in selecting
a chair, the parties who appointed each arbitrator should not be consulted. (76) He
notes that this may be controversial, but he is convinced that the more that
arbitrators consult with their appointing parties, the less likely there will be
agreement between the arbitrators. Consultation, he concludes, jeopardises
independence ‘without producing any compensating advantage’. (77) If a co-
arbitrator is insisting on party consultation, Aksen suggests that the arbitrators
should remove themselves from the process of chair selection and let the parties
negotiate directly, which, he says, works surprisingly often. The parties should
not be able to ‘veto’ agreement between the co-arbitrators on a chair. (78) If they
sought to do so, they would be agreeing on a change of tribunal which would
have other implications. Nor should the parties be able to interview a
prospective chair. (79) This should not be necessary as the supposed advantages
of the parties' particular criteria should already have been covered with the
information provided to the party arbitrators on their selection.

Conversely, some believe that the first opportunity to select a chairperson should
lie with the parties, and that the arbitrators or institution should only choose as a
last resort. (80) Born notes that the practice of conferring with nominating parties
in selecting a chair is ‘widely accepted’. (81) He notes that Article 5(2) of the IBA
Ethics Guidelines expressly permit it. He says that, provided that it happens with
both parties' knowledge, it is typically very constructive. Without more
information about the dispute (which can only come from the parties), it may be
difficult to choose an appropriate chair. The parties' involvement also increases
the chance that the parties will have confidence in the chair.

5.2.7. Professional Institution or Trade Association

Some parties prefer that an industry ‘club’ resolve their dispute, ensuring that
experienced practitioners are the arbitrators. This may speed up the process,
because technical concepts do not have to be explained. (82) This may be most
appropriate where the dispute is likely to be highly technical in an area where
the institution has expertise and legal-procedural points are less important. (83)
page "267" A contrary suggestion is that, an arbitral institution is commonly a
preferable appointing authority, as they have daily involvement in arbitration
and a better understanding of who would be an appropriate arbitrator. (84) Again
the problem is that the parties have to select a mechanism when the nature of
the dispute is unknown unless the agreement is a compromis, hence it may be
preferable to defer to an arbitral institution and trust it to select an industry
expert where appropriate.

5.2.8. A National Court

Where no agreement on appointing the arbitrators can be reached, parties may
have to rely on the lex arbitri, which traditionally allows for appointment by a
national court. (85) This is typically a fallback method where no other way of
appointing the arbitrators is available, such as in ad hoc arbitrations or when the
appointing authority otherwise cannot or will not act. (86) Many national laws
have now shifted to default appointment by a national arbitration institution
instead of a court with the aim of more informed choices being made. (87) The
relevant national court, if applicable, is almost always that of the seat, as
provided for in the lex arbitri, although there are a few notable exceptions in
which other courts may be consulted. Born warns, however, that it should only
be in very rare circumstances that a court should appoint an arbitrator in a
dispute whose seat is not in the same jurisdiction as the court. (88) He gives four
reasons. First, there is a risk of conflicting judicial orders from multiple courts if
a party also seeks judicial assistance in the courts of the seat of the arbitration.
Second, courts should normally appoint an arbitrator familiar with the law of the
seat. This will be hard to do for a court which is not in the seat of the arbitration.
Third, the foreign arbitral seat may have qualification or capacity limits or
independence requirements which the court is not aware of. Finally,
international arbitration aims to get disputes out of national courts. Judicial
assistance is acceptable where there are no other options, but a court should not
be involved when it is not absolutely necessary. (89)

page "268"

A national court may also be called upon where one party fails to nominate its
arbitrator and no institutional mechanism has been provided for. (90) This is
typically due to the respondent's failure to nominate, and leaves the claimant
with the choice between abandoning its claim or asking the court to appoint the
respondent's arbitrator. (91) In practice, national courts, which may have little
expertise in appointment of arbitrators, at times allowing the applicant party to
present possible candidates. The court may then select one of these candidates.
(92) The consent of the other party is, at this point, irrelevant. (93) A further

instance in which court intervention may be required is where the parties

agreement is ‘indefinite or internally inconsistent’. (94)

Disadvantages of court intervention include likely delays and uncertainty. (95) It

would be particularly important to know the normal time it would take for the
court to make an appointment. There are major differences between
jurisdictions on such issues. Further, many national judges will have limited
expertise and experience in arbitration, (96) as well as the possibility of
‘unintended parochialism’. (97) There is also the risk of compromising
confidentiality by going to a national court. (98)

Although delay is common, the lex arbitri typically seeks to make the
appointment as quickly as possible. For example, Article 11(5) of the Model Law
provides for no right of appeal to the selection of the arbitrator, although
annulment challenges are still possible. Problems can also arise if the seat of the
arbitration is not specified and there is a dispute about which court is the
appropriate one to go to for assistance in appointment. Some national courts
have a discretionary power to appoint arbitrators in such circumstances. (99)
Born says that this is one of the rare cases in which a national court which is not
in the seat of the arbitration would be justified in appointing an arbitrator. (100)
This occurred in the case of National page "269"Iranian Oil Co v. State of Israel.
(101) A problem may still arise if more than one court purported to have this

power.

If a court is called upon to select an arbitrator, what criteria does it use? National
legislation typically provides only limited guidance. (102) The Model Law, for
example, simply says that courts should have ‘due regard’ to any qualifications
required, and the advisability of having an arbitrator who does not have the
same nationality as one of the parties. (103)

5.2.9. UNCITRAL Rules: Designation by the Secretary-General of the PCA

Under the UNCITRAL Rules 2010, the parties may ask the Secretary-General of
the PCA to designate an appointing authority. The PCA is also the fallback
appointing authority if the parties are otherwise unable to agree on one, (104) or
the designated appointing authority fails to act. (105) The PCA may itself act as the
appointing authority if the parties request it to do so. Where it so acts, the PCA
typically uses a list system, and does not limit itself to any particular panel to
select candidates. (106)

If the parties are unable to agree on the number of arbitrators, the UNCITRAL
Rules provide for three arbitrators to be appointed. (107) The UNCITRAL Rules
also give the appointing authority the important power to deprive a party of its
right of appointment in ‘exceptional circumstances’. (108) No guidance is provided
as to what those circumstances might be, but Kee suggests that they could
include regular failure to agree on matters under Articles 8, 9 and 10 of the
Rules, spurious arbitrator challenges, and generally the use of delaying tactics.
(109)

5.3. Procedure for Appointment In Multi-Party Arbitrations

A large number of arbitrations involve more than two parties. In ICC arbitrations,
for example, over 30% have multiple parties. (110) A problem arises where
multiple page "270" parties on one side (e.g., two respondents), do not have
identical interests and cannot agree to appoint one arbitrator between them. An
example given by Craig, Park and Paulsson is the owner of a construction project
suing both its contractor and a sub-contractor, who may have conflicting
interests in any given case. (111) There are both practical and due process
considerations in such circumstances.

The seminal case highlighting the challenges in this area is the Dutco decision.
(112) In Dutco, a contract to construct a cement plant in Oman was entered into by

three parties: BKMI, Siemens and Dutco. The arbitration clause provided for ICC
arbitration with three arbitrators. As usual, this allowed for each side to
nominate one arbitrator, with each of those two arbitrators to nominate the
chair. A dispute arose in which Dutco initiated proceedings against the other two
parties, making distinct claims against them. Dutco nominated its arbitrator, but
the two respondents did not agree to appoint one arbitrator jointly, claiming
they had different interests. The ICC insisted that the respondents should
nominate one arbitrator jointly, which they eventually did ‘under protest’. The
respondents later sought to have the award set aside, claiming that the principle
of equal treatment of the parties had been violated by not allowing each of them
to appoint their arbitrator. They claimed that this meant that the tribunal was
improperly constituted and that enforcing the decision would be contrary to
international public policy. (113) The Paris Court of Appeal rejected their claim,
holding that the tribunal was constituted in a manner consistent with the
common intention of the parties in agreeing to an arbitration clause in a multi-
party contract and that there was no other issue of international public policy.
(114) The Cour de Cassation overturned that decision. It did not disagree with the

Court of Appeal's construction of the arbitration agreement, but it did find that
there was a violation of the parties' right to equality in the constitution of the
tribunal. It held that this right could not be waived in an arbitration agreement
that had been concluded before any dispute had arisen.

The decision gave rise to a huge body of commentary. Poudret and Besson state
that it is indisputable that the manner of the proceeding violated the principle of
equal treatment of the parties. (115) But although submitting to a multi-party
arbitration does not waive one's right to appointment, Poudret and Besson state
that it should not be concluded that this right cannot be waived in some
circumstances. To the authors, the parties can choose to waive the right (because
of party autonomy), although the policy issues are complicated by the principle
that mandatory due process rights are not typically waivable and problems
therefore arise where jurisdiction is dependent on such waiver. Hanotiau has
noted that the page "271" facts of Dutco are fairly unusual: the claimant must
have summoned at least two defendants with divergent interests and the
arbitration clause must provide for three arbitrators. (116) Poudret and Besson
note that nothing problematic arises when the case is to be heard by a sole
arbitrator. The process is simple: the parties can agree on an arbitrator and in
the absence of agreement, the arbitrator is appointed by the arbitration
institution or a court. (117) It is not clear that the case would be followed outside
of France. (118) In a situation where the multiple defendants operate as a group or
a consortium, a recent ad hoc tribunal held that the multiple defendants only had
the right to appoint one arbitrator between them. (119) Nevertheless, the
conceptual challenge is highly tenable and at least requires a practical response
where possible. This is discussed in the following section.

5.3.1. Multi-party Appointment Procedure Post-Dutco

Dutco has been an influential decision and has led to many institutions changing
their rules. (120) Institutional rules will often now provide that, in Dutco situations
where multiple claimants or respondents cannot agree on a joint nomination, the
institution is to appoint all of the arbitrators. (121) This is discussed further in
section 7.5.

5.4. Selecting the Number of Arbitrators: Pros and Cons of Single- and Multi-
Member Tribunals

Before selecting specific arbitrators, the number of arbitrators to sit on the panel
needs to be determined. The parties are free to choose however many they
believe is appropriate. (122) In deciding on how many arbitrators should
constitute the page "272" tribunal, the choice in practice is between one or three.
(123) Obviously a two or four-member tribunal is undesirable because of the risk

of disagreement between them. Some countries even have a prohibition on an

even number of arbitrators. (124) Others have provisions which convert an even-
numbered tribunal into an odd-numbered tribunal. (125) Likewise, a tribunal of
five or more is unlikely to add value, and tends to be used only in State-to-State
arbitrations or in multi-party situations. (126) As noted in sections 5.2.3 and 5.2.4
above, if the parties are unable to agree on the number, the administering
institution may decide on their behalf, and lex arbitri and rules will typically have
default numbers.

It is difficult to see a strong preference in the statistics for either single or multi-
person tribunals, (127) but conventional wisdom has it that, where the value of the
dispute is high enough, three arbitrators are better than one. (128) That said, the
suitability of each approach will depend in large part on the nature of the parties
and the dispute. The respective merits are also impacted upon by the selection
method. As noted above, if multi-person tribunals simply encourage search for
predisposition, the value is diminished. There is a big difference between a three-
member independent tribunal appointed by an institution and the more typical
tribunal where each party appoints their own arbitrator with the party-
appointed arbitrators mutually selecting a Chair.

One of the difficulties in deciding on the right number of tribunal members is

that this is often done at the drafting stage in the contract when the nature of the
particular dispute and the amount involved is not known. (129) On the one hand, it
is generally desirable to decide the key procedural questions at the outset, given
that at the time of a dispute parties will often fail to agree even on procedures
that others would see as being in their mutual best interest. On the other hand,
selecting page "273" tribunal size at the outset could give rise to mismatches in
many cases. This applies whether there is a direct choice of the number or choice
of rules or lex arbitri that have a default number. Even if the contract has selected
three arbitrators, it is appropriate to consider when a dispute arises whether it
will be preferable to seek agreement for a sole arbitrator. It is also difficult to set
a criterion such as amount in dispute that can be used to automatically
determine the number of arbitrators. No figure is ideal and complexity and value
may point in different directions.

The obvious advantage of a sole arbitrator is ‘speed and economy’: it is easier to

arrange meetings and hearings; the fees are lower; deliberation time is likely to
be shorter, as it does not involve exchange of views of a panel that itself often has
coordination problems. (130) This may explain why the default position of many
institutional rules is to appoint a sole arbitrator. (131) Disadvantages may include:
an increased risk of a ‘rogue decision’ or undue predisposition because the
arbitrator does not have the benefit of two other views to balance his or her own,
and the risk that the parties will be unable to agree on who to select as a sole
arbitrator. (132) In cases where general attitudes of arbitrators will be crucial,
such as the extent to which indirect expropriation norms should interfere with
governmental social welfare measures, it may also be hard for the parties or an
institution to make the selection, knowing that the very choice affects the
outcome. As Born has put it, ‘the task of a sole arbitrator is a lonely one and can
occasionally lead to inadequately-considered results’. (133)

Advantages of a panel of three arbitrators include: the ability of each party to

nominate its own arbitrator (meaning that a party's unique concerns pertaining
to its legal culture and tradition can be considered); (134) the balancing and
exchange of different views in deliberation (and this testing of views means the
‘quality of justice’ is likely to be improved); a wider range of expertise; (135) and
less likelihood of a rogue decision. (136) It was for these reasons that UNCITRAL
decided to provide for a default panel of three in the Model Law, a position
retained in the revised page "274" UNCITRAL Rules. (137) The main disadvantage
is that a three-member panel is more expensive. The cost means that a three-
member panel is probably inappropriate where the amount in dispute is small.
(138) There is also likely to be delay given the need to find times for hearings and

deliberations that fit the schedules of all. This is exacerbated where the
arbitrators are in high demand, meaning that timeliness and quality may need to
be traded off against each other.

Another important aspect of multi-person tribunals that will impact upon the
cost/benefit matrix is the quality of the chair. A highly effective chair will bring
out the benefits flowing from a robust exchange of a range of views, but an
overbearing chair can negate the very value of a multi-person tribunal and a
weak chair can allow disruptive counsel or a rogue party-appointed arbitrator to
continually frustrate the process.

If the parties make no choice as to the number of arbitrators, the parties might
select an institution that has experience and has established appropriate
guidelines in determining an optimum number. (139) Some might choose one that
utilises a case-by-case discretion. Although parties should carefully consider the
number of arbitrators, one commentator has recently made the observation that,
in the vast majority of cases, arbitrations are properly conducted and give solid
awards regardless of the size of the tribunal. (140) To the extent that this is
accurate, it at least suggests that there should be no automaticity in selecting
multi-member tribunals even in larger value cases. Parties should always
remember that it is their choice and an informed choice at any stage is to be
preferred.

5.5. Who to Choose: Qualities in an Arbitrator

Once the number of arbitrators and a procedure for appointment have been
decided, the parties or relevant body will have to choose the particular
individual(s) to act as arbitrators. It is constantly repeated that, where this
choice lies with the parties, it is the most important decision the parties make in
arbitration. (141) There are some obvious factors to look for in an arbitrator.
Availability, integrity, intelligence, collegiality, past arbitration experience and
knowledge relevant to the page "275" subject matter of the dispute are all vital.
(142) Likewise, all arbitrators need to be able to communicate well, command

respect and be courteous and tactful. (143) Other factors may include cultural
sensitivity, ability to understand technical evidence, efficient attitudes to
procedural matters including document production, active or passive approach
during hearings, diplomacy, and business acumen and commercial disposition.
There are also intellectual issues of previous positions such as attitudes to
probity of evidence, approach to choice of applicable law, drawing of adverse
inferences, attitudes to interpretation of treaties, statutes, rules and contracts
and the like. Substantive views may also be relevant. For example, where
investment arbitration is concerned, an arbitrator's attitude to general
exempting provisions from non-discrimination or anti-expropriation norms may
be the key to the outcome of the case.

In a survey of arbitration practitioners the following were stated as key factors

affecting choice of arbitrator:

1. Reputation.
2. Expertise in the subject matter.
3. Recommendations from external counsel.
4. Language proficiency.
5. Knowledge of the law applicable to the dispute. (144)

Lord Hacking has also pointed out that ‘[a] lthough good intellect is essential, the
role of an arbitrator is also a practical one’. (145) For a sole or a presiding
arbitrator, there is a need to manage the process. It is important to understand
that a fair and efficient process is arguably as important as a just outcome.
Ideally, both should be optimised. Because so many procedural issues are
contested, often with threats against enforceability, an effective arbitrator needs
to be able to steer the process through such trials. For a presiding arbitrator, this
can involve the delicate task of allowing all views to be heard as to procedural
and substantive matters, but moving the process forward in a timely manner
when consensus is not possible or likely.

Obviously different people will have differing attributes. Different cases may
require differing priorities where not all attributes can be optimised. In that
context, arbitrators do not necessarily need to have expertise in the particular
applicable law or in the subject matter of the dispute as it is essentially an
adjudicative function. page "276" Because there are so many relevant factors,
competence will vary and some arbitrators will be better suited to some parts of
a case than others. As Born has noted, there is no perfect arbitrator for all
disputes: different disputes will better suit different arbitrators. (146) One
difficulty in selecting an arbitrator is that, as most arbitrations are private,
parties and counsel with little experience in arbitration may find it impossible to
know whether a candidate possesses the requisite qualities. (147) This can, in part,
be overcome by consulting with experienced arbitration practitioners and,
within certain limits, interviewing candidates (discussed in section 5.6).
In addition to the above-mentioned attributes, some others are more
contentious. Some of the potentially debateable aspects of who to select as an
arbitrator are discussed below.

5.5.1. Nationality

There are two aspects of nationality, the first being the possibility of choosing an
arbitrator with the same nationality as a party, the second being the choice of a
neutral nationality where a sole arbitrator or tribunal chair is concerned. A
related question to the first is whether to choose an arbitrator with the same
nationality as the applicable law where that is known.

As to the parties' own selections, many parties have a natural inclination to

choose an arbitrator who shares their nationality, or at least comes from a
similar legal system. This concern is reflected in many institutional rule systems
(see section 5.2.4). While this preference for the known is understandable, a few
qualifying remarks can be made. Lowenfeld has commented that, in his
experience, nationality plays a much smaller role in arbitrators' conduct and
deliberations than is commonly believed. (148) He adds that, in practice, counsel
rightly tend to choose arbitrators based on their skills and reputation, rather
than their nationality. (149) Where neutrality is concerned, this feature appears
intuitively desirable but insistence on supposedly neutral nationality can lead to
undesirable results, such as where a sole arbitrator has no familiarity with the
law applicable to the merits of the dispute (as this is often the law of one of the
parties). (150) Further, it is by no means a guarantee of independence. It does,
however, create an appearance of independence. (151)

page "277"

5.5.2. Qualifications, Expertise and Experience

Parties may wish to consider whether their arbitrators should have specific
qualifications, legal or otherwise. Any agreement on this in the arbitration clause
will be binding. This can be problematic if there is a wide range of potential
disputes, not all calling for these attributes. As discussed in section 5.2.4, the lex
arbitri may impose certain qualification requirements, but these are increasingly
rare.

In appointing the chair or a sole arbitrator, past experience as an arbitrator is

particularly important, as he or she essentially takes control of the proceedings.
(152) However, if this is slavishly followed there can never be additions to the pool

of potential chairs. There has to be a first time for everything. Institutions have
particular challenges in this regard, trying to meet the needs of the parties while
still having subordinate systemic concerns to support adequately trained
arbitrators over the longer term.

As to expertise, the ICC typically chooses a lawyer as chair on the basis that
procedural issues inevitably arise in any arbitration. (153) In a three-member
panel, it is by no means necessary to limit the selection of all arbitrators to
lawyers. Indeed, this might be excessive or, as one commentator has put it,
‘having a belt and suspenders'. (154) For many disputes, it will be far more
sensible to ensure that at least one member of the panel has expertise in the
substantive matters of the dispute, such as experience in the construction,
mining or shipping industry. This is particularly important in highly technical
disputes. Even where experts give evidence, such an arbitrator can clarify points
when the experts are not present and ensure that evidence is led on all the
pertinent questions. (155) However, in such circumstances, the chair needs to
ensure that there is still scope for debate and discussion and not any de facto and
a priori delegation to each tribunal expert of matters within their own field of
expertise.

5.5.3. Language

An arbitration clause may contain requirements that arbitrators be fluent in a

particular language. (156) This may be regarded as inherent in the arbitration
agreement's specification of the language of the arbitration. That said, it might be
possible for an interpreter to be used. (157)

page "278"

While arbitrators need to be competent in the language of arbitration, if it is not

their first language there is an issue as to how to determine whether they are
sufficiently proficient. This is not directly tested. Care needs to be taken to
ensure that the expertise is sufficient to deal with complex documents and
testimony in foreign languages and issues of legal interpretation of foreign
language texts. A related issue is that in addition to having a working knowledge
of the language of the arbitration itself, it is preferable to also know the language
of the underlying documentary evidence where this is crucial (which may be
different from the language of the arbitration). (158)

5.5.4. Availability

It is important to consider the likely future availability of an arbitrator. Parties

should avoid the trap of ignoring availability concerns in pursuit of a star
arbitrator. The world's smartest, most commercially-minded arbitrator is of little
use if he or she does not have the time to read the parties' submissions.

However, discerning availability is difficult at the outset for two reasons. First,
the arbitrator will not know enough about the intricacies of the case. Secondly,
the arbitrator will not know what future work requests are made for their
services within the case. This relates to substantive issues but also a host of
permutations as to procedural matters depending on future agreement or
contest between the parties. Even though so much is unknown at the outset, an
ethical arbitrator will be able to give a reasonable indication of their normal
practice in being able to engage in the work, in particular when seeking to set
aside a lengthy time for hearings. They should also be open about their current
time constraints.
5.5.5. Impartiality, Independence and Predisposition to One's Case

As will be seen below, one limitation on the parties' choice of arbitrators is that
the arbitrator must be impartial and independent. (159) A distinction is sometimes
drawn between ‘neutrality’ and independence/impartiality. (160) That is, a party
may choose an arbitrator who shares its nationality, or is from a similar socio-
political system. (161) page "279" This is different from the more stringent
requirements of impartiality and independence. (162)

As noted above however, as cases become larger, there is a greater incentive to

try and select arbitrators who are not only independent and expert but who also
may be predisposed to a particular outcome. Martin Hunter has famously said
that he looks for ‘the maximum predisposition towards my client, but with the
minimum appearance of bias.’ (163) It is accepted that there is no in principle
problem with selecting an arbitrator based on broad predispositions. There is
however a difference between the ethical right of counsel to adopt this approach,
and the fairness and efficiency of such a tendency. As noted, if only one party
does so, this is unfair; if both do so, it is ineffective and counterbalancing
behaviour. The most extreme form of predisposition is an arbitrator that will
always find for the appointing party. One should not pretend that this is not a
potential issue. Besides the ethical issues, this can also be a basis for challenge as
to impartiality. This is discussed further in section 5.9.6.6.

Given the common view that selecting based on predictions of intellectual

predisposition is ethical and that one' s client will be disadvantaged if only the
other side engages in such analysis, some comments are appropriate as to
process. What exactly should parties look for? Simply putting the arbitrator's
name into a search engine is surprisingly informative. Counsel can do research
on prospective arbitrators' publications to see what view they take on key areas
and also look at their reported awards. The latter can show procedural as well as
substantive preferences. Some substantive issues may be particularly clear-cut,
such as whether good faith claims contrary to express contractual provisions are
valued. Counsel may also get a sense of whether an arbitrator is a ‘broad picture’
or a ‘black letter’ lawyer. The former tends to be more business-minded and
equity-oriented, favouring substance over form. The latter might insist on a
stricter interpretation of the relevant principles. This is something of a
psychological distinction that exists among adjudicators in all jurisdictions. (164)
There can be legitimate differences in view as to what is optimal and these
differences underlie other debates in arbitration such as between implied
consent and strict jurisdictional approaches. Where strategic selection is
concerned, counsel can make an early assessment of their case and decide which
of these approaches best supports their arguments. However, a word of warning
should again be made. The aim is not to simply have one arbitrator favourable to
one's case on the tribunal. That on its own is useless in a three-member page
"280" tribunal. The aim is to ultimately have at least a majority. (165) There are
two aspects to this, the first being the likely impact if an unduly predisposed
arbitrator is selected and, second, the ability of the selected arbitrator to
influence chair selection and influence deliberations. As to the former, while
parties are free to appoint an arbitrator who they believe will be sympathetic to
their case, practitioners seem to agree that it is often unhelpful to appoint
someone obviously predisposed one way or the other. This is because that
arbitrator loses credibility with the other members of the tribunal. (166) Other
arbitrators may overcompensate. The predisposed arbitrator is unlikely to be
influential in deliberations. As Reed and Yanos have put it, ‘it is wiser to appoint
a truly independent party-appointed arbitrator who will have an influential voice
on the tribunal.’ (167)

Parties may also wish to consider whether their arbitrator is likely to work well
with the other arbitrators, especially the chair. (168) In a negative sense, if a
personality clash arises between arbitrators, there is a risk of delay and
increased costs. (169) In a positive sense, the greater the collegiality, the more
influential each arbitrator may prove to be.

5.5.6. Transparency and Performance Indicators

Some argue that there should be published data and evaluations of arbitrators to
better inform choice. The aim is laudatory but the problems of adequate data and
unintended consequences suggest that any organisation attempting to develop
such a system should proceed with caution.

One suggestion is that institutions might invite arbitrators to include in their CVs
an indication of the length of arbitrations they are involved in and the number of
pending matters they have. (170) It is certainly desirable for parties to have a clear
page "281" understanding of the available time that a prospective arbitrator has
but there may be problems with a simple statement of the length of cases.
Obviously the length depends upon the issues involved and in many cases the
parties' own choices as to the number of witnesses they wish to call, whether
they wish for bifurcated proceedings and whether they wish for proceedings to
be held in abeyance while settlement negotiations ensue. Hence broad
conclusions from simplistic data may be problematic. While time limits and
efficiency are certainly desirable, there is no direct correlation between
shortness per se and fairness for all concerned.

Even more contentious is the question of feedback on arbitrators. Arguments in

favour would be that accountability and feedback can only help encourage high
standards and proper preparation and help future appointment decisions.
However, it is hard to assess validly, with the loser obviously having an incentive
to complain. Certainly a loser who gives a glowing testimonial is providing very
believable testimony. Another problem is that such feedback forms cannot be
tested by institutions and arbitrators would be justifiably fearful of losing work
after adverse comments even where these would not have been justified.
Importantly, that might be a disincentive to being fearless, e.g., encouraging
arbitrators to be too polite to a guerrilla counsel, or refraining from using costs
awards in response to tardy behaviour. In a case where an arbitrator thinks both
parties have behaved badly, the downside is further exacerbated. If there were to
be published reports on arbitrator performance, one can even imagine threats of
defamation proceedings alleging that the data was obtained from malevolent
parties. There are particular problems with small data sets and with any system
that calls for evaluation after the result is known as opposed to before, the latter
being the accepted optimal where analogous teacher evaluation is concerned.

5.6. Interviewing Prospective Arbitrators

5.6.1. Advantages and Disadvantages of Interviews

Parties may attempt to assess a candidate's availability and suitability by way of

an interview. As Lowenfeld has put it, it may be desirable to interview someone
unknown to determine whether they were ‘quick or dull, realistic or dreamy, a
scholar mired in books or someone who understood or could be made to
understand the ways of business.’ (171) It also may be the best way for parties to
establish whether any conflicts exist. (172) Further, the client might find it
reassuring to see and hear the potential arbitrator in person and confirm a
recommendation of counsel. (173) page "282" For these reasons, many counsel
now see an interview as part of their ‘due diligence’ in preparing for an
arbitration. (174)

However, it is problematic to put a prospective nominee in a position akin to

applying for a job. It entails a danger of a strong financial incentive for the
prospective arbitrator not to appear unfavourable to key aspects of the case. Any
wide-ranging or social interview on an ex parte basis also risks creating an
appearance of bias. (175) Advocates should be aware that many English and
continental European arbitrators will refuse to participate in interviews, seeing
them as demeaning or at least compromising the appearance of impartiality. (176)
Gerald Aksen is also critical of interviewing. (177) In his experience, the parties'
reasonable desire to familiarise themselves with the interviewee too often
descends into trying to convince him or her of the merits of the case. Despite this,
the emerging consensus seems to be that interviews are acceptable in most
circumstances, provided that they are subject to strict limitations. (178) This
consensus is evidenced by the IBA Guidelines on Conflicts of Interest and CIArb
Guidelines on Interviewing Prospective Arbitrators. (179)

5.6.2. Advisable Limitations for Interviews

Essentially, the following balance needs to be struck: parties are entitled to find
out about an arbitrator's availability and qualifications, fee structure and
approach to efficiency issues, but they should not be allowed to inquire about the
arbitrator's page "283" view of the merits of the case or otherwise create an
impression of bias. (180) As Gerald Aksen has phrased it, the interview is aimed at
making parties comfortable with him or her personally and with qualifications,
but not with the prospective arbitrator's views. (181) With this balancing act in
mind, commentators have suggested a number of parameters and limitations for
interviews. (182) The following guidelines now seem generally accepted:

1. No discussion of the specifics or the merits of the case should take place. (183)
This appears essential to ensure that the interview does not become an
expedition in assessing whether the arbitrator is sympathetic to the party's
case. Even questions about legal familiarity can too easily try and tease out
 such information.
2. Discussion of the case should be limited to what is necessary to assess
availability and competence. This includes names of the parties, likely third
parties, the general nature of the dispute, the applicable law and procedural
rules, the language of the arbitration, its seat and likely length. (184)
3. Reed and Yanos suggest that an interview can include questions about the
arbitrator's view on document disclosure and other procedural matters. (185)
4. The interview should take place at the arbitrator's office. (186) An arbitrator
should not be ‘summoned’ to attend the party's office, (187) as this risks
creating perceptions of impartiality. Likewise, an interview over lunch or
involving hospitality, regardless of who pays, would be ill-advised. (188) page
"284" If the party is in a foreign country, any meeting should be on neutral
territory.
5. A time limit should be placed on the interview. (189) This ensures control over
the process and minimises perceptions of bias. Gerald Aksen insists on a
maximum length of thirty minutes. (190) Unsurprisingly, the ICC refused
confirmation of an arbitrator who had spent fifty to sixty hours reviewing
the case with a party before appointment. (191)
6. It should be made clear that appointment does not entail any special
obligations to that party. (192)
7. Only counsel may ask questions, not the parties. (193)
8. Discussion of past arbitral experience should not include the names of
parties. Aksen insists on this to ensure confidentiality.
9. More contentious is whether a record should be made of the meeting for this
then to possibly be disclosed to the other party. (194) Those in favour suggest
that this ensures transparency and minimises perceptions of bias. Born has
criticised this as excessive. (195) Friedman has suggested that tape-recording
and disclosure is intrusive and demeaning and may lead to more litigation
from disgruntled parties. (196) Poudret and Besson suggest that, at the very
least, the fact that the interview occurred should be disclosed. (197)
10. If the interview is with a prospective sole arbitrator, no interview should
take place unless both parties or their counsel are present. (198)
11. page "285"The basis upon which the interview is to be conducted, including
conforming to the above guidelines, should be made clear in writing by the
arbitrator prior to the meeting. (199)
12. Likewise, the interviewing team should make clear who will be attending
the interview and who will be leading it. (200)

One complication, Aksen notes, is that it is much easier for an experienced

arbitrator to apply such strict stipulations than it is for a new arbitrator. The
competitive pressures of international arbitration mean that it can be difficult for
an aspirational arbitrator to refuse an interview or demand strict guidelines.

5.6.3. When should an Interview be Refused?

There are probably some circumstances in which an interview would be

inappropriate. Obviously, an interview is inappropriate if a party indicates it is
unwilling to comply with the above limitations or if it is clear that the
interviewer is seeking a partisan arbitrator. Another situation may be where the
party seeking the interview is known to have had enough prior understanding of
the potential nominee's attributes. An example would be where they have been
involved in a previous case where the arbitrator sat. In such circumstances, a
brief conversation (possibly by email) to assess availability and potential
conflicts would suffice. An interview may also be difficult to justify for well-
known arbitrators where the party is familiar with international arbitration. On
this basis, Gerald Aksen refuses to conduct interviews with US or multinational
corporations with experience in international arbitration. (201)

5.6.4. Alternatives to Interviews

It should be noted that interviews are not the only way to find out key
information about prospective arbitrators. The extensive network of arbitration
practitioners could readily lead to widespread confidential attestations as to a
person's attributes as an arbitrator. (202) Likewise, many organisations collect the
CVs of arbitrators. (203) page "286" Furthermore, a short email to assess
availability and conflicts is less problematic than an interview.

5.7. Accepting Appointment

It is one thing for a party or appointing authority to have decided on its preferred
candidate. A separate consideration is the circumstances in which an arbitrator
may accept appointment. Acceptance is the essential final stage in constituting
the tribunal. (204)

5.7.1. Assessing whether Acceptance is Permissible

IBA Ethics Rule Article 2.2 indicates that a prospective arbitrator shall only
accept appointment if fully satisfied of competence. Yet it is not easy to
determine what makes for appropriate competence in an individual case. The
issues overlap with the attributes discussed above in terms of who to nominate.
The key difference is that where acceptance is concerned, a proposed arbitrator
must consider whether they meet minimum standards. Where nomination is
concerned, one is looking for optimal standards.

Absent a specific direction by the parties or in the lex arbitri, there are no
minimum professional qualifications or experience required. If arbitral
experience was necessary to having competence, no one could ever accept their
first appointment. There are a number of elements that seem relevant, however.
They must have the time to devote and be diligent in that task. They must be able
to understand written and oral arguments about facts and law. In addition, the
abstract concern for competence or a prospective arbitrator's need to self-assess
in that regard, also needs to be looked at alongside party autonomy where a
party has invited a particular arbitrator to accept appointment. As noted by
Schwartz, the very confidence that the appointor has in the nominee is relevant
to assessment of competence and to the choice of arbitration in the first place.
(205)

A prospective arbitrator also has to decide whether he or she has the requisite
independence and impartiality. (206) Gerald Aksen, before accepting an
appointment as an arbitrator, insists on the following information to ensure that
he has the requisite impartiality and competence:

(1) The names of all of the parties, including a complete listing of the parties'
parents and subsidiaries
(2) The names of the firms representing the parties
(3) The names of the co-arbitrators (if selected); and page "287"
(4) The identity of the applicable arbitration rules (along with the language of
the arbitration clause if it is an ad hoc one). (207)

Once this information is received, the following options are available, decline the
appointment; disclose information relevant to appearance of bias but which you
believe would not affect impartiality; do not disclose the information. Disclosure
is discussed in more detail below in section 5.10.

The most significant issue is usually availability. In terms of availability, if a

person is approached to be chair, it is desirable to not only consider their own
availability but the extent to which their free time overlaps with that of the
party-appointed arbitrators. While experienced arbitrators may know that more
than 40% of cases eventually settle, it would not be reasonable to overbook on
the presumption that time will become available if settlements flow at the
anticipated rate. While that is the norm in domestic common law litigation, this
flows from the fact that there are a fixed number of judges and there is then a
need to try and set down a realistic workload for them. Where arbitration is
concerned, to consciously overbook is to accept incompatible commitments if the
matters proceed. A general statistical settlement norm cannot be presumed to
apply to a relatively small statistical sample of even the most popular
international arbitrators. Karl-Heinz Böck-stiegel has suggested that arbitrators
‘should only accept an appointment if, taking into account the normal
development of the pending cases and the suggested new case, we can be sure of
having enough time available to deal thoroughly with all the work and meetings
coming up in the various proceedings.’ (208)

5.7.2. Acceptance Procedure

Little has been written about the precise form that acceptance should take. In
practice, the procedure for acceptance is usually quite a short process once the
appropriate arbitrator has been identified. Typically, the party simply provides
written notice (209) to the other party (and the institution if applicable and
required by the rules) of its choice of arbitrator. This will often be accompanied
by a signed statement of independence or impartiality by the arbitrator, as well
as disclosure of any information that could affect independence or impartiality.
(210) Unless there is any objection, the appointment is then effected. (211)
page "288"

In ad hoc cases it will often be that the first approach is an in principle offer
calling for advice from the arbitrator as to his or her terms and conditions. There
might then be instances where the party or parties might not wish to accept a
particular term, whether as to general fees and expenses, cancellation fees or
indemnity or the like.

5.8. Timeliness, Urgency and Tribunal Constitution

It has been suggested that it takes approximately four months to constitute the
arbitral tribunal in the average case which seems excessive when efficiency is of
key importance. (212) It has been observed that the typical thirty day rule to
nominate an arbitrator and the extra time to select a chair or have a nomination
by an institution could be cut down significantly. (213) A particular problem arises
where provisional measures are sought. Where these are urgent, the time taken
to constitute a tribunal is particularly problematic. Ideally if an institution is
involved, some special mechanism for urgent appointment is desirable.

Traditionally, parties had to rely on national courts if urgent measures were

required. (214) A number of institutional rules now provide mechanisms for
urgent decisions. These mechanisms include:

• A ‘pre-arbitral’ phase for early decisions by a referee. (215) This does not usurp
the jurisdiction of a court or arbitral tribunal, but rather is a fast mechanism
which can provide a foundation for final settlement. Under the ICC Pre-
Arbitral Referee Procedure, the referee's appointment can be challenged, (216)
just like an arbitrator, but this is dealt with as quickly as possible.
• Summary arbitral proceedings under the NAI Rules. (217) Under these rules,
arbitrators in such proceedings can be challenged in the same manner as
ordinary proceedings. (218)
• Expedited formation of the tribunal under the LCIA Rules. (219) If the
application is made, the LCIA court ‘may, in its complete discretion, abridge or
curtail any time limit under these Rules for the formation of the Arbitral
Tribunal, including service of the Response and of any matters or documents
adjudged to be missing from the Request.’
• page "289"Appointment of an ‘emergency arbitrator’, typically within one
business day of request for urgent relief. (220) Emergency arbitrators must
make the same disclosures as other arbitrators, but any challenge usually has
to be made within one day. (221) Any decisions made by the emergency
arbitrator can be modified or vacated by the tribunal when it is constituted.
(222)

Another question is whether the parties should select simultaneously or

concurrently. One view of equality is that each should select without being aware
of the attributes of the other party's selection. The contrary argument is that the
respondent needs to know whether the claimant has chosen a true independent
expert or instead someone who may, through past decisions or writings, have a
predisposition to a particular view. In the latter event, the respondent might
argue that true equality should allow them the same right which would only be
availed of where necessary.

5.9. Challenge and Removal of an Arbitrator

Some legal systems would leave it to the parties to determine the rules as to
challenge, removal and replacement of arbitrators. Some see these rules as more
fundamental and hence impose requirements in the lex arbitri that would take
precedence over any agreement of the parties.

The rules and procedure of challenging arbitrator selection have to balance

various competing objectives. On the one hand, parties should not have to
participate in an arbitration in which one of the arbitrators has a conflict of
interest or otherwise lacks independence. Stressing the importance of this,
Paulsson has noted that the ongoing legitimacy of international arbitration itself
depends on parties having confidence in arbitrators being ‘immune from
favouritism’. (223) On the other hand, any challenge causes delay, and there is a
need to minimise ‘tactical challenges’ that have little genuine basis. Such
challenges tend to exploit the fact that arbitrators err on the side of caution and
over-disclose what may be innocuous facts, and the natural inclination of
institutions to be over-cautious so that the award is not set aside, (224) and so that
their processes run smoothly and speedily. Many commentators believe that
such opportunistic challenges are increasing. (225) page "290" There are even
reports of extreme cases where a party or its counsel has telephoned an
arbitrator and recorded the conversation, hoping to create evidence of
impropriety on the basis of ex parte communication. (226)

Some commentators believe that challenges to arbitrators are increasing

significantly and may pose a threat to the efficiency and legitimacy of arbitration.
(227) Although the problem is said to be increasing, (228) successful challenges are,

in practice, fairly rare. (229) Hence care should be taken to avoid unduly
interfering with potentially legitimate challenges simply in order to avoid
guerrilla tactics. This is particularly relevant to questions of disclosure. There are
also differing tactical considerations between claimants and respondents. Craig,
Park and Paulsson note that, as a matter of strategy, claimants should challenge
as infrequently as possible because of the delay it causes and the possibility that
the replacement candidate's independence will be equally questionable. (230)
Respondents should also be careful with spurious challenges as technical
barriers tend to create the impression that the party has less in the way of
arguments on the merits.

5.9.1. The Principal Ground for Challenge: Lack of Independence or

Impartiality

Section 5.5.5 above identified independence and impartiality as positive

attributes when considering arbitrator selection. The corollary is that, typically,
arbitrators can be challenged when their independence or impartiality are called
into question. The obligations of independence and impartiality come from
various sources, page "291" including national laws (231) and arbitral rules. (232) It
is generally thought that the Model Law requirement that an arbitrator be
independent and impartial is a mandatory provision. (233) Lew, Mistelis and Kröll
say that independence and impartiality are part of the ‘Magna Carta’ of
international commercial arbitration. (234) The concepts are sometimes indirectly
applicable. The ICSID Rules do not expressly mention impartiality or
independence, although arbitrators are required to sign a declaration stating
that they will ‘judge fairly as between the parties, according to the applicable
law, and shall not accept any instruction or compensation’ other than under the
ICSID Convention. (235) A failure to be independent and impartial may also ground
a due process challenge to enforcement under public policy grounds. Hence the
principles may be similar regardless of the wording of the applicable rules,
although these are of course the primary elements for consideration.

Before looking in detail at the content of this ground of challenge, it should be

noted that, even if the ground appears to be made out, it is not always wise for a
party to challenge, except where the concerns are significant, for three reasons.
First, a challenge will typically increase delays and costs. This will also be
affected by the rules on replacement. If the opposing party can simply replace
one predisposed arbitrator with another likely to be so, little may be achieved.
Second, it was noted above that nominating a clearly partial arbitrator can have a
perverse effect: it can make the other arbitrators, and importantly the chair,
sceptical of everything that side puts forward. If the other side has fallen into the
trap of doing this, pointing this out by way of a challenge may not always be a
good strategy. (236) Third, the rate of success in challenge applications is quite low
across most institutions and courts.

page "292"

5.9.2. Definition of Independence and Impartiality

Much has been written as to the exact meaning of the terms and what, if
anything, is the difference between the two. The question has analytical
significance at least, because some rules use both terms while others simply
mention impartiality. The 1998 ICC Rules were a notable example, which
provided for challenge for ‘lack of independence or otherwise’. (237) The 2012
rules now provide for challenge for independence or impartiality or otherwise.
(238) Other statutes and rules also vary in their precise formulation. Under the

English Arbitration Act, for example, dependence is only a disqualifying factor if

it amounts to partiality. (239)

To some, the terms ‘independence’ and ‘impartiality’ are validly used

interchangeably. (240) Others consider them to have different meanings. (241) The
better view is that the notion of impartiality would encompass independence,
hence nothing should turn on whether rules mention both or only one. That is
not a uniform view and in any event, attention needs to be given to the meaning
of each term to understand what they cover.

Dependence tends to refer to a relationship between the arbitrator and a party

or counsel. (242) Partiality, by contrast, tends to refer to favouring one of the
parties or prejudice in relation to the subject matter. (243) Because of this,
independence analysis tends to be employed using an objective test, because it is
based on the existence of a relationship and not a state of mind. (244) Partiality, by
contrast, page "293" deals with actual or apparent bias by the arbitrator and thus
can have a subjective element, (245) although legal systems will often rely
predominantly on objective facts to determine the likelihood of such a state of
mind. This objective approach would be particularly relevant where arbitrator
challenges are concerned as adjudicators would rarely cross-examine an
arbitrator to elucidate actual bias. In any event, a person who lacks
independence may be expected to be subject to undue pressure and hence to be
lacking impartiality.

It has been suggested that the two concepts have different ‘life-spans’. (246)
Independence must remain for the entire proceedings. The suggestion is that a
literal use of the term impartiality means that at some point it comes to an end
when an arbitrator reaches a decision on who should win the case. The better
view might be that partiality is about predisposition and not the actual
determination, which must be based on an impartial analysis of facts and law.

No consensus exists in the precise test that should be applied in assessing the
merit of a challenge. More important than the terms themselves are the
applicable standard of proof and the evidence that is appropriate for such
determinations as fact situations will rarely be clear-cut. In many rule systems,
the threshold to be applied under the statute or rules is whether there are
‘justifiable doubts' about independence or impartiality. (247) It seems that this test
is construed objectively. That is, are there reasonable doubts as tested by a fair-
minded, rational, objective observer? (248) Even then, much may turn on whether
the applicable standard is that there may be doubts, there are likely to be doubts,
or there would be doubts. This is discussed further below in the context of the
IBA Guidelines.

5.9.3. Actual and Apparent Bias

Although the standard required by most rule systems is ‘justifiable doubts’, many
national courts, especially in common law countries, continue to refer to the
long-held distinction between ‘actual bias’ and ‘appearance of bias’ in assessing
impartiality and independence. All jurisdictions will remove an arbitrator if
‘actual bias’ is proven. (249) However, the situation is more complicated with
‘apparent’ bias (which many European lawyers call ‘objective bias’). While most
jurisdictions will remove an arbitrator for mere appearance of bias, (250) the
precise test used varies.

page "294"

Within apparent bias, three different tests exist across different jurisdictions. (251)
The first and easiest to satisfy is the ‘reasonable apprehension’ test. Under this
test, the question is whether a fair-minded and informed observer would have a
reasonable apprehension that the arbitrator was biased. (252) The second is the
‘real possibility test’, under which a fair-minded and informed observer would
conclude that there was a real possibility that the arbitrator was biased. (253) This
appears to be the prevailing approach in several common law States. (254) The
third test is the more demanding ‘real danger’ test, under which the court must
find that there is a real danger of bias before any finding of apparent bias can be
made. (255) Luttrell has argued persuasively in favour of this latter test as the best
way to minimise tactical challenges that disrupt arbitration. (256) Superimposed
upon this set of options is the above-mentioned distinction between a standard
that requires only a possibility, or one requiring a likelihood, or one demanding
that there would be such a conclusion. Even that is subject in each case to the
evidentiary standard of proof under the applicable law.

5.9.4. The Relevance of the IBA Guidelines on Conflicts of Interest

IBA Guidelines were published in 2004 with the aim of bringing greater clarity
and uniformity of approach to when an arbitrator should disclose certain
information and what should be regarded as disqualifying factors. (257) The
Guidelines are in the form of ‘General Standards’, plus an explanatory text and in
addition, a ‘red’, ‘orange’ and ‘green’ list of specific examples to further explain
the General Standards. Naturally, the Guidelines do not change existing lex arbitri
and arbitral rules or party stipulations, page "295" but they can be influential in
application and interpretation. As with any such attempt to identify key
principles in contentious areas, these can themselves be contentious. The
Guidelines are not binding unless the parties have agreed for them to be so.
Further, they are almost never used as the sole source of guidance. That said,
they are now commonly used by counsel and most institutions to inform and
assess arguments. (258) Their authority lies in the fact that they were produced by
a working group of eminent arbitrators from common law and civil law
jurisdictions. The IBA Guidelines are generally said to be a thoughtful and
welcome aid to conflict of interest problems. That said, the drafters faced
obvious problems in developing specific rules for infinitely diverse fact
situations, as they themselves acknowledged. (259) Their use by courts is rare, (260)
but not unheard of. (261) The Svea Court of Appeal recently stated that they ‘serve
as important guidelines for counsel and arbitrators and also have some
relevance as background material [for] the Court of Appeal’. (262) Likewise, the
Swiss Federal Supreme Court has said that they are ‘a valuable working tool’ that
‘should impact on the practice of the courts and the institutions administering
arbitration proceedings’. (263) Other courts in England, (264) the United States, (265)
the Netherlands, (266) Belgium, (267) Germany (268) and Austria (269) have mentioned
the Guidelines. (270)

page "296"

While the publication of the IBA Guidelines clarifies the situation, it has been
suggested that the presence of such guidelines may indeed be an incentive for
increased challenges, simply by articulating standards around which
relationships can be assessed. That would depend on the content and whether it
provides broader grounds for challenge. At the very least, the guidelines make it
easier to identify and articulate the reasons for decisions on challenges.
Naturally, the content of the IBA's Orange List has received the most criticism.
This is natural with any attempt to define dividing lines in grey areas. Extremes
are always easier to articulate and for which to construct illustrative examples.
The Orange List has been criticised for providing insufficient guidance, but as
Redfern and Hunter point out, any attempt to create a comprehensive list will
likely be incomplete. (271) Despite criticisms and some institutional ambivalence,
the Guidelines have, nonetheless, become generally accepted as a non-binding
set of principles with which most arbitrators seek to comply. (272) They are
considered by many arbitral institutions. Redfern and Hunter say that, in their
experience, they are commonly relied on by challenging parties, and are at least
one significant factor in the minds of the deciding authorities. (273) The important
thing is not to use the colour coded examples rigidly and in isolation of the
general principles enunciated in the Guidelines.

5.9.5. Factual Circumstances Giving Rise to Lack of Independence or

Impartiality

In assessing whether a given case shows a lack of independence or impartiality,

regard should be had to case law and commentary, including the IBA Guidelines.
Some national laws also provide specific examples. (274)

The main circumstances that can give rise to a challenge can be summarised as
follows:

– Connection with one of the parties or counsel

– Issue conflict
– Prejudgment
– Serious errors or misconduct of the arbitrator.

An extensive list of published cases is contained in Appendix 1. The aim is not to

attempt any synthesis of principles but instead, to provide an initial guide to the
kinds of situations that have arisen and the way these have been treated.
Obviously much will depend on the facts of each case and the views of one
adjudicator may page "297" not be accepted by others. The following sections
provide further thoughts on these specific categories, and can be read in
conjunction with Appendix 1.

5.9.6. Connection with One of the Parties

The fact that there is some connection between a party and an arbitrator does
not in and of itself establish a lack of independence or impartiality. (275) Some
connection between the party and the arbitrator must be acceptable. To hold
otherwise would be to defeat the purpose of party-appointed arbitrators. (276) In
addition, because arbitrators are drawn from the private sector, years of
commercial, scholarly and organisational activity will often lead to some at least
indirect links between arbitrators and entities related to them on the one hand
and parties or their related entities on the other hand.
That said, there is a long-held legal principle that no person should be a judge in
his or her own cause (nemo debet esse judex in propria causa). The closer the
relationship between an arbitrator and a party, the more this principle is of
concern. Similarly, any direct interest in the outcome of the case will rightly be
subject to challenge. (277) Whether or not a connection with one of the parties
amounts to lack of independence/impartiality is essentially a question of degree
resolved on a case-by-case basis.

5.9.6.1. Some Easy Cases

Extreme examples which clearly compromise independence include: the

arbitrator being a manager or board member of one of the parties or its
subsidiary; (278) the arbitrator having a significant financial interest in a party;
and the arbitrator being a regular advisor to a party. (279) Likewise, non-financial
involvement in the case, such as working as the project manager on a
construction project in dispute, or acting as the lawyer who drafted the contract
in dispute, are likely to lead to disqualification. (280) A close family relationship
with a party or counsel is also an automatic ground for disqualification. For
example, an arbitrator was prevented page "298" from accepting an appointment
because his wife worked as an assistant for one of the party's counsel. (281) In
another case, a challenge was sustained when the son of one of the arbitrators
joined the law firm acting for one of the parties. (282)

In addition to the propriety of particular linkages, there is the related question as

to whether objections may legitimately be waived. The IBA Guidelines state that,
notwithstanding the general principle of party autonomy, parties cannot consent
to obviously biased arbitrators (although there was some debate on this point in
the Working Group). (283) The Guidelines include a non-waivable red category,
which include an example of an arbitrator who is an employee of one of the
parties or who is a regular advisor to one of the parties from which substantial
income is derived. (284) At the other end of the spectrum are clearly permissible
situations, such as: having previously published a general legal opinion on a
matter relevant in the arbitration; other lawyers at the arbitrator's firm having
previously acted against one of the parties; the arbitrator being a member of the
same professional or social organisation as one of the parties; and a limited,
controlled interview prior to appointment. (285)

5.9.6.2. Friends and Acquaintances

One difficult area is when the arbitrator is a friend of the appointing party or its
counsel. This will be a question of degree. Merely being an acquaintance, say via
contact at professional conferences, is usually insufficient to disqualify the
arbitrator. But, in cases of close friendship with regular contact, the appointment
might be difficult to justify. (286) Relevant factors in drawing the line will be the
length and nature of the friendship as well as the regularity and quality of
contact. (287) There is also the converse scenario where an arbitrator is an ex
spouse of one of the counsel.

5.9.6.3. Minor Shareholding

Minor shareholding does not per se show lack of independence. (288) The
Guidelines include in the green category a minor shareholding in one of the
parties if it is page "299" publicly listed. (289) This should be particularly so in
jurisdictions where the shareholding is through a general mutual fund. (290)
However, any financial interest that creates a ‘relationship of subordination’
between the party and the arbitrator is a good ground for challenge. (291) One
example may be where the arbitrator is a director of a company which is a
portfolio investor in a party. (292) The situation would also be different if
shareholding is in a private company as opposed to a publicly listed company as
this affects the closeness and the means of acquiring the shareholding.

5.9.6.4. Past Commercial and Arbitral Relationships

Great difficulty arises with past relationships. Here there are two conflicting
policy aims. On the one hand, a past relationship that does not add to the
likelihood of a biased decision in the current case should not be problematic. On
the other hand, justice should be seen to be done so that the parties feel
confident that there was a truly independent assessment. As noted, the difficulty
is that leading commercial arbitrators have usually had many years of experience
through a range of organisations and through which contact has been made
directly and indirectly with numerous entities and individuals. If appointments
were limited to arbitrators without even tenuous connections, the pool of
available experts would be greatly diminished. Hence some judgment needs to
be exercised at the margin.

One case where lack of independence has been suggested to arise is where the
arbitrator has previously acted as counsel to the party appointing him or her. (293)
Craig, Park and Paulsson note that some commentators see this as ‘an absolute
bar’ to appointment. The US Supreme Court vacated an award where the
presiding arbitrator failed to disclose that he had been a consultant for one of the
parties up until one year prior to the arbitration. (294) The preferred view is that
each case must be assessed on its merits and that such a past relationship should
not necessarily be enough for disqualification. (295) An obvious example would be
where the arbitrator has acted as counsel for both sides on previous occasions.
In the more typical case where an arbitrator has only acted for one side, relevant
factors in assessing this should include the frequency of the past work, how long
ago it occurred and the likelihood that the arbitrator will be affected by the
prospect of future work from page "300" the party. (296) The nature of the work
may also be important. For example, if a party undertook a losing case on a legal
question on the advice of counsel, that counsel as arbitrator on the same point
may feel pressured to vindicate the earlier position.

A similar problem arises out of the fact that many arbitrators, and their firms,
take on work both as counsel and as arbitrators. Thus an arbitrator may be
serving in a case where one of the parties' counsel has rendered an award for or
against that arbitrator or his or her firm in an earlier case. This arose in SGS v
Pakistan, (297) where counsel for Pakistan had previously handed down an award
in favour of a party represented by the firm of the arbitrator in the SGS case. The
arbitrator was challenged on the basis that the arbitrator may feel indebted to
Pakistan's counsel due to the earlier decision. The challenge was rejected, with
the panel noting that the arbitration community is small and that such inevitable
overlap is not, without more, sufficient for disqualification. (298)

5.9.6.5. Connection via the Arbitrator's Law Firm

One common issue is when one of the arbitrators is partner of a large law firm,
and another partner has a connection with one of the parties or counsel. (299) The
IBA Guidelines seek to move away from any presumption that a conflict of
interest of an arbitrator's partner disqualifies the arbitrator by association. (300)
Craig Park and Paulsson claim that the general trend is towards not allowing
partners in a large firm to act as arbitrators on cases where one of their partners
has advised one of the parties, even on unrelated matters. This is on the basis
that there is a theoretical financial conflict of interest – the partner will profit
from his or her colleague's ongoing work for the party. It is preferable to look at
matters on a case-by-case basis. Any hard and fast rule will simply be an
incentive for leading arbitrators to set up boutique practices at an earlier stage in
their careers. Some also adopt new forms of relationship with their previous
firms, trying to shelter themselves from any equity interest. These arrangements
may also have to be examined on a substance over form basis, but may raise
confidentiality issues and those seeking to challenge will have difficulty in
knowing enough about the arrangements so as to frame the challenge.

page "301"

Published cases on these types of connections go both ways. Invariably, they turn
on the specific facts of the case. The SCC recently sustained a challenge to an
arbitrator because his firm had worked for one of the parties, even though the
work was unrelated to the dispute at hand and had occurred several years
earlier. (301) Likewise, the Paris Court of Appeal recently held that there was an
impression of bias where the arbitrator had for many years worked at the law
firm representing one of the parties. (302) In one ICSID case, the President of an
annulment committee was challenged on the basis that another partner in his
law firm was giving advice to one of the parties. The advice was unrelated and
nearing completion and the President had no involvement in it. The challenge
was dismissed, with the tribunal noting that the President had disclosed the
information immediately, that the President had no direct relationship with the
party, that the work was unrelated to the case, that the work was about a minor
element of a specific transaction, not general strategic advice and that the work
was soon coming to an end. (303) A number of other similar cases are summarised
in Appendix 1.

A related issue is barristers who share chambers. Some debate exists over
whether this can give rise to a conflict. That is, what happens where the
arbitrator shares chambers with one of the parties' counsel? This appears on the
IBA Guidelines Orange List, so the facts of each particular case will be important.
The position in England (where the issue is most likely to arise) seems to be that,
on its own, sharing chambers is not a disqualifying factor. (304) This has been
criticised as insensitive to foreign parties unfamiliar with the primarily English
practice of barristers' chambers. (305) A recent ICSID case came to a contrary
conclusion, although this seems to have been based largely on the fact that the
connection had not been disclosed, (306) even though non-disclosure should not
itself be a ground for challenge unless it alone shows a lack of impartiality in the
relevant circumstances. In any event, it seems advisable to disclose shared
chambers, especially to non-British parties. (307)

page "302"

A similar problem concerns affiliations between law firms. If the law firm of one
of the arbitrators has some affiliation with one of the parties' law firms, does a
partiality/dependence issue arise? Again, the particular facts need to be
scrutinised. The IBA Guidelines suggest that whether or not the firms share
revenue is a key factor. (308) The closer the affiliation and risk of partiality, the
more it should be a disqualifying factor. Where large firms are concerned, a more
limited association may be a stepping stone to a merger and may thus be of
significant concern. For mid-sized firms, there may not be much in the way of
profit sharing but cross referrals are an important objective which could also
impact upon the legitimate concerns of reasonable observers.

5.9.6.6. Repeat Appointments

A similar issue can arise where the arbitrator has previously served as an
arbitrator in a case involving one of the parties. Previous appointment by the
same party or firm appears on the IBA Guidelines' Orange List. On its own, this is
usually insufficient for disqualification. (309) However, in 2008, the ICC refused to
confirm the appointment of an arbitrator who had served as the claimant's
party-appointed arbitrator in three previous cases. (310) One would hope that
parties who are happy with the efficiency and expertise of a particular arbitrator
are not ipso facto barred from nominating them in the future. The overall
efficiency of arbitration is enhanced if the best people can expect multiple
appointments. However, repeated appointment by the same party raises the
problems of expectations and strong financial dependence on ongoing work.
While repeated appointments may be a problem in most cases, with specialty
areas such as maritime and commodity disputes this may be inevitable given the
small pool of experts that dominate the fields. (311)

In the recent Suez, case, an ICSID tribunal was called upon to decide a challenge
to one of the arbitrators, who had decided against the respondent State in an
earlier case. The Respondent State argued that inconsistencies in the earlier
award showed a manifest lack of independence. This challenge was dismissed,
with the tribunal noting that there had been no allegation of the arbitrator
having an interest in the outcome of the case and that there was no evidence of a
lack of independence in the case in the lead up to the challenge. The tribunal also
page "303" noted that the test for independence was an objective one. (312) By
contrast, if the decision in the earlier case effectively prejudges liability for a
party in the later case, disqualification is appropriate. (313)

5.9.6.7. Private Communications with a Party

Ex parte communication between the tribunal and one of the parties can create
an impression of bias if not done appropriately. As a general rule, any
communication between the tribunal and a party should be made known to the
other party to minimise risks of impartiality. It is highly inappropriate, for
example, for an arbitrator to have private meetings with one party's counsel, as
occurred in one LCIA arbitration. (314)

One situation in which problems can arise is inappropriately conducted ‘med-

arb’, i.e., where the tribunal seeks to encourage settlement via mediation.
Because of the risk of compromising impartiality, this practice is avoided in
many countries. However, it has a strong history in China and Korea, (315) and is
permitted by the lex arbitri of Singapore (316) and Hong Kong (317) where the
parties have specifically consented. The key issue is as to how any mediation is
conducted. Mediation theory often advocates that parties be spoken to
separately and in confidence to help them overcome impasses. There are obvious
problems, however, if an arbitrator exercises such a mediation function and the
mediation is not successful. If this mediation includes an arbitrator meeting
privately with the parties, issues of apparent bias can arise. (318)

A recent example of a problematic ex parte communication came in the case of

Gao Haiyan (319) where the Hong Kong Court of First Instance refused to enforce a
Xian Arbitration Commission award on public policy grounds. In Gao Haiyan, one
of the arbitrators, along with the arbitral institution's Secretary-General, met
with a person who had close ties with one of the parties over dinner. This
occurred several months before the award was handed down. They informed
him that an award would be made in favour of the party, but that it would have
to compensate other parties. The Court rejected the attempt to frame the dinner
as a mediation, concluding there was an appearance of bias.

page "304"

5.9.6.8. Nationality

Common nationality between a party and an arbitrator, without more, will

usually not be enough to show partiality. This view is supported by decisions of
national courts (320) and arbitral institutions. (321) In a similar vein, the LCIA
rejected a challenge in an arbitration involving an Arabic party, where the
arbitrator had a long-standing interest in Arabic studies and affairs. (322)

5.9.6.9. Government Employees in Developing Countries

One problem that may arise more often in cases where an arbitrator is from a
developing country with a centralised economy is where the arbitrator is a
government employee or director of a State enterprise. For example, can a
government employee hear a case in which his State or one of its entities is a
party? In some countries, such employees may be part of a very small pool of
appropriately trained people. Here there are competing policies of ensuring
independence and encouraging arbitrators from developing countries, especially
if the party is from a developing country. The ICC has tended to be fairly flexible
in allowing such appointments, with the rationale that to do otherwise would
essentially bar many potential arbitrators from developing countries. (323) Some
time ago, Werner criticised this approach as outdated and unfair. (324)

5.9.6.10. General Relevant Connection Factors

What general principles can be discerned from the above cases? The Suez
decision used a four point test to qualitatively assess whether a connection was
sufficient to show lack of impartiality. (325) The four points, which provide useful
general guidance, were:

– Proximity: how closely connected is the challenged arbitrator to the party by

reason of the alleged connection? page "305"
– Intensity: how intense and frequent are their interactions?
– Dependence: to what extent is the arbitrator dependent on the party for
benefits through the alleged connection?
– Materiality: to what extent are any of these benefits significant and thus likely
to influence the arbitrator's judgment in some way? (326)

5.9.7. Issue Conflict

Issue conflict is typically where a lawyer is sitting in one case as an arbitrator

deciding on a point of law which, in another case, is simultaneously being
decided on where the lawyer is counsel. (327) This may be particularly prevalent
in investment arbitration due to the discrete number of legal concepts that can
be the point of dispute, the fact that awards are published and the small number
of practitioners in the field. The issue was touched on above in relation to
commercial connections.

This specific question arose in Republic of Ghana v Telecom Malaysia, where a

Dutch court held that an eminent French arbitrator, Professor Gaillard, should
not be part of the tribunal because he was counsel in another case dealing with
similar points of law. (328) The Court held that, even if the arbitrator did not have
an actual issue conflict, he had an appearance of one. It saw the problem as being
that in his capacity as counsel, Professor Gaillard would be putting forward all
possible pro-investor arguments on expropriation, while simultaneously having
to act objectively as an arbitrator in the other case. The Court decided that the
arbitrator could, however, remain on the tribunal if he resigned his position as
counsel within ten days. From a policy perspective, much may depend on the
circumstances and the Court's logic is not unassailable. At a general level, the
concerns might look similar to cases where an arbitrator has previously written
on a particular legal point. Some may worry that they might not wish to
undermine their previous scholarly opinion with a contrary arbitral decision.
However, authors of the leading works may often be the best people for
appointment. Where issue conflict is concerned, the likely order may have some
relevance. If the arbitral award was to predate the counsel work, an arbitral
decision contrary to the counsel brief would render that work less than fully
effective, as counsel would be demonstrably putting arguments that he or she did
not believe in as evidenced by the award. Conversely, if the counsel work
predated the award, it would be easier to come to a contrary conclusion that is
justified by the needs to have first honoured the counsel brief. Resigning the
counsel position as suggested above may have its own legal and ethical
implications outside of the scope of this book.

page "306"

A similar issue arises where an arbitrator is sitting in multiple related

proceedings. This may occur because a party to multiple arbitrations may wish
to appoint the same arbitrator for each of them. Timing issues are less relevant
as the concern will always be that the later decisions will be coloured by the
decision first taken, although this would be acceptable if all hearings were
roughly concurrent, with conclusions being reached on each matter around the
same time. While there is no problem with this in principle, the arbitrator will
have to be careful not to use evidence from one arbitration in another arbitration
unless proper disclosure and processes are followed. (329) The situation is more
complicated where one of the cases has finished and the arbitrator has
apparently made up his or her mind on the issue. In such circumstances, the
arbitrator may be accused of prejudgment (see below). (330) Bedjaoui has noted
that in related proceedings where the facts are identical but the legal issues are
different, there should be no issue of impartiality or independence. (331)

5.9.8. Prejudgment

Arbitrators may be challenged on the basis of prejudgment when their

comments, either prior to or during the arbitration, indicate that they have not
come to the case with an open mind and have a fixed view on the issues to be
decided. At its most extreme, the arbitrator may have publicly commented on the
merits of the dispute or on one of the parties prior to being appointed. In the US-
Canada softwood lumber dispute, an arbitrator was removed because of a speech
he had given eighteen months earlier, which indicated that the US claim had no
merit. (332) Another example may be a procedural order which expresses views
about a matter on which the parties have not yet made submissions, such as the
ownership of contested property. (333)

In a recent investment arbitration, the respondent seized on comments by the

arbitrator in the context of a hypothetical question put to an expert witness. The
arbitrator commented that the facts made it clear there had been major harm or
change in the expectations of the investment. This was central to the case. The
challenge was rejected by the LCIA Court, which, applying an objective test of
impartiality, noted that the context of the hypothetical questions showed that
page "307" there was no apprehended bias. (334) US cases have added that, even
where such questions show how the arbitrator is thinking, it is an acceptable
part of the evaluation of evidence. (335) This is certainly the better view. As noted
in section 6.2.8.6, giving a party a meaningful opportunity to fully present its case
may at times invite a warning from a tribunal as to the hurdles that need to be
met. If a tribunal is disposed to hold against a party unless a successful argument
is put on such an issue, the party would certainly like to know that in advance.
As noted above, this issue may also arise where an arbitrator has been a frequent
author or speaker taking a position on a particular issue. Craig Park and Paulsson
suggest that a distinction may be made between party-appointed and presiding
arbitrators on this point, at least in practice. The ICC, they say, may be prepared
to allow such an arbitrator to be party-appointed, but would be very hesitant to
appoint a chair who had published a detailed or extreme view on a particular
issue. (336) The distinction seems sensible. Once again, we expect the best
arbitrators to engage in some degree of scholarship and it would be unfortunate
to provide incentives against them doing so. Furthermore, counsel can look for
predisposition in other ways such as previous decisions. However, where the
chair is concerned, it is better that the institution not be seen to have swayed the
likely result by the appointment itself. One issue is how much research would
need to be undertaken in terms of respective curriculum vitae and past decisions
to ensure that the chair appointment seemed sufficiently open-minded.

Arbitrators should also be careful when speaking about pending cases. During a
recent ICSID case involving Ecuador, an eminent American arbitrator gave an
interview to a magazine about general issues in international arbitration. (337) He
commented that there was a need for acceptance of international arbitration by
States. He cited Ecuador's refusal to follow orders of two ICSID tribunals as an
example of the problem. Ecuador then successfully challenged the arbitration.
The Secretary-General of the Permanent Court of Arbitration, who heard the
matter, said that there was no absolute prohibition on arbitrators talking about
pending cases, but the arbitrator's words in this case were sufficient to create an
impression of bias. Arbitrators should also take care to ensure that their
assistants or law clerks do not make any adverse public statements about a
pending case. (338)

page "308"

5.9.9. Arbitrator's Conduct or Errors within the Case as Evidence of Bias

An arbitrator may also be challenged on the basis of conduct within the case
which is said to be so unfair that it, in and of itself, shows a lack of impartiality. It
seems that such arguments will only rarely succeed. The previous section
included some cases that would also fall under this heading. National courts have
stated that a procedural defect or materially wrong decision is insufficient to
show justifiable doubts about impartiality or independence, unless the decision
is repeated or is particularly serious. (339) Likewise, the ICC has said that such
challenges are only upheld if the conduct is so manifestly improper as to raise
concerns about due process. (340) LCIA decisions are consistent with this. (341) A
recent Swiss case provides a good example. (342) The parties had agreed on
specific procedures with time deadlines. The claimant requested an extension of
a deadline which was denied. Later, the respondent was granted a time extension
for the submission of certain documents. The claimant argued this showed
partiality. The argument was rejected by the ICC Court and the Swiss Federal
Supreme Court. (343) However, the situation is different if the arbitrator's conduct
or errors show clear hostility towards one party. In a Singaporean arbitration,
the arbitrator wrote to one of the parties using language which was said to be
sarcastic almost to the point of hostility. The arbitrator also accused the party of
delaying tactics. This was sufficient to raise justifiable doubts about the
arbitrator's impartiality. (344)

5.9.10. Other Grounds for Challenge

It should be noted that the grounds for challenge are not necessarily limited to
lack of independence and impartiality. The ICC Rules, for example, say that
challenge can be for ‘lack of impartiality or independence or otherwise’. (345)
There are, of course, many factual bases upon which an arbitrator might be
challenged. For example, the arbitrator does not possess the qualifications
agreed to by the parties in their arbitration agreement. (346) Likewise, an
arbitrator being physically or page "309" mentally unable to conduct the
arbitration will justify a removal. (347) An arbitrator could also be challenged for
serious incompetence like regularly failing to attend hearings. In one recent ICC
case, the ICC refused to confirm a party's nomination because of the arbitrator's
performance in a previous case which was extensively delayed. (348) Some lex
arbitri specifically allow for removal of an arbitrator in circumstances of
excessive delay. (349) Others leave it to more general grounds such as Article
14(1) of the UNCITRAL Model Law. An arbitrator who wrongly claimed required
foreign language proficiency might also be a concern. The ICC recently refused to
confirm a party-nominated arbitrator who was not fluent in the languages
required by the arbitration agreement. (350)

5.10. Disclosure

Associated with the duties of independence and impartiality is an obligation to

disclose any information that may give rise to doubts about independence or
impartiality. The obligation arises under many national laws (351) and arbitral
rules. (352)

5.10.1. Disclosure Procedure

In practice, disclosure tends to occur informally when a prospective arbitrator is

first approached by a party. If selected, the arbitrator then discloses the
information to the other party and any institution. (353) It is typically accompanied
by a declaration by the arbitrator that he or she is independent and impartial and
is aware of no other facts which could call this into question. (354)

5.10.2. Threshold and Content of Disclosure

The threshold suggested by Redfern and Hunter is for disclosure ‘of all facts that
could reasonably be considered to be grounds for disqualification’. (355) In
addition page "310" to broad policy considerations, directions can be found in lex
arbitri or recommended guidelines. For example, Article 12(1) of the UNCITRAL
Model Law requires disclosure of facts that are ‘likely to give rise to justifiable
doubts …’. That is an awkward standard given the need to allow the parties
themselves to form a view. Taken literally, an arbitrator would only disclose in
those circumstances where the arbitrator felt that the grounds for
disqualification were made out. That should not be the proper interpretation.
Article 11(2) of the ICC Rules 2012 calls for disclosure of facts or circumstances
which ‘might’ call into question independence in the eyes of the parties or give
rise to reasonable doubts. (356) This is a preferable articulation of the standard.
The ICC, which requires arbitrators to sign a statement of independence and
disclose any relevant facts, makes the following recommendation: arbitrators
should take into account ‘whether there exists any past or present relationship,
direct or indirect, with any of the parties, their counsel, whether financial,
professional or of another kind, which might call into question the arbitrator's
independence in the eyes of the parties’. (357) The IBA Guidelines provide the
following standard: ‘if facts or circumstances exist that may, in the eyes of the
parties, give rise to doubts as to the arbitrator's impartiality or independence,
the arbitrator shall disclose such facts or circumstances’. (358) The broad test was
apparently devised at the request of numerous arbitral institutions. (359)

The following is a brief selection of some key examples in which the arbitrator's
lack of disclosure was deemed important in assessing a challenge. One caution is
that to some, non-disclosure should not itself be a cause for disqualification.
Hence cases are best indirect guides to matters that should have been disclosed.
As noted, a more extensive list of challenge cases appears in Appendix 1. The US
Supreme Court has suggested that arbitrators ‘should err on the side of
disclosure’ because ‘it is better that the relationship be disclosed at the outset
when the parties are free to reject the arbitrator or accept him with knowledge
of the relationship.’ (360) The US Supreme Court also noted, however, that “an
arbitrator's business relationships may be diverse, involving more or less remote
commercial connection with great numbers of people’ and as such, an arbitrator
‘cannot be expected to provide the parties with his complete and unexpurgated
business biography’, and is not required to disclose matters that are merely
‘trivial’. Some balance and judgment is required. The US court's suggestion to the
effect that arbitrators should err on the side of disclosure is certainly reasonable
when both page "311" parties choose to reject a particular arbitrator based on
the facts disclosed. More problematic is the case where a relatively minor matter
is disclosed and the other party chooses to use this as a pretext to challenge a
party-appointed arbitrator. Because this can be expected to occur whenever
there is a perceived tactical advantage in doing so, it is overly simplistic to
recommend as much disclosure as possible. While a prospective arbitrator is not
as yet constituted as sole or co-arbitrator, nevertheless it should not be
unreasonable for them to think about what is fair and reasonable disclosure and
the circumstances where an unduly cautious approach might simply frustrate
the entitlements of an appointing party and the ability to commence the
proceedings with appropriate speed.

However, the cautionary suggestion that the parties' rights to at least consider
the matter must be respected is a most sensible direction. The arbitrator should
not paternalistically decide for the parties but should give them an opportunity
to consider any matter that might be legitimately of concern to an independent
observer. The obligation to notify may be higher, the more serious the factors
under consideration. French courts have considered that the duty is to be
determined ‘with regard to the notoriety of the situation giving rise to the
challenge and to its reasonably foreseeable impact on the arbitrator's decision.’
(361) Broad disclosure at the earliest opportunity also has the advantage of

maintaining confidence in arbitrators. If undisclosed facts are found out later,

particularly by the losing party, that may raise more significant suspicions than if
there had been timely disclosure.

Arbitrators from different legal cultures may have different views on what is
disclosable. It appears that US-trained lawyers may see more matters as
disclosable than those trained in civil systems. (362) This was one of the concerns
that led to the publication of the IBA's Guidelines. Section 5.9.4 deals with the
IBA Guidelines which set out a Red, Orange and Green list as a guide to both
issues of concern and as to when disclosure would be required. Many would
argue that it should readily be accepted that there is no duty to disclose
circumstances contained in the Green list in the IBA Guidelines. Conversely, there
seems no justification not to disclose an Orange list circumstance even if it is
expected to be acceptable. That said, some institutions do not subscribe to the
IBA Guidelines and have asserted that they may improperly limit disclosure. (363)
Cárdenas and Rivkin rightly criticise this approach, noting that it is not enough to
demand comprehensive disclosure, as judgment must always be exercised at the
margins. (364) There remains a need to balance page "312" transparency and
challenge rights on the one hand, with appropriate limits on an ability to abuse
such processes for delay purposes.

The judgment that needs to be exercised on a case-by-case basis may need to

consider other rights and obligations. For example, a duty of disclosure may even
clash with duties of confidentiality if it relates to involvement in another case
where there may have been an arguable relationship with a party or a
predisposition on a particular issue. Nevertheless, it is difficult to prioritise
conflicting obligations from two separate arbitrations. In most such cases a
careful description of the nature of the previous event without identification of
the parties ought to be enough to meet both obligations concurrently. (365)

Another issue is whether duties of disclosure can be negated by actions of the

parties, for example, by waiver or acquiescence. That would depend on their
actual knowledge. The IBA Guidelines Waivable Red list contemplates informed
acceptance. If parties select an arbitrator with full knowledge of the relevant
issues, then they have waived and further disclosure seems pointless. That
would only be the case if both parties had the requisite knowledge. Conversely,
the parties could not be said to have waived their rights at an earlier point of
time when they were unaware of the details. The situation might be different if it
was a circumstance that the parties could on reasonable inquiry have identified
for themselves, particularly where there is some information in their possession
that would at least lead them to have an appropriate suspicion of a proscribed
relationship. That is a more problematic contention. Without such grounds for
suspicion, not all legal systems would oblige general inquiries by the parties and
many could legitimately suggest that only actual waiver rather than constructive
waiver should suffice. An arbitrator should certainly disclose where there is only
a belief that the parties could be aware of the linkage and no actual evidence that
this is actually known.
5.10.3. Consequences of Failure to Disclose Relevant Material

It has at times been suggested that a failure to disclose relevant material is, in
and of itself, evidence of partiality, even if the facts would not have led to a
challenge or refusal to appoint. (366) However, the IBA Guidelines express a
contrary position. (367) The better view would be to look at the facts on a case-by-
case basis. A conscious and unjustifiable failure to disclose key information
suggests lack of suitability in any event, even if not displaying partiality per se
but instead, a wish to take on the page "313" appointment unchallenged.
Conversely, a failure to know of a modest shareholding in a publicly listed
company via an independently run superannuation account does not readily lend
itself to assertions of partiality.

Article 4.1 of the IBA Rules of Ethics states:

Failure to make such … disclosure creates an appearance of bias, and may of

itself be a ground for disqualification even though the non-disclosed facts or
circumstances would not of themselves justify disqualification.

This language is merely permissive, although it allows for the paradigm case of
rejection for non-disclosure alone as articulated above. However, the subsequent
IBA Guidelines on Conflicts of Interest express the view that non-disclosure is a
separate issue to the question of whether the arbitrator lacks sufficient
independence and impartiality. (368) All would agree, however, that in some
circumstances non-disclosure may indeed be relevant to that question as it might
lead an independent observer to be concerned about partiality. For example, an
arbitrator who is habitually appointed as a representative of a party and failed to
disclose it would naturally lead a reasonable observer to worry about the
reasons for the nondisclosure and the reasons for the continuous appointment.

The stricter approach was adopted in Commonwealth Coatings Corp v Continental

Casualty Co, where a US court held that a failure to disclose constituted grounds
to disqualify an arbitrator. (369) Ball also speaks of a US-Iran Claims Tribunal case
in which an arbitrator withdrew due to failure to disclose that he was a director
of an affiliate company that supplied a key witness for one of the parties. (370)
That could be explained by the link itself. The key policy question is when, if
ever, a failure to disclose something which itself is not enough to ground a
challenge, becomes a valid ground because of non-disclosure alone. In the case of
AT&T Corporation and Lucent Technologies Inc v Saudi Cable Company, (371) the
English Court of Appeal found that an arbitrator's failure to disclose that he was
a non-executive director of one of the parties' competitors was not necessarily
evidence of ‘bias’ under the common law or ‘misconduct’ (for failure to comply
with ICC Rules). Rather, the court analysed these issues on their merits and
concluded there was no bias or misconduct. The Court emphasised that the
nondisclosure was innocent and thus could not, on its own, bring the arbitrator's
impartiality into question. In an SCC arbitration, (372) the Claimant-appointed
arbitrator inadvertently failed to disclose that a partner at his firm was
representing the claimant's parent company in ongoing litigation. The challenge
succeeded. Although the reasons for the decision were not published, given the
relative page "314" remoteness of the connection, it would seem that the lack of
disclosure was an important consideration. In another notable Swedish
arbitration, the arbitrator was removed for failing to disclose that he had
previously worked part-time as a consultant for the law firm representing one of
the parties. (373)

If a challenge is brought after the award is rendered, it may be sought to be

utilised as a ground for annulment or blocking of enforcement. It is not clear
whether a failure to disclose should be seen as grounds for setting aside the
award. Fouchard notes that a failure to disclose facts which may affect his or her
independence or impartiality may lead to problems in enforcement on the basis
of public policy. (374) An argument in favour would be that an arbitrator who has
failed to disclose proper information has prevented the parties exercising their
choice of arbitrator in a meaningful manner. A contrary argument is that guiding
principles such as the IBA Guidelines draw a distinction between factors actually
showing a lack of independence and impartiality and the mere failure to disclose
possible factors that might not ultimately support a finding of lack of
independence or impartiality. Because the duty to disclose is a broad one, aiming
to give the parties as much information as possible, it is readily contemplated
that there is a duty to disclose factors which ultimately may not prove to be
problematic. The very act of disclosing suggests that the arbitrator has formed a
view that appointment is not problematic. If the situation was thought to be
problematic, the appointment would not have been accepted. If the factors do
not show sufficient lack of impartiality or independence, then a failure to
disclose such a fact might be argued to be insufficient to justify annulment of an
award. The English case of AT&T mentioned above, (375) considered that lack of
disclosure was only one factor to consider. Poudret and Besson argue in favour
of the English approach which has also been followed in Switzerland. (376) The
Svea Court of Appeal has also held that an award should not be overturned
simply because of a failure to disclose in circumstances which should not have
diminished the confidence in the arbitrator's impartiality. The arbitrator had
failed to disclose that he had acted as an arbitrator in a dispute involving an
affiliate of a party and had acted as a representative for an affiliate in an
unrelated legal matter. (377)

As virtually all duties could also be seen to be implied terms of the contract
between the arbitrator and the parties, a failure to disclose could render an
arbitrator potentially liable for damages subject to any immunities that apply
and subject to the argument that non-disclosure predated the contract and must
be page "315" dealt with under other legal claims such as misrepresentation or
culpa in contrahendo. (378)

5.10.4. Ongoing Duties of Disclosure, Independence and Impartiality

The duty of disclosure is an ongoing one and any new facts that arise or old facts
that freshly come to light that may raise doubt about independence/impartiality
must be disclosed. (379) There is debate as to whether the standard for disclosure
should be lessened as the arbitration goes on, to factor in the inconvenience of
changing arbitrators late in the arbitration. (380) Some commentators suggest that
disclosure at the end of the arbitration should be judged at a weaker standard.
(381) The better view is that the cost/benefit determination in the event of a

challenge may vary but not the duty of disclosure. If the link may be of concern
but can be outweighed by efficiency issues, the latter should not be determined
in isolation by the tribunal. In multi-person tribunals, there is also a value in
having the particular arbitrator disclose to co-arbitrators and seek their views.

There is also debate as to whether obligations of independence and impartiality

should continue, not only for the full length of the arbitration, but also for the
whole period of time that the award can be challenged. The IBA Working Group
that developed the Guidelines considered this, but decided against it. (382) If a
matter was sent back to the tribunal for revision, reconsideration or addition, the
duty to disclose should then arise once more.

5.10.5. Duty to Inquire into Potential Conflicts?

As noted, there is reasonable consensus that there is a broad duty of disclosure

that requires an arbitrator to give the parties an opportunity to consider their
position. Less clear is the extent of research that an arbitrator should conduct in
order to determine whether any potential conflictual situations have arisen or
may arise. At page "316" the very least, an arbitrator should not be entitled to
refuse to investigate in order to be sheltered from knowledge of a potential
conflict. Such an approach would be contrary to IBA Guidelines General Standard
7(b).

Nevertheless there do need to be reasonable limits, which are hard to identify in

practice. Identifying the extent of any investigatory obligation is particularly
difficult in the modern corporate world with so many international businesses,
subsidiaries and holding companies in corporate groups, joint venture
arrangements, affiliations amongst law firms and differential ways of regulating
and commercially conducting legal practices such as barristers' chambers in
common law countries. There is also the need to consider lawyers that switch
firms as well as the firms themselves. There may be a need to consider people
who studied together, people who have co-written articles, or who socialise
regularly at arbitration conferences.

One approach is to simply say that broad disclosure obligations raise as a

corollary an obligation to make the widest searches to see if disclosure situations
have arisen. That would be highly problematic given the realities of the complex
modern world mentioned above and the significant delay and expense that
would be occasioned by a search for any possible situation worthy of
consideration. For many arbitrators, it will at times be impossible to
comprehensively check for connections. This is particularly so where a leading
arbitrator has moved from a position of partnership in a large law firm to an
individual arbitral practice. The previous firm would be under no obligation to
undertake a conflicts check on behalf of the arbitrator. In any event, once the
arbitrator has severed links with the firm, there seems no reasonable basis for a
challenge without further connecting factors. Because of such concerns, James
Carter argues forcefully that the arbitrator ought to have a right to limit the
efforts to a reasonable inquiry and a reasonable disclosure. (383)

Rule 5.1 of IBA Ethics Rules states:

When approached with a view to appointment, a prospective arbitrator should

make sufficient inquiries in order to inform himself whether there may be any
justifiable doubts regarding his impartiality or independence; whether he is
competent to determine the issues in dispute; and whether he is able to give the
arbitration the time and attention required.

While phrases such as ‘sufficient’ and ‘reasonable’ are difficult to define further,
it is unlikely that a more precise articulation is possible. It is more likely that a
case-by-case assessment of reasonableness will need to be made, subject to the
overriding view that an arbitrator should err on the side of caution and hence
take whatever investigative steps that might legitimately be considered
reasonable. (384)

page "317"

A related question is how much inquiry there needs to be as to the elements of

the case itself in order to determine whether this may throw up specific
questions as to impartiality or independence. It would seem unreasonable to
expect an arbitrator to inquire as to the names of all potential counsel and
witnesses and the names of key corporate officers of the party seeking to appoint
so that disclosure of relationships might be made. (385)

5.10.6. Timing of Challenge

While the criteria for impartiality and independence are the same at the
appointment and subsequent stages, in situations at the margin the final decision
may vary depending on the time at which the challenge is brought. It will often
be the case that it may be harder to challenge once the proceedings are on foot
simply because the benefits of bringing in a new arbitrator whose independence
and impartiality are above challenge, need to be weighed against the practical
problems of commencing again. Not only will there be costs and delays but there
will be issues as to whether certain matters need to be reheard and whether
matters already heard may be relevant evidence in subsequent proceedings. (386)
This would not be determinative in extreme cases, but most challenges are at the
margin and in grey areas. Often they are motivated by a wish that justice be seen
to be done. If that comes with a cost, it may be that it will at times not be
warranted. Timing also involves questions of late challenges and waiver,
discussed in the following section.

Craig, Park and Paulsson note that the ICC appears more likely to accept a
challenge before appointment than after appointment. The Court may be
suspicious that later challenges are more strategic than substantive. (387) Another
reason is that institutions may be risk averse at the early stage, preferring an
arbitrator beyond challenge to ensure smooth progress. One reason for
institutions to set higher standards for independence and impartiality when they
make appointments is that ideally, all appointments should be wholly above
reproach on this basis. Where party appointments are concerned, however, that
needs to be counterbalanced with the individual party's right to select their own
arbitrator. Hence there is a need for tests of independence and impartiality
which set evidentiary standards page "318" by which challenges can be resolved.
From a policy perspective, it is only when the concerns are of a sufficient level
that the parties' choice should be undermined. Where an institutional
appointment is concerned, direct party choice is irrelevant and the institution is
simply seeking to appoint someone above reproach who will not be viewed with
caution by one party alone.

5.11. Waiver of Challenge Rights

Previous sections touched on party agreement to accept otherwise debateable

circumstances. In that regard the IBA Guidelines distinguish between a waivable
and non-waivable red list, implying that all orange situations are waivable. (388)
This section addresses behaviour by an individual party that might waive
challenge rights.

Most legal systems take the view that undue delay in challenging an arbitrator
constitutes a form of waiver. The Model Law Article 13(2) states that objections
on independence/impartiality have to be made within fifteen days of
appointment/ becoming aware of the relevant facts. (389) Likewise, most arbitral
rules provide time limits for bringing challenges. (390) Considerations of waiver
more readily apply where a party has the relevant information at hand but
chooses not to act upon it. A related situation where legal systems sometimes
deny rights is where a party did not have the information but would have had it
on reasonable and timely inquiry. In most such cases, however, the parties ought
to be able to rely on the arbitrator's duty to disclose. There ought not to be an
overriding obligation to make inquiries to determine if the arbitrator is holding
something back. (391) The situation might be different if a party had a reasonable
suspicion but chose not to make the necessary inquiries.

One difficult question that may arise is whether there are certain matters that
simply cannot be waived. In one US case, it was held that waiver could extend to
page "319" known corruption by the tribunal. (392) There is a tension here
between the principles of party autonomy and arbitrator neutrality. (393) The IBA
Guidelines on Conflicts of Interest take the position that certain serious cases of
partiality and dependence cannot be waived. (394) These include where the
arbitrator has a significant financial interest in one of the parties or the outcome
of the case and where the arbitrator acts as the lawyer for one of the parties. A
difficult question is whether such principles should protect a party who ran the
risk of a favourable award in the face of something extreme, such as known
corruption, but then wishes to turn around and challenge an adverse decision.

5.12. Challenge Procedure

Challenge procedures vary across institutions and lex arbitri. Below is a brief
outline of various approaches. Key policy questions include the proper forum for
the challenge, whether the tribunal itself is the forum, if so whether the
challenged arbitrator may sit on the challenge and, generally, what standard of
proof, evidence and procedures should apply to challenges. National laws will
generally provide for challenge before courts, although if parties have selected
procedural rules under the lex arbitri entitlement, this should take precedence as
the particular forum for challenge is not itself a mandatory procedural norm.
Parties can thus effectively choose between court or institutional or tribunal
challenge. It is often said that challenge via an institution is preferable to court
challenge, because institutions have greater expertise and generally deal with
challenges more quickly than national courts. (395)

Challenge before the tribunal itself raises conflicting fairness and efficiency
questions. It will usually be quicker and cheaper to raise the matter with the
tribunal but there are issues in asking the impugned arbitrator who accepted the
appointment to effectively say that he or she was wrong to do so. Leaving it to
the other arbitrators is no solution with sole person tribunals and raises
problems where remaining arbitrators do not agree in multi-person tribunals.
There may also be cases where the remaining arbitrators are sufficiently close to
the impugned arbitrator that they are not independent or impartial on the
challenge question itself.

page "320"

5.12.1. Arbitral Rules

Institutional rules tend to adopt one of two approaches. Some provide for a
complaint to be made to the tribunal itself, with review rights in a national court.
Others provide for a complaint directly to the institution. (396) A number of
organisations use special appointees to deal with challenge decisions. These
include LCIA and CPR Institute for Dispute Resolution. Others such as the ICC
make determinations administratively. (397) Those that provide specialist bodies
are more inclined to provide written reasoned decisions. This is the case with
LCIA and CPR. This is discussed further below. Below is an outline of the various
procedures adopted by some of the major institutions and under the UNCITRAL
Rules. A representative sample of institutions has been selected and the process
is similar in other major institutions.

5.12.2. ICC

Challenges must be made within thirty days of appointment or on becoming

aware of the circumstances giving rise to the challenge. (398) Challenge must be in
writing specifying facts and circumstances giving rise to challenge. The
Secretariat then gives the arbitrator, the other party, and any other members of
the tribunal the right to comment within a suitable time. Any comments are
given to the arbitrators and the parties. (399) The ICC Court, in a plenary session,
then decides on the challenge. (400) Reasons are not published. (401) This is because
the decision on challenges is taken by vote in a plenary session. Thus the reasons
for any given vote cannot be reduced to a single set of reasons. (402)
Article 11(4) of the ICC Rules 2012 purports to make decisions of the
International Court of Arbitration final. Such an institutional rule cannot bar a
national court's jurisdiction if it purports to be mandatory, although acceptance
of such an arbitral rule may constitute a waiver of a review right in some legal
systems at least. That may depend upon whether the review rights are
considered to page "321" be mandatory. (403) Mandated lack of reasons makes it
harder for the aggrieved party to find grounds for challenge in any event.

5.12.3. LCIA

Written challenge must be made within fifteen days of establishment of the

tribunal or within fifteen days of becoming aware of the circumstances giving
rise to challenge. If the arbitrator does not withdraw, the LCIA makes a decision
within fifteen days. (404) The LCIA Court taking the decision is typically made up
of a three-person panel. If the parties request, an oral hearing can take place. (405)
Unlike the ICC, it publishes its reasons, although in a redacted form. (406)

5.12.4. SCC

Upon becoming aware of the circumstances giving rise to the challenge, a party
has fifteen days to submit a written statement, outlining the reasons for the
challenge to the SCC Secretariat. If this time limit expires, it is considered a
waiver. Both the challenged arbitrator and the other party are invited to
comment on the challenge. The SCC Board then makes a decision on the
challenge. (407)

5.12.5. SIAC

Parties have fourteen days to challenge an arbitrator after appointment or after

becoming aware of the relevant circumstances. This must be filed with the
Registrar and sent to the other party. The Registrar has the power to suspend
proceedings while the challenge is pending. A Committee of the SIAC Board then
makes a decision on the challenge. (408)

page "322"

5.12.6. UNCITRAL Rules

Challenge must be notified to other parties and all members of the tribunal. A
fifteen day time period to do this exists (either from appointment or awareness
of relevant facts). If the parties do not all agree on the challenge, or the arbitrator
does not withdraw, the challenge is decided on by the appointing authority. (409)
The time limit is applied strictly. (410)

5.12.7. ICSID

Challenge must be made ‘promptly, and in any event before the proceeding is
declared closed’. (411) If the circumstances giving rise to challenge are discovered
after close of proceedings, then the remedy lies in seeking annulment under rule
50. (412) This is still an effective challenge, although the criteria may differ. The
meaning of ‘promptly’ is flexible, but it is not without limits. In Suez and others v
Argentina, a fifty-two day delay after becoming aware of the circumstances
giving rise to the challenge was deemed too long. (413) The challenge must be
made in writing to the Secretary-General of ICSID. It is then passed on to the
other party and the tribunal. The challenged arbitrator can then provide any
explanations he or she wishes to make, ‘without delay’. Proceedings are
suspended while the challenge is pending. (414)

Precisely who makes the decision depends on the nature of the challenge. A
challenge to one arbitrator on a three-person panel is decided by the other two
arbitrators. If they cannot agree, the decision is made by the Chairman of the
Administrative Council. Challenges against a sole arbitrator or the majority of a
tribunal are also decided on by the Chairman of the Administrative Council. (415)
The decisions are usually published.

5.12.8. National Laws and Challenges before Courts

Most national laws allow challenges to be made either during the course of
proceedings or in an application to set aside an award. (416) The challenge may be
made page "323" directly after appointment, after the award or, in cases of
institutional arbitration, it may be possible to challenge the institution's ruling
on a challenge in court. (417) A few remarks should be made about court review of
a ruling on a challenge made by an institution or by some other procedure
agreed to by the parties. The availability of such review varies across
jurisdictions. In some national laws, the right to court review of an institutional
decision is regarded as mandatory and cannot be excluded by the parties in the
agreement. (418) In other jurisdictions, courts refuse to review such decisions, at
least directly. (419) In Switzerland, the Federal Supreme Court refuses to directly
review challenge decisions made by private bodies. However, it does hear such
objections as part of an application to set aside the award on the basis that the
tribunal was irregularly constituted. (420) Some lex arbitri impose limitations on
challenges to arbitrators. For example, some restrict appeals to situations where
the information has only recently come to light. Some laws also prevent any
appeal from the initial court ruling on the challenge and also allow the
arbitration to proceed while the challenge is pending. (421)

A brief comment should be made about the position in the US. The US Federal
Arbitration Act provides for no recourse to the courts to challenge an arbitrator's
appointment until after the award. (422) Hence a party may simply have to record
its objection and wait for the award to be rendered before it can act further. (423)
This does not, however, preclude the use of institutional rules, general contract
principles (424) and equity (425) to challenge an arbitrator prior to the page "324"
award. (426) Furthermore, the party does not necessarily have to wait until the
final award. Challenging a ‘final’ partial award can be used to remove the
arbitrator. (427)

5.12.9. The Provision of Reasons for Decisions on Challenges

Where written decisions are rendered, the next question is whether these are
published. As noted, the LCIA publishes redacted versions and the ICC does not.
On the one hand, publication can promote uniformity, provide guidance for
prospective nominees and also discourage institutions from being unduly
conservative. On the other hand, the reputational rights of challenged arbitrators
ought to be considered as well, particularly where an overly conservative
institution upholds challenges.

There is a range of conflicting policy issues arising from this. From the
arbitrator's perspective, a lack of reasons gives them no meaningful opportunity
to determine whether the decision was correct or not. In an extreme case, if a
challenge is improperly accepted, an arbitrator might argue a breach of contract
with the institution and/or the parties, although an institution would argue
implied rights even if these were not expressed. Furthermore, an arbitrator
might argue that a lack of reasoning fails to give guidance as to the difference
between acceptable and non-acceptable situations. (428) Finally, and perhaps most
significantly, if reasons need not be given, it is easier for an institution to prefer
its own interest to that of the arbitrator. For example, if the view was taken that
there is no real lack of independence or impartiality but circumstances make this
arguable, it is easier to reject an arbitrator when no reasons need to be provided.
By definition, if on balance an arbitrator ought not to be rejected, providing
reasons when rejection occurs nonetheless would be intellectually challenging.
As discussed above, the ICC defends its decision not to publish reasons, because
it is impossible to do so under its procedure of deciding challenges, where a vote
is taken in its plenary session. The ICC says that this procedure provides a
representative view of all its members. (429) It also argues page "325" that if the
decision were made by its Court, then it would appear to be exercising
jurisdictional powers. (430)

5.13. Agreement to Remove an Arbitrator by All Parties

Under many rule systems, parties can agree to dismiss an arbitrator. (431) Most
national laws allow for this. (432) No reason needs to be given for such a dismissal.
(433) Born notes that, in practice, this only tends to occur when the arbitrator

appears to be unable to act or at least unable to act efficiently. (434) Most

institutional rules are silent and some appear to deny parties the right to act
jointly to dismiss an arbitrator. It should be noted, by contrast, that a party does
not usually have the right to remove an arbitrator it has nominated. The only
option available to such a party is to follow the usual challenge procedure. This
may be restricted, as most institutional rules prevent parties from challenging
their own party-appointed arbitrator unless new information has come to light.
(435) This is another aspect of the waiver policy issues discussed above.

Even if the parties can remove an arbitrator, there are still contractual questions
and fee entitlements that may arise, although an arbitrator accepting
appointment under laws allowing such removal, would be subject to implied
terms in that regard unless there were contradictory clauses in the contract.
Even then, the procedural rights might prevail, with contractual issues dealt with
separately.
5.14. Resignation of an Arbitrator

5.14.1. Resignation in the Face of a Challenge

Some national laws and institutional rules permit an arbitrator to withdraw in

the face of a challenge, regardless of the view they take as to its merits. (436) While
rules page "326" may permit resignation in such circumstances, they do not
mandate it and an arbitrator is perfectly entitled to rely on the challenge
procedures to make an informed adjudicatory ruling that properly considers the
legitimate interests of all concerned. Given the arbitrator's duty to treat parties
equally and the duty to accept the mandate, the better view is that an arbitrator
should not withdraw automatically in the face of a challenge, even if this is
permitted under the relevant rules. It is far preferable that the arbitrator
exercise judgment. At the very least, spurious challenges should be resisted. (437)

Perhaps a more challenging question is whether an arbitrator with power to do

so, ought to resign if the view is taken that the challenge has tenable merit but on
balance is not the preferable view. On the one hand, the arbitrator might wish to
resign in any event so that justice can be seen to be done. On the other hand, the
arbitrator might take the view that the other party deserves maintenance of the
tribunal as selected in all cases where the challenge is not on balance to be
supported. A relevant factor in favour of the latter approach is also the time and
inconvenience caused by the need to find a replacement arbitrator. Such
problems are also exacerbated where there is a sole arbitrator. In multi-member
tribunals, it may often be possible to replace one arbitrator without the need to
backtrack in the procedure. (438) The competing factors become even more
complex if resignation leads to an effective entitlement to re-plead a case before
a new arbitrator. (439) This is discussed further in sections 5.15.3 and 5.16 below.

5.14.2. Resignation without Party Request or Consent

The above sections dealt with situations where a party wishes to remove an
arbitrator and the arbitrator may not want to resign. But an arbitrator may wish
to resign even though the parties want him or her to continue. This raises the
question: in what circumstances should an arbitrator be permitted to resign?
The matter is governed by the arbitrator's contract, the lex arbitri and the
procedural rules agreed to by the parties. (440) These rule systems attempt to deal
with the following policy objectives. On the one hand, arbitrators obviously
should not be compelled to continue where they have a reasonable excuse. On
the other hand, once a mandate is accepted, it should not be unilaterally rejected
without valid excuse. page "327" Furthermore, resignation of an arbitrator
disrupts proceedings and causes delay. (441) There is also a need to ensure that
unethical arbitrators do not resign for partisan reasons, (442) including a wish to
disrupt and delay.

5.14.3. The Right to Resign and the Duty to Complete the Mandate

In some jurisdictions court permission must be obtained if an arbitrator wishes

to resign contrary to the parties' wishes. (443) One Swiss court said the arbitrator
could be liable in damages for unjustified withdrawal. (444) Article 813 of the
Italian Code of Civil Procedure goes so far as to expressly indicate that
arbitrators are liable in damages if they renounce an appointment ‘without just
cause’. Born states that, whether or not express provisions exist in the governing
law or rules, an obligation to complete the mandate is implied by accepting the
appointment. (445) Resignation without good cause is a breach of that obligation.

National laws and institutional rules vary as to the preconditions for an

arbitrator to resign. They range from those which express no conditions (446) to
those which provide for a requirement for permission from the parties or an
independent authority such as a court. (447) The Model Law allows for resignation
of any arbitrator who becomes unable to perform his or her functions without
delay. (448) Model Law negotiators considered that there was little point in
demanding just cause for resignation given that one could not compel a reluctant
arbitrator. (449) That reasoning only deals with a lack of a remedy and does not
explain why the duty to complete a mandate without just cause for resignation
should not be expressly articulated. Many other aspects of optimal arbitral
procedure also have inadequate remedies for breaches.

Interestingly, the drafters of the 1976 UNCITRAL Rules rejected proposals

requiring a statement of reasons for resignation. (450) It seems hard to justify such
a position. Given the fact that an arbitrator has a duty to complete a mandate and
may page "328" only resign in certain circumstances, there seems to be little
justification in allowing the resignation without a commensurate obligation to
provide an explanation to the parties. Only if an explanation is given can the
parties determine whether the arbitrator was within rights or not in terms of the
resignation. It will undeniably be the case that in some circumstances great
sensitivity would be needed in drafting the reasons, particularly when these
include criticisms of the behaviour of the parties. Nevertheless, a professional
arbitrator needs to be accountable for their behaviour. (451)

5.14.4. Circumstances Justifying Resignation

Circumstances that obviously justify resignation would be matters outside of the

control of the arbitrator that, if known at the outset, would have meant that the
nomination ought not to have been accepted. The general category involves
changes in circumstances beyond the tribunal's control. In such circumstances
the tribunal member is simply saying that they no longer feel capable of
completing the mandate in a timely manner owing to these changed events.
Obvious examples include personal illness or serious family problems. (452) Most
rule systems give no guidance for less obvious cases. (453) Born lists a number of
circumstances where resignation might be permissible based on exceptional and
unforeseeable changes in the arbitration. These include:

1. Change in the Seat.

2. A serious and unforeseeable change in the workload.
3. Addition of new parties or claims materially altering the arbitration or which
give rise to potential conflicts.
4. Failure to cooperate with the arbitrator in an appropriate manner.
5. Evidence of misconduct. (454)

Another situation in which resignation appears permissible is where the parties

have suspended arbitral proceedings for a long period of time. The ICC has
accepted resignations in these circumstances, apparently on the basis that it is
unreasonable for arbitrators to be left ‘on hold’ for lengthy periods of time. (455)
page "329" In many cases, leading arbitrators will find the delayed proceedings
overlap with other legitimate commitments making one or other impossible to
perform. (456) The ICC has also accepted resignations in rare cases of threatening
or abusive conduct by one of the parties towards an arbitrator. (457) Further, it
appears that a resignation will be justified if it appears that an arbitrator is
unlikely to be paid. (458)

One key question is whether any changes in circumstances voluntarily agreed to

by the arbitrator could also justify resignation without liability for
inconvenience. This would be unlikely to be so where the change in
circumstances was fully voluntary. For example, if a tribunal member wished to
leave the arbitration profession and undertakes full time employment
somewhere, it would be preferable for the arbitrator to complete current
matters before embarking upon a new career. In some cases, however, the
starting date would be beyond the control of the arbitrator. An example would be
an arbitrator who was offered a prestigious Supreme Court or similar
appointment in their domestic jurisdiction. (459) Inevitably there will be some
grey area situations. For example, an arbitrator might be approached to become
a partner at a leading firm where that position would place him or her in a
conflict relationship with one of the parties to the arbitration. Even if the
arbitrator postpones the appointment, some might still see this as an improper
relationship. Furthermore, an arbitrator should not be expected to reject such a
career opportunity permanently, simply so as to avoid a potential conflict. The
facts should be disclosed but if the parties wished the arbitrator to proceed, this
should generally be respected.

If the arbitrator is a member of a particular firm that is contemplating a merger

which would place the arbitrator in a potentially conflicted situation, the
situation is more complex as the duty to complete the mandate cannot
necessarily override the arbitrator's general professional rights and
responsibilities vis-à-vis the firm, the latter which may be beyond the
arbitrator's control. A more contentious question is whether an arbitrator would
be entitled to resign based on a potential conflict, where the arbitrator's firm
wishes to do some new legal work for one of the parties or a related entity to one
of the parties. If the conflict simply arises from a general merger of law firms,
arbitral duties should not stand in the way of an entire firm's commercial
evolution. The situation is more complex when the firm simply wishes to engage
in some new work on a profitable basis and considers the new work more
valuable than the fees generated from the existing arbitral page "330" contract.
(460) The firm itself should be concerned about timing in such circumstances.
Even if resignation is permitted, care should be taken to ensure that the
procedure is as efficient and timely as possible in terms of the replacement.
There is also a need to ensure that due process is followed and the replacement
arbitrator has sufficient time and access to complete the new mandate
effectively. This will all depend on the procedures on replacement discussed in
section 5.15.3 below.

5.14.5. Consequences of Wrongful Resignation

If an arbitrator resigns without good cause, there is usually no basis for obliging
the arbitrator to continue to serve on the panel. (461) However, an arbitrator may
be liable for resulting damages. The likely damages would be for delay and
increased expenses. (462) A wrongful resignation may lead to liability
notwithstanding immunity provisions, (463) discussed in section 5.18, depending
on the way immunity provisions are drafted and interpreted and depending on
the applicable law.

5.15. How to Replace an Arbitrator

The key policy question is whether an arbitrator who is challenged successfully,

or who dies, becomes ill, resigns or is otherwise removed, needs to be replaced.
This is subject to party agreement. In most cases, it is also dependent on residual
norms in the lex arbitri and/or arbitral rules. The two options are to replace the
arbitrator or for the panel to proceed as a truncated tribunal.

5.15.1. Mechanisms for Replacing an Arbitrator

If one member of a three-member tribunal resigns or is removed, then he or she

must usually be replaced (with the exception of the possibility of a truncated
tribunal as discussed in section 5.16 below). (464) Most arbitral rules provide a
mechanism for replacing an arbitrator. Generally, the new appointment is made
in the page "331" same way as the original appointment. (465) ICSID Arbitration
Rule 11(2) is an exception, which provides for the Chairman of ICSID's
Administrative Council to nominate the replacement. This is designed to prevent
abuse of the challenge/ replacement procedure.

Redfern and Hunter suggest that problems are most likely to arise in
replacement in a purely ad hoc arbitration. If the parties cannot agree on a new
arbitrator or method of selection, then it may be necessary to go to a national
court, though this is rarely a timely solution. (466) Well crafted ad hoc rules will
avoid this, as for example, the UNCITRAL Rules which essentially provide that
replacement is to be made in the same way that the arbitrator was appointed.
(467) There are also cases which suggest that, even in the absence of such rules,

this is the preferred approach. Thus, a New York court held that, where the rules
and arbitration agreement were silent, a party-appointed arbitrator who
resigned was to be replaced by that same party, rather than the court. (468)

5.15.2. Insuring against a Vacancy

In the past, it was not uncommon to take out insurance against the lives of the
arbitrators to cover the costs of repeating oral hearings were an arbitrator to die.
Redfern and Hunter note that, despite possible financial advantages, a major
disadvantage is that this will often require the arbitrators to undergo a medical
examination (with the underlying implication that the arbitrator may not survive
the hearings!) (469)

5.15.3. Procedures after Replacement

5.15.3.1. Continuation or Recommencement of Proceedings?

A key issue is to what extent the tribunal has to ‘retrace its steps’. (470) Should oral
hearings begin again from the start? Redfern and Hunter suggest that this should
be avoided if possible. The new arbitrator can read transcripts and this should
normally suffice. If no transcript is available, then the matter must be discussed
by the parties and the tribunal and agreement needs to be sought. Parties will
page "332" normally, by that stage, be keen to proceed with minimum repetition,
so it is often up to the new arbitrator to decide how much repetition is necessary.
Redfern and Hunter say that it is hard to make a specific rule on what approach
the new arbitrator should adopt. (471) Transcripts may not help where the key
question is to determine which witnesses are truthful in circumstances where
transcripts of cross-examination do not allow for easy differentiation. Arbitral
rules typically provide a discretion. Article 15 of the UNCITRAL Rules 2010 state
that the arbitration should resume from the point at which the replaced
arbitrator left, but allows the tribunal to decide otherwise. (472)

5.15.3.2. Effect of Removal on Measures Already Taken by the Tribunal

When an arbitrator is removed, there may also be a question as to the status of

measures the tribunal had already taken before the removal of that arbitrator.
Procedural decisions and partial awards already handed down are of particular
importance. The first issue to determine in answering this is who should be
responsible for deciding on the status of such measures. Bedjaoui argues that the
matter should be determined by the tribunal (possibly now with a new
arbitrator) and not by a national court. (473) The second issue is whether those
measures should be invalidated because they were made by a tribunal
containing an arbitrator that has now been removed. Lalive, Poudret and
Reymond have argued that such measures should not be reviewed after removal
of an arbitrator. (474) Redfern and Hunter suggest it is good practice for the
replacement arbitrator to record in writing his or her assent to any procedural
directions that have been given prior to his or her appointment. (475)

5.16. Truncated Tribunals

If an arbitrator resigns late in proceedings, or refuses to participate in

deliberations or sign an award, replacing that arbitrator can be burdensome and
frustrating for the parties. One possible solution in this situation is for the
remaining two arbitrators to render the award as a ‘truncated’ tribunal, without
the other arbitrator's page "333" participation. A number of arbitrations have
proceeded in this manner, (476) but much debate exists as to when, if ever, a
truncated tribunal is valid.

5.16.1. Policy Arguments

There are policy concerns either way. Replacement involves delay and concerns
about the role of the new arbitrator vis-á-vis old evidence. A truncated tribunal
has a composition different from the initial choice and will often have a differing
dynamic as to methods of reaching majority views, given that a truncated
tribunal will often have an even number of arbitrators. A number of other
concerns have been raised about truncated tribunals. (477) First, it is generally
thought that, as an element of due process, the award should be made by the
whole tribunal. (478) The reasoning behind this is that the principle of equal
treatment of the parties may be violated where one party-appointed arbitrator is
not part of the proceedings. (479) This does not arise where it is the Chair that has
left, but in this case, there may be less chance of agreement between two party-
appointed arbitrators. Secondly, it has been argued that a truncated tribunal
award violates the parties' agreement as to a three-member tribunal. (480) There
is then a query as to whether this can give rise to concerns about enforceability.
Such an argument should not prevail where as must be the case, a truncated
tribunal can be traced to an agreement between the parties as to the Seat and/or
applicable arbitral rules. An exception would be where an institution or tribunal
chooses a Seat whose lex arbitri allows for truncated tribunals rather than the
parties themselves, although even here they must have consented to this
methodology and hence arguably to the outcome.

Against the concerns, there are obvious practical arguments in favour of

truncated tribunals. A late withdrawal causes inconvenience and delay. Beyond
this, page "334" late resignation can be a deliberate ploy by a partisan arbitrator
who, seeing that the award is likely to be unfavourable, seeks to obstruct
proceedings. For these reasons, many experienced arbitrators support truncated
tribunals. (481)

5.16.2. Circumstances Justifying a Truncated Tribunal

With these arguments in mind, in what circumstances will a truncated tribunal

be permitted? One situation is where the parties have agreed to procedural rules
which expressly permit truncated tribunals. Many rules appear to allow for it.
The LCIA and WIPO permit truncated tribunals, but leave a discretion to the
remaining arbitrators as to whether the arbitrator should be replaced instead.
(482) The ICC Rules 2012 Article 15(5) leaves discretion to its Court of Arbitration,

but requires it to take into account the views of the parties and remaining
arbitrators. (483) Most institutions follow this approach. (484) In a similar vein, the
2010 UNCITRAL Rules gives the appointing authority the discretion to authorise
a truncated tribunal, but only after the close of the hearing and where it has
decided that a party should be deprived of its right of re-appointment. (485) Born
notes that there is little judicial authority on the enforceability of awards made
under these rules, but concludes that, on balance, they should be enforceable.
(486)
Even without the lex arbitri or arbitral rules expressly permitting it, if an
arbitrator has a legitimate reason for being unable to continue, it may be that the
parties could agree to give the remaining arbitrators full powers of
determination from the evidence already presented. Where there is no such
express mandate, page "335" consideration would need to be given to the
presence of implied consent to ensure that a challenge might not be made under
Article V of the New York Convention on the basis that the award is outside of
the terms of the arbitration agreement, having been presented by a differently
constituted tribunal to that originally selected. (487)

One contentious situation is where the arbitrator has been part of proceedings
and deliberation but, at the last minute, refuses to sign the award. Awards by
truncated tribunals have been overturned by national courts in such
circumstances on the basis that the tribunal was not properly constituted in
accordance with the arbitration agreement. (488) In these cases, it did not appear
relevant that the resignation was a partisan abuse of process. Despite these
decisions, the prevailing view appears to be that a truncated tribunal is
permitted in such circumstances. (489) Some national laws now expressly
recognise the validity of a truncated tribunal. (490) It is submitted that, even where
the lex arbitri is silent, a truncated tribunal should be permitted where the
arbitrator has engaged in such a late unilateral withdrawal. There are a number
of arguments to support this view. Lew, Mistelis and Kröll note that, in such
circumstances, there can be no concerns about due process and equal treatment,
because the arbitrator has had every opportunity to participate in the rendering
of the award. (491) Furthermore, where the arbitral rules require resignation to be
approved by the institution or a court, a wrongful unilateral resignation may be
said to have no effect. Thus the tribunal is still, in a sense, properly constituted.
(492) But beyond these doctrinal and technical arguments, it simply seems

counter-intuitive to allow a partisan arbitrator to frustrate the whole arbitration

by resigning. (493) Holtzmann has written that it is hard to imagine many national
courts refusing to enforce an award on public policy grounds because of the late
resignation of an arbitrator or an arbitrator's refusal to sign the award. So long
as all three arbitrators have had the opportunity to participate, there should be
no problem of enforcement. (494) There is no concern in substance page "336" as
the remaining arbitrators would be of the same view, hence they form a majority
in any event. If that was not the case, the third arbitrator effectively has the
casting vote and there would be no reason to resign at the eleventh hour.

In the Himpurna case, such an abuse of process was not permitted. (495) In that
case, a co-arbitrator nominated by Indonesia resigned at a crucial late stage of
deliberations. The arbitral tribunal comprehensively set forth in its Final Award
the authorities in international law supporting the ability of the two remaining
members of the tribunal to proceed to a final award despite the inability of the
third arbitrator to participate. In particular, the tribunal relied upon the writings
of Judge Stephen Schwebel, former President of the International Court of Justice,
to the effect that: withdrawal of an arbitrator from an international arbitral
tribunal which is not approved or authorised by the tribunal is wrong under
customary international law and the general principles of law recognised and
applied in the practice of international arbitration. It generally will constitute a
violation of the treaty or contract constituting the tribunal, if not in relation to
express terms, then because the intention of the parties normally cannot be
deemed to have authorised such withdrawal. Interestingly, the Himpurna
tribunal suggested that proceeding as a truncated tribunal was not only its right,
but indeed its obligation. (496) A number of other cases are consistent with this
position. (497)

Redfern and Hunter wisely advise that, despite the above authorities, a decision
to proceed as a truncated arbitral tribunal should be taken with caution. (498)

5.17. Fees, Liability and Immunity

This book does not aim to comprehensively deal with all questions of arbitrator
fees, liability and immunity. Some commentary is included to the extent that this
can impact on procedural issues.

5.17.1. Fees

One of the fundamental rights of an arbitrator is entitlement to payment for

services, by way of fees and reimbursement of expenses. While that is easy to
state, page "337" the rights and obligations in that regard can emanate from a
range of sources. It can emanate from an agreement between the arbitrator and
the parties. An institution may be involved. There may be rules limiting the
amount of fees or the arbitrator's discretionary ability to make orders in that
regard as part of the award. There may be ethical issues given an arbitrator's
entitlement to promote their own commercial benefits, but their concurrent duty
to organise an efficient hearing, including ensuring reasonable costs. While an
arbitrator is entitled to receive fees and be reimbursed expenses under various
rules or via express agreement or implied terms in the contract with the parties,
some arbitrators are even bound to remit those fees to an employer that does
not allow for dual employment. An example would be judges in certain
jurisdictions who are barred from undertaking any other paid work besides their
judicial office. (499) A partner of a law firm may present the fees as firm billings. A
careful analysis of the arbitrator's entitlement to fees and expenses must thus
consider a range of other duties and legal constraints. The following sections
look at who may determine fees, methods of determining them, negotiations,
liability, types of fees and ramifications of failures to pay.

5.17.2. Who Determines Fees

There is an obvious conflict between a duty of an arbitrator to promote efficiency

and hence reduce total transaction costs of an arbitration and their own
entitlement to maximise the returns from their professional endeavours. In some
legal systems this means that arbitrators are not considered to be entitled to
determine their own fees. (500)

In an ad hoc arbitration, fees are determined by negotiation and agreement

between the arbitrator and the parties. Fees should be determined at the outset
otherwise there can be conflicts at a later stage. This does not mean an exact
figure needs to be identified but instead, an appropriate methodology agreed to
by all relevant parties. In such circumstances an arbitrator has particular duties
to act with the utmost good faith and where necessary educate the parties as to
accepted parameters. It is particularly important to have all parties present for
ad hoc discussions and do everything possible to ensure that one party is not
induced to try and curry favour by offering to support higher fees than the other.
(501) It would be particularly inappropriate to allow one party to contribute a

higher amount to support the aspirations of an ad hoc arbitrator.

Born suggests that, in principle, an arbitrator should not be allowed to

unilaterally and finally decide on a fee absent agreement about a method of
calculation as page "338" this would violate a principle that no one may be a
judge in their own cause. (502) One should certainly be concerned about any such
unilateral right but this scenario is unlikely to occur readily in practice. This is
because parties who select particular institutional or ad hoc rules and/or who
select the Seat and its lex arbitri have consented to whatever methodology of fee
determination is articulated. In such circumstances it would be wrong to
describe the determination as unilateral. Furthermore, an informed party would
naturally discuss fees when considering appointment. Finally, to suggest that the
problem is that an arbitrator should not be a judge in their own cause may not be
the most appropriate policy factor. A person is a judge in their own cause when
they have a binding adjudicatory right in the context of a conflict. If the rules
allow them to set a fee, then they are merely applying a power expressly granted.
If they are setting a reasonable fee based on an entitlement to quantum meruit or
similar principle, that is still subject to review by whatever adjudicatory body
can deal with contract disputes between the arbitrator and the parties. (503)

UNCITRAL Rules 2010 Article 41 allows the tribunal to fix ‘reasonable’ fees in the
context of the amount in dispute, complexity, time spent and other relevant
circumstances. The rule also directs arbitrators to take into account schedules of
fees of an appointing authority if relevant. In such circumstances, a party may
request the arbitrator to consult the appointing authority which may then
‘comment’ on the arbitrator's determination. The language suggests that the
arbitrator is not bound by either the schedule or the comments of an appointing
authority. (504) The ICC system takes a different approach, providing that the ICC
Court fixes fees based on similar factors as alluded to in relation to the
UNCITRAL Rules and subject to scale parameters. Where an institution
determines the final fees, that would then form the amount as specified in the
award. (505) The LCIA Rules provide for fees to be fixed by the LCIA Court, largely
based on an hourly rate system with various figures depending on the amount
involved. (506)

Some systems allow the tribunal to determine fees but provide for a supervisory
jurisdiction by a national court. (507) In such circumstances it is necessary to
consider whether the court's power is to determine fees de novo or only to
exercise a more restrictive standard of review to determine whether the fees as
charged were excessive. It is not clear whether a national court without such an
express mandate could still have a power to reduce the amount of arbitrator's
fees. That might be theoretically possible given that the contract would be
subject to some national law page "339" that may impose limits of
reasonableness on commercial fees. Even so, any arbitrator following an
institutional scale or international parameters for ad hoc fees should feel safe
from reasonable challenge. The situation will be different where there is a failure
to perform on the arbitrator's part. Refund of fees already paid may be one
contractual remedy available to the parties. (508) In some circumstances the
arbitrator might be setting the level of fees at the award stage. A question may
arise as to whether the principle of finality of awards precludes parties from
challenging such fees as set by the arbitrator. While an award is generally seen as
final, the better view is that this aspect cannot be sheltered from challenge for
that reason. (509)

The better view is that fee determinations by tribunals (as opposed to cost
allocations) are not awards and are not subject to award supervision processes.
(510)

5.17.3. Methods of Determining Fees and Expenses

Arbitral fees are generally either determined on a time basis or on the basis of a
formula aligned to the amounts claimed in the dispute. These are clearly the two
key models where institutional fees and expenses are determined. Of the two, the
‘time spent’ method is now the most common method used in international
arbitration. (511) In a wide-ranging survey conducted in 2000, Gotanda found that
the majority of arbitrators set their fees based on the amount of work performed.
(512) The same is likely in an ad hoc arbitration although it is of course possible

that there will simply be a negotiated flat fee without any discussion of a formula
underlying it. Contingency fees, based on the outcome in the dispute, would be
highly unethical and would compromise notions of impartiality.

From the perspective of economic incentives, any method of determining fees is

problematic unless arbitrators retain high standards of ethical responsibility and
ensure that they do all that is necessary and appropriate for a fair and efficient
resolution of the dispute. The problems with the two key methods of charging
are the same as the general problem for lawyers choosing between time-based
and value-based costing. Time-based costing provides no strong incentive for
expediency and indeed will often encourage lawyers to work a file more than
might actually be necessary. Further, the time-keeping can itself be time
consuming page "340" and is, in any case, not easy for parties to monitor. (513)
Conversely, value-based fees do not encourage thoroughness, as the less time
that is spent, the more the effective hourly rate for the lawyer. While an ad
valorem method imposes an upper limit on the arbitral costs which could be of
value to the parties, most institutional rules allow for representations to be made
to increase the fees in appropriate circumstances.

Some institutional rules provide a range of fees depending on the complexity of

the case. (514) While it makes sense to presume at the outset that cases may range
in complexity notwithstanding identical amounts in dispute, the likely corollary
of a fee range is an incentive for arbitrators to assert greater complexity.
Experienced institutions ought to be well able to review such assertions.
However, the institutions themselves are subject to conflicting economic
incentives. An institution would wish to be seen to be controlling fees and
offering efficient arbitral services. Hence their biases might be in favour of
rejecting legitimate assertions of atypical complexity. Conversely, institutions
would wish the leading arbitrators to be willing to accept nominations. If an
institution developed a reputation for parsimony, that would be
counterproductive. On the one hand, it might be suggested that arbitration relies
on the integrity of arbitrators and it is offensive to consider such hypothetical
economic incentives and disincentives as articulated above. On the other hand, it
would be foolish for practicing lawyers and managers of law firms to not
consider the economic incentives and disincentives in deciding how to run a
commercial practice. Similarly, arbitrators should not feel affronted if these
issues are raised for their consideration, at the very least to ensure that
subconscious incentives do not impact upon ethical demands.

Ideally, an arbitrator will consider these issues at the outset when considering
whether to accept appointment and raise issues of remuneration in a timely
manner if it is felt that the standard of remuneration as proposed will not allow
the thoroughness the tribunal believes to be the fairest in the circumstances. An
example is ICSID arbitration where the institution sets a maximum daily fee that
is significantly lower than the amount normally charged by the world's leading
arbitrators. (515) The ICSID Arbitration Rules allow for a higher fee by way of
consent of the parties, a matter which could thus be addressed at the outset.

In some cases it will not be possible to give adequate attention to all

remuneration issues from the outset. As noted above, an arbitrator might have
been approached to take on an ad hoc arbitration at a stipulated figure in total.
Commonly this will occur where parties are anxious to try and conduct their
arbitration as cheaply as possible. An arbitrator might agree to do so well before
claims, defences and cross-claims are completed and before the parties have
indicated page "341" their views as to their preferred length of oral hearings and
number of witnesses. It is conceivable that circumstances could arise where an
arbitrator has agreed on a modest fee, only to find that a very elaborate
arbitration is called for by the parties. In such circumstances fine judgment might
be needed. On the one hand, it could be argued that the parties have consciously
chosen to have a less elaborate arbitration, thus justifying the arbitrator
imposing procedural parameters within the figure struck. On the other hand, in
some circumstances at least, an arbitrator who committed to completing a
mandate under mandatory due process norms, might need to accept that in
hindsight, the figure was lower than it ought to have been, but the duty to afford
due process remains unqualified. In those circumstances a highly ethical
arbitrator would conduct the arbitration according to the justice of the
circumstances and would accept that he or she has had a lower remuneration
than in hindsight they would have wished.

It is also important to be very clear in the method of determining fees. For

example, if a daily fee is being paid rather than an hourly rate, it is necessary to
define how many hours constitute a day or whether a full daily rate is permitted
as long as the day was fully allocated, regardless of the number of hours in total.
If an actual fee has not been negotiated, then one would need to look to the
particular contract and what fees would be implied under the applicable law. All
legal systems would seek to equate to some determination of a reasonable fee to
be determined by evidence or would identify an independent assessor with
binding authority. (516)

5.17.4. Multi-member Tribunals

In an ad hoc multi-member tribunal there is a possibility that time-based fees

may vary between the tribunal members where they negotiate individual
contracts. This might annoy an arbitrator on the lesser rate. The fact that
different persons work at different speeds may add to the uncertainty as to the
total arbitral cost (517) and may be of concern to the parties at the end when there
is a significant difference between the fees of individual arbitrators. While
differential fees based on individual negotiations between a party and a party-
appointed arbitrator would seem undesirable, the practice is permitted subject
to party consent. (518)

5.17.5. Tribunal Chair Fees

A related question is what extra fees should be payable to the tribunal president
or chair. Because most arbitral systems empower the president or chair to make
page "342" procedural rulings and provide a first draft of the award, some
differential in fees would be appropriate. This is easier to deal with under time-
based costing where each person simply determines what work they do. Under a
fee schedule which looks to the amount in dispute, some specific allocation must
be made to cover the extra work of the president or chair. The norm with ICC
arbitration is to is to award the chairperson 40% of the total, with 30% to each of
two co-arbitrators. (519)

5.17.6. Negotiation of Fees between Arbitrators and Parties

Where institutional arbitration is concerned, the fees are generally set by the
institution itself and are published with the Rules. This not only adds certainty
and independent control over the total fees but also obviates the need for a
negotiation between the arbitrators and the parties. Given that an arbitrator
once appointed has a duty to complete the mandate expeditiously and efficiently,
there is some tension if at the start, an arbitrator might be seeking to negotiate
the best remuneration possible.

Where an arbitrator negotiates directly with the parties the potential for
conflicts of interests is readily apparent. This was touched on above. A desirable
ethical model is well set out in AAA/ABA Code of Ethics, Canon VIIB(1) and (3).
This state:

“B. Certain practices relating to payments are generally recognised as tending to

preserve the integrity and fairness of the arbitration process. These practices
include:
(1) Before the arbitrator finally accepts appointment, the basis of payment,
including any cancellation fee, compensation in the event of withdrawal and
compensation for study and preparation time, and all other charges, should
be established. Except for arrangements for the compensation of party-
appointed arbitrators, all parties should be informed in writing of the terms
established…
(3) Arbitrators should not, absent extraordinary circumstances, request
increases in the basis of their compensation during the course of a
proceeding.”

The latter is further discussed in the following section. Where negotiation occurs
directly, if an arbitrator believes that extra fees would be appropriate, then the
parties might be approached with a view to agreeing on higher remuneration. As
noted above, Lew, Mistelis and Kröll argue persuasively that it should not be
possible for one party alone to agree to the higher fees as this would allow for an
appearance of partiality. (520)

page "343"

Aksen believes that, in ad hoc arbitrations where there is no institution to set

fees, fees should be discussed earlier rather than later. (521) Failure to do this ‘has
shipwrecked arbitrations and needlessly tarnished reputations.’ (522) In providing
guidance on negotiating with arbitrators over fees, Redfern and Hunter note the
concern that parties may not wish to offend the arbitrators and thus jeopardise
their case. They advise parties that there is no reason not to negotiate a lower
fee, provided it is done with courtesy. It is best, they suggest, if all parties
negotiate with the arbitrators as a united front. (523)

5.17.7. Renegotiation of Fees

While it is natural to ask for fees to be determined at the outset to avoid later
conflicts of interest, there still needs to be an opportunity for renegotiation when
significantly changed circumstances arise, particularly where these are caused
by the parties and not the arbitrator. However, it has been held in the US that it is
contrary to public policy to seek additional compensation after appointment. The
concern is that parties should not be put in a position where they feel compelled
to accede to demands for fear of adverse consequences. (524)

5.17.8. Joint Liability of Parties for Fees and Expenses

In addition to the setting of the fees themselves, there is also the determination
of what proportion they are to be paid in. There is a need to distinguish between
the initial liability of the parties for fees and expenses and the ultimate position
after the rendering of the award. The starting position is that the parties are
jointly liable for an arbitrator's fees and expenses. A party who is forced to pay
the other party's share of the fees will typically have a right of recourse for the
money so paid, although timing and method of recourse may depend on
applicable rules and the arbitration agreement. Section 6.11 looks at the proper
procedural response where a party fails to provide its share of the fees. The base
situation of joint liability may be varied at the time of the final award if the
arbitrator determines that one party deserves compensation for the fees and
expenses as already paid.

page "344"

5.17.9. Advances of Fees and Expenses

In institutional arbitration, the institution will typically call for advances of fees
and expenses at various stages to ensure that there is no need for debt recovery
by the institution and/or arbitrators. (525) Experienced arbitrators will typically
take a similar approach in ad hoc arbitrations. (526) While that is the norm,
institutions rarely ensure that advances on fees and expenses are at a sufficient
level and rate to always ensure that the moneys at hand are in advance of the
fees and expenses actually incurred. Indeed that is only a meaningful notion
under time costing. Where ad valorem fees are concerned, these are based on a
presumption of completion of the mandate. They may also depend on an
institutional determination as to where within the parameters, the case sits in
terms of complexity.

In an ad hoc arbitration, one practical suggestion made by Redfern and Hunter is

to open a separate bank account for payment of arbitrator's fees and expenses.
The money is then paid to the arbitrators as and when it falls due. The account is
typically opened in the name of the arbitral tribunal, with the presiding
arbitrator as signatory. This is preferable to the money being directly paid to the
arbitrators. Any interest earned on the bank account is properly that of the
parties. (527) Some institutions will also handle funds for ad hoc arbitration in
return for a fee.

5.17.10. Expenses

In addition to fees, parties and arbitrators need to work out how to pay for the
arbitrators' expenses. A number of issues need to be agreed upon early. The first
is what class of air, train or similar travel will be covered. Agreeing on this early
can save embarrassment later. Secondly, it should be confirmed that these costs
will be reimbursed immediately. Redfern and Hunter note that there is no
justification for requiring the arbitrators to fund the cash flow of the arbitration.
(528) Consumption tax liability also needs to be clarified.

Another issue is how the arbitrators' subsistence expenses are to be covered.

Various methods are available. The reimbursement method means that
arbitrators keep a detailed record of expenses which are reimbursed from time
to time. A limit is sometimes imposed on, say, the amount of hotel expenses. (529)
Under the per diem method, arbitrators are paid a fixed daily amount to cover
their expenses. The ICC uses this method. Redfern and Hunter believe that this is
the preferable method. It avoids the detailed record-keeping and prying into the
private life of the arbitrator to decide whether a given expense was appropriate.
(530) As long as the per diem is page "345" reasonable, it also reduces the costs of
the record-keeping and accounting themselves. There can also be variations on
these two methods. ICSID, for example, provides a combination of the two. (531)

5.17.11. Failure to Pay Fees

While it would be clear that a failure to pay fees would be a breach on the part of
the parties, a question arises whether an arbitrator's duty to complete the
mandate expeditiously is subject to a right to withhold services until the relevant
fees are paid. Some lex arbitri expressly allow an arbitrator to withhold the
award in such circumstances. (532) Under common law, an arbitrator is thought to
have a lien over the award in relation to payment of fees. Such a lien will not be
very beneficial if the parties believe the respondent is likely to succeed. The
claimant may not wish to pay the fees to pursue an unsuccessful application. The
respondent has no need to do so as there is no adverse order to be concerned
with. (533)

There are policy arguments either way. In an ad hoc arbitration with no specific
direction in any rules adopted, the argument in favour is that from a practical
point of view, withholding the award may be the best means to encourage
payment of fees by a reluctant party. The situation might be more problematic in
an ad hoc arbitration if a party asserted that an arbitrator's demands for
prepayment were excessive. That is unlikely to be a concern where an institution
is involved and has a clear and consistent formula for timing and level of
payments. There are a number of contrary policy arguments. First, the reluctant
party may know that it is likely to lose, hence a withholding of an award is
actually in their best interests. Second, some might argue that the duty to
complete one's mandate is separate to one's entitlement to sue for fees.

The ramifications may also vary depending on whether it is the claimant or

respondent that fails to pay its fees. Unlike many domestic litigation systems,
international arbitrators do not have a general power to make summary
decisions on the merits simply as a result of one party failing to take part in the
process. (534) Thus where a respondent fails to pay its fees and indeed fails to take
part in the proceedings in any manner, the claimant must still prove its case. In
such circumstances it may also need to pay the fees of the respondent to allow
the hearing to proceed. Because of the policy concerns, the innocent party is
commonly asked to pay the share of the party in breach to allow the arbitration
to continue, with the commensurate right to add this to damages in due course
or allow a cross claim for page "346" failure to pay agreed amounts. Whether and
when this is possible is discussed in section 6.11.

The situation may be different where the failure is on the part of the claimant.
This is simply because the person bringing the claim must take appropriate steps
from time to time in that regard, otherwise a view can be taken that the claim is
withdrawn or abandoned. That might be said to arise with failure to comply with
any procedural orders, including a failure to pay a share of fees. For example,
Article 26.5 of the SIAC Rules 2007 indicates that if either party fails to make the
advances or deposits as directed ‘the Tribunal may, following consultation with
the Registrar, refuse to hear the claims or counterclaims, whichever is applicable,
by the non-complying party, although it may proceed to determine the claims or
counterclaims by any party who has complied with orders.’ In this sense a
respondent bringing a counterclaim is effectively in the same position as the
primary claimant in terms of its own claims.

5.17.12. Cancellation Fees

The ultimate aim of any dispute resolution system is the amicable settlement of
disputes. The ideal outcome is for a mutually agreeable solution obviating the
need for a judgment or award. Indeed in most domestic litigation systems, the
bulk of cases are settled in this way. This appears to be the trend in arbitration.
From the perspective of the domestic judge, settlement is also an ideal
circumstance, alleviating total workload, helping with a backlog of cases and not
interfering with the judge's remuneration as he or she is a paid public servant on
an annual salary.

While international arbitration should also aspire to mutually agreeable

solutions, the impact of a settlement on an arbitrator's schedule and
remuneration is more problematic. Leading arbitrators are often booked well in
advance for arbitral work and will block their diaries to accommodate agreed
hearing schedules. If the parties settle at the eleventh hour, it is usually not
possible for the arbitrator to fill the gap with another case as might be possible
with domestic litigation where it is common to set a number of cases down on
one day on the basis that many will settle and hence not clash. This is
particularly problematic where an arbitrator has agreed to a daily or hourly fee.
One solution is to provide for cancellation fees in the contract between the
arbitrator and the parties. Similar fees can be described as commitment fees,
where a tribunal is booked and entitled to remuneration whether the hearing
proceeds or not. (535) The Chartered Institute of Arbitrators describes this as ‘a
non-returnable booking fee’. (536)

page "347"

One issue is whether cancellation or commitment fees may be argued to be a

disincentive to settlement. It is difficult to determine whether they would also
increase the costs of an arbitration. Absent such an entitlement, a tribunal might
simply raise its fees to account for the number of cancellations it expects in a
given year. (537) Because arbitration fees are a relatively small percentage of total
costs, cancellation fees are unlikely to be a significant disincentive to settlement.
While these should thus be seen as perfectly acceptable, care should be taken to
ensure that a cancellation fee is not excessive, as a domestic court might treat it
as an unacceptable penalty or abuse of an arbitrator's position, at least under
some contract systems. (538) Certain laws preventing liquidated damages may also
be relevant. (539) Liquidated damages are generally acceptable under common law
systems while penalty clauses are not. A cancellation fee that is a percentage of a
reasonable fee if the hearing proceeded, could be argued to be a fair measure of
liquidated damages in any event given the difficulty in getting a replacement
arbitration.
While it seems fair for an arbitrator to offer to undertake an arbitration subject
to entitlement to a cancellation fee, an arbitrator would also be acting
improperly if pressure was placed on an unwilling party to agree. For example, in
ICT Pty Ltd v Sea Containers Ltd an arbitrator was removed because of an
appearance of bias when a reluctant party was being pressured into acceptance
of a cancellation fee. (540) Norjarl involved a request by arbitrators to receive a
commitment fee in advance of the hearing. (541) The appointment had been
accepted without such a stipulation. One party sought a declaration that the
arbitrators were entitled to conclude an agreement with just one of them as to
such fees. The other party sought the removal of the arbitrators for misconduct
on such basis. The court concluded that the arbitrators must proceed if there is
no agreement as to a commitment fee but that it was not misconduct to seek
such a fee as long as it was agreed by both parties.

It would be difficult to argue that the cancellation fee was an implied entitlement
even where not distinctly negotiated. (542) Any agreement as to a cancellation or
commitment fee should occur prior to acceptance of appointment. (543) An page
"348" association's standard terms might also cover the question of a
commitment fee. (544) While many institutional rules do not expressly deal with
cancellation or commitment fees, broad rules ought to be interpreted to allow
these to be agreed.

5.17.13. Entitlement to Fees after Resignation or Removal

There may be questions as to entitlement to both unpaid historical fees and

future fees where a tribunal member resigns or is removed from office. This
would generally depend on the contract between the party and arbitrators,
perhaps impacted upon by specific provisions in the procedural rules. Where
express or implied rights under contract are concerned, much would depend on
whether the resignation was valid or not. If not, an arbitrator would be in breach
of contract and potentially be liable to damages subject to any immunity. If the
resignation is in breach, the arbitrator might even have lost the right to historical
fees and would certainly have no right to future fees. If an arbitrator has been
removed for lack of ongoing impartiality and/or independence, entitlements to
fees for the work to date may depend on the circumstances underlying the
challenge and whether the arbitrator's behaviour could be said to be culpable
and a breach of the original agreement to arbitrate.

The situation may be more complex if the parties simply agree to remove a
tribunal member under a broad autonomy power. In such circumstances it may
even be arguable that given the tribunal's duty to complete a mandate and its
right to expect parties to support this, removal by agreement could give rise to
damages that would in substance look like some implied cancellation fee
commensurate with the balance of the fees as yet unearned. The contrary
argument would be that where the power to terminate by agreement is
expressed or implied in the lex arbitri or institutional or ad hoc rules, all parties
have consented to the arbitration contract on the implied understanding that
various powers might be exercised from time to time. In many cases, these
questions must be resolved by first principles or from the basis of general
contractual norms. In other cases some direction and support is given by
applicable lex arbitri. For example, the Arbitration Act 1996 (UK) allows an
arbitrator to apply to court for a ruling on entitlement to fees and expenses
where the parties are not able to agree. (545) Another question is whether the
remaining arbitrators are entitled to their full fees in default of continuance by a
co-arbitrator. The better view is that this should be so.

page "349"

5.17.14. Excessive Fees

There are two questions in considering the notion of excessive fees. The first is
how one determines whether there have in fact been excessive fees charged and
secondly what implications flow as a result. As to the first, in the absence of any
clear stipulations in the lex arbitri or rules, arguments as to whether fees are
excessive may depend upon applicable principles of contractual law. This section
does not seek to survey possible provisions of that nature, although arbitrators
and parties should be aware that such arguments might arise. In some cases the
rules provide a mechanism for challenge. (546)

As to the ramifications where excessive fees are found to have been charged,
even if it is clear that an arbitrator has charged excessive fees, this is not a
ground for challenging the award itself. The excess charge is a breach of the
contract between the arbitrator and the parties, and not a fatal flaw in the award.
While that ought to be the proper position in relation to the award as a whole,
the same does not hold where part of the award is the apportionment of such
excess fees. Such a determination is part of an award and ought to be
challengeable on any valid grounds. (547)

5.18. Liability and Immunity

It is generally agreed that, as an arbitrator is acting in a quasi-judicial function,

he or she has some degree of immunity from civil liability. The rationale is the
preservation of independence and integrity – the arbitrator can decide free from
fear. The immunity also stems from the parties' having given the arbitrator the
power to decide the dispute between them (although most lex arbitri provide for
immunity even in the absence of party agreement). Immunity cannot be
contracted out by the parties alone. (548) It is important to consider liability and
immunity concurrently. Poudret and Besson rightly point out that one should
first consider the basis of an obligation and then consider any exclusionary
provisions that undermine it. (549)

5.18.1. Policy Issues

The fact that there may be different policy views as to the desirability of liability
of arbitrators explains why different national arbitral laws have differing
approaches. page "350" While we tend to think of immunity in the sense of a bar
to a damages claim, another aspect of immunity is whether an arbitrator is
immune from testifying in annulment or enforcement proceedings.
There are no uniform laws applicable worldwide as to the liability or immunity
of arbitrators. There are two main reasons for this. The first is that there are
significant policy arguments for and against imposing either liability or
immunity. A second and related reason that needs reiteration here is that
different legal systems do not have a common view as to the essential nature of
the arbitral function. Some legal systems lean more towards seeing it as being
essentially a creature of contract. Others see the arbitrator's role as essentially
one of status as an independent adjudicator. The latter approach is more likely to
favour immunity.

It is useful to canvass policy arguments for and against judicial immunity and
then consider whether arbitration raises different considerations in whole or in
part. On the one hand, most legal systems assert the need to provide for
immunity of adjudicators to encourage fearless and robust decision making. On
the other hand, economic theory would suggest that without any sanctions over
professional conduct, there may be insufficient incentive to optimise behaviour
from both fairness and efficiency perspectives and also to ensure that conflicts of
interest do not arise. That is not unassailably so, as economic theory recognises
that there are a range of incentives beyond financial ones. These would include
moral and social incentives. An arbitrator's own ethical values may lead to
optimal behaviour. The respect or otherwise of peers and the social community
in which arbitration is conducted would also be a strong incentive to high quality
behaviour. The argument as to promoting fearless and robust decision making
would appear to be as applicable to arbitrators as to judges. It is also thought
that being under constant threat of litigation would not promote better
behaviour by arbitrators but would more likely produce undesirable outcomes
such as excessive defensiveness and hence delay and would be an incentive to
guerrilla tactics by counsel. (550) Those supporting immunity also argue that this
best supports the ideal of finality of arbitral awards. If a different outcome can be
achieved by suing an arbitrator, this encourages ongoing disputes. Allowing a
party to argue that a different outcome would have arisen if the arbitrator
behaved in a different way effectively operates contrary to this duty of finality.
(551) If the process is inadequate, then annulment or challenges to enforcement

are the appropriate remedy. The finality argument is complex given that
challenges to the award itself are perfectly acceptable.

One situation where a damages claim would not breach finality is where the
negligent behaviour was one which allowed a jurisdictional time limit to expire.
For example, if an arbitrator was only empowered to adjudicate if he or she did
so within a specified time after a notice of arbitration, if the arbitrator fails to do
so and denies the claimant due process, it makes no sense to justify an immunity
on the page "351" basis of finality of a decision never properly taken. Nor can
annulment help in such circumstances. Born suggests that the lack of a
comprehensive appeal mechanism in arbitration provides another reason to
grant an immunity lest an arbitrator be irreversibly influenced by open or
implicit threats as to liability. (552) While that is a valid observation, the converse
is also arguable. In the absence of an appropriate appeal mechanism, increasing
accountability of arbitrators may better induce optimal behaviour. The situation
will of course vary depending on whether the threats of litigation are based on
valid concerns or whether they are improper tactical exercises to shift an
arbitrator's properly preferred position. For this reason, a blanket rule either
way is unlikely to be optimal in all circumstances. Born also argues that
immunity should apply to situations of gross negligence as the threats to
independence and impartiality arise with almost equal force in such
circumstances. (553) That, of course, may depend on what one defines as gross
negligence. When that borders on conscious recklessness, the policy justification
is diminished. Even where that is not the case, if gross negligence is only used to
describe the most unmeritorious circumstances of unprofessional behaviour, it is
hard to see why these need to be protected so as to promote independence and
impartiality. Once again, the conclusion one might be drawn to could well
depend on whether the allegation is valid or not. One ought to have little
sympathy for an arbitrator who all would agree has acted in a grossly negligent
fashion. However, one can well understand Born's concern when the stakes are
high and counsel sees tactical value in flagging unmeritorious claims of gross
negligence with commensurately gross assertions as to damages.

Poudret and Besson also suggest that there ought to be a distinction made
between actions of a juridical nature and those of a more administrative nature.
As noted above, the policy arguments in favour of immunity in respect of the
former category break down somewhat if an arbitrator simply fails to meet
simple directions such as rendering an award within a specified timeline,
providing an award which meets form requirements, completing a mandate or
undertaking appropriate disclosure. (554) Consent/jurisdiction issues can be
raised in this debate as well. On the one hand, because arbitrators are in a
contractual relationship with the parties and are promising to complete their
mandate and satisfy certain duties, one can naturally envisage that in
appropriate circumstances, the parties might seek damages from an arbitrator
for breach of the duties. From their perspective, that would be their natural right
under their contract entitlements. On the other hand, an arbitral position is not
merely a commercial contractual arrangement. Sovereign nations cede authority
to arbitrators to perform an essential judicial function. Where domestic judges
are concerned, even though they must have the highest standards of professional
behaviour, it is considered better to provide them with page "352" immunity
from suit to promote dispassionate and fearless employment of their judicial
functions rather than to use the threat of litigation as an incentive to more
professional behaviour.

It is also important to consider the commercial implications of making

arbitrators readily liable to suit. One court has noted that this could reduce the
availability of suitably skilled arbitrators. (555) In turn this is argued to be
inefficient given the other advantages of arbitration, in particular worldwide
enforceability through the New York Convention. Such a disincentive could
either deter sufficient numbers, thus undermining arbitration as a whole, or
would be a stronger deterrent on the best persons remaining in the profession,
thus reducing the quality of arbitrators. In either event this would encourage
disputants to revert to litigation with commensurate strains on the court system.
In an international environment where there are problems of enforceability of
judgments, clashes in jurisdiction and differential interpretations of such
concepts as forum non conveniens, that has other negative transaction cost
effects.

Nevertheless, some are dubious from an economic perspective as to whether

open liability would be a disincentive to persons taking on arbitration duties.
The arguments in favour of immunity to the effect that without such a provision
it might be hard to find suitable arbitrators is an argument that seems
unsustainable under the Coase theorem. (556) The theorem suggests that informed
parties will simply bargain around whatever liability rule society imposes. Such
negotiations will aspire to the most efficient allocation of risk regardless of the
initial scenario set by the applicable law. This would suggest that arbitrators
without immunity would simply take out appropriate insurance and build it into
their assessment of a required fee. (557) While these factors seem valid, care
should be taken with too ready an analogy to judges. For example, even though
judges have immunity from suit by the litigants, there is government control and
sanctions over improper behaviour. While judges generally have independence,
it is possible to take action against those that fail to act diligently, at the very
least redirecting them to more menial cases.

An added complication where arbitration is concerned is that there are both

direct and indirect sanctions in relation to arbitration behaviour. There is the
possibility of the parties agreeing to revoke an arbitrator's mandate, subject to
an obligation to pay compensation, although even this might be debateable
depending on whether the revocation is based on alleged loss-causing behaviour
by the arbitrator. Similarly, an institution may revoke an arbitrator's mandate,
for example, for failing to efficiently pursue the task. Applications might be made
to a supervisory court in the interim stages or via an application to annul the
award. Enforcement of the award may be blocked on the basis of arbitrator
behaviour. Finally there are the more direct actions that might be brought by one
or other of the parties page "353" against the arbitrator in a personal capacity
and in respect of which immunity and liability is generally discussed. Such an
action may be impacted upon by provisions in the lex arbitri, in institutional
rules or exempting provisions in contracts between the arbitrator and the
parties. Where the latter are concerned there is a need to consider the validity
and interpretation of such contractual provisions. That is itself a conflicts of law
question requiring consideration to be given to the applicable law. There is then
a need to see whether any mandatory norms of the relevant jurisdiction apply to
protect the person that is consuming these professional services from such
exclusion of liability provisions.

In addition, arbitrators are held accountable in different ways to judges simply

because they are selected. If an arbitrator behaved poorly in a particular case,
that may become part of institutional knowledge and impact adversely on their
likelihood of selection in future cases. Because of the broad range of responses,
there may be even less of an argument in favour of direct liability as a means to
promote optimal behaviour for arbitrators as opposed to judges. From a policy
perspective it is also necessary to consider the advantages and disadvantages of
the alternative solution. If there is no recourse as against an arbitrator, then
there is a greater incentive to challenge the award. Poudret and Besson cite
Lalive, Poudret and Raymond (558) in suggesting that because applications to set
aside awards and liability suits against arbitrators seek to protect different
interests, the latter should not be allowed to operate as an indirect challenge to
the award. (559) They suggest consequentially that there should be restrictions on
liability claims based on jurisdictional determinations.

5.18.2. Arbitral Laws and Rules as to Liability and Immunity

Arbitral rules and statutes rarely provide expressly for liability for damages by
an arbitrator. (560) Instead, most direct attention to immunity, impliedly
identifying a residual liability which would be based on contract and perhaps
tortious principles. (561) Express limitations on liability will invariably contain
exceptions in relation to fraud, bad faith or deliberate wrongdoing.

The UNCITRAL Model Law does not deal with the question of arbitrator liability
and immunity. The Arbitration Act 1996 (UK) provides an indemnity unless the
act or omission is ‘in bad faith’. (562) Furthermore, section 4(1) indicates page
"354" that this provision is mandatory. Section 29(3) excludes from the
immunity granted, situations of resignation of an arbitrator. If the parties cannot
agree on the implications, an application is then made to the court under section
25. The indemnity provided under the Arbitration Act 1996 (UK) is to be
contrasted with express liability imposed under the Austrian Code of Civil
Procedure. Article 584(2) indicates that:

An arbitrator who does not fulfil in time or at all the obligations assumed by his
acceptance of office is liable to the parties for all the loss caused by his wrongful
refusal or delay, without prejudice to the parties' rights to claim rescission of the
arbitration agreement.

Notwithstanding this broad language, it has been suggested that it still may be
limited to situations of gross negligence. (563) The German Court has taken a
middle position, excluding the indemnity in cases of gross negligence. (564)

Institutional rules typically protect both the institution and that of the arbitrator.
ICC Rules 2012 Article 40 states that: ‘The arbitrators, any person appointed by
the arbitral tribunal, the emergency arbitrator, the Court and its members, the
ICC and its employees, and the ICC National Committees and Groups and their
employees and representatives shall not be liable to any person for any act or
omission in connection with the arbitration, except to the extent such limitation
of liability is prohibited by applicable law.’ It thus provides a full immunity for
arbitrators, the ICC Court, its members, the ICC itself, its employees and the ICC
National Committees. The 2010 UNCITRAL Rules also provide for immunity,
unlike the 1976 version. (565) The Rules state that parties waive all claims against
the arbitrators, save for intentional wrongdoing. Article 31.1 of the LCIA Rules,
after stating a general principle that no arbitrator ‘shall be liable to any party
howsoever for any act or omission in connection with any arbitration conducted’
under the auspices of the LCIA, excludes from the indemnity ‘conscious and
deliberate wrongdoing…’. The World Intellectual Property Organization (WIPO)
also has a similar standard for immunity where ‘[e]xcept in respect of deliberate
wrongdoing, the arbitrator or arbitrators … shall not be liable to a party for any
act or omission in connection with the arbitration’. The IBA Rules of Ethics 1987
indicate ‘international arbitrators should in principle be granted immunity from
suit under national laws, except in extreme cases of wilful or reckless disregard
of their legal obligations.’ Even where a wide-ranging indemnity does not
expressly provide any exceptions, most legal systems would likely interpret such
a provision to cover only negligent and not fraudulent actions. (566)

page "355"

Where institutional and ad hoc rules are concerned, consideration should also be
given as to whether these are subject to immunity provisions in national laws. If
the parties have selected rules which provide for greater immunity than in
national laws, that should be accepted. The situation would be quite different if
the arbitrator selected the protective procedural rules under a particular broad
discretion.

Where the lex arbitri provides immunity protection for arbitrators, the question
may arise as to whether an arbitrator can increase the level of protection
through contractual provisions. That would depend on the wording of the
particular statutory provision. In most cases they seek to set minimum levels of
protection and do not aim to put a ceiling on immunity. Nevertheless, in the
extreme, if an arbitrator sought to provide contractually for immunity over
fraudulent behaviour, most legal systems would consider such an agreement to
be contrary to public policy. Another relevant question is whether any statutory
limitation on immunity is a mandatory limit which cannot be extended. These
questions are more likely to apply in ad hoc arbitrations where an arbitrator may
typically have a pro forma contract with very broadly drafted exclusion of
liability provisions. Whether such provisions are valid and enforceable would
depend on the applicable law. Even a clear expression in the arbitral rules will
not necessarily be determinative. If liability is based on contract subject to an
applicable substantive law, that might proscribe exclusions of liability by
professionals. The same situation arises with express contractual limitations on
liability. As noted, even in situations where an arbitrator is afforded a broad
immunity, this may be lost in certain circumstances. An example would be where
an arbitrator improperly seeks to resign the appointment. In many systems, such
an arbitrator could not rely on a broad immunity as a shield against damages
claims for the expenses caused by the wrongful resignation.

Where rules are silent on where they give arbitrators a choice to limit liability, it
would be expected that many arbitrators would seek contractual immunity at
the time of appointment. Here it may still be the case that in some jurisdictions,
liability for gross negligence or intentional wrongs cannot be excluded
contractually. There is also a question as to whether such a broad exclusion of
liability is ethical. (567) Different courts vary as to whether they would conclude
that arbitrators have immunity either by reason of a judicial-style office or as an
implied term of a contract. (568)

page "356"
It is also important to note that immunities are granted to arbitrators when they
operate as such. It does not apply to other functions such as mediators,
conciliators or professionals making an expert determination. (569) The situation
may be debateable where an arbitrator performs a mediation function within the
arbitration itself. The indemnity ought to apply as a matter of course if the lex
arbitri expressly mandates such a dual function and provides a broad indemnity.
If the lex arbitri is silent, it remains contentious whether an arbitrator has any
duty to support settlement and if so whether that could trigger the immunity in
such circumstances. The duties and powers in that regard are discussed in
section 2.9.15.

5.18.3. Liability and Damages

In cases where liability is possible, there will then be the question of assessment
of damages. Where damages claims are concerned, these could be based in
contract as is the expectation in civilian legal systems or through the law of tort
as is the historical approach in common law countries, or both. (570) In some cases
that would involve effectively rehearing the matter to determine who ought to
have won and what the proper costs would be. In other circumstances it would
simply be exercising judgment in the context of the applicable burden of proof of
the liability action. Lew Mistelis and Kröll sensibly suggest that it would be hard
to evaluate damages incurred through delays (571) but this could apply in the
same way as damages for late payment under any commercial contract.

A particularly difficult circumstance to evaluate would be a failure to disclose

matters giving rise to justifiable doubts as to independence or impartiality. One
could readily see the damages equating to fees already paid. (572) More difficult
would be evaluating whether a different outcome on the merits would have been
achieved with an arbitrator not subject to an apprehension of bias. It would seem
difficult to prove such an assertion. Another complex area for potential damages
claims would relate to allegedly improper imposition of provisional measures.
While these can certainly cause serious financial hardship to parties page "357"
both when granted or when refused, it is more difficult to show liability and
damage caused. Damages claims are often made more difficult because of the
common allegation that an interim measure may have adversely affected a
company's viability when there are many other circumstances that could be
blamed. The liability is also made more difficult because of the balancing test
commonly employed. A related issue to the assessment of damages is whether an
arbitrator is entitled to legal costs in defending such claims. That will depend on
the particular jurisdiction involved. For example, the US does not generally allow
for cost recovery from the winning party against the loser. Born has argued that
even in such circumstances an arbitrator ought to be entitled to fees as a matter
of public policy in order to support the purposes of the Arbitration Act and the
New York Convention. (573)

In extreme circumstances, an arbitrator may be subject to criminal liability for

egregious behaviour. An example might be accepting bribes. Very few national
statutes provide expressly for arbitral liability in such circumstances although
general criminal statutes would commonly apply in any event. Some countries
purport to provide for broad-ranging extraterritorial effect of such provisions.

5.18.4. Loss of Fees

The previous sub-section considered a direct damages claim against an

arbitrator. The converse situation is where the allegedly improper behaviour
provides a defence against a claim by the arbitrator for remuneration. Some lex
arbitri expressly indicate that fee entitlement may be lost on removal for cause.
(574) Born observes that from a policy perspective, forfeiture of remuneration

might be argued to contradict principles of immunity. He concludes, however,

that this is a sensible limitation on immunity. It provides a natural response to an
unjustifiable claim for remuneration while at the same time shielding an
arbitrator from the threat of open-ended financial liability. (575)

page "358"

Appendix 1. Decisions on Challenges to Arbitrators

This table provides a brief overview of prominent cases in which an arbitrator

has been challenged. (576) It is not intended to be a comprehensive account of the
facts or decisions, nor is it a full list of decisions in each jurisdiction. Rather, it is
intended to be a representative sample to be used as a quick reference point for
practitioners looking for cases relevant to their work. Obviously each case may
depend on particular facts and observers may disagree with the conclusions.
Nevertheless, it may be a useful way to get a feel for matters that are
questionable.

Case Facts giving rise to challenge Decision

Belgium
Eureko v • Judge Schwebel was wrongly • Challenge rejected.
Poland.577 reported to have been advising law
firm Sidley Austin in an unrelated
case against Poland. His Washington
DC office was also in the same
building as Sidley Austin.
France
Tesco v • Chairman worked at law firm of • Challenge
Neoelectra.578 counsel for the respondent until eight succeeded.
years before the arbitration. He had
been an occasional consultant since
then. He was a ‘Facebook friend’ of
respondent's counsel. This had not
been disclosed.
Société Annahold v • Just after award was rendered, • Challenge
L'Oreal.579 arbitrator began work as a succeeded.
consultant for one of the parties.
KFTCIC v Icori • Arbitrator, who was a barrister, • Challenge rejected.
Estero.580 shared chambers with one of the
party's counsel.
Marteau v CIGP.581 • Arbitrator worked at an accounting • Challenge
firm which was acting as auditors for succeeded.
a subsidiary of the party which
appointed the arbitrator.
Transgrain • Arbitrator was an executive officer • Challenge
France.582 of a company which was involved in succeeded
an ongoing dispute with one of the
parties to the arbitration.
Milan Presse v • Arbitrator was the step-father of • Challenge accepted.
Media Sud.583 one of the party's counsel.
Richy v • Arbitration concerned a dispute • Challenge accepted.
Warlaumont.584 between a doctor and a patient. Sole
arbitrator was head of clinical
surgery at a hospital. The respondent
was a doctor working in the same
department, and was thus the
arbitrator's subordinate.
Qatar v • Dispute involved multiple parties • Challenge rejected.
Creighton.585 and arbitrations. Arbitrator was
involved in initial arbitration
between C and its subcontractors.
Same arbitrator acted in related
arbitration between C and Qatar. It
appeared that decisions made in the
first arbitration would not prejudice
Qatar, who challenged the arbitrator,
in the second arbitration.
Setec Bâtiment.586 • Construction dispute involved • Challenge
multiple parties and arbitrations. succeeded.
Arbitrator was appointed by
construction site owner in
arbitration against main contractor.
Owner appointed same arbitrator in
second arbitration against engineer.
First award dealt with engineer's
liability.
Uni-Inter v • Arbitrator had made earlier • Challenge rejected.
Maillard.587 statements of opinion on legal issues
relevant to the arbitration.
Dubai v • Arbitrator stated in preliminary • Challenge rejected.
Halcrow.588 award that the judicial system in
Dubai was “somewhat
autocratic”.589 There were also
apparent errors in the arbitrator's
 determination of the applicable law.
Stolz v Ets • At same time as arbitration, • Challenge
Letierce.590 arbitrator was being personally paid succeeded.
to provide advice and technical
assistance to one of the parties.
Germany
Billerbeck.591 • Arbitration held in 1967 in the • Challenge rejected.
German Democratic Republic
between Swiss and East German
parties. All arbitrators were East
German. Swiss party alleged bias due
to nationality of arbitrators.
X v Y.592 • Chairman was co-editor of journal • Challenge rejected.
which contained a paper by
respondent's counsel. Paper
concerned arbitration and
competition law, which was relevant
to the dispute. Chairman claimed not
to have read the paper.
D v E.593 • Chairman made sarcastic comment • Challenge rejected.
to respondent's counsel and wrote
letter to counsel pressing her to
withdraw allegations about the
comment at risk of perjury. Letter
cast doubt on respondent's motives.
Hong Kong
Logy • Replacement arbitrator appointed • Challenge rejected.
Enterprises.594 by CIETAC was a director of a
Chinese government agency. A
certificate from that agency was
important evidence. The tribunal
appeared to make various errors in
its award.
Suen Wah Ling.595 • Six years prior to award, arbitrator • Challenge rejected.
(a barrister) had advised Suen on the
matter. At the first hearing, they did
not recognise each other, although
the arbitrator asked Suen whether he
looked familiar. Suen said he did not
and the case proceeded. Suen lost
and later realised the connection
with the arbitrator.
Jung Science.596 • Chairperson failed to disclose that • Challenge rejected.
he knew respondent's solicitor. Both
were well known in arbitration
circles in Asia, had served on
committees together and met at
 various functions.
ICC
• ICC 2009 Case • Co-arbitrator was acting as co- • Challenge
1.597 counsel with the respondent's succeeded.
counsel in another arbitration.
Arbitrator had allegedly been
involved in two other arbitrations
with that counsel and not disclosed
this.
• ICC 2009 Case • Chairman had acted as co-counsel • Challenge
2.598 with one of the party's counsel in a succeeded.
case which had settled (though the
file was not yet closed). It was
conceivable they would have further
work on that case together. This had
not been disclosed.
• ICC 2007 Case • Co-arbitrator disclosed, after • Challenge
1.599 confirmation, that he had given legal succeeded
advice to a party in the past and had
been involved in a project related to
the dispute.
• ICC 2007 Case • Sole arbitrator was member of law • Challenge
2.600 firm with alliance to accountants who succeeded
had acted as claimant's auditors.
• ICC 2007 Case • Respondent challenged Chairman, • Challenge
3.601 because a foreign office of the succeeded.
Chairman's law firm was acting for a
party in unrelated litigation against
the respondent's parent company.
• ICC 2007 Case • Arbitrator had acted as counsel for • Challenge
4.602 one of the parties in unrelated succeeded.
proceedings. Several of those
proceedings were ongoing.
• ICC 2007 Case • Co-arbitrator appointed by state • Challenge
5.603 party failed to disclose that he was succeeded.
part of that state's legal advisory
department.
• ICC 2007 Case • Respondent challenged arbitrator • Challenge
6.604 for failing to disclose that he had succeeded.
been Chairman in another
arbitration, which did not progress to
hearing, that concerned the same
construction project.
ICSID
• Canfor v United • This was part of the US-Canada • Challenge
States (Challenge softwood lumber dispute. Arbitrator succeeded.
l).605 had made a speech 18 months prior
to appointment about the softwood
lumber dispute with the US. It
included the following: “[t]his will be
the fourth time we have been
challenged. We have won every
single challenge … and yet they
continue to challenge … because they
know the harassment is just as bad as
the process”.
• Canfor v United • US requested that a Consolidation • Challenge rejected.
States (Challenge Tribunal hear several cases together.
2).606 The decision was to be made by the
Consolidation Tribunal. Investor
argued that, as this would mean more
remuneration for the arbitrators,
they had a significant financial
interest in the outcome.
• Hrvatska v • One of respondent's counsel was a • Counsel was
Slovenia.607 door tenant in the same chambers as prohibited from
the Chairman. participating.
• EDF International • Co-arbitrator was a director of UBS. • Challenge rejected.
SA et al v UBS had recommended to its
Argentina.608 customers to invest in EDF, the
parent of one of the parties. This had
not been disclosed.
• Azurix v • Arbitrator's law firm had, in a • Challenge rejected.
Argentina.609 separate arbitration, appointed its
counsel in the instant case as
arbitrator.
• Participaciones • Arbitrator had previously rendered • Challenge rejected.
Inversiones an award which determined whether
Portuarias v withdrawal of a concession by Gabon
Gabon.610 was expropriation. The decision was
unpublished, but the question of
withdrawing concessions was a live
issue in the new arbitration.
• Amco Asia v • Claimant-appointed arbitrator had, • Challenge
Indonesia.611 seven years earlier, given tax advice rejected.612
to claimants' parent company.
Arbitrator also had office and profit-
sharing arrangement with claimant's
lawyers until six months before the
arbitration.
• Vivdendi • Partner at arbitrator's law firm had • Challenge rejected.
Universal v given tax advice to the claimant's
Argentina.613 corporate predecessor. Arbitrator
 was not involved in this advice.
• SGS v • Arbitrator A had acted as counsel in • Challenge rejected.
Pakistan.614 another ICSID arbitration three years
earlier. In that case, Arbitrator B was
an arbitrator. The award in that case
went in favour of Arbitrator A's
client. Arbitrator B now appeared as
counsel before Arbitrator A in the
instant case.
• Generation • Arbitrator had previously worked • Challenge rejected.
Ukraine v for a World Bank agency. As part of
Ukraine.615 this, he had been part of studies and
investment policy reviews of Ukraine
for the OECD. It was alleged that he
had developed connections with
Ukrainian political officials as a
result.
• Suez v Argentina • Arbitrator, had previously made an • Challenge
(No 1).616 award against Argentina in the rejected.617
Vivendi case. Argentina claimed this
award had inconsistencies which
revealed bias.
Tanzania Electric • Arbitrator's law clerk posted • Arbitrator
618 comments on a blog about legal withdrew after panel
issues relevant to the case. was unable to decide
on challenge.
Iran-US Claims
Tribunal
• Re Judge • Iran alleged that Arbitrator had • Challenge
Mangard.619 criticised the death penalty in Iran rejected.620
and, by implication, the Iranian
regime. There was very little
evidence of this criticism having been
made.
• Amoco Oil • Arbitrator was the sole director of a • Arbitrator
Company.621 subsidiary of Morgan Stanley. withdrew, but denied
Morgan Stanley employees were bias.
called as expert witnesses in the
arbitration.
• Re Judge • Arbitrator's dissenting opinion • Challenge rejected.
Broms.622 relayed the discussions of the
arbitrators on the merits of the
application. The US agent argued this
breached procedural rules on secrecy
and showed pro-Iranian bias.
LCIA
• National Grid PLC • Arbitrator made comment to expert • Challenge rejected.
v The Republic of witness to the effect that the facts
Argentina.623 were clear and that “we know the
facts generally speaking and there
has been major harm or major
change in the expectations of the
investment”.624 This was said to be
central to the dispute.
• LCIA Challenge • Arbitrator's law firm had working • Challenge
Decision I (29 May relationship with companies succeeded.
1996).625 associated with respondent. The firm
also acted for a bank involved in a
transaction relevant to the
arbitration.
• LCIA Challenge • Respondent-appointed arbitrator • Challenge rejected.
Decision 2 (5 June shared chambers in London with
1997).626 counsel for respondent.
• LCIA Challenge • Claimant-appointed arbitrator had • Challenge rejected.
Decision 3 (23 previously acted as expert witness
October 1997).627 for claimant's counsel.
• LCIA Challenge • Claimant asserted that sole • Challenge rejected.
Decision 4 (22 My arbitrator had, on the whole, failed to
1998).628 act with reasonable diligence.
• LCIA Challenge • British claimant appointed an • Challenge rejected.
Decision 5 (30 arbitrator who was said to be a de
September facto British national, although not a
1998).629 citizen.
• LCIA Challenge • Claimant's counsel was of the same • Challenge rejected.
Decision 6(10 nationality as the sole arbitrator.
November
1999).630
• LCIA Challenge • Arbitrator's firm had previously • Challenge
Decision 6A (27 advised respondent on contracts succeeded.
January 2000).631 which were the focus of the
arbitration. Arbitrator was not a
member of the firm at the time of
that advice.
• LCIA Challenge • Respondent made a range of • Challenge rejected
Decision 7 (22 June complaints about conduct of
2001).632 proceedings which it said showed
“persistent bias”, and suggested the
finding of limited jurisdiction showed
bias.
• LCIA Challenge • Respondent cited a range of • Challenge rejected.
Decision8 (3 conduct, including scheduling of
October 2001).633 hearings at inconvenient times and
alleged failure to consider certain
 evidence
• LCIA Challenge • Sole arbitrator was member of a • Challenge
Decision 9 (22 professional organisation said to succeeded.
November have link to respondent. When
2001).634 claimant raised this, arbitrator
reacted angrily, asserting the
claimant was acting “viciously”.
• LCIA Challenge • Arbitrator conducted hearing in • Challenge rejected
Decision 10 (13 absence of claimant. The hearing had (but arbitrator was
February been cancelled as neither party not paid for
2002).635 wished to proceed. Respondent later
said it would attend, and claimant
was given less than 24 hours notice.
• LCIA Challenge • Party-appointed arbitrator • Challenge
Decision 11 (1 My disclosed content of award and succeeded.
2002).636 outcome of arbitration to the party
that appointed him, prior to
publication of the award.
• LCIA Challenge • Respondent claimed arbitrator gave • Challenge rejected.
Decision 12 (1 directions without consulting co-
October 2002).637 arbitrators; favoured the claimant in
structure and scheduling of
arbitration; improperly raised issues
regarding jurisdiction; refused an
oral hearing without reasons.
• LCIA Challenge • Respondent's counsel resigned. • Challenge rejected.
Decision 13 (3 My Respondent applied for time
2003).638 extension to file documents as a
result. Extensions were granted, but
not for the entire period sought by
the respondent.
• LCIA Challenge • Respondent criticised a range of the • Challenge rejected.
Decision 14 (18 arbitrator's conduct, including
December uncertain and unpredictable emails,
2003).639 unfair procedural orders
• LCIA Challenge • Four years prior to the arbitration, • Challenge rejected.
Decision 15 (24 arbitrator had played ‘supervisory’
December role in law firm on a project for the
2003).640 respondent. Arbitrator was no longer
with the firm. The project appeared
relevant to the arbitration.
• LCIA Challenge • Arbitrator treated respondent's • Challenge rejected.
Decision 16 (18 counterclaim as withdrawn because
February respondent failed to deposit requisite
2004).641 funds with the LCIA.
• LCIA Challenge • Dispute concerned English • Challenge rejected.
Decision 17 (14 insurance law. Non-English
August 2003).642 arbitrator without qualifications in
English law was appointed. A party
challenged this due to lack of
arbitrator's experience in the area.
• LCIA Challenge • Sole arbitrator met privately with • Challenge
Decision 18 (21 claimant's counsel during hearing succeeded.
October 2005).643 breaks; deleted parts of the
transcript without both parties'
consent; accused respondent's
counsel of entering his retiring room
when arbitrator was not present.
• LCIA Challenge • Arbitration agreement provided • Challenge rejected.
Decision 19 (28 that arbitrator was not to be British
October 2005).644 or based in England. Arbitrator was
door tenant in London (indicating he
visited from time to time).
• LCIA Challenge • Sole arbitrator was expert on Arab • Challenge rejected.
Decision 20 and Islamic law, with long-standing
(2005).645 interest in Arab studies. One party
was Kuwaiti, the other Western.
Western party alleged possible bias.
• LCIA Reference • Respondent-appointed arbitrator • Challenge rejected
No. UN3476.646 had advised the respondent on a
procurement and construction
contract four years prior to the
arbitration.
• LCIA Reference • Chair published the parties to the • Challenge rejected.
No. 5665.647 arbitration on his online CV. Chair
had also used “intemperate language”
with Claimant's counsel. Claimant
further alleged that a key piece of
evidence was excluded without
proper consideration.
• LCIA Reference • A majority of the tribunal, in a • Challenge rejected
No. 3488.648 procedural order, expressed a
tentative, preliminary view on an
issue relevant to the outcome of the
arbitration.
• LCIA Reference • Claimant-nominated arbitrator • Challenge rejected
No. 81007/81008/ issued a dissenting opinion in an
81025.649 order on provisional relief, which
touched on issues concerning the
merits of the Respondent's claim
• LCIA Reference • Arbitrator made a “peremptory • Challenge rejected
No. 7932.650 order” at the Claimant's request.
Arbitrator had only considered the
“gist” of the Respondent's
 submissions on this, but not the
actual submissions.
• LCIA Reference • Claimant-appointed arbitrator was • Challenge rejected,
No. 81132.651 a barrister. He had been instructed although the Chair
by the Claimant's law firm on two voluntarily stood
matters previously and acted against down.
them on two other occasions.
Arbitrator was also a member of an
association established by the
Claimant's law firm. Further,
arbitrator had been instructed by the
Chair in previous matters.
• LCIA Reference • Respondent-appointed arbitrator • Challenge
No. 81160.652 had acted for and against the succeeded
Respondents in previous cases. He
had been instructed on numerous
occasions by Respondent's counsel,
and it was apparent that this
relationship would continue in the
future.
• LCIA Reference • Arbitration had been delayed • Challenge rejected
No. 81210.653 extensively due to the Respondent
losing its counsel and the illness of its
main in-house lawyer. Respondent
lacked resources and experience with
arbitration. Chair sought to proceed
with arbitration, which Respondent
said showed bias.
• LCIA Reference • Arbitrator had met Respondent's • Challenge rejected
No. 8I224.654 counsel on one previous occasion.
Apparently the arbitrator had not
been paid for that meeting, which the
Respondent said may have given rise
to some bitterness towards its
counsel. Respondent was also
concerned about comments made
during a hearing which indicated
prejudgment. Specifically, the
arbitrator had prompted the
Claimant with arguments in its
favour, but challenged the
Respondent in all of its arguments.
• LCIA Reference • Respondent refused to pay certain • Challenge rejected
No. 7990.655 advances to cover the costs of the
arbitration. In consultation with the
LCIA, the arbitrator ordered that the
Claimant pay these costs in
accordance with the rules. After
 missing the deadlines, the arbitrator
refused to give a further extension
for payment. Claimant alleged this
showed bias.
Netherlands
• Aegon • Two members of the tribunal in a • Challenge
Verzekering medical insurance dispute were succeeded.
Leven.656 doctors. Both had physically
examined the claimant. Claimant said
this “investigative” conduct
compromised impartiality.
Permanent Court
of Arbitration
• Telekom • Arbitrator in an investment • Challenge
Malaysia.657 arbitration was acting as counsel for succeeded, but
an investor in a similar but unrelated arbitrator was
investment arbitration which dealt permitted to resign
with similar points of law. from role as counsel
and continue as
arbitrator.
• Vito G Gallo v • This was a NAFTA case. One • Challenge
Canada arbitrator was an independent succeeded.
consultant to a law firm which had
been retained by Mexico to advise it
on trade and investment issues.
Mexico, being a party to NAFTA, was
a potential party in the proceedings.
• Perenco Ecuador • Arbitrator gave an interview about • Challenge
Ltd v Ecuador.658 the need for acceptance of succeeded.
international arbitration by states.
He cited Ecuador's failure to follow
orders of two ICSID tribunals as an
example of the problem.
SCC
• SCC Arbitration • Claimant-appointed arbitrator had • Challenge rejected.
137/ 2008.659 ongoing business relationship with a
firm at which one of the claimant's
witnesses worked and had been in
contact with the witness on some
occasions. Chairman had also worked
with the witness in matters where
witness was expert witness. This was
not disclosed until one day before the
hearing.
• SCC Arbitration • Chairman's spouse had acted as • Challenge rejected.
001/ 2010.660 counsel in a matter where the
claimant-appointed arbitrator was
 opposing counsel. Claimant-
appointed arbitrator had strongly
criticised the spouse during this
matter and believed the spouse was
strongly averse to him. Chairman
claimed to be unaware of this.
• SCC Arbitration • Russian party-appointed arbitrator • Challenge rejected.
045/ 2008.661 was challenged because he had a
long-term personal relationship with
two high-level Russian government
officials. The Russian Government
and these officials apparently had
interests aligned with the respondent
in the case.
• SCC Arbitration • Respondent-appointed arbitrator • Challenge
207/ 2009.662 was a director of a large corporation. succeeded.
The majority of the corporation's
shares were held by a party
connected to the respondent.
Arbitrator had also been a legal
advisor to the respondents on four
occasions over past 15 years. Partner
in arbitrator's firm had been official
legal advisor to four successive heads
of the respondent. Respondent also
provided free or low cost office space
to arbitrator's law firm.
• SCC Arbitration • Claimant-appointed arbitrator had • Challenge
018/ 2009.663 been appointed by the claimant on succeeded.
several prior occasions. He was also a
former chairman and current ethics
board member of an organisation of
which two of the claimant's
subsidiaries were members.
• SCC Arbitration • Claimant-appointed arbitrator's law • Challenge
068/ 2010.664 firm had worked on three matters succeeded.
involving the claimant. Two of the
matters had been handled by another
office and finished three years
earlier. In the third matter, the
arbitrator was personally involved in
a transaction involving the claimant,
apparently advising on purchasing
shares in the claimant. That matter
had finished less than two years
before the arbitration.
• SCC Arbitration • Claimant's counsel was acting in • Challenge
058/ 2008.665 court proceedings in which succeeded.
 respondent-appointed arbitrator was
personally the respondent.
• SCC Arbitration • Chairperson disclosed that his firm • Challenge
053/ 2005.666 had been given a number of succeeded.
assignments by the claimant in
recent years, including the current
year. He had not been personally
involved in any of the cases.
• SCC Arbitration • Law firm representing the • Challenge rejected.
078/2005 respondent had employed one of the
(Challenge l).667 arbitrators for 12 years.
• SCC Arbitration • One of the arbitrators had for four • Challenge rejected.
078/2005 years been a partner of the law firm
(Challenge 2).668 where claimant's counsel worked.
Arbitrator had apparently left the
firm by the time counsel commenced
there. They had never met.
• SCC Arbitration • Arbitrator worked for a company • Challenge rejected.
019/ 2006.669 that in the past had done work for
the respondent (apparently as a
supplier). Arbitrator said he had no
knowledge about the internal
workings of the respondent.
• SCC Arbitration • Chinese respondent appointed a • Challenge rejected.
60/1999 CIETAC panellist and chief judge of
(Challenge 1).670 court in city where respondent was
based. Respondent's counsel was also
based in that city. Claimant argued
that Chinese courts practice local
protectionism.
• SCC Arbitration • Claimant-appointed arbitrator • Challenge
60/1999 disclosed that a partner at his law succeeded.
(Challenge 2).671 firm had from time to time advised
the claimant on retainer. That
partner was not involved in the
instant case.
• SCC Arbitration • Claimant-appointed arbitrator • Challenge
60/ 2001.672 inadvertently failed to disclose that a succeeded.
partner at his firm was representing
the claimant's parent company in
ongoing litigation.
• SCC Arbitration • Respondent-appointed arbitrator's • Challenge rejected.
87/ 2000.673 cousin was board member of
respondent's parent company.
Singapore
• Turner.674 • Arbitrator used language in a letter • Challenge
 to one of the parties which was succeeded.
“sarcastic, to the point of being
hostile” and had an offensive tone. He
accused one of the parties of delaying
tactics.
Sweden
• Jilkén v • Chairman had previously worked as • Challenge
Ericsson.675 a part-time consultant for law firm succeeded.
Mannheimer Swartling. The same
firm represented Ericsson in the
arbitration. He had not disclosed this.
• KPMG AB v • Arbitrator's law firm had accepted
ProfilGruppen AB to work on a substantial matter
676 against one of the parties. Work on
this case commenced before the
arbitral award was rendered. The
arbitrator worked in the same office
as those working against the party
and was aware the firm had been
asked to this work, but it is not clear
he knew the work had been accepted.
Switzerland
• X SA v Y AS.677 • Claimant's counsel and sole • Challenge rejected.
arbitrator had practiced together in
the US. There was also evidence that
claimant's counsel had worked with
and attended a social event with the
arbitrator's daughter, and may have
briefly met the arbitrator in his home
in that context.
• X & Y v V & W.678 • Tribunal invited comment from • Challenge rejected.
claimant on admissibility of
respondent's counterclaims.
Tribunal, apparently forgetting about
this invitation, admitted
counterclaims before claimant had
responded. Tribunal later reviewed
issue in light of claimant's
submissions and came to same
conclusion in admitting the
counterclaims. Claimant said this
showed partiality.
• X SpA v Y BV.679 • Tribunal established procedural • Challenge rejected
rules with time deadlines. Claimant
requested a time extension which
was denied. Later, respondent was
granted time extension in submitting
 certain documents. Claimant said this
showed partiality.
• Ligier & Diffusia v • Key witness was a former client of • Challenge
Alfa Lancia.680 the sole arbitrator. Arbitrator was succeeded.
likely to act for witness again in the
future.
• Centroza v • Arbitrator's wife was assistant to • Challenge
Orbis.681 counsel for the respondent. succeeded.
• X v Y.682 • In its award, the tribunal, using • Challenge rejected.
“personal” and “almost ironic
language”, explained why X had acted
in bad faith. The award was in
French, but the majority of the panel
were not native French speakers. X
largely succeeded on the merits.
• A v B.683 • Arbitrator was followed by private • Challenge rejected.
detectives, apparently in relation to
the arbitration. Angry about this, the
arbitrator asked at the next hearing
which party was behind it. Neither
party came forward. Claimant
believed that arbitrator thought
claimant was responsible for
surveillance.
UK decisions
• ASM Shipping Ltd • Chairperson was involved in an • Challenge
v TIMI Ltd.684 earlier arbitration involving the succeeded on
respondent's main witness. In that chairperson, but
arbitration, the witness's credibility failed or co-
was a key issue. Respondent argued arbitrators.
that, in addition to the chair, the
remaining two arbitrators were not
impartial by “contamination”
• Catalina v • In an arbitration between a • Challenge
Norma.685 Portuguese and Norwegian party, succeeded.
arbitrator apparently commented
that all Portuguese people tended to
be liars, while Norwegians tended to
be honest.
• Laker • Arbitrator shared chambers with • Challenge rejected.
Airways.686 one of the parties' counsel.
• AT &T • Arbitrator inadvertently failed to • Challenge rejected.
Corporation.687 disclose that he held shares in AT & T
and that he was a director of a
company that failed in a bid to work
on the project which was the subject
of the dispute.
• Rustal Trading • Arbitrator failed to disclose that he • Challenge rejected.
Ltd v Gill & Duffus had acted against one of the party's
SA.688 consultants in an arbitration two
years earlier.
• Fletamentos.689 • Arbitrator made it clear that he • Challenge rejected.
thought counsel for one of the parties
was incompetent, but did not express
hostility to the party itself.
United States
• National • During the arbitration, respondent- • Challenge rejected.
Shipping.690 appointed arbitrator engaged
counsel for the respondent to
represent him in an unrelated
arbitration. The work only involved
sending a letter and was not paid.
This was not disclosed. Award was
made in favour of respondent.
• AIMCOR.691 • Chairperson was CEO of a company • Challenge
with 50 offices worldwide. Late in the succeeded.
arbitration, it was revealed that his
company had a contract with the
claimant's parent company to
transport petroleum coke. The
company had earned US $ 275, 000
doing this. A 2-1 award was rendered
in favour of the claimant.
• Positive • Arbitrator and respondent's • Challenge rejected.
Software.692 counsel had both been lawyers
advising Intel in a large patent case.
They belonged to different firms.
Around 35 lawyers were advising
Intel on the matter. They never
worked together or even met. This
had not been disclosed.
• Ovalar • Chairman was CEO of a multi- • Challenge
Makine.693 billion dollar company. An affiliated succeeded.
company had entered into a $275,
000 with a company that had
acquired one of the parties to the
arbitration. The Chairman claimed
that a “Chinese Wall” had been set up
between him and the transaction.
• Lucent • Claimant-appointed arbitrator • Challenge rejected.
Technologies.694 disclosed that he had been retained
as counsel for the claimant five years
earlier. Chairperson disclosed that
his firm also did work for claimant.
 Somehow the respondent never
received this information. It was later
revealed that these two arbitrators
had for many years owned a plane
together.
• Commonwealth • Chairperson, an engineer, had done • Challenge
Coatings v consulting work for one of the parties succeeded.696
Continental on the project that was the subject
Casualty.695 matter of the dispute. He had not
disclosed this. A unanimous award
was made in favour of the party for
whom the Chairperson had
consulted.

577. Judgment of 23 December 2006, Eureko BV v. République de Pologne,

Brussels Court of First Instance, RG 2006/1542/A. The decision was affirmed on
appeal. See Judgment of 29 October 2007, République de Pologne v. Eureko BV,
Brussels Court of Appeal.

578. Judgment of 10 March 2011, Tesco v. Neoelectra, Cour d'appel de Paris, Case
No. 09/28537.

579. Judgment of 9 April 1992, Cour d'appel de Paris, Revue de l'Arbitrage

(1986): 483.

580. Judgment of 28 June 1991, KFTCIC v. Icori Estero, Cour d'appel de Paris,
Revue de l'Arbitrage (1992): 568.

581. Judgment of 30 November 1999, Cour d'appel de Paris, Revue de l'Arbitrage

(2000): 299.

582. Judgment of 12 December 1999, Commercial Agraria Hermanos Lucena v.

Transgrain France, Cour d'appel de Paris, Revue de l'Arbitrage (1998): 699.

583. Judgment of 12 January 1999, Milan Presse v. Media Sud Communication,

Cour d'appel de Paris, Revue de l'Arbitrage (1999): 381.

584. Judgment of 20 October 1994, Edgard Richy v. Warlaumont et Union des

Assurances de Paris, Cour d'appel de Paris, Revue de l'Arbitrage (1996): 422.

585. Judgment of 16 March 1999, État du Qatar v. Société Creighton Ltd, Cour de
cassation, Revue de l'Arbitrage (1999): 308.

586. Judgment of 13 January 1986, Setec Bâtiment v Sicca, Tribunal de grande

instance de Paris, Revue de l'Arbitrage (1987): 63.

587. Judgment of 5 July 1990, Cour d'appel de Paris, Revue de l'Arbitrage (1991):
359.
588. Judgment of 1 April 1993, Etat de Dubai et société Dubai Drydocks v. Halcrow
& F McWilliams, Tribunal de grande instance de Paris, Revue de l'Arbitrage
(1993): 455–461.

589. In the original French: ‘quelque peu autocratique’.

590. Judgment of 15 January 1988, Société des Equipements Industriels Stolz SA v.

Ets Letierce, Tribunal de grande instance de Paris, Revue de l'Arbitrage (1988):
316.

591. Judgment of 3 May 1967, Billerbeck Cie v. Berghau-Handel GmbH, Supreme

Court of the Federal Republic of Germany, Yearbook of Commercial Arbitration 1
(1976): 200.

592. Judgment of 4 October 2007, X v. Y, Oberlandesgericht Frankfurt am Main

(Higher Regional Court of Frankfurt).

593. Judgment of January 2007, D v. E. See Sam Luttrell, Bias Challenges in

International Commercial Arbitration: The Need for a ‘Real Danger’ Test (Alphen
aan den Rijn: Kluwer Law International, 2009), 106.

594. Logy Enterprises Ltd v. Haikou City Bounded Areas Wansen Products Trading
Co. [1997] 2 HKC 481.

595. Suen Wah Ling t/a Kong Luen Construction Engineering Co. v. China Harbour
Engineering Co. (Group) [2008] HKCU 570.

596. Jung Science Information Technology Co. Ltd v. ZTE Corporation [2008] 4
HKLRD 776.

597. Jason Fry and Simon Greenberg, ‘The Arbitral Tribunal: Applications of
Articles 7–12 of the ICC Rules in Recent Cases’, ICC International Court of
Arbitration Bulletin 20, no. 2 (2009): 24.

598. Ibid., 25.

599. A.M. Whitesell, ‘Independence in ICC Arbitration: ICC Court Practice

Concerning the Appointment, Confirmation, Challenge and Replacement of
Arbitrators’ ICC International Court of Arbitration Special Supplement (2007), 29.

600. Ibid.

601. Ibid.

602. Ibid., 30.

603. Ibid.

604. Ibid.
605. Canfor Corporation v. United States; Tembec et al. v. United States and
Terminal Forest Products Ltd v. United States, Consolidated NAFTA Arbitration,
Order of 7 September 2005.

606. Canfor Corporation v. United States; Tembec et al. v. United States and
Terminal Forest Products Ltd v. United States, Consolidated NAFTA Arbitration,
Order of 7 September 2005.

607. Hrvatska Elektroprivreda dd v. Republic of Slovenia, ICSID Case No.

ARB/05/24.

608. EDF International SA, SAUR International SA and León Participaciones

Argentinas SA v. Argentine Republic, ICSID Case No. ARB/03/23, Challenge
Decision Regarding Professor Gabrielle Kaufmann-Kohler of 25 June 2008.

609. Azurix v Argentine Republic, ICSID Case No. ARB/01/12, Annulment

Proceeding, Decision on the Application for Annulment, 1 September 2009.

610. Participaciones Inversiones Portuarias SARI v. Gabonese Republic, ICSID Case

No. ARB/08/17, Decision on the Proposal to Disqualify an Arbitrator of 12
November 2009.

611. Amco Asia v Indonesia, ICSID Case No. ARB/81/1, Decision on Proposal to
Disqualify an Arbitrator (24 June 1982, unpublished. See summary of case in Sam
Luttrell, Bias Challenges in International Commercial Arbitration: The Need for a
‘Real Danger’ Test (Alphen aan den Rijn: Kluwer Law International, 2009), 225–
226.

612. But see criticism of this decision in Compañia de Aguas del Aconquija S.A. &
Vivendi Universal S.A. v. Argentine Republic, ICSID Case No. ARB/97/3, Decision
on the Challenge of the President of the Committee (3 October 2001).

613. Compañia de Aguas del Aconquija S.A. & Vivendi Universal S.A. v. Argentine
Republic, ICSID Case No. ARB/97/3, Decision on the Challenge of the President of
the Committee (3 October 2001).

614. SGS Société Générate de Surveillance S.A. v. Islamic Republic of Pakistan, ICSID
Case No. ARB/01/13, Decision on Claimant's Proposal to Disqualify Arbitrator
(19 December 2002).

615. Generation Ukraine v. Ukraine, ICSID Case No. ARB/00/9 (16 September
2003).

616. Suex and others v. Republic of Argentina, Disqualification Decision, ICSID

Case No. ARB/03/17.

617. This was partly due to delay in making the challenge, but the Tribunal also
stated that the claim would have been rejected in any case.
618. Tanzania Electric Supply Co. Ltd v. Independent Power Tanzania Ltd. ICSID
Case No. ARB/98/8 (12 July 2001). See also Lars Markert, ‘Challenging
Arbitrators in Investment Arbitration: The Challenging Search for Relevant
Standards and Ethical Guidelines’, Contemporary Asia Arbitration Journal 3, no. 2
(2010): 264–265.

619. Re Judge Mangard ‘Challenge Decision’, 1 Iran–US Claims Tribunal Reports

111.

620. In an extraordinary subsequent development, two of the Iranian arbitrators

physically attacked Judge Mangard when entering the Tribunal Chamber. This
was followed by fairly clear death threats. The US agent then challenged the
impartiality of the two arbitrators. The arbitrators withdrew, so no decision was
made, but the challenge would have almost certainly been successful.

621. See Sam Luttrell, Bias Challenges in International Commercial Arbitration:

The Need for a ‘Real Danger’ Test (Alphen aan den Rijn: Kluwer Law
International, 2009), 101–102.

622. Re Judge Broms, Decision of the Appointing Authority to the Iran–US Claims
Tribunal, 7 May 2001. See also Sam Luttrell, Bias Challenges in International
Commercial Arbitration: The Need for a ‘Real Danger’ Test (Alphen aan den Rijn:
Kluwer Law International, 2009), 102.

623. LCIA Case No. UN7949, Decision on the Challenge to Mr Judd L. Kessler, 3
December 2007.

624. This is a working translation of the arbitrator's phrasing in Spanish:

‘sabemos los hechos en general, de que hubo un daño importante o hubo cambio
muy importante en las expectativas de la inversion’.

625. Geoff Nicholas & Constantine Partasides, ‘LCIA Court Decisions on

Challenges to Arbitrators: A Proposal to Publish’, Arbitration International 23, no.
1 (2007): 26.

626. Ibid.

627. Ibid., 29.

628. Ibid., 37.

629. Ibid., 30–31.

630. Ibid., 26.

631. Ibid., 27.

632. Ibid., 34.

633. Ibid.

634. Ibid., 36.

635. Ibid., 26.

636. Ibid., 37.

637. Ibid., 26.

638. Ibid.

639. Ibid., 34.

640. Ibid., 27.

641. Ibid., 34.

642. Ibid., 41.

643. Ibid., 37.

644. Ibid., 31.

645. Ibid., 36.

646. ‘LCIA Court Decisions on Challenges to Arbitrators Reference No. UN3476,

24 December 2004’, Arbitration International 27, no. 3 (2011): 367–370.

647. ‘LCIA Court Decisions on Challenges to Arbitrators Reference No. 5665, 30

August 2006’, Arbitration International 27, no. 3 (2011): 395–412.

648. ‘LCIA Court Decisions on Challenges to Arbitrators Reference No. 3488, 11

July 2007’, Arbitration International 27, no. 3 (2011): 413–419.

649. ‘LCIA Court Decisions on Challenges to Arbitrators Reference No.

81007/81008/81024/81025’, Arbitration International 27, no. 3 (2011): 425–
432.

650. ‘LCIA Court Decisions on Challenges to Arbitrators Reference No. 7932, 17

June 2008’, Arbitration International 27, no. 3 (2011): 433–438.

651. ‘LCIA Court Decisions on Challenges to Arbitrators Reference No. 81132, 15

November 2008’, Arbitration International, 27 no. 3 (2011): 439–441.

652. ‘LCIA Court Decisions on Challenges to Arbitrators Reference No. 81160, 28

August 2009’, Arbitration International, 27 no. 3 (2011): 442–454.

653. ‘LCIA Court Decisions on Challenges to Arbitrators Reference No. 81210, 16

November 2009’, Arbitration International, 27 no. 3 (2011): 455–460.

654. ‘LCIA Court Decisions on Challenges to Arbitrators Reference No. 81224, 15

March 2010’, Arbitration International 27, no. 3 (2011): 461–470.

655. ‘LCIA Court Decisions on Challenges to Arbitrators Reference No. 7990, 21

May 2010’, Arbitration International 27, no. 3 (2011): 471–473.

656. Judgment of June 2007, N. v. Aegon Verzekering Leven, Supreme Court of the
Netherlands.

657. Judgment of 18 October 2004, District Court of The Hague, civil law section,
provision measures judge, Challenge No. 13/2004, Petition No. HA/ RK 2004.

658. Perenco Ecuador Limited v. Republic of Ecuador and Empresa Estatal Petróles
del Ecuador, ICSID Case No. ARB/08/16, PCA Case No IR-2009/1, Decision of 8
December 2009.

659. Niklas Lindstrom, ‘Challenge to Arbitrators – Decisions by the SCC Board

during 2008 – 2010, <www.sccinstitute.com/library/articles.aspx3>, 7–8.

660. Ibid., 9.

661. Ibid., 10.

662. Ibid., 14–16.

663. Ibid., 12.

664. Ibid., 16–19.

665. Ibid., 19–20.

666. Helena Jung,‘SCC Practice: Challenges to Arbitrators – SCC Board decisions

2005–2007’, <www.sccinstitute.se/filearchive/2/28190/04-Art32-J...>, 5.

667. Ibid., 7–8.

668. Helena Jung,‘SCC Practice: Challenges to Arbitrators – SCC Board decisions

2005–2007’, <www.sccinstitute.se/filearchive/2/28190/04-Art32-J...>, 7–8.

669. Ibid., 11.

670. Marie Ohrstrom, ‘Decisions of the SCC Institute Regarding Challenge of

Arbitrators’, <www.sccinstitute.se/filearchive/2/21256/Decisions_...>, 39.

671. Ibid.

672. Ibid., 48.

673. Ibid., 46.

674. Turner (East Asia) Pte Ltd v. Builders Federal (Hong Kong) Ltd & Anor (No 2)
[1988] SGHC 47.

675. Judgment of the Supreme Court of Sweden, [2007] 3 SIAR 167.

676. Judgment of 27 September 2011, KPMG AB v. ProfilGruppen AB, Svea Court

of Appeal, Case No. T 1085-11.

677. Judgment of 28 July 2010, X SA v. Y SA, Swiss Federal Supreme Court, Case
No. 4A_233/2010.

678. Judgment of 6 January 2010, Swiss Federal Supreme Court, ASA Bulletin 28,
no. 4 (2010): 772–777.

679. Judgment of 19 February 2009, Swiss Federal Supreme Court, ASA Bulletin
27, no. 4 (2009).

680. Judgment of 16 September 1988, Société Ligier et Société Diffusia v. Alfa

Lancia industriale, Swiss Federal Supreme Court, Revue de l'Arbitrage (1989):
505–509.

681. Judgment of 26 October 1966, Swiss Federal Court. See Sam Luttrell, Bias
Challenges in International Commercial Arbitration: The Need for a ‘Real Danger’
Test (Alphen aan den Rijn: Kluwer Law International, 2009), 113.

682. Judgment of 26 September 2007, Swiss Federal Supreme Court, ASA Bulletin
26, no. 1: 152–164.

683. Judgment of 14 December 2004, Swiss Federal Supreme Court, ASA Bulletin
23, no. 2 (2005): 321–336.

684. [2006] 1 CLC 656.

685. Re the Owners of the Steamship ‘Catalina’ and the Owners of the Steamship
‘Norma’ [1938] 61 Lloyd's Reports 362–363.

686. Laker Airways Incorporated v FLS Aerospace Limited [1999] 2 Lloyd's Law
Reports 45, 26 April 1999, QB.

687. [2000] All ER (Comm) 625.

688. [2000] 1 Lloyd's Reports 14.

689. Fletamentos Maritimos SA v. Effjohn International BV (No 2) [1997] 2 LIL Re.

690. National Shipping Co of Saudi Arabia v. Transamerican Shipping Corp US Dist

Lexis 18725.
691. Applied Industrial Materials Corp v. Ovalar Makine Ve Ticaret Sanayi, 492 F.
3d 132 CA 2 (NY).

692. Positive Software Solutions Inc. v. New Century Mortgage Corporation 476 F.
3d 278 (5th Cir 2007).

693. Applied Indus Materials Corp. v. Ovalar Makine Ticaret Ve Sanayi AS 492 F. 3d
132 (2d Cir 2007).

694. Lucent Technologies Inc. v. Tatung Co., 379 F. 3d 24 (2nd Cir 2004).

695. 393 US 145 (1968).

696. This was a 5-4 decision, and the judges who formed the majority did not use
the same reasoning.

page "359"

1 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,
1999), 453. See Model Law Arts 10(1) and 11(2); Swiss Private International
Law Art. 179; Belgian Judicial Code Art. 1682; Netherlands Code of Civil
Procedure Art. 1027; ss 15 and 16 of the English Arbitration Act; German Code of
Civil Procedure Art. 1035.
2 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 251.

3 Orkun Akseli, ‘Appointment of Arbitrators as Specified in the Agreement to

Arbitrate’, Journal of International Arbitration 20, no. 3 (2003): 248.

4 US Federal Arbitration Act s. 5; English Arbitration Act s. 15(1); UNCITRAL

Model Law Art. 11(2); Swiss Private International Law Art. 179; French New
Code of Civil Procedure Art. 1508(2); German Code of Civil Procedure s. 1035(1).
5 Jan Paulsson is one of few to suggest the abolition of party-appointed
arbitrators. See Jan Pauls-son, ‘Moral Hazard in International Dispute
Resolution’, Inaugural Lecture as Holder of the Michael R. Klein Distinguished
Scholar Chair, University of Miami School of Law, 29 April 2010, 8. The text of the
address is available at <www.arbitration-
icca.org/media/0/12773749999020/pa...>. Professor Paulsson's paper provides
a number of interesting examples of egregious behaviour by party-appointed
arbitrators. See also the response to Professor Paulsson's paper in Alexis Mourre,
‘Are Unilateral Appointments Defensible? On Jan Paulsson's Moral Hazard in
International Arbitration’, in Liber Amicorum Eric Bergsten: International
Arbitration and International Commercial Law: Synergy, Convergence and
Evolution, ed. Kröll et al. (Alphen aan den Rijn, Kluwer Law International, 2011).
6 Jan Paulsson, ‘Ethics, Elitism, Eligibility’, Journal of International Arbitration 14,

no. 4 (1997): 18.

7 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on
International Commercial Arbitration (The Hague: Kluwer Law International,
1999), 463 & 474. See Art. V(1)(d) of the New York Convention.
8 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 251.

9 ICC Rules 2012 Art. 12(3); LCIA Rules Arts 5.5 and 7.2; AAA Rules Art. 6(3);

CIETAC Rules Arts 24–26; ACICA Rules 9.2; HKIAC Rules Art. 8.1(a); SIAC Rules
Art. 8.2; SCC Rules Art. 13(3).
10 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1439.

11 See Orkun Akseli, ‘Appointment of Arbitrators as Specified in the Agreement

to Arbitrate’, Journal of International Arbitration 20, no. 3 (2003): 251; Gary B.

Born, International Commercial Arbitration (The Hague: Kluwer Law
International, 2009), 624.
12 ICC Rules 2012 Art. 5(2).
13 Section 17 English Arbitration Act 1996.
14 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 469.
15 For example CIETAC Rules Art. 24 provides that parties ‘shall nominate

arbitrators from the Panel of Arbitrators provided by CIETAC’, but allows

arbitrators who are not on this panel to be selected if they are confirmed by the
Chairman of CIETAC.
16 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 458, citing the case of Philipp Brothers v Drexel, 1990, Revue de l'Arbitrage,
497.
17 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 456.
18 See also Netherlands Code of Civil Procedure Art. 1023; Belgian Judicial Code

1680; Italian Code of Civil Procedure 812.

19 For discussion of this, see Jan Paulsson, ‘Moral Hazard in International Dispute

Resolution’, Inaugural Lecture as Holder of the Michael R. Klein Distinguished

Scholar Chair, University of Miami School of Law, 29 April 2010, 10.
20 UNCITRAL Rules Art. 6(7); LCIA Rules Art. 6.1.
21 W. Laurence Craig, William W. Park & Jan Paulsson, International Chamber of

Commerce Arbitration, 3rd edn (Dobbs Ferry: Oceana, 2000), 193.

22 [2011] UKSC 40.
23 Pierre Lalive, ‘On the Neutrality of the Arbitrator and the Place of Arbitration’,

in Swiss Essays on International Arbitration (1984), 27.

24 Ibid., 27–28.
25 Chinese Arbitration Law Art. 13. See Lawrence Boo, ‘Appointment of the

Arbitral Tribunal’, in The Asian Leading Arbitrators' Guide to International

Arbitration, ed. Michael Pryles & Michael Moser (Huntington: JurisNet LLC,
2007), 85.
26 Section 3 of the Rules for the Implementation of the Saudi Arabian Arbitration

Regulation.
27 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1444.

28 See Belgian Judicial Code Art. 1680; Spanish Arbitration Act Art. 13; Italian
Code of Civil Procedure Art. 812; Korean Arbitration Act Art. 5(6).
29 US judges, for example, are prevented by the ABA Code of Judicial Conduct

Canon 5(E).
30 French Code of Civil Procedure Art. 1450.
31 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1448, n. 498.

32 Belgian Judicial Code Art. 1678.
33 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1452.

34 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 258; Emmanuel Gaillard & John Savage
(eds), Fouchard, Gaillard, Goldman on International Commercial Arbitration (The
Hague: Kluwer Law International, 1999), 458.
35 Lawrence Boo, ‘Appointment of the Arbitral Tribunal’, in The Asian Leading

Arbitrators' Guide to International Arbitration, ed. Michael Pryles & Michael

Moser (Huntington: JurisNet LLC, 2007), 80–81.
36 Owners of The ‘Myron’ v. Tradax Export [1970] 1 QB 527.
37 Such agreements have been enforced by national courts. See, e.g., Aviall Inc. v.

Ryder System Inc., 913 F. Supp. 826 (SDNY 1996). Of course, this does not
necessarily mean that the person will be willing or able to arbitrate. Where the
named arbitrator does or cannot accept the nomination, the arbitral rules or lex
arbitri have to be relied on to find a replacement if they are broad enough in that
regard. See Orkun Akseli, ‘Appointment of Arbitrators as Specified in the
Agreement to Arbitrate’, Journal of International Arbitration 20, no. 3 (2003):
248.
38 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1457.

39 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 259.

40 Ibid., 259–260.
41 Lawrence Boo, ‘Appointment of the Arbitral Tribunal’, in The Asian Leading

Arbitrators' Guide to International Arbitration, ed. Michael Pryles & Michael

Moser (Huntington: JurisNet LLC, 2007), 82; Qrkun Akseli, ‘Appointment of
Arbitrators as Specified in the Agreement to Arbitrate’, Journal of International
Arbitration 20, no. 3 (2003): 249.
42 Ibid., 81.
43 Judgment of 20 February 2011, Oberlandesgericht Dresden (Higher Regional

Court of Dresden), 11 Sch 2/00.

44 See, e.g., English Arbitration Act 1996 s. 15(3).
45 See, e.g., UNCITRAL Model Law Art. 10(2); Belgian Judicial Code Art. 1682;

Swedish Arbitration Act s. 13.

46 For example, Netherlands Code of Civil Procedure Art. 1026(2).
47 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 252.

48 ICC Rules 2012 Art. 12(3).
49 See Christopher Kee, ‘The Evolving Role of an Appointing Authority’, in Liber

Amicorum Eric Bergsten: International Arbitration and International Commercial

Law: Synergy, Convergence and Evolution, ed. Kröll et al. (Alphen aan den Rijn:
Kluwer Law International, 2011). The cost tends to be around USD 1,000.
50 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1416. See also AAA Commercial Rules, rule 11.
51 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1413.

52 Born suggests that the LCIA tends to try to constitute the tribunal too quickly,

without consulting the parties. Gary B. Born, International Commercial

Arbitration (The Hague: Kluwer Law International, 2009), 1415.
53 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1416–1417.

54 See Robert Coulson, ‘Survey of International Arbitration Procedures’, in
Towards a Science of International Arbitration: Collected Empirical Research, ed.
Christopher R. Drahozal & Richard W. Naimark (The Hague: Kluwer Law
International, 2005), 95, 97–98.
55 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1417.

56 W. Laurence Craig, William W. Park & Jan Paulsson, International Chamber of

Commerce Arbitration, 3rd edn (Dobbs Ferry: Oceana, 2000), 197.

57 Ibid.
58 Ibid., 190–191.
59 ICC Rules 2012 Art. 12(2); LCIA Rules Art. 5.4; SIAC Rules Art. 6.1; Swiss Rules

2012 Art. 6.2.

60 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1362–1363.

61 Jason Fry & Simon Greenberg, ‘The Arbitral Tribunal: Applications of Articles

7–12 of the ICC Rules in Recent Cases’, ICC International Court of Arbitration
Bulletin 20, no. 2 (2009): 15.
62 Ibid.
63 Christopher Newmark, ‘Controlling Time and Costs in Arbitration’, in The

Leading Arbitrators' Guide to International Arbitration, ed. Lawrence W. Newman

& Richard D. Hill (Huntington: Juris Pub, 2008), 91.
64 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 252–253.

65 Orkun Akseli, ‘Appointment of Arbitrators as Specified in the Agreement to

Arbitrate’, Journal of International Arbitration 20, no. 3 (2003): 249.

66 Ibid.
67 Ibid.
68 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1356.

69 See, e.g., ICC Rules Art 12(4); HKIAC Rules 8.1(b); ACICA Rules Art 10.3.
70 Gerald Aksen, ‘The Tribunal's Appointment’, in The Leading Arbitrators' Guide

to International Arbitration, ed. Lawrence W. Newman & Richard D. Hill

(Huntington: Juris Pub, 2008), 36.
71 Ibid.
72 Ibid., 37.
73 Ibid., 36.
74 W. Laurence Craig, William W. Park & Jan Paulsson, International Chamber of

Commerce Arbitration, 3rd edn (Dobbs Ferry: Oceana, 2000), 192.

75 Ibid., 193.
76 Gerald Aksen, ‘The Tribunal's Appointment’, in The Leading Arbitrators' Guide
to International Arbitration, ed. Lawrence W. Newman & Richard D. Hill
(Huntington: Juris Pub, 2008), 38.
77 Ibid., 39.
78 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1404.

79 Ibid., 1405.
80 Christopher R. Seppälä, ‘Obtaining the Right International Arbitral Tribunal: A

Practitioner's View’, Mealey's International Arbitration Report 22, no. 10 (2007):

34.
81 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1403.

82 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 254.

83 Ibid.
84 Qrkun Akseli, ‘Appointment of Arbitrators as Specified in the Agreement to

Arbitrate’, Journal of International Arbitration 20, no. 3 (2003): 250.

85 See, e.g., French Code of Civil Procedure Art. 1452; UNCITRAL Model Law Art.

11; Swiss Law on Private International Law Art. 179(2).

86 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1426; Orkun Akseli, ‘Appointment of Arbitrators as

Specified in the Agreement to Arbitrate', Journal of International Arbitration 20,
no. 3 (2003): 250. For a case in which the ICC declined to act as the appointing
authority, see Astra Footwear Industry v. Harwyn International, Inc. 442 F. Supp.
907 (SDNY 1978).
87 A recent example is in Australia, where ACICA has been made the appointing

authority under Australia's arbitration law.

88 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1433.

89 Ibid., 1434.
90 See, e.g., UNCITRAL Model Law Art. 11(4); US Federal Arbitration Act s. 5.
91 Orkun Akseli, ‘Appointment of Arbitrators as Specified in the Agreement to

Arbitrate’, Journal of International Arbitration 20, no. 3 (2003): 251.

92 Ibid.
93 Ibid.
94 Ibid. See CAE Industries Ltd v. Aerospace Holdings Co., 741 F. Supp. 388, 392–

393 (SDNY 1989).

95 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 251.

96 Orkun Akseli, ‘Appointment of Arbitrators as Specified in the Agreement to

Arbitrate’, Journal of International Arbitration 20, no. 3 (2003): 252.

97 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1418. Aguilar Alvarez talks of the risk of court decisions
being influenced by each country's favourable or unfavourable attitude to
arbitration. See Guiliermo Aguilar Alvarez, ‘The Challenge of Arbitrators’,
Arbitration International 6, no. 3 (1990): 205.
98 Guiliermo Aguilar Alvarez, ‘The Challenge of Arbitrators’, Arbitration
International 6, no. 3 (1990): 208.
99 See, e.g., English Arbitration Act 1996 ss 2(4) and 18.
100 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law
International, 2009), 1435.
101 Judgment of 29 March 2001, Paris Cour d'appel (Paris Court of Appeal),

reported in Mealey's International Arbitration Report 17, no. 6 (2002).

102 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1431.

103 UNCITRAL Model Law Art. 11(5).
104 UNCITRAL Rules 2010 Art. 6(1).
105 UNCITRAL Rules 2010 Art. 6(3).
106 Thomas H. Webster, Handbook of UNCITRAL Arbitration (London: Sweet &

Maxwell, 2011), 103.

107 UNCITRAL Rules 2010 Art. 7(1).
108 UNCITRAL Rules 2010 Art. 14(2).
109 Christopher Kee, ‘The Evolving Role of an Appointing Authority’, in Liber

Amicorum Eric Bergsten: International Arbitration and International Commercial

Law: Synergy, Convergence and Evolution, ed. Kröll et al. (Alphen aan den Rijn:
Kluwer Law International, 2011), 310.
110 W Laurence Craig, William W. Park & Jan Paulsson, International Chamber of

Commerce Arbitration, 3rd edn (Dobbs Ferry: Oceana, 2000), 199.

111 Ibid.
112 Decision of 7 January 1992, Sociétés BKMI et Siemens v. société Dutco, Cour de

cassation, Revue de l'Arbitrage (1992) 470.

113 See French New Code of Civil Procedure Arts 1514 and 1520(5).
114 Judgment of 5 May 1989, Paris Cour d'appel (Paris Court of Appeal) Revue de

l'Arbitrage (1989), 723.

115 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 202.

116 Bernard Hanotiau, Complex Arbitrations (The Hague: Kluwer Law
International, 2005), 201.
117 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 201.

118 Bernard Hanotiau, Complex Arbitrations (The Hague: Kluwer Law
International, 2005), 201.
119 Ibid., 207.
120 Yves Derains & Eric A. Schwartz, A Guide to the ICC Rules of Arbitration, 2nd

edn (The Hague: Kluwer Law International, 2005), 181.

121 Bernard Hanotiau, Complex Arbitrations (The Hague: Kluwer Law
International, 2005), 201. See ICC Rules 2012 Art. 12(6) and (8); LCIA Rules Art.
8.1; AAA Rules Art. 6(5); WIPO Rules Art. 18; Stockholm Arbitration Rules Art.
13. See also W. Laurence Craig, William W. Park & Jan Paulsson, International
Chamber of Commerce Arbitration, 3rd edn (Dobbs Ferry: Oceana, 2000), 198–
201.
122 See, e.g., Art. 10(1) of the Model Law; s. 15(1) English Arbitration Act; Swiss

Law on Private International Law Art. 179(1); Netherlands Code of Civil

Procedure Art. 1026(2); Swedish Arbitration Act s. 12; German Code of Civil
Procedure s. 1034(1); Spanish Arbitration Act Art. 12.
123 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 248.

124 Netherlands Code of Civil Procedure Art. 1026(1); Belgian Judicial Code Art.
1681. See also Gary B. Born, International Commercial Arbitration (The Hague:
Kluwer Law International, 2009), 1352.
125 See, e.g., s. 15(2) of the English Arbitration Act, which provides for such a

conversion, unless otherwise agreed by the parties.

126 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1358.

127 Around61% of ICC cases in 2008 had a three-member panel. In 93.5% of

cases, the decision to choose three arbitrators was made by the parties. See Jason
Fry & Simon Greenberg, ‘The Arbitral Tribunal: Applications of Articles 7–12 of
the ICC Rules in Recent Cases’, ICC International Court of Arbitration Bulletin 20,
no. 2 (2009): 15.
128 Wendy Miles, ‘International Arbitrator Appointment’, Dispute Resolution
Journal 57, no. 3 (2002): 40; White & Case and Queen Mary School of
International Arbitration, ‘2010 International Arbitration Survey: Choices in
International Arbitration’, <www.arbitrationon-
line.org/docs/2010_International...>, October 2010. See also Jennifer Kirby,
‘With Arbitrators, Less Can Be More: Why Conventional Wisdom on the Benefits
of Having Three Arbitrators May Be Overrated’, Journal of International
Arbitration 26, no. 3 (2009): 337.
129 See Toby Landau, ‘Composition and Establishment of the Tribunal’, American

Review of International Arbitration 9 (1998): 50.

130 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 248. See also Gary B. Born, International
Commercial Arbitration (The Hague: Kluwer Law International, 2009), 1354,
1358.
131 See ICC Rules 2012 Art. 12(2); LCIA Rules Art. 5.4; ICDR Rules Art. 5.
132 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 249.

133 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1359.

134 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 250.

135 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1359.

136 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 249. See also Gary B. Born, International
Commercial Arbitration (The Hague: Kluwer Law International, 2009), 1355.
Born notes that an ‘idiosyncratic’ decision is less likely.
137 Lawrence Boo, ‘Appointment of the Arbitral Tribunal’, in The Asian Leading

Arbitrators' Guide to International Arbitration, ed. Michael Pryles & Michael

Moser (Huntington: JurisNet LLC, 2007), 78–79.
138 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 250.

139 Christopher Newmark, ‘Controlling Time and Costs in Arbitration’, in The

Leading Arbitrators' Guide to International Arbitration, ed. Lawrence W. Newman

& Richard D. Hill (Huntington: Juris Pub, 2008), 93.
140 Jennifer Kirby, ‘With Arbitrators, Less Can Be More: Why the Conventional

Wisdom on the Benefits of Having Three Arbitrators May Be Overrated’, Journal

of International Arbitration 26, no. 3 (2009): 337.
141 Gary B. Born, International Arbitration: Cases & Materials (Frederick, MD:
Aspen, 2010), 631.
142 Gerald Aksen, ‘The Tribunal's Appointment’, in The Leading Arbitrators' Guide

to International Arbitration, ed. Lawrence W. Newman & Richard D. Hill

(Huntington: Juris Pub, 2008), 37; Gary B. Born, International Commercial
Arbitration (The Hague: Kluwer Law International, 2009), 1389.
143 Eric A. Schwartz, ‘The Rights and Duties of ICC Arbitrators’, in The Status of

the Arbitrator, ICC International Court of Arbitration Bulletin: 1995 Special

Supplement, ed. Jean-François Bourque (Paris: ICC Publishing, 1995), 69, n. 14.
144 Emilia Onyema, ‘Empirically Determined Factors in Appointing Arbitrators in

International Arbitration’, Arbitration 73, no. 2 (2007): 205–206.

145 Lord Hacking, ‘Arbitration Is Only as Good as Its Arbitrators’, in Liber
Amicorum Eric Bergsten: International Arbitration and International Commercial
Law: Synergy, Convergence and Evolution, ed. Kröll et al. (Alphen aan den Rijn:
Kluwer Law International, 2011).
146 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1386.

147 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 261.

148 Lowenfeld, ‘The Party-Appointed Arbitrator: Further Reflections’, Leading

Arbitrators' Guide, 46.

149 Ibid.
150 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 262.

151 Ibid., 263.
152 Ibid., 246.
153 W. Laurence Craig, William W. Park & Jan Paulsson, International Chamber of

Commerce Arbitration, 3rd edn (Dobbs Ferry: Oceana, 2000), 194.

154 Jennifer Kirby, ‘With Arbitrators, Less Can Be More: Why the Conventional

Wisdom of Having Three Arbitrators May Be Overrated’, Journal of International

Arbitration 26, no. 3 (2009): 354.
155 See Ibid.
156 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 459.
157 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1454.

158 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 260.

159 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 464.
160 See, e.g., Doak Bishop & Lucy Reed, ‘Practical Guidelines for Interviewing,

Selecting and Challenging Party-Appointed Arbitrators in International

Commercial Arbitration’, Arbitration International 14, no. 4 (1998): 400–401.
161 Lucy F. Reed & Alexander A. Yanos, ‘Pre-hearing Advocacy in International

Arbitration’, in The Art of Advocacy in International Arbitration, ed. Doak Bishop

& Edward G. Kehoe, 2nd edn (Huntington, NY: JurisNet, 2010), 340.
162 W. Laurence Craig, William W. Park & Jan Paulsson, International Chamber of
Commerce Arbitration, 3rd edn (Dobbs Ferry: Oceana, 2000), 196.
163 Martin Hunter, ‘Ethics of an International Arbitrator’, Arbitration 53 (1987):

219, 233; Ahmed S. El-Kosheri & Karim Y. Youssef, ‘L'indépendence des arbitres
internationaux: le point de vue d'un arbitre’, ICC Special Supplement 2007, 51
refers to a teacher who suggests to his students that the only factor to consider in
appointing an arbitrator is how sympathetic he or she will be to one's case.
164 Charles J. Moxley Jr, ‘Selecting the Ideal Arbitrator’, Dispute Resolution Journal

60, no. 3 (2005): 25–26.

165 See Christopher Seppälä, ‘Obtaining the Right International Arbitral Tribunal:

A Practitioner's View’, Mealey's International Arbitration Report 22, no. 10

(2007).
166 See, e.g., Horacio A. Grigera Naón, ‘Factors to Consider in Choosing an

Efficient Arbitrator’, in Improving the Efficiency of Arbitration Agreements and

Awards: 40 Years of Application of the New York Convention, ICCA Congress Series
No. 9, ed. Albert Jan van den Berg (The Hague: Kluwer Law International, 1999):
307–308; W. Laurence Craig, William W. Park & Jan Paulsson, International
Chamber of Commerce Arbitration, 3rd edn (Dobbs Ferry: Oceana, 2000), 196.
167 Lucy F. Reed & Alexander A. Yanos, ‘Pre-Hearing Advocacy in International

Arbitration’, in The Art of Advocacy in International Arbitration, ed. Doak Bishop

& Edward G. Kehoe, 2nd edn (Huntington, NY: JurisNet, 2010), 339.
168 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1389.

169 Peter Michaelson, ‘Enhancing Arbitrator Selection: Using Personality
Screening to Supplement Conventional Selection Criteria for Tripartite
Arbitration Tribunals’, Arbitration 76, no. 1 (2010): 99.
170 Paulo Michele Patocchi & Harold Frey Brentano, ‘The Provisional Timetable

in International Arbitration’, in Global Reflections on International Law,

Commerce and Dispute Resolution: Liber Amicorum in Honour of Robert Briner, ed.
Gerald Aksen et al. (ICC Publishing, 2005), 63.
171 Andreas F. Lowenfeld, ‘The Party-Appointed Arbitrator in International
Controversies: Some Reflections’, Mealey's International Arbitration Report 11,
no. 11 (1996): 31.
172 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1393.

173 Lucy F. Reed & Alexander A. Yanos, ‘Pre-hearing Advocacy in International

Arbitration’, in The Art of Advocacy in International Arbitration, ed. Doak Bishop

& Edward G. Kehoe, 2nd edn (Huntington, NY: JurisNet, 2010), 340.
174 Gerald Aksen, ‘The Tribunal's Appointment’, in The Leading Arbitrators' Guide

to International Arbitration, ed. Lawrence W. Newman & Richard D. Hill

(Huntington: Juris Pub, 2008), 35–36. See also Lucy F. Reed & Alexander A.
Yanos, ‘Pre-hearing Advocacy in International Arbitration’, in The Art of Advocacy
in International Arbitration, ed. Doak Bishop & Edward G. Kehoe, 2nd edn
(Huntington, NY: JurisNet, 2010), 340. Reed and Yanos call interviewing a
‘standard’ practice.
175 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 334.

176 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 264; Doak Bishop & Lucy Reed,
‘Practical Guidelines for Interviewing, Selecting and Challenging Party-Appointed
Arbitrators in International Commercial Arbitration’, Arbitration International
14, no. 4 (1998): 423.
177 Gerald Aksen, ‘The Tribunal's Appointment’, in The Leading Arbitrators' Guide

to International Arbitration, ed. Lawrence W. Newman & Richard D. Hill

(Huntington: Juris Pub, 2008), 35–36.
178 See Gerald Aksen, ‘The Tribunal's Appointment’, in The Leading Arbitrators'

Guide to International Arbitration, ed. Lawrence W. Newman & Richard D. Hill

(Huntington: Juris Pub, 2008); Jean-François Poudret & Sébastien Besson,
Comparative Law of International Arbitration, 2nd edn (London: Sweet &
Maxwell, 2007), 334; Nigel Blackaby et al., Redfern and Hunter on International
Arbitration, 5th edn (Oxford: Oxford University Press, 2009), 264–265; W.
Laurence Craig, William W. Park & Jan Paulsson, International Chamber of
Commerce Arbitration, 3rd edn (Dobbs Ferry: Oceana, 2000), 213.
179 IBA Guidelines on Conflicts of Interest, 24; CIArb, ‘Practice Guideline 16: The

Interviewing of Prospective Arbitrators’, para. 1.3.

180 W. Laurence Craig, William W. Park & Jan Paulsson, International Chamber of

Commerce Arbitration, 3rd edn (Dobbs Ferry: Oceana, 2000), 213. This is
reflected in the IBA Guidelines on Conflicts of Interest, 24.
181 Gerald Aksen, ‘The Tribunal's Appointment’, in The Leading Arbitrators' Guide

to International Arbitration, ed. Lawrence W. Newman & Richard D. Hill

(Huntington: Juris Pub, 2008).
182 Ibid.; CIArb Guidelines; Doak Bishop & Lucy Reed, ‘Practical Guidelines for

Interviewing, Selecting and Challenging Party-Appointed Arbitrators in

International Commercial Arbitration’ , Arbitration International 14, no. 4
(1998); Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th
edn (Oxford: Oxford University Press, 2009), 264–265.
183 Gerald Aksen, ‘The Tribunal's Appointment’, in The Leading Arbitrators' Guide

to International Arbitration, ed. Lawrence W. Newman & Richard D. Hill

(Huntington: Juris Pub, 2008), 35–36.
184 CIArb Guideline No 10.
185 Lucy F. Reed & Alexander A. Yanos, ‘Pre-hearing Advocacy in International

Arbitration’, in The Art of Advocacy in International Arbitration, ed. Doak Bishop

& Edward G. Kehoe, 2nd edn (Huntington, NY: JurisNet, 2010), 340.
186 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1394; Nigel Blackaby et al., Redfern and Hunter on

International Arbitration, 5th edn (Oxford: Oxford University Press, 2009), 265;
CIArb Guideline No. 15.
187 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 265; CIArb Guidelines, Guideline No. 15.
188 Ibid.
189 Doak Bishop & Lucy Reed, ‘Practical Guidelines for Interviewing, Selecting

and Challenging Party-Appointed Arbitrators in International Commercial

Arbitration’, Arbitration International 14, no. 4 (1998): 424.
190 Gerald Aksen, ‘The Tribunal's Appointment’, in The Leading Arbitrators' Guide

to International Arbitration, ed. Lawrence W. Newman & Richard D. Hill

(Huntington: Juris Pub, 2008); Nigel Blackaby et al., Redfern and Hunter on
International Arbitration, 5th edn (Oxford: Oxford University Press, 2009), 265.
191 Dominique Hascher, ‘ICC Practice in Relation to the Appointment,
Confirmation, Challenge and Replacement of Arbitrators’, ICC International Court
of Arbitration Bulletin 6, no. 2 (1995): 7–8.
192 CIArb Guidelines, No. 3.
193 Gerald Aksen, ‘The Tribunal's Appointment’, in The Leading Arbitrators' Guide

to International Arbitration, ed. Lawrence W. Newman & Richard D. Hill

(Huntington: Juris Pub, 2008); Redfern and Hunter (2009), 265.
194 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 265.

195 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1393.

196 Mark Friedman, ‘Regulating Judgment: A Comment on the Chartered Institute

of Arbitrators Guidelines on the Interviewing of Prospective Arbitrators’, Dispute

Resolution International (2008): 288, 289–290.
197 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 334.

198 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 265; CIArb Guideline No. 4.

199 CIArb Guideline No. 1; Doak Bishop & Lucy Reed, ‘Practical Guidelines for

Interviewing, Selecting and Challenging Party-Appointed Arbitrators in

International Commercial Arbitration’, Arbitration International 14, no. 4 (1998):
425.
200 CI Arb Guideline No. 6; Lowenfeld, ‘The Party-Appointed Arbitrator: Some

Reflections’ (1995): 62.

201 Gerald Aksen, ‘The Tribunal's Appointment’, in The Leading Arbitrators' Guide

to International Arbitration, ed. Lawrence W. Newman & Richard D. Hill

(Huntington: Juris Pub, 2008), 35–36.
202 Gerald Aksen, ‘Reflections of an International Arbitrator’, Arbitration
International 23, no. 2 (2007): 256.
203 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1391.

204 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 529–530.
205 Eric A. Schwartz, ‘The Rights and Duties of ICC Arbitrators’, in The Status of

the Arbitrator, ICC International Court of Arbitration Bulletin: 1995 Special

Supplement, ed. Jean-François Bourque (Paris: ICC Publishing, 1995), 70.
206 IBA Guidelines on Conflicts of Interest, 5.
207 Gerald Aksen, ‘The Tribunal's Appointment’, in The Leading Arbitrators' Guide

to International Arbitration, ed. Lawrence W. Newman & Richard D. Hill

(Huntington: Juris Pub, 2008), 33.
208 Karl-Heinz Böckstiegel, ‘Case Management by Arbitrators: Experiences and

Suggestions’, in Global Reflections on International Law, Commerce and Dispute

Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al.
(ICC Publishing, 2005), 116.
209 Some national laws require acceptance to be in writing. See, e.g., Netherlands

Code of Civil Procedure Art. 1029(1).

210 See section 5.10 below.
211 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1388.

212 David W. Rivkin, ‘Towards a New Paradigm in International Arbitration: The
Town Elder Model Revisited’, Arbitration International 24, no. 3 (2008): 379.
213 Ibid.
214 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 242.

215 See ICC Rules for Pre-Arbitral Referee Procedure and section 8.6.
216 ICC Rules for Pre-Arbitral Referee Procedure Art. 4.4.
217 See Netherlands Arbitration Institute Rules Art. 42a.
218 Netherlands Arbitration Institute Rules Arts 19 and 42f. See also section 6.4.
219 See LCIA Rules Art. 9.
220 See ICDR Rules Art. 37; SIAC Rules, Schedule 1; ACICA Rules, Schedule 2; SCC

Rules, Appendix II.

221 See SIAC Rules, Schedule 1 Art. 3; ICDR Rules Art. 37(c); ACICA Rules,

Schedule 2 Art. 2.1; SCC Rules, Appendix II Art. 4(3).

222 See, e.g., SCC Rules, Appendix II Art. 9(5); ACICA Rules, Schedule 2 Art. 5.2.
223 Jan Paulsson, ‘Ethics, Elitism, Eligibility’, Journal of International Arbitration

14, no. 4 (1997): 13.

224 Markham Ball, ‘Probity Deconstructed: How Helpful, Really, Are the New

International Bar Association Guidelines on Conflicts of Interest in International

Arbitration?’, Arbitration International 21, no. 3 (2005): 326–327.
225 See, e.g., Markham Ball, ‘Probity Deconstructed: How Helpful, Really, Are the

New International Bar Association Guidelines on Conflicts of Interest in

International Arbitration?’, Arbitration International 21, no. 3 (2005): 326–327.
Certainly, the number of challenges – opportunistic or otherwise – has been
increasing in institutional arbitrations. See Gary B. Born, International
Arbitration: Cases and Materials (Frederick, MD: Aspen, 2011), 705.
226 Sam Luttrell, Bias Challenges in International Commercial Arbitration: The

Need for a ‘Real Danger’ Test (Alphen aan den Rijn: Kluwer Law International,
2009), 4.
227 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 272–273; Markham Ball, ‘Probity

Deconstructed: How Helpful, Really, Are the New International Bar Association
Guidelines on Conflicts of Interest in International Arbitration?’, Arbitration
International 21, no. 3 (2005): 326.
228 IBA Guidelines on Conflicts (2004), 7.
229 See, e.g., Jason Fry & Simon Greenberg, ‘The Arbitral Tribunal: Applications of

Articles 7–12 of the ICC Rules in Recent Cases’, ICC International Court of
Arbitration Bulletin 20, no. 2 (2009): 23, n. 23. From 1998 to 2008, 9,605
arbitrators were appointed in ICC arbitrations. There were only 316 challenges,
and only 21 of those were accepted. Those statistics do not take into account
situations where confirmation of the arbitrator was refused by the ICC (as
distinct from a challenge post-appointment), but the numbers are nonetheless
very low. Mark-ham Ball, ‘Probity Deconstructed: How Helpful, Really, Are the
New International Bar Association Guidelines on Conflicts of Interest in
International Arbitration?’, Arbitration International 21, no. 3 (2005): 326.
230 W. Laurence Craig, William W. Park & Jan Paulsson, International Chamber of

Commerce Arbitration, 3rd edn (Dobbs Ferry: Oceana, 2000), 212.

231 UNCITRAL Model Law Art. 12; US Federal Arbitration Act s. 10(a)(2);
Swedish Arbitration Act s. 8; English Arbitration Act ss 33(1) and 68; Belgian
Judicial Code Art. 1690(1); Netherlands Code of Civil Procedure Art. 1033(1)
Some jurisdictions, such as France, provide relatively little indication of grounds
for challenge. See French New Code of Civil Procedure Art. 1456. The Swiss
Private International Law is silent on impartiality.
232 UNCITRAL Rules Art. 10.1; ICC Rules 2012 Art. 14; ACICA Rules Art. 13;

HKIAC Rules Art. 11.1.

233 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 267; Howard M. Holtzmann & Joseph E.
Neuhaus, A Guide to the UNCITRAL Model Law on International Commercial
Arbitration: Legislative History and Commentary (Deventer: Kluwer Law and
Taxation Publishers, 1989), 390.
234 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 95.
235 Doak Bishop & Lucy Reed, ‘Practical Guidelines for Interviewing, Selecting

and Challenging Party-Appointed Arbitrators in International Commercial

Arbitration’, Arbitration International 14, no. 4 (1998): 398; ICSID Rules, Rule
6(2).
236 See Christoper Seppälä, ‘Obtaining the Right International Arbitral Tribunal:

A Practitioner's View’, Mealey's International Arbitration Report 22, no. 10

(2007).
237 Dominique Hascher, former Deputy Secretary-General and General Counsel

of the ICC, has said that this formulation is broad enough to include impartiality.
See Hascher, ‘ICC Practice in Relation to the Appointment, Confirmation,
Challenge and Replacement of Arbitrators’, ICC International Court of Arbitration
Bulletin 6, no. 4 (1995): 6.
238 ICC Rules 2012 Art. 14(1).
239 Departmental Advisory Committee on Arbitration Law (Chairman, the Rt Hon

Lord Justice Saville), Report on the Arbitration Bill, paras 101–104 (February
1996).
240 One notable example of this is Art. 14(1) of the ICSID Convention: the English

and French versions refer to independence, while the Spanish version refers to
impartiality.
241 See, e.g., Markham Ball, ‘Probity Deconstructed: How Helpful, Really, Are the

New International Bar Association Guidelines on Conflicts of Interest in

International Arbitration?’, Arbitration International 21, no. 3 (2005): 330; Sam
Luttrell, Bias Challenges in International Commercial Arbitration: The Need for a
‘Real Danger' Test (Alphen aan den Rijn: Kluwer Law International, 2009), 21.
242 See IBA Rules of Ethics for International Arbitrators, para. 3.1; Gary B. Born,

International Commercial Arbitration (The Hague: Kluwer Law International,

2009), 1618; Markham Ball, ‘Probity Deconstructed: How Helpful, Really, Are the
New International Bar Association Guidelines on Conflicts of Interest in
International Arbitration?’, Arbitration International 21, no. 3 (2005): 330;
Mohammed Bedjaoui, ‘Challenge of Arbitrators’, in International Arbitration in a
Changing World, ICCA Congress Series No. 6, ed. Albert Jan van den Berg
(Deventer: Kluwer Law and Taxation Publishers, 1994), 91.
243 See IBA Rules of Ethics for International Arbitrators, para. 3.1; Gary B. Born,

International Commercial Arbitration (The Hague: Kluwer Law International,

2009), 1618; Markham Ball, ‘Probity Deconstructed: How Helpful, Really, Are the
New International Bar Association Guidelines on Conflicts of Interest in
International Arbitration?’, Arbitration International 21, no. 3 (2005): 330.
244 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 564.
245 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 267–268.

246 Sam Luttrell, Bias Challenges in International Commercial Arbitration: The

Need for a ‘Real Danger’ Test (Alphen aan den Rijn: Kluwer Law International,
2009), 21.
247 See, e.g., UNCITRAL Model Law Art. 12(2).
248 See Decision of 11 January 1995, XXII Yearbook of Commercial Arbitration

227 (1997).
249 Sam Luttrell, Bias Challenges in International Commercial Arbitration: The

Need for a ‘Real Danger’ Test (Alphen aan den Rijn: Kluwer Law International,
2009), 7.
250 However, there are conflicting cases in investment arbitration. See Compañia

de Aguas del Aconquija S.A. & Vivendi Universal S.A. v. Argentine Republic, ICSID
Case No. ARB/97/3, Decision on the Challenge of the President of the Committee
(3 October 2001); Amco Asia Corp v. Indonesia ICSID Case No. ARB/81/8,
Decision on the Proposal to Disqualify an Arbitrator (24 June 1982). See also
Constantine Partasides, ‘The Selection, Appointment and Challenge of
Arbitrators’, Vindobona Journal of International Commercial Law and Arbitration
5, no. 2 (2001): 222–223.
251 See Sam Luttrell, Bias Challenges in International Commercial Arbitration: The

Need for a ‘Real Danger’ Test (Alphen aan den Rijn: Kluwer Law International,
2009), 8.
252 This test originally comes from the English case R v. Sussex Justices, Ex parte

McCarthy [1924] 1 KB 256.

253 This test comes from Porter v. Magill [2002] 2 AC 357.
254 Sam Luttrell, Bias Challenges in International Commercial Arbitration: The

Need for a ‘Real Danger’ Test (Alphen aan den Rijn: Kluwer Law International,
2009), 8. For a thorough discussion of the US case law, see Shore, ‘Disclosure and
Impartiality: An Arbitrator's Responsibility Vis-à-Vis Legal Standards', Dispute
Resolution Journal 57, no. 4 (2006): 34.
255 R v. Gough [1993] UKHL 1.
256 See generally, Sam Luttrell, Bias Challenges in International Commercial

Arbitration: The Need for a ‘Real Danger’ Test (Alphen aan den Rijn: Kluwer Law
International, 2009), Australia has adopted this test in its arbitration legislation.
See s. 18A Australian International Arbitration Act.
257 A number of other codes of ethics exist for domestic and international

arbitrators. The ABA and AAA had published a code of ethics a few months prior
to the IBA. Broadly, the ABA/AAA code is similar to that of the IBA. In 2002,
California published the Ethics Standards for Neutral Arbitrators in Contractual
Arbitration, which is compulsory in domestic arbitrations. See Markham Ball,
‘Probity Deconstructed: How Helpful, Really, Are the New International Bar
Association Guidelines on Conflicts of Interest in International Arbitration?’,
Arbitration International 21, no. 3 (2005): 325–326.
258 The Guidelines were cited in 106 of the 187 challenge decisions made by the

ICC between 2004 and 2009. See Jason Fry & Simon Greenberg, ‘The Arbitral
Tribunal: Applications of Articles 7–12 of the ICC Rules in Recent Cases' ICC
International Court of Arbitration Bulletin 20, no. 2 (2009): 17; Niklas Lindstrom,
‘Challenge to Arbitrators – Decisions by the SCC Board during 2008–2010,
<www.sccinstitute.com/library/articles.aspx3>.
259 Markham Ball, ‘Probity Deconstructed: How Helpful, Really, Are the New

International Bar Association Guidelines on Conflicts of Interest in International

Arbitration?’, Arbitration International 21, no. 3 (2005): 324–325.
260 See Matthias Scherer, ‘First Reference to the IBA Guidelines on Conflicts of

Interest in International Arbitration – Case Note on Swiss Supreme Court

Decisions’, ASA Bulletin 26, no. 3 (2008): 588–589.
261 See, e.g., Decision of 20 March 2008, Swiss Federal Supreme Court, 1st Civil

Chamber, Case 4A_506/2007 ASA Bulletin 26, no. 3 (2008).

262 Judgment of 10 December 2008, Svea Court of Appeal, Case T 10321-06, cited

in Niklas Lindstrom, ‘Challenge to Arbitrators – Decisions by the SCC Board

during 2008–2010, <www.sccinstitute.com/library/articles.aspx3> accessed 11
June 2011.
263 Judgment of 20 March 2008, No. 4A_506/207, Swiss Federal Supreme Court,

ASA Bulletin 28, no. 3 (2010): 565.

264 ASM Shipping Ltd of India v. TTMI Ltd of England [2005] EWHC 2238 (Comm).
265 Applied Industrial Materials Corp. v. Ovalar Makine Ve Ticaret Sanayi,
unreported, Case No. 05 CV 10540 (RPP), SDNY; Applied Industrial Materials
Corp. v. Ovalar Makine Ve Ticaret Sanayi, 492 F. 3d 132 CA 2 (NY).
266 Judgment of 18 October 2004, District Court of The Hague, civil law section,

provision measures judge, Challenge No. 13/2004, Petition No. HA/RK 2004.
267 Judgment of 29 October 2007, République de Pologne v. Eureko BV, Brussels

Court of Appeal.
268 Judgment of 4 October 2007, X v. Y, Oberlandesgericht Frankfurt am Main

(Higher Regional Court of Frankfurt).

269 Judgment of 24 July 2007, Case No. 16 Nc 2/07w, Vienna Commercial Court.
270 For further details on judgments mentioning the IBA Guidelines, see IBA

Conflicts of Interest Subcommittee, ‘The IBA Guidelines on Conflicts of Interest in

International Arbitration: The First Five Years 2004–2009’, Dispute Resolution
International 4, no. 1 (2010).
271 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 271.

272 Ibid.
273 Ibid.
274 See Swedish Arbitration Act 1999 s. 8; Chinese Arbitration Law Art. 34.
275 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 281.

276 Jan Paulsson, ‘Ethics, Elitism, Eligibility’, Journal of International Arbitration

14, no. 4 (1997): 18.

277 W. Laurence Craig, William W. Park & Jan Paulsson, International Chamber of

Commerce Arbitration, 3rd edn (Dobbs Ferry: Oceana, 2000), 225.

278 See, e.g., SCC Case 3 (U 045/2008), cited in Niklas Lindstrom, ‘Challenge to

Arbitrators – Decisions by the SCC Board during 2008–2010,

<www.sccinstitute.com/library/articles.aspx3> accessed 11 June 2011.
279 See IBA Guidelines on Conflicts of Interest, 21. These examples appear on the

‘non-waivable red list’.

280 Doak Bishop & Lucy Reed, ‘Practical Guidelines for Interviewing, Selecting
and Challenging Party-Appointed Arbitrators in International Commercial
Arbitration’, Arbitration International 14, no. 4 (1998): 410.
281 Decision of 26 October 1966, Centrozap v. Orbis, Swiss Federal Tribunal, ATF

92 I 271.
282 Niklas Lindstrom, ‘Challenge to Arbitrators – Decisions by the SCC Board

during 2008–2010, <www.sccinstitute.com/library/articles.aspx3> accessed 11

June 2011.
283 Markham Ball, ‘Probity Deconstructed: How Helpful, Really, Are the New

International Bar Association Guidelines on Conflicts of Interest in International

Arbitration?’, Arbitration International 21, no. 3 (2005): 332. Ball notes that
under many national laws, parties could clearly consent to such arbitrators.
284 IBA Guidelines on Conflicts of Interest, 20.
285 All of these examples appear on the ‘Green List’ of the IBA Guidelines on

Conflicts of Interest.
286 Doak Bishop & Lucy Reed, ‘Practical Guidelines for Interviewing, Selecting

and Challenging Party-Appointed Arbitrators in International Commercial

Arbitration’, Arbitration International 14, no. 4 (1998): 418.
287 Ibid., 418–419.
288 AT&T Corporation v. Saudi Cable Co. [2000] 2 Lloyd's Rep 127. This appears

on the ‘Orange List’ of the IBA Guidelines on Conflicts of Interest, 23.

289 IBA Guidelines on Conflicts of Interest, 24.
290 Suez, Sociedad General de Aguas de Barcelona S.A., & Vivendi Universal S.A v.

Argentine Republic, ICSID Case No. ARB/03/19, Decision on a Second Proposal

for the Disqualification of a Member of the Arbitral Tribunal (12 May 2008),
para. 33.
291 W. Laurence Craig, William W. Park & Jan Paulsson, International Chamber of

Commerce Arbitration, 3rd edn (Dobbs Ferry: Oceana, 2000), 211.

292 Suez, Sociedad General de Aguas de Barcelona S.A., & Vivendi Universal S.A. v.

Argentine Republic, ICSID Case No. ARB/03/19, Decision on a Second Proposal

for the Disqualification of a Member of the Arbitral Tribunal (12 May 2008),
para. 33.
293 W. Laurence Craig, William W. Park & Jan Paulsson, International Chamber of

Commerce Arbitration, 3rd edn (Dobbs Ferry: Oceana, 2000), 210.

294 Commonwealth Coatings Corp. v. Continental Casualty Co. 393 US 145(1968).
295 See IBA Rules on Ethics s. 3.4.
296 Doak Bishop & Lucy Reed, ‘Practical Guidelines for Interviewing, Selecting

and Challenging Party-Appointed Arbitrators in International Commercial

Arbitration’, Arbitration International 14, no. 4 (1998): 417.
297 SGS Société Générate de Surveillance S.A. v. Islamic Republic of Pakistan, ICSID

Case No ARB/01/13, Decision on Claimant's Proposal to Disqualify Arbitrator

(19 December 2002).
298 Ibid., 404.
299 See Jan Paulsson, ‘Ethics, Elitism, Eligibility’, Journal of International
Arbitration 14, no. 4 (1997): 17–18; Markham Ball, ‘Probity Deconstructed: How
Helpful, Really, are the New International Bar Association Guidelines on Conflicts
of Interest in International Arbitration?’, Arbitration International 21, no. 3
(2005): 339.
300 See IBA Guidelines on Conflict of Interest (2004), 14. See also Markham Ball,
‘Probity Deconstructed: How Helpful, Really, Are the New International Bar
Association Guidelines on Conflicts of Interest in International Arbitration?’,
Arbitration International 21, no. 3 (2005): 327.
301 SCC Arbitration Case V (068/2010), cited in Niklas Lindstrom, ‘Challenge to

Arbitrators – Decisions by the SCC Board during 2008–2010,

<www.sccinstitute.com/library/articles.aspx3>, 16–19.
302 Judgment of 10 March 2011, Tesco v. Neoelectra, Cour d'appel de Paris, Case

No. 09/28537.
303 Compañia de Aguas del Aconquija SA and Vivendi Universal v. Argentine

Republic.
304 Ste Icori, v. Kuwait Foreign Trading Contracting and Investment Co., Tribunal

de grande instance de Paris, 24 February 1992, 1994 Rev. Arb 557; Laker
Airways Incorporated v. FLS Aerospace Limited [1999] 2 Lloyd's Law Reports 45,
26 April 1999, QB.
305 Armen H. Merjian, ‘Caveat Arbiter: Laker Airways and the Appointment of

Barristers as Arbitrators in Cases Involving Barrister-Advocates from the Same

Chambers’, Journal of International Arbitration 17, no. 1 (2000).
306 Hrvatska Elektroprivreda dd v. Republic of Slovenia, ICSID Case No.
ARB/05/24. See Jeff Waincymer, ‘Reconciling Conflicting Rights in International
Arbitration: The Right to Choice of Counsel and the Right to an Independent and
Impartial Tribunal’, Arbitration International 26, no. 4 (2010).
307 See David W. Brown, ‘Arbitrators, Impartiality and English Law: Did Rix J

Really Get It Wrong in Laker Airways?, Journal of International Arbitration 18, no.
1 (2001)L 125.
308 See IBA Guidelines on Conflicts of Interest, sections 3.2.2 & 4.31.
309 Doak Bishop & Lucy Reed, ‘Practical Guidelines for Interviewing, Selecting

and Challenging Party-Appointed Arbitrators in International Commercial

Arbitration’, Arbitration International 14, no. 4 (1998): 421.
310 Jason Fry & Simon Greenberg, ‘The Arbitral Tribunal: Applications of Articles

7–12 of the ICC Rules in Recent Cases’, ICC International Court of Arbitration
Bulletin 20, no. 2 (2009): 20.
311 Philip Yang, ‘The Organisation of International Arbitration Proceedings’, in

The Asian Leading Arbitrators' Guide to International Arbitration, ed. Michael

Pryles & Michael Moser (Huntington: JurisNet LLC, 2007), 172.
312 Suex and others v. Republic of Argentina, Disqualification Decision, ICSID Case

No. ARB/03/17.
313 Dominique Hascher, ‘ICC Practice in Relation to the Appointment,
Confirmation, Challenge and Replacement of Arbitrators’, ICC International Court
of Arbitration Bulletin 6, no. 4 (1995): 10, n. 32.
314 See LCIA Challenge Decision 18 in Geoff Nicholas & Constantine Partasides,

‘LCIA Court Decisions on Challenges to Arbitrators: A Proposal to Publish’,

Arbitration International 23, no. 1 (2007): 37.
315 See Gabrielle Kauffmann-Kohler & Fan Kun, ‘Integrating Mediation into

Arbitration: Why It Works in China’, Journal of International Arbitration 25, no. 4

(2008).
316 Singapore International Arbitration Act s. 17.
317 Section 33 Hong Kong Arbitration Ordinance 2010.
318 See further section 8.11.7.
319 Gao Haiyan Xie Heping v. Keeneye Holdings Limited & New Purple Golden
Resources Development Limited [2011] HKEC 514.
320 See Judgment of 3 May 1967, Billerbeck Cie v. Berghau-Handel GmbH,
Supreme Court of the Federal Republic of Germany, Yearbook of Commercial
Arbitration 1 (1976): 200.
321 LCIA Challenge Decision 6 in Geoff Nicholas & Constantine Partasides, ‘LCIA

Court Decisions on Challenges to Arbitrators: A Proposal to Publish’, Arbitration

International 23, no. 1 (2007): 26.
322 LCIA Challenge Decision 20 in Geoff Nicholas & Constantine Partasides, ‘LCIA

Court Decisions on Challenges to Arbitrators: A Proposal to Publish’, Arbitration

International 23, no. 1 (2007): 36.
323 W. Laurence Craig, William W. Park & Jan Paulsson, International Chamber of

Commerce Arbitration, 3rd edn (Dobbs Ferry: Oceana, 2000), 211–212.

324 Jacques Werner, ‘The Independence of Arbitrators in Totalitarian States:

Tackling the Tough Issues’, Journal of International Arbitration 14, no. 1 (1997):
144.
325 Suez, Sociedad General de Aguas de Barcelona S.A., & Vivendi Universal S.A. v.

Argentine Republic, ICSID Case No. ARB/03/19, Decision on a Second Proposal

for the Disqualification of a Member of the Arbitral Tribunal (12 May 2008),
para. 35.
326 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 281–282.

327 Ibid., 283.
328 Republic of Ghana v Telekom Malaysia, District Court of The Hague, civil law

section, provision measures judge, Challenge No. 13/2004, Petition No. HA/RK
2004, 18 October 2004.
329 W. Laurence Craig, William W. Park & Jan Paulsson, International Chamber of

Commerce Arbitration, 3rd edn (Dobbs Ferry: Oceana, 2000), 233.

330 W. Laurence Craig, William W. Park & Jan Paulsson, International Chamber of

Commerce Arbitration, 3rd edn (Dobbs Ferry: Oceana, 2000), 233.

331 Mohammed Bedjaoui, ‘Challenge of Arbitrators’, in International Arbitration

in a Changing World, ICCA Congress Series No. 6, ed. Albert Jan van den Berg
(Deventer: Kluwer Law and Taxation Publishers, 1994), 94.
332 Canfor Corporation v. United States; Tembec et al. v. United States and
Terminal Forest Products Ltd v. United States, Consolidated NAFTA Arbitration,
Order of 7 September 2005.
333 William W. Park, ‘Arbitrator Integrity: The Transient and the Permanent’, San

Diego Law Review 46, no. 3 (2009): 683–684.

334 National Grid PLC v. The Republic of Argentina, LCIA Case No. UN7949,

Decision on the Challenge to Mr Judd L. Kessler, 3 December 2007, para. 38.

335 Fairchild & Co, Inc. v. Richmond F & PR Co., 516 F. Supp. 1305.
336 W. Laurence Craig, William W. Park & Jan Paulsson, International Chamber of

Commerce Arbitration, 3rd edn (Dobbs Ferry: Oceana, 2000), 231.

337 Perenco Ecuador Limited v. Republic of Ecuador and Empresa Estatal Petróles

del Ecuador, ICSID Case No. ARB/08/16, PCA Case No IR-2009/1, Decision of 8
December 2009.
338 In Tanzania Electric Supply Co. Ltd v. Independent Power Tanzania Ltd, ICSID

Case No. ARB/98/8 (12 July 2001), an arbitrator resigned after his law clerk
posted views on legal issues relevant to the case on a blog. See Lars Markert,
‘Challenging Arbitrators in Investment Arbitration: The Challenging Search for
Relevant Standards and Ethical Guidelines’, Contemporary Asia Arbitration
Journal 3, no. 2 (2010): 264–265.
339 See, e.g., Judgment of 6 January 2010, X & Y v. V & W, Swiss Federal Supreme

Court, ASA Bulletin 28, no. 4: 776.

340 Jason Fry & Simon Greenberg, ‘The Arbitral Tribunal: Applications of Articles

7–12 of the ICC Rules in Recent Cases’, ICC International Court of Arbitration
Bulletin 20, no. 2 (2009): 25.
341 See, e.g., LCIA Challenge Decision 13 and 14, Geoff Nicholas & Constantine

Partasides, ‘LCIA Court Decisions on Challenges to Arbitrators: A Proposal to

Publish’, Arbitration International 23, no. 1 (2007): 26 and 34.
342 Judgment of 19 February 2009, Swiss Federal Supreme Court, ASA Bulletin

27, no. 4 (2009).

343 Ibid., 810–811.
344 Turner (East Asia) Pte Ltd v. Builders Federal (Hong Kong) Ltd &Anor (No 2)

[1988] SGHC 47.

345 ICC Rules 2012 Art. 14(1).
346 UNCITRAL Model Law Art. 12(2); English Arbitration Act 1996 s. 24(1)(b).

See also Gary B. Born, International Commercial Arbitration (The Hague: Kluwer
Law International, 2009), 1556.
347 See, e.g., LCIA Rules Art. 10; English Arbitration Act 1996 ss 24(1)(c) and

73(1).
348 Jason Fry & Simon Greenberg, ‘The Arbitral Tribunal: Applications of Articles

7–12 of the ICC Rules in Recent Cases’, ICC International Court of Arbitration
Bulletin 20, no. 2 (2009): 19.
349 Swedish Arbitration Act s. 17; English Arbitration Act 1996 s. 24(1)(d)(ii).
350 Jason Fry & Simon Greenberg, ‘The Arbitral Tribunal: Applications of Articles

7–12 of the ICC Rules in Recent Cases’, ICC International Court of Arbitration
Bulletin 20, no. 2 (2009): 22.
351 See, e.g., UNCITRAL Model Law Art. 12(1).
352 ICC Rules 2012 Art. 11(2); LCIA Rules Art. 5.3; Swiss Rules 2012 Art. 9.2;

ACICA Rules Art. 13.1; HKIAC Rules Art. 11.3; SIAC Rules Art. 10.4.
353 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 269.

354 See, e.g., ICC Rules 2012 Art. 11(2).
355 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 268.

356 The modern approach to broad disclosure for the benefit of the parties is to

be contrasted with the earlier formulation in some rules that left it to the opinion
of the arbitrators. See, e.g., 1975 ICC Arbitration Rules.
357 W. Laurence Craig, William W. Park & Jan Paulsson, International Chamber of

Commerce Arbitration, 3rd edn (Dobbs Ferry: Oceana 2000), 195.

358 IBA Guidelines on Conflicts of Interest, 9.
359 See Markham Ball, ‘Probity Deconstructed: How Helpful, Really, Are the New

International Bar Association Guidelines on Conflicts of Interest in International

Arbitration?’, Arbitration International 21, no. 3 (2005): 337.
360 Commonwealth Coatings Corp. v. Continental Casualty 393 US 145, 151–152

(1968).
361 Judgment of 25 March 1999, Cour de cassation, Société SIAB v. Société

Valmont, reported in Revue de l'Arbitrage (1999), no. 2: 319.

362 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn
(Oxford: Oxford University Press, 2009), 270.
363 Emilio Cárdenas & David W. Rivkin, ‘A Growing Challenge for Ethics in

International Arbitration’, in Global Reflections on International Law, Commerce

and Dispute Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald
Aksen, Karl-Heinz Böckstiegel, Michael J. Mustill, Paolo Michele Patocchi & Anne
Marie Whitesell (Paris: ICC Publishing, 2005), 199.
364 Ibid.
365 Eric A. Schwartz, ‘The Rights and Duties of ICC Arbitrators’, in The Status of

the Arbitrator, ICC International Court of Arbitration Bulletin: 1995 Special

Supplement, ed. Jean-François Bourque (Paris: ICC Publishing, 1995), 93.
366 See W. Laurence Craig, William W. Park & Jan Paulsson, International
Chamber of Commerce Arbitration, 3rd edn (Dobbs Ferry: Oceana, 2000), 215.
Suez, Sociedad General de Aguas de Barcelona S.A., & Vivendi Universal S.A. v.
Argentine Republic, ICSID Case No. ARB/03/19, Decision on a Second Proposal
for the Disqualification of a Member of the Arbitral Tribunal (12 May 2008),
para. 44.
367 IBA Guidelines on Conflict of Interest, 18.
368 Ibid.
369 393 US 145 (1968).
370 Markham Ball, ‘Probity Deconstructed: How Helpful, Really, Are the New

International Bar Association Guidelines on Conflicts of Interest in International

Arbitration?’, Arbitration International 21, no. 3 (2005): 335.
371 [2000] EWCA Civ 154.
372 Marie Ohrstrom, ‘Decisions of the SCC Institute Regarding Challenge of

Arbitrators’, <www.sccinstitute.se/filearchive/2/21256/Decisions_...>, 48.

373 Jilkén v. Ericsson, Judgment of the Supreme Court of Sweden, [2007] 3 SIAR

167.
374 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 465–66.
375 [2000] EWCA Civ 154.
376 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 362.

377 Judgment of 7 December 2006, The Rappler Invest v. TNK Trade Case, Svea

Court of Appeal, Sweden, Case No. T5004-04.

378 Less clear would be the method of determining whether damages are
warranted depending on the ultimate outcome. Damages would be easy to
calculate where there were direct wasted fees in a successful challenge
application. If delay could be identified, the successful party might be able to
claim damages equal to time value of money, although that might already be built
into an interest award against the losing party. Where circumstances only come
to light after the award has been rendered the situation is far more difficult. If
enforcement of the award cannot be successfully challenged on this basis, it
would be difficult to prove causation of damage through lack of disclosure. If
enforcement was blocked on this basis, the wasted costs would more naturally
be seen to flow from the breach of the duty.
379 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 269.

380 Ibid.
381 See Ibid.
382 IBA Guidelines on Conflicts of Interest, 7.
383 James H. Carter, ‘The Rights and Duties of the Arbitrator: Six Aspects of the

Rule of Reasonableness’, in The Status of the Arbitrator, ICC International Court of

Arbitration Bulletin: 1995 Special Supplement, ed. Jean-François Bourque (Paris:
ICC Publishing, 1995), 27.
384 Case law on the duty to inquire is limited. In one US case, the court did not go

so far as to say that a general duty to inquire into conflicts exists. It did, however,
hold that arbitrators have a duty to inquire in situations where they have reason
to believe that some conflict might exist. For conflicting US decisions as to the
duty to investigate see Catherine Rogers, ‘The Ethics of International
Arbitrators’, in The Leading Arbitrators' Guide to International Arbitration, ed.
Lawrence W. Newman & Richard D. Hill (Huntington: Juris Pub, 2008), 641.
385 Applied Indus Materials Corp. v. Ovalar Makine Ticaret Ve Sanayi AS 492 F. 3d.

132 (2d Cir, 2007).

386 Michael F. Hoellering, ‘The Role of the International Arbitrator’, Dispute

Resolution Journal 51, no. 2 (1996): 103. Hoellering notes in such circumstances
the stricter standards are applied under challenges in the AAA-ICDR system.
387 W. Laurence Craig, William W. Park & Jan Paulsson, International Chamber of

Commerce Arbitration, 3rd edn (Dobbs Ferry: Oceana, 2000), 204.

388 Craig, Park and Paulsson note that, where the parties agree, it may even be

possible for one party to nominate its own lawyer or employee, even though this
essentially makes that arbitrator an advocate for one of the parties. W. Laurence
Craig, William W. Park & Jan Paulsson, International Chamber of Commerce
Arbitration, 3rd edn (Dobbs Ferry: Oceana, 2000), 195– 196. This goes against
the position under the IBA Guidelines. IBA Guidelines on Conflicts of Interest, 9.
389 See also Swedish Arbitration Act s. 10; Belgian Judicial Code Art. 1690(2);

Swiss Private International Law Art. 180(2); Judgment of 9 February 1998, Swiss
Federal Supreme Court, ASA Bulletin 16, no. 3 (1998): 646; Ghiradosi v. Minister
of Highways (1996) 56 DLR (2d) 469; Fidelity Fed Bank FSB v. Durga Ma Corp.
386 F. 3d 1306 (9th Cir. 2006). Most procedural rules impose a similar
requirement of promptness.
390 ICC Rules 2012 Art. 14(2); LCIA Rules Art. 10.4; UNCITRAL Rules 2010 Art.

13; HKIAC Rules Art. 11.5; SIAC Rules Art. 12.1; SCC Rules Art. 15(2); ACICA
Rules Art. 14.1; ICSID Rules Art. 27.
391 Catherine Rogers, ‘The Ethics of International Arbitrators’, in The Leading

Arbitrators' Guide to International Arbitration, ed. Lawrence W. Newman &

Richard D. Hill (Huntington: Juris Pub, 2008), 644.
392 AAOT Foreign Econ Ass'n (VO) Technostroyexport v. International Dev and

Trade Serv, Inc. 139 F. 3d 980 (2d Cir 1998). A contrary view would be that
honesty of arbitrators is a mandatory norm of procedure which cannot be
waived.
393 See Markham Ball, ‘Probity Deconstructed: How Helpful, Really, Are the New

International Bar Association Guidelines on Conflicts of Interest in International

Arbitration?’, Arbitration International 21, no. 3 (2005): 332–333.
394 Ibid., 332. Ball questions the IBA Working Group's assertion that ‘most

common law countries’ agree that there can be no waiver of ‘real or possible
bias’.
395 See, e.g., Gary B. Born, International Commercial Arbitration (The Hague:
Kluwer Law International, 2009), 1553.
396 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 277.

397 Emilio Cárdenas & David W. Rivkin, ‘A Growing Challenge for Ethics in

International Arbitration’, in Global Reflections on International Law, Commerce

and Dispute Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald
Aksen et al. (Paris: ICC Publishing, 2005), 200.
398 ICC Rules 2012 Art. 14(2).
399 Jason Fry & Simon Greenberg, ‘The Arbitral Tribunal: Applications of Articles

7–12 of the ICC Rules in Recent Cases’, ICC International Court of Arbitration
Bulletin 20, no. 2 (2009): 23.
400 ICC Rules 2012 Art. 14(3).
401 ICC Rules 2012 Art. 11(4).
402 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 278.

403 See, e.g., LCIA Rules Art. 29.2.
404 LCIA Rules Art. 10.4.
405 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 279.

406 This followed the suggestion of Geoff Nicholas & Constantine Partasides,

‘LCIA Court Decisions on Challenges to Arbitrators: A Proposal to Publish’

Arbitration International 23, no. 1 (2007). For an example of a decision, see
National Grid PLC v. The Republic of Argentina, LCIA Case No. UN7949, Decision
on the Challenge to Mr Judd L. Kessler, 3 December 2007.
407 Niklas Lindstrom, ‘Challenge to Arbitrators – Decisions by the SCC Board

during 2008–2010, <www.sccinstitute.com/library/articles.aspx3>, 4–5.

408 See SIAC Rules Arts 10–14.
409 UNCITRAL Rules 2010 Art. 13.
410 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 279.

411 ICSID Rules Art. 9(1).
412 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 280, n. 113.

413 Suez and Others v. Republic of Argentina, Disqualification Decision, ICSID Case

No. ARB/ 03/17 at paras 22–26.

414 ICSID Arbitration Rules Art. 9(6). See also Mohammed Bedjaoui, ‘Challenge of

Arbitrators’, in International Arbitration in a Changing World, ICCA Congress

Series No. 6, ed. Albert Jan van den Berg (Deventer: Kluwer Law and Taxation
Publishers, 1994), 100–101.
415 See ICSID Arbitration Rules Art. 9(4).
416 UNCITRAL Model Law Art. 13; English Arbitration Act 1996 s. 24; Swiss

Private International Law Act Art. 180; Netherlands Arbitration Act 1986 Arts
1034 and 1035.
417 See, e.g., UNCITRAL Model Law Art. 13(3).
418 English Arbitration Act 1996 s. 24(2) and Schedule 1. Germany also regards

such review as a mandatory rule. See Decision of 24 February 1999, Bayerisches

Oberstes Landesgericht [2000] NJW-RR 360.
419 France, Switzerland and Singapore are notable examples. See Raffineries de
pétrole d'Homs v. CCI, Decision of 15 May 1985, Revue de l'Arbitrage 147; Opinter
France v. S.A.R.L. Dacomex Cass 2e Civ, 7 October 1987, Revue de l'Arbitrage 479.
See also Swiss Private International Law Art. 197 and the decision of the Swiss
Federal Tribunal 120 II 155.
420 See, e.g., Judgment of 29 October 2010, Valverde v. CONI et al., Swiss Federal

Supreme Court, ASA Bulletin 29, no. 1 (2011): 88.

421 UNCITRAL Model Law Art. 13; English Arbitration Act 1996 s. 24(3).
422 Morelite Construction Corp. v. New York City District Carpenters Benefit Funds

748 F. 2d 79 (1984); Michaels v. Mariforum Shipping, S.A., 624 F. 2d 411 (2d Cir,
1980). See US Federal Arbitration Act s. 10(a)(2). See also Yulia Andreeva, ‘How
Challenging Is the Challenge, or Can U.S. Courts Remove Arbitrators before an
Arbitration Has Come to an End?’, American Review of International Arbitration
19 (2008): 132.
423 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 278.

424 See Jefferson-Pilot Life Insurance Co v. LeafRe Reinsurance Co. 2000 WL

1724661 (N.D. Ill. 2000). In that case, the failure to appoint an arbitrator with the
qualifications required in the arbitration clause justified removal prior to an
award. See also Yulia Andreeva, ‘How Challenging Is the Challenge, or Can U.S.
Courts Remove Arbitrators before an Arbitration Has Come to an End?’,
American Review of International Arbitration 19 (2008): 133.
425 See York Hannover Holding A. G. v. AAA, 1993 US Dist Crt (SDNY, 7 May 1993).

In that case, the Court noted the possibility of using its general equitable powers
to intervene in an arbitration, but ultimately found no factual basis for doing so.
426 See Yulia Andreeva, ‘How Challenging Is the Challenge, or Can U.S. Courts

Remove Arbitrators Before an Arbitration Has Come to an End?’, American

Review of International Arbitration 19 (2008): 128. One case even allowed a
challenge on general arbitration policy grounds. See Metropolitan Property and
Cas. Ins. Co. v. J.C. Penney Cos. Ins. Co., 780 F. Supp. 885 (D. Conn. 1991). But this
has been criticised: In re Universal Serv. Fund Tel. Billing Practices Litig. 370 F.
Supp. 2d 1135, 1140 (D. Kan. 2005).
427 Applied Indus Materials Corp. v. Ovalar Makine Ticaret Ve Sanayi A.S., 492 F.

3d 132 (2d Cir. 2007). See also, Yulia Andreeva, ‘How Challenging Is the
Challenge, or Can U.S. Courts Remove Arbitrators Before an Arbitration Has
Come to an End?’, American Review of International Arbitration 19 (2008): 135–
136.
428 Bola Ajibola, ‘Recourse against the Arbitrator: During the Arbitral
Proceedings’, in The Status of the Arbitrator, ICC International Court of Arbitration
Bulletin: 1995 Special Supplement, ed. Jean-François Bourque (Paris: ICC
Publishing, 1995), 112.
429 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 279.

430 Ibid.
431 See, e.g., ICDR Rules Art. 9.
432 See Gary B. Born, International Commercial Arbitration (The Hague: Kluwer

Law International, 2009), 1563–1566 and 1611. For examples of national laws,
see UNCITRAL Model Law Art. 14(1); English Arbitration Act s. 23(3) and (4);
Swiss Private International Law Art. 179(1); French New Code of Civil Procedure
Art. 1458; Austrian Code of Civil Procedure Art. 590; Finnish Arbitration Act s.
19(1); Netherlands Code of Civil Procedure Art. 1031(1).
433 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1612.

434 Ibid.
435 LCIA Rules Art. 10.3; ICDR Rules Art. 8; HKIAC Rules Art. 11.4; SIAC Rules Art.

11.4;SCC Rules Art. 15(1); ACICA Rules Art. 13.3.

436 UNCITRAL Model Law Art. 13(2); German Code of Civil Procedure Art.

1037(2); Japanese Arbitration Law Art. 21(2); UNCITRAL Rules 2010 Art. 11(3);
ICDR Rules Art. 8(3); LCIA Rules Art. 10(4); Swiss Rules 2012 Art. 11.2.
437 Eric A. Schwartz, ‘The Rights and Duties of ICC Arbitrators’, in The Status of

the Arbitrator, ICC International Court of Arbitration Bulletin: 1995 Special

Supplement, ed. Jean-François Bourque (Paris: ICC Publishing, 1995), 93; Gary B.
Born, International Commercial Arbitration (The Hague: Kluwer Law
International, 2009), 1637.
438 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1637.

439 Ibid.
440 Codes of ethics may also express a requirement not to resign without good

cause. See, e.g., AAA/ABA Code of Ethics Canon I(H).

441 Yves Derains & Eric A. Schwartz, A Guide to the ICC Rules of Arbitration, 2nd

edn (The Hague: Kluwer Law International, 2005), 193.

442 Ibid.
443 See, e.g., Dutch Code of Civil Procedure Art. 1029(2). The Austrian Code of

Civil Procedure Art. 579 refers to ‘reasonable cause’.

444 Ivan Milutinovic PIM v. Deutsche Babcock AG – Swiss Fed. Trib.
445 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1634.

446 See ICSID Rules Art. 8(2); ICDR Rules Art. 10; ACICA Rules Art. 15.
447 Belgian Judicial Code Art. 1689 which states that arbitrators cannot resign

without court approval; Netherlands Code of Civil Procedure Art. 1029(3) and
(4) which requires approval by the parties, a designated appointed authority or a
court. See also ICC Rules 2012 Art. 15(1) which requires resignation to be
accepted by the ICC Court of Arbitration.
448 UNCITRAL Model Law Art. 14(1).
449 Howard M. Holtzmann & Joseph E. Neuhaus, A Guide to the UNCITRAL Model

Law on International Commercial Arbitration: Legislative History and

Commentary (Deventer: Kluwer Law and Taxation Publishers, 1989), 464.
450 Summary Record of the Fifth Meeting of the Committee of the Whole (II),

UNCITRAL, 9th Session, UN Doc. A/CN.9/9/C.2/SR.5, 5 paras 31–32 (1976)

available at <www.uncitral.org>.
451 Indeed Born notes that most conscientious practitioners would provide

reasons in any event. See Gary B. Born, International Commercial Arbitration

(The Hague: Kluwer Law International, 2009), 1638.
452 Illness is said to be the most common reason for resignation in ICC
arbitrations. See Yves Derains & Eric A. Schwartz, A Guide to the ICC Rules of
Arbitration, 2nd edn (The Hague: Kluwer Law International, 2005), 194.
453 See Yves Derains & Eric A. Schwartz, A Guide to the ICC Rules of Arbitration,

2nd edn (The Hague: Kluwer Law International, 2005), 193.

454 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law
International, 2009), 1636–1637.
455 Yves Derains & Eric A. Schwartz, A Guide to the ICC Rules of Arbitration, 2nd

edn (The Hague: Kluwer Law International, 2005), 194.

456 A related question albeit with differing policy issues, is whether a busy

arbitrator can reject extension applications on the basis that the arbitrators
schedule would then make the arbitration unduly delayed.
457 Yves Derains & Eric A. Schwartz, A Guide to the ICC Rules of Arbitration, 2nd

edn (The Hague: Kluwer Law International, 2005), 194–195.

458 See Decision of 6 May 2004 and 21 April 2005, Paris Cour d'Appel (Paris

Court of Appeal) Louis, Philippe & Rachel Malecki v. Adena Inc and David, Donna
and Carolyn Long. This case primarily concerned truncated tribunals, but the
circumstances in which the arbitrator was permitted to resign are informative.
459 See Yves Derains & Eric A. Schwartz, A Guide to the ICC Rules of Arbitration,

2nd edn (The Hague: Kluwer Law International, 2005), 193.

460 Born notes that different considerations would apply in such circumstances

but does not proffer a view as to an ethical conclusion. Gary B. Born,

International Commercial Arbitration (The Hague: Kluwer Law International,
2009), 1637, n. 241.
461 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1635.

462 Ibid., 1613.
463 Ibid.
464 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 287.

465 Ibid.; Mohammed Bedjaoui, ‘Challenge of Arbitrators’, in International
Arbitration in a Changing World, ICCA Congress Series No. 6, ed. Albert Jan van
den Berg (Deventer: Kluwer Law and Taxation Publishers, 1994), 101.
466 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 287.

467 UNCITRAL Rules 2010 Art. 14.
468 Northwestern National Insurance Co. v. Insco Ltd, 2011 WL 1833303 (SDNY

12 May, 2011).
469 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 292.

470 Ibid., 290.
471 Ibid., 290–291.
472 Similarly, WIPO Art. 34, ICDR Art. 11.2 and ICC Rules 2012 Art. 12(4) all

provide for a discretion as to whether to repeat oral hearings.

473 Mohammed Bedjaoui, ‘Challenge of Arbitrators’, in International Arbitration

in a Changing World, ICCA Congress Series No. 6, ed. Albert Jan van den Berg
(Deventer: Kluwer Law and Taxation Publishers, 1994), 101.
474 Pierre Lalive, Jean-Francois Poudret & Claude Reymond, Le droit de
l'arbitrage interne et international en Suisse (Lausanne: Payot, 1989), 334.
475 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th ed.

(Oxford: Oxford University Press, 2009), 290.

476 See Howard M. Holtzmann et al., ‘Working Group I: Preventing Delay and

Disruption of Arbitration: III Conduct by a Party-Appointed Arbitrator During the

Arbitral Proceedings: Topic 8: Resignation of a Party-Appointed Arbitrator’, in
Preventing Delay and Disruption of Arbitration/Effective Proceedings in
Construction Cases, ICCA Congress Series No. 5, ed. Albert Jan van den Berg
(Deventer: Kluwer Law and Taxation Publishers, 1991), 241–247 & 270–274.
477 See Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative

International Commercial Arbitration (The Hague: Kluwer Law International,

2003), 323.
478 Ibid.; Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 746. See also Belgian Judicial Code Art. 1701(1) which expressly states
that all the arbitrators must take part in deliberation.
479 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 323; Stephen M. Schwebel, International Arbitration: Three Salient
Problems (Cambridge: Grotius Publications Limited, 1987), 251. Schwebel
discusses truncated tribunals in the context of the Iran– US Claims Tribunal.
480 Tadeusz Szurski, ‘The Constitution of the Arbitral Tribunal’, ICCA Congress

Series No. 9, 332.

481 See, e.g., Pieter Sanders, ‘Has the Moment Come to Revise the Arbitration

Rules of UNCITRAL?’, Arbitration International 20, no. 3 (2004): 262–263; Serge

Lazareff, ‘The Constitution of the Arbitral Tribunal’, in Improving the Efficiency of
Arbitration Agreements and Awards: 40 Years of Application of the New York
Convention, ICCA Congress Series No. 9, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 1999): 326– 327.
482 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 289.

483 But note that the Rules only allow for a truncated tribunal where the

arbitrator resigns or is removed after the close of proceedings. Derains &

Schwartz suggest that this rule does not cover situations where the arbitrator
simply refuses to participate from an earlier stage. See Yves Derains & Eric A.
Schwartz, A Guide to the ICC Rules of Arbitration, 2nd edn (The Hague: Kluwer
Law International, 2005), 195.
484 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 290.

485 UNCITRAL Rules 2010 Art. 14.2. See also Christopher Kee, ‘The Evolving Role

of an Appointing Authority’, in Liber Amicorum Eric Bergsten: International

Arbitration and International Commercial Law: Synergy, Convergence and
Evolution, ed. Kröll et al. (Alphen aan den Rijn: Kluwer Law International, 2011),
311–312. Kee observes that the appointing authority would be very unlikely to
authorise a truncated tribunal if the remaining arbitrators were not in favour of
it.
486 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1589.

487 Stephen M. Schwebel, International Arbitration: Three Salient Problems
(Cambridge: Grotius, 1987), 146.
488 See Decision of 1 July 1997, Paris Cour d'appel, Agence Transcongolaise des

Communications – Chemin de fer Congo Océan v Compagnie Miniére de l'Ogooue –

Comilog SA (Gabon), XXIVa Yearbook of Commercial Arbitration 281 (1999). See
also the decision of the Swiss Federal Tribunal, Ivan Milutinovic PIM v. Deutsche
Babcock AG. Note that that decision was under the old Swiss law. Schwebel has
questioned whether the same result would arise under the new law. See
Schwebel, ‘The Validity of an Arbitral Award rendered by a Truncated Tribunal’,
ICC Bulletin 6, no. 2 (1995): 19, 22.
489 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 324.
490 Swedish Arbitration Act s. 30; German Code of Civil Procedure Art. 1052(2).
491 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 323.
492 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 325.
493 Ibid., 323.
494 Howard M. Holtzmann, ‘How to Prevent Delay and Disruption of Arbitration:

Lessons of the 1990 ICCA Stockholm Congress’, in Preventing Delay and

Disruption of Arbitration/ Effective Proceedings in Construction Cases, ICCA
Congress Series No. 5, ed. Albert Jan van den Berg (Deventer: Kluwer Law and
Taxation Publishers, 1991), 28. See also Emmanuel Gaillard & John Savage (eds),
Fouchard, Gaillard, Goldman on International Commercial Arbitration (The Hague:
Kluwer Law International, 1999), 749.
495 Himpurna California Energy Ltd v. Republic of Indonesia (2000) XXV Yearbook

of Commercial Arbitration 186.

496 Ibid., at 194.
497 See Republic of Columbia v. Cauca Company 23 S. Ct. 704, 190 US 524;

Uiterwyk Corporation (et al.) Claimants v The Government of the Islamic Republic
of Iran (et al.) 19 Iran–US Claims Tribunal Reports 107.
498 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 290.

499 See, e.g., English Arbitration Act 1996 s. 93(4).
500 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 372. Poudret and Besson
refer to the German situation which considers such a determination contrary to
public policy, while Italian law indicates that an arbitrator's determination does
not bind the parties if they do not accept it.
501 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1648.

502 Ibid., 1650.
503 In some cases a national arbitration statute will expressly give a court

supervisory powers. See, e.g., Swedish Arbitration Act ss 37 and 41.

504 David D. Caron, Matti Pellonpää & Lee M. Caplan, The UNCITRAL Arbitration

Rules: A Commentary, 2nd edn (Oxford: Oxford University Press, 2006), 945; Gary
B. Born, International Commercial Arbitration (The Hague: Kluwer Law
International, 2009), 1647.
505 See, e.g., ICC Rules 2012 Art. 37(4).
506 LCIA Rules, Schedule of Arbitration Fees and Costs.
507 See, e.g., English Arbitration Act s. 28(2).
508 Philippe Fouchard, ‘Relationships between the Arbitrator and the Parties and

the Arbitral Institution’, in The Status of the Arbitrator, ICC International Court of
Arbitration Bulletin: 1995 Special Supplement, ed. Jean-François Bourque (Paris:
ICC Publishing, 1995), 20.
509 Ibid.
510 Paris Court of Appeal, 17 March 2005, SNCI v Interim Nation, Revue de

l'arbitrage (2005): 790; Swiss Supreme Court 4A-391/2010 and 4A-399/2010.

The latter case involved a tribunal staying proceedings when fees were not paid.
511 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 307.

512 John Yukio Gotanda, ‘Setting Arbitrators' Fees: An International Survey’,

Vanderbilt Journal of Transnational Law 33, no. 4 (2000): 782.

513 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 307.

514 See, e.g., ICC Rules Appendix II, Arts 1(10) and 2(2); HKIAC Rules, Schedule of

Fees; JCAA Rules Art. 70; DIS Rules s. 40(2); SCC Rules, Appendix III Schedule of
Costs; Swiss International Arbitration Rules 2012 Art. 39.
515 ICSID Rules effective 1 January 2008 provide for a fixed daily fee of USD

3,000 calculated on the basis of an eight hour day. ICSID Administrative and
Financial Regulations, Reg. 14.
516 For the latter see English Arbitration Act 1996 s. 28; Hong Kong Arbitration

Ordinance 2010 s. 78; Swedish Arbitration Act s. 37.

517 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 284.
518 See, e.g., IBA Ethics Rules Art. 6 which provides ‘unless the parties agree

otherwise or a party defaults, an arbitrator shall make no unilateral

arrangements for fees or expenses’.
519 Yves Derains & Eric A. Schwartz, A Guide to the ICC Rules of Arbitration, 2nd

edn (The Hague: Kluwer Law International, 2005), 358.

520 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 205.
521 Gerald Aksen, ‘The Tribunal's Appointment’, in The Leading Arbitrators' Guide

to International Arbitration, ed. Lawrence W. Newman & Richard D. Hill

(Huntington: Juris Pub, 2008), 40.
522 Ibid.
523 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 308.

524 Double-M Construction Corp. v. Central School District No 1 Town of Highlands

Orange County (1978) 402 NYS 2d 442.

525 See, e.g., ICC Rules 2012 Art. 4(4); ICDR Rules Art. 33.1; LCIA Rules Art. 1.1(f).
526 This is supported under UNCITRAL Rules Art. 41.1.
527 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 311.

528 Ibid., 308–309.
529 Ibid., 309.
530 Ibid.
531 Ibid., 310.
532 See, e.g., Arbitration Act 1996 (UK) s. 56. A contrary position is taken in the

Swedish Law on Arbitration, 4 March 1999 Art. 40.

533 Murray L. Smith, ‘Contractual Obligations Owed by and to Arbitrators: Model
Terms of Appointment’, Arbitration International 8, no. 1 (1992): 30.
534 The situation might be different in the rare case where parties expressly

accorded such a power to a tribunal.

535 John Yukio Gotanda, ‘Setting Arbitrators' Fees: An International Survey’,

Vanderbilt Journal of Transnational Law 33, no. 4 (2000): 790.

536 Chartered Institute of Arbitrators, Guidelines for Arbitrators as to How to

Formulate their Terms of Remuneration: Matters to be Considered, s. 2 (2000) at

<www.arbitrators.org/Services/services.htm>.
537 John Yukio Gotanda, ‘Setting Arbitrators' Fees: An International Survey’,

Vanderbilt Journal of Transnational Law 33, no. 4 (2000): 823.

538 See, e.g., ICT Pty Ltd v. Sea Containers Ltd, reported in Mealey's International

Arbitration Report 17, no. 3 (2002) B 1; K/S Norjarl A/S v. Hyundai Heavy
Industries Co. Ltd [1991] 1 Lloyd's Rep 524; and section 2.3 above which noted
that the contract with the arbitrator will have its own applicable law that might
provide some protective mechanisms for the disputing parties.
539 John Yukio Gotanda, ‘Setting Arbitrators' Fees: An International Survey’,

Vanderbilt Journal of Transnational Law 33, no. 4 (2000): 816.

540 ICT Pty Ltd v. Sea Containers Ltd, reported in Mealey's International
Arbitration Report 17, no. 3 (2002) B 1.
541 Norjarl K/S A/S v. Hyundai Heavy Industries Co. Ltd [1992] 1 QB 863.
542 Such an argument was rejected in Case/S Norjarl A/S v. Hyundai Heavy Indus

Co Ltd [1999] 1 Lloyd's Re (QB) affirmed on appeal Case/S Norjarl A/S v. Hyundai
Heavy Indus Co Ltd [1992] 1 QB 862 (Court of Appeal).
543 Ibid.
544 See, e.g., terms of the London Maritime Arbitrators Association, LMAA Terms:

First Schedule s. B(1)(a) to (e) 1977, <www/lmas.org.uk/default.html>.

545 English Arbitration Act 1996 s. 25(3)(b).
546 For example see UNCITRAL Rules 2010 Art. 41.
547 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 371.

548 Lawrence Boo, ‘Appointment of the Arbitral Tribunal’, in The Asian Leading

Arbitrators' Guide to International Arbitration, ed. Michael Pryles & Michael

Moser (Huntington: JurisNet LLC, 2007), 104.
549 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 373.

550 See generally Hausmaninger, ‘Civil Liability of Arbitrators – Comparative

Analysis and Proposal for Reform’, Journal of International Arbitration 7, no. 4

(1990).
551 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 288.
552 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1662.

553 Ibid.
554 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 375–376.

555 Arenson v. Casson Beckman Rutley & Co. [1975] 3 All ER 901, 922 (Lord

Salmon).
556 RH Coase, ‘The Problem of Social Cost’, Journal of Law and Economics 3
(1960).
557 See generally Julian D.M. Lew (ed.), The Immunity of Arbitrators, 1990; Julian

D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative International

Commercial Arbitration (The Hague: Kluwer Law International, 2003), 290.
558 P. Lalive, J.-F. Poudret & C. Reymond, Le Droit de L'Arbitrage Interne et

International en Suisse (Lausanne: Payot, 1989).

559 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 373.

560 An example of an exception is Austrian Code of Civil Procedure Art. 594(4)

which provides for liability for damages caused by culpable refusal or delay.
561 General immunity from liability has been included in the arbitration laws of

Australia, Hong Kong, Ireland, Malaysia and Singapore.

562 English Arbitration Act 1996 s. 29(1). This enshrines the position adopted by

the House of Lords in Sutcliffe v. Thackrah [1974] AC 725 and Arenson v. Arenson
[1976] 1 Lloyd's Rep 179. Australian International Arbitration Act s. 28; Swiss
Rules 2012 Art. 45.1.
563 Platte in S. Rigler et al. (eds), Arbitration Law of Austria: Practice and

Procedure Arbitrators' Liability 686 (2007), cited in Gary B. Born, International

Commercial Arbitration (The Hague: Kluwer Law International, 2009), 1658.
564 Judgment of 8 February 1907, RGZ 65, 175–176 (Reichsgericht).
565 UNCITRAL Arbitration Rules 2010 Art. 16.
566 Hausmaninger, ‘Civil Liability of Arbitrators – Comparative Analysis and

Proposals for Reform’, Journal of International Arbitration 7, no. 4 (1990). See

generally Julian Lew, The Immunity of Arbitrators (London: Lloyd's of London
Press, 1990). See also Eric Robine, ‘The Liability of Arbitrators and Arbitral
Institutions in International Arbitration under French Law’, Arbitration
International 5, no. 4 (1989); David W. Shenton, ‘The Liability of Arbitrators,
England’ International Business Lawyer 8, no. 12 (1980).
567 Bernard Hanotiau. ‘Misdeeds, Wrongful Conduct and Illegality in Arbitral

Proceedings’, in International Commercial Arbitration: Important Contemporary

Questions, ICCA Congress Series No. 11, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 2003), 283.
568 Countries that have adopted this approach include Canada, France and

Germany. See Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 293. The US has probably gone furthest in establishing broad immunity
based on a quasi-judicial status. See, e.g., Hoosac Tunnel Dock & Elevator Co. v.
O'Brien, 137 Mass 426 (1884); Babylon Milk & Cream Co. v. Horvitz, 151 NYS 2d
221 (NY S. Ct. 1956). The US is the most extreme as common law has even
provided immunity for bad faith or intentional conspiracy. (Hoosac Tunnel Dock
& Elevator Co., supra). The common law position in the US may be shifting,
however. See, e.g., EC Ernst, Inc. v. Manhattan Constr Co. of Texas, 551 F. 2d 1026
(5th Cir 1977).
569 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1640.

570 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 288, n. 46.
571 Ibid., 294.
572 For a case where fees and arbitral costs were successfully obtained in such
circumstances see Judgment of 12 May 1993, Raoul Duval v. V, Tribunal de
Grande Instance Paris, Revue de l'Arbitrage (1996): 411, confirmed by the Cour
d'Appel de Paris in Judgment of 12 October 1995, Revue de l'Arbitrage (1995):
324 and the Cour de Cassation 16 December 1997 unpublished, cited in Julian
D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative International
Commercial Arbitration (The Hague: Kluwer Law International, 2003), 296, n. 83.
573 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1662.

574 See, e.g., Singapore Arbitration Act s. 16(4).
575 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1641.

576 There are a number of good publications which summarise case law on

arbitrator challenges. One of the best, which covers bias challenges in

international arbitration generally, is Sam Luttrell, Bias Challenges in
International Commercial Arbitration: The Need for a ‘Real Danger’ Test (Alphen
aan den Rijn: Kluwer Law International, 2009).

Part II: The Process of an Arbitration,

Chapter 6: Establishing the
Procedural Framework
Jeff Waincymer,

6.1. Introduction

Chapter 2 analysed at a general level, a range of potentially conflicting powers

and duties of a tribunal that can fundamentally shape the fairness, efficiency and
result of the proceedings. This chapter looks at these issues in the context of
establishing a procedural framework. What attitude should a tribunal take to
managing the process from the very beginning through to the conclusion of
proceedings? This introductory section only seeks to make some general
observations about control and case management. Specific elements are dealt
with in separate sections in this and the following chapter, as many express rules
may limit such general powers or somehow circumscribe them.
It was noted previously that an arbitrator has the broadest discretion to
establish arbitral procedure, subject to consent of the parties and subject to
mandatory procedural norms. While that is easy to state, how a tribunal
determines consent and identifies the constraints on it of mandatory procedural
norms and further, how it determines when and why it may wish to depart from
the preferences of the parties are far more contentious. The latter question is
discussed in sections 6.2.4 and 6.2.5 below. A second key issue is as to timing.
The way powers and duties ought to apply will depend on the particular stage
under consideration. On the one hand, it is desirable that arbitrators turn their
attention at the earliest possible time to all procedural matters that impact upon
the fairness and efficiency of the process. The parties ought to know the
procedural framework as soon as possible so that they can direct their energies
in the appropriate direction. On the other hand, if a tribunal attempts to be too
prescriptive too early, it may constrain the parties prior to it having enough
information about the essence of the case.

Another important issue is that of neutrality and equal treatment. Neutrality

raises challenges where parties come from different legal cultures, given the
page "383" significant differences in approach to procedure and evidence
between legal families. That itself poses challenges for equality of treatment.
Equality also poses challenges on such issues as time allocation, number of
submissions and number of witnesses. Other key challenges are how to balance
or promote flexibility on the one hand and certainty and consistency on the
other. This has led to some differences in view as to the degree to which
guidelines or checklists should be developed by institutions and associations to
assist arbitrators. This is discussed further in section 6.5.

One of the most controversial questions is how proactive a tribunal should be

and who is essentially in charge, the tribunal or the parties. The question is not
only important conceptually, but also guides the tribunal's attitude on many
practical considerations. While this chapter has a number of sections dealing
with specific issues, most are in part dependent on overriding attitudes to case
management or proactivity by the tribunal. The following sections deal with this
general question in some detail in the hope that preliminary conclusions will
guide the responses to individual issues.

6.2. Proactive Management and Party Autonomy

6.2.1. Introduction

In recent times, much has been written about how proactive a tribunal should be
and who should essentially be in charge, the tribunal or the parties. (1) The two
overriding questions are, when is it legitimate to constrain an individual party's
wish to present evidence and arguments in a particular manner and when can a
tribunal legitimately ignore a consensus procedural decision by the parties. (2) As
with many debates, the initial forays can tend to be polarised. One person's
desirable proactivity is another person's intrusive prejudice. In reality, every
procedural decision by a tribunal, whether to do something or not do something,
if appropriately thought out, is a conscious activity. Because there are so many
complex duties that can only be stated in general terms and because various
duties can pull in opposite directions in many complex factual circumstances, an
arbitrator must exercise judgment constantly and is in that sense always
proactive. Such proactivity is generally about constant concern to reconcile due
process, efficiency and consent. At the margins, arbitrators will always be faced
with complex challenges when trying to reconcile these factors.

Calls for greater proactivity should also be looked at in the context of the stimuli
to such developments. Proactivity is primarily about aiming to minimise page
"384" costs and delay. The suggestion is that due process and fairness have to be
considered alongside issues of efficiency and expediency, more so now that many
lex arbitri and rules impose positive obligations on tribunals to achieve those
ends. These issues also need to be considered in the context of the growing
phenomenon of disruptive parties and counsel who may seek to manipulate the
processes in order to frustrate a fair and efficient resolution.

Inherently practical questions such as control over time and cost have to be
resolved in the context of some very significant and contentious questions that
go to the foundation of arbitration itself. If the parties must be afforded at least a
reasonable or adequate opportunity to present their case, when and why should
a tribunal be able to impose a priori constraints on what they are able to do? If
arbitration is based on consent, how can a tribunal ever make procedural
decisions contrary to the parties' wishes? What due process constraints are
imposed on party consent under the lex arbitri, or generally under principles of
public policy? Sections 2.2 and 2.10.3 dealt respectively with the fundamental
role of party autonomy in arbitration and the conceptual reasons why an
arbitrator may at times be entitled to depart from the wishes of the parties. The
latter issue is particularly challenging, as it seems inherently problematic to
ignore the wishes of the parties in a consent-based system.

As is the case with most complex questions in arbitration, complexity arises

because of the tension between fundamental but potentially conflicting goals. On
the one hand, due process, equal treatment and an adequate opportunity to
present the case must be afforded in all circumstances, regardless of party
preferences. On the other hand, fairness need not be considered in a vacuum, but
must in many circumstances consider the preferences of the parties. Any
consensus between the parties as to what is fair must be a relevant factor even if
it would not be determinative in every instance. The duty to render an
enforceable award also makes the tribunal continually mindful that where the
arbitral procedure was not in accordance with the agreement of the parties, this
can be a ground for annulment or refusal of recognition under provisions such as
UNCITRAL Model Law Articles 34(2)(a)(iv) and 36(1)(a)(iv) and the New York
Convention Article V(1)(d). It is within the context of these issues that it is
appropriate to consider the proper ambit of tribunal proactivity.

6.2.2. Applicable Rules, Proactivity and Autonomy

Some might argue that the ambit of reasonable proactivity should simply be
identified by analysing the procedural norms in the applicable laws and rules.
Some lex arbitri indirectly call for more proactivity by demanding timeliness as a
mandatory norm. There is a possible difference in nuance between rules that call
for arbitrators to avoid delay and those which present it in a more positive and
hence stronger manner. For example, Article 21 of the Swedish Arbitration Act
1999 states that ‘(t)he arbitrator shall handle the dispute in an impartial,
practical and speedy manner.’ Article 16.2 of the ICDR Rules requires the
tribunal to ‘conduct the page "385" proceedings with a view to expediting the
resolution of the dispute.’ Some lex arbitri mix both positive and negative
formulations. The English Arbitration Act 1996 speaks of ‘avoiding unnecessary
delay or expense’ in section 33(1)(b) but calls on arbitrators ‘to use all
reasonable dispatch in conducting the proceedings or making an award’ in
section 24(1)(d)(ii). Yet that itself may be circular as section 33 is also framed in
terms of fairness to the parties. Party autonomy could be seen as an essential
element of such fairness. (3) The Spanish Arbitration Act includes a preamble
calling for arbitration to evolve and notes the need for the swiftness typical of
arbitration to accommodate practical demands. (4) Observations by courts may
also give guidance. The Swiss Federal Supreme Court has considered that ‘one of
the objectives of arbitration is to enable parties to obtain a quick solution to the
differences referred to arbitration. Parties that agree to go to arbitration are
therefore bound by the rules of good faith to avoid all that could delay the
normal progress of arbitration proceedings without necessity.’ (5)

An important approach is to consider the hierarchy of norms once these are

identified, although this can also be circular. For example, Article 19 of the ICC
Rules 2012 shifts the order as compared to Article 19 of the Model Law. The
latter makes party agreement the first principle, with the tribunal having a
residual discretion. Article 19 of the ICC Rules 2012 indicates that the proceeding
shall be governed by the rules and only where they are silent, by rules which the
parties or failing them the tribunal may settle on. On plain meaning this might
suggest that the ICC Rules take precedence over party autonomy and by
necessary implication, that a tribunal can rely on the rules to override party
agreement. Yet this ignores the fact that the ICC Rules were only drafted that way
for two reasons, first because they rightly presume that the rules are only
relevant because of party agreement to apply them and secondly, because of a
presumption that the ICC is an institution that does not wish to be bound to any
variation of its proposed model, at least without itself being consulted. (6)

Where the ICC Rules are concerned, some provisions expressly empower an
arbitrator to act contrary to the parties' wishes. For example, Article 18(2) of the
ICC Rules 2012 provides that an arbitrator may conduct hearings at any location
after consultation with the parties. The tribunal must consult, but by implication
need not defer. Other rules may operate as indirect constraints on party
autonomy. For example, Article 22(1) imposes obligations on all parties to make
every effort to conduct the arbitration in an expeditious and cost-effective
manner. page "386" Contextually, this could be argued to be a constraint on the
broader powers in Article 19. Yet Article 22(2) again gives party agreement a
paramount position and in respect of procedural measures, calls on the tribunal
‘to ensure effective case management’. This is all subject to mandatory due
process norms in Article 22(4). Article 22(5) is also relevant when the parties
undertake to comply with any order made by the tribunal, although such orders
must be limited to matters on which the tribunal can validly direct the parties. (7)

Other rules seem to articulate a tribunal's overriding control in other ways. For
example, Article 14.1 of the LCIA Rules indicates that the parties' agreements on
procedure are to be consistent with the arbitral tribunal's general duties as to
mandatory due process norms and its duty ‘to adopt procedures suitable to the
circumstances of the arbitration, avoiding unnecessary delay or expense, so as to
provide a fair and efficient means for the final resolution of the party's dispute.’
LCIA Rule 14.2 allows the parties to circumscribe the tribunal's discretion but
also indicates that ‘at all times the parties shall do everything necessary for the
fair, efficient and expeditious conduct of the arbitration.’ That might be
presumed to imply deference to a tribunal's rulings in that regard,
notwithstanding the party's contrary wishes.

6.2.3. Party Autonomy to vary Institutional Rules

Another challenge to party autonomy arises where party agreement seeks to

both utilise an institution but also alter the institutional rules or practices
without the institution's prior consent. Parties to an arbitration agreement can
only bind each other. When they select a particular institution, it is not based on
an irrevocable institutional offer to accept all cases of any nature at any time and
certainly is not based on an irrevocable offer to utilise a procedure contrary to its
norms.

However, one cannot presume that an institution and/or the tribunal established
through it would necessarily refuse to accept a particular modification agreed to
by the parties. Some institutional rules allow the parties to freely derogate from
any or all provisions. (8) While a number of rules expressly allow parties to make
choices, in other circumstances the presumption is that the institution has made
an offer to provide services on condition that the rules are properly adhered to in
full. (9) This is discussed in more detail in section 3.9.2. The same problem will not
arise in ad hoc arbitration. Article 1(1) of the UNCITRAL Rules 2010 expressly
allows for modification by the parties.

page "387"

6.2.4. When Can Proactivity Override Party Preferences?

For the purposes of this chapter, it is appropriate to distinguish between two

stages, party autonomy before the commencement of an arbitration or at least
before the appointment of a tribunal and decisions thereafter. The first situation
is outside the scope of this chapter and is relatively easy. Parties are free to agree
on any qualifications or pre-conditions to their wish to arbitrate as long as the
agreement itself is valid in law and would be recognised and supported by the lex
arbitri. In the latter case, the key conceptual differences are between mandatory
norms of due process, such as Article 18 of the UNCITRAL Model Law and non-
mandatory provisions. (10) This latter category may be modified by procedural
elements that were articulated in the arbitration agreement or subsequent
agreements that were known to the arbitrator at the time of the invitation to
accept appointment. In such circumstances, the arbitrator has only agreed to
accept the position on the understanding that those procedures will apply. The
arbitrator should honour that agreement and honour the conditional mandate. If
the arbitrator is not happy with pre-appointment procedural agreements, the
appointment should be declined.

The situation is fundamentally different after the tribunal has been appointed
and after an institution has become involved where that is applicable. This
scenario is far more complicated as there are now a range of intertwining rights
and obligations between the parties, the arbitrator and the institution. The added
complexity is that there is no consensus as to the exact nature of those
relationships. Section 2.3 dealt with the question as to whether arbitrators are in
fact in a contractual relationship with the parties and even whether they become
parties to the arbitration agreement. Whether one takes the view that arbitrators
have rights flowing from such contracts or flowing from the inherent status of
their role or a combination of the two, it becomes axiomatic that an agreement
between the parties themselves cannot bind others, including the tribunal,
unless the tribunal has somehow consented to be so bound.

Most laws will contain a provision such as Article 19 of the Model Law which
makes tribunal discretion residual, and subject to agreement of the parties. The
conceptual and practical ambit of provisions such as Article 19 was considered
by the drafters of the Model Law. Holtzmann and Neuhaus note that the
Secretariat suggested that the Working Group amend Article 19 requiring
procedural agreements between the parties to occur before arbitral appointment
so that the arbitrator or arbitrators know the rules under which they are
expected to work. The proposal was rejected by the Working Group, who
considered that party autonomy should be a continuing right. Differing views
were presented before the Commission to either page "388" clarify or revise the
draft, but in the end, no amendment was made. While there were differing views,
‘it was noted that in any case the arbitrators could not be forced to accept any
procedures with which they disagreed, since they could always resign rather
than carry out the unwanted procedural stipulations. Moreover, if the matter
was of strong concern, the timing of any agreement on procedure could be
regulated by agreement between the parties and the arbitrators.’ (11)

The resignation right is not so clear. Acceptance of a mandate under a particular

known lex arbitri or rule system that provides for party autonomy contemplates
arbitral discretions subject to the overriding rights of the parties. Thus the
entitlement to resign must be based on a rule of reason where circumstances
indicate that the exercise of party autonomy on that specific issue is
unreasonable. If the agreement seeks to overturn mandatory norms, and the
parties cannot be convinced otherwise, an arbitrator would be entitled to resign.
It might also be suggested that in some circumstances at least, the agreement can
be ignored and the mandate continued with. The justification is that original
intent is determinative and always comes with an express or implied
understanding that it cannot overturn mandatory procedural norms. In those
circumstances, a properly appointed arbitrator might simply ignore the
improper later direction. The parties still have an overriding power to bring the
arbitration to an end and remove the arbitrator, subject to the obligation to pay
outstanding fees and perhaps damages for breach of the arbitral contract.

It was argued previously that it is more problematic if the agreement changes an

earlier agreed process, but does not clearly offend against a mandatory norm.
The difficult scenario is where the parties preferred trade-off between fairness
and efficiency differentiates significantly from that of the arbitrator, but not in a
way which unassailably interferes with due process. Even then it was argued
that if the parties purport to change the arrangement contrary to the reasonable
expectations that the tribunal would have had at the time of appointment and the
arbitrator is unwilling to accept the change, the arbitrator is again entitled to
resign without any liability for failing to complete the mandate and in most cases
with an entitlement for fees to date. There are good reasons for this. Party
autonomy cannot force an arbitrator to take unreasonable steps from his or her
own perspective. For example, the parties cannot agree on a particular hearing
date that is subject to a prior commitment by the arbitrators and try and force
them to change their commitments. As has been observed, a strict literal
application of Article 19 would lead to some farcical scenarios. Michael Pryles
gives the example of parties agreeing to a five-year period for the exchange of
memorials. (12) One suggestion is that there could be said to be an implied term
‘that the parties should not act or behave in page "389" such a way as to
prejudice the arbitrator’. (13) They must only be able to exercise such a power
‘within assumptions reasonably held by the arbitrators at the time when they
accept their mandates’, (14) or ‘within usual or common parameters for
commercial arbitrations of the type and nature of the arbitration before the
arbitral tribunal.’ (15) To the extent that there should then be limitations on party
autonomy under Article 19 type provisions, the better view is that autonomy is
not unlimited, but is circumscribed by the need for the agreement to be within
the reasonable and legitimate expectations of the tribunal at the time of
appointment. (16)

Even more troubling hypothetical scenarios are where the parties have come to
an agreement that the tribunal simply does not believe is the most suitable
means for dealing with the issues at hand, although this might not be common in
practice. One such problem facing an arbitrator is where counsel for the parties
advocate particular procedural norms simply because of their familiarity with
particular litigation techniques. Many experienced practitioners have decried the
judicialisation of arbitral procedures as a result of this phenomenon. (17) A pure
reference to party consent might imply that the parties' preference should win.
However, that consent was to empower an expert adjudicator to exercise fair
and reasonable judgment from time to time. That articulation might incline to
the view that in many cases, a tribunal could reasonably overrule party
determinations on the basis of fairness and efficiency. Particular factual
scenarios may always be problematic as the most extreme cases can be dealt
with on the basis that they are contrary to a mandatory law of due process or
contrary to the representations or reasonable expectations made at the time of
appointment. One of the difficulties with such a test is that it is always
debateable and the initial determination is made by the page "390" tribunal
itself. While that is not a classic conflict of interest situation, the tribunal must be
aware that it is difficult to dispassionately evaluate a proposition of the parties
that simply did not appeal to the tribunal.

Carter argues that notwithstanding the priority given to party autonomy in

various arbitral rules, ‘it seems implicit in those Rules that an arbitrator has the
right, to be exercised prudently and presumably not frequently, to impose
procedures in the interests of expedition and fairness, even when both parties
(or, more likely, both sets of counsel) may wish some other procedures.’ Carter
argues for some ‘recognition of an arbitrator's inherent right to arrive at a just
decision with a reasonable degree of independence in raising and defining the
necessary issues.’ (18) Reiner has also suggested that there may be limits to what
the parties may legitimately agree to procedurally in terms of detail. Thus they
can agree on the rules but can they necessarily agree on the number of days for
the hearing? (19) Redfern and Hunter speak of the gradual shift in control of
proceedings from the parties, who at first are the only ones who understand
their case, to the tribunal that ultimately determines which facts it regards as
relevant and which questions of law it regards as important and hence how the
case should be presented. (20) Born also suggests that the duty to cooperate can
be a counter to the parties' general autonomy in cases where they seek to have
the matter dealt with in ways contrary to the arbitrator's preference. (21) This
relates to issues of proactivity. The duty may also be an indirect way to allow
tribunal control over counsel behaviour.

These observations are important but it is easier to state broad propositions as

to a tribunal's entitlement than to identify some criterion by which it can
determine when to override party consent. As noted throughout, there are likely
to be grey areas where a tribunal feels conflicted between the duty to uphold due
process obligations and a duty to promote fairness and efficiency, while at the
same time supporting the duty to respect the autonomy of the parties. Examples
of the contentious cases include party agreements to:

1. lengthen the total time period until a final award;

2. have longer oral hearings than the tribunal thinks is appropriate;
3. have more witnesses than are thought to be necessary;
4. have excessive document production processes;
5. agree on a documents-only determination when the central issue is one such
as which of two key conflicting witnesses is believed to be telling the truth;
page "391"
6. deny the tribunal the ability to appoint its own expert;
7. grant each other extensions because both are ill-prepared.

If the parties have notified their intent to adopt an unreasonable procedure from
the outset, this should be raised by the arbitrator before accepting appointment.
If the agreement of the parties is subsequent to appointment, appropriate
counselling by experienced arbitrators will often resolve the issue. This is
particularly so when it occurs face-to-face in a preliminary or final hearing. It is
much more difficult to take an intransigent position when looking a leading
arbitrator in the eye. Because of the duty to educate, tribunals could consider
party autonomy in the context of guided autonomy where they help parties
understand what needs to be done for a fair and efficient resolution of the
dispute. As has been suggested, if wise counsel does not prevail, it may in
appropriate circumstances give justifiable cause for resignation. The most
challenging situation is where an arbitrator purports to continue with the
mandate, overriding the parties' direction.

In a practical sense at least, what is permissible by an arbitrator is also

dependent on the extent to which a court is prepared to interfere with
procedural determinations contrary to the will of the parties. While lex arbitri
may indicate a hierarchy between party autonomy and tribunal discretion, the
divergent conceptual perspectives suggest that the optimal approach is for a
dialectic relationship between them, with regular consultation and soliciting of
the parties' views and suggestions. (22) Nevertheless, Herrmann also observes
that absent party agreement, it is ultimately for the tribunal to take a decision
and there is no obligation to consult in every single case. (23) Conversely,
Newmark suggests that as a general rule, if parties want to adopt a procedure
contrary to the tribunal's preferences, as long as this does not offend against
mandatory norms, this should be accepted out of respect for party autonomy. In
circumstances where the parties themselves cannot agree, the tribunal should
look to the lex arbitri and/or arbitral rules to determine what procedural
mandate it has been given. The mandate as given emanates from the original
consent of the parties and hence can be employed as the tribunal sees fit,
consistent with respect for party autonomy. (24)

Another factor which might be referred to in order to suggest a stronger position

for party autonomy than proactive entitlements of a tribunal is the fact that
modern rules call on a tribunal to consult with parties before setting a
procedural timetable. There is no equivalent obligation on parties to consult with
an arbitrator prior to coming to a procedural decision as per their entitlement
page "392" under Article 19 of the Model Law or equivalent provisions in other
lex arbitri. (25) However, those rules simply refer to an obligation to consult and
not to defer, hence suggesting that in many or most cases the tribunal can have
the final word.

Finally, in relation to party autonomy, one question is whether either party can
resile from an agreement unilaterally. Normally that would not be so. One
contentious situation would be where the parties have agreed to a procedure
and the tribunal being concerned as to its fairness or efficiency seeks to convince
the parties to change their agreement and only one agrees to do so. The
appropriate action would depend on the tribunal's power. There are situations
where a party may wish to vary an earlier agreement because it was
subsequently felt to offend against mandatory principles of due process. If the
facts show this to be the case, the tribunal should not feel bound by the
agreement. In such circumstances, the party ought to be able to resile from the
agreement unless that was a condition of the arbitration clause itself, in which
case the entire agreement may be flawed if it was seen as conditional.
6.2.5. Practical Considerations in Determining Whether to Override Party
Preference

Where a tribunal has discretion to override an agreement of the parties, it is

appropriate to consider the factors that may be relevant to such a discretion. One
factor likely to be relevant is the reason why the parties came to such an
agreement. For example, where parties seem to wish to lengthen the overall time
periods, this may be appropriate given new developments, complexity,
contemporaneous legal proceedings or a wish to explore settlement options. (26)
At the other extreme might be situations where counsel has simply not been
diligent in properly preparing the matter.

Circumstances of the parties must also be considered in the context of the

tribunal's own circumstances, although care needs to be taken with potential
conflicts of interest in terms of scheduling and reward for service issues. An
obvious example would be an arbitration where the arbitrators will only be paid
on completion and where the parties are requesting to extend that time simply
because of their tardiness. Another scenario is when the parties are calling for a
lengthier hearing time in an arbitration where fees are on an ad valorem basis.
Under the reasonable expectations test, if the parties have good reason to extend
the time, such as with a view to settling and saving cost, the tribunal ought to
defer. Stated differently, a procedural variation agreed to by the parties that is
inherently in their page "393" interest in seeking an overall fair and efficient
resolution would rarely, if ever, fall outside the legitimate expectations of an
arbitrator on appointment and hence should generally be deferred to, even if it
modestly alters the cost benefit value of the work of the arbitrator. If the
variation is more extreme, it is harder to say that it would be within legitimate
expectations. Payment methodology may be a factor in those expectations. If the
parties wish to take longer than the arbitrator thinks is appropriate and the
arbitrator is paid on an hourly or daily basis, there may be a greater argument in
favour of deference to their wishes. Even that may be problematic if the extra
time is unreasonable and adversely impacts upon an arbitrator's ability to
complete other mandates optimally.

6.2.6. How Proactive should a Tribunal be When Unconstrained by the

Parties?

The next issue is how proactive a tribunal should generally be in directing the
parties. One argument in favour of proactivity is that most arbitral rules and
guidelines when revised, seek to increase the entitlement or even obligation on
arbitrators to adopt such approaches. That is certainly the corollary of requiring
attention to be given to timeliness and efficiency. The Queen Mary survey
suggested that it was pervasive throughout the results that the parties preferred
proactive arbitrators. Similarly they preferred proactive arbitration institutions.
(27) One reason why proactivity is important in arbitration is that so much is

delegated to arbitral discretion, particularly as to time lines and stages. Systems

that entail passivity of adjudicators will typically provide fixed time lines to
ensure expediency. Conversely, those with broad discretions must expect that
adjudicators will actively seek deadlines and other procedures appropriate to
the circumstances of the individual case. (28)

It is also relevant that subject to due process and other limits, parties can agree
on issues at any time, including up to the end of proceedings. It is important to
provide an ongoing opportunity for consent-based decisions as well as to
support consent-based decisions already agreed to. For example, as each
procedural matter becomes relevant, an arbitrator might offer the parties an
opportunity to come to an agreement on the matter and only make a direction if
necessary. If the parties reach an agreement, the advantages of this approach are
obvious. A downside might arise where one party habitually aims at dilatory
tactics. An arbitrator who adds time for the parties to try and come to an
agreement where agreement is never reached, is giving the dilatory party an
extra opportunity to stall the process. While that may be a concern,
appropriately set time limits for those negotiation processes should ensure that
the overall delays are not unreasonable.

page "394"

There is still a conceptual question as to the proper role of an arbitrator. At one

extreme is the view that it is for the parties to do all necessary research and
prepare the case and that a tribunal can expect to be taken through all necessary
elements at the hearing and in the context of final written submissions. At the
other extreme, numerous authors assert that it is important that arbitrators
become sufficiently familiar with the dispute at the earliest stage so as to be
proactive, properly set out terms of reference and timetables, (29) and otherwise
guide the parties. Over time, all legal systems have gravitated to a position where
case management is a core value, but tempered by the overriding pursuit of
justice. Both historical extremes have been seen as suboptimal. If common law
adjudicators are too laissez faire, the system becomes too costly. If civilian
adjudicators are too controlling, a true opportunity to present one's case might
be unduly restrained.

It is not about being proactive or not per se that matters, but the degree of
intervention that is most appropriate to give effect to the due process rights of
the parties in the context of efficiency obligations that also apply. In that sense at
least, there can be said to be a duty to at least consider these issues, flowing from
the general duty to afford due process. While some would question an interactive
approach, Schneider makes the apt observation that in many procedural
decisions, this is occurring in any event. This is so where terms of reference are
drawn, when posing questions, when deciding on the relevance of certain
evidence being challenged and when questions are posed to expert witnesses. (30)
For that reason, the ideal degree of proactivity cannot simply be resolved in
relation to a duty of equality as every intervention is inherently unequal. (31)

6.2.7. Proactivity and Enforceability

One possible undesirable corollary of the duty to render an enforceable award

and practical concerns about challenges to enforceability may be to provide
incentives for arbitration to unduly mimic judicial proceedings, simply to
preclude challenges. That would be undesirable. Other undesirable outcomes
would be excessively long and expensive procedures that simply aim to give each
party everything it asks for, again to minimise the scope for challenges. A further
undesirable incentive is for arbitrators to give undue attention to crafting
awards in ways which are less open to challenges, at least where this dissuades
them from utilising page "395" principles and arguments that they would
otherwise wish to apply. A final undesirable incentive would be for arbitrators to
feel constrained from making observations about inappropriate behaviour of a
party, lest that be used as a ground for challenge.

One complex situation is where a tribunal suspects that the likely enforcement
court may interfere with enforcement on grounds that would not be uniformly
respected by the arbitral community. A related concern is where a party seeks to
put undue pressure on an arbitrator by threatening challenges to enforcement
unless particular procedural determinations are made. The preferable view
would be for an arbitrator to simply make procedural determinations that are
reasonable in and of themselves, which should also serve to pre-empt legitimate
challenges to enforcement. The tribunal should ignore threats that are
dependent upon unmeritorious approaches by enforcement or annulment courts
and which aim to encourage suboptimal and self-serving procedural
determinations. However, when considering contentious issues of proactivity,
the more borderline the scenario, the more likely that there will be court
scrutiny, in some cases by judges who do not have pro-arbitration attitudes. (32)

6.2.8. Elements of Proactivity

It is important to quickly shift the proactivity debate from the general conceptual
level to a practical level. At the conceptual level it is easy for any counsel to
casuistically argue that any interference with their fundamental freedom of
expression interferes with their ability to present their case. At the practical
level, however, many such arguments are spurious and the tribunal may wish to
direct the parties' counsel in a range of circumstances. It is therefore appropriate
to consider some practical examples, roughly in their likely chronological order.

6.2.8.1. Proactivity and Appointments

The first proactive thing a prospective arbitrator can do to promote timeliness is

to refuse appointments when there is too strong a chance that they will be
unable to complete their work in a timely fashion due to other commitments. (33)
While there can be a reluctance to reject an overture as to appointment on the
basis that another may not be forthcoming, if a prospective arbitrator explains
the reasoning and indicates when their schedule will become more open, this
should only enhance their reputation. Taking on too much work and then not
handling it in a timely page "396" manner is also likely to be a more significant
problem in terms of ongoing reputation, (34) although the impressionistic view
that this is a significant problem with the most famous arbitrators who
nonetheless are regularly reappointed, suggests that the ‘market’ for arbitrator
selection is not a perfect one.
One reason why counsel might habitually select arbitrators who are unlikely to
be able to handle the matter in a timely fashion is that they are themselves
unwilling to take the reputational risk of appointing a less experienced
arbitrator. Respondents may even consciously prefer to have arbitrators who are
unable to handle the matter in a timely fashion. Even a claimant may wish to do
so if a respondent cannot cope with the uncertainty of delayed proceedings and
might be more disposed to settle if a case would drag on and disrupt their
liquidity, business planning or reporting.

6.2.8.2. Proactivity and Preparation

The next opportunity is when first establishing the procedure. The ICC
Techniques publication suggests that the tribunal should consider informing the
parties that it will proactively manage the procedure to assist them in resolving
the dispute efficiently. (35) This would include highlighting matters that have not
been sufficiently addressed by the parties so that they can be forewarned of the
tribunal's views and adequately present their case as best as it can be put. At the
very least, the better view is that an arbitrator should be sufficiently on top of the
issues and materials as they are presented so that sensible time frames can be
prepared, adequate terms of reference can be drafted where they are to be used,
requests for production of documents can be dealt with sensibly and appropriate
decisions can be made as to the appropriate number of witnesses and the length
of the hearing. Appropriate preparation and planning should also allow the
hearings to be tailored more effectively towards the key issues. (36)

6.2.8.3. Proactivity and Order of Processes

A number of steps should be non-contentious. Control and case management

powers operate throughout the hearing. Hence an arbitrator is not obliged to set
a comprehensive framework from the outset but may instead make
determinations from time to time as they seem appropriate. This also includes
the power to review determinations previously made. While arbitration is
inevitably divided into distinct stages, it is important to understand that these
are all elements of a unified process to identify the correct result within the
fairest and most efficient framework. Document production and oral evidence
are not conceptually distinct page "397" elements, but are instead, different
forms of often related evidence that ultimately need to be integrated within that
process. The point to the observation is that giving each party a true opportunity
to both present its case and respond to that of its opponent requires thought to
be given to the way the various stages will best be handled and integrated.

Proactivity can thus call for sensible consideration of the order of the procedure.
In some cases the traditional order might be reversed, for example, hearing
issues of quantum in advance of liability where there is a suggestion that the
form of relief sought is inappropriate, for example, where the contract precludes
loss of profits but this is the sole claim or the sole form of damages conceivable
on the facts. While the typical case involves claims and defences, document
production, then witness statements and then hearings, if for example, it is
obvious that a case turns on which of two conflicting witnesses is more credible,
it makes a lot of sense to conduct a quick hearing, allowing each side to cross-
examine the other's key witness. (37)

In some cases there may be a large number of claims between the parties but the
majority of these may be dependent on a gateway issue. This is typically the case
with breakdowns in long term relationships where there is an initial dispute
alleging breach of an obligation, which then escalates into a number of self help
and retaliatory measures, the validity of which may depend on the conclusion as
to the initial liability. If certain issues would obviate the need for expensive
evidence on other matters, the presumption is that these should be dealt with
first and discretely. A countervailing concern is that if the later issue may need to
be dealt with anyway, separate consideration of the preliminary point may lead
to wasted time and travel costs for necessary witnesses. Poudret and Besson,
commenting on procedural determinations, also suggest that where a question
will inevitably or at least probably arise, it should be immediately addressed to
promote foreseeability and prevent the applicable rule being dictated solely by
the desired result. (38) Another possibility is to order claims in terms of
importance. David Rivkin uses the example of a hypothetical with a hundred
claims where a tribunal might call for the most significant ones to be handled
discretely in the hope that settlement of the rest will occur after the key ones are
resolved, or their resolution might show the way forward for the parties as to the
balance. (39)

6.2.8.4. Proactivity and Cost Minimisation

Particular aspects of cost minimisation and efficiency are discussed in various

sections throughout this book. This section only seeks to make some general
page "398" observations on some select issues. Chapter 2 noted a duty of
efficiency, now commonly enshrined in lex arbitri and arbitral rules. Rule
changes have tended to follow demand from consumers. A recent study by the
Corporate Counsel of International Arbitration Group (CCIAG) considered that
international arbitration ‘takes too long’ and ‘costs too much’. (40) While
complaints as to cost must always be considered, it is always important to
consider how reducing cost may impact upon justice and what level of costs
could be seen as ‘right’ in the context of individual disputes. A particular
difficulty is that the main costs tend to be counsel costs. One practitioner has
impressionistically suggested that the vast bulk of total costs are in the parties'
own legal fees. In considering a hypothetical USD 20,000,000 dispute with a total
of USD 2,000,000 in costs, he thought the break-up would be legal fees 80%, legal
costs 2%, party costs 3%, witness and expert fees and costs 5%, arbitrator fees
and costs 9% and administrative fee 1%. (41) If this is an accurate picture across
the board, it diminishes the cost differences between ad hoc and institutional
arbitration. (42) The difficulty with this scenario is that tribunals do not have
direct control over counsel fees. Even here a tribunal can have a significant
indirect impact on counsel costs in terms of what they are directed to do or not
do. Because there is a duty of efficiency, a tribunal must consider these issues. In
Pillar v. Edwards (43) a court considered that an arbitrator breached his duty to
adopt expeditious and efficient proceeding when total costs and fees accrued to
nearly four times the value of the original claim and where there were a range of
procedural measures that could have been adopted to avoid this result.

In some situations, cost reduction could be non-contentious. Proactivity could

simply be a matter of ensuring that there is no unnecessary duplication. For
example, there is a need in oral hearings with written submissions to have the
oral stage work seamlessly with the written phase and not duplicate it. If using
written submissions, a tribunal should ordain that as much as reasonably
possible be done in the paperwork as it is less costly than oral hearings. Oral
hearings can then be for cross-examination and redirect examination, primarily
in testing and bringing the witnesses to face conflicting evidence. Ideally, oral
hearings should be less concerned with expanding on what was already said, as
this would not be necessary if well said.

Other aspects of proactivity are whether a tribunal should encourage the parties
to seek an agreed statement of facts. In principle, that should generally be
desirable, although there are more conceptual challenges if a tribunal uses a
threat of adverse costs awards where even a successful party puts its opponent
to unnecessary proof on matters that the tribunal believes ought to have been
page "399" conceded. That should not be problematic if the tribunal gives the
parties sufficient advance warning. At the very least, the parties could come to a
contrary agreement demanding that costs simply follow the event.

Another feature of a proactive arbitrator is one who ensures that proper

attention is given to whether typical elements of procedural design are truly
needed. David Rivkin has advocated the use of what he describes as a Town
Elder model. Using the historical analogy of business persons taking disputes to a
wise person who they trusted for a speedy resolution of their dispute, his
suggestion is that arbitration should begin with this minimalist notion and only
add additional procedures that are necessary for the particular dispute. The
tribunal should not presume that the historical norm of multiple written
submissions, document production, witness statements and oral hearings should
apply as a matter of course. (44) When considering what is necessary under a
Town Elder model, it is appropriate to consider whether a step is necessary in
the pursuit of truth or in aid of the appearance of due process or both. David
Rivkin has wisely counselled the arbitration community to consider a dispute in
the context of the parties broader business activities. He suggests persuasively
that under the modern hectic pace of international commerce, taking two years
to resolve a dispute is somewhat at variance with the aspirations of parties and
minimises the chance that they could find new ways to work together. (45)

An important and highly controversial question is whether notions of

proportionality should apply in arbitration, that is, should the degree of relevant
evidence permitted be reasonably proportional in cost to the amount in dispute
even if other evidence might also be material? When discussing the duty of
efficiency, in sections 2.7.9 and 6.2.8 it was suggested that many modern systems
accept that legal adjudication must take cognisance of proportionality, if for no
other reason than that a correct decision that costs more than the amount in
dispute is neither efficient nor fair. This is of course highly contentious and some
lex arbitri would allow for setting aside in some circumstances. There may be
other means used to provide low cost incentives that do not directly bar
presentations. One such option is to warn the parties that there will be a cap on
recoverable costs or at least careful scrutiny of the cost of each element,
regardless of the substantive outcome. Even these are contentious aspects of
proactivity, absent directions from the parties.

There have been suggestions to use performance indicators and statistics to

evaluate arbitrators and institutions. (46) While the aspiration of transparency
and full information is a laudable one, there are significant problems with any
measurement process. Typically these will lead to undesirable and unintended
consequences when participants seek to improve formal results in the measures
adopted without page "400" sufficient concern for negative effects. An example
would be speedy arbitrations that cut corners. Of course participants can be
asked to evaluate the overall process but after the result is known, the objectivity
becomes compromised.

Efficiency issues are also more problematic when one party is not cooperative.
This is discussed further in section 6.2.8.12.

6.2.8.5. Proactivity and Time Efficiency

One of the most significant aspects of proactivity is in relation to timing and time
efficiency. Many domestic litigation systems involve lengthy delays before
matters can come before a judge. One of the hoped for advantages of arbitration
was that it would be more timely. Eric Schwartz has suggested that, in
arbitration, ‘time is of the essence’. (47) ICC Techniques points out that controlling
time is a key way to control cost. As noted, arbitrators are now often expressly
given duties of expediency and efficiency. Unfortunately, an analysis of arbitral
practice over recent decades show that timeliness is often not the case. Gary
Born suggests that for many complex matters, arbitration will often take
between eighteen and thirty-six months to reach a final award. (48) Relative
timing also depends very much on developments in domestic litigation
processes. These range from jurisdictions where it can take many years to get
before a court to others that have moved to significantly reduced timelines.

There are many reasons why arbitral proceedings might take more time than is
ideal if a tribunal does not devote sufficient attention to planning expeditious
proceedings. First, any international dispute will add potential timing problems
where face-to-face hearings are required and different schedules must be
coordinated. International arbitrators are not designated public servants who
can be directed to continue a hearing once it is commenced, although it is
important to note that that model is primarily the common law style. It is
common in civilian litigation systems for staged hearings. In arbitration, leading
arbitrators with busy international practices can inadvertently delay
proceedings simply because they have narrower windows of opportunity where
a lengthier hearing schedule is required. A multi-person tribunal must also seek
to accommodate the availability of the parties, their counsel and key witnesses.
The fact that some parties may be working in a foreign language also requires
appropriate time to be given so that the interests of justice are supported. One
disadvantage of arbitration in clear-cut cases is that there is no process for
summary judgment which many domestic litigation systems utilise so that
spurious defences cannot delay a deserving complainant.

Other reasons why arbitration may be delayed include the need to make
determinations on discretionary procedural matters and possible challenges to
page "401" impartiality and independence of arbitrators which importantly, can
occur at any stage of the proceedings and not simply at the outset. (49) There may
also be challenges as to the validity of the arbitration agreement and general
jurisdictional challenges. Multi-party situations often take longer to resolve
because of greater uncertainty in arbitral theory and rules. The need to find
compromises between civil law and common law evidentiary traditions often
add to the time and expense where elements of each are adopted so that no party
can claim that they were denied their traditional means of presenting their case.
For example, large arbitrations see multiple rounds of written submissions as
per the civilian tradition, but still have direct examination, cross-examination
and at times re-examination of witnesses on top of that. Some also allow post-
hearing written submissions.

The very discretionary nature of many key elements of the arbitral process also
means that extra applications are often needed to determine what procedural
processes will be adopted. Examples include applications for production of
documents and related challenges about confidentiality and privilege. Another
reason for delay is that an arbitrator concerned to render an enforceable award
may afford more leniency in the face of a range of procedural challenges along
the way that might be responded to more forcefully in a domestic litigation
environment. This is particularly so where extensions are sought.

Finally, there are problems of dilatory tactics by the parties. This is discussed
below in section 6.2.8.12. It is hard to deal with extension applications and
missed deadlines by respondents that are likely to be aiming to delay payment as
responsive behaviour still has to ensure they have had an adequate opportunity
to present the case. While it is more commonly the case that a reluctant
respondent may wish to delay proceedings, in some cases it is the claimant that
is tardy. More often than not this will be a result of inexperience with arbitration
and a lack of application of sufficient resources by counsel and/or its client. At
other times it may be because the amount of ultimate damages may benefit from
a protracted dispute, hence the delay is strategic. This might arise where a new
venture is likely to display greater profitability after a number of years and
hence make calculation of damages appear less speculative. It may even be that
an applicable interest rate may exceed a commercial rate. It is difficult to deal
with a claimant who is not proceeding with the arbitration with due diligence
and efficiency. Lex arbitri may be silent on this issue or may have restrictive
grounds for dismissing an application for want of prosecution. (50) Section 41(3)
of the English Arbitration Act 1996 allows a claim to be dismissed for inordinate
and inexcusable delay on the part of the Claimant in pursuing this claim.

One should also be careful in presuming that a timely arbitration will necessarily
be cheaper. For example, if very strict time frames are imposed in a complex
matter, a large law firm might simply respond by putting a very big team on the
case to ensure that nothing is missed. The costs to that party would not vary. If
there page "402" is a small law firm on the other side, they may be unable to
cope with the reduced timeframe. A lengthier timeframe can even allow for more
care to be taken in terms of promoting efficiency. If a more relaxed timeframe
was allowed, one coordinating lawyer from the large firm might be able to
identify the necessary steps more confidently and adopt a more reasonable
approach. While that might not be the norm, the point is to not simplistically
presume how parties and counsel are likely to respond, but give some thought to
this, particularly as it is readily accepted that the bulk of the costs of arbitration
are in representation fees, not arbitrator fees and expenses.

6.2.8.6. Proactivity and the Duty to Warn Parties as to Matters of Concern

The next question is whether tribunals have a duty to warn parties as to the
matters that are central to their deliberations and/or as to the inadequacies of
arguments and evidence as already presented, at least where further information
would assist. While leading arbitrators have legitimately differing views, the
trend seems to be for greater intervention by arbitrators with a view to meeting
the goals of fairness and efficiency. Preamble 3 of the IBA Rules on Evidence
states:

Each arbitral tribunal is encouraged to identify to the parties as soon as it

considers it to be appropriate, the issues that it may regard as relevant and
material to the outcome of the case, including issues where a preliminary
determination may be appropriate.

Efficiency would suggest that the tribunal should advise the parties as to the
matters of greatest concern as early as possible so that no work is wasted.
However, if a tribunal does so too early, it may be misguided in its views. New
issues and facts may emerge over time and careful reflection may alter what
were originally perceived to be the key issues. In many cases, the tribunal might
wish to at least see one round of submissions by the parties before trying to
narrow their attention in this way. Alternatively, this might occur when the
parties are seeking to agree on terms of reference. For an arbitrator to undertake
this process successfully, there is also a need to be prudent in discussions with
the parties, ensuring that they understand that the tribunal retains an open
mind. (51) Counsel should feel comfortable in seeking to argue that the tribunal
has erred in its selection of key issues. (52)

Fixing gaps in submissions in a timely and efficient manner is also important and
may raise similar challenges. If written submissions and witness statements are
inadequate, thought might be given to asking specific questions. It is acceptable
for page "403" an arbitrator to ask questions or call for documents and witnesses
in cases where the evidentiary record presented by the parties is considered
inadequate. Similar proactivity concerns arise where a tribunal is uncomfortable
with a witnesses’ evidence. Evasive witnesses can simply lead to conclusions as
to veracity. At times it may be appropriate to provide a judicious warning about
the possibility of drawing adverse inferences, although here the language needs
to be very circumspect and not allow for suggestions of prejudgment. If the
witness is not being taken to the matters of importance, the tribunal might give a
direction to do so.

6.2.8.7. Proactivity, Suboptimal Counsel and Inequality of Parties

An educative function may arise as a corollary of allowing the parties an

adequate opportunity to present a case. A tribunal may need to explain to
inexperienced counsel the way arbitration procedures ought to evolve. If there is
a significant imbalance between the parties, this might also be a corollary of
treating the parties equally. Duties of efficiency may also involve giving
appropriate support to the parties, as might the duty of commerciality and the
duties to cooperate and act in good faith. The need to educate about legal
principles may be more significant with investment arbitration where parties
and counsel may be less familiar with proper interpretation of public
international law norms and the way they impact upon joint venture agreements.

Where both parties are inexperienced, the arbitrator might hope to undertake
educative obligations without fear of being perceived as treating the parties
unequally, although some could still see this as being too intrusive. Where the
educative function flows from differences in ability between counsel, that is
more sensitive again, as an arbitrator must not appear to be favouring the less
experienced side. Where the educative role is concerned, there is also a
difference between true guidance and paternalism. There is often a fine line
between seeking to educate the parties and autocratically directing the process.
(53)

A tribunal need not ‘correct natural inequality’ in the context of equal treatment
of the parties. (54) If the duty or power arises from a proactive entitlement to
clarify ambiguous claims, this may give an unfair advantage to less competent
counsel, although Wetter also makes the point that not asking a question is as
relevant as asking one. (55) The behaviour should be acceptable where it is truly
seeking to resolve the ambiguity for the benefit of the tribunal. The problems in
page "404" dealing with incompetent counsel who fail to take points that seem
obvious and the challenges this poses for an arbitrator were noted by Blessing.
He felt that a case-by-case solution may be more appropriate than identification
of a general principle. (56) There are good reasons for this view, not the least
because all arbitral interventions guide parties to some degree at least. It would
be almost impossible to ensure than any kind of intervention by a tribunal does
not aid a party in raising issues or presenting arguments or evidence it otherwise
would not have chosen to do. (57) For example, asking for clarification from a
witness who has inadequately particularised the elements of a claim can only
help the party with the burden of proof to meet its obligations. Asking whether a
particular claim is accepted or denied is likely to induce the latter response,
although this can be justified on the basis that the tribunal has an overriding
obligation to identify those matters in dispute and those matters which are
conceded where there is any doubt.

A distinction also needs to be drawn between guiding counsel as to matters

already raised or interventions which encourage introduction of new ideas not
included in claims or counterclaims. While this remains a contentious question,
the majority view would be that a tribunal ought not alert a party to a defence
that might be arguable on the facts but which was not raised. This would even be
so in respect of an irrefutable argument such as that an applicable statute of
limitation time period has expired, particularly under legal systems that allow
parties to waive their entitlement to such defences. The situation would be more
contentious under legal systems that would see certain defences and set-off
rights as applying automatically. This issue is discussed further in section 13.9
dealing with the duty to know the law and section 6.2.8.10 dealing with the
power to decide on matters not addressed by the parties.

Even equal treatment will not necessarily shield an arbitrator from criticism.
Calvo refers to an ICC case where a sole arbitrator, a retired French judge, was
addressed by two inexperienced lawyers and began to give advice to both
lawyers on ICC arbitration. He was challenged. The court rejected the challenge
but apparently told the arbitrator in confidence, through the Secretariat, to
refrain from such advice, at least in writing.” (58)

A final question is whether an obligation to take action in response to inadequate

counsel goes so far as to say in an extreme case that an arbitrator should advise
the party to select different counsel in the face of general incompetence. There
seems no reason in principle why an arbitrator should not be able to do so in the
most extreme circumstances where the parties' rights are not being meaningfully
exercised. Of course, such a factual scenario should be clear-cut.

page "405"

6.2.8.8. Proactivity and Preliminary Views

One of the most contentious issues is whether a tribunal should give an

indication of its preliminary views under principles of proactivity. A tribunal
might argue that this is the best way to afford the parties a meaningful
opportunity to address the things that matter. Yet in such circumstances, a party
might accuse a tribunal of prejudging the matter simply through its directions as
to the important matters for consideration. A tribunal would need to be very
careful with the language it uses to this end. In many instances an arbitrator can
be proactive without giving an indication of the likely result. For example,
directions may be given as to which factual issues might benefit from further
evidence. The tribunal might make it clear that on certain issues, it believes it has
enough documentation and argument to be confident in its views and hence,
further information that merely corroborates existing material will not change
the outcome. Conversely, the tribunal might indicate that in other areas it is
having difficulty understanding and reconciling conflicting evidence and would
benefit from further material or at least further elaboration.

Böckstiegel suggests that a tribunal may appropriately indicate its preliminary

views so as to concentrate the parties' attention on the key matters in dispute.
(59) He referred to a Procedural Order by which the arbitrators ‘informed the
parties of their present inclinations and asked for certain specific information
and comments in this regard before the tribunal came to its final decision’. (60) In
this sense the proactive approach is sometimes described as an interactive
approach or dialectic, where the parties are given an indication of preliminary
views and then given an opportunity to adapt arguments and complete evidence
so as to best influence those views. (61) Schneider argues that the parties should
not be left in the dark about an arbitrator's thinking but should be invited to aid
the tribunal in its evolution of its thinking. (62) A proactive approach is also
supported by Bernardini. (63) While some might suggest that in doing so a
tribunal displays some prejudgment of the issue, this should not be so where the
tribunal is only pointing out the state of evidence and the natural inferences from
the body of evidence at hand. This is particularly so where there are significant
evidentiary gaps. Raeschke-Kessler takes a different view to Böckstiegel,
asserting that a tribunal should only discuss its preliminary view with consent of
all parties, otherwise this could be seen as a violation of the obligation to remain
impartial and independent. (64) There is certainly a risk of such page "406" an
accusation, but in principle, the complaint would not appear to be valid.
Ultimately, every arbitrator is inevitably partial as to a conclusion, in that a
decision is made in favour of one party over the other. Directing people as to the
key elements that still need to be presented is more properly to be seen as an
honest articulation of a particular stage of an impartial process, rather than an
interference with impartiality per se.

There can be no unassailable view either way as norms of fundamental due

process can be brought in aid of opposing arguments. Those opposing an
indication of a preliminary view may be concerned with equality of treatment of
the parties as any warning will typically help the party who has as yet failed to
satisfy any onus of proof. To those arguing the contrary position, a meaningful
opportunity to present a case should inexorably include an indication of the
nature of the case as seen by the tribunal and not simply as articulated by
counsel. Inevitably, tribunals will need to consider these issues on a case-by-case
basis. A tribunal might simply warn the parties that there is an inadequate body
of evidence but that principles of burden of proof will lead to a particular
outcome unless a more significant body of evidence is provided. Even that is
contentious as the opponent of the party who would fail on a burden of proof
issue may feel disadvantaged by such an intervention.

The situation is more contentious where a preliminary inclination is notified in

the context of a significant evidentiary record. Here the question is whether this
is prejudging or truly giving the parties the fullest opportunity by pointing them
to the things that have to be done to win. To a tribunal brave enough to change
its mind, it should be the latter, but a challenge might ask what is the reason for
indicating the inclination rather than the tribunal merely saying that it is
particularly in need of more guidance on this issue. Inclinations can usually be
framed as more neutral two-way questions. For example, the tribunal might ask,
‘should it take the contextual view on the evidence that … or conversely should
we adopt a literal interpretation because … ?’ A tribunal can also avoid giving a
preliminary inclination by simply drawing the parties' attention to the parts of
their argument where the tribunal is as yet unconvinced either way and where it
would benefit from further responses to the factors that are interfering with the
tribunal’s ready acceptance of the argument as put by either side. Examples of
questions of this nature might include: ‘How could we avoid the following
unintended consequences were we to accept your broad proposition?’ ‘How does
your argument as to the weight of authority sit with the opposing authorities
presented by counsel for the other side?’

Regardless of these conceptual views, how effective proactive arbitration may be

can in part depend on the attitude of the parties and hence Raeschke-Kessler's
observation in favour of agreement by the parties to this approach is a sensible
practical suggestion. If the parties are both cooperative, most would agree that it
is an ideal form of interaction. Conversely, if one party is obstructive, looking for
grounds to challenge or delay, an arbitrator may be constrained for fear that
observations will be misconstrued and improperly utilised in subsequent
challenges. However, an arbitrator cannot simply be dissuaded from an optimal
procedure as a result of guerrilla tactics by one party alone.

page "407"

6.2.8.9. Proactivity and Fraud and Corruption

As to the duty to investigate and/or disclose fraud or corruption, the situation

may vary depending on the nature of the criminality and who is responsible, but
for the purposes of the discussion, let us assume that it is a criminal activity that
relates directly to the arbitration itself. Some take the view that there is no such
obligation. For example, a Working Group of the ICC considered such disclosure
to be contrary to the nature of arbitration and the trust that parties place in an
arbitrator. (65) Redfern and Hunter rightly criticise this observation if it was also
thought to apply where an arbitrator observes clear criminal behaviour. A clear
example would be where one arbitrator knows that a bribe has been accepted by
a co-arbitrator. It would be nonsensical to suggest that there is no duty of
disclosure. Such a bribe undermines the very nature of the arbitral agreement
itself.

Nevertheless, it is difficult to determine whether a tribunal has an overriding

duty in terms of investigating and reporting on fraudulent or criminal activity. A
range of permutations make the issue particularly difficult. Such a duty of
disclosure might arise under a particular domestic statute that can properly be
seen to cover the arbitrator. In complying with such a statute, the arbitrator
would be acting in a personal capacity and is not doing so as part of the inherent
status as arbitrator. One situation is where the tribunal has suspicions about the
underlying transaction but neither party wishes the tribunal to explore this
issue. In ICC Case No 1110 of 1963, the arbitrator, Judge Lagergren was advised
that there had been bribery but was asked by the parties to ignore it. He refused
to do so. (66) A tribunal may also need to be alert to misuse of arbitral proceedings
for ulterior purposes. An example would be money laundering where there is an
award by consent. (67) Where an arbitrator concludes that an arbitration is being
conducted for an illegitimate purpose, Born suggests that after ascertaining the
truth of this, appropriate steps should be taken, including resignation of the
mandate or dismissing the arbitration sua sponte after hearing the parties. (68) He
does not indicate which of the two would be more appropriate.

Another situation is where one party makes allegations against the other. Here
there is a problem in determining what evidentiary standard to apply if there is a
duty to consider the allegation. Allied to that is the possibility that the person
making the allegation did so dishonestly and with a view to gaining strategic
advantages. A further issue relates to a duty of enforceability where some aspect
of the illegality would offend against international public policy. If the allegation
was included in the claims or cross-claims and adjudicated upon, then the award
page "408" should be upheld. A related issue is whether foreign corrupt practices
legislation constitutes a mandatory substantive norm. (69) A final issue is the
relationship between the powers and rights of an arbitrator vis-à-vis the
arbitration itself and their entitlement under their contract with the parties,
including the entitlement to resign in appropriate circumstances.

While strongly advocating consideration of illegality by tribunals, Cremades and

Cairns suggest the following guidelines:

1. Ensure due process, including making those suspected fully informed of the
suspicions and the basis of the allegations.
2. Appropriately place the burden of proof, to which should be added apply the
appropriate standards of proof.
3. Do not examine other contracts outside of the ambit of the arbitration
agreement.
4. Do not permit tactical abuse.
5. Provide for confidentiality.

The authors suggest that there would be strong public policy objections to
legislation mandating disclosure by arbitrators. They suggest that notification
might also be compromising the right to a fair and independent determination of
civil rights under Article 6 of the European Convention on Human Rights
although that seems unlikely if the authority before whom notice is given then
conducts an appropriate review and analysis. The issue needs to be further
discussed and analysed, however, given the great variety of national statutes and
rules and the fact that in many jurisdictions, the rules of professional secrecy
apply to members of the bar but not to arbitrators as such. The authors also refer
to treatment of legal privileges, but do not indicate exactly when these should be
respected or ignored. (70)

The geographical location of the alleged fraud may also be relevant. For example,
notification to authorities of offences within the Seat might be considered to be
an implied part of an arbitrator's mandate, ensuring that the Seat can promote
honest arbitral practices. The matter becomes further complicated when the
offence has not been committed at the seat of the arbitration.

6.2.8.10. Proactivity and Matters not Addressed by Either Party

Another issue is whether a tribunal should raise matters not addressed by either
of the parties. In all cases, the parties should be informed of the tribunal's
concern, but there is still an overriding question as to whether and in what
circumstances this may occur. It is suggested that an arbitrator should be
proactive if there is a hint of fraud or illegality not raised by either party, (71) but
not where waivable claims such page "409" as a statute of limitations is involved.
Historically, commentators often looked at this question from a comparative law
perspective by analogy from their system's preferred approach to judicial
behaviour. The fully adversarial common law saw judges historically taking a
back seat as compared to inquisitorial civilian judges. Those extremes were
never clear-cut and are no longer true in litigation in any system. In any event,
they are not determinative of the due process elements where an arbitrator's
duties are concerned. Even common law courts have supported such a duty. As
one judge observed:

If an arbitrator considers that the parties or their experts have missed the real
point … then it is not only a matter of obvious prudence, butt the arbitrator is
obliged, in common fairness or, as it is sometimes described, as a matter of
natural justice, to put the point to them so that they have an opportunity of
dealing with it … (72)

A tribunal should not make a decision on facts which the parties do not presume
are in dispute. For example, in Pacol Ltd an award was set aside when a tribunal
held the respondent was not liable even though it had admitted liability. (73) The
situation is different if a tribunal simply draws a different inference to that of the
parties from evidence presented to it. Petrochilos suggests that the extent to
which a tribunal may depart from party's submissions is a question of degree. (74)

6.2.8.11. Proactivity and Applicable Law

In addition to the general question as to the degree of proactivity to be preferred,

there are also controversial aspects in relation to which matters should or
should not be addressed. A tribunal might consider directing the parties in
relation to factual and/or legal questions. Where the latter are concerned, there
is a distinct question as to whether the tribunal should seek to point the parties
towards what it believes to be the relevant legal system and legal principles that
are applicable. This relates to another contentious issue as to whether a tribunal
has a duty to know the law or should rely on contentions of the parties. This is
discussed in section 13.19. Proactivity as to applicable law is not controversial
when the tribunal is merely exercising its discretion to identify applicable
substantive law absent a determination by the parties. It has practical
significance, however, where the parties argue in relation to a particular law that
the tribunal may feel to be inappropriate or not comprehensive.

page "410"

An all too common unfortunate example is where a party in an international sale

of goods dispute, argues under domestic contract principles when the
transaction is automatically subject to the Vienna Convention on the
International Sale of Goods. (75) There is simply no consistency in approach when
viewing reported cases, although these tend to primarily be court-based
decisions. Some consider the behaviour of the parties to constitute a waiver
under procedural norms. Some consider the behaviour to be an implied
exclusion of the CISG based on its own provisions to that effect. Some take the
contrary view, arguing that waiver or exclusion must be conscious activities and
not responses from the ignorance of counsel, which seems the better view.
Furthermore, some would take the view that even if the claimant's behaviour
constituted waiver or an intent to exclude, a respondent that merely answers
allegations made is not presenting any view as to its preferred legal position and
hence cannot consummate any agreement as to joint waiver. The only sensible
solution for arbitrators consistent with due process norms is to warn the parties
that an inappropriate law is being argued and allow them to either correct their
mistake or come to a new agreement as to applicable law.

6.2.8.12. Proactivity and Guerrilla Tactics

The arbitrator's obligation to afford a full or at least adequate opportunity to

present a case and concerns to minimise the opportunity for procedural
challenges can at times play into the hands of disreputable disputants and
encourage what are sometimes described as guerrilla tactics. It is natural that a
party with a weak case will consider the financial advantage of delaying tactics.
This may depend on likely attitudes to the award of interest but can be
influenced by cash flow considerations if a party simply does not have the funds
to pay an adverse award. In such circumstances a respondent might look for a
myriad of defences and counterclaims, urge maximum document production and
seek a number of stages in written submissions and oral hearings. (76) A claimant
with a weak case may wish it to drag on in the hope that some nuisance value
settlement offer will be made. Even claimants with meritorious cases may seek
to strategically delay if it is felt that lengthy and expensive proceedings will be an
insurmountable burden on respondent.

In addition, a party simply may not involve itself in the proceedings, but this
poses fewer problems in the context of this section. Where the claimant refuses
to engage in the process, it will eventually be dismissed. Where the respondent
refuses to be involved, the tribunal must proceed in any event.

page "411"

Section 2.10.2 indicated that there is a right to cooperation from the parties and
a duty upon them to conduct themselves in good faith and professionally.
Nevertheless, an arbitrator needs to consider how to approach the situation
where such behaviour is not forthcoming. Obviously, no rigid blueprint can be
provided and this is a particular case where judgment must be exercised on a
case-by-case basis. Nevertheless, a few observations may be appropriate at a
more general level. A key problem in determining how tribunals ought to
respond to guerrilla or dilatory tactics is to note that these cannot easily be
defined and hence identified. Different legal families have different attitudes to
how robustly counsel may seek to manipulate procedural elements to gain
tactical advantages. In addition, most guerrilla tactics will be dressed up in the
guise of fundamental due process requests. They invariably have a tenable form,
but on careful reflection, an unmeritorious substance. That very evaluation as to
a lack of substance must be made in the context of mandatory due process
norms.

Dilatory guerrilla tactics take a number of forms. Proper responses will vary in
each case. In chronological order these may include:

– challenges to jurisdiction;
– applications for anti-arbitration injunctions;
– repeated challenges to arbitrator appointments;
– repeated challenges to procedural determinations and deadlines;
– failure to meet deadlines and/or repeated requests for extensions;
– withholding of key evidence;
– excessive document production requests and challenges;
– attempting to submit evidence after deadlines;
– demands for hearing adjournments and rescheduling of witnesses;
– annulment challenges and similar court applications;
– threats to block enforcement unless certain procedural steps are taken.

In all instances during the currency of the arbitral proceedings, a tribunal will
aim to be firm but fair, ensuring that the matter proceeds in a timely manner, but
preventing any valid assertion that allegedly delaying tactics were not in each
case considered on their merits.

The primary response from an arbitrator should be to attempt to ensure balance

in the exercise of discretions on the assumption that all parties are operating in
good faith. Separate attention can then be given to situations where one party's
tactics seem so outrageous as to not conform to this ideal. In such circumstances,
meritorious applications should still be accepted by the tribunal. Unmeritorious
ones should be rejected. Judicious use of costs discretions can assist, although
this will be less likely to have an impact when the total amounts in dispute far
outweigh any permissible cost penalties. Because this is seen as a growing
problem in international arbitration, another avenue is to promote commonly
accepted ethical standards against which to assess the behaviour of parties and
their counsel. This has not as yet been developed and consent to best practice
norms would be difficult to agree upon.

In some cases guerrilla tactics may emanate from one of the party-appointed
arbitrators. Paradigm examples include co-arbitrators who will refuse to sign
page "412" awards or who threaten resignation at the eleventh hour. Section
16.6.3 deals with the role of the chair in ensuring fair and efficient responses to
such circumstances. Section 16.7 deals with the powers of the majority to
proceed in the absence of collaboration by such an arbitrator.
6.2.9. Proactive Institutions

There is similarly a growing desire for institutions to be proactive to aid

efficiency and timely resolution of disputes, although here there are
philosophical and marketing differences between those of a more controlling
disposition and those offering a ‘light touch’ as an alternative model. One
advantage of greater control by institutions is that this shelters the tribunal itself
from needing to take such action in the first instance, which at times can lead to
party assertions of bias. Even though the same erroneous assertions may be
made against an institution, the latter is not doing so in the context of reviewing
the claims and submissions along the way to a binding determination.
Furthermore a strict procedural ruling by an institution can be supplemented by
a contrary application to the tribunal itself.

6.3. Time Limits

Where time limits are concerned, there is a distinction between, on the one hand,
time limits set either by the lex arbitri, by party agreement, or by the arbitral
rules and on the other hand, time limits set under broad discretions of the
tribunal. The first category entails proper compliance with the procedural
framework of the mandate. The second category involves integration of due
process and efficiency concerns from a discretionary perspective. The two
categories obviously are related as discretions are circumscribed by laws and
rules.

Where strict time limits are set by the lex arbitri or rules they should obviously
be followed. Even then it will usually be the case that extensions are permissible.
(77) Some arbitral rules set out elaborate principles for dealing with time limits.
(78) Some institutions impose time limits on the rendering of an award. For

example, the ICC Rules require the tribunal to render an award within six
months of signing the Terms of Reference or such later period as may be allowed
by the ICC Court of page "413" Arbitration. (79) An ICC tribunal must also advise
the ICC Secretariat on the close of proceedings as to when it expects to submit a
draft award. (80) The CIETAC Rules impose a six-month time limit for the
rendering of an award subject to the right to obtain an extension from the
CIETAC Commission. (81) A number of European systems also provide for time
limits in their lex arbitri as to when an award must be rendered. (82) Time limits
are also sometimes imposed for particular steps within the arbitral process such
as Answers, challenges to jurisdiction, new claims and terms of reference.

Various arbitral rules more generally require an arbitrator to deal with the
matter expeditiously. For example, Singapore cases are to be heard within one
year or the Registrar must explain why. Some lex arbitri would expressly allow
for removal of an arbitrator in circumstances of excessive delay. (83) Others leave
it to more general grounds such as Article 14(1) of the UNCITRAL Model Law.
UNCITRAL Model Law Article 14 provides for termination of the mandate if the
arbitrator ‘fails to act without undue delay’. Article 25(1) of the ICC Rules 2012
indicates that ‘the arbitrator shall proceed within as short a time as possible to
establish the facts of the case by all appropriate means’. The ICC Rules also fix a
time limit for rendering of the award at six months unless extended by the ICC
Court. (84) WIPO Arbitration Rules Article 38 requires that ‘the tribunal shall
ensure that the arbitral procedure takes place with due expedition’. Rule 9.2 of
the CPR Rules for Non-Administered Arbitration of International Disputes states:

The proceedings shall be conducted in an expeditious manner. The Tribunal is

empowered to impose time limits it considers reasonable on each stage of the
proceeding, including without limitation the time allotted to each party for
presentation of its case and for rebuttal. In setting time limits, the Tribunal
should bear in mind its obligation to manage the proceedings firmly in order to
complete proceedings as economically and expeditiously as possible.

IBA Rules of Ethics, Article 2(3) indicates that:

A prospective arbitrator shall accept an appointment only if he is able to give to

the arbitration the time and attention which the parties are reasonably entitled
to expect.

Rule 7 of the IBA Rules of Ethics indicates that ‘(a)ll arbitrators should devote
such time and attention as the parties may reasonably require having regard to
all circumstances of the case …’. The introductory note to the Rules of Ethics
indicates that arbitrators should be diligent. The duty of efficiency as to cost is
page "414" enshrined in Rule 7 of the IBA Rules of Ethics indicating that
arbitrators ‘shall do their best to conduct the arbitration in such a manner that
costs do not rise to an unreasonable proportion of the interests at stake.’ This
could be described as a separate duty to constantly consider the costs in the
context of the amount in dispute. (85)

The English Arbitration Act 1996 opens with the statement that ‘the object of
arbitration is to obtain a fair resolution of disputes by an impartial tribunal
without unnecessary delay or expense’. (86) Section 33(1)(b) of the Act requires a
tribunal to ‘adopt procedures suitable to the circumstances of the particular case,
avoiding unnecessary delay or expense, so as to provide a fair means to the
resolution of the matters falling to be determined.’ The norm is mandatory as
section 33(2) states:

The tribunal shall comply with that general duty in conducting the arbitral
proceedings, in its decisions on matters of procedure and evidence and in the
exercise of all other powers conferred on it. (emphasis added)

Some rules only provide time limits for expedited procedures. (87) Nevertheless,
in EC Ernst Inc v. Manhattan Constr Co of Texas, (88) a US court considered that a
duty to make a reasonably expeditious decision was implied where not
expressed. An English court in Norjarl considered that arbitrators owe a duty of
due diligence requiring them to proceed with the reference with reasonable
dispatch. (89)

There are advantages and disadvantages of express time limits as against general
expediency obligations. Those in favour argue that such time limits place strong
pressure on arbitrators to ensure compliance. Express time limits are clear and it
is easy to see when they have or have not been complied with. Under a consent-
based paradigm, parties are naturally free to set time limits for arbitral
proceedings, so selection of rules to that effect are a corollary of autonomy. From
an efficiency perspective, there may be much to commend in terms of setting
tight time limits. The disadvantage is that they do not take account of the
circumstances of the particular case. Even if complied with, too rigid a timeline
could mean that inadequate attention has been given to complex matters.

An arbitrator who fails to meet obligations of diligence and expediency might be

removed (90) and in some circumstances may even be liable for damages. In many
legal systems, failure to comply with a time limit for rendering of an award will
also allow for challenge of an award subsequently rendered. (91) This will arise
where page "415" strict time limits can be argued to be part of a tribunal's
express jurisdictional mandate. A failure to meet those timelines might not only
render the award invalid, but should also render the tribunal functusofficio. A
roundabout way to come to the same conclusion occurred where a Paris Court of
Appeal considered that a failure to comply with a time limit was a violation of
international public policy. (92) Holtzmann argues that arbitrators certainly have a
responsibility over the pace of proceedings and ‘an activist attitude is the most
important contribution an arbitrator can make to overcoming delays.’ (93) In some
circumstances this will be controversial, particularly if the parties are denied an
opportunity to present what the tribunal feels are irrelevant arguments.

In addition, a number of issues arise if the time limits are unlikely to be met or
have not been met. Because all rule systems must allow for the possibility of
extension in appropriate circumstances, (94) the time limits may be honoured in
the breach and might mislead the parties as to the likely completion date. It
would be very difficult for an institution to refuse to extend time limits
applicable to an arbitrator who simply indicates that he or she cannot meet the
deadline. There are no natural punitive or enforcement measures except in the
most extreme circumstances where removal is possible. Even that does not
guarantee timely completion and comes with some degree of inevitable
duplication. An arbitrator who feels unable to meet such a timeline, perhaps
because it was unreasonably short from the outset, might invite the parties to
agree on an extension, but cannot compel them to do so. In most such
circumstances, an arbitrator could still structure a procedural timeframe to meet
the deadline, warn the parties that this may necessitate a compressed hearing
and necessitates an award with minimal reasoning and warn that this is the
corollary of the timeframe they have selected. That might be dangerous in a
jurisdiction that might block enforcement on the basis of inadequate reasoning.
(95)

Because an arbitrator must be concerned with fundamental principles of due

process, including the need to give parties an adequate opportunity to be heard,
in an extreme case where parties seek to impose unduly restrictive time frames,
an arbitrator should simply decline to accept the appointment, explaining to the
parties the reason and again inviting them to revise their determination. Some
perhaps unlikely situations would pose casuistic annoying challenges even if
page "416" they would ultimately be rejected. For example, in an ICC arbitration
with an express six-month time limit, could a respondent argue that it only
agreed to arbitration under the ICC Rules and only agreed to be bound by an
award which met the time limits? No doubt the ICC Court would grant an
appropriate extension to avoid spurious arguments, but circumstances could be
envisaged where parties truly selected expeditious rules for sound commercial
reasons, such as in the case of an almost insolvent company with major cash flow
concerns. (96)

The situation is even more problematic if time limits are contained in the
arbitration agreement itself. At the extreme, a party might again argue that the
agreement to be subject to a binding award was conditional on it meeting the
deadlines. Conversely, depending on the terminology, such time limits might not
necessarily be seen as crucial, but could instead be seen as aspirational goals
where failure to comply does not undermine the award. This is a question of
construction of the arbitration agreement. The argument would be that the
agreement to arbitrate is unconditional and the stipulation as to time is a best
endeavour stipulation in that regard. Nevertheless, Poudret and Besson note
conflicting views by national courts. (97) They distinguish the case where the
tribunal has not met the deadline for an award, with the situation where there is
a time limit for initiation of the arbitration. They do not indicate whether the
principle ought to be different. From a consent paradigm, it ought to be at least
arguable that a party that only agrees to arbitrate if a case is brought within a
time limit, is also only agreeing to be bound by a decision if the time frame was
met. If the parties have expressed a range of stipulations within an arbitral
mandate it is problematic to second guess some parts of their stipulations,
holding them to be something less than conditions precedent to jurisdiction. The
converse argument would be that the parties have clearly expressed a wish to
arbitrate and if the time limit on the award took away power, there may be a
vacuum in terms of dispute settlement or costly debates as to whether litigation
could then ensue. In such circumstances, it might be open to conclude that their
implied intent was to not treat the time indication as a condition precedent to
jurisdiction. Because of such competing arguments, and because general
principles cannot be determinative of the way to interpret the intent behind all
articulations of arbitration agreements, best practice would suggest that the
parties are very clear as to the intent behind any designated time limits, (98) and
the consequences they envisage if the time limits are not met. Where there is
page "417" insufficient clarity, a tribunal should look at all relevant
circumstances to aid in understanding the likely intent. (99)

If a time limit is imposed and an application is made to remove an arbitrator for

failing to comply with the time limit, a problem arises in terms of the powers of a
truncated or replacement tribunal to subsequently render an award. If there is
no power to extend, how can a truncated tribunal or a replacement arbitrator
meet the deadline that has expired? Some lex arbitri allow for extensions in such
circumstances. The functus officio problem has led Belgian law to adopt a more
practical solution to the effect that after six months has elapsed, the parties may
apply to the court to set a time limit for rendering of the award. (100) Section
50(4) of the English Arbitration Act 1996 allows a court to extend the time even
if the time previously fixed has expired. Other lex arbitri seem to have gaps in
that regard. This has led Poudret and Besson to rightly observe that if the parties
wish to subject an arbitration to contractual time limits, they should ensure that
they do so within a framework where extensions may be granted. (101) These
concerns were the main reason why the UNCITRAL Model Law chose not to
impose time limits. (102) The Model Law drafters considered that time limits
would prevent flexibility and unduly constrain arbitrators. (103)

Where tribunal imposed time limits are concerned, decisions as to time limits
and order of proceedings can often involve other conceptual challenges. For
example, simultaneous filing of written submissions can ensure shortened
proceedings but may be argued by some respondents to be unfair. While
simultaneous submission appears superficially equal in terms of treatment, a
claimant can take as long as it likes to prepare a case within any applicable
statute of limitations and knows the essence of the claim it wishes to bring.
Conversely, a respondent receiving a notice of arbitration is often only then
aware of the rudimentary gist of the claim and might reasonably want to read
and synthesise claimant's written submissions before drafting responses. While
party agreement can always resolve this, it is often the case that a respondent
might tactically wish to slow down proceedings in any event in the hope that this
will encourage more generous settlement concessions by a deserving claimant.
Where discretionary time limits are concerned, supervisory courts would
require time limits to be clear as to what is required and to be reasonable in the
circumstances. (104)

page "418"

Another time control mechanism that is conceptually challenging is the framing

of time limits for hearings. Does it flow automatically that each party should be
given identical time or should the time allowed be in proportion to the matters
they have raised? If the latter, would that be an undesirable incentive for parties
to raise a multiplicity of issues in order to gain extra time? In this scenario, a
recent development is what is known as chess clock arbitration or the
Böckstiegel method, in honour of the leading arbitrator who has done most to
inspire its utilisation. This involves setting an overall time limit with equal rights
to each party, but indicating that the parties may themselves choose how to best
use their own allotted time. This is discussed in section 9.6.

6.3.1. Calculation of Time Limits

If the lex arbitri or arbitral rules set out how time is to be measured, these
principles should of course apply. Provisions in the rules have to be looked at
alongside procedural orders where it is the tribunal's time limits that are in
issue. In determining whether time requirements have been met, where the
tribunal has drafted requirements with arguable ambiguity, the benefit of the
doubt ought to be given to the parties. (105) Examples include when to calculate
starting and finishing dates and whether days only include working days.

6.3.2. Enforcement of Time Limits, Extensions and Missed Deadlines

An important issue with time limits is what enforcement powers a tribunal may
have or what other ramifications there can properly be when a party fails to
meet a deadline. (106) This involves two issues. Given the obligation to provide
each party with a full or reasonable opportunity to present its case, can the strict
application of time limits be permitted and if so, in what circumstances? The
second and related issue is whether there are to be any express provisions
indicating what responses a tribunal may make and whether the matter is
ultimately a discretionary one, albeit subject to the due process norms alluded
to.

Where a tribunal has a power to extend a deadline, two key factors are the
reason for the delay and the impact on the parties as to a fair and efficient
outcome from either rejection or acceptance of the extension application. In
some cases, the page "419" tribunal may even have some conflict in deciding on
such a question where counsel asserts that it was unable to meet the deadline
because it was unreasonably short in the first place, or where rescheduling
causes most inconvenience to the arbitrator. A typical example of a troubling
request is if a tribunal sets a final deadline for submission of evidence and a
party then seeks to submit evidence at a later date, arguing that a failure to allow
this would be a denial of a reasonable opportunity to present a case. In response,
a tribunal can legitimately note that the key word is ‘opportunity’ in Article 18 of
the UNCITRAL Model Law and provisions to similar effect. As long as the parties
have had adequate warning and a sufficient time for preparation, the fact that a
party has failed to avail itself of the opportunity through its own fault should not
ground an allegation that the tribunal has failed to afford it the appropriate
opportunity.

Less clear is the situation where a party can legitimately argue that it only
became aware of the new evidence after the deadline and it had no reasonable
ability to know of the existence of the evidence beforehand. One might readily
expect differences in view on this issue, with some arbitrators disposed to allow
the relevant evidence, while others might argue that all disputes have to be dealt
with within a reasonable timeframe and such problems are an inevitable by-
product of this need. The debate can be circular, as the power to set time limits
invariably includes the power to extend such limits. (107) If there is a duty to deal
reasonably with applications to extend, then blanket rulings against such
extensions would be problematic.

Even where the applicant is at fault, when considering the consequences of a

failure to meet time limits, a tribunal will inevitably accept certain breaches as
there will be little in the way of adverse consequences for the other party or for
the sanctity of the proceedings as a whole. Examples would be very short delays
in filing of submissions, documents or witness statements. Conversely, a late
submission of a new claim, counterclaim or witness statement which would
require an impending hearing to be postponed can more legitimately be rejected.
More difficult situations can be envisaged where it is harder to assess the true
reason for the delay. Consider for example a party who at the eleventh hour
before oral hearings asks for an adjournment because key witnesses are now
unavailable. Circumstances may also vary as to whether the tribunal is
suspicious or not as to the veracity of the assertion.

Where due process challenges are concerned, there is always the threat of
annulment proceedings or defences to enforcement. If a tribunal need only
concern itself with the behaviour of a reasonable annulment or enforcement
court, most issues do not need to be separately considered from the perspective
of the duty to render an enforceable award. The tribunal need only follow the
applicable procedural rules and the concerns of a reasonable enforcement court
will have page "420" been met. On this basis, tribunals ought to be confident that
strict but reasonable time frames to try and conclude the case efficiently and
expeditiously may safely be utilised. That is further supported by rules requiring
the tribunal to have regard to complexity and value when seeking an expeditious
and cost-effective procedure. Nevertheless, while it is easy to state such
principles, it is somewhat more difficult to apply them in practice, particularly
when faced with persistent threats from aggressive counsel.

6.4. Expedited or Fast-Track Arbitration

Arbitration has traditionally been viewed as a relatively fast and cheap method
of dispute resolution. However, an arbitration of typical complexity (i.e., one
which is not overly complicated) may commonly run for two or three years from
the time of the original notice to the provision of a final award. Although this may
seem acceptable to practitioners frequently involved in arbitration, business
people may be surprised at the anticipated length of proceedings (and may elect
to settle instead). Recent surveys have indicated that corporate counsel would
generally hope that a dispute of moderate complexity in the USD 5 million to USD
10 million range should take less than a year, while many business persons felt
that two to three months would be even more appropriate. (108) Furthermore,
there tends not to be a method of summary determination of simple disputes in
arbitration that is often available from courts, meaning that a straightforward
dispute is dealt with according to the same procedural framework as one that is
far more intricate.

An attempt to respond to these issues is the development of expedited or ‘fast-

track’ arbitration. In a fast-track arbitration the dispute is dealt with in an
abridged timeframe with modified procedural rules that are designed to result in
an award being rendered within weeks or months rather than years. The
procedure is particularly useful ‘when ongoing commercial relationships are
concerned’. (109) Arbitral proceedings may be expedited in two broad ways. The
first is simply through proactive time management by the tribunal. This will
involve ensuring tribunal and counsel availability, identifying truly necessary
procedures and not presuming that time-consuming processes such as document
production, written witness statements or dual rounds of written submissions
should apply as a matter of course, constraining documents to truly material
matters and generally responding to dilatory tactics. This was discussed
generally in section 6.2 and throughout the book. This section is instead dealing
with the alternate approach to expedited page "421" proceedings, being the use
of specialised rules that seek to direct a faster and more cost-effective process.
There are several principal potential difficulties with this form of fast-track
arbitration. First, not all disputes are suited to fast-track proceedings. In
particular, disputes requiring significant technical or expert input, or which
necessarily involve significant document production or discovery, will not fit the
fast-track mould even if the amounts in dispute are low. Second, completion of
the arbitration in a shorter period than usual requires some compromise in
relation to the scope of submissions, responses to submissions and presentation
of witnesses, experts and other evidence. This raises due process challenges and
the conceptual question of proportionality as an element of procedural justice.
Third, and practically perhaps most importantly, for fast-track proceedings to
achieve their goal, the respondent and claimant generally need to be cooperating
with the tribunal to meet shortened deadlines. Effectively, such provisions work
best when both parties desire a swift outcome and avoid dilatory tactics.

6.4.1. Consent Aspects

Parties may consent to fast-track procedures in their arbitration agreement or at

the time of the dispute. Agreement at the time of the dispute has the benefit that
both parties can be expected to cooperate with the tribunal's stipulated time
frames, although that is less likely as at the time of the dispute, one party will
often have a vested interest in delay. Conversely, agreeing in advance of a
dispute may be problematic if the rules selected are not fit for the dispute that
emerges. A compromise is to stipulate that expedited procedures apply in certain
circumstances, for example, if the amount in dispute is below a specified figure.
That itself can be problematic as complexity is not necessarily proportional to
the amount involved although the proper cost and effort from a commercial
perspective may well be looked at in that context. A further problem in using a
stipulated claims figure as the determinant of whether normal or expedited rules
are to be used means that the claimant has an effective choice depending on the
level at which it makes its claim. While there is a disincentive to inflate claims
due to some institutional rules indicating that fees will be higher as a result, the
possibility that there may be tactical misuse in some circumstances at least
remains.

In any event, consent is the key. Express fast-track arbitration is an option that
the parties must choose to take up. One way of doing so is by specifying the fast-
track or expedited procedures published by various arbitration institutions. (110)
Alternatively, parties may request the tribunal or institution to set up an
expedited page "422" timeframe, (111) while still retaining the general framework
set out in the normal rules. (112)

6.4.2. Institutional Approaches to Accelerated Proceedings

While this section concentrates on comprehensive expedited rules, in some cases

timing is dealt with by mandated duties of efficiency and expediency and/or by
expedited procedures for particular stages, such as empanelment of a tribunal.
(113) Where more comprehensive rules are concerned, the key models include

those promulgated by such regimes as the Swiss Rules, the Singapore

International Arbitration Centre, the Hong Kong International Arbitration Centre
and the International Institute for Conflict Prevention & Resolution (CPR). Some
features of these regimes are set out below. If rules designate expedited
procedures in this way, a question may then arise as to whether the parties can
override that stipulation. That would depend on whether those provisions were
seen as mandatory. As discussed in section 3.9.2, while the parties can agree on
any rules they like, they cannot impose a modification on an institution itself
without that institution's consent. Institutional views may vary, hence
clarification needs to be sought from the relevant institution where such a
modification is proposed.

The Swiss Rules of International Arbitration 2012 require that any dispute
intended to be governed by the Swiss Rules and which has a value of less than
one million Swiss francs must be dealt with under its ‘Expedited Procedure’. (114)
This provides for shortened time limits, limits on the number of submissions and
responses (one statement of claim, one statement of defence and one defence to
counterclaim), a single oral hearing for all submissions, witnesses and experts, a
time limit for the entire proceeding (from transmission of the file to the tribunal
until delivery of a final award) of six months. The award may be given with
reasons set out in a summary form only or parties may agree that no reasons
need be given. In such cases the tribunal will consist of only one arbitrator unless
the parties have agreed to a greater number. Where the arbitration agreement
calls for three arbitrators, the parties will be encouraged to accept a single
arbitrator. (115)

The SIAC rules provide that in certain circumstances, including where the
amount in dispute does not exceed SGD 5,000,000 (as compared with, for
example, a threshold of USD 250,000 applied by the HKIAC Rules), an expedited
procedure will apply where it is requested by just one of the parties (i.e., party
consent is not necessarily required) and is agreed to by the Chairman. (116) The
HKIAC rules page "423" provide for expedited arbitration by documentary
hearings only, unless the arbitrators consider that oral hearings are necessary.
(117) The SIAC Rules take the opposite position: oral hearings and evidence-taking

will occur unless the parties agree to a documents-only process. (118) The CPR
Rules set out a comprehensive regime that is independent of its normal
arbitration rules, rather than including provisions that adapt the standard rules
as is done by most institutions. The Rules provide as a general context that the
proceedings provide the parties with a ‘reasonable, but not exhaustive’
opportunity for each of the parties to present its case. (119) The rules allow for the
parties to set a deadline for finalisation of the award. (120) The Rules provide the
Tribunal with a broad mandate to manage the proceedings in order to expedite
them; it may limit evidence given and time for testimony and will ‘control the
proceedings as is necessary in the discretion of the Arbitral Tribunal to arrive at
a speedy, just Award’. (121) The tribunal may award interim measures simply on
the application of a party, (122) and includes specific provisions for such aspects as
disclosure of documents. The tribunal also has broad discretion in shaping the
form and scope of oral hearings, ‘including the application of appropriate
standards for evidence’. (123) The ICC, in contrast with this comprehensive
approach, sets out in Article 32 simply a mechanism for shortening time limits
otherwise stipulated in the ICC Rules.
6.4.3. Due Process

As noted above, the fast-track proceedings will provide for shortened time
frames for submission of arguments and evidence. However, in narrowing the
window in which parties must submit their cases, the tribunal must be careful to
ensure that each party continues to have an opportunity, and an equal and
adequate opportunity, to present its case and evidence. A problem can arise with
limitations on the number of submissions. Due process requires that a party has
an opportunity to respond to the submissions of another party, especially where
those submissions include expert evidence. (124) Failure to do so may result in the
award being page "424" challenged on due process grounds. To minimise this
risk, ‘the setting of strict deadlines for responding to submissions appears to be a
much safer approach than denying the opportunity to comment at all’. (125)
However, parties which have chosen fast-track arbitration should expect
limitations on their opportunities to present their arguments, both in terms of
the number of occasions on which they can present new arguments, and the
length of those arguments. The proceedings may be held without oral
submissions, or they may be held without experts, or parties may be prohibited
from cross-examining witnesses. At times it might be suggested that to reduce
the time spent by the arbitrators in the preparation of an award, the parties
should agree that the award be provided without reasons. When considering this
possibility, arbitrators should check whether an award given without reasons
would conflict with any due process requirements in relevant enforcement
countries and whether waiver would apply in such circumstances.

Some complicated disputes are so intricate that they cannot be resolved by a

fast-track method even where selected. Where the parties have chosen fast-track
arbitration, the clause may be unworkable as it is not possible to implement it in
accordance with the parties' consent. However, some arbitral rules may contain
a ‘safety-valve’ provision allowing a tribunal to revert to normal procedures
where it determines that a dispute is not, or is no longer, capable of resolution
using fast-track arbitration. For example, Article 13 of the HKIAC Short-Form
Arbitration Rules provides that:

If, during the course of the arbitration, the Arbitrator concludes that the dispute
is incapable of proper resolution in accordance with Articles 3–12 above, he shall
conduct the arbitration in accordance with the Hong Kong international
Arbitration Centre's Domestic Arbitration Rules save insofar as these have
already been complied with. The arbitration shall unless otherwise directed by
the Arbitrator, continue from the point already reached.

6.4.4. Practical Considerations

Parties and arbitrators can take some steps to try to maximise the likelihood that
an intended fast-track arbitration yields a result quickly. Parties might try to
choose an arbitrator with technical knowledge of the context of the dispute, as
the parties will not have the same opportunity to educate the arbitrator that they
would have in a normal proceeding. On the one hand, they should also seek an
arbitrator with the legal skills to quickly understand legal intricacies. On the
other hand, such arbitrators are likely to be in high demand, and the parties
would do well to try to find arbitrators (and counsel) who are able to devote a
significant amount of their time to the dispute. This will enable the various
participants to make themselves available to meet tight time constraints. Parties
may be well-advised to consider the page "425" availability of an arbitrator as a
primary criterion in his or her selection, and ‘nominees not convinced that they
can comply with the shortened time limits should decline the nomination’. (126)
This is consistent with the view propounded in section 6.2 that proactivity
generally requires willing and able arbitrators with sufficient time and interest
in coming to grips with a case at the earliest available opportunity.

Challenges to arbitrators by the opposing party should be dealt with quickly in

order to accelerate the proceeding toward the hearing stage and also to ensure
that the compressed nature of the hearing is not interpreted by a party as unfair
treatment by a biased arbitrator. Parties should be ready with replacement
arbitrators should their first choices be challenged. (127)

6.5. Utilising Guides to Procedure

It has been suggested throughout that an optimal procedural framework must of

necessity provide a trade-off between fairness and efficiency. Within the concept
of efficiency, there are also inevitable trade-offs between flexibility and certainty.
A significant challenge relates to the degree to which issues should be left to the
broad discretion of arbitrators or whether detailed procedural models should be
developed.

Obviously the most significant regulatory models are those contained within lex
arbitri and arbitral rules and which are hence binding. Nevertheless, these are
often stated in general terms and contain many open-ended discretions. There is
a debate as to whether guides should be promulgated to assist in promoting
optimal use of those discretions. This relates to institutions as well as tribunals,
although institutional practice can be expected to display more consistency in
approach. Where institutions are concerned, not only are the rules important but
also the practices that tend to be adopted. Institutions use the same personnel to
deal with multiple cases. Internal directions and consultation aim to promote
consistency. Institutions can also be expected to tend to more circumscribed
approaches to the utilisation of discretions with a view to consistency and
minimising claims of bias. They will also tend towards a plain meaning
interpretation of their rules for similar reasons.

Where tribunals are concerned, there have also been a number of guides
developed in the hope that these are adopted by parties and used by tribunals as
aids in the exercise of discretions. The value of detailed guides may depend on
one's perspective and the identity of the users. The same arguments can be
raised page "426" about the need for more detail in rules and statutes. Highly
experienced arbitrators are more inclined to see this as a needless constraint,
given their familiarity with the matters that need to be dealt with. Pierre Lalive
has described ‘the modern disease of over regulation’. (128) Complaints are heard
about ‘formalism, over-lawyering, delays and inefficiency …’. (129) Flexibility and
discretion are argued to be the best means to allow the tribunal to meet the
needs of the individual case, both in terms of substantive requirements and the
experiences of the parties. For example, the procedures may vary depending on
whether both parties come from a common law or civil law background or
whether their backgrounds are different. In the latter event a tribunal may opt
for more neutral or hybrid procedures. (130) Conversely, proponents of guides
might argue that if arbitration aims to provide consistent and optimal responses,
well thought out solutions in leading guides ought to be influential. Furthermore,
given the need to expand the pool of arbitrators to include inexperienced
practitioners, notes can be a more helpful form of support, ensuring the kind of
consistency in approach that experienced practitioners have come to display
through their past dealings.

Debates of this nature are in reality debates about the optimal means of
articulating legal norms, including the relative value of rules versus discretions.
William Park addressed more generally the overriding question of rules versus
discretion, concluding that the benefits of discretion and flexibility are overrated.
(131) He argued for more detailed default provisions that set clear parameters,

absent agreement to the contrary by the parties. Park argues further that parties
can opt out of ‘soft law’ standards, hence the standards are not truly
constraining. Park also rightly observes that much of what we see as law
‘involves a continuum between generality and precision’. (132) Some of the
problems he identifies with undue reliance on discretions include failure to meet
the parties' expectations, the problem of a tribunal exercising a discretion when
it will be obvious which side is favoured by the determination, and the problem
of making choices where different legal cultures have different preferences.
Others also advocate greater use of guidelines and rules. (133) Another approach is
to have supplementary opt-in rules of a more elaborate nature. Park notes that in
practice there would be a need for arbitrators to select such rules, as commercial
drafters rarely give sufficient attention to dispute settlement in the page "427"
initial phases, while once the dispute erupts, the parties again are unlikely to
agree on many procedural issues. (134)

While arguing strongly for more detailed and objective a priori procedures, Park
notes some of the implications. Lawyers and arbitrators will need to educate
themselves about the changes. There will be increased transaction costs in
bargaining, although that would be likely to be counterbalanced by savings over
time. More detailed procedures might even encourage annulment or
enforcement challenges as it might be easier to argue that a more specific
procedural rule was breached. In the extreme, disenchantment with the specific
rules might lead to some counsel arguing in favour of litigation over arbitration.
(135)

In the context of the debate, there is a difference between details agreed to by the
parties and details recommended for and adopted by the tribunal. Unduly
detailed procedures agreed to by the parties may prove undue constraints on
tribunal power, (136) although that would not be so with a more general set such
as the IBA Rules of Evidence which still leave much to tribunal discretion. Ad hoc
stipulations by the parties can be problematic if very specific procedures are
agreed upon in the arbitration clause without reference to the particular dispute
that has arisen. (137) This also comes with its own cost, reduces flexibility and may
lead to challenges where the stipulations are not strictly adhered to. Where
procedural rules are sought to be agreed between the parties once a dispute is
known, this would be less problematic, although less likely. It is still far
preferable that the tribunal be included in the establishment of such agreed
procedures either through preliminary conferences, or at least through a request
to approve or amend any draft agreements reached. (138)

The balance of the discussion addresses the possibility of adopting or following

some of the better-known published guides and notes from bodies such as
UNCITRAL, the IBA, the ASA and the ICC. It is within the context of the efforts
made to develop guidelines on organising procedure, rules of evidence and
impartiality, that many commentators have made observations about flexibility
versus certainty. Those comments may apply equally to party adoption or
tribunal adoption of such norms. Common solutions are particularly problematic
if they are suboptimal compromises seeking to be deferential to those features of
each system that negotiators are unwilling to depart from. One advantage with
international arbitration as compared to public international law treaties is that
the harmonisation process is more iterative. Symposia, institutions, statutory
reforms and private drafts are presented with reasonable frequency, seeking to
both find 428 common ground in contentious areas and also confirm common
practices that have been developed by leading experts.

Whether guides or notes are unduly constraining can depend on the way they
are drafted. Commonly, notes and guidelines will indicate that they are not
binding rules and should not be interpreted in any rigid fashion. There is also a
question of how much detail and how controlling the language should be. Care
should certainly be taken to ensure that a guideline is not overly prescriptive and
does not force the tribunal to address issues before it is necessary to do so.
Criticisms of this nature led UNCITRAL to replace its guideline initiative with
Notes on Organising Arbitral Proceedings, adopted in 1996. (139) The earlier
example was the UNCITRAL Guidelines for Preparatory Conferences in Arbitral
Proceedings. In particular, the Notes point out that many of the matters
mentioned do not need to be considered in most arbitrations. Some of those
opposing the initiative were still concerned that UNCITRAL as a promulgating
body has such prestige that arbitrators may feel compelled to respond
automatically. (140) One of the problems with guides to procedure is that less
experienced arbitrators and practitioners may think that they comprehensively
resolve contentious questions instead of merely giving a framework for thinking
about all key issues. Selection of any arbitral rules will have already incorporated
broad principles and discretions and will often give rise to potential debates
about the inter-relationship between party autonomy, lex arbitri and provisions
in rules which will not be resolved by guidelines.

One of the most significant problems in attempting to promote harmonisation or

convergence is the fact that procedural rules are attempting to maximise
potentially conflicting norms. As noted throughout. we are concerned to deal
with both efficiency and fairness. As there can naturally be trade-offs between
the two and within each, it would simply be impossible to write any rules with
sufficient specificity that would indicate how trade-offs should be made on a
case-by-case basis. Thus, even a detailed set of rules can only identify starting
presumptions and would then inevitably give overriding discretions to vary from
the norm where it is reasonable to do so. Nevertheless, some indication of the
starting position can help parties understand the strength of their case, consider
whether to settle and if necessary give guidance as to how to prepare for an
adjudicatory process. Parties assessing their positions must benefit from
knowing whether they have a right to an oral hearing, a right to seek production
of specific documents and a right to cross-examine witnesses. Because such
evidentiary questions are so central to the outcome of individual cases, it is
understandable that this is the area where greatest privatised harmonisation has
occurred through the IBA Rules of Evidence.

It is also important to consider how standardised rules and guidelines may be

interpreted. To one leading practitioner ‘such attempts at procedural page "429"
standardisation may be counter-productive, as they are often interpreted and
applied in a formalistic and rigorous way rather than encouraging initiative.’ (141)
As is often the case, the devil is in the detail. Most adjudicators would wish to
have maximum predictability at the same time as maximum flexibility, aims
which inevitably conflict. It has also been observed that the differences in views
are more about the distribution of responsibilities between the parties and
adjudicators. (142) Another reason why a polarised and conceptual debate is
suboptimal is that the key issue is to determine which principles should be
articulated and to what extent. Flexibility requires an ability of a decision-maker
to adapt to particular circumstances. Yet that adaptation ought to be based on
some accepted criteria. Of course, there can be a multiplicity of factors that
conflict, but without a proper consideration of the factors, flexibility can too
easily become arbitrariness, and subjectivity. When such actions are undertaken
by a range of different tribunals, it also leads to inconsistency and
unpredictability. For example, a requirement to arbitrate in good faith is a clear
and desirable norm, but the way it applies in any individual instance may vary.
Nevertheless, it is better to articulate the norm if it ought to apply on policy
grounds, rather than leave it to arbitrators to decide if they wish it to apply.

The history of arbitration has also shown that the demand factor is significant.
The key example is the development of the IBA Rules and Guidelines as well as
work by the ILA. Leading arbitrators and institutions do not rush to take on new
work of this nature and such developments typically follow from lengthy periods
where concerns with the status quo become evident. However, the results can
sometimes be contentious. This is certainly the case with the preferences in the
IBA Rules of Evidence 2010. The real challenge is to articulate matters with such
precision that parties can know what is expected in advance, tribunals can shape
the process and follow their own mandate and do so in a way that is consistent
with other tribunals facing the same scenarios. Conversely, such rules and guides
should not be so detailed as to constrain adjudicators from dealing appropriately
with the special circumstances of individual cases. The real value of guidelines or
notes of this nature is that they at least force an arbitrator to consider the
desirability or otherwise of taking a procedural step. The following sections look
at the key generalist guides. Specialist guides such as the IBA Rules of Evidence
and Guidelines on Conflicts of Interest in International Arbitration are dealt with
in sections 5.9.4 and 10.3.6.1, and Chapters 11 and 12 generally.

page "430"

6.5.1. UNCITRAL Notes

Particular aspects of the UNCITRAL Notes on Organising Arbitral Proceedings

are dealt with under the relevant topic headings. Only general comments are
made in this section. The notes are not intended to be binding and unless
expressly selected by the parties, cannot alter any other agreement that they
have made on procedural matters. The UNCITRAL Notes are also not
incompatible with use of any particular institutional rules, although any express
prescriptions in such rules that conflict with a suggestion in the Notes would
take precedence, unless the notes were expressly agreed to by the parties, in
which case it would be a question of resolving conflicting intent. In most aspects,
the Notes are more in the way of a checklist of important factors that ought to be
considered in any event. (143)

The Introduction to the Notes makes clear that not all of the matters identified
will be relevant in each case. Furthermore, it makes the sensible suggestion that
it is advisable not to raise a matter prematurely, that is, before it is clear that a
decision is needed on the matter. However, while it can seem reasonable to defer
decision on procedural matters until they arise, once that has happened the
tribunal must choose between conflicting positions, which can be a problem.

Annex 8 to the UNCITRAL Notes contains a list of matters for possible

consideration in organising arbitral proceedings. The items listed are as follows:

List of Matters for Possible Consideration in Organizing Arbitral Proceedings

1. Set of arbitration rules

If the parties have not agreed on a set of arbitration rules, would they wish
to do so?
2. Language of proceedings
(a) Possible need for translation of documents, in full or in part
(b) Possible need for interpretation of oral presentations
(c) Cost of translation and interpretation
3. Place of arbitration
(a) Determination of the place of arbitration, if not already agreed upon by
the parties
(b) Possibility of meetings outside the place of arbitration
4. Administrative services that may be needed for the arbitral tribunal to carry
out its functions
5. Deposits in respect of costs
(a) Amount to be deposited
(b) Management of deposits
(c) Supplementary deposits
page "431"
6. Confidentiality of information relating to the arbitration; possible agreement
thereon
7. Routing of written communications among the parties and the arbitrators
8. Telefax and other electronic means of sending documents
(a) Telefax
(b) Other electronic means (eg electronic mail and magnetic or optical disk)
9. Arrangements for the exchange of written submissions
(a) Scheduling of written submissions
(b) Consecutive or simultaneous submissions
10. Practical details concerning written submission and evidence (eg method of
submission. copies, numbering, references)
11. Defining points at issue; order of deciding issues; defining relief or remedy
sought
(a) Should a list of points at issue be prepared
(b) In which order should the points at issue be decided
(c) Is there a need to define more precisely the relief or remedy sought
12. Possible settlement negotiations and their effect on scheduling proceedings
13. Documentary evidence
(a) Time limits for submission of documentary evidence intended to be
submitted by the parties; consequences of late submission
(b) Whether the arbitral tribunal intends to require a party to produce
documentary evidence
(c) Should assertions about the origin and receipt of documents and about
the correctness of photocopies be assumed as accurate
(d) Are the parties willing to submit jointly a set of documentary evidence
(e) Should voluminous and complicated documentary evidence be
presented through summaries, tabulations, charts, extracts or samples
14. Physical evidence other than documents
(a) What arrangements should be made if physical evidence will be
submitted
(b) What arrangements should be made if an on-site inspection is
necessary
15. Witnesses
(a) Advance notice about a witness whom a party intends to present;
written witnesses' statements
(b) Manner of taking oral evidence of witnesses
 (i) Order in which questions will be asked and the manner in which
the hearing of witnesses will be conducted
(ii) Whether oral testimony will be given under oath or affirmation
and, if so, in what form an oath or affirmation should be made
(iii) May witnesses be in the bearing room when they are not testifying
page "432"
(c) The order in which the witnesses will be called
(d) Interviewing witnesses prior to their appearance at a hearing
(e) Hearing representatives of a party
16. Experts and expert witnesses
(a) Expert appointed by the arbitral tribunal
(i) The expert's terms of reference
(ii) The opportunity of the parties to comment on the expert's report,
including by presenting expert testimony
(b) Expert opinion presented by a party (expert witness)
17. Hearing
(a) Decision whether to hold hearings
(b) Whether one period of hearings should be held or separate periods of
hearings
(c) Setting dates for hearings
(d) Whether there should be a limit on the aggregate amount of time each
party will have for oral argument: and questioning witnesses
(e) The order in which the parties will present their arguments and
evidence
(f) Length of hearings
(g) Arrangements for a record of the hearings
(h) Whether and when the parties are permitted to submit notes
summarizing their oral arguments
18. Multi-party arbitration
19. Possible requirements concerning filing or delivering the award Who should
take steps to fulfil any requirement?

6.5.2. ICC Techniques and ICC Rules 2012 Appendix IV

6.5.2.1. ICC Techniques

In 2007 the ICC published a booklet Techniques for Controlling Time and Costs in
Arbitration. (144) As the introduction suggests, the bulk of the costs of arbitration
are incurred by parties in presenting their cases. ICC analysis of cases in 2003
and 2004 showed that these costs comprised 82%, with arbitrators' fees and
expenses comprising 16%, while administrative expenses of the ICC were 2%.
The ratios may vary given that some institutions have higher scales for
arbitrators' fees and ad hoc arbitration might allow for even higher fees in
certain circumstances. Nevertheless, the broad differential is likely to be
consistently felt. Hence, the ICC considered that to reduce the overall costs,
special emphasis needed to be placed on reducing parties' presentation costs.
The length of proceedings is proportional page "433" to cost, hence the ICC belief
in the value of a publication aiming to control both time and costs in arbitration.

The Techniques make clear that they are to provide guidance on procedures that
the parties may agree upon. They are not intended to be comprehensive and
other solutions may be appropriate. The suggestion is that the document may be
given to the parties with a view to helping them reach agreement. In the event
that parties cannot agree, the Techniques may assist the tribunal in considering
the appropriate application of its own discretion. The introduction further
suggests that it is not intended to be prescriptive or even regarded as a code of
best practice. The aim is to provide both ideas and a checklist for appropriate
consideration. It warns against suboptimal practices and provides alternatives.
The Techniques indicate that they aim to embody two underlying principles;
first, that the parties and tribunal should seek to agree on appropriate specific
procedures early in the proceedings; secondly, that the tribunal should work
proactively with the parties. Because agreement begins with the arbitration
clause itself, the Techniques contain suggestions about the way this can help
constrain time and cost.

Where initiation is concerned, parties are invited to consider selecting

arbitrators and counsel with appropriate experience and time availability,
including arbitrators with strong case management skills; consider appointing a
sole arbitrator to reduce the time of appointment; and perhaps delegating
appointment to the ICC. Tribunals are also invited to use their discretion in
apportioning costs to promote efficient behaviour; empower the chairperson to
decide procedural matters alone; consider impact on time and cost if there is a
suggestion to appoint a secretary; consider alternatives to physical meetings;
consider whether there is a need for a hearing, fix the hearing date to avoid
problems at a later stage, use prehearing conferences to organise the process
efficiently; use IT; consider bifurcation where appropriate; provide an
environment where the parties may hopefully settle their dispute; invite the
parties to set out their case in full early in the proceedings; avoid repetition; limit
the length and number of submissions; manage document production; avoid
document duplication; highlight important documents; minimise length and
number of hearings; choose the best location; consider using telephone and
video conferencing, in particular for procedural hearings; provide cut-off dates
for evidence; consider which witnesses should appear; minimise the number of
witness statements and limit the number of witnesses; only use experts where
appropriate; consider the use of a single expert; organise expert meetings;
consider use of written witness statements of direct evidence; limit cross-
examination; limit closing submissions; and ensure appropriate time for
deliberation and completion of the award.

6.5.2.2. ICC Rules Appendix IV

Article 22(2) of the ICC Rules of Arbitration 2012 calls for the tribunal to ensure
effective case management, subject to the views of the parties. Appendix IV of the
Rules includes examples of some case management techniques that can be used
to control time and cost. The Appendix incorporates the key elements previously
page "434" published in ICC Techniques. These include consideration of
bifurcation, identifying issues; considering documents-only issues, controls on
document production requests and obligations to produce, limits on written
submissions and oral evidence, avoidance of repetition, IT techniques, pre-
hearing conferences and support of settlement.

6.5.2. Other Guides

6.5.3.1. ASA

The Swiss Arbitration Association has published a document ‘International

Arbitration “Do's and Don'ts“‘. (145) As the title implies, it aims to set out some
recommended practices and pitfalls to avoid as well as setting up the pros and
cons for the approach to various stages, providing information on arbitration
clauses, arbitration laws and institutions and generally outlining the essential
features of the arbitration process for those less experienced.

6.5.3.2. Counsel Protocols

While guides and protocols are normally for institutions, tribunals or the parties,
it is important to consider the role of counsel in seeking to promote time and
cost savings in arbitration. As has been noted, the bulk of the costs of arbitration
are in fact in counsels' fees and expenses. The above-mentioned guides look at
the way tribunals and parties could constrain counsel so as to minimise costs.
Section 6.2 above looked at proactivity in that regard, where the wishes of the
tribunal may conflict with the wishes of counsel. These are reactive and
restrictive regulatory issues, but there is also scope for a more positive approach
by counsel in seeking to achieve desired outcomes. This can of course be left to
individual counsel, but can also involve collaboration between counsel or guides
and protocols developed by individual counsel or bar associations. One approach
outside of the scope of this book is collaborative lawyering, where counsel for
both parties undertake to try and assist them in an agreed outcome and in many
cases, agree not to act as litigators if a settlement is not achieved.

An innovative guide aimed at counsel in the arbitration field is the Debevoise &
Plimpton LLP Protocol to Promote Efficiency in International Arbitration. (146) It
states:

International arbitration can provide significant advantages for parties to cross-

border disputes, such as a neutral forum, input into selecting the decision-maker
and nearly worldwide enforceability of awards. With page "435" seemingly
greater frequency, however, parties to international arbitrations express
concerns about increased length and cost of the arbitration process. These
concerns have caused some parties to question the value of international
arbitration as an efficient dispute resolution mechanism.
To respond to these concerns, the international arbitration practitioners at
Debevoise & Plimpton LLP have developed this Protocol to Promote Efficiency in
International Arbitration. This Protocol identifies specific procedures that
generally make an arbitration more efficient. Through this Protocol, we express
our commitment to explore with our clients how such procedures may be
applied in each case. In each arbitration, parties, counsel and arbitrators should
take maximum advantage of the flexibility inherent in international arbitration
and should use only the procedures that are warranted for that particular case.
The procedures set out here are therefore not meant to be inflexible rules.
However, through their consideration, we believe that we can improve the
arbitration process and thereby enable our clients to enjoy the advantages of
international arbitration.

Formation of the Tribunal

1. Before appointing arbitrators, we will ask them to confirm their availability

for hearings on an efficient and reasonably expeditious schedule.

2. We will ask arbitrators for a commitment that the award will be issued within
three months of the merits hearing or post-hearing briefs, if any.

3. We will work with out opposing counsel to appoint a sole arbitrator for
smaller disputes or where issues do not need the analysis of three arbitrators.

Establishing the Case and the Procedure

4. We will encourage consolidation and joinder of parties and disputes to avoid

multiple proceedings when possible.

5. When possible, we will include a detailed statement of claim with the request
for arbitration, so that briefing can proceed promptly once the procedural
calendar is established.

6. We will propose and encourage the arbitral tribunal to adopt procedures that
are appropriate for the particular case and that are designed to lead to an
efficient resolution. We will use our experience in crafting such procedures, and
we will not simply adopt procedures that follow the format of prior cases.

7. We will request the arbitral tribunal to hold an early procedural conference,

usually in-person, to establish procedures for the case. Although in-person
meetings may cost more because of travel time and expense, they often
ultimately save costs by allowing a more complete discussion of the procedural
issues that may arise. We will seek to set the merits hearing date, as well as all
other procedural deadlines, in this first procedural conference.

8. We will request our clients and opposing clients to attend any procedural
meetings and hearings with the arbitral tribunal, so that they can have page
"436" meaningful input on the procedures being adopted and consider what is
best for the parties at that time.
9. When appropriate to the needs of the case, we will consider a fast track
schedule with fixed deadlines.

10. We will explore whether bifurcation or a determination of preliminary

issues may lead to a quicker and more efficient resolution.

Evidence

11. We will limit and focus requests for the production of documents. We believe
that the standards set forth in the IBA Rules of Evidence generally provide an
appropriate balance of interest.

12. We will work with opposing counsel to determine the most cost-effective
means of dealing with electronic documents.

13. We will, when possible, make filings electronically and encourage paperless
arbitrations. When cost-effective, we will use hyperlinks between documentary
exhibits and their references in memoranda.

14. We will use written witness statements as direct testimony to focus the
evidence and hearings.

15. We will avoid having multiple witnesses testify about the same facts.

16. We will encourage meetings of experts, either before or after their reports
are drafted, to identify points of agreement and to narrow points of
disagreement before the hearing.

17. We will generally brief legal issues and consider presenting experts on issues
of law only when the tribunal and counsel are not qualified to act under that law.

18. We will divide the presentation of exhibits between core exhibits and
supplementary exhibits that provide necessary support for the claim or defense
but are unlikely to be referenced at a hearing.

The Hearing

19. We will consider the use of videoconferencing for testimony of witnesses

who are located far from the hearing venue and whose testimony is expected to
be less than two hours.

20. We will consider the use of a chess-clock process (fixed time limits) for
hearings.

21. We will not automatically request post-hearing briefs, but we will consider
in each case whether they would be helpful in promoting the efficient resolution
of the issues. When post-hearing briefs are appropriate, we will ask the arbitral
tribunal to identify the issues on which it may benefit from further exposition,
and then seek to limit the briefing to such issues.
22. We will also consider alternative briefing formats, such as the use of detailed
outlines rather than narrative briefs, to focus the issues and to make the briefs
more useful to the tribunal.

page "437"

Settlement Considerations

23. We will investigate routes to settlement, including by suggesting mediation,

when appropriate, either at the outset of the case or after an exchange of
submissions has further clarified the issues.

24. Where applicable rules or law permit, we will consider making a “without
prejudice except as to costs” settlement offer at an early stage. This will not only
protect our client's costs position, but it may lead the opposing party to consider
potential outcomes more seriously.

25. When appropriate, we will ask arbitrators to provide preliminary views that
could facilitate settlement.

6.6. The Role of the Chair

While this section is dealing with the role of the chair or presiding member of the
tribunal and hence is considering multi-person tribunals, many of the
observations are also relevant to a sole arbitrator in terms of procedural
management and the relationship between the tribunal and the parties, their
counsel and an institution if involved. Where there are multi-member tribunals,
one person will be designated as presiding. Such person is commonly spoken of
as the Chair, Chairperson, Chairman, President of the tribunal or presiding
arbitrator. Such a position is expressly designated in a range of arbitral statutes.
(147)

In a practical sense the Chair has significant powers. The Chair will need to
consider the additional functions that flow from that role when determining
availability and expertise. Once an appointment is accepted, the Chair will
typically control the establishment and implementation of the arbitral process.
Where there are oral hearings, the Chair will control the process, typically
establishing time frames and giving advice to witnesses as to what they may
expect to occur. In due course, the Chair will coordinate the deliberations. If a
multi-member tribunal has an even number, the Chair will typically have a
casting vote. The Chair will also typically draft the award for consideration by the
co-arbitrators. In some cases the Chair may seek permission to appoint a
secretary to support him or her in these tasks.

6.6.1. The Chair and Procedural Determinations

It is often the case that the Chair may make procedural orders alone. It is
normally appropriate to get such an agreement at the outset unless expressly
mandated in the laws or rules. Usually agreement will be called for. For example,
Article 29 of the page "438" UNCITRAL Model Law indicates that ‘questions of
procedure may be decided by the presiding arbitrator, if so authorised by the
parties or all members of the tribunal.’ At times, the Chair will be empowered to
make such orders but only after consultation with his or her tribunal colleagues.
At other times it may be that the obligation to consult is not applicable in
situations of urgency. (148) A midway position is articulated in the UNCITRAL
Rules 2010 where Article 33(2) allows the presiding arbitrator to make
procedural decisions where there is no majority or where so authorised by the
tribunal, subject to the entire tribunal having a power of revision.

Where the power to make procedural rulings is dependent on the concurrence of

co-arbitrators, this can pose a problem where a particularly parochial arbitrator
has been appointed. (149) Kaplan and Mills suggest the following clause that might
be utilised where party agreement and/or tribunal concurrence is required
before a chair can make decisions on procedure:

The Chair may make procedural (interlocutory) orders unless either side
requests that the same shall be considered by the whole tribunal. Any procedural
(interlocutory) orders signed by the Chair shall be deemed to be that of the
whole Tribunal if it states that the matter has been considered by the whole
Tribunal, whether face to face or on the telephone or in writing or in any
combination thereof. (150)

While rules may allow the presiding member to make decisions on questions of
procedure, there is then a need to consider what is meant by that term. As argued
in section 1.1.2 this should be determined by the way the case will be conducted
and not be dependent on the way different legal systems classify issues as either
procedural or substantive. For example, the fact that some legal systems
consider statutes of limitations to be procedural matters does not mean that a
Chair should decide on applicability alone where that is a matter in dispute
between the parties. Typical procedural decisions include logistics such as
designating who is to book rooms, organise interpreters, stenographers, setting
and extending time limits and identifying number of rounds of submissions. Even
then a Chairperson may often wish to consult co-arbitrators before making
decisions on such matters. In many instances it would be impractical to do
otherwise, in particular in setting time frames and hearing dates.

Another way to consider the distinction between matters appropriate for

separate determination by a Chair is to say that such procedural orders or
directions may not be matters that would constitute a partial award, as that is
only within the page "439" jurisdiction of the tribunal as a whole. It is important
to understand that the mere naming of a direction as an ‘order’ does not prevent
an argument that it is in essence a partial award, that is, a final determination of
an issue in the dispute. Time limits and extensions would naturally fit the
concept of procedure but choice of applicable law is instead a matter more
appropriate for the entire tribunal. As a general rule, interim measures should
not be dealt with by a Chair alone unless express agreement has been reached to
that effect. Any question which may impact upon the substantive outcome should
not be seen as a purely procedural matter for Chairperson determination,
although even that needs qualification. (151) Nevertheless, some rules also allow
the Chair to make the ultimate decision on the merits where there is no majority,
but this is an express provision dealing with a different issue. (152)

6.6.2. The Chair and Collaboration

A key role of the chair is to establish a harmonious and efficient working

relationship between the arbitrators. Where two arbitrators have had a close
relationship over the years, a third arbitrator who is unfamiliar to them needs to
feel properly included. A Chair should be careful to ensure that communications
with individual co-arbitrators do not give rise to a concern by a co-arbitrator that
there is an internal clique.

6.6.3. The Chair, Time Management and Hearings

Where there is a multi-person tribunal, it should be the Chair that raises issues of
proactivity generally and ensures that the parties are fully consulted. Proactivity
would involve aiming to ensure that only appropriate steps are utilised, rather
than following the fullest multi-staged process as a matter of course. The
relevant issues were discussed in section 6.2 and are not repeated here.

Organising time is particularly important. A sought after arbitrator will

commonly have many matters on at the same time. For that person to function
well as a Chair and at the same time perform other duties impeccably, it is
desirable to have an appropriate timetable of all key stages in all cases, with the
ability to promptly follow up any slippage by any party in meeting a particular
deadline. While it is generally easy to set up a schedule, it is particularly difficult
to keep it up to date as and when each case may have to revise deadlines. Having
an easy record of each individual case's schedule and a combined schedule
should help a Chair keep a page "440" good understanding of how each case is
progressing and an overview as to how the entire workload can best be
progressed. The Chair should help all parties understand what is required of
them. This occurs throughout the process, from preliminary conferences through
to final hearings. For example, where witnesses are involved, a Chair would
explain directly or via counsel how their evidence in chief is submitted, prepare
them for the impending cross-examination and advise as to the entitlement of
their counsel to re-examine.

Because it is particularly important for the Chair to gain the utmost respect and
be beyond reproach for neutrality, some suggest that it can be desirable to take a
passive role during the hearing, and allow party-appointed arbitrators to be the
ones to pose the most challenging questions to witnesses. (153) It has also been
observed that ‘humour does not travel well cross-culturally, nor does it read well
in a transcript in the cold light of a distant courtroom.’ (154) The recommended
demeanour is friendly, while at the same time being serious and professional
rather than informal. (155)

6.6.4. The Chair and Deliberations

Specific obligations will include ensuring that co-arbitrators have an appropriate
opportunity for input. Experienced and expert presiding members also face a
tension between ensuring that the most appropriate outcome is achieved while
at the same time ensuring that co-arbitrators truly have an independent
decision-making function. It is desirable that the Chair organise some time for
deliberations after the close of the hearing, making it clear that discussions at
that stage are only preliminary. (156) As suggested in Chapter 16, it can be too easy
to let complex damages issues slip through deliberations, without easy ways to
adequately thrash out complex issues.

A Chair may also need to carefully manage an obstructionist co-arbitrator. Co-

arbitrators need to feel that their views will be respected while at the same time
being made aware that a minority position can only be presented for such time
as the majority is as yet unclear of their final conclusion. If the co-arbitrator is
seeking to push a minority view ad nauseam, fair and firm time limits should be
imposed. If the co-arbitrator refuses to be involved in deliberations, the Chair
needs to promote page "441" transparency and give the person every
opportunity to attend the meetings of the majority. (157)

6.7. The Role of the Party-Appointed Arbitrator

Where party-appointed arbitrators are concerned, there is a natural tension

between duties of impartiality and independence, including duties not to be
predisposed or biased and the very ability of a party to nominate a person that it
hopes will lead to a favourable outcome. The potential for party appointment is
one of the most significant differences between the adjudicatory role of an
arbitrator and that of a domestic judge.

Not all rule systems direct party appointment of an arbitrator. In some

circumstances the tribunal may be appointed by an independent authority.
Where some institutions are concerned, parties may nominate an arbitrator but
the institution makes the appointment after considering questions of
independence and impartiality.

Historically, concerns arose as a result of the difference between domestic US

arbitration norms and the approach expected under international arbitration.
Where US domestic arbitration is concerned, there is a difference between
‘neutral’ and ‘non-neutral’ arbitrators under the AAA Commercial Arbitration
Rules and the Uniform Arbitration Act. (158) International arbitral rules and lex
arbitri do not allow for US domestic style norms and simply rely on the
obligation of independence and impartiality to indicate the overriding and
continuing obligations of all arbitrators whether party-appointed or not.

Born considered that party-appointed arbitrators ‘are an essential means of

ensuring the expert, efficient and internationally-neutral arbitral procedure
which is a central object of the parties' agreement to arbitrate.’ (159) Party
involvement in selection can ensure appropriate expertise and ability,
availability, appropriate language skills, understanding of cultural, economic and
political circumstances and conforms with party autonomy and adds to the
confidence in the process. (160)

page "442"

While party-appointed arbitrators are generally accepted, it is more challenging

to consider exactly how they should behave. In terms of describing how their
role may differ, Born sees their role as acting as guarantors for the respective
parties' rights to present their cases and have them fully understood and
considered. (161) HKIAC Rules Article 3.2 states that the arbitrator ‘shall not act as
advocate for any party’. A similar provision is contained in Article 5.2 of the LCIA
Rules. A party-appointed international arbitrator also has no special rights to
engage in ex parte communications during the course of the hearing. All agree
that there can be no bias in favour of the appointor. The party-appointed
arbitrator must bring an open mind to the facts and law in coming to an
independent judgment. (162)

The first challenge with a party-appointed arbitrator arises in relation to pre-

appointment interviews. To the extent that they are entitled to have any distinct
role, it would seem logical that a pre-appointment interview can seek to ensure
that they understand the proper ambit of that role. Yet if it goes so far as to warn
them of a need to test witness evidence, explain the party's position to co-
arbitrators and such other activist functions, it is very hard to distinguish this
from a contractual negotiation about the degree of advocacy to be employed. The
next potential challenge for a party-appointed arbitrator is in relation to
selection of a Chair. It is acceptable for such an appointee to consult with their
appointor in relation to potentially suitable presiding arbitrators. Nevertheless,
such discussions cannot deal with questions of predisposition or likely
disposition on the merits.

The next stage is during the currency of the proceedings. Commentators describe
it as being appropriate for a party-appointed arbitrator to be ‘sympathetic’ to the
appointer's position, ensuring that it is clearly understood. (163) The party-
appointed arbitrator can also ensure that the case is ‘adequately heard’. (164) They
may help ‘explain the parties' position’. (165) A party-appointed arbitrator
engaging in these aspects of sympathy is required to undertake certain actions.
(166) This may also include elucidating information from witnesses. The next

concern would relate to deliberations. To say that a party-appointed arbitrator

cannot be an advocate is itself problematic. In a tripartite tribunal, each must be
able to argue forcefully for any view they hold. If it happens to be a view as
propounded by the appointor, the party-appointed arbitrator should be equally
able to robustly argue the position in the tribunal deliberations.

page "443"

Some elements would not be controversial. Where the losing party does not
speak the language of the arbitration as a first language, a party-appointed
arbitrator can ensure that the award is sufficiently well drafted so that the loser
will understand why the case was lost. (167) A party-appointed arbitrator can also
ensure that the award addresses each of the loser's arguments in a
comprehensive way, lest it be thought that its position was not properly
understood. Actions on the part of a party-appointed arbitrator to improve the
quality of reasoning once a decision has already been made would certainly not
be contentious.

While it is relatively easy to state the broad obligations as to impartiality, the

next question is what to do if the obligation is breached. On one view, partial
behaviour may prove to be self-defeating. Böckstiegel says ‘While one may well
consider it a responsibility of the party-appointed arbitrator to make sure that
the written and oral submissions of the party that appointed him are taken well
into account in the deliberations of the tribunal, experience shows that a party-
appointed arbitrator playing the role of the advocate of the party within the
tribunal is self-defeating. If the chairman of a tribunal gets that latter impression
from the behaviour of a party-appointed arbitrator, this arbitrator's influence on
the deliberation and decision process of the tribunal will decrease considerably.
Thus, even from the party's point of view it is not in that party' s interest to select
and pressure an arbitrator to that effect.’ (168) That is true although the premise is
that the behaviour is transparent and unseemly. A skilled partisan arbitrator
might instead exert influence by subtle choices as to when to intervene in
deliberations or with questions to witnesses. Regardless of the likelihood or not
of such behaviour being self-defeating, it ought to be prevented wherever
possible. The presiding arbitrator may well have a private word with a party-
appointed arbitrator who is wrongly adopting an advocacy role, although the
language used will need to be particularly sensitive if there is no confidence that
confidentiality will be maintained. Domke also raised concern about the pressure
that a partial party-appointed arbitrator may put upon the neutral arbitrator.
This might encourage compromise solutions which would not be fair. (169)
Another extreme situation is if a neutral suggests a reduced award to try and
discourage a dissent. (170) At the extreme, such behaviour would mean that the
neutral is effectively engaged in the negotiation of an arbitral solution. An page
"444" excessively sympathetic party-appointed arbitrator might also encourage
the presiding arbitrator to make unilateral procedural decisions.

A contentious aspect of the role of a party-appointed arbitrator is to be vigilant

in terms of the behaviour of the other party appointee. If that person is thought
to be overstepping the mark, what duties does the first arbitrator have in such
circumstances? If it is felt that there are grounds for challenge as to impartiality,
can these concerns be communicated and if so, to whom? In most cases, an
arbitrator would not have express power to make a challenge against another
arbitrator, although some lex arbitri may be drafted broadly enough to
potentially encompass this.

6.8. The Role of a Secretary to the Tribunal

Subject to a contrary agreement of the parties, a tribunal may appoint a secretary

to assist it in certain tasks. There is some debate as to the proper ambit of those
tasks. Section 2.7.12 dealing with the duty of the tribunal to complete the
mandate, noted that a commensurate duty is a duty not to delegate unduly. Thus,
to properly understand the matters that can be left to be dealt with by a tribunal
secretary, one must consider the tribunal's own mandate and the limits on its
powers of delegation. Even if an arbitrator has not appointed an official
secretary, partners in law firms can use firm staff and individuals can use
research assistants in any event. Hence it is important that the norms provide
realistic guidance in such circumstances. Some institutions seek to circumscribe
the role of a tribunal secretary. For example, the ICC in a Secretariat Note seeks
to outline permitted duties and requires prior consultation and agreement of the
parties. (171) The ICC Note limits duties to administrative tasks.

A secretary should only be appointed with the agreement of the parties. The
parties should be entitled to approve the person nominated, a fee schedule if this
will ultimately be included in a cost award or otherwise be charged to the
parties, and have a clear understanding of the limits of the duties of the secretary
when appointed. A tribunal secretary or other assistant to one or more tribunal
members can naturally engage in secretarial and clerical work, whether this be
organising travel, hearing rooms, typing, arranging statements, exhibits and
filing documents. A secretarial assistant may also conduct legal research and
summarise evidentiary materials to help point tribunal members towards the
areas where key deliberations are required, although the tribunal must still
exercise independent and adequate judgment at all times. Care should be taken
to ensure that the tribunal does not improperly delegate the decision as to what
is or is not important if evidentiary materials are to be mustered by a secretary.

It is accepted that it is permissible for a secretary or arbitrator assistant to draft

the introductory portion of an award such as those parts outlining the identities
of the parties and counsel and if included, the procedural history and a brief
outline of page "445" the non-controversial facts. The procedural background
simply states what has occurred and is a summary of the material in
correspondence files or perhaps an auxiliary file developed by the tribunal. An
outline of the facts merely aims to be a non-contentious description of certain
agreed or uncontested facts. Where terms of reference have outlined the issues,
the secretary can incorporate appropriate extracts.

The essential parts of the award should be drafted by the tribunal and should not
be delegated to a secretary or any other person. More contentious is the question
of whether a secretary or other assistant can draft recommendations as to parts
of the award for tribunal evaluation. It is commonly suggested that this is
inappropriate. Nevertheless, if this was proscribed, there would be little to stop
thoughts being proffered as part of a research assistant's separate briefing note.
Even if it was expressly permitted, it still must come with a qualification that the
assistant cannot take over the essential adjudicative functions of the tribunal.
Here the important question is whether the tribunal has independently formed
its view as to the key evidentiary matters, as to the applicable law and has come
to form its own fully informed and reasoned view as to the basis of the
determinations and orders it wishes to make. The most controversial view would
be to suggest that if appropriate directions are given to the assistant, then it may
not be inherently improper for that person to be allowed to do a first draft of at
least some aspects. For example, if the tribunal already understands which
testimony of which witness was seen as determinative, is there harm in an
assistant extracting all other key aspects of testimony from all relevant witnesses
for completeness? Similarly, an assistant can collect an outline of various legal
submissions from both parties as to questions of law. This is not presented as a
recommended view, but instead, as an articulation of arguments about activities
that probably occur more often than they are spoken about.

In addition to debateable questions of propriety, however, there are also

questions of quality. An important aspect of the role of any adjudicator is to
explain to the loser why a particular result occurred. Tribunals should be careful
of delegating drafting tasks that detract from their ability to give a clear and
compelling set of reasons as to why the particular commercial outcome resulted
from the evidence and law that was proffered to it. Perhaps the most important
issue is to confirm that no matter what activities are delegated, the responsibility
is entirely on the tribunal to ensure that the award is presented as well as it can
be. At most, any delegated work can be seen as a recommended draft, in relation
to which independent judgment must be exercised by the tribunal. Complex
work should never be delegated if it cannot be meaningfully evaluated without
the same work being done by the tribunal.

6.9. Procedural Conferences and Timetables

A key aspect of the tribunal's role is to confer with the parties in setting and
explaining the intended procedures. This may involve a number of conferences
page "446" and meetings. It is important to utilise preliminary and case
management conferences for a range of reasons. Tribunals should do so to
ensure that proceedings are both efficient and fair and party autonomy is
maximised. Issues can be identified and procedural steps and timelines
articulated. Attention can be given to whether some preliminary determinations
should be made, whether as to jurisdictional challenges or as to applicable law.

Problems with preparatory meetings include the fact that a party can simply
seek to use them to increase delay or look for opportunities to challenge based
on those procedural determinations. It is important to be aware that once a
dispute has arisen, the parties are less likely to agree on even quite reasonable
and perhaps innocuous suggestions from the tribunal. A tribunal thus needs to
be concerned as to whether preliminary meetings will be used as an opportunity
for guerrilla tactics. Even without such an extreme attitude, raising matters in
the hope of clarifying all procedural issues could simply invite disputes that
would otherwise not arise. (172) Thus an arbitrator raising a very long checklist of
issues at a preliminary meeting might find that this simply exacerbates tensions
and slows down the process unnecessarily. A proposed text may be a desirable
alternative.

It is appropriate to distinguish between preliminary, procedural and prehearing

conferences and preliminary hearings. Conferences involve attempts to solicit
the views of the parties and seek a consensus position as to the way to conduct
the hearings. A preliminary hearing is where the parties make formal
submissions on a contested procedural question, calling for an order or direction
by the tribunal itself. The two are not mutually exclusive and a tribunal could
commence with an informal meeting and then at an appropriate stage, call for
submissions on a matter about which the parties cannot agree. (173) At the very
least, the tribunal should make clear what the exact nature and process of each
meeting will be. There is also a difference between what is described as a
preliminary conference or meeting on the one hand, and a pre-hearing meeting
or conference on the other. This section is dealing with preliminary meetings,
being the first opportunity to set a procedural framework. Conversely, a pre-
hearing meeting or conference is one that typically occurs shortly before any oral
hearing to finalise all key issues such as logistics, order of witnesses, witness
sequestration, chess clock rules and the like. This is discussed further in section
9.5.1.

An arbitrator's general discretionary power to conduct proceedings as the

tribunal sees fit is broad enough to encompass preliminary conferences.
Nevertheless, some arbitral rules expressly provide for these. (174) For example,
the ICC Rules require arbitrators to consult the parties before developing a
provisional page "447" timetable that it aims to follow throughout the
proceedings. (175) The tribunal should still consult with the parties even if the
relevant lex arbitri does not expressly call for this. (176) The UNCITRAL Rules
2010 direct the tribunal ‘as soon as practicable after its constitution and after
inviting the parties to express their views’ to establish a provisional timetable.
(177) Other rules expressly refer to such conferences in a permissive fashion. (178)

The ICDR Rules state the policy behind such conferences being to expedite the
subsequent proceedings. (179) A natural opportunity for a preliminary meeting is
where the arbitral rules call for the drafting of a memorandum of issues or terms
of reference.

Consideration of a draft timetable proposed by a party would meet minimal

consultation requirements, (180) although if the tribunal proposes something
quite different, there ought to be separate consultation on such a proposal.
Furthermore, party drafts can be wasteful when they just lead to competing
drafts with no consensus. However, there is no ‘one size that fits all’. Herrmann
has suggested that using expressions such as preparatory conference is
presumptive in signifying some meeting, whether face-to-face or via
teleconference. (181) The key issue is advance planning and rulings and not
necessarily an actual meeting except where prescribed.

6.9.1. Meetings with Parties

One advantage of in-person preliminary conferences is that they allow the

parties to meet each other face-to-face and to meet the tribunal. It is often
desirable to have a face-to-face meeting at the earliest opportunity so
arbitrators, counsel and perhaps the parties can put faces to names and develop
as harmonious a working relationship as possible. (182) This also helps parties
best understand the process, and page "448" feel that their wishes are
appropriately taken into account. (183) An important advantage is that parties are
more likely to agree to more timely and cost-effective procedures. This also helps
the parties to develop confidence in the arbitrator's resolve to deal with the
matter fairly and efficiently. Another advantage of having client representatives
present at case management conferences is so that they can be informed as to
future input required from them. (184) A tribunal is generally entitled to
specifically require attendance of client representatives. (185) There is also great
value in providing opportunities for settlement.

6.9.2. Consultations Generally

The following comments relate more generally to consultations per se, whether
the parties are present or not. Consultations ensure that procedural directions
are more likely to be met, given that the parties and their counsel can also
indicate their own scheduling commitments. Consultation also ensures that
proactivity does not degenerate into hierarchical paternalism. Being invited to be
actively involved helps the parties understand their entitlement to set the
framework where agreement is possible amongst them. This will also help to
turn the minds of parties from diverse backgrounds to the need to provide for
appropriate compromise solutions. Early and regular involvement of the parties
should also hopefully preclude procedural challenges at a later stage. Such a
process also has a valuable educative role, helping parties less familiar with
arbitration to understand how it is to be employed and to have parties from
different legal cultural backgrounds understand the norms that may be applied.
An early procedural conference will help a tribunal understand if there are
differences in attitude based on differing legal cultures of the parties and/or
their counsel and whether there are differing levels of expertise in arbitral
experience between counsel. (186) This is particularly important when there is no
automatic method of dealing with key issues such as evidence and production of
documents. A tribunal might be reluctant to be making rulings at such early
conferences, hoping instead to foster consensus and a non-adversarial ambience
in procedural deliberations. (187)

Video conferences and telephone conferences may be alternatives to face-to-face

meetings and a useful means of saving costs, although they are unlikely to be as
suitable for engendering good social interaction as outlined above. It is even page
"449" appropriate to consider the methodology by which procedural
determinations are made in deciding whether face-to-face meetings are
desirable. Psychological insights would suggest that it is far easier to disagree
with a procedural suggestion made in writing or via an international telephone
conference than in person. The mere physical presence of an arbitrator of
renown can have a very strong influence on the way parties behave.
Furthermore, a face-to-face meeting can be more interactive in relation to body
language and in terms of questioning potential oppositional statements.

6.9.3. Timing of Preliminary Conferences

There are important issues of timing involved. An arbitrator needs to exercise

significant judgment when seeking to establish an initial procedural framework.
On the one hand, it seems reasonable that there be as much clarity as early as
possible to ensure that the parties exercise their resources in the most efficient
manner. On the other hand, trying to be too prescriptive at the outset may be
misguided if decisions are made before the key issues in the case become clearer
after written submissions are provided. In this context, some advocate pre-
hearing conferences at the earliest opportunity so that parties get to meet each
other and are best encouraged to work collaboratively on procedural matters
and perhaps settle their dispute. (188) Others assert that conferences should wait
until detailed claims and cross-claims have been exchanged so that the tribunal
can articulate the process after having sufficient knowledge of the real issues.
(189)

The timing and purpose of preliminary conferences may also vary depending on
whether the case is ad hoc or institutional. The key differences where an
institution is involved is that the parties will almost certainly have been asked to
outline their claims and counterclaims through some form of Request for
Arbitration and Answer to the Request, in some instances with a view to
preparation of Terms of Reference or a Memorandum of Issues. If the parties
have carefully articulated their key contentions, the tribunal can come to a
preliminary conference with a good understanding of the case at hand. ICC
Techniques indicate that if the parties set out their respective cases in sufficient
detail, a case management conference can be held concurrently with a meeting to
resolve the terms of reference. (190) Conversely in many ad hoc arbitrations, there
might only have been a short Notice of Arbitration before there is a need to
establish a procedural framework. This would be particularly so if the ad hoc
arbitration had not selected something such as the UNCITRAL Rules. Either way,
if a tribunal is to make effective use of preliminary page "450" meetings, the
arbitrator should devote sufficient time and attention to understanding the case
and the way it should best be resolved procedurally.

6.9.4. Organising the Preliminary Conference

Where there is no prior agreement on important procedural questions, the

tribunal will need to consider whether these matters should simply be addressed
at a preliminary conference or whether submissions should be sought in advance
as to a preferred procedural model by each party. There are advantages and
disadvantages to each approach. Leaving it to the hearing may be less certain and
the hearing can at times go off the rails, although as noted, counsel find it harder
to take unreasonable positions when face-to-face with an experienced expert
tribunal. Another approach is to send parties' counsel a procedural
questionnaire inviting them to state their preferences on potential elements of
the process. (191)

It is generally desirable for a multi-person tribunal to meet before a preliminary

conference to get to know each other where necessary and discuss their
preliminary views about organisational issues. The same is so for a meeting of
the parties, given the need to consult them as to their views. It is far preferable to
consider a consensus view of the parties than to try and bring together divergent
positions.

As noted above, given that an important function of face-to-face conferences is to

have the parties themselves feel involved, get to understand contrary positions
and at least consider the possibility of settlement, it is desirable that both parties
and counsel are present. Where the parties are concerned, it is highly desirable
that they are represented by someone who can make decisions and who does not
have to refer to higher authorities. (192) This may be more problematic where a
State is a party in investment disputes.

It may be appropriate to consider adjourning such meetings from time to time so

that parties can consult with counsel and each other. This is particularly so given
the educative function of such meetings for persons less experienced with
arbitration and the different procedural options that could legitimately be
employed. True consent should be informed consent. This is also important
when parties come from different cultural backgrounds, given that an agreed
procedure is likely to be a hybrid or compromise.

Where the parties have come to agreement, it may be appropriate for the chair to
prepare or have the parties prepare a Consent Order for the tribunal. (193)

page "451"

6.9.5. Matters to Consider at a Preliminary Conference

There is a significant difference between a preliminary conference that merely

identifies the parties preferred procedures and stages and one which instead,
invites consideration of the issues in dispute and the way they should best be
handled from fairness and efficiency perspectives. David Rivkin has argued
persuasively that the arbitration community should not simply follow historical
norms but should look at each case afresh to decide exactly which procedures
are necessary and appropriate. That means asking whether documents,
document production, oral hearings and other costly and time-consuming stages
are in fact appropriate for the case at hand. (194) It is sometimes the case that it is
just as appropriate to designate what should not occur as it is to designate what
should occur. Khanna argues persuasively against the use of boilerplate
‘Procedural Order No 1’ by experienced arbitrators or those who obtain such
proformas from published works. Instead, the ideal starting point suggested for a
preliminary meeting is where counsel are asked to address the tribunal as to the
merits and issues involved so that the tribunal can better frame a procedural
schedule relating to those issues and the way they are best dealt with. (195) This
may not be practical in complex and emerging matters but the idea is sensible if
practical.

Without being too prescriptive, the tribunal might consider whether to set an
agenda for the conference. It might also be appropriate to draft Terms of
Procedure. Depending on the stage of the proceedings, it may also be appropriate
to call for short outlines of each party's legal and factual arguments. If so, it
should be made clear that these are without prejudice to any later or different
submissions. (196) The tribunal might also consider inviting the parties to adopt a
set of arbitration rules if they have not already done so, although there should be
no presumption that this is required. A preliminary conference may also be an
opportunity to invite the parties to consider expedited proceedings. If parties are
interested in that approach, a set of appropriate expedited rules may be
recommended.

It is particularly important for the parties to understand the approach to

evidence from the outset. At that stage, parties will naturally think about the role
of document production but the entire gamut of evidence should be taken into
account by competent counsel in deciding on the best strategy to persuade the
tribunal of the outcome they desire. This would imply early attention to the role
of expert witnesses, whether party or tribunal appointment. Often the impact of
a procedural direction depends on the way a tribunal will approach other issues.
page "452" For example, whether evidence in chief and even cross-examination
will be limited to matters dealt with in the written statements will impact upon
the way they are drafted. Simply speaking about document production without
an indication of how challenges are to be conducted, does not really indicate how
documentary evidence is to be dealt with. Even identifying the need for written
witness statements without an indication of format, content and counsel
involvement may be insufficient for effective case management. This is
particularly so where there is a mismatch in experience or significant differences
in cultural background. (197)

The tribunal might also discuss the method of cost allocation and whether it will
be prepared to consider unreasonable behaviour as well as the result in making
such cost determinations. Examples of unreasonable behaviour could include
dilatory behaviour, excessive and unnecessary document production requests,
excessive numbers of witnesses and cross-examination, and refusals to meet
deadlines. If the point is not made as to the potential use of costs powers, the
incentive effects of such an approach will be limited.

An early preliminary conference can also be an opportunity to explore the

possibility of agreed statements of facts. Some arbitrators will seek to require the
parties to at least attempt such an agreement. A mid-position is for an arbitrator
to at times ask whether a particular fact is truly likely to be in dispute,
notwithstanding that it is asserted to be so in the statements of case and defence.
An example would be a case dealing with force majeure or frustration where a
storm was alleged to have frustrated the ability of a ship to depart with delivery
of goods. Are the parties truly contesting whether the storm occurred and its
exact severity or are they instead merely contesting its foreseeability and
alternative means by which the supplier could have met its contractual
obligations? Early questions by the tribunal in such circumstances display no
bias and help the parties consider which matters they truly wish to spend time
and money contesting. One non-contentious but commonly useful aspect of
agreed facts is a timeline of key events and an agreed bundle of relevant
documents. It may also be useful to have an outline of key personnel and their
status, particularly where there are a large number of witnesses and/or persons
whose documents are to be utilised.

The tribunal should be careful to separate out matters that are properly
considered terms of its own appointment from the balance of the issues in a
preliminary conference. Questions such as advance payments of fees and
cancellation fee entitlements are not matters of fundamental due process and are
within the legitimate negotiating entitlements of parties and prospective
arbitrators. However, it will inevitably be the case that some procedural
suggestions will have implications for tribunal remuneration and workload and
may not easily be separated in this way.

page "453"

Sections 6.5.2.1, 6.5.2.2 and 6.5.1 above outlined ICC Techniques, revised ICC
Rules and the UNCITRAL Notes and similar checklists that can aid a tribunal in
setting an agenda. These are central to conference design. The most significant
questions are likely to be:

1. Are there preliminary issues that should be determined such as challenges to

jurisdiction, applicable law or requests for consolidation or joinder? Are there
any cross or counter claims?
2. Where written submissions are concerned, how many rounds should there
be? Should these be simultaneous or sequential?
3. At what stage should parties produce the documents on which they intend to
rely? What, if any, document production requests ought to be permissible by
one party as against the other? How will challenges to such requests be dealt
with?
4. What physical form will the documentary evidence be presented in and how
can this best lead to an efficient record for the benefit of all parties including
the tribunal?
5. Will there be attempts to seek evidence from third parties and if so, what
involvement will the tribunal be asked to have?
6. Will witnesses present evidence? Will there be written witness statements
and reply statements? Parties may also be asked to confirm that witnesses
within their control will attend to give evidence. Even if this does not
eventuate, this allows for adverse inferences at a later stage and minimises
the need for subpoenas.
7. Will there be expert witnesses and will there be tribunal-appointed or party-
appointed experts or both?
8. In what order will witnesses give testimony? What guidance will be given to
experts? Will there be pre-hearing meetings and conferences called to refine
such evidence?
9. Will there be one or more hearings? If so, how much time is likely to be
needed for any hearing and when is it appropriate to set aside a time?

The Iran–United States Claims Tribunal supplemented the broad discretion as to

conduct of proceedings with Note 4 providing for pre-hearing conferences.
Internal guidelines of the tribunal dated 28 January 1982 detail an illustrative
list of matters that may be considered at a pre-hearing conference. The non-
exhaustive list includes:

(a) clarification of the issues presented and the relief sought;

(b) identification of any issues to be considered as preliminary questions;
(c) status of any settlement discussions;
(d) whether any further written statements, including any reply or a joinder, is
requested by the arbitrating parties or required by the arbitral tribunal…;
(e) fixing a schedule for submission by each arbitrating party of a summary of
the documents or lists of witnesses or other evidence it intends to present…;
page "454"
(f) fixing a schedule for submission of any documents, exhibits or other
evidence which the arbitral tribunal may then require …;
(g) whether voluminous and complicated data should be presented through
summaries, tabulations, charts, graphs or extracts in order to save time and
costs;
(h) desirability of appointing an expert by the arbitral tribunal, and if so the
expert's qualifications and terms of reference; whether the arbitrating
parties intend to present experts, and, if so, the qualification of and the areas
of expertise to be covered by any such expert;
(i) determining what documentary evidence will require translation;
(j) fixing a schedule of hearings;
(k) other appropriate matters. (198)

Again, such checklists can be useful or can lead to additional problems. A party
can simply seek to use them to increase delay or look for opportunities to
challenge based on those procedural determinations. Once a dispute has arisen,
the parties are less likely to agree on even quite reasonable and perhaps
innocuous suggestions from the tribunal. A tribunal thus needs to be concerned
as to whether preliminary meetings will be used as an opportunity for guerrilla
tactics. Even without such an extreme attitude, raising matters in the hope of
clarifying all procedural issues could simply invite disputes that would otherwise
not arise. (199) Thus an arbitrator raising a very long checklist of issues at a
preliminary meeting might find that this simply exacerbates tensions and slows
down the process unnecessarily.

There is also an issue of variations to matters previously determined. Due

process and an appropriate opportunity to present a case requires a proper
understanding of the way complex matters are to be dealt with. Once the ground
rules are established in this way, they should not be changed without good
reason, as a change may itself be seen to be unfair. (200) However, early
procedural determinations should allow for variations once the essential nature
of the case becomes more obvious. It is particularly important that early
directions do not need the agreement of all parties to be modified for changed
circumstances. (201) Parties should of course be consulted when such changes are
contemplated.

page "455"

6.9.6. Timetables
Current versions of some key rules call for the development of a provisional
timetable. Examples include the ICC Rules 2012, (202) the Swiss Rules 2012 (203)
and the UNCITRAL Rules 2010. (204) A provisional timetable is a crucial document
in that it not only outlines key dates but also outlines key stages, many of which
are discretionary elements of arbitral procedure. Hence in devising a provisional
timetable, an arbitrator is typically making important choices about the way the
proceedings will be conducted and is not simply setting time periods. A tribunal
must decide what stages are appropriate and only then can consider when they
should occur. In this sense, while a provisional timetable is an inherently
practical matter, what is desirable is dependent on some important conceptual
issues, namely the extent of the arbitrator's power to control proceedings and in
particular, the proper approach where the views of the parties and the arbitrator
diverge.

There is also a difference between a timetable established prior to the parties'

articulating their respective cases and one that is established afterwards. The
latter can aim to set out a framework for the entire proceedings. In the
alternative scenario, the timetable instead is primarily about organising the
parties to present their cases in a timely and efficient manner.

The previous section noted rules that indicate that parties are to be ‘consulted’.
These imply that in those cases the tribunal has the final word and further imply
that unreasonable preferences of the parties could rightly be ignored. (205) This
may also imply that the parties cannot impose a timetable that the tribunal
believes to be either unfair or inefficient in the context of the circumstances that
existed when the appointment was made. (206) Given that such a timetable should
be developed in consultation with the parties, it has the advantage of having the
parties brought together at an early stage which may facilitate settlement and
helps all in planning the workload, as the timetable will not only deal with key
stages but also indicate what work must be done by whom at what point in time.
Consultation with the parties will also allow the tribunal to seek further
information that may be necessary before deciding the timetable. For example,
the tribunal might seek the parties' advice as to whether an agreed statement of
facts is likely to be possible. If consultations lead to a consensus position, then
this is itself an agreement of the parties as to their legal entitlement. (207) It has
been suggested that the page "456" provisional timetable should state if it is
agreed by all the parties, (208) but should not be signed lest it be seen as a binding
part of the arbitration agreement. (209)

Even if the parties have a firm agreement as to the proposed timetable, a tribunal
must still be able to indicate when the proposed agreement is impractical,
unreasonable or has inappropriate gaps. One problem in relying too heavily on
the wishes of the parties is that in many cases, neither may be able to assess
what is reasonable. In addition to the lack of experience of many counsel as to
arbitration and its logistical challenges, many claimants would want unduly fast
proceedings based on a view that there is no defence on the merits and time is
money, while many respondents who expect to lose may wish to delay
proceedings significantly to delay any payment and/or force a favourable
settlement.
In some circumstances it may be necessary to make a procedural ruling on a
particular matter before deciding on the overall timeframe. An example would be
cases where there are claims against multiple parties, only some of whom have
signed the arbitration agreement. Where there is any kind of challenge to
jurisdiction, the tribunal will need to be sensitive in terms of the timetable.
Drawing up a timetable that purports to go beyond the determination of
jurisdiction might look to some as if there was a prejudgment on the issue. The
better view is that if the interests of justice and efficiency would require parties
to lock in certain dates in case jurisdiction is found, it may be preferable to set a
broader timeframe. (210)

6.9.7. Timetables and Time Limits

Section 6.3 above dealt with time limits generally. This section is limited to the
setting of a timetable. Nevertheless, when setting the timetable, the tribunal
needs to be aware of any overriding time limits for the making of the award. This
is complicated under those lex arbitri or rules that provide a time limit but allow
for extension. The key example is the ICC Rules where extensions are habitually
granted. The question is then whether a tribunal should set the first provisional
timetable with a view to complying with the time limit or can ignore it from the
outset where circumstances make it clear that it would not be appropriate or
possible to complete within the period designated. The ICC practice is to allow
provisional timetables that extend beyond the six-month time limit articulated
page "457" in Article 30 of the 2012 Rules, no doubt on the assumption that
extensions would be granted in due course. (211)

Time limits might be set by way of fixed dates or by set periods as from the
completion of the previous stage. There are advantages and disadvantages either
way. Specified dates are clearer and avoid disputes about the triggering event.
However, the alternative approach of specifying a time following a previous
event, will stay relevant if the previous event takes longer than expected. Fixed
time limits invariably need to all be modified if there is some slippage at a prior
stage. If stipulated time periods are utilised rather than fixed dates, attention
needs to be given to the calculation of the starting and finishing dates and also
how weekends and public holidays will be dealt with.

In setting a provisional timetable there is also a natural tension between two

alternative objectives. The first would be that time frames are realistic, giving
appropriately generous deadlines to meet the vagaries of modern commitments
so as to avoid missed deadlines, extension applications and the need to recast
timetables from time to time. However, it is desirable that there is sufficient
firmness and urgency to ensure that the matter is concluded as expeditiously as
could reasonably be expected. ICC Techniques suggest that ‘(a) provisional
timetable with the shortest times that are realistic should be established.’ (212) If
timelines are generated, the tribunal should also indicate the consequences
where these are not met. If the timetable set is too ambitious it is likely that time
will slip, which defeats the initial purpose. (213) Setting a fair but tight timetable
will be an important way to promote efficiency, calling for appropriate work at
the earliest opportunity by counsel, which in turn will ensure that each party will
properly address the strengths and weaknesses of its case, identify points of
agreement and the remaining points that are truly in issue and hence allow for a
more targeted dispute. This will also aid in settlement. A provisional timetable in
default proceedings also has the added value of providing ongoing notice to the
defaulting party of future stages where it might wish to rejoin the proceedings.

Philip suggests that in the absence of special circumstances, a reasonable

timetable in most cases is to allow four to six weeks for each submission, utilise
consecutive submissions, allow each party two briefs and provide for document
requests and witness statements with briefs, a further four weeks for further
submission of documents and three to five days for a hearing. Under this
timeframe, hearings are concluded within seven to eight months, after which the
award can page "458" then be completed. (214) This should not be seen as even a
presumptive model, but instead, a guide to typical timelines if all of these stages
are warranted. Added to this timeframe is the time taken to appoint the tribunal,
hold the preparatory meeting and where appropriate, settle terms of reference.
Many see a year as a maximum timeframe to aim at in normal circumstances. (215)

A tribunal needs to ensure that there is both an appropriate timetable and also
that delays and extension requests are appropriately dealt with and the parties
are guided in preparing for any hearings in an adequate and timely manner.
Where a schedule is developed, it should be made clear that it is provisional and
can be modified where circumstances require, but otherwise should be seen as
binding on the parties. If timelines are generated, the tribunal should also
indicate the consequences where these are not met. Many counsel will tend to
ignore tribunal timelines more than judicial guidelines unless a firm hand is
established at the outset. The efficiency benefits of such a schedule can be
undermined if a tribunal will too readily depart from it when the parties simply
fail to adhere without just cause. However, some flexibility will always be
necessary.

Where States or State institutions are concerned, extra leeway may be inevitable
as representatives cannot usually make clear commitments without seeking
supervisory approval and bureaucratic processes tend to take longer than
similar processes in commercial enterprises. (216)

6.9.8. Draft Timetables

A provisional timetable need not set out each and every stage of the proposed
arbitration. Because of the desirability of setting a framework at the earliest
opportunity, in some cases it only makes sense to set a timetable for part of the
process, to be completed when further material is available and the tribunal is
best able to identify the optimal procedure. There is no presumption that a
matter not included cannot be part of the proceedings. Nevertheless, if a party or
the tribunal wishes to add a step, this must be determined after appropriate
consultation. While it would generally be inappropriate to consider items not
mentioned to be excluded by implication, (217) if the parties have expressly agreed
on a procedure, one of them might seek to argue that the agreement was indeed
comprehensive and conditional page "459" on other matters not being included.
In most cases such a conclusion would not seem appropriate.

The requirement that the tribunal consult the parties or provide them with an
opportunity to express their views does not necessarily imply that the tribunal
should present them with a draft timetable but it may instead call on them to
make comments about timetabling per se. There are advantages and
disadvantages in providing a draft. Key advantages are that an experienced
tribunal will be better able to establish a sensible timeline. Parties might more
readily accept a suggestion from the tribunal than one from the opposing party.
However, if not specifically tailored with careful thought for the issues in the
instant case, a draft timetable from the tribunal can too readily be a pro forma
that might invite the parties to adopt certain processes that might otherwise be
unnecessary. An example is a blanket inclusion of document production and
production contest stages. Experienced arbitrators should be careful of using
standard models from past cases and less experienced arbitrators should be
wary of simply transplanting a timetable found in an article or reported case (or
this book!). Obviously the need is to develop a timetable that is fair and
reasonable in the circumstances of the instant case.

Instead of proffering a draft timetable with items the parties would presume to
be desirable, an alternative might be to invite counsel to consider the checklist in
the UNCITRAL Notes on Organising Arbitral Proceedings and discuss amongst
themselves which matters ought to be included and covered. An alternative
would be for the tribunal to have a discussion with the parties around such a
checklist, indicating the tribunal's views as to the steps it believes are necessary
and the advantages and disadvantages of optional steps. As noted, this may be a
more useful and targeted discussion if the parties are first asked to articulate the
key issues and facts in contention.

The following is a list of matters to be considered in the provisional timetable.

Again the appropriate caution is against including all items as a matter of course,
as parties will tend to accept the tribunal's suggestion and might include
expensive and time-consuming stage that might otherwise be unnecessary in the
instant case. With that caveat in mind, the following topics should at least be
considered by the tribunal:

1. Submissions.
2. Documentary evidence.
3. Preliminary determinations.
4. Identification of witnesses.
5. Applications for interim measures.
6. Witness statements and pre-hearing conferences.
7. Responsive witness statements (if any).
8. Dates for hearings.
9. Pre-hearing witness conferences.
10. Joint expert witness reports.
11. Post-hearing briefs.
page "460"

6.10. Service and Communication

Am early task of the tribunal is to establish the means of service and

communication to the extent that these have not already been agreed upon by
the parties or are not designated in the lex arbitri or arbitral rules. These issues
were discussed in sections 4.7 and 4.8 above as they are central to
commencement of proceedings. In some cases, tribunals may vary stipulated
processes where warranted at later stages.

6.11. Advances and Deposits on Costs

Various rules will provide for each party to pay a share of an advance for costs in
relation to the arbitration. This is standard under institutional rules. A tribunal in
an ad hoc arbitration may also make such a direction as an early procedural
order. In some legal systems it may be considered inappropriate for a tribunal to
demand advance deposits in relation to its own fees absent an agreement to that
effect with the parties at the time of appointment. That would be dependent on
the way that agreement was interpreted but if the parties had selected rules that
so empowered a tribunal, the power can naturally be employed. (218) Where
annulment applications are made in relation to ICSID arbitration, it is the
applicant that must pay the full advances. (219)

An important issue is what ramifications flow if one or either party fails to

comply with such directions. It is appropriate to analyse ICC rules and practices
as more consideration and analysis has occurred in that context. Article 36 of the
ICC Rules 2012 establishes a detailed regime dealing with advances to cover the
costs of the arbitration. The claimant may be asked to pay a provisional advance
that aims to cover the costs of the arbitration until the Terms of Reference have
been drawn up. As soon as practicable, the Court then fixes the advance based on
the institute's schedule. Article 36(2) directs that the advance shall be payable in
equal shares by the claimant and the respondent. Joinder and multi-party
scenarios are treated separately. In such circumstances the Court determines
what advances shall be payable. (220) Separate advances may be fixed in relation
to any counterclaims. Adjustments may be made at any time during the
arbitration. (221)

The ICC Rules do not provide any remedy or direction as to the process to follow
or ramifications of a party failing to produce the required sum for the advance of
costs save that Article 36(5) indicates that in all cases any party page "461" shall
be free to pay any other party's share of any advance on costs should such other
party fail to pay its share. That would normally be a claimant paying a
respondent's share, but in some cases after a claimant has abandoned
proceedings, a respondent may wish to have a final award together with a costs
order denying the claims on a res judicata basis. Generally, even though they
have failed to pay the required advance, respondents who refuse to pay are not
denied the ability to defend themselves in arbitral proceedings. The only
proceedings they will be barred from are those which they instigate (such as a
counterclaim) which requires a separate advance of costs. The counterclaim can
be barred if such advances are not paid. Article 36(6) indicates that where a
request for an advance has not been complied with and after consultation with
the tribunal, the Secretary-General may direct the tribunal to suspend its work
and set a time limit of not less than fifteen days on the expiry of which the
relevant claims shall be considered as withdrawn. An objection can be made
within that time period to be decided by the Court. If claims are then withdrawn,
they can be reintroduced subsequently. (222) Article 36(6) needs to be read
alongside Article 36(5) allowing for either party to pay the other's share. It is
only where this does not occur that claims may be treated as abandoned for non-
payment.

Tribunal procedural orders are likely to be to similar effect where ad hoc

arbitration is concerned, although they are not bound to do so. Article 43 of the
UNCITRAL Rules 2010 allows a tribunal on its establishment to request the
parties to deposit an equal amount as an advance for the costs as defined in
Article 40(2)(a) to (c). This covers the costs of the tribunal, travel and other
expenses of arbitrators and the costs of expert advice and assistance required by
the tribunal. Supplementary deposits may be called for during the course of the
proceedings. Where there is an appointing authority and the parties so request
and the authority consents, the amount of deposits or supplementary deposits
can only be fixed after consultation with the appointing authority which can
make comments to the tribunal. Where deposits have been ordered, if not paid in
full within thirty days after receipt of the request, the tribunal is to inform the
parties after which one or more of them may make the required payment. If not
then made, the tribunal may order the suspension or termination of the
proceedings. The Model Law is silent on the proper response where a party
refuses to pay its share of costs.

The logic behind an institution being entitled to treat the claims as abandoned, or
an ad hoc arbitrator making procedural orders to similar effect is that there is a
separate contract between the parties and the institution on the one hand and
the tribunal itself. Neither the tribunal nor the institution should be forced to
carry on its activities if professional fees will not be paid. More troubling is the
question of the inter-party rights where one has been required to pay the other's
advance on costs. There is no consensus view as to what should happen. There
may be differences depending on whether it is institutional or ad hoc arbitration,
and depending on the particular articulation in the relevant rules. Party
autonomy would always be paramount but parties will rarely give any express
directions in that regard. Where page "462" the rules are unclear, tribunals have
taken fundamentally different views depending on the way they consider that
arbitral consent impacts upon this question. The better view is that fee
determinations by tribunals (as opposed to cost allocations) are not awards and
are not subject to award supervision processes. (223)

There are three different approaches in the literature to obtaining the required
cost advancement from a non-paying party. The first approach is the contractual
approach. The contractual approach is based on the fundamental underlying
obligations of the parties in accepting arbitration and rules which direct
advances. Because the non-paying party has initially agreed to rules as their
dispute resolution mechanism, some will argue that this creates reciprocal
contractual promises to pay the advance on costs as required. A subsequent
failure to pay thus results in a contractual breach itself that can be dealt with
under the arbitration agreement if it is broad enough. (224) Unfortunately this
does not solve the problem of the party having to pay the full advance as only
after it is paid can the tribunal proceed. Furthermore, non-payment would be a
new breach of the arbitration agreement that ought to be introduced as a new
claim. That itself could be problematic depending on timing and may even
require further advances on costs. While a tribunal could technically order
specific performance of a contractual obligation, if a party ignored the original
direction, it may be likely to ignore such a ruling in any event. Importantly, a
party cannot be excluded from the arbitration on this basis. Nor can adverse
inferences be drawn as to the merits. The arbitration agreement might also not
be broad enough to encompass such a claim. Other problems may arise where
the non-paying party cannot be said to have consented to such rules. An example
would be where the respondent claims that the negotiations broke down and
there was never an agreement to arbitrate or any commercial agreement for that
matter. This approach has been successfully argued in a number of unpublished
ICC awards, including Partial Award dated 2 September 1996 in ICC case 7289.
In other proceedings, tribunals have suggested that such contractual obligations
never arise. Where ICC arbitration is concerned, the logic against is that the
agreement is between the ICC Court and the parties and is therefore procedural,
not contractual as between the parties. (225) Other tribunals have taken the view
that there are no contractual promises between the parties but instead
obligations between the individual parties and the institution. From the
institution's perspective it certainly does not seek to take on a power to take
legal action page "463" for its fees, but instead requires the innocent party to pay
the other's share or the case will be considered abandoned. Thus the
institutional model is never about fee recovery but instead about sufficient
advances before the work is done and abandonment when the advances are not
paid.

The second approach to non-payment of an advance is known as the interim

approach. This approach involves the paying party making a request to the
arbitral tribunal to apply an interim or provisional measure requiring the non-
paying party to cover their portion of the cost advance or to reimburse the
paying party for the full amount paid. That would in turn depend on the interim
measure powers of the particular tribunal and in some cases, whether domestic
courts would have concurrent or controlling jurisdiction. Even where a tribunal
has jurisdiction, there may be a problem in getting the claim before the tribunal
in a timely manner. Court orders occurred in two cases where the parties refused
to pay their advance because the respondents were also refusing to pay. The ICC
dismissed the claims against the respondents but both parties obtained interim
measures from the respondents' respective domestic courts ordering each
respondent to pay the required cost advance. (226) This is, however, very
inefficient and time-consuming. Furthermore it brings in issues pertaining to
domestic courts and their view on whether they have jurisdiction. Forcing
parties to go to court for orders, even if permitted by national law, adds time and
expense and would have enforcement problems if the recalcitrant party was not
in that jurisdiction. Furthermore, because different national courts would take
differing views, the solution does not apply equally to all potential parties. An
interim measure of protection is also not an appropriate use of such a power as it
is really a final determination of a legal obligation and not a measure of
protection as to a separate merits determination. A contrary argument is that it
could be seen as an interim measure as all advances or deposits of costs are
provisional vis-à-vis the tribunal's final costs order.

If an interim measure approach is taken, it should be remembered that this is

always discretionary. A tribunal might consider that there is no need for a
separate order in advance of the final costs ruling, that there is questionable
merit in the claims or they are inflated, setting higher advances than would
otherwise be reasonable or where there are legitimate challenges to jurisdiction.
(227) An interim measure approach may also need to consider the balancing of

interests between the parties.

page "464"

The third approach is known as the provisional approach. This approach hinges
on the wording of provisions such as Article 36(1) of the ICC Rules 2012 which
states:

After receipt of the Request, the Secretary-General may request the claimant to
pay a provisional advance in an amount intended to cover the costs of arbitration
until the Terms of Reference have been drawn up.

The paying party can request the ICC to determine a provisional advance for it to
pay in lieu of total costs. This provisional amount will allow the arbitral tribunal
to continue beyond the reference stage and allow for an application based on the
failure of the non-paying party to pay and obtain an order requiring the
respondent to pay the advance. This is made possible because Article 28(1) of
the ICC Rules 2012 allows for interim awards before the terms of reference are
drawn up. It has been suggested that this approach enables the claimant to: ‘(1)
pay less fees than the total advance on costs, (2) circumvent any procedural
issues, (3) get the merits before the arbitral tribunal, and (4) reserve the
argument for the final award, should the interim measure petition fail’. (228) An
interim measure approach is also a discretionary matter.

It has also been suggested by some tribunals and courts that where a respondent
fails to pay the cost advance required under the ICC Rules they have simply
waived their rights in arbitration and therefore their inaction allows the claimant
to pursue the defaulting party through the courts. There is both arbitral (229) and
judicial (230) support for such a position but it is a strained conclusion as it is
tantamount to an agreement to give claimant the option as to the applicable
forum. It would also seem unfair if the respondent has no funds to contribute, in
which case it is hardly conscious waiver. Considering the arbitration agreement
at an end based on an allegation of fundamental breach would also require
attention to the applicable law of the agreement and whether non-payment of an
advance could itself be described as fundamental. Even if these hurdles could be
overcome, resort to litigation may be wholly unsatisfactory, with problems as to
the appropriate forum and enforceability of judgments.

Often scholarly commentary predominantly assumes that a non-paying

respondent is declining payment in order to delay proceedings or for another
illegitimate purpose. However it has been recognised that there are a number of
situations where there may be a legitimate reason for refusing to pay the cost
advance. These include:

[W]hen it challenges the jurisdiction of the Arbitral Tribunal … [and] when it has
no prospect of recovering its costs from the claimant, either because the latter
has filed for bankruptcy, or receivership, or is otherwise in a financial page "465"
position which makes reimbursement of the costs incurred by the respondent
from the outset unlikely, if not impossible. (231)

These issues however can still be easily dealt with if the respondents do pay
their cost advance as per the ICC Rules. In the first situation if jurisdiction is not
found, then the cost advance would be recovered by the respondent as the
advance will not have been required. In the latter case the respondent can ask
that some sort of security for costs be given by the claimant in order to make
sure the advance could be reclaimed at the end of the proceedings. It is therefore
suggested that in any case where a respondent is genuinely and legitimately
concerned about the claimant there are arbitral provisions and principles to deal
with this under the rules. Therefore the intention behind the refusal to pay need
not be investigated.

A problem may arise if a respondent pays the advance in relation to its

counterclaim but refuses to pay its share in relation to the primary claims. In a
practical sense, the claimant can simply refuse to pay its advance on the
counterclaim, thus forcing each to pay the full advances on the matters it wishes
to bring before the tribunal.

Many other institutions have reacted to the non-payment of cost advances by

respondents by inserting express provisions within their institutional rules in
order to deal with the situation. Where a party fails to make a payment and
another party makes the payment instead (e.g., in order to keep the proceedings
on foot), the Stockholm Chamber of Commerce Rules allow arbitrators to make a
separate award for reimbursement of the payment. (232) Article 24.3 of the LCIA
Arbitration Rules provides that the party paying the substitute payment shall be
entitled to recover that amount as a debt immediately due from the defaulting
party.

6.12. Terms of Reference

Historically, terms of reference were a corollary of legal systems that did not
enforce arbitration agreements purporting to deal with prospective disputes, but
instead, required a submission to arbitration once the dispute evolved. (233) While
this is no longer a justification, some would see value in terms of reference being
drawn with a view to aid the efficiency of the process. Others would consider it
an unnecessary expense. Arguments for and against are outlined in section
6.12.1 below.

Some institutions require the use Terms of Reference, the most notable being the
ICC. Parties are, of course, free to utilise Terms of Reference regardless of page
"466" whether these are demanded by institutional rules. (234) An arbitrator's
broad procedural discretion would also allow the use of Terms of Reference
where it was thought appropriate. Documents to similar effect may at times be
utilised. These may be similar in content and style to minutes of preparatory
conference determinations. (235)

6.12.1. Should Terms of Reference be Used?

While some critics suggest that the use of terms of reference is needlessly
bureaucratic and actually counter-productive in terms of cost and time, Gary
Born suggests that this is unfounded as the techniques required by the process
are inherently useful and, in any case, many of its components would be adopted
by a prudent tribunal even if it was not operating under the ICC Rules. (236) The
debate is complex as there are potential advantages and potential risks with the
use of Terms of Reference where their use is discretionary. These can be heavily
impacted upon by the attitude of the participants, in particular the tribunal. It is
useful to outline the arguments for and against, as much as anything, to warn
users how to optimise the process.

There are some important suggested advantages with the use of Terms of
Reference. While the parties have framed claims and counterclaims in their
opening documents, these may not be clearly and succinctly expressed, in which
case an arbitrator can make sure the parties understand and address every key
issue by articulating them clearly within the Terms of Reference. Terms can also
simplify matters which were unduly repetitive. By framing the key aspects of the
dispute from the outset, the parties are properly guided as to how to engage in
their ongoing preparation and presentation of material. Properly drafted Terms
of Reference remove any element of surprise. Forcing the parties to consider
issues at the outset and consult on agreed terms, concentrates their attention on
the issues at an early stage, and helps establish a relationship between them.
Well drafted terms of reference will ensure that each party addresses each key
issue and will do so in an order that is determined to be logical by the tribunal.

Where Terms of Reference force the parties to consider aspects of the case that
they may have glossed over, that will not only help in preparation, but may also
help them to critically evaluate the strengths and weaknesses of their own case.
The terms of reference process can also aid settlement, particularly when the
parties are forced to consider counterclaims and defences alongside their own
page "467" claims. (237) Another suggested advantage of terms of reference is to
allow the parties to meet to discuss the issues, although this could be utilised in
any event in a preliminary conference. Gerald Aksen also notes that the terms of
reference process forces parties to address the real issues when the request for
arbitration and answer might be drafted in haste; and provides comfort to an
enforcing court that the parties were engaged in the process. (238)

Terms of reference will also help the arbitrator in setting an appropriate

procedural framework. Another advantage of identifying issues at the earliest
stage is that it helps limit document production entitlements and circumscribes
the relevance of witnesses and the breadth of their testimony. Terms of
reference also support scrutiny of the award where this is part of an institutional
function. The Terms provide a frame of reference to consider whether the award
is ultra or infra petita. (239) Another suggested advantage of terms of reference is
that when signed by the parties, they constitute an agreement on their terms and
in some instances can correct defects in the original arbitration agreement. (240)
Depending on the way they are drafted and interpreted, terms of reference can at
times be seen as constituting a distinct arbitration agreement. Once the Terms of
Reference are agreed, that provides strong evidence of consent to that particular
methodology and hence helps protect the award from challenge at the
enforcement stage. (241)

There are some potential disadvantages of Terms of Reference as well. They may
unduly constrain the process, given the difficulty of amending the terms and
admitting new claims. (242) This is discussed further in section 6.13.3.
Disadvantages also include the need to undertake the work itself with attendant
costs. There is the potential for disruptive parties to disagree on draft terms to
try and delay the process and present the arbitrator with threats as to future
challenges based on disputed articulation of the terms themselves. There are
also timing issues. Imposing strict time limits on development of terms of
reference seems intuitively desirable to shape the process from the earliest
stages, but this can have negative effects. For example, it might lead to delay in
other stages such as document production, which might have helped to clarify
the issues in dispute. Calling for terms to be drafted at an early stage might mean
that the parties could take a scattergun approach to issues to be determined, lest
they are seen to have given away an argument that on further research might
prove better than originally thought.

page "468"

A further problem with terms of reference may arise if they articulate a

particular procedural process that the tribunal then wishes to alter based on its
broad discretion. If a tribunal does so, it may be challenged for failing to follow
the terms of reference. (243) Alternatively, if the tribunal was reluctant to make
the change for fear of such a challenge, its ability to modify the procedure is
unduly constrained. This can best be handled by separating terms of reference
and procedural determinations.

The final potential problem with Terms of Reference is that they might not be
appropriately drafted. (244) Where the arbitrator seeks to paraphrase the
essential elements of the pleadings and does so in a contentious manner, this
might inappropriately redirect the arbitration and lead to potential grounds of
challenge in due course. (245) On the one hand, this should not be a concern, as
Terms of Reference signed by the parties can be said to have been consented to.
On the other hand, it is conceivable that a party might ultimately discover that
the deference it gave to the tribunal's draft Terms of Reference was
inappropriately given. In some cases, a question will be whether unconditional
acceptance of terms of reference waives the right to challenge an issue not
articulated in the terms, including a challenge to jurisdiction. Where there is a
question as to jurisdiction, the mere involvement in signing Terms of Reference
does not generally constitute a submission to jurisdiction. (246)

Many of these problems are either overstated or can be readily avoided. For
parties engaged in arbitration in good faith, the costs issue should not be
significant. If the pleadings have already carefully articulated the key issues, then
it should not take much time to convert these into Terms of Reference.
Conversely, if the pleadings do not articulate the key issues well, then the added
costs in drafting terms of reference are likely to lead to significant savings in
later processes simply because of the added clarity. Where disruptive parties are
concerned, this is now a significant potential problem in many arbitrations, but
should not in and of itself preclude the use of any particular process that
otherwise might be fair and efficient. The problem of a disruptive party should
be met head on in relation to that party's behaviour and should not be allowed to
redirect what would otherwise be fair and reasonable processes.

The ultimate utility of terms of reference may largely depend on the view the
tribunal has of them. There are likely to be many arbitrators who do not believe
in their value and simply produce terms to comply with institutional rules that
require them. Tribunals with this mindset are most likely to defer to the
articulation of issues by the parties and ensure that there is a well drafted
blanket provision page "469" allowing for new claims and issues to be dealt with.
This will cause little harm, but will also provide little benefit to offset the
expense. Unclear pleadings will underpin unclear terms. This is discussed further
in the following section dealing with recommended inclusions in terms.

6.13. What Should be Included in Terms of Reference

Where terms of reference are to be used, this will commonly be finalised at the
first preliminary conference at the same time as a First Procedural Order and
provisional timetable is established. In some cases the timetable may be deferred
till a short time thereafter if necessary. (247) Procedural orders should be
separated from terms of reference as they will generally be provisional in nature,
whereas the terms of reference have a more permanent controlling function. As
noted above, a tribunal may wish to vary procedures, which becomes
problematic if contained in terms of reference.

Using the ICC Rules 2012 as a model, Article 23(1) requires the tribunal to draft
Terms of Reference as soon as it has received the file from the Secretariat on the
basis of the documents submitted or in the presence of the parties and in the
light of their most recent submissions. Article 23(2) indicates that within two
months of the date on which the file has been transmitted to it, the tribunal shall
transmit to the court the terms of reference signed by it and the parties. An
extension may be granted. The court is not given a review power over the Terms
of Reference. That is to be contrasted with the court's authority to review the
award. It will have an approval power if any party refuses to take part in drafting
or signing the terms. (248)

In addition to providing details of the tribunal, the parties and their

representatives and the addresses for notices and communications, the Terms of
Reference are to include particulars of the place of arbitration; particulars of the
applicable procedural rules, including if appropriate, reference to the power
conferred on the tribunal to act as amiable compositeur or to decide ex aequo et
bono; a summary of the parties' respective claims and the relief sought by each
party, with an indication to the extent possible of the amounts claimed or
counterclaimed; and ‘unless the arbitral tribunal considers it inappropriate, a list
of issues to be determined’. Some of these details may have important
consequences at later stages. For example, the incorrect identification of a party
which is a member of a group of companies may affect the successful party's
abilities to enforce the award.

While ICC Rules provide for the terms of reference to be prepared either on the
basis of the parties' submitted documents or in the presence of the parties, in
practice, Derains and Schwartz suggest that the most practical approach is for
the page "470" tribunal to prepare a draft of the terms of reference based on the
parties' written submissions, circulate it for comment, and then arrange a
meeting at which the terms of reference can be finalised and then signed by the
parties and arbitrators. (249) Conversely, it has been suggested that best practice
is to ask each party to provide a short summary for insertion into the Terms. (250)
Inviting the parties to also prepare their own versions of the summary of claims
can help in avoiding any later allegation of bias in the description of the claims.
Such an invitation may also reduce time spent by the parties trying to change the
wording of a tribunal's draft terms of reference in order to ensure that their own
claims are described as favourably as possible. It is certainly less contentious for
the parties to summarise what they want rather than to invite debates as to
whether the tribunal has accurately captured the respective positions. However,
terms of reference are often most useful when earlier pleadings are convoluted
and do not clearly delineate the particular issues to be determined. Where those
pleadings are not elegantly drawn, slavishly copying them adds little in the way
of clarification through the use of Terms of Reference. Parties will too readily
repeat contentions in Statements of Case and Defence, in which case there is
little, if any, value added by the process. While it is not inherently wrong for
tribunals to invite the parties to set out the issues in their own words, the more
they do so, the less this stage may constitute proactive case management. (251)
Hence, regardless of who prepares the initial summary, the role of the tribunal is
to ensure that the final form becomes a useful framework for directing the
parties and the tribunal and where appropriate, allowing for easy institutional
scrutiny.

6.13.1. Drafting a List of Issues

Whether and how to draft a list of issues is the most challenging matter. The ICC
Rules suggested list of issues is an optional aspect of the terms of reference
requirement, because the list can become an area of contention between the
parties. Derains and Schwartz point out that ‘issues’ should not be confused with
‘claims’ and that issues consist of ‘fact and substantive or procedural law which
at the time of drafting appear to be relevant to the adjudication of the parties'
claims’. (252) Preparation of the list of issues can be difficult at this early stage of
the arbitration, page "471" both because the issues may only become fully
apparent as the proceedings unfold, and because parties may object to certain
issues being included or described in a particular way. (253) However, the list of
issues is not intended to be comprehensive and it is understood that more issues
may arise in the course of the arbitration. (254) Accordingly, the ICC Rules provide
that the list of issues is to be prepared unless the tribunal considers it
inappropriate to do so. If necessary, then, the list of issues can be omitted where
it becomes too contentious, but to do so would be contrary to the objectives of
the terms of reference and accordingly many tribunals tend to include the list.
(255)

One advantage of a list of issues may be to bring together claims and

counterclaims in a logical chronological order for each party to follow in making
their submissions. Individual parties will not be primarily concerned with the
most logical overall structure of the tribunals task when drafting their individual
assertions. Another advantage may be where the tribunal can list sub-issues not
expressly addressed by the parties, but which are essential elements of their
contentions. This will ensure that each party properly addresses the matters of
concern and neither can say that the tribunal's logic took them by surprise.

As always, there are potential disadvantages as well. Some question whether

there is anything meaningful to add by constructing a list of issues where the
terms of reference must have already summarised the claims and counterclaims.
Problems can arise where the parties disagree about the definition of issues and
hence delay the process further. This is one reason for giving the tribunal a
discretion whether to undertake this task. (256) Others are more concerned to
avoid grounds for challenge and dispute. Some arbitrators, concerned with the
possibility of challenges to their articulation, might feel that it is habitually more
appropriate to stipulate that the tribunal will simply decide ‘such issues as may
arise in the course of the dispute as shall be set out in the submissions of the
parties', or words to similar effect. (257) This is discussed further in section 6.13.3
dealing with new claims. One of the problems in attempting to devise a list is
where the final award does not strictly marry up to the issues outlined in the
terms. There are many reasons why this could possibly eventuate. New issues
can emerge as the proceedings develop. Issues that were thought important may
lose their importance depending on the outcome of other determinations.
Nevertheless, where there is an page "472" apparent mismatch on the face of the
documents, there is the risk that an aggrieved party will argue that the final
award was infra petita or ultra petita. Where institutional scrutiny is involved,
because institutional counsel would be less familiar with the case materials than
counsel of the parties, the terms of reference will be the natural document on
which they base their analysis.
If an attempt is made to articulate issues, the arbitrator is trying to distil one
logical set of issues that will frame resolution of the dispute from disparate
claims and counterclaims in various pleadings. Inevitably, the arbitrator is
considering what is important in the context of the dispute as pleaded. In doing
so, however, the arbitrator must refrain from appearing to take any preliminary
view on the merits, even in determining which issues seem most important from
the outset. Some tribunals ask the parties to prepare their own drafts of issues.
(258) Other tribunals seek to merely extract the claims as articulated in the

Request and Answer. Others seek to synthesise those documents into a clearer
and more coherent form.

As noted, it is also standard to include a general provision indicating that the

issues will be those raised from time to time by the parties so that the Terms do
not unduly constrain the inclusion of new matters. (259) Some suggested drafts are
outlined in section 6.13.3 below. It is also worth repeating that it is desirable to
leave procedural directions out of the terms of reference as such directions may
naturally need to be modified from time to time. (260)

6.13.2. Signature of Terms of Reference

Institutional rules that call for terms of reference or memoranda of issues

commonly require these to be signed by the arbitrator(s) and the parties so that
there is a clear evidentiary record of agreement. When the terms of reference are
signed, arbitrators must be careful to ensure that the individuals signing the
terms of reference have the power to enter into arbitration agreements on behalf
of their principals. This is especially important where persons are signing as
attorneys for a party, as some jurisdictions have special requirements where
attorneys purport to enter into arbitration agreements on behalf of the donor of
the power. (261)

page "473"

It is important to consider the implications, if any, if the necessary parties refuse

to sign the terms of reference. Where a party refuses to sign, an indication as to
the basis of this decision will give the tribunal guidance as to how best to
proceed. For example, if a party is unhappy with the drafting and no compromise
seems possible, care might well be taken to ensure that the complaining party
properly presents material in relation to its preferred articulation of the terms.
Where the ICC is concerned, if one of the parties refuses to take part or to sign,
the draft is submitted to the Court of Arbitration for its approval. (262) The court
then fixes a time limit for approval by the party in default. If the time limit
expires without the party having signed, the arbitration may proceed. ICC
Techniques suggests that if a party refuses to take part in drawing up the Terms
or refuses to sign them, the tribunal should ensure that the final form for
approval does not contain provisions requiring agreement of the parties or
decisions of the tribunal. (263)

Goldsmith raises the question as to what should occur if one arbitrator refuses to
sign the terms in a multi-arbitrator panel. (264) That would be a barrier to the
continuation of that arbitrator in the proceedings. If an arbitrator refuses to sign,
that would be reported to the relevant institution who may take action as is the
case with any failure of an arbitrator to perform his or her duties.

6.13.3. Admissibility of New Claims outside of the Terms of Reference

A process which aims to clarify claims and issues to promote certainty, also
needs to determine what degree of flexibility to allow in terms of admissibility of
new claims. From a policy perspective, if it is too easy to change claims, the
process was probably wasteful. If it is too hard, it is probably too constricting. If
it is unclear when changes are permitted, this undermines the very certainty that
was aimed for. Because of this, some parties may be wary of the terms of
reference process if they feel that signing the document limits their rights or may
be construed as constituting a waiver of rights or expression of intent, or would
otherwise affect the tribunal's mandate. The terms of reference are not
necessarily intended to do this, and a qualification stating this can be included
where necessary in the terms of reference. Furthermore, Derains and Schwartz
point out that ‘the scope of the Arbitral Tribunal's mandate flows […] from the
claims that have been properly placed before the arbitrators in accordance with
page "474" the parties' agreements and the Rules and not from the description of
those claims in the Terms of Reference’. (265)

As noted above, the first issue is the timing of the terms vis-à-vis the initial
articulation of claims. Article 23(1) of the ICC Rules 2012 provides that the terms
of reference are to be prepared in the light of the parties' ‘most recent’
submissions. It is not clear at what point the tribunal may refuse to accept any
further submissions, although it is open to the tribunal to clarify this by setting a
deadline after which further submissions may not be made. (266) Fouchard,
Gaillard, Goldman suggest that the claims may be changed at any time prior to
the terms of reference being signed, other than where the change includes the
addition of a new party, in which case a new request for arbitration must be
made. (267) Derains and Schwartz note that arbitrators will generally incorporate
new aspects of submissions or new claims into the terms of reference given that
it can become difficult to amend the terms of reference once they are signed. (268)

The balance of this section looks at attempts to change claims after terms have
been finalised. ICC Rules 2012 Article 23(4) requires the tribunal's discretion to
add new claims after the Terms of Reference are complete. The tribunal is to
consider ‘the nature of such new claims, the stage of the arbitration and other
relevant circumstances’. It is only to be expected that the power would be
articulated in such a general fashion. There is then a need to consider when and
why it should be employed. This raises major policy tensions in terms of fairness
versus efficiency. It is not simply a question of application by a party. The
tribunal may need to give separate attention to whether matters have been
raised in subsequent submissions that are arguably outside the Terms. If so the
tribunal should seek the response of the parties. (269) Even if there is a debate as
to the validity of matters outside the Terms, proper consultation with the parties
may alleviate the problem. There may also be questions of waiver under Article
39 of the ICC Rules 2012. (270)
It is necessary to first consider the drafting of the terms and the degree to which
it can be said that there truly is a new claim. Because new issues and the relative
importance of issues will typically vary as cases are developed, it is important
that there is not too much rigidity in the Terms requiring agreement of all parties
before such revised issues can be dealt with. In some cases it is not clear whether
a change is needed. New arguments as to existing claims do not need approval.
This is discussed further in the following section. At other times, the terms may
be ambiguous. A party will therefore need to refer to the tribunal to page "475"
check whether the proposed new claim actually does fall within the terms, or
otherwise satisfy the tribunal that the new claim should be incorporated. To the
extent that the tribunal has discretion, a policy issue is whether the tribunal
should consider whether a party should have raised matters earlier or whether
the tribunal should only consider any potential prejudice to the opposing party.
A concern for the parties alone might suggest allowance of a new claim on the
basis that the parties consented to arbitration and if there is no prejudice, all
relevant matters should be dealt with. This is also affected by whether any
potential prejudice can be compensated for in other ways, such as costs or
reduced damages or interest. Conversely, a concern for a more systemic
encouragement of efficiency might wish to warn other prospective disputants to
act expeditiously in order to promote respect for arbitration per se.

In considering the exercise of any discretion, a tribunal might also consider

whether in the event that an amendment is denied, the aggrieved party could
start a new case and in turn ask for consolidation between the two. There might,
however, be differing views as to whether this factor aids or hinders the
favourable exercise of a variation discretion. Whether new claims ought to be
permitted also depends on how one views Terms of Reference. To the extent that
these are seen as a form of modified consent, then they should not be able to be
varied unilaterally. Conversely, to the extent that they are merely seen as a
method of streamlining the matters in issue, the more that modification should
at least be contemplated, particularly in complex matters.

An unduly restrictive approach to changes will also encourage undesirable

responses. The more there is a difficulty in bringing new claims, the more the
parties will be encouraged to describe their claims broadly. At the extreme, this
will lead to different problems such as challenges on the grounds that the claims
are too vague or requests for further particulars. It will be likely to add to overall
cost. Thus, terms tend to be drafted broadly and empower an arbitrator to
consider all matters flowing from the claims and counterclaims that are deemed
appropriate and necessary to these issues. Yet a form of drafting that is overly
concerned to allow modifications is less likely to perform the primary function of
clarifying and constraining the issues truly in dispute. It is also important that
due process must still be upheld, with the parties given an adequate opportunity
to address and respond to the issues. (271) Hence even the broadest wording must
still allow a party opposing a change to explain why this might raise due process
concerns.

There are different types of changes to terms of reference that might be

contemplated. Tribunals have adopted ‘widely varying’ techniques for
determining the question of admission of new claims, but as an example, Derains
and Schwartz suggest that a claim based on facts relevant to existing claims
would be more likely to be admitted than one requiring the submission of new
evidence. (272) The most challenging are changes that raise statute of limitations
issues or other jurisdictional concerns.

page "476"

The Terms of Reference themselves can indicate whether new claims are to be
allowed and if so on what criteria. For example, an ICC Commission Working
Group Report suggested the following clause for such purposes:

The issues to be determined shall be those resulting from the parties'

submissions and which are relevant to adjudication of the parties' respective
claims and defences. In particular, the arbitral tribunal may have to consider the
following issues (but not necessarily all of these and only these, and not in the
following order): (273)

Carter also cites another model provided at an ICC Educational Seminar that is
even broader:

The arbitral tribunal is to resolve all issues of fact and law that shall arise from
the claims and counterclaims and pleadings as duly submitted by the parties,
including, but not limited to, the following issues, as well as any additional issues
of fact or law which the arbitral tribunal, in its own discretion, may deem
necessary to decide upon for the purpose of rendering any arbitral award in the
present arbitration: (etc). (274)

Another variant of the catch-all general reference is as suggested by Webster:

The questions of fact or law to be resolved by the arbitral tribunal shall be those
appearing in the parties submissions, statements and pleadings, and in addition
any further questions of fact or law which the tribunal, in its own discretion, may
deem necessary or appropriate to decide upon for the purpose of rendering any
arbitral awards. (275)

6.13.4. New Claims as Distinct from New Evidence and Arguments

The ICC Rules merely prevent new claims not covered by the Terms of Reference,
not new arguments and evidence. New arguments may be raised, although there
are still due process issues as to timing. The party responding to a new argument
is entitled to have appropriate warning and an adequate opportunity to respond,
which should include reasonable time for research and analysis. In ICC Case No
7047 page "477" (1994), a principle was propounded whereby a claim would
only be seen as being new and not within the limits of then Article 16 of the ICC
Rules if it raised issues of fact and of law which are completely new compared
with the issue in dispute so far. If it is, however, a claim subsequently based on
different reasoning, but still on the same facts, it is within the limits of Article 16
of the ICC Rules. (276)
Recalculating quantum should not be seen as a new claim. (277) If the currency has
changed that may constitute a new claim. (278) New evidence is a separate issue
and again depends on timing and due process. This is discussed in section
10.22.3.

6.13.5. Defences, Counterclaims and Set-Off and the Terms of Reference

Because particular institutional rules will require counterclaims and defences to

be raised prior to the time for terms of reference, these should be included in the
latter as a matter of course. The most contentious situation is that of set-off
rights where some legal systems believe these arise automatically without any
need to be pleaded. This is discussed further in section 4.4.

Some rules leave the discretion open-ended while others specify certain relevant
factors including the delay in making the request and the prejudice to the other
parties and the impact on the conduct of the proceedings. Derains argues that the
impact on proceedings should be the predominant criterion. (279) Derains also
points out that it is conceptually problematic to state that to be admitted, new
claims must fall within the scope of the arbitration clause. If it falls outside the
clause, the Tribunal ultimately has no jurisdiction but to make that
determination would have to ‘admit’ the claim to rule on it. (280) Amendments to
requests could either be simple completion of logistical requirements or could
instead constitute an essentially new request for arbitration, particularly where
new parties or claims are involved.

Any allowed amendment to claims and counterclaims must obviously meet the
criteria of allowing each party an appropriate opportunity to present its case. In
some situations it may be necessary to determine whether a party is truly
amending its claims or is merely adding additional reasons to support claims
already made. If it is a true amendment to a claim, it must be based on
application and be dealt with by the tribunal with appropriate concern for due
process. If it is page "478" merely an added reason for an existing claim, then the
party responding to that new reason cannot claim lack of due process. (281) A
change in the amount of a claim should not constitute a new claim. Depending on
the circumstances a claim in a different currency may constitute a new claim.

Additional claims can always be made by consent. Where consent is not

forthcoming, the rules can either try to define conceptually what can or cannot
be changed or instead, anything may be allowed if it comes within a defined time
period.

6.14. Written Submissions

6.14.1. The Role of Submissions

Because parties come from different geographical regions, as may the

arbitrators, it is natural that international arbitration relies heavily on written
submissions and communications. These will always be more cost-effective than
oral presentations conducted at a single location, requiring many participants to
travel and sit listening to each others submissions. International arbitrations
typically use various rounds of written submissions to present claims, relief
sought, issues within claims, and factual and legal allegations on each of the
above. Written submissions are given different names and have different form
requirements. These include Request or Notice for Arbitration and Answer,
Statements of Case and Defence, Memorials, Written Briefs or Submissions,
counter memorials, reply and rejoinder. They take the place of what are
described as pleadings in common law litigation, but they generally go well
beyond such submissions and that term can be misleading when applied to
arbitration, as some civilian lawyers would consider the term as a reference to
an oral presentation.

There are in fact two broad types of written submissions, the first covering the
articulation of claims, counterclaims, defences and issues and the second, being
attempts to synthesise the legal contentions and emerging evidence, including
oral testimony, into comprehensive documents that present the arguments
underlying the claims. Thus there is a distinction between the required written
submissions to ensure that each party knows what its opponent is contending,
an essential element of an adequate opportunity to present its case and the
second form, being an efficient presentation of all arguments so that the tribunal
can readily note and reconcile the conflicting views. There is of course a degree
of overlap given that arbitration adopts a more continental style, with more
detailed reasoning from the outset and a recommended approach of presenting
claims and relevant documents concurrently. Nevertheless, from the perspective
of the tribunal, the first category is more about due process, while the second is
about advocacy. (282) As to the first page "479" category, there must be concern to
meet the minimum standards of requests or notices of arbitration and answers
or defences to requests. Chapter 3 looked at the requirements for these
preliminary documents which frame the claims and cross-claims.

This section looks at subsequent written submissions. Once the minimum

written standards of notice of claims are met, there is no presumption that
further written submissions are required although that is the norm. In a small
number of instances, there may be no value in any further written submissions,
as the whole case could depend, for example, on which of two witnesses is
believed to be telling the truth. As noted previously, David Rivkin has sensibly
counselled against presuming that traditional stages are always needed, instead
suggesting that the presumption should be to the contrary, only utilising
elements thought to be suitable to the instant dispute.

If a decision is made to utilise written submissions at these later stages, key

issues are timing, as to when submissions should occur, the number of
submissions and the order, in particular whether they are to be consecutive or
simultaneous. The tribunal might also determine whether there should be any
form and length constraints. The number and timing of written submissions will
depend on the complexity of the matter and whether the original claims are
simply to be defended or whether there are counterclaims and/or set-off rights
asserted.
An important rule of thumb is to have whatever can be done in writing presented
in a timely manner and not simply repeated in hearings. (283) Reducing hearing
time will generally reduce cost. Directions should be given as to how written
submissions are to be tendered, including how accompanying documents are to
be presented. It is the norm to ask parties to include the key documents on which
they intend to rely together with written statements. The tribunal should also
consider giving directions on some more prosaic logistical matters such as
common headings, inclusion of tables of contents, use of dividers and
appendices, page size in hard copy and electronic submission for ease of
reference via laptop computers. Many arbitrators now prefer hard copy in
double-sided A5 size, which is both more manageable to carry and to read on
lengthy flights to and from hearings.

A tribunal might also call for a primer on technical issues to properly prepare
itself for a hearing and ensure that time is not wasted on elaboration of complex
concepts. (284)

6.14.2. The Scheduling of Written Submissions

A first question is as to number and whether each party should have the same
number of submissions or instead, whether the common style of submission,
page "480" answer and rejoinder is to be used. Due process and equality
discussed in sections 2.7.5 and 2.7.6 pointed out that there are arguments either
way. From an equality perspective, it seems intuitively logical to give each an
identical number of rounds, although this depends on how one articulates
equality in this circumstance, a matter on which different legal families have
taken varying approaches. The common law has adopted a view at variance with
numerical equality, arguing within the norm of sequential submissions that each
party only needs one response to what it sees for the first time from its
opponent. Hence it is natural for the claimant to present a first submission, a
respondent to present its answer, and then the claimant to have the sole right of
reply. Respondent has answered once to claimant, claimant has answered once
to respondent, in particular as reply rights are limited to new matters raised by
respondent. Nevertheless, Pierre Karrer suggests that in international
arbitration, the norm is to offer a full double exchange unless both are
comfortable with the Anglo-American system where a rejoinder is only offered in
exceptional circumstances. (285) An English court refused to overturn an award
where a claimant had been prevented from having a right of reply. The Court also
held that the norm in international arbitration was to allow for an equal number
of submissions. (286) Some rules still provide for one exchange plus a reply by
claimant. (287)

Another important question is whether to call for sequential submissions or

concurrent submissions. Sequential submissions would call for the claimant to
first prepare its written submissions and the respondent to review these before
compiling its defence submissions. Where a counterclaim or set-off is involved,
the respondent would go first on those issues, allowing the claimant time to
prepare responses. Because considerations of fairness and equality might vary
depending on the circumstances, no blanket rule as to ordering of written
submissions would be appropriate, although it is likely that sequential
submissions are more common. In many circumstances, sequential submissions
will be the natural way to treat parties with equality and allow them a
meaningful opportunity to meet the opponents case. A respondent cannot know
the essence of a claim and how to respond to it without seeing how it is
articulated. The same could be said for defences to counterclaims. Even this is
complicated by comparativist perspectives. Under the common law, the claimant
usually speaks last. Under the civil law, it is usually the respondent that has the
final say.

In some circumstances, however, concurrent submissions would be fairer. For

example, circumstances may arise where the same disputed facts lead to both
claims and counterclaims. Here it would be reasonable to ask each party to
present their submissions at the same time as to why they are entitled to relief
based on those disputed facts. Each party would then be given equivalent time to
respond to the other's primary assertions. Here it would only be numerical
equality that could be justified on the basis of equal rights to reply. In some
cases, even if the page "481" respondent is merely seeking to negate the
claimant's contentions, these may be sufficiently clear from the notice of
arbitration itself and hence simultaneous submissions may remain appropriate.
For example, a simple factual allegation against the quality of the respondent's
goods would be a natural situation where each could be expected to present
their arguments at the same time, provided that adequate particulars of the
alleged defects have been contained in the request or Answer. (288) Simultaneous
submissions may also be appropriate where there is legitimate disagreement as
to who should properly be seen as the claimant. (289)

Simultaneous submissions could be expected to be more timely, although this

alone should not be the basis for adopting this approach. Such an order should
only be made in cases where each party ought truly be able to adequately
present its argument at the same time. In addition, simultaneous submissions
will not always save time as they raise a greater risk that parties argue at cross-
purposes, perhaps addressing different issues, thus requiring further and/or
more detailed replies in due course. (290)

If simultaneous submissions have been ordered, it may be necessary to consider

the possibility that one party will not comply with the deadline as stipulated. It
would be unfair for that party to receive the other party's submissions and have
an opportunity to respond as if sequential service had been ordered. In such
circumstances, the tribunal could direct that submissions must first be made to
the tribunal and an institution if applicable, and that these only be exchanged
between the parties when both have been received.

6.14.3. Word Limits of Submissions

There are fairness and efficiency arguments for and against imposing word limits
on the parties in relation to written submissions. Because an arbitrator has an
obligation to try and promote an efficient process, and because some
inexperienced or avaricious counsel working on a time-costing basis might
present excessive submissions, there can be some value in arbitrators imposing
reasonable limits. However, because inexperienced counsel may take longer to
make their points, particularly in a foreign language, and because there is a
difference in legal culture as to how detailed written submissions tend to be, an
arbitrator should be careful that any constraints imposed do not allow a party to
raise a legitimate challenge that page "482" its ability to prepare its case was
interfered with. Limits should thus err on the side of generosity.

The appropriate trade-off also needs to take into account that written
submissions are in support of the hearings and are not a distinct methodology.
Because written submissions are efficient and can lead to time savings at the
hearing, unduly constraining them may simply lead to greater costs at the
hearing stage. It is also important to consider orders as to submission length in
conjunction with likely orders as to hearings. In setting any limits both for the
length of submissions and the length of the hearing, a tribunal would ideally try
and consider which kind of material can best be presented in the more efficient
written form and which material would best be explored in the oral hearings.

6.14.4. Post-hearing Briefs

While post-hearing submissions would seem unusual to common lawyers, from

the civilian perspective, it can make sense to tie all submissions together in one
document after all stages have been completed. In that regard, the logic goes,
some time is needed to reflect on the transcript and tie the final evidentiary
record to the submissions made from the outset. Post-hearing briefs work best if
they actually shorten the hearing, for example, by obviating the need for closing
submissions and allowing Counsel to reflect on and synthesise transcripts. Post-
hearing briefs might also support equal treatment if late admission of evidence
has been accepted. Post-hearing submissions may also be appropriate where a
tribunal seeks clarification on matters that were not able to be adequately
covered at the hearing.

While the potential value is high, the added cost is significant and there is also
the time factor in terms of delay as to completion of the award. If post-hearing
briefs are too readily accepted, Counsel may be inefficient in preparing for the
hearing on the assumption they will be able to fix problems at a later stage. An
added problem is that the more a busy tribunal is invited to delay its
deliberations until post-hearing briefs are received, the more it may forget
important aspects of the quality of testimony where conflicting oral evidence is
presented. There is also the logistical issue that a multi-person tribunal may not
get another opportunity for face-to-face deliberations. If post-hearing briefs are
allowed, this can constrain tentative deliberations at the hearing as the
arbitrators would not wish to be presenting concrete views as to the result prior
to viewing the impending briefs.

Arbitrators concerned to appear fair and open minded may too readily agree to
post-hearing briefs. These should only be allowed where they will truly assist the
tribunal in resolving the matter. In some cases an oral discussion of key issues at
the conclusion of the hearing with counsel may be undertaken in lieu of post-
hearing submissions. (291) If a tribunal is likely to be reluctant to allow post-
hearing briefs, it page "483" may be preferable to make this clear before the
hearing, with the invitation to apply for such an entitlement in exceptional
circumstances. Conversely, if there is a clear difference in oral experience
between counsel, allowing post-hearing briefs may afford those with less
experience of the oral tradition the opportunity to make comprehensive and
efficient submissions.

If post-hearing briefs are to be allowed, the tribunal should make sure that it has
adequately noted its responses to the oral hearing while at the same time not
prejudging the matter, that it imposes appropriate limits both as to time and
length in post-hearing briefs and indicates whether these should be limited to
additional arguments to be cross referenced with earlier submissions or instead,
should constitute a new document encompassing all previous arguments and
evidence. In the latter event, it may even be useful in some cases to invite
presentation of a tracked change version that highlights the new material. This
would be valuable for arbitrators who are fully familiar with the earlier
submissions and only wish to see what extra points are being made after the
hearing. To simply limit the post-hearing submissions to such information can
still waste time if the arbitrator must then cross reference assertions back to
earlier documentation.

It is generally advisable to get a statement of costs from each party prior to

making a determination on the merits so that a determination can be made in the
award and so that neither party has an incentive to presume its opponent's costs
are excessive. This can often be sought in post-hearing briefs if it had not been
sought before. Unless a tribunal directs and allows it, post-hearing submissions
should not otherwise contain new material but should simply synthesise the
arguments in the light of what transpired at the hearing itself. If a post-hearing
submission violates these principles, the tribunal should promptly indicate what
it intends to do. The tribunal may also wish to list specific questions that it would
particularly like to have addressed. Unfortunately, there may be a reluctance to
employ such an approach for fear that there is a presumption of prejudgment on
the remaining matters. As Rivkin has pointed out, such an accusation cannot
legitimately be made as against an arbitrator who has reviewed all submissions,
witness statements and has sat through oral testimony and oral summations. (292)

The tribunal will also need to determine whether such post-hearing briefs should
be sequential or simultaneous. In most cases, post-hearing briefs would be
simultaneous, as each party is simply collating all previous material. In some
cases, however, a matter may have been touched on at the hearing that was not
appropriately dealt with, with orders allowing one party to make submissions
and the other to respond. In such circumstances, sequential ordering may be
fairer. There is always the chance that one or both parties could call for reply
submissions in either case.

page "484"

6.15. Default Proceedings

6.15.1. Default Proceedings and Arbitration

Default proceedings for the purposes of international arbitration can be defined

as the partial or total failure of a party in a dispute to participate in the
proceedings. (293) A fundamental aspect of default proceedings in arbitration is
that there is nothing akin to summary judgment flowing purely as a result of
non-appearance by a responding party. (294) This flows from the fact that parties
have agreed to resolve their disputes via adjudicated arbitration on the merits
and not via automatic judgments as may be established for other policy reasons
by domestic litigation systems. (295) There is also no evidentiary presumption that
claims are correct in the absence of formal denial, as non-appearance may be due
to a range of reasons. Hence numerous rules and lex arbitri expressly indicate
that a failure to appear or to present a case is not deemed to be an admission.
(296)

Where a tribunal is to proceed in the absence of a party and make a fair and
efficient decision on the merits, there are then important due process questions
that involve at times contentious discretionary determinations. For example,
how often and by what methods should the non-attending party be notified of
developments? Should the tribunal test the evidence of the claimant, including
asking testing questions of witnesses? Should a tribunal raise jurisdictional
concerns sua sponte? Should a transcript be kept as a matter of course so that
there is a clear record in the event of a challenge? Most rules are silent on such
questions and the proper response is more generally an amalgam of efficiency
concerns subject to due process considerations and the aim of promotion of
enforceability.

6.15.2. Rules and Laws as to Default Proceedings

Most laws and rules provide that a tribunal is to proceed in the absence of a
respondent and render an award in due course. (297) Where a claimant is
concerned, page "485" it will usually be the case that if it does not proceed, the
tribunal will conclude that the case has been abandoned. In some such cases, the
respondent will want to proceed at least as to costs, and perhaps to obtain a final
award on the merits for res judicata effect. In either case, the sole proceeding
party may need to advance all of the costs pending an award in that regard. (298)

While the rules clearly express the right to proceed, there is generally little
specific guidance as to how this is to occur. Hence it is left to general discretions,
subject to mandatory due process norms. There can then be debate as to the way
these mandatory norms should apply in the absence of a party. These need to be
considered in the broader context of tribunal proactivity and also, in the context
of potential challenges. Given a tribunal's duty to render an enforceable award,
particular care ought to be taken when one party refuses to attend. In such
circumstances the tribunal is on notice that there is more likelihood of a
challenge to enforcement and perhaps an annulment application. However, due
process is always the key in such cases and it is too easy for a non-appearing
challenging party to overstate the obligation. Due process need only be as to an
opportunity to present. Ultimately, the New York Convention's procedural
requirements will not prevent the continuation of proceedings due to a non-
appearing party as long as a reasonable opportunity was provided. (299)

Article 25 of the UNCITRAL Model Law is a clear example. It provides that if

without showing sufficient cause, the claimant fails to communicate its statement
of claim then the tribunal shall terminate the proceedings. Conversely, if the
respondent fails to communicate the statement of defence the tribunal shall
continue the proceedings without treating the failure as an admission of
claimant's allegations. Article 25(3) applies to both parties, indicating that if any
party fails to appear at a hearing or fails to produce documentary evidence, the
arbitral tribunal may continue the proceedings and make the award on the
evidence before it. Where the claimant chooses not to appear, in some
circumstances the tribunal may consider that the claim is abandoned. This would
be so if the claimant has concurrently failed to provide advances on costs after
repeated requests to do so. Again a respondent could advance the costs if an
award was important to it.

Some rules speak in mandatory language while others provide discretion. The
ICDR Rules simply provide discretion where the respondent fails to
communicate a statement of defence. (300) The ICSID Rules leave it to a party to
make an application to the tribunal to proceed. (301) ICC Rules 2012 do not
provide discretion, instead requiring the arbitration to proceed. (302)

page "486"

6.15.3. Assistance by the Tribunal in Default Proceedings

There are important questions as to what a tribunal should do in default

proceedings. One challenging aspect is that whatever it does, whether adopting a
passive or active role, a casuistic argument can be raised that it has offended one
or other party's due process rights. It is appropriate to consider the various
stages to better discern best practice and the validity of challenges that might
colour behaviour.

6.15.3.1. Notice and Opportunity

Due process normally suggests that a tribunal should not engage in ex parte
communications with a party. Yet this cannot be a principle fully applicable in
default proceedings or they could not take place. If one party fails to attend, the
arbitrator can of course communicate discreetly with the party that is present.
Nevertheless, there should be a clear evidentiary record showing what occurred.
The tribunal must ensure that adequate notice has occurred and further, that it
gives consideration to any reasons the party gives as to why it cannot appear at
that stage. The notification should also be meaningful. The tribunal should notify
the defaulting party regularly as to progress with enough detail of arguments to
allow it to decide if it wants to reappear. The absent party must as a result be
given constant notice of all elements of the proceedings: that is, arguments,
evidence, submissions – as well as meeting times and changes to any scheduling
to ensure that there can be no objections based on proper notice. Grace periods
for non-participating parties, that is ‘sufficient time to mend (ones) ways’, (303)
are also one method of ensuring the non-present party has a real opportunity to
present a case and that the proceeding complies with fairness requirements. Due
process requires an opportunity to present a case. It does not require a reluctant
party to be present. As long as the absent party has been duly notified, it has thus
been afforded an opportunity to present its case. If it chooses not to avail itself of
that right, there is no interference with due process.

Chapter 3 looked generally at the question of notice. Note was taken of a Swedish
case where a court considered that notice to the last known address of
respondent where changes had been notified to corporate authorities but not the
claimant, was not sufficient to support enforcement. While the case is
contentious, it makes good sense that a tribunal take whatever reasonable steps
are available to ensure that actual notice has occurred in the context of default
proceedings. There should be a clear evidentiary record showing that the
unwilling party was given every opportunity to attend and present its case, was
adequately notified of the page "487" timeframe, and was continually invited to
take part and offer its own views as to a fair and reasonable process.

At times the tribunal will need to accommodate legitimate problems facing the
defaulting party via procedural determinations. Due process implications would
vary depending on whether a party chooses not to appear or simply is prevented
from doing so. An example of the latter would be a party who is unable to appear
simply because of a failure to obtain a governmental travel permit. The party is
actually a willing participant but is being prevented from engaging in a particular
step. This is not properly characterised as a default proceeding. Here there is a
difference between technical appearance where written submissions and video
presentations are possible and physical appearance which may be blocked in the
above circumstances. In the latter event if the matter was one that could
reasonably be decided on the documents or via video or phone link then the
party would be appearing fully in any event. If not, the tribunal may need to
explore other options to afford the party a meaningful opportunity.

There may be a range of other challenging procedural requests from time to time
some of which are not truly default issues. A respondent may argue for later
dates for the hearings on the basis of unavailability. A respondent might
reappear near the end of the process and ask the tribunal to rehear or extend the
process. In such circumstances the tribunal can reject this if it can conclude that
a reasonable opportunity had been afforded. In some cases if there is no serious
prejudice to the other party, late attendance may be permitted.

6.15.3.2. Written Records

Default proceedings must take a vigorous approach to demonstrating fairness

and impartiality. Where a party is not present for any reason, in order to allow a
non-participating respondent's full notice of the relevant arguments, it is
appropriate to require these to be committed in writing so that they can be sent.
(304) A tribunal might also call for a transcript even if not sought by either party.
6.15.3.3. Jurisdictional Questions

It is suggested that the tribunal should consider jurisdictional questions itself

whenever there are default proceedings, to ensure validity of the submission to
arbitration. (305) This would logically be so if a tribunal should similarly raise
these issues sua sponte where both parties attend, although that is contentious.
Jurisdictional challenges are considered further in section 8.1.

page "488"

6.15.3.4. Testing the Evidence and Legal Arguments

Continuation in the absence of a party will not result in an automatic award in

favour of the party present. (306) Each party still carries the burden of proving the
elements required for its case. While that is easy to state, there are some
important conceptual questions as to the degree of assistance that should be
provided in such circumstances which are impacted upon by questions of
proactivity generally. For example, what if the claimant has presented a prima
facie case? Should the tribunal still test this or should it conclude that a refusal to
negate a prima facie case is the respondent's own fault and claimant deserves an
award without further delay? Conversely, if the claimant's prima facie case is
inadequate, can the tribunal seek further information, perhaps helping it to make
its case? Even the notion of a prima facie case is itself problematic in arbitration,
being more of a common law procedural construct. In most instances of
international arbitration, a strict analysis of the presence or absence of a prima
facie case does not apply and an arbitrator ought to be entitled to test
submissions without going so far as to advocate contrary positions.

While a tribunal should not advocate for the non-attending party, it should also
be mindful that the party present may be overstating submissions on law or fact
and may be selective in concentrating on favourable elements. In contested
proceedings with a competent opponent, such matters will typically be pointed
out by the latter without any effort on the tribunal's part. Absent an opposing
party, a tribunal may reasonably feel that it should itself test whether it is being
given a comprehensive and accurate picture.

6.15.3.5. Can an Arbitrator Call for Further Evidence of behalf of the Non-
present Party?

Wetter has argued that a tribunal should not simply limit itself to identifying
whether a prima facie case was presented by a claimant. The duty to achieve
justice in relation to legitimate expectations, coupled with the res judicata effect
of an award, led him to argue that tribunals may request additional
documentation and legal argument from a claimant. (307) A tribunal ought to be
entitled to engage in its own inquiries and research to make a proper
determination on the merits, although this might well be limited to matters that
would at first sight appear to be tenable defences and criticisms of the claimant's
logic. In BP v. Libya the tribunal considered that it was appropriate to undertake
‘an independent examination of the legal issues deemed relevant by it, and to
engage in considerable legal research page "489" going beyond the confines of
the materials relied upon by the Claimant.’ (308) In some cases this is clarified by
the rules. (309) Nevertheless, arbitrators, should operate with full disclosure. Any
independent findings should be presented to both parties for assessment and
critique. (310)

These questions relate in part, to the question of the duty to know and apply the
law and the power to consider legal matters not argued. In Bogdanov v. Moldova
(311) a BIT arbitration conducted under SCC Rules saw the tribunal apply the

principle of iura novit curia subject to the duty not to take the parties by surprise.
This is discussed further in section 13.19. In Goetz there was a quote from the ICJ
to the effect that: [t]he vigilance which the court can exercise when aided by the
presence of both parties to the proceedings has a counterpart in the care it has to
devote to the proper administration of justice in a case in which only one party is
present. (312)

6.15.4. Enforcement and Due Process

As noted, a party who chooses not to attend will often seek to block enforcement
of the award. Non-appearance is not a waiver of a right to seek to annul or
challenge enforcement. A concern to maximise communication with that party,
give it every opportunity and still thoughtfully evaluate the claims as presented if
they remain an unwilling participant, can only aid in enforceability if and when it
is tested. As noted, in terms of promoting enforceability, it also makes sense to
have as much of the proceeding as possible conducted in writing, with timely
presentation of material to the non-responding party so that the tribunal's
activities are clear and the merits issues facing the non-represented party and its
opportunity for involvement are obvious.

While it is appropriate to ignore unmeritorious threats as to enforcement

challenges, some attention to the practices of a likely enforcement court could be
appropriate in such circumstances. Thus an arbitrator might look to decisions
from such a jurisdiction to see if a blueprint can be identified as to how to
conduct arbitrations in such circumstances to maximise enforceability. (313) In the
end a tribunal must still simply behave reasonably. If an enforcement norm that
is unique to a particular country goes against what would otherwise be best
practice, it can be page "490" ignored. If the enforcement norm is part of best
practice, it should be respected for the latter reason in any event.

6.16. Online Arbitration

There are two distinct questions to consider. The first relates to arbitrations
centred around online contracts. A great deal of business is done online.
Unavoidably, disputes arise from this business, and when they do they tend to
present several characteristic issues. First, the parties to an online transaction
(which could be as simple as a consumer purchase from an e-retailer) may be in
different countries. Second, the value of the transaction may be quite low.
Whereas a dispute of this kind might be dealt with, for example, through a
regulator if both parties were in the same country, the international element of
these transactions may require that an aggrieved party commence court
proceedings in the other party's country if it hopes to receive any redress. Of
course, the frequently relatively low value makes this course of action untenable.
In any case, the international element of online transactions, and the fact that the
nature of e-commerce would make it hard for transacting parties to foresee the
countries that would be relevant to a future dispute, makes arbitration a
particularly suitable method of dispute resolution. (314) This book is not dealing
with consumer transactions that may be resolved via arbitration. Where
business to business (B2B) international disputes are concerned, special
questions would include the form requirement for arbitration clauses and the
connecting factors where applicable law is to be determined.

The second aspect deals with attempts to conduct international arbitrations

online. Online arbitration describes a process of dispute resolution that takes
place entirely (315) online. This could arise from an online contract but may be a
separately selected process, although to date, the process tends to be applied
mainly to disputes that relate to online activity, or that involve contracts entered
into online and which contain arbitration agreements or disputes that relate to
the registration of domain names. Where online arbitration is used with more
traditional contracts, this will generally relate to relatively uncomplicated
subject matter and involve lower-value claims, and accordingly the benefits of
swift and cheap dispute resolution may be seen as outweighing the procedural
and analytical compromises that accompany online arbitration, such as a general
lack of page "491" hearings. (316) Nevertheless, there is nothing to stop the use of
online arbitration in more significant matters in similar circumstances where
documents-only hearings are appropriate.

6.16.1. Procedures

Given that online arbitration tends to deal with small disputes in an expedited
and somewhat standardised way and at very low cost, it is typically
institutionalised. (317) In the electronic sphere it does not always purport to be
binding, leading to questions as to whether it would then be arbitration at all.
(318) Again this book is only concerned with procedures for binding international

arbitration.

The particular rules for the proceedings need to deal with time-lines and
evidence. An example of a system that is not intended to deal solely with e-
commerce disputes (i.e., it is designed for use in more traditional sale of goods
transactions, albeit within a limited scope of transaction types) is the ‘ICDR
Protocol for Manufacturer/Supplier Disputes' developed by the American
Arbitration Association through its International Centre for Dispute Resolution
(ICDR), and which is intended for use by manufacturers in their contracts with
suppliers. A trial version of the model piloted by General Electric in Italy results
in binding awards and caps the value of claims at USD 10,000. The rules require
that, following electronic lodgement of a claim, the respondent has twelve days
in which to file a defence. The parties then attempt negotiation, and if no
settlement is reached within a further twelve days, arbitration commences under
the direction of a single arbitrator appointed by the ICDR. The arbitrator
considers the substance of the dispute ‘based upon the documents submitted by
the parties' and only in ‘extraordinary circumstances [may] the arbitrator may
request additional evidence from the parties'. (319) The arbitrator must deliver an
award within thirty days of appointment. The rules also include provisions
relating to the substantive law and the aspects that must be addressed in the
award. The rules do not provide for any hearings, and communications are to be
conducted by email unless otherwise agreed by the parties or directed by the
arbitrator.

page "492"

While it may achieve its goals of speed and economy, critics would assert that the
pursuit of these same factors necessarily compromise the online arbitral
tribunal's opportunity for rigorous substantive analysis and tend to require
procedural rules that somewhat curtail some aspects of due process. (320)
However, if such expedited proceedings were not available, the costs of
traditional arbitration or litigation mean that an aggrieved party with a low-
value claim might have no effective opportunity for redress at all. The key issue
to consider in framing online arbitration is the balance between pursuing
procedural fairness (with its associated costs) and the need for time and costs to
be proportional to the magnitude of the dispute. In the case of an e-commerce
consumer dispute with a value of a few hundred dollars or euros, financial
practicalities may mean that there are only ‘two choices: no justice or rough
justice, the sort of rough justice found in simplified arbitration procedures and in
small claims courts' (321) intended to ‘avoid crass disrespect of the contract or
basic legal obligations in a consumer transaction’ (322) rather than to engage in a
deeper, and more costly (if less legally thorough) arbitral process. As Julia Hörnle
observes, it ‘has to be recognised that online arbitration that complies with due
process standards will only be proportionate for certain disputes’. (323) In the case
of B2B disputes, frameworks such as the ICDR Protocol described above provide
a greater degree of party participation and legal analysis, as this makes practical
sense in the context of the increased value of the dispute and financial resources
of the parties.

page "493"

1 See, e.g., V.V. Veeder, ‘Whose Arbitration Is It Anyway: The Parties or the
Arbitration Tribunal – An Interesting Question?’, in The Leading Arbitrators'
Guide to International Arbitration, 2nd edn, ed. Lawrence W. Newman & Richard
D. Hill (New York: Juris Publishing, Inc., 2008), 337–358.
2 A tribunal may also wish to prevent inappropriate procedural agreements

between the parties.

3 Michael Pryles, ‘Limits to Party Autonomy in Arbitral Procedure’, Journal of

International Arbitration 24, no. 3 (2007): 334.

4 See the ‘Exposción de Motivos’ in the Spanish Arbitration Act 2003.
5 Edok SA and Others v. Hydromechaniki Sàrl and Eupalinaos SA, 10 May 1982,
ATF/BGE108 Ia 197 at 201, translation provided by Paulo Michele Patocchi &
Harold Frey-Brentano, ‘The Provisional Timetable in International Arbitration’,
in Global Reflections on International Law, Commerce and Dispute Resolution:
Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al. (Paris: ICC
Publishing, 2005), 576.
6 See the following section dealing with the parties' rights to vary aspects of

institutional arbitration.
7 The power for anyone to modify procedures after tribunal appointment may

also be further circumscribed in cases where Terms of Reference are used. For
example, Art. 23 of the ICC Rules 2012 calls for particulars of the applicable
procedural rules to be identified in the terms and for the tribunal to set out a
provisional timetable. Yet Art. 24(4) contemplates that modifications may be
made.
8 ICDR Rules Art. 1.1(a).
9 Yves Derains & Eric A. Schwartz, A Guide to the ICC Rules of Arbitration, 2nd edn

(The Hague: Kluwer Law International, 2005), 388.

10 While Art. 18 of the Model Law is not expressed to be mandatory, it is

regularly accepted as such. Howard M. Holtzmann & Joseph E. Neuhaus, A Guide

to the UNCITRAL Model Law on International Commercial Arbitration: Legislative
History and Commentary (Deventer: Kluwer Law and Taxation Publishers, 1989),
583; Michael Pryles, ‘Limits to Party Autonomy in Arbitral Procedure’, Journal of
International Arbitration 24, no. 3 (2007): 329.
11 Howard M. Holtzmann & Joseph E. Neuhaus, A Guide to the UNCITRAL Model

Law on International Commercial Arbitration: Legislative History and

Commentary (Deventer: Kluwer Law and Taxation Publishers, 1989), 583. See
also Michael Pryles, ‘Limits to Party Autonomy in Arbitral Procedure’, Journal of
International Arbitration 24, no. 3 (2007): 332.
12 Michael Pryles, ‘Limits to Party Autonomy in Arbitral Procedure’, Journal of

International Arbitration 24, no. 3 (2007): 333.

13 Philip Yang, ‘The Organisation of International Arbitration Proceedings’, in

The Asian Leading Arbitrator's Guide to International Arbitration, ed. Michael

Pryles & Michael J. Moser (New York: JurisNet, 2007), 219.
14 Michael Pryles, ‘Limits to Party Autonomy in Arbitral Procedure’, Journal of

International Arbitration 24, no. 3 (2007): 333.

15 Ibid.; Eric A. Schwartz, ‘The Rights and Duties of ICC Arbitrators’, in The Status

of the Arbitrator, ICC International Court of Arbitration Bulletin: 1995 Special

Supplement, ed. Jean-François Bourque (Paris: ICC Publishing, 1995), 89.
16 Even that is a potentially simplistic proposition. For example, if parties agree

on a documents-only arbitration in a Model Law Seat and appointment is

accepted, the parties could assert that the arbitrator has accepted the ongoing
role of Art. 19 of the UNCITRAL Model Law which on plain meaning, binds the
tribunal to subsequent procedural agreements by the parties, including the right
to demand a hearing. Such an argument would be even stronger if rules were
agreed to which give each party the unilateral right to call for a hearing. The
converse argument that the specific agreement prevails over the general
discretionary power and that the specific agreement itself varies Art. 19 as a
non-mandatory provision may be preferable, but at least some uncertainty
remains. If the parties are clear enough in their intentions, they can circumscribe
their powers under provisions such as Art. 19 as it is not a mandatory provision.
17 See, e.g., Serge Lazareff, ‘Arbitrators and Arbitral Procedures: Summing Up’, in
Arbitration in the Next Decade: ICC International Court of Arbitration Bulletin:
1999 Special Supplement, ed. Robert Briner (Paris: ICC Publishing, 1999), 73. See
also Lucy Reed & Johnathan Sutcliffe, ‘The “Americanisation” of International
Arbitration?’, Mealey's International Arbitration Report 16 (2001): 5–34.
18 James H. Carter, ‘The Rights and Duties of the Arbitrator: Six Aspects of the

Rule of Reasonableness’, in The Status of the Arbitrator, ICC International Court of

Arbitration Bulletin: 1995 Special Supplement, ed. Jean-François Bourque (Paris:
ICC Publishing, 1995), 32.
19 Robert Reiner cited in James H. Carter, ‘Rights and Obligations of the
Arbitrator’, Dispute Resolution Journal 52, no. 1 (1997): 66, n. 18.
20 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 315.

21 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1652.

22 Gerold Herrmann, ‘The Arbitration Agreement as the Foundation of
Arbitration and Its Recognition by the Courts’, in International Arbitration in a
Changing World, ICCA Congress Series No. 6, ed. Albert Jan van den Berg (The
Hague: Kluwer Law International, 1994), 42.
23 Ibid.
24 Christopher Newmark, ‘Controlling Time and Costs in Arbitration’, in The

Leading Arbitrators' Guide to International Arbitration, ed. Lawrence W. Newman

& Richard D. Hill (Huntington: Juris Pub, 2008), 85.
25 Paolo Michele Patocchi & Harold Frey-Brentano, ‘The Provisional Timetable in

International Arbitration’, in Global Reflections on International Law, Commerce

and Dispute Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald
Aksen et al. (Paris: ICC Publishing, 2005), 591.
26 Michael Pryles, ‘Limits to Party Autonomy in Arbitral Procedure’, Journal of

International Arbitration 24, no. 3 (2007): 337.

27 See Queen Mary School of International Arbitration, ‘International Arbitration:

Corporate Attitudes and Practices’,

<www.arbitrationonline.org/docs/IAstudy_2008.pdf>, 2008.
28 Rolf Stürner, ‘The Principles of Transnational Civil Procedure: An Introduction

to Their Basic Conceptions’, RabelsZ 69, no. 2 (2005): 228.

29 Paolo Michele Patocchi & Harold Frey-Brentano, ‘The Provisional Timetable in

International Arbitration’, in Global Reflections on International Law, Commerce

and Dispute Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald
Aksen et al. (Paris: ICC Publishing, 2005), 579.
30 Michael E. Schneider, ‘Combining Arbitration with Conciliation’, in Planning

Efficient Arbitration Proceedings/The Law Applicable in International Arbitration,

ICCA Congress Series No. 7, ed. Albert Jan van den Berg (The Hague:Kluwer Law
International, 1996), 60–61.
31 J. Gillis Wetter, ‘Procedures for Avoiding Unexpected Legal Issues’, in Planning

Efficient Arbitration Proceedings/The Law Applicable in International Arbitration,

ICCA Congress Series No. 7, ed. Albert Jan van den Berg (The Hague: Kluwer Law
International, 1996), 94–95.
32 Judith Gill, ‘Applications for the Early Disposition of Claims in Arbitration

Proceedings’, in Fifty Years of the New York Convention, ICCA International

Arbitration Conference, ICCA Congress Series No. 14 (Dublin 2009), ed. Albert
Jan van den Berg (Kluwer Law International, 2009), 521.
33 Karl-Heinz Böckstiegel, ‘Case Management by Arbitrators: Experiences and
Suggestions’, in Global Reflections on International Law, Commerce and Dispute
Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al.
(Paris: ICC Publishing, 2005), 116.
34 Ibid., 117.
35 ICC, ‘Techniques for Controlling Time and Costs in Arbitration’,
<www.iccwbo.org/uploadedFiles/TimeCost_E.pdf>, 2007, para. 33.
36 Allan Philip, ‘The Duties of an Arbitrator’, in The Leading Arbitrators' Guide to

International Arbitration, 2nd edn, ed. L. Newman & R. Hill (Huntington: Juris
Publishing, 2008), 79.
37 David W. Rivkin, ‘Towards a New Paradigm in International Arbitration: The

Town Elder Model Revisited’, Arbitration International 24, no. 3 (2008): 383.
38 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 464.

39 David W. Rivkin, ‘Towards a New Paradigm in International Arbitration: The

Town Elder Model Revisited’, Arbitration International 24, no. 3 (2008): 381.
40 See Lucy Reed, ‘More on Corporate Criticism of International Arbitration’,

<http://kluwerarbitrationblog.com/blog/2010/07/15/m...>, 16 July 2010.

41 Gerald Aksen, ‘Arbitrability of Disputes’, in Global Reflections on International

Law, Commerce and Dispute Resolution: Liber Amicorum in Honour of Robert

Briner, ed. Gerald Aksen et al. (Paris: ICC Publishing, 2005), 17.
42 Ibid., 19.
43 R C Pillar & Sons v. Edwards and Another [2001] All E.R. (D) 232.
44 David W. Rivkin, ‘Towards a New Paradigm in International Arbitration: The

Town Elder Model Revisited’, Arbitration International 24, no. 3 (2008): 375.
45 Ibid., 377–378.
46 Jean-Claude Najar, ‘Inside Out: A User's View of International Arbitration’,

Speech delivered at Clayton Utz and the University of Sydney International

Commercial Arbitration Lecture, Sydney, 6 November 2008, quoted in Professor
Doug Jones, ‘International Dispute Resolution in the Global Financial Crisis’, The
Arbitrator and Mediator (October 2009): 45.
47 Eric A. Schwartz, ‘The Rights and Duties of ICC Arbitrators’, in The Status of the

Arbitrator, ICC International Court of Arbitration Bulletin: 1995 Special

Supplement, ed. Jean-François Bourque (Paris: ICC Publishing, 1995), 77.
48 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 86.

49 Consider for example an arbitrator who is a partner in a law firm whose firm

merges with a firm representing one of the parties, some years into the process.
50 See Hong Kong Arbitration Ordinance (Cap 609), s. 59.
51 Michael E. Schneider, ‘Combining Arbitration with Conciliation’, in Planning

Efficient Arbitration Proceedings/The Law Applicable in International Arbitration,

ICCA Congress Series No. 7, ed. Albert Jan van den Berg (The Hague: Kluwer Law
International, 1996), 75.
52 Christopher Newmark, ‘Controlling Time and Costs in Arbitration’, in The

Leading Arbitrators' Guide to International Arbitration, ed. Lawrence W. Newman

& Richard D. Hill (Huntington: Juris Pub, 2008), 86.
53 V.V. Veeder, ‘Whose Arbitration Is It Anyway: The Parties or the Arbitral

Tribunal?’, in The Leading Arbitrators' Guide to International Arbitration, ed.

Lawrence W. Newman & Richard D. Hill (New York: Juris Publishing, Inc., 2004),
347.
54 Fabian von Schlabrendorff & Audley Sheppard, ‘Conflict of Legal Privileges in

International Arbitration: An Attempt to Find a Holistic Solution’, in Global

Reflections on International Law, Commerce and Dispute Resolution: Liber
Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al. (Paris: ICC
Publishing, 2005), 767.
55 J. Gillis Wetter, ‘Procedures for Avoiding Unexpected Legal Issues’, in Planning

Efficient Arbitration Proceedings/The Law Applicable in International Arbitration,

ICCA Congress Series No. 7, ed. Albert Jan van den Berg (The Hague: Kluwer Law
International, 1996), 92.
56 Marc Blessing, ‘The ICC Arbitral Process – Part III: The Procedure Before the

Arbitral Tribunals’, ICC International Court of Arbitration Bulletin 3, no. 2 (1992):

43.
57 Michael E Schneider, ‘Combining Arbitration with Conciliation’, in Planning

Efficient Arbitration Proceedings/The Law Applicable in International Arbitration,

ICCA Congress Series No. 7, ed. Albert Jan van den Berg (The Hague: Kluwer Law
International, 1996), 60–61.
58 Michel A. Calvo, ‘The Appointments, Duties and Rights of The ICC Arbitrators

(Revisited under the New ICC Rules)’, International Business Law Journal 3, no. 3
(1999): 373.
59 Karl-Heinz Böckstiegel, ‘The Role of the Arbitrators in Investment Treaty

Arbitration', in International Commercial Arbitration: Important Contemporary

Questions, ICCA Congress Series No. 11, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 2003), 366.
60 Ibid., 367.
61 Michael E Schneider, ‘Combining Arbitration with Conciliation’, in Planning

Efficient Arbitration Proceedings/The Law Applicable in International Arbitration,

ICCA Congress Series No. 7, ed. Albert Jan van den Berg (The Hague:Kluwer Law
International, 1996), 60.
62 Ibid., 75.
63 Piero Bernardini, ‘The Role of the International Arbitrator’, Arbitration
International 20, no. 2 (2004): 115.
64 Hilmar Raeschke-Kessler, ‘The Arbitrator as Settlement Facilitator’,
Arbitration International 21, no. 4 (2005): 529.
65 ICC Report of Working Group on Criminal Law and Arbitration (Doc.
420/492).
66 The view is supported by Cremades and Cairns. See Bernardo M. Cremades &

David J.A. Cairns, ‘Corruption, International Public Policy and the Duties of
Arbitrators’, Dispute Resolution Journal 58, no. 4 (2004): 82.
67 Bernardo M. Cremades & David J.A. Cairns, ‘Corruption, International Public

Policy and the Duties of Arbitrators’, Dispute Resolution Journal 58, no. 4 (2004):
84.
68 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1627.

69 See section 13.8.
70 Legal privilege is considered in section 10.17.10.
71 See A. Mourre, ‘Arbitration in Criminal Law’, Arbitration International 22, no. 1

(2006): 95; D Hilder & V. Pavic, ‘Arbitration and Crime’, Journal of International
Arbitration 25, no. 4 (2008): 461; M. Wirth, ‘Production of Documents and Fraud
in International Arbitration’, in Written Evidence and Discovery in International
Arbitration, ed. P. Giovannini & A. Mourre (Paris: ICC Publication, 2009), 177; B.
Cremades & D. Cairns, ‘Transnational Public Policy in International Arbitral
Decision-Making: The Cases of Bribery, Money Laundering and Fraud’, in
Arbitration, Money Laundering and Fraud, ed. Kristine Karsten & Andrew
Berkeley (Paris: ICC Publication, 2003), 65.
72 The Vaneira case [1984] 2 Lloyd's Rep 66, 76 per Ackner LJ.
73 Pacol Ltd v. Joint Stock Company Rossakhar [2000] 1 Lloyd's Rep 109.
74 G. Petrochilos, Procedural Law in International Arbitration (Oxford: Oxford

University Press, 2004), para. 4.89.

75 Lisa Spagnolo, ‘Green Eggs and Ham: The CISG, Path Dependence, and the

Behavioural Economics of Lawyers' Choices of Law in International Sales

Contracts', Journal of Private International Law 6, no. 2 (2010): 417–464.
76 See generally, L. Yves Fortier, ‘The Minimum Requirements of Due Process in

Taking Measures Against Dilatory Tactics: Arbitral Discretion in International

Commercial Arbitration – “A Few Plain Rules and a Few Strong Instincts’”, in
Improving the Efficiency of Arbitration and Awards: 40 Years of Application of the
New York Convention, ICCA Congress Series No. 9, ed. Albert Jan van den Berg
(The Hague: Kluwer Law International, 1999), 396–409.
77 Section 41 of the English Arbitration Act 1996 contains detailed timing

provisions. In other jurisdictions, there are conceptual questions as to whether

there are inherent limits on such powers. See Patocchi and Brentano for
comment on the situation in Italy where judicial powers are more limited in that
regard. Nevertheless, the authors argue that the situation should be seen
differently where arbitration is concerned. Paolo Michele Patocchi & Harold
Frey-Brentano, ‘The Provisional Timetable in International Arbitration’, in Global
Reflections on International Law, Commerce and Dispute Resolution: Liber
Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al. (Paris: ICC
Publishing, 2005), 63.
78 Swiss Rules 2012 Art. 2.
79 ICC Rules 2012 Art. 30.
80 ICC Rules 2012 Art. 27.
81 CIETAC Rules 2012 Art. 46.1.
82 French New Code of Civil Procedure Art. 1463; Italian Code of Civil Procedure

Art. 820; Spanish Arbitration Act 2003 Art. 37(2).

83 Swedish Arbitration Act s. 17; Dutch Code of Civil Procedure Art. 1031(2);

English Arbitration Act 1996 s. 24(1)(d)(ii).

84 ICC Rules 2012 Art. 30.
85 Martin Hunter & Jan Paulsson, ‘A Code of Ethics for Arbitrators in
International Commercial Arbitration?’, International Business Lawyer 13, no. 4
(1985): 157–158.
86 English Arbitration Act 1996 s. 1(a).
87 See, e.g., Swiss Rules 2012 Art. 42.1(d).
88 551 F. 2d 1026, 1033 (5th Cir 1977).
89 K/S Norjarl AS/S v. Hyundai Heavy Industries Co. Ltd [1992] QB 863, 878

(Leggatt LJ) and 884 (Stuart Smith LJ).

90 See, e.g., UNCITRAL Model Law Art. 14; English Arbitration Act 1996 ss

33(1)(b) and 24(2); Swedish Arbitration Act s. 17.

91 Jean-François Poudret & Sébastien Besson, Comparative Law of International
Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 380.
92 Judgment of 22 September 1995, Dubois & Vanderwalle v. Boots Srites BV, Cour

d' appel de Paris, Revue de l'Arbitrage (1996): 100. The court considered that the
tribunal violated international public policy by taking a power which they did
not have, contrary to Art. 1502(5) of French NCCP.
93 Howard M. Holtzmann, ‘Delays in International Arbitration – Role of the

Arbitrator’, International Business Lawyer 14, no. 4 (1986): 120.

94 See, e.g., ICC Rules 2012 Art. 30(2).
95 See the earlier position taken in Australia in Oil Basins Ltd v. BHP Billiton Ltd

(2007) 18 VR 346 and Gordian Runoff Ltd v. Westport Insurance Corporation

[2010] NSWCA 57, but note the more recent decisions in Thoroughvision Pty Ltd
v. Sky Channel Pty Ltd & Anor [2010] VSC 139 and Northbuild Construction Pty Ltd
v. Discovery Beach Project Pty Ltd [2010] QSC 94. See also the recent High Court
of Australia decision in Westport Insurance Corporation v. Gordian Runoff Ltd
[2011] HCA 37. This is discussed further in section 16.9.
96 A related issue is whether an arbitrator who accepts an appointment under

rules with award time limits and who has no legitimate excuse for failing to meet
the deadlines could be liable to the parties. Most arbitrators would ensure that
there were appropriate indemnities and waivers in their direct contractual
agreements with the parties. Often, however, arbitrators do not generate
separate express contractual agreements with the parties and arguments as to
implied indemnities might ensue.
97 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 380–381.

98 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 133.

99 Interpretation of arbitration agreements is considered generally in sections

3.2.3–4.
100 Belgian Judicial Code Art. 1698(2).
101 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 379–380.

102 See Howard M. Holtzmann & Joseph E. Neuhaus, A Guide to the UNCITRAL

Model Law on International Commercial Arbitration: Legislative History and

Commentary (Deventer: Kluwer Law and Taxation Publishers, 1989), 422.
103 L. Yves Fortier, ‘The Minimum Requirements of Due Process in Taking

Measures against Dilatory Tactics: Arbitral Discretion in International

Commercial Arbitration: “A Few Plain Rules and a Few Strong Instincts“‘, in
Improving the Efficiency of Arbitration Agreements and Awards: 40 Years of
Application of the New York Convention, ICCA Congress Series No. 9, ed. Albert Jan
van den Berg (The Hague: Kluwer Law International, 1999), 403.
104 See the cases referred to in Paolo Michele Patocchi & Harold Frey-Brentano,

‘The Provisional Timetable in International Arbitration’, in Global Reflections on

International Law, Commerce and Dispute Resolution: Liber Amicorum in Honour
of Robert Briner, ed. Gerald Aksen et al. (Paris: ICC Publishing, 2005), 585.
105 Paolo Michele Patocchi & Harold Frey-Brentano, ‘The Provisional Timetable

in International Arbitration’, in Global Reflections on International Law,

Commerce and Dispute Resolution: Liber Amicorum in Honour of Robert Briner, ed.
Gerald Aksen et al. (Paris: ICC Publishing, 2005), 585.
106 English law expressly gives the tribunal power to enforce a time limit. See
English Arbitration Act 1996 s. 41.
107 Paolo Michele Patocchi & Harold Frey-Brentano, ‘The Provisional Timetable

in International Arbitration’, in Global Reflections on International Law,

Commerce and Dispute Resolution: Liber Amicorum in Honour of Robert Briner, ed.
Gerald Aksen et al. (Paris: ICC Publishing, 2005), 584.
108 CPR Institute, International Dispute Negotiations (IDN) Podcast 66, ‘As Long

As a Piece of String: Timing Expectations in International Arbitration’ (March

2009) cited in Peter Morton, ‘Can a World Exist Where Expedited Arbitration
Becomes the Default Procedure?’, Arbitration International 26, no. 1 (2010): 103,
104.
109 Stephen R. Bond et al., ‘ICC Rules of Arbitration, Miscellaneous, Article 32

[Modified Time Limits]’, in Concise International Arbitration, ed. Loukas A.

Mistelis (The Hague: Kluwer Law International, 2010), 386.
110 For example, the ‘Short-Form Arbitration Rules' of the Hong Kong
International Arbitration Centre, which set out specific timelines for the
submission of various documents and for the delivery of the award within one
month of the last submission (whether oral or written).
111 See, e.g., ICC Rules 2012 Art. 38. Also, LCIA Rules Art. 9 provides for
expedited establishment of the tribunal when a party applies for this in
circumstances of ‘exceptional urgency’.
112 Nigel Blackaby et al., Redfern and Hunter on International Arbitration (Oxford:

Oxford University Press, 2009), 434.

113 See, e.g., LCIA Rules.
114 As set out at Swiss Rules 2012 Art. 42.
115 Swiss Rules 2012 Art. 42.2(c).
116 SIAC Rules Art. 5.1.
117 HKIAC Rules Art. 38.2(c).
118 HKIAC Rules Art. 5.2(c).
119 CPR Rules Art. 1.5.
120 CPR Rules Art. 1.5.1.
121 CPR Rules Art. 1.6.
122 CPR Rules Art. 9.1.
123 CPR Rules Art. 13.1.
124 Welser and Klausegger note that ‘denial of the opportunity to comment on

expert written reports or oral statements’ has been held by State courts to
constitute a violation of due process; however, ‘the implicit limitation of the
deadline for submissions through the arbitrator's announcement that a decision
was to be made by a certain date and based on the documents made available up
until that date’ has been held by courts not to violate due process requirements;
see generally Irene Welser & Christian Klausegger, ‘Chapter II: The Arbitrator
and the Arbitration Procedure – Fast Track Arbitration: Just Fast or Something
Different?’, in Austrian Arbitration Yearbook 2009, ed. Christian Klausegger et al.
(München: C.H. Beck; Bern: Stämpfli & Manz, 2009), 269.
125 Ibid., 270.
126 Stephen R. Bond et al., ‘ICC Rules of Arbitration, Miscellaneous, Article 32

[Modified Time Limits]’, in Concise International Arbitration, ed. Loukas A.

Mistelis (The Hague: Kluwer Law International, 2010), 386.
127 Irene Welser & Christian Klausegger, ‘Chapter II: The Arbitration and The
Arbitration Procedure – Fast Track Arbitration: Just Fast or Something
Different?’, in Austrian Arbitration Yearbook 2009, ed. Christian Klausegger et al.
(München: C.H. Beck; Bern: Stämpfli & Manz, 2009), 267.
128 Pierre Lalive, ‘Arbitration – The Civilized Solution’, ASA Bulletin 16, no. 3

(1998): 484.
129 Wolfgang Peter, ‘Witness Conferencing’, Arbitration International 18, no. 1

(2002): 47.
130 See UNCITRAL Secretariat commentary to Art. 19 of the Model Law cited in

Howard M. Holtzmann & Joseph E. Neuhaus, A Guide to the UNCITRAL Model Law
on International Commercial Arbitration: Legislative History and Commentary
(Deventer: Kluwer Law and Taxation Publishers, 1989), 584.
131 William W. Park, ‘Arbitration's Protean Nature: The Value of Rules and the

Risks of Discretion’, Mealey's International Arbitration Report 19, no. 5 (2004): 3;

See also William W Park, ‘The 2002 Freshfields Lecture: Arbitration's Protean
Nature: The Value of Rules and the Risks of Discretion’, Arbitration International
19, no. 3 (2003): 279.
132 William W. Park, ‘Arbitration's Protean Nature: The Value of Rules and the

Risks of Discretion’, Mealey's International Arbitration Report 19, no. 5 (2004): 9.

133 See, e.g., Nathalie Voser, ‘Harmonisation by Promulgating Rules of Best

International Practice in International Arbitration’, SchiedsVZ 3 (2005): 113–118.

134 William W. Park, ‘Arbitration's Protean Nature: The Value of Rules and the

Risks of Discretion’, Mealey's International Arbitration Report 19, no. 5 (2004): 7.

135 Ibid., 10.
136 See Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th

edn (Oxford: Oxford University Press, 2009), 363–365.

137 E. David & B. Tavender, ‘Before and During Hearings, Here Is How to Manage

Arbitration Matters before They Get Out of Hand’, Alternatives to High Cost of
Litigation 28, no. 7 (2010): 145.
138 Ibid.
139 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 464.

140 Roberto Ceccon, ‘UNCITRAL Notes on Organizing Arbitral Proceedings and

the Conduct of Evidence: A New Approach to International Arbitration’, Journal

of International Arbitration 14, no. 2 (1997): 67.
141 Arthur Marriott, ‘Arbitrators and Settlement’, in New Horizons in
International Commercial Arbitration and Beyond, ICCA Congress Series No. 12,
ed. Albert Jan van den Berg (The Hague: Kluwer Law International, 2005), 533.
142 Claude Reymond, ‘The Distribution of Responsibilities between the Parties

and the Arbitral Tribunal’, in Alan Redfern et al., ‘The Standards and Burden of
Proof in International Arbitration’, Arbitration International 10, no. 3 (1994):
324.
143 A number of commercial publications have also sought to assist with
checklists. See, e.g., Grant Hanessian and Lawrence Newman, International
Arbitration Checklists, 2nd edn, (Huntington, NY: JurisNet, 2009). Individual
arbitrators have also allowed some of their personal checklists to be widely
quoted and hence become influential. One example is Martin Hunter's
questionnaire.
144 ICC, ‘Techniques for Controlling Time and Costs in Arbitration’,
<www.iccwbo.org/uploadedFiles/TimeCost_E.pdf>, 2007.
145 ASA, ‘ASA Special Series No. 31: International Arbitration Do's and Don'ts',
6th edn <www.arbitration-ch.org/publications/>, June 2009.
146 (c) Debevoise & Plimpton LLP 2010.
147 See, e.g., UNCITRAL Model Law Art. 11(3)(a); English Arbitration Act 1996 s.

15; French New Code of Civil Procedure Art. 1513; Swedish Arbitration Act s. 20;
Spanish Arbitration Act 2003 Art. 15; Belgian Judicial Code Art. 1685; Dutch
Code of Civil Procedure Art. 1026.
148 See, e.g., ICC Case No. 4873, cited in Sigvard Jarvin, ‘The Sources and Limits of

the Arbitrator's Powers’, Arbitration International 2, no. 2 (1986): 146.

149 Philip Yang, ‘The Organisation of International Arbitration Proceedings’, in

The Asian Leading Arbitrators' Guide to International Arbitration, ed. Michael

Pryles & Michael Moser (Huntington: JurisNet LLC, 2007), 176.
150 Neil Kaplan & Karen Mills, ‘The Role of the Chair in International Commercial

Arbitration’, in The Asian Leading Arbitrators' Guide to International Arbitration,

ed. Michael Pryles & Michael Moser (Huntington: JurisNet LLC, 2007), 129–130.
151 On the distinction between orders and awards see further section 16.2.
152 See ICC Rules 2012 Art. 31(1); HKIAC Rules Art. 29.1; LCIA Rules Art. 26.3;

SCC Rules Art. 35(1); SIAC Rules Art. 28.5; Swiss Rules 2012 Art. 31.1. The
UNCITRAL Rules 2010 Art. 33.2 only allows for a decision by the presiding
arbitrator alone on questions of procedure.
153 Neil Kaplan & Karen Mills, ‘The Role of the Chair in International Commercial

Arbitration’, in The Asian Leading Arbitrators' Guide to International Arbitration,

ed. Michael Pryles & Michael Moser (Huntington: JurisNet LLC, 2007), 139.
154 Ibid., 146.
155 ASA, ‘ASA Special Series No. 31: International Arbitration Do's and Don'ts',

6th edn <www.arbitration-ch.org/publications/>, June 2009, 21.

156 Neil Kaplan & Karen Mills, ‘The Role of the Chair in International Commercial

Arbitration’, in The Asian Leading Arbitrators' Guide to International Arbitration,

ed. Michael Pryles & Michael Moser (Huntington: JurisNet LLC, 2007), 141.
157 Karl-Heinz Böckstiegel, ‘Case Management by Arbitrators: Experiences and

Suggestions’, in Global Reflections on International Law, Commerce and Dispute

Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al.
(Paris: ICC Publishing, 2005), 121.
158 For example, Canon VII of the ABA/AAA Code allowed arbitrators to be

‘predisposed’ towards the appointing party and communicate with them subject
to overriding obligations to act in good faith and with integrity and fairness. Even
where non-neutral arbitrators are permitted, there are still ethical obligations to
follow due process and to hear both sides before making a final determination. It
is outside the scope of this book to consider non-neutral arbitrators in the
domestic American context. For further discussion of this issue see James H.
Carter, ‘Improving Life with the Party-Appointed Arbitrator: Clearer Conduct
Guidelines for “Non-neutrals’”, American Review of International Arbitration 11
(2000): 295.
159 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1502.

160 Ibid., 1501–1504.
161 Ibid., 1505.
162 James H. Carter, ‘Improving Life with the Party-Appointed Arbitrator: Clearer

Conduct Guidelines for “Non-neutrals’”, American Review of International

Arbitration 11 (2000): 295.
163 Stephen R. Bond, ‘The International Arbitrator: From the Perspective of the

ICC International Court of Arbitration’, Northwestern Journal of International Law

and Business 12 (1991): 7.
164 Andreas Lowenfeld, ‘The Party-Appointed Arbitrator: Further Reflections’, in

The Leading Arbitrators' Guide to International Arbitration, ed. Lawrence W.

Newman & Richard D. Hill (New York: Juris Publishing, Inc., 2004), 46.
165 Jose Rosell, ‘The Challenge of Arbitrators’, Croatian Arbitration Handbook 10

(2003): 155.
166 David J. Branson, ‘American Party-Appointed Arbitrators – Not the Three

Monkeys’, University of Dayton Law Review 30 (2004): 47.

167 A.A. de Fina, ‘The Party Appointed Arbitrator in International Arbitrations –

Role and Selection’, Arbitration International 15, no. 4 (1999): 387.

168 Karl-Heinz Böckstiegel, ‘The Role of the Arbitrators in Investment Treaty

Arbitration’, in International Commercial Arbitration: Important Contemporary

Questions, ICCA Congress Series No. 11, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 2003), 371. The potentially counter-productive result
of over-zeal in a party-appointed arbitrator is also confirmed in Andreas F.
Lowenfeld, ‘The Party-Appointed Arbitrator in International Controversies:
Some Reflections’, Texas International Law Journal 30 (1995): 60.
169 Martin Domke, The Law and Practice of Commercial Arbitration (Mundelein:

Callaghan, 1968), 198.

170 Andreas Lowenfeld, ‘The Party-Appointed Arbitrator: Further Reflections’, in

The Leading Arbitrators' Guide to International Arbitration, ed. Lawrence W.

Newman & Richard D. Hill (New York: Juris Publishing, Inc., 2004), 48.
171 ICC Secretariat, ‘Note Concerning the Appointment of Administrative
Secretaries by Arbitral Tribunals', ICC International Court of Arbitration Bulletin
6, no. 2 (1995): 77.
172 Gerold Herrmann, ‘Power of Arbitrators to Determine Procedures under the

UNCITRAL Model Laws’, in Planning Efficient Arbitration Proceedings/The Law

Applicable in International Arbitration, ICCA Congress Series No. 7, ed. Albert Jan
van den Berg (The Hague: Kluwer Law International, 1996), 53.
173 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 372.

174 See ICC Rules 2012 Art. 24; SIAC Rules Art. 16.3.
175 ICC Rules 2012 Art. 24.
176 Gerold Herrmann, ‘Power of Arbitrators to Determine Procedures under the

UNCITRAL Model Laws’, in Planning Efficient Arbitration Proceedings/The Law

Applicable in International Arbitration, ICCA Congress Series No. 7, ed. Albert Jan
van den Berg (The Hague: Kluwer Law International, 1996), 42.
177 See UNCITRAL Rules 2010 Art. 17(2); HKIAC Rules Art. 14.3; Swiss Rules

2012 Art. 15.3; SCC Rules Art. 23.

178 See, e.g., ICDR Rules Art. 16.1.
179 ICDR Rules Art. 16.2.
180 Paolo Michele Patocchi & Harold Frey-Brentano, ‘The Provisional Timetable

in International Arbitration’, in Global Reflections on International Law,

Commerce and Dispute Resolution: Liber Amicorum in Honour of Robert Briner, ed.
Gerald Aksen et al. (Paris: ICC Publishing, 2005), 590.
181 Gerold Herrmann, ‘Power of Arbitrators to Determine Procedures under the
UNCITRAL Model Laws’, in Planning Efficient Arbitration Proceedings/The Law
Applicable in International Arbitration, ICCA Congress Series No. 7, ed. Albert Jan
van den Berg (The Hague: Kluwer Law International, 1996), 53.
182 Karl-Heinz Böckstiegel, ‘Case Management by Arbitrators: Experiences and

Suggestions’, in Global Reflections on International Law, Commerce and Dispute

Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al.
(Paris: ICC Publishing, 2005), 118.
183 Piero Bernardini, ‘International Arbitration and A-National Rules of Law’, ICC

International Court of Arbitration Bulletin 15, no. 2 (2004): 119.

184 ICC, ‘Techniques for Controlling Time and Costs in Arbitration’,
<www.iccwbo.org/uploadedFiles/TimeCost_E.pdf>, 2007, para. 34.
185 Ibid.
186 Karl-Heinz Böckstiegel, ‘Case Management by Arbitrators: Experiences and

Suggestions’, in Global Reflections on International Law, Commerce and Dispute

Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al.
(Paris: ICC Publishing, 2005), 118.
187 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 373.

188 ICSID Arbitration Rules Art. 20(1) states that the tribunal shall endeavour to

ascertain the views of the parties on procedural questions as early as possible.

189 E. David & B. Tavender, ‘Before and During Hearings, Here Is How to Manage

Arbitration Matters before They Get Out of Hand’, Alternatives to High Cost of
Litigation 28, no. 7 (2010): 143.
190 ICC, ‘Techniques for Controlling Time and Costs in Arbitration’,
<www.iccwbo.org/uploadedFiles/TimeCost_E.pdf>, 2007, para. 22.
191 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 374.

192 Ibid., 372.
193 Neil Kaplan & Karen Mills, ‘The Role of the Chair in International Commercial

Arbitration’, in The Asian Leading Arbitrators' Guide to International Arbitration,

ed. Michael Pryles & Michael Moser (Huntington: JurisNet LLC, 2007), 128.
194 David W. Rivkin, ‘Towards a New Paradigm in International Arbitration: The

Town Elder Model Revisited’, Arbitration International 24, no. 3 (2008): 378.
195 Devika Khanna, ‘Early, Active and Effective Case Management in Arbitration:

A Call to Reject Procedural Order No. 1’, International Arbitration Law Review 13,
no. 6 (2010): 237, 240.
196 This approach is taken by Professor Gabrielle Kaufmann-Kohler and is
outlined in Neil Kaplan & Karen Mills, ‘The Role of the Chair in International
Commercial Arbitration’, in The Asian Leading Arbitrators' Guide to International
Arbitration, ed. Michael Pryles & Michael Moser (Huntington: JurisNet LLC,
2007), 128–129.
197 Paolo Michele Patocchi & Harold Frey-Brentano, ‘The Provisional Timetable

in International Arbitration’, in Global Reflections on International Law,

Commerce and Dispute Resolution: Liber Amicorum in Honour of Robert Briner, ed.
Gerald Aksen et al. (Paris: ICC Publishing, 2005), 582.
198 Award of 28 January 1982, Iran-US Claims Tribunal Reports (1981–1982),

98.
199 Gerold Herrmann, ‘Power of Arbitrators to Determine Procedures under the

UNCITRAL Model Laws’, in Planning Efficient Arbitration Proceedings/The Law

Applicable in International Arbitration, ICCA Congress Series No. 7, ed. Albert Jan
van den Berg (The Hague: Kluwer Law International, 1996), 53.
200 Piero Bernardini, ‘International Arbitration and A-National Rules of Law’, ICC

International Court of Arbitration Bulletin 15, no. 2 (2004): 121.

201 Karl-Heinz Böckstiegel, ‘Case Management by Arbitrators: Experiences and

Suggestions’, in Global Reflections on International Law, Commerce and Dispute

Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al.
(Paris: ICC Publishing, 2005), 120.
202 ICC Rules 2012 Art. 24.
203 Swiss Rules 2012 Art. 15.3.
204 UNCITRAL Rules 2010 Art. 17.2.
205 The relevant rules are: UNCITRAL Rules 2010 Art. 17.2; HKIAC Rules Art.

14.3; Swiss Rules 2012 Art. 15.3; SCC Rules Art. 23; ICC Rules 2012 Art. 24; SIAC
Rules Art. 16.3.
206 Michael Pryles, ‘Limits to Party Autonomy in Arbitral Procedure’, Journal of

International Arbitration 24, no. 3 (2007): 335.

207 Paolo Michele Patocchi & Harold Frey-Brentano, ‘The Provisional Timetable

in International Arbitration’, in Global Reflections on International Law,

Commerce and Dispute Resolution: Liber Amicorum in Honour of Robert Briner, ed.
Gerald Aksen et al. (Paris: ICC Publishing, 2005), 591.
208 Ibid.
209 W. Laurence Craig, William W. Park & Jan Paulsson, Annotated Guide to the

1998 ICC Arbitration Rules with Commentary (Oxford: Oxford University Press,
1998), 120.
210 Paolo Michele Patocchi & Harold Frey-Brentano, ‘The Provisional Timetable

in International Arbitration’, in Global Reflections on International Law,

Commerce and Dispute Resolution: Liber Amicorum in Honour of Robert Briner, ed.
Gerald Aksen et al. (Paris: ICC Publishing, 2005), 589.
211 Paolo Michele Patocchi & Harold Frey-Brentano, ‘The Provisional Timetable

in International Arbitration’, in Global Reflections on International Law,

Commerce and Dispute Resolution: Liber Amicorum in Honour of Robert Briner, ed.
Gerald Aksen et al. (Paris: ICC Publishing, 2005), 63.
212 ICC, ‘Techniques for Controlling Time and Costs in Arbitration’,
<www.iccwbo.org/uploadedFiles/TimeCost_E.pdf>, 2007, para. 32.
213 Philip Yang, ‘The Organisation of International Arbitration Proceedings’, in

The Asian Leading Arbitrator's Guide to International Arbitration, ed. Michael

Pryles & Michael J. Moser (New York: JurisNet, 2007), 205.
214 Allan Philip, ‘The Duties of an Arbitrator’, in The Leading Arbitrators' Guide to

International Arbitration, 2nd edn, ed. L. Newman & R. Hill (Huntington: Juris
Publishing, 2008), 74.
215 Ibid.
216 Karl-Heinz Böckstiegel, ‘Case Management by Arbitrators: Experiences and

Suggestions’, in Global Reflections on International Law, Commerce and Dispute

Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al.
(Paris: ICC Publishing, 2005), 119.
217 Paolo Michele Patocchi & Harold Frey-Brentano, ‘The Provisional Timetable

in International Arbitration’, in Global Reflections on International Law,

Commerce and Dispute Resolution: Liber Amicorum in Honour of Robert Briner, ed.
Gerald Aksen et al. (Paris: ICC Publishing, 2005), 583.
218 The tribunal can exercise a lien over the award if fees have not been paid but
this will only be of benefit in the context of the party who expects to win.
219 ICSID Administrative and Financial Regulations, Reg. 14(3)(e).
220 ICC Rules 2012 Art. 36(4).
221 Matthew Secomb, ‘Awards and Orders Dealing with the Advance on Costs in

ICC Arbitration: Theoretical Questions and Practical Problems’, ICC Bull 14, no. 1
(2003): 59; Ibrahim Fadlallah, ‘Payment of the Advance to Cover Costs in ICC
Arbitration: The Parties' Reciprocal Obligations’, ICC Bull 14, no. 1 (2003): 53.
222 Statutes of limitations may apply. P
223 Paris Court of Appeal, 17 March 2005, SNCIv. Interim Nation, Revue de

l'arbitrage (2005): 790; Swiss Supreme Court 4A-391/2010 and 4A-399/2010.

The latter case involved a tribunal staying proceedings when fees were not paid.
224 The approach taken in ICC Case No. 10526 was that this is a contractual

matter between the parties. If a claimant brings a claim but refuses to pay its
share of the advance, a respondent might also be able to sue for breach of
contract in that scenario. See judgment of Tribunal de Commerce de Beaune 8
July 1994, Wenko Wenselaar v. S.A. GB Industries (1995) Rev Arb 132.
225 This was the case in Partial Award No. 2 of 1 June 2004 ICC Case No. 12491

and in Interim Award, dated 26 March 2002, in ICC Case No. 11405.
226 Fertalage Industries (Algeria) v. Société Kaltenbach Thurin, S.A. (France),

Tribunal de grande instance de Beauvais, 9 April 1998 (unreported) and Ulrich

Schubert v. H.O. Engineering, Inc., Docket No. L-4310-90 (N.J. Super. Ct. Middlesex
Co. 4 March 1994). Another French court has considered that it has no
jurisdiction to make orders of this nature: SARL Sifamos v. SA Grammer AG,
Tribunal de Commerce de Paris (18 December 1998) Rev Arb (2002): 997 cited
in Yves Derains & Eric A. Schwartz, A Guide to the ICC Rules of Arbitration, 2nd
edn (The Hague: Kluwer Law International, 2005), 345.
227 Yves Derains & Eric A. Schwartz, A Guide to the ICC Rules of Arbitration, 2nd

edn (The Hague: Kluwer Law International, 2005), 346.

228 See Lisa Nieuwveld, ‘ICC Advance on Costs: Strategical Games’, International

Law Practicium 20, no. 2(2007): 114.

229 ICC Case No. 9667 (1998), Clunet (2000): 1096.
230 Bundesgerichtshof, III ZR 169/83, NJW (1985), pp. 1903–1904 and Société

TRH Graphics v.Société Offset Aubin, Cour de cassation (19 November 1991), Rev.
arb. (1992), 462.
231 See Micha Bühler, ‘Costs in ICC Arbitration: A Practitioner's View’, Am. Rev.

Int. Arb. 3 (1992): 116, 148.

232 Stockholm Chamber of Commerce Arbitration Rules Art. 45(4).
233 Wetter observes that terms of reference are a relic from when French law

required a compromis for arbitration to have validity. See J. Gillis Wetter, ‘The
Conduct of the Arbitration’, Journal of International Arbitration 2, no. 2 (1985):
7–38; W. Laurence Craig, William W. Park & Jan Paulsson, International Chamber
of Commerce Arbitration, 3rd edn (Oxford: Oxford University Press, 2000), 273–
274.
234 Terms of reference are also described as Acte de Mission.
235 Pierre A. Karrer, ‘Pros and Cons of Terms of Reference and Specific
Procedural Agreements in Arbitration Clauses: Storm in to Calm the Sea’, in
Planning Efficient Arbitration Proceedings/The Law Applicable in International
Arbitration, ICCA Congress Series No. 7, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 1996), 73–74.
236 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1818.

237 Gerald Asken et al. (eds), Global Reflections on International Law, Commerce

and Dispute Resolution: Liber Amicorum in Honour of Robert Briner (Paris: ICC
Publishing, 2005), 565–566.
238 Ibid., 23.
239 Thomas Webster, ‘Terms of Reference and French Annulment Proceedings’,

Journal of International Arbitration 20, no. 6 (2003): 565.

240 See comments by Jarvin in Howard M. Holtzmann & Giorgio Bernini,
‘Hypothetical Case for Use in a Comparative Study of Arbitration Practice in
Various Legal Systems’, in Comparative Arbitration Practice and Public Policy in
Arbitration, ICCA Congress Series No. 3, ed. Pieter Sanders (The Hague: Kluwer
Law International, 1987), 60.
241 See Yves Derains & Eric A. Schwartz, Guide to the ICC Rules of Arbitration, 2nd

edn (The Hague: Kluwer Law International, 2005), 257–258.

242 Jean-Claude Goldsmith, ‘How to Draft Terms of Reference’, Arbitration
International 3, no. 4 (1987): 298–299.
243 See, e.g., Judgment of 25 April 1985, Diaeti v. Eurodis, Sofidis et al., Cour

d'appel de Paris, cited in Jean-Claude Goldsmith, ‘How to Draft Terms of

Reference’, Arbitration International 3, no. 4 (1987): 300, n. 4.
244 Ibid., 298.
245 Ibid.
246 Judgment of 12 July 1984, Arab Republic of Egypt v. Southern Pacific
Properties and STP (Middle East) Ltd, Cour d'appel de Paris, cited in Jean-Claude
Goldsmith, ‘How to Draft Terms of Reference’, Arbitration International 3, no. 4
(1987): 303, n. 5.
247 ICC Rules 2012 Art. 24(1) calls for the timetable when drawing the terms ‘or

as soon as possible thereafter’.

248 ICC Rules 2012 Art. 23(3).
249 Yves Derains & Eric A. Schwartz, A Guide to the ICC Rules of Arbitration, 2nd

edn (The Hague: Kluwer Law International, 2005), 256.

250 Neil Kaplan & Karen Mills, ‘The Role of the Chair in International Commercial

Arbitration’, in The Asian Leading Arbitrators' Guide to International Arbitration,

ed. Michael Pryles & Michael Moser (Huntington: JurisNet LLC, 2007), 130.
251 Paolo Michele Patocchi & Harold Frey-Brentano, ‘The Provisional Timetable

in International Arbitration’, in Global Reflections on International Law,

Commerce and Dispute Resolution: Liber Amicorum in Honour of Robert Briner, ed.
Gerald Aksen et al. (Paris: ICC Publishing, 2005), 579, fn. 19.
252 ICC Working Group on Terms of Reference, ‘Terms of Reference under the

1988 ICC Arbitration Rules: A Practical Guide’, ICC Court of Arbitration Bulletin 3,
no. 1 (1992): 35.
253 J.L. Greenblatt & P. Griffin, ‘Towards the Harmonization of International

Arbitration Rules: Comparative Analysis of the Rules of the ICC, AAA, LCIA and
CIET’, Arbitration International 17, no. 1 (2001): 107.
254 Yves Derains & Eric A. Schwartz, A Guide to the ICC Rules of Arbitration, 2nd

edn (The Hague: Kluwer Law International, 2005), 253.

255 Ibid.
256 J.L. Greenblatt & P. Griffin, ‘Towards the Harmonization of International
Arbitration Rules: Comparative Analysis of the Rules of the ICC, AAA, LCIA and
CIET’, Arbitration International 17, no. 1 (2001): 107.
257 This wording was postulated by Kaplan and Mills but only as a possibility in

exceptionally complicated cases. Neil Kaplan & Karen Mills, ‘The Role of the Chair
in International Commercial Arbitration’, in The Asian Leading Arbitrators' Guide
to International Arbitration, ed. Michael Pryles & Michael Moser (Huntington:
JurisNet LLC, 2007), 131.
258 Neil Kaplan & Karen Mills, ‘The Role of the Chair in International Commercial

Arbitration’, in The Asian Leading Arbitrators' Guide to International Arbitration,

ed. Michael Pryles & Michael Moser (Huntington: JurisNet LLC, 2007), 130.
259 See Michael Schneider, ‘Terms of Reference, the New 1998 ICC Rules of

Arbitration’, ICC International Court of Arbitration Special Supplement (1997):

26–36; Thomas Webster, ‘Terms of Reference and French Annulment
Proceedings’, Journal of International Arbitration 20, no. 6 (2003): 564.
260 Pierre A. Karrer, ‘Pros and Cons of Terms of Reference and Specific
Procedural Agreements in Arbitration Clauses: Storm in to Calm the Sea’, in
Planning Efficient Arbitration Proceedings/The Law Applicable in International
Arbitration, ICCA Congress Series No. 7, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 1996), 80.
261 Yves Derains & Eric A. Schwartz, A Guide to the ICC Rules of Arbitration, 2nd

edn (The Hague: Kluwer Law International, 2005), 256.

262 ICC Rules 2012 Art. 23(3).
263 ICC, ‘Techniques for Controlling Time and Costs in Arbitration’,
<www.iccwbo.org/uploadedFiles/TimeCost_E.pdf>, 2007, para. 30.
264 Jean-Claude Goldsmith, ‘How to Draft Terms of Reference’, Arbitration
International 3, no. 4 (1987): 299.
265 Yves Derains & Eric A. Schwartz, A Guide to the ICC Rules of Arbitration, 2nd

edn (The Hague: Kluwer Law International, 2005), 256–257.

266 Ibid., 255.
267 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 658.
268 Yves Derains & Eric A. Schwartz, A Guide to the ICC Rules of Arbitration, 2nd

edn (The Hague: Kluwer Law International, 2005), 254.

269 Thomas Webster, ‘Terms of Reference and French Annulment Proceedings’,

Journal of International Arbitration 20, no. 6 (2003): 586.

270 Ibid.
271 J. Gillis Wetter, ‘The Internationalisation of International Arbitration: Looking

Ahead to the Next Ten Years’, Arbitration International 11, no. 2 (1995): 117–
118.
272 Yves Derains & Eric A. Schwartz, A Guide to the ICC Rules of Arbitration, 2nd

edn (The Hague: Kluwer Law International, 2005), 267. For further discussion of
techniques employed by tribunals in this regard, see Yves Derains & Eric A.
Schwartz, A Guide to the ICC Rules of Arbitration, 2nd edn (The Hague: Kluwer
Law International, 2005), 267–268.
273 Report of the ICC Commission on Arbitration. © International Chamber of

Commerce (ICC). Reproduced with permission of the ICC. Originally published in

English and French in the ICC International Court of Arbitration Bulletin Vol. 3
No. 1 (1992). The Bulletin is available from the ICC Business Bookstore
<www.iccbooks.com> and in the ICC Dispute Resolution Library
<www.iccdrl.com>. ICC Working Group on Terms of Reference, ‘Terms of
Reference under the 1988 ICC Arbitration Rules: A Practical Guide’, ICC Court of
Arbitration Bulletin 3, no. 1 (1992): 23.
274 © International Chamber of Commerce (ICC). Reproduced with permission of

the ICC.
275 Thomas Webster, ‘Terms of Reference and French Annulment Proceedings’,

Journal of International Arbitration 20, no. 6 (2003): 582.

276 See also ICC Case No. 6223 (1991) (1997) 8 ICC International Arbitration

Bulletin 62, 70.

277 Andreas Reiner, ‘Terms of Reference: The Function of the International Court

of Arbitration and Application of Article 16 by the Arbitrators’, ICC International

Court of Arbitration Bulletin 7, no. 2 (1996): 59. See also Emmanuel Gaillard &
John Savage (eds), Fouchard, Gaillard, Goldman on International Commercial
Arbitration (The Hague: Kluwer Law International, 1999), 670.
278 Ibid.
279 Yves Derains, ‘Amendments to the Claims and New Claims: Where to Draw

the Line?’, in Arbitral Procedure at the Dawn of the New Millennium, Report of the
International Colloquium of CEPANI, 15 October 2004 (Brussels: Bruylant, 2005),
66, 71.
280 Ibid.
281 Swiss Supreme Court, Tribunal Fédéral, First Civil Chamber, 30 December

1994, No. 4P115/ 1994, (1996) XXI Yearbook of Commercial Arbitration 172
refused to set aside an award simply because the claimant had changed its case
theory as to the basis of damages. The court considered this was not a new claim
but instead a new legal justification of the facts.
282 Having said that, from counsel's perspective, advocacy should be seen as a

seamless exercise and the case strategy should be forcefully put from the outset.
283 Andrew Burr & Pierre A. Karrer, ‘“Chess Clock” Arbitration and Time
Management – Techniques in International Commercial Arbitration: From the
Perspective of the Arbitrator and Counsel’, Construction Law Journal 26, no. 2
(2010): 69.
284 Ibid., 68–69.
285 Pierre A. Karrer, ‘Freedom of an Arbitral Tribunal to Conduct Proceedings’,

ICC International Court of Arbitration Bulletin 10, no. 1 (1999): 22.

286 Margulead Ltd v. Exide Technologies, [2004] EWHC 1019 (QB, Commercial

Court).
287 See LCIA Rules Art. 15.
288 Jack J. Coe, ‘Pre-Hearing Techniques to Promote Speed and Cost-
Effectiveness: Some Thoughts Concerning Arbitral Process Design’, Pepperdine
Dispute Resolution Law Journal 2, no. 1 (2002): 64.
289 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 381.

290 Philipp Habegger & Anna von Mühlendahl, ‘UNCITRAL Notes on Organising

Arbitral Proceedings: Time for an Update?’, in Liber Amicorum Eric Bergsten.

International Arbitration and International Commercial Law: Synergy,
Convergence and Evolution, ed. S Kröll et al. (Alphen aan den Rijn: Walder Wyss,
2011), 216.
291 Neil Kaplan & Karen Mills, ‘The Role of the Chair in International Commercial
Arbitration’, in The Asian Leading Arbitrators' Guide to International Arbitration,
ed. Michael Pryles & Michael Moser (Huntington: JurisNet LLC, 2007), 126.
292 David W. Rivkin, ‘Towards a New Paradigm in International Arbitration: The

Town Elder Model Revisited’, Arbitration International 24, no. 3 (2008): 385.
293 It may be difficult to establish exactly when proceedings can properly be

described as default proceedings as parties might be involved intermittently or

may feign involvement while always asking for extensions.
294 Some rules now establish a summary mechanism. This can only arise by

agreement to the process or the relevant rules and still involves a view on the
merits by the tribunal. This is discussed in section 8.7.
295 Liberian Eastern Timber Corp. (LETCO) v. Government of the Republic of

Liberia, ICSID Case No. ARB/83/2, Award of 31 March 1986, ILM 26, (1987): 647;
Goetz and Others v. Republic of Burundi 10 February 1999, Award, ICSID Case No.
ARB/95/3, in Albert Jan van den Berg (ed.), Yearbook of Commercial Arbitration,
Volume XXVI (The Hague: Kluwer Law International, 2001): 24, 28, para. 7.
296 For example, UNCITRAL Model Law Art. 25(b).
297 For example, Swiss Rules 2012 Art. 28, LCIA Rules Art. 15.8.
298 Advances on costs are discussed in section 6.11.
299 See ICSID Case No. ARB/83/2 Liberian Eastern Timber Corp (LETCO) v.

Government of the Republic of Liberia, ILM 26 (1987): 647.

300 ICDR International Arbitration Rules Art. 23.1.
301 ICSID Rules Art. 42(1) and ICSID Convention Art. 45(2).
302 ICC Rules 2012 Art. 6.3 provides that ‘If any of the parties refuses or fails to

take part in the arbitration or any stage thereof, the arbitration shall proceed
notwithstanding such refusal or failure.’
303 Michael Mustill, Stewart Boyd & Neil Andrews, Commercial Arbitration, 3rd

edn (Butter-worths Law, 2008).

304 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 324–325.

305 Judith Butchers & Philip Kimbrough, ‘The Arbitral Tribunal's Role in Default

Proceedings’, Arbitration International 22, no. 2 (2006): 237.

306 For example, Art. 25(b) of the UNCITRAL Model Law provides that if the

respondent fails to communicate its statement of defence the arbitral tribunal

shall continue the proceedings, ‘without treating such failure in itself as an
admission of the claimant's allegations’.
307 J. Gillis Wetter, ‘Procedures for Avoiding Unexpected Legal Issues’, in
Planning Efficient Arbitration Proceedings/The Law Applicable in International
Arbitration, ICCA Congress Series No. 7, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 1996), 96.
308 British Petroleum Company (Libya) Ltd, wholly owned by the British Petroleum

Company Ltd (U.K.) v. The Government of the Libyan Arab Republic, Award on the
Merits of 10 October 1973; in Pieter Sanders (ed.), (1980) V YB Com Arb 147.
309 See, e.g., LCIA Rules Art. 22.1(c) which expressly allows for independent

research as to law.
310 Judith Butchers & Philip Kimbrough, ‘The Arbitral Tribunal's Role in Default

Proceedings’, Arbitration International 22, no. 2 (2006): 239.

311 Bogdanov v. Moldova 22 September 2005, SCC Case No. 93/2004.
312 Goetz and Others v. Republic of Burundi 10 February 1999, Award, ICSID Case

No. ARB/95/ 3, in Albert Jan van den Berg (ed.), Yearbook of Commercial
Arbitration, Volume XXVI (The Hague: Kluwer Law International, 2001): 24, 28,
para. 7 citing Military and parliamentary activities in and against Nicaragua
(Nicaragua v. United States of America), ICJ Reports (1986); (2001) XXVI YB Com
Arb 24, 27.
313 Günther J. Horvath, ‘The Duty of the Tribunal to Render an Enforceable

Award’, Journal of International Arbitration 18, no. 2 (2001): 144.

314 Online dispute resolution techniques exist which are not arbitration; for

example, negotiation (including automated negotiation) and mediation.

However, these methods are not binding and are beyond the scope of this
discussion.
315 Some aspects of the proceedings may occasionally need to be offline – for

example, Gabrielle Kaufmann-Kohler notes that ‘compliance with due process

guarantees may sometimes, especially in evidentiary proceedings, make it
necessary for the tribunal to use other than IT means’ Gabrielle Kaufmann-
Kohler, ‘Online Dispute Resolution and Its Significance for International Dispute
Resolution’, in Global Reflections on International Law, Commerce and Dispute
Resolution, Publication 693 (Paris: ICC Publishing, 2005), 451.
316 An alternative view and one which is set out in detail by Julia Hörnle in Cross

Border Internet Dispute Resolution (New York: Cambridge University Press,

2009), is that the accession to online arbitration agreements is less a product of
reasoned consent by the parties and more one of the less powerful party having
no option but to accept an arbitration clause specifying online (and therefore
procedurally and substantively abridged) arbitration. The book concentrates
more attention on consumer matters in leading to such conclusions.
317 Gabrielle Kaufmann-Kohler, ‘Online Dispute Resolution and Its Significance

for International Dispute Resolution’, in Global Reflections on International Law,

Commerce and Dispute Resolution, Publication 693 (Paris: ICC Publishing, 2005),
443. Kaufmann-Kohler further notes at 437 that as at 2005 ‘over a hundred
dispute resolution providers offer their services worldwide’.
318 For example, the Uniform Domain-Name Dispute-Resolution Policy (UDRP)

process established by the Internet Corporation for Assigned Names and

Numbers (ICANN) provides a dispute resolution procedure comparable to
arbitration but which does not prohibit recourse to courts.
319 ICDR Protocol for Manufacturer/Supplier Disputes, Rule 4.
320 For a detailed discussion of due process concerns with various models of

online arbitration and dispute resolution but again concentrating on consumer

issues, see Julia Hörnle, Cross Border Internet Dispute Resolution (New York:
Cambridge University Press, 2009), and for a consideration of the extent to
which due process concerns can be feasibly reconciled with the efficiency
imperatives of online transaction disputes see Thomas Schultz, ‘Internet
Disputes, Fairness in Arbitration and Transnationalism: A Reply to Julia Hörnle’,
Int J Law Info Tech 19, no. 2(2011): 153.
321 Thomas Schultz, ‘Internet Disputes, Fairness in Arbitration and
Transnationalism: A Reply to Julia Hörnle’, Int J Law Info Tech 19, no. 2 (2011):
159.
322 Ibid., 156.
323 Julia Hörnle, Cross Border Internet Dispute Resolution (New York: Cambridge

University Press, 2009), 251.

Part II: The Process of an Arbitration,
Chapter 7: Complex Arbitration*

Jeff Waincymer,

Chapter 7. Complex Arbitration (*)

7.1. Introduction

One of the most challenging areas for arbitration design, both theoretical and
practical, is the treatment of multi-party, multi-claim and multiple proceedings
disputes. There are a range of questions that typically arise. (1) These include,
who are the parties to the relevant contract and/or arbitration clause; when
may non-signatories be included; when can several contracts be described as a
single economic transaction such that one arbitration agreement will apply to
all; how broadly may a tribunal interpret an arbitration clause to encompass
claims under other contracts; where there are separate proceedings, when and
why may they be consolidated; where consolidation is not possible, how are the
proceedings otherwise to be coordinated; to what extent are decisions in one
proceeding binding on the other both as to final outcome and as to questions of
fact and law decided along the way; when can third parties be joined to an
existing arbitration; how will multi-person tribunals be appointed where there
are more than two parties involved; to what extent do questions of
enforceability inform these questions?

This chapter separates the discussion into a discrete analysis of multi-party and
multiple claim scenarios as questions of interpretation of arbitration
agreements and secondly, applications for joinder and consolidation. The
division is in part artificial as similar policy issues arise in each case and these
can be differing routes to the same ultimate outcome, being a single award or
series of awards covering all persons and claims. Determining the optimal
approach may also page "495" involve a comparison and ranking of the
alternatives. There is also the question of whether there is some unifying theory
that can be applied in resolving each issue. Nevertheless, there are distinct laws
and principles that apply and in many cases even different persons making the
determination. This is because an analysis flowing from the arbitration
agreement itself flows from separability and competence and is primarily
determined by the tribunal. Joinder and consolidation applications are more
purely procedural and such determinations are at times to be made by
institutions or courts rather than the tribunal itself.

The chapter begins with some overarching discussion of policy, as well as

autonomy and conflict of laws questions before turning to the various forms of
complex arbitration. Specific issues of policy, autonomy, rules and conflict of
laws will be revisited in more detail on a topic-by-topic basis. Counterclaims and
set-off rights were also discussed in section 4.4 in the context of the opening
stages of proceedings.

7.2. Policy Issues in Complex Arbitration

There will obviously be numerous permutations that may apply on a case-by-

case basis in complex scenarios where multiple claims and persons may be
involved. Policy issues may vary significantly within each scenario.
Nevertheless, it is useful to consider some overriding policy questions before
considering how to respond to various permutations as some factors will be
common. The following discussion looks at fairness and efficiency questions in
that context and in respect of both multi-person and multi-claim scenarios.

A preliminary observation is that a policy analysis of questions such as whether

the law should allow an arbitration clause to be extended, a third party joined or
proceedings consolidated, should not only consider these questions expressly,
but should also consider the merits of alternative scenarios and solutions. Any
such application is dealing with a potential claim that may, if denied, be brought
in other fora. A related aspect of the policy debate is to consider how to best deal
with situations where even if a third party is not included, its commercial
interests may be adversely affected by a case which is heard in its absence.
While the tribunal's decision does not bind third parties, it may make orders
which effectively undermine the value of the third party's alleged rights. This
would certainly be the case if there is a dispute over physical possession of an
asset and a decision is made to order specific performance in favour of a named
party.

7.2.1. A Single Efficient Forum

There are a number of efficiency arguments in favour of multi-party, multi-

claim, consolidated or joined proceedings. It seems sensible to presume that
resolving all questions in the one proceeding is likely to be more timely and less
costly than separate proceedings. A single proceeding also removes the
possibility of page "496" conflicting awards. Derivative proceedings would be
particularly wasteful when the findings of the first tribunal would need to be
reconsidered afresh. While restrictive principles of res judicata are unlikely to
apply where parties are technically distinct, in some cases related companies
may be so close as to be argued to be identical parties. There are also scenarios
where there would be timing problems in identifying damages if inter-related
claims were separated, for example, when a contractor is threatened with claims
by a subcontractor which have not yet crystallised, but where recourse is sought
against the owner. (2) A tribunal may also get a better understanding of what
truly occurred if all relevant parties are present, although that may be less
relevant as they can give evidence as witnesses in any event.

Conversely, it is important to understand that there can also be efficiency losses

where multi-party, multi-claim, consolidation and joinder are involved. There
can be problems with tribunal composition, parties being physically present in
relation to issues that truly do not concern them and greater difficulty in
coordination. This can all be affected by the stage at which applications for
inclusion are made. The later in the proceedings, the more likely that there will
be problems with the impact on previous steps and due process generally.

7.2.2. Resolving Admissibility via a Consent-Based Paradigm

Even if the efficiency argument is persuasive in some cases at least, inclusion of

multiple persons and claims cannot occur without some legal basis for this.
Ultimately it is a question of consent, either by reason of express agreement of
the parties, or by some implied indication of intent and/or under the laws and
rules made applicable by their arbitration agreement. Even that is contentious
in some multi-party scenarios where it is harder to fit a theory of inclusion into
a traditional consent paradigm. Examples considered below include lifting the
corporate veil or ostensible authority under agency law. Nevertheless, it is
generally the case that consent is the key and this is the position maintained
throughout this chapter. Even then, many of the differences in view as to
admissibility of complex claims can be explained by differences in approach to
the determination of consent. A tribunal that works from a strong albeit
rebuttable presumption that, a priori, parties would always want an efficient
resolution of their disputes is likely to find that admissibility is justified in many
instances. (3) Conversely, some tribunals would want to see some express
provision allowing for admissibility within the arbitration clause, the lex arbitri
or the procedural rules selected by the parties. Between the extremes, page
"497" different arbitrators might wish to see differing levels of direct or indirect
evidence of actual intent before making conclusions either way.

7.2.3. Procedural Rules as Guides to Consent

If the primary question is that of consent, one must begin with the arbitration
agreement as supplanted by the lex arbitri and any procedural rules agreed
upon. Where the latter expressly deals with the admissibility question and
where the parties have consented to those rules applying, evidence of consent is
clear. (4) Nevertheless, the consent logic flowing from express references in the
lex arbitri varies significantly if an institution, appointing authority or tribunal
selects the Seat rather than the parties, as this is one step removed from the
parties actual or implied consent. At times the choices made may have a
fundamental impact on admissibility of claims. For example, an appointing
authority has a particular challenge where it has the discretion to nominate a
Seat and where it is aware that a party wishes to bring a set-off defence. Should
one select a Seat that is favourably disposed to such claims? Even here, if the
likely choices by the independent entity would be known, inferences as to the
parties legitimate expectations can still be contended for.

Even where procedural rules are not clear, consent might be found through
waiver or acquiescence, for example, via a claimant not objecting to a claim by
the respondent when it is pleaded. Consent could also be inferred if a party
objects to a matter being raised in court on the basis of an allegation that it is
subject to an arbitration clause. A tribunal already invested with jurisdiction
under that clause might then consider that the assertion in court is effectively
evidence of an agreement to arbitrate. (5)

7.2.4. Efficiency and Implied Consent

While some authors and practitioners work from a presumption that business
people should be taken to have intended to have efficient proceedings and
efficient solutions to any jurisdictional questions, (a reasonable presumption in
and of itself), that will at best be a rebuttable presumption. (6) To the extent that
a tribunal page "498" is interested in considering subjective evidence, the
presumption is weaker as it presumes certain objective features of the parties
that may not be evident in the instant case. Even if the entire analysis is to be
limited to objective evidence and inferences, an efficiency paradigm might be
more complex than would at first appear to be the case. For example, would all
parties simply wish to allow counterclaims and set-off on the basis of efficient
resolution of the disputes between all parties and as a bar to duplicity of
proceedings? One countervailing criterion is that by accepting a counterclaim or
set-off, the amount in dispute and hence the costs of arbitration will increase.
Most rule systems will add the amount of the claim and reverse claims together
to determine the advance on costs unless they truly overlap in substance. This
can be a particular concern if it is foreseeable that at times there will be inflated
counterclaims that are tactically aimed at frightening the claimant into taking a
reduced settlement. Disparity in costs is not only problematic from a general
fairness perspective but impacts significantly on any consent analysis. This is
further complicated by the fact that tribunals have discretions as to costs and
may or may not seek to use these to counter the potential stumbling blocks.
Berger notes the use of possible counterclaims as delaying tactics or as
retaliatory devices. Ulmer notes the practical inspiration for many
counterclaims to have something to bargain with and set parameters for
arbitrators who might try and find mid-position solutions. (7) Craig, Park and
Paulsson point out that the requirement to include this in the advance of costs
can act as a deterrent to such strategic claims (8) although, if there is enough in
dispute, a costs advance obligation would have little deterrent effect,
particularly where this might provide for differential hardship in cases where
the parties are in vastly differing financial circumstances.

In addition to the costs implications of multiple claims there is also the question
of tribunal composition. Consider an extreme example where the primary claim
is essentially about complex questions of law where the parties have selected a
tribunal that is expert on such issues. Now envisage reverse claims that deal
only with challenges as to the quality of professional building or engineering
activities. The parties might prefer different experts for the latter claims. In
some cases preferred arbitrators for the second dispute might be professional
engineers, builders or architects who are not legally trained and thus may be
unsuitable for the first dispute. While this is not necessarily so, the example
simply highlights the fact that one cannot necessarily presume that the wish to
bring finality to all disputes between the parties and the wish to avoid
duplication in costs will necessarily mean that an existing tribunal formed in
response to the first dispute can confidently presume implied intent to allow it
to claim a mandate over a broad category of claims.

page "499"

A realistic assessment of the likely thinking of the parties at the outset might
raise some counter-intuitive hypotheses. Parties do not hope to have disputes
from the outset. More often than not, if an individual party envisages the
possibility of a future dispute it would most likely be envisaged as either one
brought on reasonable grounds by it or one brought on unreasonable grounds
by the other contracting party. This is because if the other party's claim was
reasonable, the first party would believe it would honour it without the need for
an adjudicated dispute. The contentious case, therefore, is to consider how each
might have wished at the outset to defend against claims they do not agree with.
In this event, there is at least a possible hypothesis that they might be presumed
from the outset to want whatever strategic advantage that may be permissible
to a defendant, subject to ethical and good faith duties. As Hanotiau has pointed
out in the context of multi-contract and multi-party arbitrations, ‘(t)he absence
of co-ordination of dispute resolution clauses, therefore, is not necessarily
pathological. It is sometimes intended deliberately. The same goes for the
possible refusal to consolidate the proceedings.’ (9) Tribunals should thus be
particularly careful not to stretch existing principles to try and promote
efficiency in the face of some of arbitration's more intractable problems. Those
problem areas may well be a small but important group of cases where parties
might genuinely prefer litigation over arbitration or where these matters have
to be raised and dealt with carefully in the parties agreement to arbitrate,
particularly in multi-contract circumstances. It is naturally the case that courts
have greater opportunity to allow counterclaims, consolidation or joinder than
do tribunals. That perspective, suggested by Leboulanger (10) is supported by
Poudret and Besson. (11)

Having said this, much can be gained from a consideration of efficiency

perspectives within ambiguous or uncertain consent scenarios. Parties can
certainly be presumed to want an efficient resolution of meritorious disputes
from the outset. Even if some business people would not want this, arbitration
ought to be built on good faith approaches. Hence if a party is trying to
destabilise proceedings or promote inefficiency, perhaps in order to frustrate a
claimant, such tactical endeavours in the face of a dispute should not colour the
determination of implied good faith consent from the outset. Attention should
be focused on a priori implied consent rather than behaviour after the dispute
has arisen, although the latter may help our understanding of ambiguous
original intent. Where subsequent behaviour is concerned, in some legal
systems there will be a further need to consider whether there has been an
abuse of rights through a party either blocking or relying on an extension of the
powers to cover multiple claims. Tribunals should not be quick to reach such
conclusions. Many tactical considerations are perfectly reasonable and would
not offend against notions of good faith.

page "500"

Converse presumptions of intent include that if set-offs are not readily allowed,
there may be additional expense, delay and even financial disaster for a party
truly entitled to a net benefit in circumstances where they had paid out on one
claim and the other party is insolvent before they can be forced to pay on the
reverse claim. Craig, Park and Paulsson also note that in these circumstances
there may be increased pressure on the respondent to settle for less than a
reasonable amount. (12) Thus there should be a concern to enhance the
interpretation of the arbitration agreement, where necessary, with a careful
assessment of all factors that might help a tribunal draw conclusions as to the
likely a priori intent of the parties. Implied intent to promote efficient solutions
is an important working hypothesis, as long as it is seen as one factor that needs
to be looked at alongside others and within the paradigm of consent, not
tribunal paternalism. Efficiency should simply be one of a range of factors to be
considered by tribunals that are faced with uncertain guidance from the
arbitration agreement and the rules themselves. It is important to understand
how it should be utilised. It is right for tribunals to consider efficiency factors as
a means of identifying a good faith a priori intent of the parties. It should not
simply be that tribunals look at efficiency per se from their own post-dispute
perspective, regardless of other evidence of parties intent.

7.2.5. Applicable Law and Conflicts Methodologies

Some would resolve multi-party and multi-contract issues via a broad-ranging

consent analysis while others would be strongly guided by a conflicts analysis.
Any conflicts analysis can see differences depending on whether a direct choice
or conflicts rules approach applies and can also see differences in classification.
For example, in a multi-contract scenario, if the question is pursued from the
perspective of the ambit of any one of the arbitration agreements, then this is a
partly procedural, partly substantive issue given the dual nature of arbitration
agreements and the need to interpret them. Applicable law issues were dealt
with in section 3.2.2. Conversely, where there is a distinct application for joinder
or consolidation, this is a procedural question in the context of the laws and
rules as agreed. (13) Treating it as either an interpretation question or a
procedural matter can also alter the person making the decision. In virtually all
cases of interpretation of the arbitration agreement, that would be a matter for
the tribunal. Conversely, in some legal systems, applications for consolidation
must be made to a court. Other questions of applicable law and conflicts
methodologies are left to discrete sections. The balance of the chapter now turns
to these discrete topics.

page "501"

7.3. Multi-Party Scenarios

This section deals with multi-party disputes. Because of globalisation of trade

and commerce, multi-site and multi-party manufacturing processes, increased
foreign investment and complex corporate groups, there are now a significant
number of commercial endeavours that inevitably involve the activities of more
than two parties. This is a significant practical issue. In 2002, the LCIA reported
that more than 50% of its cases involved more than two parties. (14) It has since
been suggested that approximately 40% of arbitration cases do indeed involve
more than two parties. (15) A far greater percentage have implications for third
persons even if they are not directly involved as parties.

While the commercial trends are clear, early views of arbitration argued that it
was essentially bipolar and was not naturally able to accommodate multi-party
scenarios. (16) Even where that was thought to be the better view theoretically,
the negative implications for the utility of arbitration became obvious and some
practitioners and institutions sought to provide new mechanisms to
accommodate such scenarios. Other practitioners and scholars argued for
power within existing broad discretions or under theories of consent, good faith
and estoppel. It is beyond the scope of this book to attempt a definitive and
comprehensive critical analysis of the various theories by which multi-party
situations are dealt with in arbitration. There have of course been excellent
treatises and articles led by those of Professor Hanotiau (17) and numerous
conferences and conference presentations that have highlighted the issues in
great detail and with great sophistication. (18) In any page "502" event, this book
is concerned more with issues of procedure and evidence and it is in that
context that the topic is discussed, although some observations are also made
about the theories themselves, given that they are essentially theories as to the
ambit of arbitration agreements and notions of consent, in no small part
procedural and evidentiary matters.

There are indeed many questions of procedure and evidence that arise which
make the topic contentious in theory and challenging in practice. Many of these
questions have indeed informed scholars and practitioners in their suggestions
about the proper conceptual treatment of multi-party scenarios, proper
approaches to drafting and ideal elements in rules. (19) Questions include how
should a tribunal approach such scenarios? What if the rules or lex arbitri are
silent? Is it simply a matter of consent? How is consent integrated with what is
found in the law and rules? What flows from the various permutations of
express consent in arbitration agreements? How are these to be interpreted? To
what extent is contemporaneous evidence relevant to determine what the
signatories truly intended vis-à-vis third parties and vice versa? Where
applicable tests look to the behaviour and/or the degree of connection between
the third party and the existing parties and/or various claims and
circumstances, by what evidence and analysis will a tribunal make such
determinations? At what stage will such determinations be made? If it is to
occur before tribunal composition, who is to make the determination and on the
basis of what evidence? Would those determinations be binding on a
subsequent tribunal? If the questions are to be raised after establishment of the
tribunal, how will this impact retrospectively on tribunal composition, given
that the proposed third party has not been privy to its establishment?

It is important to distinguish between the preliminary question as to who may

be a party to an arbitration and subsequent questions as to how an arbitration
may proceed from the outset when more than two parties are involved. Many
commentators speak of ‘extension’ of the arbitration agreement to third parties
although it is widely acknowledged that this is an unfortunate expression. The
real question is whether the third party is properly within the agreement.
Hence, it is better to speak of inclusion rather than extension. Nevertheless,
because many use the term as a simple short-hand expression, this will at times
be used here as well. Another issue is whether it can ever be said that all
relevant persons must be included to make a particular proceeding valid. It is
certainly possible for parties to agree that disputes can only be brought to
arbitration when all signatories are to be included. In other cases, a failure to
include all relevant persons may simply impact upon the decision on the merits.

page "503"

7.3.1. Typology of Multi-party Scenarios

Answers to the questions posed in the previous section may vary depending on
a range of factors, including who is trying to be involved as a party or who is
trying to involve another against their will. In a major study, Voser has listed a
number of scenarios where disputes may typically involve more than two
parties. These include construction and major industrial projects, in particular
through the involvement of sub-contractors; guarantees, insurance and
reinsurance; supply chains, back-to-back purchases and commodities
transactions where the same asset may be sold many times, including defective
product claims; joint venture, merger and acquisition and shareholder
transactions; trust arbitrations; and sports-related disputes. (20) Third-party
problems also typically arise with transport and software.

The various permutations can be classified in ways other than in respect of

subject matter and commercial endeavour and the different classifications may
themselves throw up theoretical and practical concerns. While there are
different scenarios, each ultimately requires valid consent to justify the third-
party involvement, although even that is contentious. In some cases, consent
might be deemed via some estoppel. Some scholars also assert a non-consensual
basis for inclusion. (21) These questions are addressed when considering the
individual grounds that are asserted to support inclusion. Furthermore, because
there are differing notions of consent and the evidence on which it can be based,
different theories can be applied to multi-party situations, which in part
explains differing views as to their optimal treatment. Less contentious is the
fact that each scenario must meet the same mandatory procedural rules of due
process.

An important distinction is between cases where multi-parties are required to

be involved as an essential element of an express arbitration agreement and
others where there is a disputed issue as to whether multiple parties can be
brought within the one arbitration. A distinction might also be made between
persons who are mentioned in an arbitration clause but unlike other persons,
did not sign it and alternatively, those who are neither mentioned nor are
signatories. There may even be a situation where there is a signature of a person
but no mention of them or the entity that they represent. This might arise where
an arbitration clause is signed by an employee of a related party to the named
party or endorsed by a governmental official in an investment joint venture.
Another possible classification is between those persons that have the essential
character of claimants and those that have the essential character of
respondents. Such a classification will be relevant to tribunal page "504"
selection and early documentary stages where formal requests and answers are
required. In some cases, it is easy to divide multiple parties into claimant and
respondent groups with common interests and arguments while in many other
situations, multiple parties may have varying interests and unique claims or
defences, regardless of whether they concur with other parties on some issues.
This raises added problems with tribunal selection.

In addition to the differing interests of multiple parties, the nature of their

claims against each other may also be classified differently. One typical scenario
is where a respondent wishes to join a third party on the basis that it asserts no
primary liability to claimant, but in the event that there is, the joined party is
argued to be ultimately responsible. These are described as claims for recourse.
(22) Another scenario is a cross-claim, being a claim by one respondent against

another. Another scenario is described as intervention, where a third party

seeks to join existing arbitral proceedings. (23) Motivation may also be relevant.
Unwilling non-signatories may be sought to be brought in because the signatory
is thought to have insufficient funds or because there would be separate claims
against a non-signatory, perhaps under guarantees that would more efficiently
be dealt with at one and the same time. (24) Scenarios could also be classified
depending on whether extension notions are being argued by a claimant in
arbitration against an unwilling respondent, whether they are being used by a
respondent in court proceedings to argue that the claim itself should instead be
brought in arbitration or whether the non-signatory is itself a claimant in
arbitration. A non-signatory might also wish to be party to arbitral proceedings
where it believes the determination will significantly affect its interests, for
example, where decisions are to be made on treatment of intellectual property
over which it claims some entitlement.

Another important distinction is between cases where multiple parties are all
sought to be involved from the outset and conversely, cases where a party seeks
or is sought to be added to an existing arbitration. Claimants will typically join
multiple respondents from the outset, either because of an assertion of joint and
several liability, or because of uncertainties as to causation. In the latter
scenario, the claimant runs the risk of being successful against one party but
failing against others, with possible commensurate cost orders as a result.
Joinder arises where a third party is added to arbitral proceedings that have
already commenced. The latter scenario still must consider questions of consent
and due process but adds other theoretical and practical problems in terms of
tribunal composition and impact on the prior events in the existing arbitration.
Other timing and evidentiary page "505" questions may also be relevant. There
may be differences as to whether the multi-parties are mentioned in the
arbitration agreement or not. Some may be mentioned but did not sign the
contract. Some may be related companies from the outset, while others may
have been established after the contract was begun. Some may be legitimate
commercial entities while others may be asserted to be mere shells created to
avoid arbitral liability or shift assets out of enforcement reach. Claims may be
brought at the outset, before tribunal establishment or before, during or after
key hearings. The inter-party claims and defences may relate solely to the same
issues or to similar issues. There may be significant differences as to the
knowledge of each party of the involvement of others in the essential
commercial transaction.

The point to the above discussion is simply that attention may need to be given
to any legitimate criterion by which separate classification may occur and that
general theories may not lead to common conclusions in each factual
permutation even within certain categories.

7.3.2. Policy Issues in Multi-party Scenarios

Previous sections looked generally at some policy questions in relation to

complex arbitrations. Some additional comments are appropriate in relation to
the discrete topic of multi-party scenarios. As noted, arguments in favour of
admissibility or consolidation include general efficiency, reducing the
transaction costs of parallel proceedings, overall timeliness, and the avoidance
of some of the pitfalls flowing from the composition of multiple tribunals where
overlap may raise questions of prejudice or undue influence. Because of the
many permutations of counterclaims and cross-claims and numerous scenarios
where multiple parties do not neatly fall into twin camps of claimant and
respondent interests, it is quite possible that there will be particular issues that
are not of interest to each and every participant. From their perspective, this can
needlessly extend the proceedings and add to costs if they nevertheless have to
witness issues being addressed that are beyond their concern. Where there are
claims of recourse by a respondent or cross-claims between respondents, a
claimant may need to sit through arguments that have no immediate relevance
to it. It simply seeks to show that the primary respondent owes it relief. If it is
successful in that regard and the respondent has the ability to comply, the
latter's attempt to seek reimbursement is a commercial side issue from
claimant's perspective, albeit one that may benefit it if respondent is
impecunious. However, where joint signatories are concerned, claimant must
have impliedly consented to this scenario on the understanding that single
proceedings would have overriding efficiency and fairness gains. Having an
excessive number of parties involved may also confuse matters. A tribunal may
be wrongly coloured in its view as to the reasonableness of one party simply
when considering the actions of a related entity. Many third persons can best
assist as witnesses without their own counsel making submissions and adding
to cross-examination time. Efficiency questions are also impacted upon by
discretionary decisions as to page "506" costs and interest. If multi-party
proceedings delay finality and there are no interest awards, that may be more
problematic. A similar scenario arises if a tribunal is reluctant to award costs to
the winning party. Where a primary respondent is held liable but is successful in
its claim for recourse, it may also be more problematic as to how it will be
treated in terms of allocation of costs. (25) There are also added logistical
concerns with finding mutually convenient times for key stages. There can be
particular problems if a third party is joined to existing proceedings where a
schedule was developed based on the wishes of the initial parties and where this
is not of equal convenience to the joined party.

While there are thus efficiency arguments for and against, a sophisticated
analysis would look at the cost benefit of alternative scenarios, one where the
third party is included and the other where claims by or against the third party
are conducted in other proceedings. With this caveat in mind, it might be
presumed in most cases that a well organised joint proceeding would be more
efficient than the alternative, notwithstanding some undesirable transaction
costs. This can be impacted on by the proactivity and planning of the tribunal
aiming to reduce transaction costs.

7.3.3. Consent

As noted throughout, all arbitration is based on consent. Hence any justifiable

theory of the proper treatment of multi-party scenarios must be consistent with
the evidence of consent in the instant case. Mere efficiency is not sufficient,
although efficiency may be factored in by some as an element of implied consent
where the parties have not clarified their attitude clearly. As noted in Chapter 3,
consent may arise from express or implied terms in the arbitration agreement,
from the lex arbitri invigorated by selection of the Seat and/or by selection of
institutional or ad hoc rules that expressly or impliedly cover such scenarios.
There may also be individual specific agreements on procedural issues. Within
this broad context, the adjudicator must be able to say that each of the relevant
persons can be taken to have consented to mutual rights and obligations under
the relevant arbitration agreement. Mantilla-Serrano concludes that consent is
the essential issue in each case. (26) Derains also sees it as essentially dependent
on intent. (27) Stephen Bond also argues strongly that the question is essentially
that of consent. (28) Rubins considers the issue to be one of establishing a
‘meeting of the minds’ but page "507" acknowledges that contractual consent is
not purely subjective, nor can it be described as fixed and formulaic. (29)

While it is difficult to generalise, Rubins highlights an in principle difference

between the common law and civil law approaches to consent. He suggests that
common law legal systems tend to view the question of binding non-signatories
to an arbitration agreement as one of general contract law, ‘permitting the
application of a wide range of exceptions to the rule of privity’. (30) As will be
discussed further in this chapter, these exceptions may include incorporation by
reference, agency, estoppel and piercing the corporate veil. Rubins contrasts this
with a strict formal approach maintained in many civil law jurisdictions. (31) The
Dutch, German and Russian courts for example, have traditionally shared the
view that arbitral agreements should be limited in their reach to the parties that
sign them. (32) At this stage it should simply be noted that while most argue that
consent is the key, some of the bases involve third persons who subjectively
would never have wished to be involved. An example is a shareholder using an
asset free shell company to protect against recourse. In some cases, extension
applies because of that third person's behaviour and the signatories' reasonable
expectations. There may then be differences in view as to whether this is a form
of deemed consent or whether it is better to describe this as a distinct category
where consent is not determinative. This is again discussed after individual
scenarios are analysed.

Important questions are who is to determine parties' intent, by what evidence,

what presumptions might apply, for example, separate legal identity and privity
of contract, and what applicable law such as in relation to contract,
corporations, agency, public and international law rights of States should be
brought to bear on this analysis? (33) These matters are considered in detail in
section 7.7 which seeks to make concluding remarks after analysing individual
bases for extension and after considering some specific logistical and due
process concerns where extension is contemplated.

The final preliminary observation is that original consent is the key where
consent is the basis for inclusion. (34) After all, a claimant need not obtain
additional consent to bring a case against a respondent who is already within
the arbitration agreement. Similarly, a respondent in an arbitration agreement
with a third party could bring a new case against it based on the consent in that
original agreement. page "508" Hence the consent involved in extension, joinder
and consolidation cases is aiming to identify consent to that possibility from the
outset. The analysis of actual implied intent is also best undertaken in the
context of seeing intent as presumed to be in good faith. (35) The simplest case is
where multiple parties are all party to the same arbitration agreement. Here
they have either expressed an agreement to multi-party arbitration or this could
be implied from the common signature of the agreement. A contrary view to the
effect that in some cases there needs to be common issues of fact or law or
appropriate connection of the multiple claims is not to be preferred. (36)

7.4. Lex Arbitri and Rules

Arbitral laws and rules may seek to express principles of inclusion, leave the
question to general procedural powers or constrain a tribunal as to certain
powers. Because of the conceptual challenges, most arbitral laws do not deal
expressly with multi-party situations. The questions are thus more commonly
dealt with under general principles.

A further problem is that many rules are built on a bipolar model, simply adding
the notion that there may be more than one claimant or respondent
respectively. (37) Yet as noted above, in many instances it is difficult to easily
classify multiple parties into claimant and respondent camps.

7.5. Tribunal Appointment and Multi-Parties

Before looking at the reasons for including multi-parties, one important due
process hurdle should be examined as it has had a significant impact on the
policy analysis and has led to challenges and rule changes. This relates to the
selection of multi-person tribunals. It is important to keep in mind that Article
V(1)(d) of the New York Convention provides as a ground for refusal of
recognition and enforcement of the award that the composition of the arbitral
tribunal was not in accordance with the parties' agreement. Due process and
public policy norms also require equal treatment which would encompass equal
treatment as to tribunal appointment. Ensuring as a preliminary matter that the
appointment of arbitrators accords with the parties' wishes will thus avoid
costly disputes about the enforceability of an award. Difficult issues of
appointment primarily arise with multi-person tribunals where multiple parties
are involved. Where there is to be a sole arbitrator, the parties will either agree
on a single appointment or not regardless of page "509" how many voices must
be heard. If not, there will be a fallback mechanism with appointment by an
independent authority. There may still be problems if a new party is sought to
be brought in after a sole arbitrator is already appointed, where the new party
would argue that they have been denied a right to be involved in the selection
process, particularly where there was a consensus among the prior parties that
the new party would not have concurred in. Nevertheless, the key concerns
remain with multi-person tribunals. Historically, rules have been simply
inadequate to deal with multi-party scenarios. Arbitral rules for multi-person
tribunals were traditionally built on a bipolar model, with two parties each
typically selecting one arbitrator, and the two appointees then selecting a Chair.
Even where rules do seek to expressly deal with multi-party scenarios, they are
premised on the view that the parties can nevertheless be divided into a
claimant group and a respondent group. This is clearly simplistic as on a range
of issues, parties seeking to all defend claims may have strongly varying
interests. (38)

There is no ideal model when more than two parties are involved in selecting a
multi-person tribunal. Such appointment raises difficult policy questions. From
the perspective of simple numerical equality, if there is an imbalance between
parties on either side of the dispute, that form of equality can never be
optimised in multi-party scenarios. Furthermore, and as noted, there remains a
problem of identifying which side each party is truly on when their interests
may vary from issue to issue. Where substantive equality is concerned, one
might look to see what are the true similarities in relation to their interests and
issues, but this would involve contentious qualitative assessment early on in
proceedings, with no natural independent adjudicator at that stage. Due process
and enforceability issues must also be considered, with practitioners and rule
makers all taking note of the Dutco decision where two respondents could not
agree on a joint appointment, leading to nomination by an institution. The
claimant's appointment was accepted. The respondents successfully challenged
enforcement on the basis that they were denied equal treatment at the
appointment stage. (39)

Where agreement is not possible, one possibility is to have more than three
arbitrators where there are multiple parties involved, although there still is no
mechanism that would be beyond criticism. Simply providing each party with
one selection would lead to an imbalance when there are distinct claimant and
respondent groups of differing numbers. That would be particularly problematic
if parties on the majority side were able to appoint some or all parochial
arbitrators as page "510" this might guarantee a favourable outcome. Another
option is to allow each truly distinct party to appoint their own arbitrator where
those on one technical side have conflicting interests. Such an approach will
often lead to an even number of arbitrators and the need for one to have a
casting vote. It will add to the expense. The most significant problem is that
there would be a need to make a determination at the outset as to whether the
interests truly conflict or not. There is no easy way for such a decision to be
made in advance of the tribunal's appointment, the very occurrence that flows
from the necessary determination. It would be problematic for an institution to
make such a determination. While institutions will at times make significant
preliminary decisions such as in relation to jurisdiction, these are generally
provisional and allow for a revision by the tribunal. In addition, a decision would
have to be made on allegations in early pleadings and not be based on
assessment of evidence. It would be too easy for multiple parties to plead
conflicting views to generate an entitlement to an extra arbitrator. It would thus
be highly problematic.

While these inherent problems are significant, if there truly is jurisdiction, then
some process must apply. The first question is whether a multi-party arbitration
clause provides for an express appointment mechanism. If so, that should
obviously be employed. Some arbitration clauses may indicate that if two or
more respondents are unable to agree on a joint nomination, the proceedings
against them must be separated. (40) Even specific agreements for joint
appointments may be problematic where a supervisory court employs logic
such as in Dutco to the effect that the right to equal treatment is a matter of
public policy which cannot be waived in advance. (41) Poudret and Besson
criticise the view that this cannot be waived before the dispute has arisen. (42)
Acceptance of waiver is based on party autonomy, although the argument is
circular. There seems no valid reason to distinguish between pre-dispute waiver
and post-dispute waiver. Either the mandatory due process argument prevails
or it does not. There is no reason to argue that it operates differentially in the
two scenarios. Nevertheless, waiver may not be the best logic. Express
mechanisms in the arbitration agreement should be upheld unless their terms
are problematic from a substantive equality perspective. Provisions that are
conditions of the arbitral agreement cannot readily be discarded on due process
grounds without undermining the entire agreement. If they are so problematic,
then a reasonable prospective appointee would not accept the mandate.

Consent may also be found through the laws and rules applicable. Post Dutco,
many rules now provide that if such a joint appointment cannot be made, the
page "511" institution or appointing authority will appoint all members. (43) This
is a second-best solution in the context of removing from all parties one of the
suggested values of arbitration, being the right to appoint one's arbitrator,
although that has been criticised by some at least in terms of its incentive to
parochialism. There may even be inequality in the post Dutco situation in the
sense that the unitary party on one side has done nothing to create the failure,
while the parties on the other may have failed to agree in good faith on a joint
appointment. As Voser points out, the post Dutco responses also do not accept
the case's logic in full in terms of the court's view that parties could not waive
their right to appoint their own arbitrator in advance. She rightly makes the
point that accepting new institutional rules that allow for fallback institutional
appointment of all three is such a waiver. (44) There are also differences between
rules that simply provide a discretion to the institution to appoint all members
and those that stipulate this as mandatory. Where an institution merely warns
the parties that it will make an appointment in their stead, this may encourage
them to reach agreement. A further problem with the post Dutco solution is that
it could even be a disincentive to claimants bringing action against a number of
respondents. In such circumstances, the claimant would be aware that if the
respondents cannot agree on an arbitrator, it will itself lose its right to make an
appointment. It would be problematic if because of this fear, the claimant
strategically brought proceedings against one respondent, had the tribunal
constituted and then immediately brought a claim against a second respondent,
seeking consolidation. A further problem with the Dutco logic that there can be
no advance waiver is that a failure to recognise such a waiver could itself be a
failure to respect the choice of procedure of the parties and hence be a ground
for blocking enforcement under Article V(1)(d). (45) Of course, the Dutco logic is
dependent on Article V(1)(b) in the context of the right to participate in the
constitution of the tribunal. Hence there is again no easy solution.

Where an institution makes an appointment, in some cases it will disregard the

previous claimant's nomination, but in other cases will reconfirm claimant's
appointee. (46) In considering whether it is appropriate to retain the single
party's nomination where multiple opposing parties cannot agree on their
selection, Voser page "512" divides situations into three categories. The first two
are where the multiple parties have common or similar interests respectively,
and a third category arises where they have conflicting interests. She argues that
in the first two categories the single appointment should be upheld as the failure
to make a joint nomination ‘must be considered an attempt to obstruct the
smooth commencement of the arbitration.’ (47) While there is much to be said in
support of the single party's nomination being retained in most instances, there
are problems with any rule of reason to that end. First, it would be difficult to
analyse and categorise the parties' interests as either common, similar or
conflicting in advance of hearing all claims and cross-claims. In most situations,
multiple parties will have common interests on some issues, but will
fundamentally divide on others. Even closely related entities are often treated as
separate cost and profit centres and may have differing legal issues in the
arbitration that could call for different appointments. (48) In many multi-party
scenarios, the parties may also have differing views as to choice of arbitrator, at
the very least as between those who take an expansive view of extension of
arbitration agreements and those who do not. Secondly and relatedly, one
cannot presume that both parties are equally attempting to obstruct the process
by failing to agree. One might want a parochial appointment while the other
might want to appoint a person of impeccable character. One might have
experience of arbitral processes and the other be ignorant, refusing to agree on
a non-contentious appointment. There can be no confident presumption that
each of the parties failing to agree is equally culpable for the failure. Finally,
where appointment is only for the member not agreed, there may be a debate as
to which side the various persons are on as there may be a gateway issue as to
who actually must agree with whom. Finally, where an institution would make
the preliminary determination as to inclusion, because of the implications for
composition, its prima facie determination of validity has more significant
practical implications than simple cases of challenges to jurisdiction.

A further problem with subsequent joinder of third parties is that previously

appointed tribunals who were accepted as independent and impartial could
have some relationship with the newly joined party or its counsel that could
lead to challenges. The more that an arbitrator is aware that there may be third-
party involvement, the more that a conflicts search prior to acceptance of an
appointment might include such potential third parties. (49) Conflicts situations
could be particularly problematic when there is joinder of a reluctant third party
who might then seek to select counsel with the specific aim of creating a conflict
scenario and page "513" disrupting the proceedings. As argued in section 9.7, a
tribunal ought to have power of control in such circumstances and be able to bar
the involvement of such counsel.

Section 7.13 below looks at a range of practical options where extensions,

consolidation or joinder are not possible. As to tribunal appointment, where it is
not possible to bring all the parties to the same arbitral hearing, another
possibility is to appoint the same arbitrators in different arbitrations. Hanotiau
suggests that this will minimise the risk of conflicting decisions. (50) Another
possibility is at least some common arbitrators. This option has been criticised
on the grounds that the participation of one arbitrator in parallel proceedings
with different arbitrators will have access to documents or facts to which the
other arbitrators may not. (51) Once again, there is no perfect and unassailable
solution.

7.6. Inclusion of Multi-Persons or Extension of Arbitration Agreements

As noted at the outset, commentators tend to speak of ‘extension’ when

considering when and why arbitration agreements can bind non-signatories. It
has rightly been observed that the term is misleading. Nothing is actually
extended. Instead, a determination is made as to whether a non-signatory
nevertheless falls within an arbitration agreement on some theoretically and
factually valid basis and hence can be said to be included in the agreement. (52)
While most cases will involve an attempt to bring in a non-signatory, in some
cases a non-signatory will itself seek to rely on an arbitration clause as a bar to
court proceedings, or simply as the basis for a claim. (53) In an
inclusion/extension scenario, there are actually two questions, first whether the
agreement can apply to the non-signatory and secondly whether the agreement
still applies to the original named party with whom the third party has a
particular connection. In many cases such as group of companies, this will be so,
but in other instances, for example agency, assignment, subrogation or cases
involving piercing of the corporate veil, even if a claim is validly made against
the third party, it might be seen as having overtaken and replaced the original
arrangement. If it is extended, there will then be further questions as to the
optimal procedure to adopt to promote both fairness and efficiency.

page "514"

As noted at the outset, while most accept that extension must be based on some
theory of consent, differing views may be argued for, leading to polarised views
on the proper approach to non-signatories. Different practitioners and scholars
have differing views as to the importance of actual, implied or constructive
consent or whether one can rely on particular notions of good faith and reliance
such as estoppel or abuse of rights. (54) While there are different theories and
categories of inclusion, all rely to some degree on notions of reasonableness and
good faith considered in the context of original consent. A procedure contrary to
the legitimate expectations of the original parties is for that reason
unreasonable. Instead, the categories are concerned to respond to some valid
but technical reasons why parties who in substance ought to be involved, have
some technical barrier through distinct legal forms, contractual limitations or
differences in property ownership.

There are a number of theories or fact typologies underlying claims to apply

arbitration agreements to non-signatories or parties not otherwise expressly
referred to. These include the group of companies doctrine, agency, piercing the
corporate veil, and agreement or estoppel by conduct, incorporation by
reference, assumption, third-party beneficiary, subrogation, novation and
guarantee scenarios. (55) The categories can overlap. However, noting the
commonality of features of the different categories does not necessarily point to
a solution. To some adjudicators, a legal barrier is a mere technicality and
should not prevent an efficient and comprehensive resolution of the true
substance of the dispute. To others, sophisticated parties exercising their free
will to create such complex relationships are entitled to do so in a way which
prevents arbitral jurisdiction just as they are entitled to create jurisdiction in the
first place through consent. Some adopt an entirely different methodology and
would argue for a conflicts approach looking to the proper law of the arbitration
agreement or the legal categories listed above, to determine how to deal with
each situation. (56) The following sections deal with each potential category, after
which some concluding remarks are made as to the proper approach that is
recommended.

7.6.1. Agency
Legal systems accept that in appropriate circumstances, an agent can bind a
principal. It is accepted that this can include binding a principal to an arbitration
page "515" agreement. The most common and non-controversial agency
scenario is where a corporate officer signs an arbitration agreement in that
capacity in order to bind the company. When acting in this way, the company is
bound but not the company executive. (57)

In other cases where a non-signatory is asserted to be a principal bound by the

acts of a signatory agent, the first question will be what law of agency is
applicable. Questions will arise as to whether it should be considered under
conflicts rules contained within the lex arbitri, or be based upon national laws
on legal capacity and agency of the persons involved. Because of the separability
of the arbitration agreement, there is not even consensus as to whether this is a
procedural or a substantive issue. There is thus a question whether the
separability of the arbitration agreement calls for resort to the law applicable to
the arbitration agreement itself. (58) Other choice of law options are to look to the
place where either the alleged principal or agent acted or had its business
establishment (59) or the law where acts were performed or the law of the place
where the holding out occurred leading to some detrimental reliance. Because
none of these is obviously preferable, Born argues for an international solution.
(60) He also suggests that a validation principle should apply whereby the non-

signatory should be bound if either the law governing the arbitration agreement
or the law governing the agency relationship would subject the principal to
jurisdiction. (61) A contrary view would be that a validation concept is most
justified when determining jurisdiction as between signatories under
interpretation effete utile. (62) Parties who entered an arbitration agreement can
be presumed to intend it to be effective. Similar logic is not readily maintainable
where a non-signatory is concerned. The preliminary question is whether that
party can be said to have agreed to arbitrate at all. (63)

Where agency is involved there is also a distinction between actual authority

and ostensible authority. Where the latter is concerned, similar principles are
raised to estoppel discussed below. This generally requires some culpable action
in holding out that authority is in existence where it does not actually exist.
There is usually also a need for detrimental reliance. Again there are choice of
law page "516" questions, but more fundamentally, in an arbitral context, some
would find it difficult to accept consent to arbitrate purely as a result of estoppel
type scenarios for two main reasons. The first is conceptual in that there was no
actual intent to be bound. The second is factual, in that the more the non-
signatory acts in a way that holds out authority, the more likely that they have
held out an intent to be party to the agreement in other ways. Hence there may
not be a need for a distinct category of apparent or ostensible authority. That
will not always be so, however, and there will be circumstances where the only
holding out is as to the supposed agent's authority.

In the US, non-signatories have relied upon agency theory as a basis by which to
claim the benefit of an agreement to arbitrate. This led one US Court to decide
that ‘[w]here the parties to [an arbitration] clause unmistakably intend to
arbitrate all controversies which might arise between them, their agreements
should be applied to all claims against agents or entities related to the
signatories’. (64) However the theory has not always been followed. For example,
in Merrill Lynch Inv. Managers v. Optibase Ltd (65) the claimant Optibase
commenced arbitration proceedings against Merrill Lynch and one of its sister
companies Merrill Lynch Investment Managers (MLIM). Optibase claimed
compensation for the losses it had suffered through an investment fund
recommended by Merrill Lynch, with which Optibase had an agreement to
arbitrate. MLIM served as an investment adviser for the fund, however Optibase
had no arbitration agreement with MLIM. The Court of Appeals held that
Optibase had failed to adduce facts that supported the contention that MLIM
should be forced to arbitrate in the absence of an agreement. The Court
specifically rejected the argument that MLIM could be bound to the arbitration
agreement between Optibase and Merrill Lynch by virtue of it being an agent of
Merrill Lynch.

7.6.2. Assignment

Most legal systems accept the possibility of assignment of a contract together

with its arbitration clause. Some presume that the agreement automatically
transfers with the underlying contract. (66) Some consider that this may not be so
because of separability of the arbitration agreement. Conversely, separability
also means that if the assignment of the underlying contract is invalid, it is at
least arguable that the assignment of the arbitration clause itself may be
accepted, although that may be somewhat strained logic. Separability is
intended to make true consent to arbitrate meaningful and should not be a
means to find valid consent where it otherwise does not exist.

There are a number of subsidiary questions. The first is whether the contract
itself may be validly assigned. The second relates to the evidence of intent of the
page "517" various parties to allow for arbitral rights and obligations to transfer
in this way. There are again choice of law issues as to the validity of the
assignment. (67) Born again argues for the validation principle to uphold
assignment if either the law of the arbitration agreement or the law of
assignment would lead to validity. (68) This may make sense for the assignor and
assignee, but not ipso facto for the other signatory who must separately be seen
to have intended to allow the assignment to occur. This is particularly so if the
assignor would be taken out of the picture if the assignee is included as
assignment is one of the examples like subrogation where one party is at times
seen as being replaced by another. This would be particularly problematic from
a consent perspective where an assignment might have been made to avoid
liability or shield assets, in which case the other signatory would not have been
supportive.

There appears to be increasing agreement across common law and civil law
legal systems that a contract will automatically transfer to the assignee any
rights contained within it to submit disputes to arbitration. (69) Sinclair examines
case law in a number of countries, including the UK, the US, Italy and France,
concluding that we have ‘almost arrived’ at a uniform rule of international
arbitration by which an arbitration agreement is automatically assigned
together with the main contract. (70) However, he cautions that the issue is not
completely settled and that in the absence of express approval of the assignee,
the counterparty or both, the assignment of arbitration agreement together with
the main contract may be challenged. (71) However certain US Courts have denied
the assignment of the arbitration clause for lack of an independent consent on
the part of the assignee to be bound by the arbitration clause. (72) The English
Arbitration Act does not contain an express provision with regard to the rights
of assignees. However, section 82(2) of the Act defines a party to an arbitration
agreement as including ‘any person claiming under or through a party to the
agreement’. The prevailing view in the courts is that as a matter of English law,
an assignee of a contract may become a party to the arbitration agreement
contained in that contract, whether the assignment is legal or equitable. (73)

page "518"

Assignments may occur before the commencement of arbitral proceedings or

even afterwards, in which case there may be a requirement of tribunal consent.
(74) There would also be issues as to the assignee's acceptance of the earlier

stages in the proceedings.

7.6.3. Assumption

Assumption involves conduct showing acceptance. Here there is still a need to

find implied consent from the other parties. One form of conduct indicating
assent is simply to rely on the agreement in bringing a claim, although the
person's own behaviour cannot prove consent from the other party.

7.6.4. Third-Party Beneficiaries

Some legal systems consider that rights and obligations conferred on third
parties in a contract will allow those parties to have direct rights in that regard.
Traditional notions of privity would hold against this view and would simply
consider the express treatment of the third party to be a term to be performed
between the two signatories. Most legal systems have moved away from strict
notions of privity as there are many situations where efficiency demands that
such rights be afforded. (75) Where this occurs, the question is then whether the
third-party beneficiary of the contract rights is also subject to an arbitration
clause within the contractual agreement.

As always it is a question of intent and will depend in part on the drafting of the
arbitration agreement and the surrounding circumstances. There are again
choice of law issues as to whether the treatment of third-party beneficiaries
should be as per the law applicable to the arbitration agreement or the law
applicable to the underlying contract. (76) Consideration of the contract itself can
only show the intent of the signatories. Separate attention would need to be
given to the actions of the third party. The situation may vary depending on
whether the third party is seeking to rely on the arbitration agreement or is
instead challenging jurisdiction. The distinction is simply because if the third
party brings a claim which is not rejected jurisdictionally, that may itself be
considered a separate arbitral agreement or separate evidence of consent. In
other circumstances it is simply a question of intent regardless of whether the
third party brings or defends a claim. The Iran–US Claims Tribunal accepted
such a claim in Land Serve Inc. (77)

page "519"

7.6.5. Incorporation by Reference

An agreement may be extended where a signatory enters into a separate

relationship with the non-signatory and in doing so incorporates the existing
arbitration clause by reference. That would show the consent of those two
persons but not the other original signatory. The law applicable to that second
agreement would determine whether there had in fact been incorporation by
reference. (78) Incorporation by reference cases must be looked at alongside
formal validity principles.

7.6.6. Subrogation

Subrogation involves one party having the right to take over contractual rights
of another party. This commonly arises in insurance contracts where the insurer
may pay out a claim but then wishes to take over the rights of the insured
against persons who were at fault. There are a number of instances where
tribunals and courts have considered the insurer to be party to the arbitration
agreement as a result. As between the insurer and insured, it will essentially be
a question of contractual drafting as to whether this was their intent. As to the
other signatory, that will depend on the circumstances. In most instances they
may be benefited by the inclusion, hence a priori intent may be presumed.
However, there is nothing to stop the insurer assisting the insured other than as
a distinct party. Hence, there may be circumstances where involvement as a
party would not be ideal from the perspective of the original signatory. Where
the other contracting party knows that there will be an insurance policy and can
reasonably presume rights of subrogation, this can add to the evidence of ex
ante intent.

The general view would be that the party subrogated would have the same
rights and obligations as the original party. In some legal systems, claims can
still be brought against the original promisee. (79) Where national law varies as to
whether the original promisee remains a party, this would also impact upon
analysis of consent. Even then, presumptions are not clear cut. On the one hand,
the other contracting party may have further options of recovery but on the
other, may be subject to additional cross-claims.

7.6.7. Novation

Novation occurs where one person ceases to have obligations and is replaced by
another. This typically arises in cases where a new individual assumes an page
"520" obligation to pay or perform an act that was incurred by the original party
to the contract. The general view would be that the party novated would have
the same rights and obligations as the original party.

7.6.8. Ratification

Legal systems will commonly consider that there are consummated agreements
when a particular party entitled to do so, ratifies the behaviour of others. A
party may ratify an assignment or subrogation or novation. (80) Again there
would be choice of law issues. Born again argues for a validation principle which
would be particularly sensible if the ratifying person was the other original
signatory.

7.6.9. Legal Succession

There are also cases where entities are bound by operation of law regardless of
intent. These include succession such as through mergers or similar doctrines.
(81) The dominant view is that this may lead to conclusion. (82) Problems may

arise where a contract expressly calls for consent to any variation in corporate
identity. There may also be issues with interpretation of choice of law clauses as
to whether these purport to apply to questions of succession and the like.

One situation where the law provides for a successor is insolvency. Here the
question is whether receivers, administrators or liquidators may or must engage
in the insolvent company's arbitral processes. Insolvency may also be a bar to
the arbitration itself. This is discussed in section 7.15.

7.6.10. Guarantors

Courts and tribunals have taken different views as to whether guarantors

should be considered parties to arbitration agreements in the underlying
contract. Again it is a question of intent. In some cases this would not be
possible as the guarantee would have a separate and conflicting dispute
resolution clause. As a general rule, it should be difficult to incorporate a
guarantor into the underlying contract. Well drafted guarantees are quite
distinct in their legal basis and do not afford rights to normal contractual
defences. Even where defences are possible, guarantees will generally not allow
the guarantor to raise them directly. If there are subrogation page "521" rights,
that itself would be a separate reason to consider inclusion. If the guarantor has
had a significant role in the history and conduct of the transaction such as a
holding company which is heavily involved in the transaction, it might be
included for that reason and not simply because of its specific guarantor
position.

7.6.11. Parent Companies and Significant Shareholders

Under some company law systems, shareholders may bring actions to protect
their interests. It has been suggested that where the applicable national
corporate law allows the shareholder to act on behalf of a company signatory to
an arbitration agreement, a shareholder might be permitted to invoke the
arbitration clause. (83)
7.6.12. Company Groups

There has been much debate about the circumstances, if any, where a related
non-signatory company can rely on or be included in an arbitration agreement
signed by its related entity. The most famous instance of the application of a
supposed group of companies doctrine occurred in the Dow Chemical case
reviewed by French courts. An arbitration clause was extended to other
companies in a group where they, ‘by virtue of their role in the conclusion,
performance, or termination of the contracts containing said clauses, and in
accordance with the mutual intention of all parties to the proceedings, appear to
have been veritable parties to [the] contracts or to have been principally
concerned by them and the disputes to which they may give rise.’ (84) The
tribunal considered that the group of companies constituted ‘one and the same
economic reality…’. (85) This case and cases that have followed it are highly
controversial, although it is difficult to critically evaluate decisions without a
careful analysis of the facts in issue. The group of companies approach has had
less support in other jurisdictions although one should be careful to distinguish
comments which simply suggest that a group of companies alone is not
sufficient from those that are at least prepared to consider the surrounding
circumstances. (86)

One reason why it is more controversial is that it has simply developed in the
arbitral context unlike some of the other concepts that are merely transplants
from or applications of contract, agency and abuse of rights laws. (87)
Importantly, it is page "522" something of a misnomer to describe it as a group
of companies ‘doctrine’, suggesting that simply because a company is part of a
group, extension is for that reason permissible. That would be wholly improper,
ignoring the essential requirement of consent and ignoring the purpose and
status of separate legal entities. (88) Group of company scenarios leading to
inclusion should simply be those where contemporaneous evidence and
circumstances make it reasonable to conclude that related companies were
sufficiently involved to be reasonably assumed by all to be subject to arbitral
rights and obligations. (89) It is generally a question of intent in the
circumstances. Even in Dow Chemicals the tribunal made its comments in the
context of noting that it should ‘reach its decision regarding jurisdiction, by
reference to the common intent of the parties to these proceedings, such as it
appears from the circumstances that surround the conclusion and characterise
the performance and later termination of the contracts in which they appear.’

If it is a question of intent, attention then needs to be given to the relevant

evidence, given that the issue will only arise when contested. Bernard Hanotiau
argues persuasively that the issue is not whether there is a group of companies
but instead whether both signatories and non-signatory parties intended to be
bound by the arbitration clause. (90) Mantilla-Serrano suggests that intent is to be
determined by ‘examining the parties' positions and actions’ and that this
involves a subjective analysis where the non-signatory's conduct plays a
paramount role in determining whether it has agreed to be bound by the
arbitration agreement. (91) Derains suggests that a group of companies provides
no objective rule to resolve multi-party scenarios and is nothing more than a
relevant factor in assessing intent. He notes that it is an ambiguous factor as the
mere presence of a group of companies could allow for conflicting arguments as
to whether there was true intent to be bound by an agreement signed by one
only. (92) In some cases it will show that the parties treated all companies as
being part of the one transaction. In other circumstances there may be an intent
to only contract with a subset of the group. (93) Rubins suggests that the most
important factual circumstances supporting intent are ‘the non-signatory is part
of a group of companies including at least one signatory entity; the companies
within the group are so intertwined in their activities and (more importantly)
responsibilities that they constitute a single economic reality; and the non-
signatory page "523" played an active role in the conclusion, performance,
and/or termination of the contract’, (94)

It is important to note that the Dow Chemical case dealt with the issue of
whether non-signatories may be claimants in arbitration rather than whether
non-signatories can be drawn into arbitration as respondents. In this regard,
Rubins notes that the Tribunal cited in support of the group of companies
doctrine the decision of a US arbitral tribunal that explicitly limited its reasoning
to cases where an arbitral clause is to be extended to claimant non-signatories.
(95) This ought not be presumptively easier. Because it is essentially a factual

question, there should be no presumption either for or against jurisdiction

depending on whether the non-signatory is a willing claimant or a reluctant
respondent. (96) Craig, Park, Paulsson suggest that it is more easily established
when they are claimants. This should certainly not be taken as a doctrinal
position. It is simply different because it changes the focus onto the likely intent
of the signatory rather than the non-signatory.

Nevertheless, actual consent will not be the key in many cases. Group of
company scenarios could also be dealt with via agency, including ostensible
authority, alter ego or piercing the corporate veil. In that sense it is important to
understand that it is not truly limited to separate companies but involves the
consideration of any circumstance where arguably related entities are
sufficiently similar to all justifiably be involved. There could be different legal
persons such as companies, partnerships and in some jurisdictions trusts, or
also the involvement of individuals such as dominant shareholders. It may
involve corporate officers or States or State entities. (97) It may also arise in trust
situations or estoppel scenarios where there was conduct leading to reliance.
There may also be situations of abuse of rights or acceptance by conduct. The
group of companies scenario should thus be asked to fit into one of two
conceptual models, inclusion where there is sufficient evidence of implied
consent or extension where unreasonable behaviour on the part of the non-
signatory makes it only fair and just that it be brought into the proceedings. In
some cases it is reasonable to presume that all members of a group intended to
be bound and it may only have been an oversight that the written contract was
drafted the way it was. At the other extreme, parties may have deliberately
sought to use related entities to shield themselves from any adverse arbitral
awards. It is inappropriate as some cases have done, to simply look at the
commercial relationship, the degree of control and the supposed needs of page
"524" international commercial relations. (98) Notwithstanding the better view of
the bulk of commentators, a more efficiency-oriented logic of this nature was
utilised by a US Court of Appeal in Ryan. (99) In Ryan the Court found that if a
company has to start two proceedings, one in court and one before an arbitral
tribunal in relation to ‘inherently inseparable facts’, the arbitration proceedings
would be rendered ‘meaningless and the federal policy in favour of arbitration
effectively thwarted’. (100) In deciding to extend the arbitration clause to a non-
signatory parent company, the Court stated that ‘[w]hen the charges against a
parent company and its subsidiary are based on the same facts and are
inherently inseparable, a court may refer claims against the parent to
arbitration even though the parent is not formally a party to the arbitration
agreement’. (101)

In the Jaguar case, the principle was also used by the non-signatory as a shield
to court proceedings. The Cour d'appel held that the clause can ‘extend to
parties directly involved in the performance of the contract provided that their
respective situations and activities raise the presumption that they were aware
of the existence and the scope of the arbitration clause, so that the arbitrator can
consider all economic and legal aspects of the dispute.’ (102) The principle was
applied to subsidiaries involved in contract performance in Alcatel. (103) It may
be that the articulation of the principle was overly broad as it concentrated on
the intent of the non-signatory. For it to be used as a bar to court proceedings, it
ought to be necessary to show that the claimant in those proceedings ought to
have understood the non-signatory to be party to the arbitration agreement.
Other methodologies have also been proposed. In Arthur Andersen v. Carlisle, the
US Supreme Court considered that State contract law would be determinative as
to whether a non-signatory has rights under a contract containing an arbitration
clause. (104) Trade usage was seen as a justification where the common law was
thought to otherwise exclude extension in ICC Case No 6000 (105) where an
arbitration clause was extended to a related entity that was involved in the
execution of the contract page "525" but was not a party to it. (106) In that case,
the related entity was fully involved with the conclusion, performance and
termination of the contracts in dispute. Nevertheless, the logic is problematic. It
is particularly strange to argue in favour of a group of companies theory based
on international trade usage, (107) given that separate corporate identity must
surely be one of the most significant elements of such usage.

A similar issue to a group of companies scenario is a joint venture situation.

Joint ventures may often be required as a precondition to approved foreign
investment. The very existence of the joint venture shows strong involvement
by the non-signatory, but under a consent analysis, there is again the need to
consider why only one entity in the venture executed an arbitration agreement.

7.6.13. Estoppel by Conduct, Good Faith and Related Concepts

At the outset, it was suggested that while extension/inclusion situations should

be analysed from the perspective of consent, a number of categories clearly lack
subjective intent by the third party but circumstances may nevertheless support
contentions that they should be bound. Different legal families use different
concepts to deal with what is said to be the justice in such scenarios where
persons act in unreasonable ways leading to detrimental reliance by others. In
many cases, legal systems will suggest that if a person induces a
misapprehension and causes detrimental reliance, it might be bound in those
circumstances. Civilian legal systems rely on principles such as abuse of right,
good faith or venire contra factum proprium. (108) Common law systems tend to
utilise concepts such as estoppel, acquiescence and waiver. In most cases, such
concepts are used to justify lifting the corporate veil or asserting that a third
person is an alter ego of a signatory. Specific comment on this is left to the
sections dealing with those categories. Only some general observations are
made at this stage.

From a conceptual level, it is first necessary to consider whether expanding

coverage in this way is consistent with the essential nature of arbitration. To
some, that could only be so where consent can be validly implied. Because
consent should be ab initio consent identified objectively on the assumption that
actions are deemed in good faith, extension can be said to be impliedly consent
based in some such cases. For example, this could be justified where the
evidence pointing to involvement outweighs sham elements seeking to point to
a contrary conclusion. Another scenario is where there is some inadvertence. An
example is the Deloitte case where an international accounting firm entered into
contracts containing page "526" arbitration clauses with regional affiliates. The
entitlement to use the international firm's name was conditional upon
acceptance of the agreement. One affiliate did not sign the agreement but used
the name nonetheless. Estopping the affiliate from denying consent is simply
another way of asserting that the decision to use the firm's name knowing that it
was conditional upon acceptance, is evidence of acceptance by conduct. (109)
Similarly, where a non-signatory purports to use an arbitration clause or does
not object to jurisdiction when defending a claim under it, this could either be
looked at as evidence supporting ab initio implied consent or could be looked at
as a new agreement to arbitrate. Where the non-signatory brings a claim, if the
other party does not object, then the claims and defences could be argued to be
a new arbitration agreement. The important distinction is the fact that a non-
signatory bringing a claim may not be valid proof that it always intended to be
bound, but simply that it sees a strategic value at the time the dispute is known.
Hence a tribunal needs to be more circumspect in its analysis.

In other cases there remains the need to decide the conceptual validity of
jurisdiction via estoppel and then consider whether the elements of the
inclusion principle being relied on are made out on the facts. There are
particular choice of law difficulties in applying such doctrines given that it
would be hard to determine which of the relevant States' laws ought to apply.
Born again argues for international principles of estoppel and good faith. (110) It
may be easier to justify inclusion conceptually if the law applied was no more
expansive than the law which the third party might reasonably expect to apply
to its conduct. This remains a challenging question. Even when the law is known
there are then a host of evidentiary issues. The important question, rarely
articulated in the literature, is exactly what kind of conduct ought to lead to
arbitral jurisdiction. If the assertion is that it would be unjust not to include,
what is the factual basis for the conclusion? It cannot simply be that a person
with funds who would benefit if the deal was successful, should be liable if it is
not. That could have been factored into the contract price. This is also the
essential motivation for the development of limited liability companies. Hence
the situation will be less contentious where the facts show the person led the
signatory to reasonably believe that they would accept responsibility. Where
arbitral jurisdiction and non-signatories are concerned, the behaviour has to be
looked at as against the reasons why the non-signatory was not included in the
contract in the first place. The more contentious cases are where tribunals
simply concentrate on the third person's involvement and control and draw
broad estoppel type conclusions the more these factors are present. For
example, in Hughes Masonry Co v. Greater Clark County School Building Corp (111)
a construction manager interfered in a contract between a masonry contractor
and a building owner. The contractor's page "527" agreement with the owner
contained an arbitration clause, however the manager's did not. The Court of
Appeals for the Seventh Circuit found that the contractor, in bringing a suit
against the manager, was equitably estopped from refusing to arbitrate because
the basis of its claim was that the manager had breached the duties and
responsibilities assigned and ascribed to the manager under the agreement that
contained the arbitration clause.

Hanotiau notes that equitable estoppel has only rarely been applied to compel a
non-signatory to arbitrate. (112) For example, in Thomson-CSF, S.A. v. Am.
Arbitrations Association (113) a parent company sought a declaration that it was
not bound by an arbitration agreement between one if its subsidiaries and a
supplier. The supplier cross-moved to compel arbitration with the parent. The
United States District Court for the Southern District of New York denied the
parent company's request. This decision was overturned by the Court of
Appeals. The Court of Appeals stated that the parent company could not be
estopped from denying the existence of an arbitration clause to which it is a
signatory, because no such clause existed. (114)

Again as noted above, some of the more contentious decisions could be looked
at from the perspective of objective evidence of consent outweighing the
contrary factors. The degree to which a party has been involved in the
development and performance of the underlying agreement and the extent to
which its conduct has led to reliance by the other party, are two factors
commonly referred to in cases looking at extension via conduct. It is important
that the factors considered are consent related. Stated differently, if there is to
be an estoppel, it is important to consider exactly what the third party is being
estopped from denying. Often they cannot deny that they wished to have control
over the transaction but equally, they may legitimately have wished to impose a
limited liability company to shield themselves from recourse.

Finally, Born notes the debate as to whether estoppel should apply more as a
shield or as a sword, operating differently depending on whether it is used by a
signatory or non-signatory. He argues persuasively that this should not be the
case. (115) Consent logic certainly cannot have any a priori presumptions of this
nature.
7.6.14. Corporate Veil and Alter Ego

Courts will at times ignore a corporate edifice if it is considered to be a facade or

sham. This can otherwise be described as an alter ego scenario. (116) Born
suggests page "528" that ‘(t)he essential theory of the “alter ego” doctrine is that
one party so dominates the affairs of another party, and has sufficiently misused
such control, that it is appropriate to disregard the two companies' separate
legal forms, and to treat them as a single entity.’ (117) Lifting the corporate veil
has been accepted under international law in the Case Concerning the Barcelona
Traction, Light & Power Co. (118) While a corporate veil analysis is commonly used
to bring in a person with a controlling interest, the logic may be used by such a
person in support of a stay of litigation application. The third party may be more
litigious, have deeper pockets and may be advantaged by the place of arbitration
or the selected tribunal. (119)

Where the principle is being used to include a reluctant controller, piercing the
corporate veil will be closely related to situations of abuse of rights. Corporate
veil scenarios could also be dealt with under other mechanisms of extension
such as agency, (120) third-party beneficiary, assumption or estoppel. (121)
Similarly corporate veil theories need to be distinguished from group of
companies doctrine, although the latter is often argued in such circumstances.
The logic is different however. A corporate veil approach is based on company
law theories and whether separate legal identity can be ignored. Group of
companies logic operates on contract theories and intent. (122) Besson makes the
observation that there would be complex choice of law issues that arise. Which
company law should be looked at for notions of piercing the corporate veil, that
of the signatory company or the non-signatory? (123) In deciding on applicable
law, some look to the law of the State of the company whose corporate veil is
sought to be pierced. (124) Born argues against and in favour of an international
solution. (125) A national solution can pose particular problems when the aim is
to look behind a State-controlled entity page "529" with a view to bringing the
State in as a party to proceedings. Besson has also made the important point
that a corporate veil may be pierced for liability but not necessarily for
jurisdiction. (126) That may naturally apply where only recognised courts are able
to provide remedies through a judicial determination of abuse of the corporate
form.

Where inclusion is based on notions of intent, there would rarely be sufficient

evidence to justify piercing the corporate veil to afford arbitral jurisdiction.
Under a consent theory, there needs to be appropriate evidence that all parties
intended to be bound. Yet if a party established a shell company to protect those
with deeper pockets from recourse, its subjective intent is to the contrary and
objective evidence points in the same direction. Hence, consent must arise by
some deeming theory based on fraud, abuse of rights, misrepresentation or lack
of good faith. The difficulty is to identify what constitutes misuse, fraud or other
injustice vis-à-vis arbitral jurisdiction. Simple control is not enough, although if
control and involvement is sufficient, there may be an intent logic and a group of
companies scenario rather than piercing of the corporate veil. Once again, the
mere fact that the signatory has no assets should have little, if any, significance.
The very essence of separate corporate identity is to alter the risk/reward
matrix in commercial transactions, offering contracting parties in their dealings
with corporations' greater resources and greater potential profits in return for
limited recourse in some scenarios at least.

While mere control should thus not suffice, inevitably some cases will
concentrate on these issues. In Bridas, (127) a US enforcement court dealt with an
attempt to bring in a State as liable for its State-owned entity signatory. The US
court considered a range of factors, namely, whether the national law would
consider the entity an arm of the State; the source of its funding; its level of
autonomy; whether it has a local or State-wide focus; whether it can sue and be
sued under law; whether it may hold or use property; and where corporate
relationship is concerned, common stock ownership; common directors or other
key officers; common business departments; consolidated financial statements;
source of funds; cause of incorporation; inadequacy of capital; payment of
salaries and expenses; whether there is any external business not sourced from
the parent; how the parties use the subsidiary's property; whether daily
operations are kept separate; observance of corporate formality; whether the
directors of the subsidiary can be said to act primarily in the interest of the
parent; guarantors and payment of debts; and generally whether dealings are at
arm's length. (128)

In some cases, solutions are sought in one area of arbitral practice because of
potential inadequacies in others. For example, where a corporate controller
seeks to siphon off the assets of a defendant company in the face of litigation,
interim page "530" injunctions can more readily be sought and a national court
will have jurisdiction over the shareholders as well as the corporation. The same
cannot be said where arbitration is concerned. An interim measure against the
company itself is problematic until jurisdiction is determined. No award can be
made against a controlling shareholder unless it is a party to the agreement. In
many instances a tribunal has no interim measure powers in any event. (129)

7.6.15. State Entities and States

While most would see the involvement of non-signatory States or State entities
in multi-party scenarios as raising the same questions of consent, there are
other complex questions of law that might impact upon the analysis. There
would be issues of the public law of the country concerned, both as to
constitutional allocation of powers, administrative law ambit of authority, and
the ability of State entities to bind the State. In that context there is an important
distinction between administrative law principles of delegation of powers to
grant authorisations on the one hand and conferring of contractual entitlements
including in relation to arbitration agreements on the other. A government may
properly authorise the right to enter into a discrete arbitration agreement
without taking on jurisdictional obligations itself. (130) International law
principles of attribution would not naturally apply in commercial arbitration but
might be raised in investment disputes where international law is itself stated to
be applicable. (131) The relevance of the State and international law principles
may also be complicated in privatisation scenarios where it may be a matter of
timing as to whether the substantive rights are with the government or instead
with a privatised agency. In due course it is certainly the natural corollary that
the State is sheltered where it has privatised certain functions. (132)

In addition to these legal questions, there are also distinct factual questions such
as the way decisions are actually made, the control that government ministers
and other officials have over entities and the use to which revenues are put and
the page "531" source of expenditures. Close control and consolidated revenue
can add to the factual matrix of intent. There may also be problems where the
State can simply liquidate an entity either for honourable regulatory reasons or
as a means to avoid arbitral liability. (133) Petrochilos suggests that where States
are concerned, the evidence should be as to a contractual intent rather than
simply an intent to exercise administrative supervision. (134) This distinction led
to the setting aside of the award in the Pyramids case where the Paris Court of
Appeal considered that the relevant Ministry's notation that an agreement was
‘approved, agreed and ratified’ was simply an administrative step and not an
agreement to itself be bound. (135) In ICC Case No 8035, a tribunal took a similar
view and refused to consider Libya a party simply because its representative
had noted on a suspension agreement that it was ‘approved and endorsed’. (136)
Conversely, Born criticises the decision in Pyramid as being inconsistent with
international authority and difficult to reconcile with the language used by the
Ministry. (137) Neither view is inherently preferable. Problematic drafting and
ambiguous facts will always lead to understandable divergences of view.

The recent case of Dallah has also been controversial. Enforcement was refused
in England when an ICC award issued in Paris had sought to include a non-
signatory State. (138) Dallah was a company which provided services for pilgrims
travelling to holy places in Saudi Arabia. In July 1995, Dallah signed a
memorandum of understanding with the Pakistani Government in relation to
the construction of certain housing for Pakistani pilgrims. In 1996, Dallah
executed a contract with the Awami Hajj Trust, a body which had been
established by an Ordinance promulgated by the then President of Pakistan. The
contract contained an arbitration agreement, under which all disputes were to
be referred to ICC arbitration in Paris. The Government was not a signatory to
the contract, although the contract made reference to a guarantee to be
provided by the Government and included a provision by which the Trust could
assign its rights and obligations to the Government without the permission of
Dallah. The housing project was never commenced and a change in government
occurred shortly after the contract was executed and relationships deteriorated
rapidly. In May 1998, Dallah commenced ICC arbitration proceedings against the
Government of Pakistan. Dallah page "532" successfully argued that the tribunal
had jurisdiction over the government. The tribunal awarded Dallah
approximately USD 20 million in damages and costs. Dallah endeavoured to
have the award enforced in the UK and in Paris. The UK Supreme Court was
ultimately faced with the question of whether the Government should be
considered a party to the arbitration agreement. The Court held that there was
no material sufficient to justify the tribunal's conclusion that the Government
was a party to the arbitration and therefore refused to enforce the award in the
UK.
The outcome in Dallah might be justified on the basis of the distinction between
administrative control and contractual intent alluded to above. However, if too
artificial a distinction is made, States might be put in a privileged position as
against group of companies scenarios, requiring explicit evidence of intent to be
bound. (139) That might more readily apply where the State is given rights and
obligations under the contract and there is an acknowledgment that it intends to
be bound. (140) In ICC Case No 9762, a tribunal considered that the fact that the
agreement was entered into by a Minister, subsequently replaced by another
Ministry, led to attribution to the State under international law and agency
principles. (141)

Problems may arise with State entities when key evidence of the State's
involvement occurs after the consummation of the agreement. The intention to
be bound should be found at the outset. The fact that a State has the legal power
to frustrate an existing contract it is not privy to should not give rise to an ex
post facto extension of jurisdiction. That potential is always part of sovereign
risk and the contracting parties could have made that a negotiating feature from
the outset. (142)

7.7. Concluding Remarks

The following discussion attempts some concluding observations in relation to

multi-party scenarios.

7.7.1. Consent-Based Analysis and the Nature of Consent

As noted at the outset, all arbitration is based on consent. Hence any justifiable
theory of multi-party scenarios must be consistent with the evidence of consent
in page "533" the instant case. Even when there is some consensus that consent
is required, an important question is whether that is actual subjective consent,
whether it is subjective consent proven by objective features or whether purely
objective consent is the required factor. The latter may more readily allow for
estoppel or corporate veil type scenarios. There is also a significant difference
between evidence of ex ante intent and instead evidence of subsequent
behaviour that might be explicable for a range of reasons. Original consent is the
key. (143) Furthermore, there is a fundamental difference between evidence of the
actual thinking of the persons concerned and arguments based on the likely
intentions of the parties when presumed to be acting in good faith. Arguments
based on presumed intent are particularly problematic in this context. While the
logic in favour of inclusion is that persons would prefer efficient dispute
resolution avoiding the possibility of conflicting decisions and abusive
behaviour, nevertheless the mere decision to set up a separate corporate entity
for the benefit of publicly afforded limited liability cannot be seen as abusive
and must a priori be an equally valid presumption of modern business persons.

7.7.2. Party Autonomy and Drafting

Problems can be overcome if appropriate arbitration clauses are well drafted.

Where a group of companies is concerned there would still be a need to have
every company in the group sign to be sure that there is valid jurisdiction.
Where a group of contracts is concerned, there would be a central arbitration
agreement, express indication that related contracts are to have an identical
clause and a requirement that parties to the ancillary contract agree to be bound
by the framework agreement. Thought might be given to an umbrella clause or
an umbrella arbitration agreement that all parties to the various contracts are
required to sign. (144) While that may be an ideal, it is unlikely to be practical in
most circumstances and in any event the vagaries of international transactions
will inevitably throw up permutations beyond the contemplation of the
transactional lawyers. (145) Finally, parties might wish to consider whether there
should be a cut-off point beyond which applications for consolidation or joinder
should not be made.

page "534"

7.7.3. Interpretation of Ambiguous Clauses

While the natural conclusion is to try and deal with multi-party scenarios in
drafting, complex factual permutations will still require interpretation of intent
and determinations as to the relevance of good faith. Furthermore, the more
elaborate an attempt to draft an agreement covering multi-party situations, the
more that an adjudicator may be loath to fill gaps based on good faith, arguing
that the parties went to great bother to articulate exactly which categories were
to be included. While a contrary approach is equally possible, it is salutary to
remember this potential problem with any form of elaborate legal drafting.
Notions of consent are also bound up in the question of the proper
interpretation of an arbitration agreement. While consent must be looked at in
the context of the presumption of separability, meaning a distinct intent to be
bound by the arbitration agreement itself, (146) nevertheless determinations of
consent must be looked at in the context of commercial reality. It would rarely
by the case that a party wishes to be bound by an arbitration agreement without
being bound to some underlying commercial transaction.

Where interpretation is concerned, there ought to be no presumption either in

favour of or against inclusion, consolidation or joinder. Pro-arbitration or
provalidity presumptions should not apply in inclusion scenarios. Categories
based on consent must look for that fact. The absence of a signature to an
arbitration agreement tells against any factual presumptions in favour.
Similarly, extension via good faith and justice analysis requires very strong
evidence of culpable behaviour. Hence again there can be no presumptions in
favour. There can however be pro-efficiency assumptions at least in
circumstances where it makes no real sense to separate proceedings. Effete utile
is a logical presumption with pathological clauses where the signature shows a
good faith willingness to arbitrate but where the terms are unclear. The same
logic does not hold for a non-signatory. If the analysis is truly consent based and
parties could have resolved the situation one way or another in their drafting, a
tribunal may need to turn its attention to the reasons why that has not occurred
and what reasonable expectations can be inferred from the contractual
negotiations. Privity may no longer be as binding a legal norm as it once was, but
it is still relevant evidence of intent.

In some cases the means by which a party is alleged to have consented may be
dependent on ambiguous terminology used. For example, in Arab Republic of
Egypt v. Southern Pacific Properties Ltd & Southern Pacific Properties (Middle
East) Ltd (147) an ICC Award was set aside by a French court where it had held
that Egypt was a party to an arbitration agreement by reason of the Minister of
Tourism's signature appearing at the end of the main contract under the words
‘approved, agreed and ratified’. To the extent that an adjudicator is willing to
page "535" look at all extraneous circumstances to discern what the parties
intended in good faith, the involvement of non-signatories in central roles in
bringing the transaction to fruition may support a broader analysis. There may
also be circumstances where the consent analysis is complicated in
circumstances where there might be no good faith reason to refuse consent.
Voser uses the example of a disputed ownership of property where a third party
claims better title to that of the original disputants. (148)

Where a signatory respondent seeks to raise claims against a third signatory

party, the better view is that this should be allowed for a range of reasons. First,
signatories to joint agreements must have understood that they could be
included in single proceedings. Secondly, if claimants are allowed to bring
actions against multiple respondents privy to the arbitration agreement, the
same right should be afforded respondents. That is particularly so as it will often
be simply a question of timing as to who commenced proceedings first and
hence who has the natural claimant's right to assert against as many parties as
are subject to the arbitration agreement. (149)

7.7.4. Consent to Laws and Rules

Arguments as to consent become complicated where the parties do not agree at

the time of the multi-party request but have selected a Seat or arbitral rules that
give tribunals broad powers to bring in third parties. On one view, true consent
requires consent at an earlier time. On another view, the consent to the
procedural framework acknowledges the discretion and contemplates the
inclusion. Even on the latter view, this is conceptually different as at most it is
consent to consideration of the question. If that party is nevertheless reluctant
to allow third-party involvement, its views and reasons ought to be a relevant
factor in the exercise of tribunal discretion.

7.7.5. Choice of Law

If multi-party determinations and application to non-signatories are dependent

on implied consent, problems arise with selection of applicable law. Utilising the
law of the contract or the law of the arbitral situs ‘may involve a circular
exercise that presumes its conclusion when identification of who agreed to
arbitrate constitutes the very question to be decided. In addition, the contract's
applicable law, and the law of the arbitral Seat, will be foreign to an entity that
remained a stranger to the page "536" transaction.’ (150) The relevance of choice
of law questions to this analysis, whether the application of national laws or
international principles is bound up in broader debates as to the approach one
takes to contentious arbitral matters. Rather than opt uniformly for national or
international approaches, it might be preferable to consider both under a
cumulative logic, feeling confident in inclusion where each key system would
support that conclusion. Where this would not be so, attention might then be
given as to which legal system the parties might reasonably have expected to be
determinative in these scenarios. Born argues that implied consent should be
determined by the same law as governs the arbitration agreement. Born has also
argued for an international approach rather than a domestic conflicts analysis.
(151) Adopting an international approach then requires attention as to the

methodology by which international principles are to be discerned. Here there is

a great difference between an expansive approach, on the one hand purporting
to establish principles thought to be appropriate for international markets and
on the other hand, a conservative perspective, which would only identify an
international principle where it has been recognised by a significant number of
national systems. In the latter event, there may be little difference to a national
approach where the relevant jurisdictions adopt those norms.

Whether one adopts national or international solutions to inclusion would

naturally follow one's predisposition to interpretation of arbitration agreements
generally. (152) If general principles of good faith are involved these might not be
limited to the countries of incorporation and may also be impacted upon where
there is a choice of law clause in the contract, although that agreement does not
a priori bind the non-signatory. An alternative strategy is to consider which
international principles could form part of lex mercatoria.

7.7.6. The Evidence of Consent

The various reasons why non-signatories may come within an arbitration

agreement are all fact dependent. ICC Case No 9517 noted:

The question of whether persons not named in an agreement can take

advantage of an arbitration clause incorporated therein is a matter which must
be decided on a case-by-case basis, requiring a close analysis of the
circumstances in which the agreement was made, the corporate and practical
relationship existing on one side and known to those on the other side of the
bargain, the actual or presumed intention of the parties as regards right of non-
signatories to participate in the arbitration agreement, and the extent to which
and the circumstances under which non-signatories subsequently became page
"537" involved in the performance of the agreement and in the dispute arising
from it. (153)

Tribunals and courts have looked at reference to the third party in the contract
clauses, awareness by the party of the arrangement, involvement in
negotiations, approval and control, performance of the contract, frustration of
the contract and termination. At all times the question should not simply be
what involvement the party had but whether that involvement goes far enough
to indicate implied consent on its behalf. In some cases consent of one party may
be imputed to another, for example, where consent of an agent may be imputed
to a principal. A similar scenario is where there is succession or assumption. (154)

There is also a danger in simply looking at what tribunals say about consent as
compared to analysing what they are actually doing on a case-by-case basis. Too
often consent language is simply used to mask what is in essence a subjective
view about the respective equity of the persons involved. (155) In some cases,
merely resorting to articulations by courts or tribunals in previous cases would
lead to principles that cannot truly be identified as consent. (156) While most pay
lip service to consent, much depends on the evidence that is required. As noted,
if consent is determinative, it should be ex ante consent, except in circumstances
where it can be argued that the parties have assumed the rights and obligations
under a contract with an arbitration clause. (157) This was a problem with over-
use of the Dow Chemicals verbal formulations in later cases. These comments
refer to the non-signatory's involvement in the conclusion, performance or
termination of the contract. There is nothing wrong with the tribunal's logic as
long as it is always seen within the context of intent to be bound by arbitration.
There is a significant difference between taking actions to frustrate a contract
that one is not privy to and engaging in a transaction in a way that demonstrates
an intent to be bound. A transport company can simply refuse to carry goods
subject to a sale of goods contract, but does not become privy to that contract by
reason of doing so.

page "538"

In some cases intent has been found by way of trade usage. (158) If trade usage is
to be referred to, it ought to be based on appropriate evidence. Trade usage is
highly problematic in most scenarios as the very use of limited liability
companies to shield recourse is a most central usage of international trade and
investment. While claimants may often wish to extend arbitration clauses to
non-signatories who have the funds to meet an award, it is extremely difficult to
justify this under a consent paradigm. The whole notion of limited liability
companies is to alter the risk/ reward ratio in commercial scenarios. A
corporation may afford greater potential profits to a contracting partner on the
understanding that if circumstances are less than satisfactory, there will be
limits on recourse. On this logic, there would need to be some evidence of fraud,
abuse or subterfuge that would allow the concepts to come in aid of finding a
party with deeper pockets.

7.7.7. Two-Way Consent Analysis

Hanotiau has made the important observation that one should consider whether
consent could also have been found in a reverse situation, where the parties are
swapping positions as claimant and respondent. (159) A more strained implied
consent scenario is to consider what will happen if inclusion is denied. The more
that the outcome would be seen as problematic on an ex ante basis, the easier it
is to imply consent to inclusion. The converse is true where the non-signatory
clearly gets an adjudicatory advantage in such circumstances but would not go
so far as to be seen as an abuse of rights or lack of good faith.
7.7.8. Estoppel-Based Approaches to Consent or in Place of Consent

Actual or even implied consent is not particularly relevant to some of the

scenarios such as estoppel or piercing the corporate veil. In these circumstances
the behaviour of the targeted party is such that the law will not accept its denial
of consent. Consent theories are also strained where equitable notions give rise
to contractual liability such as where there is an agent for an undisclosed
principal. One can certainly argue that there was implied consent where that
scenario arose, although that may not be so given that jurisdictional rights are a
two-way street.

Where the frustrating behaviour of the related corporation could not have been
reasonably anticipated at the outset, is this a basis for extending jurisdiction?
There page "539" is also some uncertainty as to whether the logic is based on
actual mutual intent or is instead based on reliance through the involvement of
the non-signatory. (160) While detrimental reliance may be a tenable logical
construct, one would rarely find that the extraneous circumstances were so
strong as to override the decision to not include the non-signatory. Examples of
improper behaviour leading to at least arguable justifiable confusion by a
signatory might be refraining from publicising some of the complexities of a
corporate group, particularly where assets are shielded; corporate officers who
are senior members of a range of related companies who fail to adequately
distinguish on what basis they are acting; express misrepresentation and
subsequent behaviour to strip the signatory of assets purely as a defence against
enforcement; or negotiations being conducted on behalf of the group with a last
minute and not heavily publicised change of name in the written documentation
presented for signature.

7.7.9. Confidentiality

One concern with broad inclusion is in relation to confidentiality and the

inherently private nature of arbitration. There are both conceptual and practical
reasons why this should be less of a concern. First, if there is truly a valid basis
to find implied consent to multi-party arbitration, that implied consent must
include implied agreement to be engaged with the other parties and to give
them access to confidential material. From a practical point of view, the third
party is typically aware of much of the information in any event. (161) In addition,
specific orders as to confidentiality might be made.

7.7.10. Jurisdiction and Liability

As noted by Petrochilos and Besson, (162) separate attention must be given to

extension of jurisdiction and extension of substantive liability. If jurisdiction
extends to a third party, one might normally expect that substantive rights and
obligations would flow as a matter of course. That is not necessarily so and the
converse is also page "540" not inherently true. A third party could be bound in
relation to the underlying economic transaction but may nonetheless be outside
the ambit of the arbitration clause. (163) While it may be correct to say that in
most instances where a claimant bothers to seek to introduce a non-signatory,
substantive liability will follow jurisdiction, there is no conceptual link and it
must be possible that a non-signatory could be within the jurisdictional ambit of
an arbitration clause but not itself be found to be substantively liable. Classic
examples would be construction disputes where the search is for who is truly at
fault in the building work. Differences may also apply because of differing
evidentiary standards for jurisdictional and substantive determinations and also
because of separability of the arbitration agreement, where intent to be bound
by the latter does not necessarily connote an intention to be bound by the
underlying contract. (164) Differences between jurisdictional consent and
substantive liability may also arise where parties have made an effort to draft
their arbitration clauses to bring all related disputes together in one forum but
have at the same time expressly denied certain substantive rights and
obligations as between one person and another. For example, one agreement
might create a partnership while another may deny it. The same may be so with
agency. It may even be that it is expressly stated that two of the parties are not
in any contractual relationship. Thus jurisdiction may be found but substantive
rights and remedies may vary.

7.7.11. Tribunal Consent

A question has been raised as to whether the tribunal must also consent to a
multi-party scenario. Here there are a range of permutations. If the matter has
commenced as a multi-party arbitration, acceptance of an appointment
naturally constitutes consent to that process. A second scenario is where an
arbitrator accepts an appointment in relation to a multi-party clause where
there are initially only two parties involved. Here the better view is that the
arbitrator has properly understood the implications of the multi-party clause
and has accepted that further signatories may be joined in due course, although
in extreme cases logistical issues may justify an arbitrator resigning when the
joinder is not intended on terms that would be within the arbitrator's
reasonable expectations. (165) Another scenario is where a tribunal has a general
discretion whether to allow for joinder or consolidation in page "541" all the
circumstances. Here it is not a matter of consent per se but simply the proper
exercise of the discretion.

7.7.12. Form Requirements

Even where there is a consent-based logic to extending an arbitration

agreement to a non-signatory, consideration must also be given to the form
requirements for valid arbitration, typically a writing requirement where it
exists. There may also be form requirements emanating from the particular law
dealing with the relevant transfer of rights. In some cases this might require
written agreement in any event. Where the arbitration agreement must be in
writing there are added problems with inclusion of non-signatories. In some
scenarios, such as corporate veil, the controlling party goes to great lengths not
to be party to any agreement.

As outlined in section 3.6.1.6 some lex arbitri require a signature in certain

circumstances for valid arbitration. Others simply require some form of written
agreement or mere evidence to that effect. Others have followed Option 2 of the
UNCITRAL Model Law 2006 reforms and have removed the writing
requirement. Where there is some writing requirement there are two schools of
thought. The more restrictive will assert that the full express writing
requirement must be found vis-à-vis each and every party sought to be brought
within the ambit of the agreement. The contrary school of thought is that as long
as the underlying arbitration agreement meets the writing requirement, even
where third parties are not mentioned, the underlying agreement satisfies all
matters of form. (166) The treatment of form requirements in inclusion scenarios
is distinctly problematic. If they were required in full, there would be virtually
no circumstances where inclusion would be possible. If they are effectively
ignored by arguing that there simply needs to be one valid arbitration
agreement vis-à-vis the original signatories, then the form requirement is
effectively ignored where extension parties are concerned, which goes against
the logic of the requirement itself, requiring written evidence that someone has
given away litigation rights. Because the dominant view is that form
requirements have been overly restrictive, the more supportive position has
been preferred. (167)

7.7.13. Enforceability

Questions of enforceability are particularly important in this context as

inclusion decisions are jurisdictional decisions where an arbitrator's positive
determination page "542" will be subject to challenge before a supervisory
court. However, whatever approach a tribunal takes, enforceability challenges
might be made on the basis that an agreement has wrongly been extended to a
non-signatory and hence from the latter's perspective, there is no agreement in
writing, (168) or the tribunal has decided ultra petita (169) or there are problems
with composition and agreed procedure when additional parties are involved.
(170) There is also diversity between those who would simply look for an implied

intent based on all available circumstances and those who would adopt a
conflicts analysis which could play out again before an enforcement court.
Where conflicts are concerned, there are complexities as to which law from
which jurisdiction will apply. (171) As a general rule, however, if a tribunal has
followed one of the recognised categories of inclusion of non-signatories, a
respected enforcement court is not likely to interfere. Even where enforceability
may be a problem in some jurisdictions, section 2.7.14.3 argues that because of
the possibility of numerous places of enforcement and the fact that most awards
are honoured voluntarily, potential enforcement challenges should not dissuade
a tribunal from acting in ways that are otherwise thought appropriate.

There are still other enforceability issues. Even where it is appropriate to

consider extending an arbitration agreement, the target entity must be properly
informed of the application against it and be given every opportunity to make its
submissions and be involved in the proceedings if a positive decision on
jurisdiction is taken. In Altain Khuder, enforcement was blocked when neither
the arbitration agreement nor the request for arbitration expressly referred to a
sister company of the respondent where the award made orders against both.
(172)
7.7.14. Logistics

There are important issues of timing where third-party claims are made. To the
extent that any discretion is required, often an attempt will be made to join
before the constitution of the tribunal. Documents may be filed with the ultimate
determination to be made by a constituted tribunal, although for reasons noted
above, that may be problematic in terms of multi-party appointment. In an ad
hoc arbitration there might also be recourse to a court. Where an institution is
involved, either express or broad discretions would be utilised to determine
how to deal with such requests. It is particularly important that decisions are
made in a timely page "543" manner as to third-party involvement. If parties are
not to be included, there may be a need for separate proceedings which in turn
may need to keep an eye on elapsing limitation periods. Thus the norm is to
render preliminary awards on such questions. Such decisions may themselves
be challenged in a supervisory court. (173) If there are to be separate proceedings,
thought should be given as to how they might be coordinated if that would be
desirable. This is discussed in section 7.13

Where there are multi-parties, the tribunal will need to consider in what
proportion advances on costs should be paid. Modern institutions will interpret
their rules to allow for case-by-case allocations. (174) Article 43 of the UNCITRAL
Rules 2010 merely refers to requests to parties to deposit equal amounts.
Where one or more parties does not pay their share, the party pursuing claims
against them will be required to pay their share of the advance with the
possibility of reimbursement through a subsequent award. While this will
normally be an obligation on claimant, simply because the claimant failing to
pay its share will usually lead to proceedings being seen as abandoned, in multi-
party scenarios this could also operate against respondents where there are
cross-claims between them or counterclaims against a third party.

7.8. Multiple Claims and Contracts (175)

7.8.1. Introduction

A similar logic to multiple party scenarios is at times applied when a range of

interrelated contracts are involved. Problems arise where the contracts do not
have a comprehensive mechanism to bring all of the individual contracting
parties into the one dispute resolution forum. However, the evidentiary analysis
in a group of contracts scenario will be different to that of group of companies.
Where a group of contracts is concerned the analysis will tend to look at the
wording of the arbitration agreement relied upon from one of the contracts, how
broadly it defines matters related to that contract and the extent of
inconsistency with dispute resolution clauses in other contracts sought to be
incorporated. The actions and intentions of the relevant persons may still be
relevant but will have less impact on the final decision. Conversely, where group
of company scenarios are concerned it will be the actions and intentions of the
relevant persons that will best indicate page "544" whether all wish the
arbitration agreement to apply to non-signatories. (176) Of course it is possible
that the two scenarios are combined, with multiple arguably related contracts
dealing with multiple arguably related entities. (177) Where circumstances are
mixed, there may be good faith or estoppel type arguments as above, but this
would not flow from the presence of multiple related contracts alone. Hence this
section looks at the discrete question of multiple claims and contracts between
non-related entities where consent must be found.

When a dispute arises between parties there may be a multitude of claims which
are made. In part this is because international commercial relations are
becoming ever more complex. Often there are long-term relationships involved,
perhaps with framework and ancillary contracts. More than one might give rise
to disputes. At times there are differing contracts relating to trade and
investment on the one hand and payment and guarantees on the other. Within
any contract there may also be multiple claims flowing backwards and forwards
as to the performance of each party. (178) At times these may be non-contractual
claims that nevertheless relate to the central transaction. Where the parties
have agreed to arbitrate disputes a question therefore arises as to whether all
the claims between them, or only some, can be referred to the arbitral tribunal
which has been constituted. It is thus of fundamental importance to consider
how arbitration can or should deal with the entire range of multiple claims that
might be brought between the same parties.

The question posed presumes that the parties are not in agreement, as they may
of course agree between themselves to allow or bar claims or consolidated
tribunal hearings. As a question of consent, the various persons could make this
clear either way, choosing to incorporate identical arbitration clauses and
expressly defining their ambit to cover the group of contracts concerned. To
similar effect they could enter an umbrella arbitration agreement that covers a
range of listed contracts. Conversely they could expressly reject such broader
ambit or show this intent by utilising incompatible dispute resolution clauses.
Hence the real challenge arises where the parties have been less than clear in
their intentions. This is then a question of interpreting consent in all the
circumstances, in particular the wording of the clause or clauses used. Because
of this, only general theoretical perspectives page "545" can be posed. Where
arbitrators are asked to decide on such disputed preliminary questions, there is
a need to identify the principles by which such determinations should be made.
At times, arbitrators are given some discretionary leeway, in which case they
will naturally consider the practical ramifications of their decisions. It is
immediately obvious that if all claims are not dealt with in one arbitration
between the two parties, and if two or more proceedings are commenced, there
may be much less efficiency in terms of expense and time as well as the risk of
inconsistent decisions. However, if distinct claims are brought together against
the wishes of one party, this might offend against the very foundations of
consent as the basis of arbitration. It may also lead to a tribunal dealing with an
issue in situations better suited to a differently constituted tribunal.

Even if the different claims ought to be heard before different dispute resolution
bodies, each tribunal might then have additional procedural decisions it must
make in order to promote the greatest fairness and efficiency between the
matter before it and the parallel or sequential proceedings. At the very least,
each tribunal cannot ignore as a matter of course, the existence and procedural
implications of parallel proceedings. Parallel proceedings are discussed in
section 7.13.

7.8.2. Policy Reasons behind Admissibility of Multi-contract Claims

From a policy perspective key general reasons to allow multiple claims include
efficiency (including cost savings), speed and the desirability of avoiding
conflicting decisions or conflicting evidence and the avoidance of some of the
pitfalls flowing from the composition of multiple tribunals where overlap may
raise questions of prejudice or undue influence. Arguments against allowing
multiple claims include the possibility that there was lack of real consent and
the consequent negative implications for enforceability and the encouragement
of spurious reverse claims to add to the costs of the initial hearing with a view to
promoting more favourable settlement. In addition it cannot be presumed in all
cases that involve multiple claims that consolidation will indeed be speedy and
more efficient.

Scholars and practitioners have tended to either caution against multiple claims
or advocate broad inclusion. Rather than contending for one school of thought
over the other in terms of expansive versus restrictive admissibility, the aim
instead is to look at the kinds of factors and methods that should guide the
analysis on a case-by-case basis. As argued throughout, procedural challenges in
dispute resolution are inevitably about balancing certainty against flexibility
and fairness against efficiency. We aspire to all four values but they will
inevitably conflict. Hence trade-offs need to be made, ideally on some coherent
and logical basis. In terms of a logical approach, the differing methodologies that
have been applied are a conflicts approach, an efficiency-based approach or an
approach based on a broad analysis of actual and implied consent. The working
hypothesis is to consider whether consistency would best be promoted by first
analysing the issue as a question of consent. In this way, flexibility is allowed for
by eschewing page "546" any strong evidentiary presumptions one way or
another. A tribunal would instead look at all factors in any individual case to see
how confident it can truly be as to the express or implied consent to
admissibility. Consideration of questions of fairness and efficiency and conflicts
analysis should be seen as merely means by which consent can be implied,
rather than alternative paradigms.

7.8.3. Interpretation of the Arbitration Agreement

In any consent paradigm, the starting position should be the words of the
arbitration agreement itself. At times the words of the arbitration agreement
should be supplanted with the lex arbitri and arbitral rules derived through that
agreement. If the parties are in agreement at the outset, they can articulate the
desired treatment of multiple claims in their arbitration agreement. For
example, the Model Arbitration Clause of the Netherlands Arbitration Institute
encompasses ‘all disputes arising in connection with the present contract and
further contracts resulting thereof.’ (179) Where there are multiple contracts with
differing dispute resolution clauses, another possible approach is to provide for
a clear hierarchy between them. (180) If the arbitration clause does not cover this
but the parties are in agreement at the time of the dispute, they can express
agreement by way of a distinct compromis which would itself be a revised
agreement to arbitrate.

If there is more than one contract and arbitral clause, the interplay between
each must also be considered. This draws attention to the express comments
made by the parties on the issue. Unfortunately in many instances, the drafting
is less than perfect. It is important to understand that determinations of intent
based solely on ambiguous drafting are dangerous. While this is an obvious
proposition, it is particularly important when considering multi-contract
situations and the interplay between differing permutations of dispute
settlement provisions. It also provides a caution against too ready a willingness
to come up with some theory that purports to cover all scenarios or too rigid a
set of propositions to that end.

7.8.4. Avenues of Management in Arbitral Laws and Rules

Unfortunately, if consent is the determining factor, insufficient guidance is given

by most procedural rules. As noted above, this is because most arbitral rules
simply refer to counterclaims and set-off in a procedural timing sense rather
than identifying the typology of cross-claims that can be brought.

The difficulty with any solution in institutional rules is that the solution must be
drafted before disputes have arisen. Thus it must be of a general nature and be
page "547" able to deal fairly and efficiently with all permutations of facts. Here
the difficulty is that the drafter must consider what trade-offs would be
appropriate between flexibility and certainty. If issues of consent are to be
resolved by express provisions either in statutes establishing the lex arbitri or
institutional or ad hoc rules, there are three broad possibilities. The rules could
be drafted on an opt-in basis. They could indicate that the tribunal may deal
with certain issues where the parties expressly agree. Such a provision adds
nothing in terms of initial consent and would prevent inclusion if the parties
could not agree once a dispute arose. A second approach would be to express
the view that claims from multiple contracts plus multi-contract set-off and
counterclaim rights would never be mandatory save where they are truly
dealing with inherent defences central to the initial claim, in which case they fall
within the arbitration agreement which underpins the initial claim. Hence, they
would in all other cases be subject to distinct consent of the parties on a case-by-
case basis.

An alternative approach is an opt-out provision which allows the tribunal to

consider multiple claims and set-off and counterclaim rights except where the
parties agree otherwise. Even if an opt-out approach is adopted, there is a need
to consider just how to define claims, counterclaims and set-off that are
presumptively included. Should it be any form of counterclaim or set-off
recognised by the applicable law or should it expressly be limited to claims that
could be brought within the original arbitration agreement? Even in the latter
event, should the tribunal be given a discretion not to include the reverse claim
where the circumstances of the case suggest that the benefits of separate
proceedings outweigh the savings of a consolidated hearing? An example would
be where the tribunal as initially constituted may not have appropriate
expertise to deal with the multiplicity of claims. Perhaps counterclaims or set-off
rights should be notified prior to the composition of the tribunal so the claimant
can turn its mind to similar considerations to the respondent in selecting the
tribunal. Actual pleadings can be left to differing time limits. Respondents who
refrain from giving such notification to try and gain tactical advantages would in
most cases be found out in terms of when the likely counterclaim and set-off
came to their attention.

At the other extreme, the rules might allow for the broadest category of reverse
claims. An example is Article 21.5 of the Swiss Rules of International Arbitration
2012 which states:

The arbitral tribunal shall have jurisdiction to hear a set-off defence where the
relationship out of which this defence is said to arise is not within the scope of
the arbitration clause or is the object of another arbitration agreement or forum
selection clause.

Pavić suggests that the drafters had procedural economy as their prime
consideration. (181) Wolfgang Peter suggests instead that the justification for
page "548" Article 21(5) is the right to defence. (182) Rules as to counterclaim and
set-off rights were discussed in section 4.4.

7.8.5. Group of Contracts and Presumptions of Intent

A question which has arisen is whether a claimant is confined to putting forward

claims with respect to the contract that is referred to in the arbitration
agreement or can also put forward claims with respect to other contracts. This
matter has been analysed, primarily from the perspective of French law by
Philippe Leboulanger. (183) This matter is also addressed by Bernard Hanotiau.
(184) It is important to again understand that if the parties are clear in their

intentions, this will be determinative. Hence, scholarly commentary is more

about reasonable presumptions or approaches in ambiguous circumstances.
According to Leboulanger, the classic theory of contract holds that each
individual agreement within a group of contracts is a completely independent
agreement. But he goes on to say that this traditional notion does not
correspond to current contractual practice. He says that whenever there is an
economic link between contracts, ensuing from the contracts' nature and mutual
function, these agreements should not be regarded as autonomous agreements
but should be analysed together with all the other related contracts. As an
example he refers to the ICSID award delivered in Klockner v. Cameroon. (185)
There the tribunal adopted a ‘commercial reality’ analysis and, applying the law
of the Republic of Cameroon, considered that the reciprocal obligations
constituted a single legal relationship despite the existence of separate and
successive instruments governing the rights and obligations of the parties. (186)

Leboulanger proceeds to provide guidelines for determining whether multi-

contract situations should be treated as a whole. The first criterion is to see
whether the agreements make up one single business transaction in the sense
that the obligations are undertaken for the accomplishment of a single goal and
are economically inter-dependent. Leboulanger considers that this may arise
where there is an ‘economic and operational unit “hidden” behind a multi-
contract façade …’, page "549" and where ‘the obligations undertaken under the
different agreements are reciprocal, having a common origin, identical sources
and an operational unit.’ (187) Relevant surrounding circumstances might include
whether the various agreements were concluded on the same day, have the
same purpose and duration, whether they are drafted as master and subsidiary
agreements, and contractual evidence may involve comments in preambles or
definitional sections and other cross-referencing. (188) A ‘same relationship’ or
‘same economic transaction’ test obviously allows for at least some multi-
contract situations but is also likely to lead to differing responses from
Tribunals, with some looking to an expansive interpretation, while others might
take a more circumspect approach to controversial fact situations. Even with a
broader formulation, admissibility under multi-contract situations would still
need to be linked back to an agreement to arbitrate found within one contract
that, because of the integrated nature of the various contracts, is held to be
broad enough to encompass claims under distinct contracts.

A second criterion is the wording of the contracts concerned. Leboulanger also

says that agreements may be considered to be inter-related when they were
concluded on the same date, for the same duration and for the same purpose.
Another indication of inter-relationship is the presence of a master agreement
outlining the obligations undertaken by the parties which are more particularly
described in ancillary agreements. Sometimes the recitals to an agreement will
refer to other agreements and thereby establish their inter-dependence.
Hanotiau examines a number of French cases where the courts have uniformly
considered that if two agreements between the same parties are closely
connected and one finds its origin in the other or is the compliment or
implementation of the other, the absence of an arbitration clause in one of the
contracts does not prevent disputes arising from the two agreements being
submitted to an arbitral tribunal and being decided together. (189)

Essentially, these authors and cases are addressing the evidentiary factors in
light of which a tribunal might accept that there was consent to multiple claims.
While it will often be easy to say that the various contracts were part of one
underlying economic transaction, it does not follow as a matter of logical
necessity that the parties therefore would have intended the same dispute
resolution methodology. For example, some construction matters are best left to
independent arbitrators while others may best be resolved by owner-appointed
engineers or dispute resolution boards. In some cases the parties might believe
that only matters of significance should go to arbitration, with alternative
dispute resolution being the preferred means in other circumstances. Even
where all disputes ought to go to page "550" international arbitration, the
parties might want very different arbitrators depending on which contract is
involved. Guarantee and finance disputes may call for expertise in that regard,
while the underlying construction contract might call for engineering expertise.
It may be false efficiency to have less experienced arbitrators dealing with
matters that could readily be resolved by those with greater familiarity in each
area. (190) If the essential question is that of intent, a tribunal will consider why
there were separate contracts and why these do not have a common and
consolidating dispute resolution clause. This can be impacted upon by evidence
of intent where there is a great difference between those who would consider
the negotiating circumstances and those who feel bound by applicable law or
principles of certainty to concentrate on the drafting itself. This may be
impacted upon by the law applicable to each of the contracts, including the law
applicable to the arbitration agreement under notions of separability. In this
regard it is necessary to distinguish a number of situations. There are a number
of permutations that arise in multi-contract situations which are discussed
separately below. This is not to refute the traditional presumption, but instead
point to the potential for a more nuanced analysis on a case-by-case basis.

7.8.6. Admissibility under Related Contracts

This section separates out the various permutations; namely where one contract
with an arbitration clause is argued to be closely related to other contracts with
no dispute resolution clauses; secondly where different contracts have identical
arbitration clauses; thirdly where different contracts have differing arbitration
clauses; and finally where different contracts have arbitration clauses in some
cases and choice of forum clauses in others. The analysis is divided in this way
so that the implications as to consent in each of these permutations can be
considered and then see how that would impact upon the treatment of discrete
claims and counterclaims. (191)

In addition to considering whether all claims should be allowed, tribunals must

also consider how to conduct proceedings even if some claims are rejected. They
must still consider the appropriate elements of due process within each arbitral
process, at least with an eye to what is happening with the other. In either
circumstance tribunals also have to consider the potential impact on
enforceability of their decisions as to admissibility. (192) Parallel proceedings are
considered in section 7.13.

page "551"

7.8.6.1. Claims under Closely Related Contracts without Their Own Dispute
Settlement Clauses

The first case is where one contract contains an arbitration agreement and the
second contract does not contain any dispute resolution clause. In this situation
in some cases at least, the claimant may put forward claims founded on both
contracts in the one arbitration. The lack of an arbitration agreement in the
other contract is presumably seen as more of an oversight or explained on the
basis that repetition was unnecessary given the intended closeness of the
contracts, rather than evidence that the parties prefer litigation over arbitration
for disputes arising under it. Where there is one overriding agreement, (a
framework agreement or heads of agreement), which contains an arbitration
clause and where there is no arbitration clause in related contracts emanating
from the first, most would agree that the most likely intent was to cover all
disputes under the one arbitration agreement. (193) Poudret and Besson note the
developments in French law allowing extension of an arbitration agreement to a
dispute arising from a group of contracts if there are sufficient economic links
between the various agreements and also if the aspects of the dispute are
‘inseparable’, although the authors question whether the courts which
articulated this standard were truly faced with facts that would ground such a
test. (194) Where attention is given to the closeness of the relationship, this draws
attention to principles such as ensemble économique and ensemble légale. Here
there are again a number of permutations depending upon whether the second
contract has its own dispute settlement clause or not and if so, whether it is
arbitral or court based. This section presumes that there are no such clauses in
the other contracts.

Another complex situation is where amendments are made from time to time to
extend contracts. Are these variations of the original agreement, perhaps
undermining an original arbitration clause, or are they separate promises not
subject to an arbitration agreement, or are they merely contemplated steps to be
taken in performance of the original contract and hence subject to its dispute
settlement provisions? This should again be a question of determining a priori
intent after a consideration of all relevant circumstances.

7.8.6.2. Admissibility of Claims Subject to Contracts Each with Their Own

Identical Dispute Settlement Clause

Many commentators work from the presumption that if the same parties have
two contracts with arbitration clauses in identical terms, they can be presumed
from the page "552" outset to have wanted a global settlement of mutual claims.
However, Fouchard, Gaillard and Goldman correctly note that ‘the answer
depends on the interpretation of the parties' intention at the outset.’
Nevertheless, they suggest that it is ‘generally legitimate to presume’ that the
identical clauses signify an intent to submit the entire operation to a single
tribunal. (195) Even here such a presumption might readily be rebutted. Hanotiau
contrasts ICC Award No 5989 (196) where the parties signed two related
contracts on the same day and the case of Abu Dhabi Gas Liquefaction Co Ltd v.
Eastern Bechtel Corporation. (197) Leboulanger says that it is reasonable to infer
that the parties intention was to consider the two agreements as one unified and
indivisible transaction ‘and this is the reason why the arbitration clause was
repeated, in identical terms, in each one of the agreements’. (198) The practice of
the ICC International Court of Arbitration under the pre-2012 Rules was
explained by Anne Marie Whitesell and Eduardo Silva-Romero. (199) Whitesell
and Silva-Romero observe that for the ICC Court to decide that a single
arbitration shall proceed on the basis of multiple contracts, three criteria must
be fulfilled. The first is that all contracts must have been signed by the same
parties. The second is that all contracts must relate to the same economic
transaction. Thirdly the dispute resolution clauses contained in the contracts
must be compatible. ICC Rules 2012 Article 9 now indicates that subject to the
various challenge rights, claims arising out of or in connection with more than
one contract may be made in a single arbitration irrespective of whether they
are made under one or more than one arbitration agreement under the rules.

One reason why the parties might nevertheless wish to have different tribunals
under identical arbitration clauses relates to composition, a matter addressed
above and which is simply a countervailing factor to a blanket efficiency
presumption. For example, if the claimant was unaware of the potential
respondent's claim at the time of constituting the first tribunal and believed that
it would have picked a different expert if that claim was known, from its
perspective at least, there is no necessary intent to have the same tribunal deal
with both. At most it is a question of the trade-off between efficiency and
duplication on the one hand, against optimal tribunal composition on the other.
If composition is a problem, the best solution might not be rejection of
admissibility but instead, requirement of early notification of reverse claims to
allow this to be taken into account at the time of tribunal selection as is the case
with a number of institutional rules.

page "553"

7.8.6.3. Admissibility of Claims Subject to Contracts Each with Similar But

Not Identical Dispute Settlement Clauses

This category deals with cases where the arbitration clauses are identical in
most respects, but have some key differences. Examples might be differing Seats
for each and/or differing number of arbitrators. In such circumstances Hanotiau
suggests that separate proceedings must be initiated. ‘Mere concern for the good
administration of justice cannot prevail over the intent of the parties.’ (200) Even
here is it logical to presume conclusively that they would not have wanted
consolidation in the event that claims and reverse claims were both brought? A
more expansive approach might be based on a view that where there are
differing dispute resolution clauses, it is at least arguable that the intent of each
was simply to explain what to do with single claims but not multiple claims.
Under such an alternative approach the aim might be to interpret the second
dispute settlement clause to see if it was showing exclusive intent about disputes
regardless of whether they arose by way of claim or counterclaim, or instead,
whether the only intent was in relation to primary claims. The latter argument
suggests that where parties say that certain claims will be brought in one forum,
they are only speaking of the obligations of the claimant in commencing an
action. Such clauses, the argument proceeds, say nothing about when and why
that same issue could instead be brought as a counterclaim in a matter already
brought elsewhere. Even if this view is appealing, it does not presume that there
is automatic jurisdiction to hear the counterclaim under the first clause. All it
says is there is no presumptive evidence of a lack of intent to allow this to occur.

On this logic, parties' identification of different Seats may have been relevant on
the presumption that there was only one claim, but the clause might still be
capable of being interpreted to the effect that they have not given any indication
of the preferred Seat if there were multiple claims. For example, in a
construction contract with a side loan agreement, the parties might have
selected a neutral and conveniently located seat for loan disputes but a different
seat under the construction contract, being where the building work is taking
place. This might have been simply to make it cheaper for the arbitrators to take
a view of the physical building where appropriate or because that is the Seat
that is demanded by the host State of the building works. Even with such
provisions, they may still have preferred from the outset that a claim under the
loan would simply piggyback on the construction Seat in the event of concurrent
disputes. The suggestion is not that arbitrators should always accept this as
valid, but simply that irrebuttable presumptions to the contrary from
inadequately drafted clauses make little sense within a consent paradigm.

Similarly if the two contracts call for differing numbers of arbitrators, perhaps
because disputes under one were presumed to be likely to be dealing with
bigger amounts than under the other, it might still be cheaper to consolidate the
smaller page "554" claim in front of the panel of three rather than force a three-
person hearing plus a separate single-person hearing. This will not always be
the case but to again presume a lack of consent as a matter of course flowing
from a separate arbitration agreement would not be a presumption that one
could confidently predict to be commercially sound in all circumstances. If
reverse claims would clearly save time and money and if the party arguing
against consolidation cannot articulate any fairness or efficiency factors in its
favour, that may be telling.

It is also possible to envisage cases where parties would not have intended
differing arbitration clauses to automatically block reverse claims on essentially
related matters. For example, differing clauses cannot wholly overcome the
policy arguments in relation to true defences. Arguably the second clause is only
a promise about what to do with respect to primary claims and not a waiver of a
right to raise true defences as and when needed. As Fouchard, Gaillard and
Goldman note, where cross-claims are not allowed, fairness and efficiency
arguments would also be complicated if a party in one arbitration claims that it
refused to perform its obligations because of a breach by the other party in the
matter being considered under a second arbitration. The same logic may apply
with ensemble contracts where claims under one contract still fit within the
wording of different arbitration agreements in other contracts. (201) As always,
the parties could resolve these ambiguities by carefully delineating in their
arbitration agreement which counterclaims, if any, are permitted.

7.8.6.4. Reverse Claims with Contracts Combining Arbitration Clauses and

Jurisdiction Clauses

While most authors treat the situation of differing arbitration clauses and
jurisdiction clauses together, they need to be considered separately under an
intent paradigm. It was suggested above that any presumptions flowing from
differences in arbitration clauses should be rebuttable at most. The situation is
different where arbitration and forum clauses are brought together. A separate
contract with a jurisdiction clause indicates an intent to litigate and not
arbitrate such disputes. The parties may simply be saying that for that type of
dispute they want a page "555" completely different type of adjudicator, with a
differing conflicts methodology and a different procedural model. (202)

7.8.7. Conclusions

The following broad principles are suggested. These incorporate comments

previously made as to counterclaim and set-off rights in section 4.4 and some
comments made in relation to counterclaims in this chapter.

1. The first question is to consider a priori consent to multiple claims.

2. The starting point in discerning consent to multiple claims is always the
arbitration agreement before the initial tribunal, together with the lex
arbitri and the procedural rules if any, agreed to by the parties. The
Tribunal must analyse the agreement and applicable law and rules to
discern what the parties truly intended at the outset, in the circumstances
that have in due course arisen. It is a question of construction and not a
discretionary matter, unless an express discretionary power has been
granted via the rules. If extra claims by the claimant are permitted because
these come within the arbitration clause, these should be allowed unless
there are significant concerns about the prior constitution of the tribunal
being inappropriate in the circumstances. That is unlikely to be the case. If
the claimant was unaware of the additional claim at the time of the initial
appointment, then there is no difference in its approach to that choice and
the respondent's. If the claimant was aware of the future potential claim, it
is first selecting an arbitrator it believes to be suitable. Here however the
respondent did not have the same opportunity, nor did the two arbitrators
in selecting the Chair.
3. Assuming there is only one contract involved, it is obviously the case that if
it expressly provides for or denies the opportunity for additional claims,
this is the end of the matter.
4. If it is an attempt by Claimant to add a new claim after the arbitration has
commenced, the arbitration agreement is still the gateway. Arbitral rules
deal with timing issues and provide a discretion as to late acceptance of
new claims, for example, where Terms of Reference are required, but this
would still be provided the claims otherwise fall within the subject matter
of the agreement to arbitrate.
5. However, even if a late claim is within the agreement, the lex arbitri or
procedural rules agreed to may place time limits on additional claims
and/or may add a discretion to deny otherwise admissible claims.
Difficulties may arise with late claims in ad hoc arbitrations without any
guidance in any rules. page "556"
6. Rules which merely address process issues should not be taken to be
providing clear jurisdictional tests. For example, rules which simply
indicate the time periods within which counterclaims and set-offs are to be
brought, notwithstanding that they refer to these concepts expressly, add
little if anything to the analysis as they do not purport to define the
parameters within which such claims are permissible.
7. A consent-based paradigm will vary if parties have selected the Seat and the
 rules, or if instead an institution or other appointing authority does so. In
the latter event, presumptions of intent one way or another based on
statements within the lex arbitri and the rules would be more problematic.
8. If the arbitration clause is unclear, it should be interpreted in good faith but
without presumptions either way. The tribunal should consider all relevant
factors, including good faith a priori efficient intent, but not efficiency per se
from the tribunal's post-dispute perspective, even if the two would usually
be identical. Presumptions such as effete utile do not usually apply naturally
to the question of multiple claims. Their natural utility is in relation to
express indications of arbitral consent that are nevertheless flawed and
need some recrafting. If the parties truly wanted some form of arbitration
but they express themselves in a way which might render that choice
invalid, they are presumed to have preferred validity over invalidity. Hence,
they are presumed to have wanted an arbitrator to try and find a better
meaning of their express but flawed choice. Conversely, where multiple
claims are concerned, even if there is some ambiguous reference in the
arbitration agreement, it is often not a question of validity but simply a
question of scope. The parties definitely wanted arbitration of primary
claims. The question is whether they also wanted arbitration of reverse and
additional claims. At the very least, a presumption of efficiency is a more
honest articulation of likely intent than a presumption of validity.
9. Additional claims need to be sufficiently linked to the primary arbitration
clause under which they are sought to be introduced. Merely calling some
or all of them ‘defences’, particularly as that term is viewed in domestic
legal systems, or defining them as substantive as opposed to procedural
rights, should not replace a careful analysis of the sufficiency of the linkage
to the consent to arbitration. The challenge for those wishing to have a
more expansive ambit of inclusion is to consider why we would
countenance a link to a claim that is broader than any linkage that would
naturally fit within the arbitration clause itself.
10. Multiple contract situations raise a number of permutations that would
profit by being addressed separately, simply because the consent logic
differs, although all should be dealt with under the same methodology as
outlined above. The added complication is merely that a true contextual
approach to interpretation of one clause needs to consider what is being
said in any other contractual arrangement between the parties. This not
page "557" only involves all of the foregoing questions such as choice of
interpretation method and the evidentiary basis of identifying consent, but
also must take into account the particular articulation of rights and
obligations between the differing contracts and the timing of each. It is
typical in such interpretative conflicts that adjudicators might employ such
additional presumptions as, for example, that specific rules are normally
presumed to override general rules and later rules can be taken to override
earlier ones. Once again these should not be used as fixed presumptions but
merely aids to try and discern the true intent from contradictory
documentation.
11. Once again, clear drafting will resolve any problems in multi-contract
situations.
12. In the absence of clear drafting, the inevitable ambiguities are a dangerous
basis upon which to draw confident presumptions about the presence or
lack of consent; either the presence of consent because of identical
arbitration clauses or the lack of consent because of differences. As to the
first, some cases may raise legitimate procedural justice concerns as to
composition even where clauses are identical. Where there are different
clauses, the historical drafting of model clauses simply leads to confusion.
Such clauses may say nothing more than that isolated claims must go to
different places. They may give no clear indication of what was intended for
concurrent reverse claims.
13. In these circumstances, tribunals should analyse all of the factors, accepting
that the differences in clauses are at least relevant and might require extra
caution before allowing multiple claims that appear to go against a
direction in favour of separate proceedings.
14. Turning to respondent's claims, because a counterclaim remains alive even
if the primary claim is withdrawn or invalid, it must be based on its own
independent evidence of consent. As always, such consent should be found
to emanate from the arbitration agreement itself, either directly or through
a lex arbitri or rules that expressly allow for counterclaims.
15. The rules vary from those which give no guidance as to the type of
counterclaims that may be brought to those defining the linkage. Defined
linkages are either to the contract or the arbitration agreement or the same
relationship. Even then it is important to construe these rules alongside the
arbitration agreement as there may be variances between the two. For
example, if the arbitration agreement expressly allowed for a broader range
of counterclaims than the rules, most would consider that the express
reference was the better indication of the true intent of the parties. The
same would hold if the arbitration agreement was expressed to be
narrower.
16. Where set-off is concerned, section 4.4.3.1 argued against the view that
being defences, all set-offs must be admissible on the basis that justice
demands that any defence must be allowed. They are seen as defences in
page "558" domestic systems. These systems display very different
approaches to the treatment of set-off even if all describe them as defences.
Furthermore, domestic litigation treatment is built on many policy
considerations not relevant to arbitration.
17. For arbitration, it is suggested that a similar approach should be taken to
counterclaims. Is the set-off within the agreement and/or the applicable lex
arbitri and rules? If so it should be permitted. If not, for example an
independent set-off in an ad hoc arbitration, the mere description as a
defence should not suffice. All additional claims need to be sufficiently
linked to the primary arbitration clause under which they are sought to be
introduced. Merely calling some or all of them ‘defences’, particularly as
that term is viewed in domestic legal systems, or defining them as
substantive as opposed to procedural rights, should not replace a careful
analysis of the sufficiency of the linkage to the consent to arbitration.
18. If parties agree on rules which expressly address an issue, this is again clear
 evidence of consent. For example, adoption of the Swiss Rules 2012 and
Article 21.5 is clear consent to allow a broad range of set-offs.
19. In any event, admissibility of counterclaims under multi-contract situations
would still need to be linked back to an agreement to arbitrate found within
one contract that, because of the integrated nature of the various contracts,
is held to be broad enough to encompass claims under distinct contracts.
20. Additional counterclaims and set-offs after the initial stages again draw
attention to the rules selected and the potential for time limits and/or
express discretionary powers in an institution or Tribunal.
21. Principles of consolidation build upon similar issues and are discussed
more fully in section 7.11. At this stage some general observations are made
in the context of consolidation as a means to allow additional claims. In
most cases the decision to order consolidation is discretionary.
22. Where discretions are expressly provided for additional claims or
consolidations, many matters could be considered in the exercise of the
discretion including:
• How closely are the two disputes linked in terms of their facts? Obviously
if they are sufficiently linked, then the entitlement comes about directly
under the arbitration clause and not via discretion of a tribunal, but even
circumstances that do not fit directly into the agreement can have
various degrees of connection to the primary claim.
• In terms of efficiency, are there clear transaction cost savings to be made
by having one tribunal?
• Are there questions of evidence that would best be heard by a single
tribunal, either to prevent inconsistency or to promote confidentiality?
• Do the facts show that it would be both fair and efficient to try to find the
net payment obligations, if any, between the parties rather than to
separate these out through more than one tribunal hearing? page "559"
Alternatively, could cash flow issues simply be dealt with via awards that
are timed to allow netting out mutual payment obligations?
• Is the tribunal composition adequate to deal with each of the matters
both in terms of expertise and in terms of cost benefit as to number of
arbitrators?
23. Finally, a question arises as to how to resolve those multiple claims
disputes where an analysis of the agreement and rules leaves a Tribunal in
doubt. Some suggest a conflicts approach; some suggest that efficiency
considerations are the best evidence of the presumed intent of the parties.
As to the latter, a contentious question would be whether all or any of the
aforementioned factors relevant to express discretions can also be
considered in determining the likely intent of the parties in cases not clearly
resolved by the agreement and rules. In such circumstances the concern
should be to aid the interpretation of the arbitration agreement, where
necessary, with a careful assessment of all factors that might help a tribunal
draw conclusions as to the likely a priori intent of the parties. Implied
intent to promote efficient solutions is an important working hypothesis, as
long as it is seen as one factor that needs to be looked at alongside others.
24. As noted, arguments in favour of admissibility or consolidation include
general efficiency, reducing the transaction costs of parallel proceedings,
overall timeliness, and the avoidance of some of the pitfalls flowing from
the composition of multiple tribunals where overlap may raise questions of
prejudice or undue influence. Countervailing factors include the concern
that reverse claims might be brought on spurious grounds to actually delay
proceedings, frighten the claimant into settlement and add immediate
financial burdens through the arbitral advance on costs. In addition,
admissibility or consolidation may raise questions as to the suitability of
the tribunal to deal with all of the multiple claims and might provide
undesirable tactical advantages in tribunal selection. Presumptions based
on general efficiency alone are only a small part of a commercially realistic
analysis of likely intent in the hopefully small number of troublesome cases
where the agreement and rules are still ambiguous.

7.9. Joinder and Consolidation

The following sections separately examine issues of joinder and consolidation.

The latter was also considered in the previous section where discretionary
issues pertaining to multiple claims were considered. As with most of the topics
in this chapter, policy and process issues overlap considerably. As always, there
is a need to consider question of consent and applicable laws and rules. Where
discretions apply there is a need to consider efficiency and due process norms.
Joinder and consolidation can also be looked at alongside each other as they
may constitute page "560" differing approaches to achieving similar outcomes.
Nevertheless, because arbitral laws and rules tend to treat them distinctly, they
are analysed separately below.

Under some national laws, consolidation or joinder decisions must be made by

the courts and not tribunals. (203) Some rules direct that the institution is the
appropriate entity to make decisions as to joinder or consolidation. Others
devolve these decisions on the tribunal. (204) Even where an institution makes
the initial determination, this will not aim to preclude a tribunal which can make
a final determination of validity as is the case with any jurisdictional challenge.

7.9.1. Introduction to Joinder and Policy Considerations

Problems with joinder were suggested as a concern that users have with
international arbitration. (205) While some have used the term joinder generally
in multi-party situations, it is preferable to limit its use to cases where a third
party asks or is asked to join arbitral proceedings that have already commenced.
(206) Sometimes rules or commentators also speak of joinder of proceedings to

cover what would otherwise be described as consolidation. (207) Some

commentators distinguish between joinder, being a request by a party to include
a third person and ‘intervention’ being a request by the third person to join
existing proceedings. (208) Some would define the notion of intervention
differently in the context of some involvement that is less than being a full party.
(209) One scenario of intervention of this nature would be intervention as an

amicus curiae. It is important to distinguish between these cases. The chapter

separately considers joinder, consolidation and non-party intervention but
these options must often be considered alongside each other as in some cases, a
consolidation decision may be a preferable means to reach similar outcomes to
joinder requests. Consolidation and joinder can also both be looked at from the
perspective of interpretation of the arbitration agreement where they are all
sought to be included.

As to joinder, the most likely scenario is that a respondent wishes to bring a

derivative claim against a third party or wishes to counterclaim against such a
party alongside a counterclaim against claimant. Claimant itself could seek to
join an page "561" additional party when it concludes after commencement that
an additional respondent ought to be included. From a claimant's perspective,
joinder will typically be sought when a detailed defence to the statement of
claim or subsequent document production or witness statements show why the
claims ought to also be brought against a third party or in some cases should
only have been brought against that entity. This will commonly arise when there
are multiple related entities or it was not apparent which employee acted for
which entity or which entity has side agreements to honour obligations under
the main contract or which entity technically holds rights, licences and
intellectual property. (210) These situations could be further divided into those
where the claimant ought to have known of the structure at the outset, in which
case the later joinder application will meet less sympathy, and those where
there is no fault on claimant's part. In the extreme, there may even be cases
where the other party has created a complex corporate web for no other
purpose than to wreak havoc with legal proceedings. Cases can further be
divided into those where the joined party truly would have distinct contentions
on its behalf and others where the same essential facts and law will be
considered and their inclusion simply ensures a binding and consistent
determination against all relevant parts of a corporate group with res judicata
effect.

A number of questions arise when joinder is considered. The first is whether

joinder is arguably permissible. If so, the second question is what impact there is
on existing tribunal composition. The third is the impact on the existing
proceedings and whether any previous stages must be revisited, in part to afford
due process rights to the joined party. All joinder and consolidation scenarios
must find their basis in consent, either direct consent of the persons involved or
via agreement to arbitral laws or rules which provide a discretion to do so.
Enforceability concerns can arise in relation to queries as to consent, and proper
composition of the tribunal. From a policy perspective efficiency suggests that if
all of the claims and cross-claims are sufficiently connected, then one
proceeding is likely to be preferable, although it was argued above that it may be
a complex cost/benefit analysis where one party incurs costs in being party to
matters of less interest to them. A key issue is that of potentially conflicting
decisions if multiple proceedings ensue. There is not only the prospect of
inconsistent final awards but also inconsistent interim measures, with one
tribunal demanding that a party do something while another demands that it
refrains from doing so. (211) Efficiency considerations are also dependent on the
tests that are applied. The more that these are restrictive, limiting consolidation
or joinder to identical issues, the more there are natural efficiencies and less
likelihood that one of the signatories will be forced to sit through matters of no
interest to it. Fairness considerations can be dependent on the motivation of the
person seeking to effect or block joinder. Where there are legitimate grounds to
page "562" do so that itself seems a fairness consideration. Fairness may still be
a problem even where there are significant efficiency gains. While overall cost
savings are likely, these will not flow proportionally to all parties. Indeed some
may have their costs increase as a result of consolidation or joinder.

One particular problem relates to tribunal composition. There are significant

problems in attempting to join a third party after the tribunal has been
constituted. The key issue is whether the joined party will accept the existing
tribunal or whether there would be a need for the tribunal to be reconstituted.
Where the third party itself seeks to join, it may be accepting the existing
appointment at the same time as making the joinder application. Where a
reluctant third party is concerned, it would be more likely to raise issues of due
process in being denied its equal right to participate in tribunal selection. The
degree of the relationship between the new party and existing parties and the
commonality of claims and issues should also impact upon questions of tribunal
composition and whether a new party can truly raise concerns that it was not
privy to the establishment of the tribunal. The more the joined party can be seen
as part of a group with an existing party, the less justification for complaining
about prior constitution of the tribunal, although it is hard to make confident
conclusions in that regard at preliminary stages. It may be easier for an
institution to make a prima facie joinder determination if this will occur prior to
tribunal appointment and overcome one of the significant challenges in this
arena.

Where a claimant seeks to join a third related party after discovering that the
respondent has little in the way of assets and looks for a related entity with
deeper pockets, in most cases claimant needs to stretch liability arguments to
encompass that entity with the commensurate need to add new and significant
issues. In cases where the claimant is fault free, Voser suggests that a starting
presumption ought to be to allow joinder and reconstitute the tribunal or at
least that part that would have been subject to third-party rights if included
from the outset. Conversely, where the claimant simply failed to include an
appropriate person from the outset through it own oversight, joinder should
only be allowed with consent of all parties.

7.10. Legal Provisions as to Joinder

There are two separate scenarios – requests by a third party to be included as a

party, or requests by a party to join a third person. The policy and consent
issues are the same but at times laws and rules address each separately or only
refer to one scenario.

Lew, Mistelis and Kröll suggest that generally speaking, joinder is only possible
if all parties consent. (212) Such consent could be direct or via selection of arbitral
laws or rules that provide broad discretionary powers to institutions or
tribunals. Even then, the relevant test must look for a reason to join that raises
page "563" all of the questions of the theoretical basis for multi-party
arbitration discussed in previous parts of this chapter. It was strongly argued
that a priori implied good faith consent is the gateway. Where a third party
seeks to intervene, the relevant consent should be considered from the outset
and should not require a new express agreement by the existing parties, (213)
although the situation is easy where the parties all consent at the time of a
dispute. Here there is a separate submission agreement in any event.
Nevertheless, there are numerous reasons why parties may not consent at the
time of the dispute. From the claimant's perspective, if the original respondent
has the funds to honour an award, it gains no benefit from bringing in a third
party against whom respondent may seek recourse. That party may wish to
delay or avoid liability or may only have a concern with some of the claims.
Conversely, if it is claimant seeking to join, the respondent may have concerns in
bringing a subcontractor in if it had not wished to disclose its financial
relationship with the owner. (214)

If a relevant decision maker will consider whether there is an implied

agreement to joinder or consolidation, this will raise all of the questions of the
means of discerning such intent. In the extreme, this chapter has noted in a
range of scenarios that some tribunals might conclude that parties can be
presumed to have impliedly agreed where efficiency concerns would make this
a desirable outcome. The conclusion is more likely where all persons are
signatories to the one agreement. (215) Even where an implied term is possible, it
does not necessarily suggest that consolidation or joinder should always occur.
The a priori implied intent would only be where circumstances suggest that it is
reasonable for this to occur. Analysis of consent is also impacted upon by
whether the joinder provisions are found in the lex arbitri or in ad hoc or
arbitral rules. If they are in the lex arbitri and the parties have selected the Seat,
then similar consent issues apply to where they have selected the rules. If
instead, the tribunal or an institution selected the Seat, it is more problematic to
attempt an original consent logic based on broad discretions in those laws.
Another question is what law should apply. Considering it as purely a
procedural matter would look to the lex arbitri or agreed rules. Considering it
from the perspective of consent would potentially look to the law applicable to
the arbitration agreement. There will be no difference, of course, if the law of the
arbitral Seat was selected for that purpose. (216) Another question is page "564"
whether the provision simply leads to inclusion in the same proceedings of
claims and counterclaims between applicant and third parties or whether it also
involves claims for and against the non-consenting party. (217) In addition to
express rules covering questions of joinder and consolidation, other rules may
indirectly limit the power to achieve such outcomes. An example would be a
strict requirement to identify all respondents in a request for arbitration. (218)
This could also be impacted upon by definitions of who is a party. There is also a
question as to whether form requirements need to be satisfied to allow for
joinder or consolidation. Born suggests that this would appear to apply but
notes decisions to the contrary. (219)

The New York Convention has no express provisions dealing with joinder. The
UNCITRAL Model Law also does not expressly deal with joinder. The UNCITRAL
Model Law drafters considered but rejected proposals to cover consolidation
and joinder both in the 1985 version and the 2006 revisions. (220) While it has no
specific provisions in that regard, to the extent that consolidation or joinder
flows from the parties' agreement, it comes within power in any event. Some
arbitral laws allow a request by a third party to intervene where there is an
arbitration agreement between that party and the disputants. (221) Article 35 of
the English Arbitration Act 1996 simply allows for consolidation by agreement
of all parties. Where institutions are involved, once a tribunal is appointed, the
first question is whether any subsequent joinder application must involve
institutional consideration or is left for the tribunal itself. This will obviously be
dependent on the rules, if any, in relation to these matters. Voser argues in
relation to the previous version of Article 6(2) of the ICC Rules that the gateway
analysis should still occur as the mere constitution of the tribunal should not
take away the third party's right to have the first hurdle of court scrutiny. (222)
That presumes that the intent of the rules and in turn the intent of the parties
when the Rules were chosen, was indeed to have such continuing scrutiny even
when the tribunal is present. A further question is whether some express power
given to the institution interferes with the tribunal's right to subsequently
consider the same matter. That should not be so as institutions typically take
preliminary views on such matters, allowing for the tribunal to reconsider after
hearing directly from the parties and perhaps after hearing witnesses and
perusing a broader range of documents.

page "565"

Previously, the ICC Rules did not expressly deal with respondents' cross-claims
or counterclaims against third parties. ICC practice moved from a strict position
only allowing this where all parties agreed to a modified position allowing third-
party joinder at respondent's request if:

1. the third party has signed the arbitration agreement;

2. the respondent introduces a counterclaim or cross-claim;
3. the request for joinder must have been made before the arbitrators were
appointed or confirmed. (223)

It is interesting that the shift in the ICC approach occurred without any changes
in the rules. The new ICC Rules are more expansive in this regard, although care
should be taken as there will always be situations where a party's arbitration
agreement is interpreted to call for arbitration pursuant to older versions of
institutional rules. The ICC Rules 2012 now contains a specific provision on
joinder of additional parties in Article 7. A request is made to the Secretariat, the
date of which is deemed the date of commencement of the arbitration against
the additional party. The request is subject to the provisions of Article 6(3) to
(7) which indicate that jurisdictional questions are considered by the tribunal
unless the Secretary-General refers the matter to the court for a decision
pursuant to Article 6(4). The request for joinder is to follow the requirements
that pertain to an original notice and the additional party is to submit an answer
in similar form. The additional party may also make claims against any other
party as per Article 8. Article 7(1) indicates that no additional party may be
joined after the confirmation or appointment of any arbitrator unless all parties
including the additional party otherwise agree. The Secretariat may also fix a
time limit for the submission of the request for joinder. The Article does not
indicate the criteria by which joinder decisions are to be made. ICC Rules 2012
Article 8 indicates that where there are arbitrations with multiple parties,
claims may be made by any party against any other party.

Article 4.2 of the Swiss Rules 2012 requires the tribunal to decide on joinder
requests ‘after consulting with all of the parties, including the person or persons
to be joined, taking into account all relevant circumstances.’ Article 4.2 allows
for the request to be made independently by a third party. The Rules do not
require that the third party be a signatory to the agreement or that there be
express consent by the third party where it is sought to be joined. The Swiss
Rules do not expressly require consent from existing parties. The tribunal
ultimately exercises a discretion based on ‘all circumstances it deems relevant
and applicable.’ This would include the nature of the relationship and the timing
vis-à-vis the existing proceedings. It has been suggested that the revised Hong
Kong International Arbitration Centre Administered Arbitration Rules effective
from 1 September 2008 were largely page "566" modelled on the Swiss Rules
2006 but chose not to incorporate the equivalent of Article 4.2 of the Swiss
Rules. (224)

Article 22.1(h) LCIA Arbitration Rules 1998 speaks of a ‘third person to be

joined in the arbitration of the parties.’ Article 22.1(h) deals with a request or
application by an existing party after all existing parties are given a reasonable
opportunity to state their views. Article 22.1 allows a tribunal upon the
application of a party to join a third party provided that the applicant and third
party have consented in writing. The provisions do not call for consent by the
other party to the proceedings. The power remains unless the original parties
agree otherwise in writing. While the provision only covers applications by a
party and not applications by a third party, given the requirement for consent
between them, little turns on this restrictive language as where the agreement is
there, there is no need for the separate application by the third party. Once
again there may be a debate as to whether that party's acceptance of the LCIA
Rules was consent to arbitral jurisdiction in that regard. Certainly if the rules tell
the original parties that they could remove the power and they did not do so,
there is consent to the discretionary power. Nevertheless, that consent is not
blanket consent to a particular outcome, but instead, to a reasonable
methodology. A relevant factor before the tribunal's considerations ought to be
whether that other party does in fact wish to have the third party joined or not.

Article 24(b) of the SIAC Rules entitles the tribunal ‘to allow other parties to be
joined in the arbitration with their express consent …’. It is not clear whether
the term ‘their’ references consent by all parties or only by the prospective
joined party. (225) Article 41 of the NAI Arbitration Rules allows a third party who
has an interest in the outcome of arbitral proceedings to request the tribunal to
join the proceeding or to intervene therein. All parties must have an opportunity
to be heard and the third party must accede in writing. The Netherlands
provision is limited to situations where there are two or more arbitral tribunals
in the Netherlands. Other laws following the Netherlands model include New
Zealand and Hong Kong. (226) Born suggests that the Netherlands provision was
introduced as a result of proposals from the Netherlands construction industry.
(227)

In ad hoc arbitration the tribunal will typically consider joinder within its broad
discretionary rights subject to due process mandatory norms. In some instances
of ad hoc arbitration, provisions such as Article 20 of the previous UNCITRAL
Rules allowing for amendment of claims and defences have been page "567"
used to allow for third-party joinder. (228) UNCITRAL gave considerable attention
to this issue in its 2010 Rules revision process. Draft UNCITRAL Rules Article
7bis (2) required joint appointments by multiple claimants or respondents.
Where a joint appointment did not occur, draft Article 7bis (3) gave the
appointing authority at the request of a party the discretion to make a default
appointment of the arbitrator who was to be jointly appointed or alternatively,
revoke prior appointments and appoint all three arbitrators afresh. Divergent
views were presented at the Working Group's 46th Session in February 2007.
(229) The Secretariat had sought advice from arbitral institutions on their

experiences with joinder. (230) The Working Group did not consider that the non-
applicant party needed to consent. (231) However, the Working Group seemed to
contemplate that there will be claims and counterclaims between the non-
consenting party and the third party. (232) Roos refers to the draft revised
UNCITRAL Article 15(4) from the 49th Session using differing language
requiring that the third persons are a party to the arbitration agreement and
have consented to be joined. (233) Draft Article 15(4) states first that a tribunal
may on the application allow third persons to be joined and only then states
‘and, provided …’ there is consent, it may then make an award in respect of all
parties. On plain meaning this departs from the more restricted model of Article
22.1(h) of the LCIA Rules separating the power to join and limiting the need for
consent to scenarios where the award intends to be in respect of all parties
involved. (234) Article 17(5) of the UNCITRAL Rules 2010 now provides that the
tribunal may at the request of any party allow one or more third persons to be
joined as a party provided such person is a party to the arbitration agreement.
All parties including the person or persons to be joined must be given the
opportunity to be heard. The tribunal may reject the application because of
prejudice to any of those parties.

Another question where terms of reference are involved or documents to

similar effect are used, is whether a subsequent request to join a third party
should page "568" be treated as raising a new claim requiring the tribunal's
concurrence. (235) A contrary argument would be that the new claims provisions
are discussing claims that are new between the parties as articulated in the
terms of reference. It would certainly be true that if arguments as between the
third party and either existing party are matters that both originating parties
would have to consider, then they are certainly new claims from their
perspective. In any event, even if such provisions applied, the same principles
ought to be taken into account as under a joint consideration of the joinder
application. (236) Where terms of reference have previously been completed, and
a new party is joined, there will need to be new terms of reference covering its
claims and defences as well. If the rules do not expressly cover this, the existing
provisions covering terms should be applied mutatis mutandis. It would be even
simpler to allow for an addendum to the original terms, (237) although care will
need to be taken that the addendum does not contradict elements of the original
terms.

7.11. Consolidation (238)

7.11.1. Introduction and Policy Considerations

Sometimes, after an arbitration has been commenced, a party to that arbitration

will seek to start a second arbitration concerning the same or a related contract
or legal relationship. It may be the Respondent in the first arbitration who seeks
to put forward a claim in a second arbitration rather than by way of a
counterclaim in the first arbitration. Alternatively the Claimant in the first
arbitration may, for various reasons, put forward additional claims in a second
arbitration. A question which then arises is whether the two sets of proceedings
are desirable or whether they can and should be consolidated. Consolidation
involves bringing together two or more arbitrations into the one proceeding.
Consolidation is generally considered in the context of multiple contracts but
can also arise where separate proceedings are instituted in relation to the one
agreement. Consolidation may cover circumstances where separate proceedings
have the same or different arbitrators. In some cases only one tribunal has been
constituted. Consolidation generally arises where the same parties are engaged
in more than one arbitration, although it can also be considered in multi-party
scenarios.

page "569"

7.11.2. Legal Provisions as to Consolidation

Consolidation can always occur if the parties agree to do so. But this is unlikely
because one of the parties has taken a deliberate decision to commence a second
arbitration. The legal position can be analysed in two distinct ways, the first is to
look at the arbitral laws and rules applicable to see whether there is an express
provision allowing for consolidation. If so, attention will obviously be given to
the preconditions that are articulated. Alternatively, a tribunal can approach the
question as a matter of interpretation of the arbitration agreement before it to
decide whether it is broad enough to cover the other matters. The two are not
mutually exclusive and as invariably arises also bring into play the complex
resolution of consent versus jurisdictional paradigms. If a tribunal approaches
the question from an interpretational perspective, attention may also need to be
given to the circumstances behind the second proceedings as the party
contending for a broader ambit of the first clause might be argued to have
waived the right to so argue based on its actions in bringing or defending the
second proceedings.
Some might see interpretation as a fallback only where the laws and rules do not
expressly deal with consolidation. The contrary view would be that the laws and
rules must always be considered in the context of the parties' intentions in their
arbitration agreement and the proper competence of a tribunal in relation to
each and every arbitration clause before it. On that view, in some circumstances
at least, a tribunal might interpret the clause broadly enough to cover other
contracts in circumstances where the preconditions in arbitral rules have not
been satisfied. That makes logical sense in that broadly drafted arbitration
clauses allow for multiple claims and issues to be decided if they all ‘relate to’ or
‘arise out of the essential commercial transaction. However, regulatory
preconditions to consolidation are more conservative and look to sufficient
identity between the two proceedings to justify consolidation. Even then there
are questions of tribunal appointment where that has already occurred in at
least one proceeding and the relevance of evidentiary stages where one is partly
heard.

The New York Convention has no express provisions dealing with consolidation.
The UNCITRAL Model Law also does not expressly deal with consolidation. The
UNCITRAL Model Law drafters considered but rejected proposals to cover
consolidation and joinder both in the 1985 version and the 2006 revisions. (239)
As noted, while it has no specific provisions in that regard, to the extent that
consolidation or joinder flows from the parties' agreement, it comes within the
principles in any event.

Some legal systems leave it to the courts to determine whether consolidation

should be allowed. (240) Where institutions are involved, some afford the right to
page "570" consolidate to the institution or an institutional court where that is
involved. (241) Absent specific agreement of the parties it would not seem
possible to consolidate arbitrations in different arbitral Seats. (242) It has been
suggested that compulsory consolidation negates party autonomy and may
jeopardise enforceability. (243) Born also suggests that non-consensual
consolidation or joinder is contrary to the New York Convention. (244) That may
depend on the circumstances. If the parties have selected an arbitral law or rules
that allow for an application on that basis, then they have consented to the
mechanism even if not to the conclusion as to the particular application made.

Under Article 1046 of the Netherlands Arbitration Act 1986, the President of the
District Court of Amsterdam may, upon request of a party, order consolidation
unless the parties otherwise agree. The NAI Rules have not included a similar
provision. Article 30 of the Belgian Judicial Code adds further criteria in inviting
consideration as to the degree of connection and whether the claims ‘are so
closely related that it is desirable to consolidate them and judge them together,
in order to avoid an outcome that would be incompatible, if said disputes would
have been handled separately.’ Consolidation is allowed for under Article 1126
of the NAFTA Rules and Article 33 of the 2004 US Model Bilateral Investment
Treaty. Conversely, the Departmental Advisory Committee on Arbitration Law
whose report inspired the change to English legislation, considered that it was
inappropriate to allow a tribunal to order consolidation regardless of party
agreement, seeing that power as a negation of party autonomy. Section 35 of the
Arbitration Act 1996 (UK) now provides:

35. (1) The parties are free to agree –

(a) that the arbitral proceedings shall be consolidated with other
arbitral proceedings, or
(b) that concurrent hearings shall be held,

on such terms as may be agreed.

page "571"
(2) Unless the parties agree to confer such power on the tribunal, the
tribunal has no power to order consolidation of proceedings or
concurrent hearings.

The Australian legislation is seen as being one of the most far-reaching.

Australia's International Arbitration Act allows a party to make an application
for consolidation where ‘a common question of law or fact arises in all those
proceedings … a right to relief claimed in all those proceedings are in respect of,
or arise out of, the same transaction or series of transactions … (or) for some
other reason specified in the application, it is desirable that an order be made …’
This even allows for an application when the proceedings are not before the
same tribunal. It calls on the tribunals to communicate with a view to a joint
order of consolidation and otherwise calls for their coordination where
consolidation is not to occur. Orders may be made:

(a) that the proceedings be consolidated on terms specified in the order;

(b) that the proceedings be heard at the same time or in a sequence specified in
the order;
(c) that any of the proceedings be stayed pending the determination of any
other of the proceedings.

It has been suggested that this may limit the parties' autonomy to craft their
arbitration agreement to prevent consolidation, (245) although this would depend
on how the provision is interpreted. The operation of this provision is more
limited than it looks. In the first place, section 24 is part of Division 3 of Part III
of the International Arbitration Act. Part III gives effect, in Australia, to the
UNCITRAL Model Law. Division 3 contains certain additional provisions which
are optional. Division 3 only applies if the parties to the arbitration agreement
have agreed that the division applies. Moreover if there is more than one
tribunal appointed, both tribunals must agree to the consolidation or else the
application lapses. (246) It does not appear to be a mandatory norm and in any
event, there is only an entitlement to make application and no guarantee that
either tribunal will accede to a request. If the parties have expressed sufficient
intent not to allow for consolidation in their arbitration agreement, then a
tribunal facing an application might hold that the agreement waives the right to
make that application or the contrary intent is a relevant factor in deciding
against the exercise of the discretion.
The ICC Rules 2012 Article 10 provides an express provision in relation to
consolidation. Unlike joinder, where requests are made to the Secretary-General
who will then leave it for tribunal determination unless it is thought that a court
ruling is more appropriate, consolidation decisions are taken by the court alone.
The court has no authority to do so on its own volition but requires the request
of a party. The stipulated criteria for consolidation are that: page "572"

(a) the parties have agreed to consolidation; or

(b) all of the claims in the arbitrations are made under the same arbitration
agreement; or
(c) where the claims in the arbitrations are made under more than one
arbitration agreement, the arbitrations are between the same parties, the
disputes in the arbitrations arise in connection with the same legal
relationship, and the Court finds the arbitration agreements to be
compatible.

In addition to these criteria, Article 10 indicates that the court may take into
account any circumstances it considers to be relevant including whether any
arbitrators have been confirmed or appointed in more than one of the
arbitrations and if so whether the same or different persons have been
confirmed or appointed. Article 10 further provides that if arbitrations are to be
consolidated, they will be consolidated in the arbitration that commenced first
unless otherwise agreed by all parties. (247)

This is not uniformly so. Article 4.1 of the Swiss Rules 2012 allows consolidation
where the parties are not the same. This may arise where underlying contracts
are closely connected. The Swiss Chambers' Arbitration Court may order page
"573" consolidation on its own motion and does not require request of the
parties. The parties must be consulted. Article 4 of the 2012 Rules, however,
provides little guidance as to what must be considered by these bodies, save
noting that the link between the cases in respect of which consolidation is
proposed must be the subject of deliberation, as well as calling for ‘all relevant
circumstances' surrounding the disputes. (248) The Cepani Rules allow the
tribunal itself to call on the appointments committee or chairman to order
consolidation. (249) Article 12 of the Cepani Rules allows for joinder of
proceedings (another way to describe consolidation) where several contracts
containing the Cepani arbitration clause ‘give rise to disputes that are closely
related or indivisible …’. An order can be made by the appointments committee
or the chairman of Cepani either at the request of the tribunal or a party or on
Cepani’s own motion. Consolidation is possible in the Cepani Rules even if the
parties are not the same in whole or in part. Furthermore, all that is required is
some link or connection between the disputes.

Article 11 of the SCC Rules allows for the board of the SCC Arbitration Institute
to consolidate where the two arbitrations involve the same parties and concern
the same legal relationship. JCAA Rules allow for consolidation where the claims
are ‘essentially and mutually related …’ provided that all parties consent.
Consent is not required where the multiple requests arise out of the same
agreement. The SIAC Arbitration Rules may provide the tribunal with a power to
consolidate proceedings, but only on a broad interpretation of the wording of
Article 24:

In addition and not in derogation of the powers conferred by any applicable law
of the arbitration, the Tribunal shall have the power to:…

b. upon the application of a party, allow one or more third parties to be joined
in the arbitration, provided that such person is a party to the arbitration
agreement, with the written consent of such third party, and thereafter make
a single final award or separate awards in respect of all parties.
page "574"

ICSID Rules do not allow for consolidation applications. Nevertheless, the

practice has emerged to seek to have the same arbitrators appointed and
harmonise approaches as much as possible. (250) A unique provision on
consolidation is contained in the NAFTA. The decision on consolidation is not
taken by an administering body but by a separate tribunal established under
Article 1126 to decide consolidation. Article 1126(2) of the NAFTA provides:

2. Where a Tribunal established under this Article is satisfied that claims have
been submitted to arbitration under Article 1120 that have a question of law
or fact in common, the Tribunal may, in the interests of fair and efficient
resolution of the claims, and after hearing the disputing parties, by order:
(a) assume jurisdiction over, and hear and determine together, all or part of
the claims; or
(b) assume jurisdiction over, and hear and determine one or more of the
claims, the determination of which it believes would assist in the
resolution of the others. (251)

7.12. Joinder and Consolidation Practice

Given that there may be reasons for or against joinder or consolidation, it is

preferable to give a Tribunal a discretion. In most cases any power to order
consolidation is discretionary. Many matters could be considered in the exercise
of the discretion including:

• the degree to which the second case can be linked to the connecting test
within the first arbitration agreement. Stated another way, to what extent
could the separate action have instead been brought as an element of the
primary claim.
• the desire for efficiency and the avoidance of inconsistent results.
• the nature of the two disputes and whether efficiency would in fact be served
by hearing them together (for example, if one dispute is much more complex
than the other).
• whether the parties have provided for arbitration in different venues.
• whether the lex arbitri and/or the lex causae in the two matters differ.
• potential for irreconcilable decisions and delays.
• impact on relevant evidence, whether evidence part heard in the first case or
admissibility from one to the other. page "575"
• If the applicant for joinder or consolidation has delayed unduly without just
excuse, that alone might be grounds for denial, particularly if there would be
some prejudice to another party.

These issues must be considered in the context of due process norms and
potential challenges where a later joined party might argue that due process
requires some earlier issues to be revisited. While it will normally be desirable
to consolidate related proceedings, this will not always be the case, for example,
if one matter is close to the end when the second proceedings are commenced
or if one is far more complex and costly than the other.

Where a discretion is being exercised, the tribunal or institution ought to

consider whether the non-requesting party has been prejudiced by a late
request. This could be so for a range of reasons. Not only has at least one party
not been involved in tribunal composition but existing parties may have made a
different selection if the full range of participants was known from the outset. A
party might not have defended the original request or may have approached
settlement negotiations in a different way if it understood the intended broader
involvement. The degree of the relationship between the new party and existing
parties and the commonality of claims and issues should also impact upon
questions of tribunal composition and whether a new party can truly raise
concerns that it was not privy to the establishment of the tribunal. The more the
consolidation party can be seen as part of a group with an existing party, the less
justification for complaining about prior constitution of the tribunal.

Consolidation and joinder may lead to new impartiality challenges. The

circumstances should not vary between the Chair or sole arbitrator and party-
appointed arbitrators. In the context of majority decision-making each
arbitrator has an equal a priori vote. Each vote should be based on a
presumption of independence and impartiality. Either the involvement in the
other arbitration impacts upon this or it does not. Any adverse impact relates to
decisions already made or views already formed rather than the reasons for
appointment.

Another possibility is an anti-suit injunction by a tribunal. It is not clear whether

tribunals may make such orders and even if they do, how enforceable they may
be. A more conceptual concern is the extent to which one tribunal can seek to
pre-empt a jurisdictional decision of another. (252) This is discussed in section
8.2.10.5.

7.13. Case Management of Parallel Proceedings

Because of the restrictive preconditions for application of lis pendens, res

judicata and consolidation, attention needs to be given to the best way to
coordinate the page "576" many other overlapping scenarios where related
questions of fact and law may apply. Typical examples include construction and
shipping activities where there may be a multiplicity of parties and contracts.
There are a range of practical circumstances where there may be multiple
arbitral proceedings. A complex commercial transaction may have a number of
agreements each with separate arbitration clauses. There may be a dispute as to
which dispute resolution clause applies to an individual transaction, for
example, where a battle of forms has arisen. The first award may not have dealt
with all issues or may have refused to allow certain issues to be raised. The
parties may be required to commence differing actions, for example, where
insurance or guarantees are concerned. (253) If a second arbitration is instituted,
and it is not possible to consolidate the two arbitrations, is some form of
harmonisation or coordination between the two sets of proceedings possible
and desirable?

7.13.1. Same Tribunal

Even where there is no consolidation, appointment of common arbitrators may

lead to practical harmonisation of processes although there will still be
questions of admissibility of evidence, confidentiality and duties of expediency.
Where each dispute involves a sole arbitrator, it makes sense to appoint the
same person. This situation is more difficult with multi-person tribunals if the
methodology of appointment is to allow each to select the nominee of their
choice. If they freely choose to select the same tribunal well and good but it
would be difficult to impose this upon them against their wishes where this is
possible. Philippe Leboulanger (254) describes this as ‘de facto consolidation’. By
way of illustration he refers to the decision of the English Court of Appeal which
avoided the risk of contradictory awards by appointing the same arbitrator in
two parallel proceedings. In Abu Dhabi v. Eastern Bechtel (255) the parties
referred to court the question of whether separate arbitrators or the same
arbitrator should be appointed to the two arbitrations. The Court of Appeal held
that it had power to appoint the same arbitrator to both arbitrations. While this
elegant solution largely avoids the risk of inconsistent decisions, it is still not as
efficient as consolidating the arbitrations in one proceeding. However the
Tribunal can, by appropriate orders, direct that hearings be held sequentially
and therefore manage costs. In some cases courts may have the authority to
order that proceedings in two arbitrations be heard together. (256) If a joint
tribunal is to be appointed, it is suggested that there be a coordination page
"577" conference to sort out how the parties would expect coordination to occur
and so as to clarify the arbitrator's mandate. (257)

In order to have the same personnel on two tribunals it is necessary for the
parties, and sometimes the administering authority, to make identical
appointments. This does not always happen. The second Tribunal may consist of
different personnel or there may be some overlap of personnel between the two
tribunals but not complete identity. Anne Marie Whitesell and Edwardo Silva-
Romero (258) discuss ICC practice. They say that where the parties have not
agreed to have the same tribunal in parallel proceedings and one side decides to
nominate an arbitrator already acting in a related matter to which the opposing
party objects, the Court must decide whether to confirm that arbitrator. The
Court takes into account various factors including whether the parties, counsel
and the issues to be decided are identical and the stage that the arbitral
proceedings have reached. They say that the Court assesses whether the
arbitrator would have access to information that would not be available to other
members of the arbitral tribunal and also considers whether a decision has been
rendered in one of the matters that might cause the arbitrator to prejudge the
related case. Each case is evaluated separately and decisions can therefore go
either way depending on the circumstances. Whitesell and Silva-Romero give as
an example a case where the Respondent nominated an arbitrator acting in a
related case and the Claimant objected on the ground that the arbitrator would
have access to information not available to other members of the arbitral
tribunal. Counsel in both cases were the same, the claimants were the same and
the respondents were related companies. No award had been rendered in the
related case and there were no overlapping issues. The Court decided to confirm
the co-arbitrator. In consequence the claimant changed its mind and decided to
nominate the same co-arbitrator in the second case. However, in two more
recent cases the Court decided not to confirm a co-arbitrator even though the
parties and counsel were the same. The Court was influenced by the advance
stage of the first proceedings and the possibility the co-arbitrator could obtain
privileged information in the second proceedings.

Leboulanger (259) raises a question of good faith. He asks whether a party can be
considered to be acting in good faith when, on the basis of the existence of two
distinct but identical arbitration clauses contained in two inter-related
agreements, it seeks the constitution of two distinct arbitral panels and thereby
increases the costs and creates the risk of contradictory awards. He says that an
arbitration clause is nothing but one of the clauses of an agreement and the
principle of good faith page "578" should apply to the constitution of the arbitral
tribunal, which corresponds to the performance of the obligations assumed
under the arbitration clause. In his view a party who refuses to designate the
same arbitrator in parallel arbitral proceedings might be considered in violation
of its obligation to perform, in good faith, its undertakings assumed under the
arbitration clause.

7.13.2. Exchange of Information

Where the two tribunals are not identical, the risk of inconsistent decisions may
be reduced if there is an exchange of information or documentation between the
two arbitrations. Such coordination to avoid duplication will be impacted upon
by confidentiality issues. There are questions of admissibility in the face of
confidentiality and the view that a tribunal will take of the evidence from the
other proceedings. There is also a question whether statements in the two
proceedings will be compared to try and undermine the veracity of witnesses or
submissions. The better view is that there should not be obligations of
confidentiality where the same parties and arbitrators are involved. (260) In one
situation there were parallel ICSID and ICC arbitrations. The respondent in both
cases was the same but the claimants differed. The claimant in the ICSID case
was a shareholder of the claimant in the ICC case. The two tribunals were
different and there was no common member. The Tribunal in the ICSID case
ordered the respondent to produce all the documentation in the ICC case. The
ICC tribunal issued a corresponding order requiring the respondent to produce
all the documentation in the ICSID case.

Even where the parties are the same, but the tribunals differ and contain a
common member, an interesting question may arise. Can the common arbitrator
refer to or otherwise have regard to a document produced in arbitration A in
arbitration B? If the arbitrator discloses it, is it a breach of a duty of
confidentiality? As confidentiality belongs to the parties and as the parties are
the same in both proceedings, it might be thought that there was no breach. But
disclosure is being made to the other members of the Tribunal. Bernard
Hanotiau (261) says that the principle of neutrality, independence and impartiality
of the arbitrator is of paramount concern and the duty of confidentiality will
lead the arbitrator in some cases to reach the conclusion that it is no longer
possible to fulfil the arbitrator's duties in total independence or impartiality and
may have to resign. However, in other cases the arbitrator may simply make a
full disclosure of the problem to the co-arbitrators and the parties.

page "579"

7.13.3. Joint Sittings

If confidentiality issues can be overcome there may be efficiencies through joint

sittings. Another possibility is to have a joint preliminary conference between
the two cases, perhaps leading to an agreement on joinder or consolidation. (262)

7.13.4. Other Issues

The ILA Recommendations also touch on general questions of case management.

Paragraph 6 of the Recommendation on lis pendens covers circumstances where
the issues or parties may be different, but where issues in one proceeding may
influence the second, again inviting the second tribunal to consider staying
proceedings where appropriate, as long as the objecting party's interests are not
substantially affected.

A tribunal must be concerned with the interests of the parties before it. (263)
General societal concern for the avoidance of duplication or efficient resolution
of disputes should not determine a tribunal's behaviour except where these can
be said to be part of the intent of the parties before the tribunal. A tribunal
should not take decisions for the benefit of third parties. The parties themselves
should be consulted as to mechanisms aimed at promoting efficiency where
there are parallel proceedings. (264)

If there are to be a multiplicity of proceedings, it is also necessary to consider

the impact on quantum to ensure that an identical or related claimant in both is
not overly compensated. That may be more problematic when a first award has
not been satisfied or where the type of remedy is different in nature.
Complexities may arise depending on the hierarchy of companies. As has been
pointed out, compensation to a subsidiary should reduce any claims by a holding
company for damage to its shareholding. The converse does not hold as
compensation to a holding company does not necessarily filter down to a
subsidiary. (265)

One problem if all parties are not heard in each proceeding is if events in one
could be used to support an argument of denial of an adequate right to be heard
in the other. (266) A problem in having a third party involved as a witness rather
than a party is where they have expertise but also a vested interest. Their lack of
independence may prevent them being given the status of an expert witness.

page "580"

7.14. Class Arbitration

Arbitration is conceptualised as a dispute resolution process defined by consent

and direct participation of discrete parties in the relevant dispute. However,
particularly in the US, some interest in arbitration on a ‘class-wide’ basis has
developed in recent years. The development is highly contentious. Class
litigation has emerged in the US and in other common law countries. In a ‘class-
wide’, or simply ‘class’ action, one party claims to represent a class, with the
outcome able to bind and benefit the entire class. There will be rules as to fee
contribution, notice of class members and due process concerns. Where
litigation is concerned, such actions are seen as appropriate where issues are
identical and the costs to an individual as a percentage of potential relief
benefits are too high. Pooling resources aims to promote justice without unduly
interfering with the rights of the party that the class is opposed to. Where
litigation is involved, gateway issues such as identification of a definable class
and ensuring due process protections are under court control. Civilian
jurisdictions have tended to deal with the same regulatory concerns by
affording standing to consumer organisations. (267)

A question then arises as to whether this may or ought to apply in international

arbitration. Key questions include: can parties expressly agree on class
arbitration; if the agreement is silent, can class arbitration be commenced under
a broad clause, akin to a consolidation application; what notice requirements
are appropriate; must the other class members opt-in for arbitral consent or can
there be opt-out presumptions, i.e., they are covered unless they advise to the
contrary; what due process rights do the class members have from tribunal
composition onwards; will procedural and jurisdictional issues be determined
by the courts or by the tribunal?

It is inevitable that these questions are in part analysed in the context of US

experience to date although it is important to understand the particular context
in which it has arisen as it arises from a quite unique scenario. In most cases, the
possibility of class arbitration has arisen after providers of goods and services
sought to avoid the potential of class litigation by including mandatory
arbitration clauses in their contracts. Hence they purported to avoid class
actions, but found their contracting parties seeking to argue that broad
arbitration clauses allowed for class actions in any event. US courts dealing with
these issues had to consider what attitude to take to parties in stronger
commercial positions attempting to undermine class litigation entitlements of
their customers. The courts also had to consider questions of control in the
context of pro-class action regulations, where some might wish to have more
control over tribunal discretion under mandatory substantive law notions.

Class arbitration attempts to follow the same model as litigation; the party
initiating the arbitration claims to do so as a class representative; that is, on
behalf of a group of claimants. The members of the group are not necessarily
individually page "581" named; rather, at least in the US, the members of the
group will be defined as those potential claimants meeting a certain description
(e.g., a person who is party to an arbitration agreement with the respondent on
the same terms as the class representative, and to whom facts apply that are
comparable to those applying to the class representative). If the class arbitration
goes ahead, then members of the class will be bound by the award delivered by
the arbitral tribunal despite having taken no initiative to commence the
arbitration, choose the arbitrators, or in any other way participate in the
proceeding (although a party may avoid being bound by an award if it has
expressly indicated that it does not wish to be considered as a member of the
class - that is, it has ‘opted-out’ of the class proceeding). This may seem to run
counter to some of the basic principles underpinning arbitration, and indeed in
some jurisdictions the mechanism might be considered to undermine important
legal rights. As a result, class arbitration as a procedural mechanism remains the
subject of considerable discussion and disagreement. Differing views in US state
courts (268) also show that there would not be uniform views worldwide if the
phenomenon was more broadly utilised.

Critics of class arbitration cite various factors to support the contention that
arbitration is an inappropriate mechanism for class-wide dispute resolution.
Conceptually, they would argue that in most cases, there is inadequate consent
by the class. The concerns may also be practical, such as the long delay involved
in class arbitration, (269) its necessary complexity given the large number of class
members, and its high cost as a result both of that complexity and of the
extended period of the arbitration. There is also the added costs of the
uncertainty as to its validity and hence further transaction costs while its legal
parameters are worked out. An additional concern for potential respondents is
that class arbitrations can involve very large awards with no possibility of
appeal, presenting an additional layer of risk for respondents. Given that the
class is often very large and the individuals within the class may not always be
known, and due also to the need to publicly notify potential members of the
class of the pending arbitration, confidentiality will necessarily be
compromised. Indeed, the Supplementary Rules on Class Arbitration of the
American Arbitration Association (AAA) do away with confidentiality altogether
in a class arbitration, and in cases that bear on the public interest, governments
or the public may require that the confidentiality usually associated with an
arbitration be lifted. It is also suggested that class arbitration results in a formal,
legalistic process, and that it has been colonised by statutory procedural
requirements applicable to class litigation. (270) Accordingly, detractors argue,
class arbitration undermines the perceived classical characteristics of
arbitration: its page "582" speed, informality, low cost and amenity to
commercial imperatives including confidentiality. Indeed, these kinds of factors
contributed to the US Supreme Court's position in the case of AT&T Mobility LLC
v. Concecion (271) that class arbitration is an animal which is ‘not really
arbitration’. (272)

However, proponents of class arbitration emphasise its policy benefits and

argue that many of the concerns raised against class arbitration can be
addressed with appropriate procedural measures, or are illusory. For example,
although class arbitration will be lengthier and costlier than bilateral
arbitration, it may still be quicker and cheaper than class litigation. Total costs
of class arbitration will be vastly less than individual awards where res judicata
could not apply as between different litigants. In relation to considerations of
formality and confidentiality, it has been argued that these are not essential, or
even necessarily characteristic elements of arbitration, as parties to arbitration
are free to choose procedures of varying formality, and confidentiality is not
necessarily one of the cornerstones of arbitration that it is often held up to be.
(273) In response to the suggestion that class arbitration, with its potential for

high awards and inability to appeal, presents unacceptable risk to respondents,

it is noted that arbitration has for a long time been used to determine disputes
with awards of very high amounts. (274) In addition, its supporters argue that
class arbitration (like class litigation) serves important twin policy goals. First, it
enables potential claimants whose claims are too small individually to justify the
expense of initiating arbitration to defend their interests by combining their
claims with others in a similar situation to make arbitration economically
feasible. Second, the ability of claimants to do this, even for individual claims of
very low value, (275) discourages companies from engaging in wrongdoing. This
acts as a kind of de facto regulation that is seen as beneficial in a jurisdiction like
the US that emphasises a light regulatory regime. However, in jurisdictions that
tend toward greater regulatory supervision and within which an aggrieved
party might expect relief though regulation (such as within Europe), the policy
benefits described above are less valuable and carry lesser weight in offsetting
due process concerns.

7.14.1. Autonomy and Arbitral Laws and Rules

There is no reason why a large group of individuals cannot expressly agree on a

comprehensive class arbitration methodology, although that is unlikely to occur
in page "583" practice. If an attempt was made to do so, there would still be due
process issues and the need to ensure that the agreement was not in violation of
the mandatory due process norms of the relevant Seat. Concerns may arise if
members of the group waived their rights to make submissions or to engage in
tribunal selection. Views would differ as to whether this was permissible or
whether this was antithetical to arbitration per se.

Parties can also select arbitral rules that support class-wide actions although
this would need to be on an opt-in basis. Conversely, if parties prohibit class-
wide applications in their arbitration clauses, this should be respected. Such an
agreement may contradict a national law supporting class actions for protective
reasons but unless that law covered arbitration and was thought to be
mandatory, the waiver should apply to that form of dispute resolution at least. A
contrary view is that the ban on class arbitration is illegal and severable and the
balance of the arbitration promise stands, hence allowing for class arbitration in
any event. That would still need some logic in finding arbitral consent via a
clause that certainly did not expressly support it. Some US courts, having first
found that the class action waiver in an arbitration clause is unconscionable,
have then determined that the class action waiver is not severable from the
arbitration agreement. As a result, the arbitration agreement itself has become
invalid and the parties have proceeded with class litigation rather than class
arbitration. (276) It has been suggested that arbitration agreements aimed at
avoiding class arbitration should seek to be seated in a jurisdiction that gives
them effect (or, perhaps even better, does not recognise class actions generally).
Furthermore, it may be useful to set out in the arbitration agreement the
consequences for the agreement if the class arbitration waiver is considered
unconscionable or the arbitration agreement is otherwise found to allow class
arbitration. (277)

The more common situation as noted above, is where a claimant purports to

establish a class action under a broad arbitration clause in favour of others who
have similar clauses. The difference to joinder or consolidation applications is
that the latter bring the others in as full parties with equal rights, while class
actions bring them in on a full basis as to outcome but on a representative basis
as to process. Class arbitration in the US came into the mainstream following the
2003 Supreme Court ruling in Green Tree Fin. Corp v. Bazzle. (278) In Bazzle, the
US Supreme Court considered whether class arbitration was permitted under an
arbitration agreement that contained no mention of class arbitration. The
Supreme Court determined that the arbitration agreement did not prohibit class
arbitration, and that the question related to how the arbitration should proceed
rather than to the validity of the arbitration agreement itself. Accordingly, as
interpretation of the page "584" arbitration agreement was properly a matter
for the arbitrator, the matter was returned to the arbitrator for determination.
The arbitrator interpreted the arbitration agreement as allowing class
arbitration and the arbitration proceeded on that basis. Following Bazzle, some
US arbitrators tended to interpret arbitration agreements that are silent as to
class arbitration as allowing class arbitration, and class arbitration is now
specifically accommodated in the procedural rules of the American Arbitration
Association, JAMS and the National Arbitration Forum. Indeed, the AAA's public
docket of class arbitration reportedly lists several hundred arbitrations. (279)

However, two recent Supreme Court cases have created a climate much more
hostile to class arbitration. In Stolt-Nielsen S.A. v. Animalfeeds International Corp
(280) the Supreme Court noted that Bazzle had been a plurality decision, and

espoused the view that class arbitration is a fundamentally different process to

bilateral arbitration. As a result, the Court considered class arbitration is only
permissible where the parties have consented to it. This consent was to be
evaluated according to applicable principles of contractual interpretation, and
mere silence would not be sufficient. In AT&T Mobility LLC v. Concepcion, the
Supreme Court reconfirmed its view that ‘requiring the availability of classwide
arbitration interferes with fundamental attributes of arbitration and thus
creates a scheme inconsistent with the [Federal Arbitration Act]’, (281) deepening
the climate of hostility toward class arbitration. Accordingly, the future of class
arbitration in the US is unclear.

However, it is probably worthwhile, and at least instructive, to consider the

procedural framework developed by the AAA for class arbitration. Following
Bazzle, the AAA introduced its Supplementary Rules for Class Arbitration. The
rules provide for three distinct steps in the initiation of class arbitration. First,
the tribunal considers the arbitration agreement and issues a ‘clause
construction award’ determining whether the parties' agreement allows for
class arbitration. After the award is delivered, the proceedings are stayed in
order to let any party seek judicial review of the clause construction award. In
the next stage the tribunal determines whether the arbitration should proceed
as a class arbitration. This stage is analogous to the class certification
undertaken by courts in class litigation. In making its ‘class determination
award’, the tribunal must consider a wide variety of characteristics of the
proposed action, with a view to preserving the due process rights of the parties
affected. This analysis may include consideration of the following criteria
applied in US class litigation legislation:

(a) ‘numerosity’: is the class so numerous that the joinder of separate

arbitrations on behalf of all members is impracticable?
(b) ‘commonality’: are there questions of fact or law common to the class? page
"585"
(c) ‘typicality’: are the claims of the representative parties typical of those of
the class?
(d) ‘adequacy’ of representation: will the representative parties and class
counsel fairly and adequately protect the interest of the class?
(e) ‘similarity’: has each class member signed an agreement with an arbitration
clause substantially similar to that signed by the other class members? (282)

Following this, notice is issued to the class and the dispute proceeds to
consideration of the merits. The staged procedure allows for judicial oversight
and input at various critical stages and is intended to protect the interests of the
parties and also to reduce the possibility of the award being challenged later.

A critical perspective might make the following comments about these criteria.
As to numerosity, if the parties have not expressly agreed to class arbitration,
how are theories of consent to due process arbitration consistent with a view to
saying there are too many for joinder; why couldn't a sensible arbitrator allow
joinder and/or consolidation and manage the process in a cost effective way e.g.
via representative evidence, options to add individual evidence, with warnings
about individual costs awards for time-wasting additions etc; if the arbitration
agreements and issues are only similar and not identical, by what theory of
arbitration could class arbitration absent express agreement be justified in
going beyond principles of joinder and consolidation; as to adequacy of
representation, when can this be a basis for accepting or denying jurisdiction as
opposed to determining that a party should be warned when its representation
is inadequate? Absent express agreement, it would be hard to see why an
arbitrator could find a power for mass arbitration where joinder or
consolidation could not occur if only one of the class was sought to be included.
Furthermore, if the tribunal must treat all parties to an arbitration equally or
fairly and give each an adequate opportunity to be heard, at most a tribunal can
accept a waiver of direct rights when it concludes that the procedures it adopts
and the adequacy of counsel who will appear means that these due process
norms are not violated. It would be hard to accept such a waiver without the
class expressly considering this and agreeing, in which case they are consenting
to the proposed arbitration in any event.

However, it seems equally problematic to argue for a hybrid court supervised

model. If the tribunal can find jurisdiction, then it can adequately control the
process, including making adequate decisions on competence and jurisdiction.
(283)

page "586"

7.14.2. Class Arbitration outside the US

Class arbitration has occurred outside the US but, like class litigation, it is rare.
However, versions of class arbitration that are adapted to meet the particular
needs of a jurisdiction's legal culture are being developed. In particular, Stacie
Strong argues that ‘collective’, rather than class arbitration, holds significant
promise in European jurisdictions. Under this model, a representative
organisation (e.g., a trade association or public interest group) (284) represents
claims of its members against a respondent. However, it seems unlikely that
class arbitration based on the US model will gain much acceptance outside that
jurisdiction.

However, there is precedent for a form of class arbitration (styled ‘mass

arbitration’) emerging from an investor-State dispute between Italian
bondholders and the Argentine Republic (285) flowing from sovereign debt
restructuring by Argentina. The tribunal's determination (with a 2 to 1
majority) in relation to procedural aspects in this dispute is ‘historic in its
holding that there is no impediment to mass claims under the ICSID Convention
and Arbitration Rules and that ICSID tribunals have the power under ICSID
Arbitration Rule 19 to adopt procedures to handle mass claims’. (286) Effectively,
the tribunal considered that there was no bar to mass claims in the ICSID
Arbitration Rules, and that fairness and practicality compelled the conclusion
that the mass claim procedure should be employed. The tribunal summarised
the relevant considerations as follows:

[…] the Tribunal holds that the mass aspect of Claimants' claims does not
constitute an impediment to their admissibility. In particular:

(i) The silence of the ICSID framework regarding collective proceedings is to

be interpreted as a gap and not as a qualified silence;
(ii) The Tribunal has, in principle, the power under Article 44 ICSID
Convention to fill this gap to the extent permitted under Article 44 ICSID
Convention and Rule 19 ICSID Arbitration Rules;
(iii) The procedure necessary to deal with the collective aspect of the present
proceedings concern the method of the Tribunal's examination, as well as
the manner of representation of Claimants. However, it does not affect the
object of such examination. Thus, the Tribunal remains obliged to examine
all relevant aspects of the claims relating to Claimants' rights under the BIT
as well as to Respondent's obligations thereunder subject to the Parties'
submissions;
(iv) Such procedure is admissible and acceptable under Article 44 ICSID
Convention, Rule 19 ICSID Arbitration Rules, as well as under the more
general spirit, object and aim of the ICSID Convention; page "587"
(v) Respondent's policy arguments regarding the appropriateness of ICSID
proceedings in the context of sovereign debt restructuring are irrelevant
for the determination of the admissibility of the claims. (287)

The longer-term impact of the Abaclat dispute is unclear (the dispute remained
pending at the time of writing). It has been suggested that in its award on
jurisdiction the majority of the Abaclat tribunal was ‘very clear in that it was not
setting up general rules of procedure for mass dispute resolution within the
ICSID framework. Instead, the tribunal limited itself to creating a procedure to
be used in this particular dispute. However, in so doing, the tribunal had to have
been aware that ICSID awards are often considered to have some sort of
precedential or persuasive value, particularly with respect to matters of
procedure’ and therefore ‘it may very well be that future ICSID tribunals will
look to this award as persuasive authority regarding the procedures to be used
in mass investment arbitrations.’ (288)

7.14.3. Notice Considerations

A complication facing class proceedings is the need for provision of adequate

notice to the members of the class, even though the identities of some of those
members may be unknown. The AAA Supplemental Rules (at article 6) require
that ‘class members be provided the best notice practicable under the
circumstances (the “Notice of Class Determination”). The Notice of Class
Determination shall be given to all members who can be identified through
reasonable effort.’ Additional guidance as to what constitutes ‘the best notice
practicable’ is not provided in the rules, but given that adequate notice (or the
lack of it) is likely to have a significant bearing on the extent to which the final
award is viewed as observing due process standards, an arbitrator would
therefore be well advised to consider adhering to at least the class action notice
requirements of enforcement jurisdictions (to the extent that such requirements
exist in those jurisdictions).

7.14.4. Enforcement Considerations

As noted above, much of the acceptance of class arbitration in the US stems from
the policy benefits that it is considered to bring. In other jurisdictions, the US
form of class arbitration may be viewed as offending certain fundamentals of
arbitration as well as due process rights (in particular of absent members of a
class). As a result, parties may have an arguable case before a court in a
jurisdiction unfriendly toward class action that the award should not be
enforced. Accordingly, when considering whether to pursue class arbitration,
consideration should be given page "588" to the positions on class arbitration of
countries relevant to enforcement of the eventual award, and to whether it
would be possible, for example, for a dissatisfied party to challenge the award by
arguing that the class arbitration was contrary to the public interest in that
country.

7.15. Insolvency in Arbitration

7.15.1. Introduction

Commercial activity sometimes results in insolvency, and accordingly

commercial arbitration will sometimes encounter the insolvency of a party to an
arbitration agreement. Insolvency can affect the very entitlement to arbitral
proceedings, the person who may make decisions in relation to them, the
procedural conduct, the claims that may be heard, and the nature and content of
any award. (289)

The intersection of insolvency and arbitration throws up various issues. These

stem primarily from the inherent inconsistency between the goals and nature of
these two areas of law: insolvency law is concerned with equitable management
and disposal of the insolvent party's assets among its creditors. The law allows
for the overriding of autonomy for three key reasons: ensuring maximum total
return to creditors; providing for equal return where there is an insufficiency of
assets save for certain ordained priorities; and concern for innocent third
parties, such as employees. (290) Insolvency law will typically seek to restructure
debts and defer claims where there is a reasonable prospect of the entity
surviving. At a particular point in time, an insolvency regulator will consider the
interests of the debtor to be irrelevant. Once they are considered terminal in
terms of an ability to trade out of difficulty, the concern is instead to protect
creditors and other third parties. This is particularly important with insolvent
corporations where the controlling shareholders have a conflict in trying to
maximise their own returns at the expense of corporate creditors. (291)
Conversely, arbitration rests on the private agreement page "589" between the
parties and is concerned with resolving a particular dispute considering their
interests alone. Insolvency law is therefore motivated by public interest
considerations whereas arbitration is not. Not only is autonomy contrary to the
essence of insolvency regulation but it could also be easily abused. Insolvency
laws will typically seek to ensure that an impending insolvent is not able to
preference certain creditors over others for personal gain. Insolvency laws may
seek to invalidate an entire transaction where it occurred at a point in time
when the entity was unlikely to be able to continue in viable form. Such laws
would purport to be mandatory and would typically wish to leave it to domestic
courts to rule on their application. Optimal insolvency regulation would not
wish to see the mere presence of an arbitration agreement removing key
transactions from insolvency control. However, there would be problems if a
particular country had an unduly parochial insolvency regime or purported to
have comprehensive jurisdiction contrary to legitimate rights of other
regulators. Furthermore, it might be equally problematic if an organised
insolvency could be used as a simple means to avoid prior promises to arbitrate.
Here there may be an argument that the insolvency is an abuse of right. (292)

Furthermore, certain ‘core’ aspects of insolvency law (such as, for example, a
consideration of whether an entity is insolvent) are beyond the reach of
arbitration. All would agree that the key functions of insolvency administrators
are not arbitrable. These include decisions to impose an administrator, winding-
up decisions, rescheduling of debts and determination of pro rata distributions
to creditors. (293) Instead, arbitral tribunals are typically asked to consider what
implications insolvency has for the ongoing rights and obligations of the parties
and/or for the underlying contract or transaction.

Another challenge is that insolvency law has a particular territorial dimension

from an adjudicatory perspective. As insolvency law is concerned with
protecting creditors, it may seek to restrain the activities of the insolvent entity
even beyond the borders of the country in which the insolvency proceedings
take place to prevent an insolvent from redistributing assets beyond the reach
of the courts overseeing the insolvency or preferencing creditors contrary to the
dictates of the applicable insolvency law. Where arbitration is concerned,
because the Seat of arbitration is typically a neutral venue, invariably this will
require consideration page "590" of the relevance or otherwise of foreign
insolvency laws. Some countries may consider that the observation of
insolvency laws constitute part of their international mandatory law, which has
implications for arbitrations seated in those jurisdictions or for situations where
there is an attempt to enforce an award in them. Alternatively, insolvency
regimes may be implemented in a defined international area by specific
legislation such as the European Insolvency Regulation.

For the purposes of this book, most acutely, domestic insolvency law often
purports to limit the activity of the insolvent entity in order to facilitate the
administration of the insolvency, and this may extend to suspending or staying
legal proceedings (potentially including arbitration proceedings) in which the
entity is involved. Some countries have legislated specifically to preclude an
insolvent party from engaging in arbitration. These factors (which, among
others, are discussed in more detail below) purport to limit the scope of an
arbitral tribunal's jurisdiction or competence where an insolvent party is
involved. In addition to legal considerations, there are also practical questions to
consider when dealing with an insolvent party, such as ability to pay the costs of
arbitration or even satisfy an award against it. These policy concerns imply that
a harmonised international solution is preferable but differences in domestic
regimes and differing views about who should ultimately control insolvency has
made this difficult to achieve. There is harmonisation in Europe through EU
Council Regulation 1346/2000 (294) and an UNCITRAL Model Law on Cross-
Border Insolvency more generally but only a modest number of countries have
adopted the latter. Hence in most instances, an arbitral tribunal must determine
how to proceed by consideration of an amalgam of issues including questions as
to the scope of the arbitration agreement, capacity, arbitrability, application of
mandatory laws and applicable law generally. Because views can differ on each
of these issues and lex arbitri rarely provide directions, differing tribunals have
adopted inconsistent methods and solutions.

7.15.2. Legal Issues Raised Where an Insolvent Party Is Involved in

Arbitration

There are a number of different permutations as to the impact of insolvency on

an arbitration. An arbitration could be brought by a person or corporation that
is already insolvent. The second scenario is that an arbitration is brought
against such a person. A third scenario is that the claimant becomes insolvent
after the arbitration is commenced. The fourth scenario is that the respondent
becomes insolvent after the arbitration has commenced. The final scenario is
that the parties themselves are not insolvent but some assets under
consideration by the tribunal are claimed by an insolvency regulator.

page "591"

The first and third scenarios are relatively straightforward. It generally seems
unreasonable to consider the arbitration agreement itself to be invalid simply as
a result of insolvency. As Kröll points out, this view would mean that the
administrator could not bring claims under that agreement. (295) An insolvent
entity is entitled to seek recovery of its debts whether it is insolvent before or
after commencement of the arbitral proceedings. In each case it is simply a
question of who can speak on behalf of the relevant corporation. It is not a
question of capacity in the classical sense, namely the capacity of a particular
entity to enter into arbitration but instead who can make decisions for the
corporation, its traditional board of directors or instead some insolvency
administrator. Where the law applying to that corporation would designate an
insolvency administrator as the key decision maker, a tribunal should normally
defer to that designation. If the claimant becomes insolvent after the arbitration
has commenced, it is important to ensure that the insolvency administrator
wishes to continue with the proceedings. There may also be complications in
terms of advances on costs where that might be seen as an improper priority.
That should not be so under any well developed insolvency law that grants
broad powers to the administrator to make such payments. Once an
administrator is in place, it would normally be the case that this person is the
only one who could enter into new arbitration agreements.

The key challenges arise where the insolvent is the respondent and insolvency
law is used as a means to block the arbitral proceedings themselves or
enforceability of any award that would prioritise the arbitral creditor. Of course,
the situation is more complex where a counterclaim or set-off rights are
involved. (296) Differing arbitral solutions have tended to arise because
arbitrators have applied different methodologies to dealing with these
questions. Some have considered the broad question of arbitrability and what
kinds of insolvency matters should properly be before an arbitral tribunal. Some
have sought to question whether an insolvency invalidates an arbitration
agreement. Others have instead considered questions of capacity. In each
scenario, different tribunals may have differing approaches to identifying the
relevant law. Should the laws of insolvency or corporate control and capacity be
those of the place of incorporation, the place of performance or that of the Seat?
Given a tribunal's broad discretion as to applicable law, that can again lead to
differing outcomes. While it is generally the case that there is also a tension
between tribunals that are more concerned with implied intent rather than a
pure conflicts analysis, this is less applicable in this scenario as the key issue is
to determine when conflicting domestic insolvency laws should override clear
consent to arbitrate. (297)

page "592"

7.15.3. Arbitral Law and Arbitral Agreements

As noted above, an express agreement by the parties to continue arbitration in

the face of insolvency does not solve the issue as party autonomy may need to
defer to mandatory insolvency laws. Lex arbitri rarely seek to resolve the issue.
Set out below are several of the most critical aspects of domestic law that may
seek to constrain the jurisdiction of an arbitral tribunal. This is followed by a
consideration of the choice of law analysis which a tribunal may employ to
determine the extent to which such domestic requirements will be considered
as part of the arbitration.

The English Arbitration Act 1996 introduced a provision into the insolvency
legislation allowing a trustee in bankruptcy the choice whether to adopt or
reject an arbitration agreement. If the trustee does not accept the agreement it
is still possible for the creditors and the court to agree to do so. (298) Where the
European Union is concerned, EC Council Regulation 1346/2000 provides that
the law of the country where insolvency proceedings are commenced is
determinative of the impact of such proceedings on other proceedings, including
arbitration. An exception is made for pending proceedings where it is the law of
the State where proceedings are pending that will be determinative. In
arbitrations with the Seat in the EU, a failure to apply the EU Regulation might
lead to similar annulment or enforcement concerns as arose in the competition
law context in Eco Swiss. (299)

7.15.4. Arbitrability

As noted above, ‘core’ aspects of insolvency law will generally not be arbitrable.
Born provides the following list of such ‘core’ aspects:

(a) commencement, administration and winding-up of insolvency proceedings

(including proceedings that liquidate an insolvent company);
(b) rescheduling of the insolvent company's liabilities;
(c) operating the company under some form of receivership or administration;
 or
(d) distributing pro rata payments to designated creditors. (300)

Levy notes that in any case these kinds of issues would not be referred to
arbitration because there would be no arbitration agreement governing them,
and because they do not constitute disputes having an adversarial nature. (301)
However, where the insolvency of a party is not the question before the tribunal,
and rather the tribunal faces the issue of how the insolvency of a party affects
consideration of a dispute page "593" that does not itself relate to the
insolvency, then an arbitral tribunal may be able to resolve the dispute.

7.15.5. Jurisdictional Attempts to Limit the Arbitrability of Claims Involving

Insolvent Parties

Different States have differing views as to the degree to which they wish to
centralise insolvency claims. The less they are inclined to do so, the more they
are likely to accept party autonomy where arbitration is concerned. Some
countries (e.g., Spain, (302) Latvia, (303) The Netherlands (304) and Poland (305) )
purport to render an arbitration agreement void or inoperative in the event of
insolvency of a party to it. Other jurisdictions require a stay, and provide that
arbitration may commence or continue with consent from the trustee, the
creditors' committee or by court order. Stay of proceedings may be required in
Italy although this is unclear. (306) At the most liberal extreme (307) are countries
such as Switzerland, France (308) and Germany (309) where insolvency will not
affect arbitrability under domestic law. (310) The US courts apply a rule of reason
and will generally allow an arbitration to proceed ‘unless doing so would
seriously jeopardise the objectives of the [Bankruptcy] Code.’ (311)

Where restrictive laws apply, if these do not emanate from the Seat, a tribunal
will need to consider whether they apply as mandatory substantive laws. (312)

page "594"

Less clear is whether arbitration proceedings should be stayed even if the

agreement is not invalidated by the bankruptcy. Some would argue that this is a
discretionary matter for the tribunal. Certainly the parties themselves could
agree to stay their arbitration pending the outcome of insolvency
determinations. Where the parties are not in agreement, a tribunal's
consideration of a stay application would depend on the attitude of the Seat and
whether it would consider this to be a matter of public policy. (313) While some
commentators consider that allowing a stay may be ‘contrary to the goal of
arbitration’ (314) it may also be useful in particular cases to allow a brief stay to
enable a trustee to become familiar with the issues in the arbitration. Where
such a stay is necessary to ensure that procedural fairness norms are upheld,
then it may be considered compulsory. (315) Alternatively, a stay may be required
to allow domestic courts to decide a core issue which is non-arbitrable. (316)

7.15.6. Choice of Law: To What Extent, and in What Manner, Should the
Tribunal Take into Account the Insolvency of a Party?

Where domestic insolvency laws exist that would seek to confine the tribunal's
jurisdiction (e.g., by providing that an insolvent entity cannot be a party to
arbitration), the arbitral tribunal may undertake a choice of law analysis to
determine whether it is bound by those laws. Section 13.8 looks more generally
at the problem of the application of supposedly mandatory substantive laws.
There are diverse views on this more general question. Attitudes to that
question will obviously impact on one's views about the proper methodology for
dealing with insolvency scenarios.

Because of differing views on this issue, differences between conflicts and direct
choice regimes, different approaches to characterisation and overriding
discretions as to applicable law, one would not expect consistent approaches by
tribunals. Some of the uncertainties are well shown by the 2009 dispute
between Vivendi SA and Elektrim SA. The case demonstrates the importance of
the page "595" characterisation stage. In summary, Elektrim concerned the
acquisition by Vivendi, a French company, of an interest in a Polish
telecommunications company in which Elektrim, a Polish company, had a
significant interest. The transaction failed and arbitrations were initiated in both
England and Switzerland. After the proceedings had been initiated, Elektrim was
declared insolvent (by its own petition). (317) The English proceeding related to
an alleged breach by Elektrim of an investment agreement where the subsidiary
interest was sought to be acquired. The Swiss proceeding related to a settlement
agreement of a prior dispute which itself contained an arbitration clause.
Elektrim argued in both the English and Swiss arbitrations that, due to a
provision of Polish law, it could not continue with the arbitrations. The
provision (being Article 142 of the Polish Bankruptcy and Reorganisation Law)
provides as follows:

Any arbitration clause concluded by the bankrupt shall lose its legal effect as at
the date bankruptcy is declared and any pending arbitration proceedings shall
be discontinued.

Thus Polish law purported to apply as a mandatory norm to remove the validity
of each arbitration clause. The English and Swiss tribunals took different
approaches to the characterisation of the question raised by this provision. The
Swiss approach was to characterise the issue as one relating to Elektrim's
capacity to participate in an arbitration. Swiss law provided that capacity was a
question determined according to a party's law of incorporation, and
accordingly applied the prohibitive Polish provision. Accordingly, the
arbitration was discontinued. This position was upheld by a split decision in
Swiss courts (318) when Vivendi challenged the award on the basis that the
tribunal had wrongly refused jurisdiction. However, the English tribunal applied
English law to determine the choice of law rule relevant to the effect on
arbitration proceedings of the insolvency of a party. The parties were in
agreement that the arbitration agreement was governed by English law while
the balance of the contract was subject to Polish law. The relevant choice of law
rule was found in Article 15 of the European Insolvency Regulation, which
states:

The effects of insolvency proceedings on a lawsuit pending concerning an asset

or right of which the debtor has been divested shall be governed solely by the
law of the Member State in which that lawsuit is pending.

page "596"

That provision is an exception to the general norm of the insolvency regulation

to the effect that the law of the State where insolvency proceedings has been
opened governs the effect of such proceedings. (319) As the ‘lawsuit pending’ (a
term held in the instant case to include arbitrations) was seated in England, the
tribunal used this choice of law rule to identify English law as relevant to
determining the effect of Elektrim's insolvency on the arbitration. English law
did not prevent the arbitration continuing despite the insolvency of a party, and
accordingly, the tribunal considered that the arbitration should proceed. The
trustee for Elektrim took this issue to the English courts, and, broadly speaking,
the tribunal's position was upheld.

The differing approaches in the Elektrim dispute accordingly resulted in

diametrically opposing outcomes and demonstrate the critical role of
characterisation in initially charting the tribunal's course of reasoning.
Responses to the case tend to be based on commentators' views as to whether
party autonomy should in most instances be supported. For example, it is
suggested that ‘the whole conflict of law process should be governed by the idea
that party autonomy should be restricted as little as possible’. (320) However,
there is no recipe for ensuring that the ‘correct’ characterisation is arrived at. As
a starting point, Kröll suggests considering first whether what needs to be
characterised is a ‘particular rule of law pleaded by one of the parties’, or ‘the
more general question of the effects of a foreign insolvency on an ongoing
arbitration’. (321) Markert points out that the Swiss case could have come to a
different conclusion if it began with the question of the validity of the arbitral
agreement and not capacity per se. When considering validity under Article
178(2) of SPILA, the relevant law would be that chosen by the parties, the law
governing the dispute or Swiss law. The provision allows for application of
whichever law will provide for validity. (322) This may help to guide the tribunal
toward deciding which conflict of law rule to choose, and from there the matter
becomes more straightforward. Any approach building solely on validity and
capacity will be problematic. As has been pointed out, the approach in the Swiss
case which looks at capacity will effectively apply any national law that purports
to interfere with arbitration. (323) It has also been argued that national laws that
purport to undermine page "597" arbitration agreements could run counter to a
State's obligations under Article II(3) of the New York Convention. (324) New
York Convention provisions have to be considered in the context of subject
matter being capable of settlement by arbitration.

7.15.7. The Role of the Seat

If a specific rule is contained in the lex arbitri, that would prevail. If it is

contained in the insolvency law of the Seat, there are more ambiguous
jurisdictional and consent issues that arise. If the parties selected the Seat, were
they selecting the insolvency regime or only the arbitral regime? From a
jurisdictional perspective, did the legislature in the Seat intend the insolvency
rules to apply to neutral arbitrations under its international arbitration regime?

One other variable is whether the government in the Seat has at least adopted a
provision such as the EU Insolvency Regulation or the UNCITRAL Model Law on
Cross-Border Insolvency. (325) Where the latter is concerned, it merely refers to
‘proceedings’ but does not otherwise expressly indicate whether it intends to
cover or leave out arbitration. Where provisions of this nature apply, the Seat
has indicated how it wishes to deal with the conflict between arbitral consent
and universality in insolvency regulation. Where the lex arbitri is clear that
should be determinative. Where there is deference to universality generally, but
ambiguity as to whether this applies to arbitration, one might generally
presume that jurisdiction supporting universality would not wish to distinguish
between litigation and arbitral proceedings. That is only a policy presumption
and it is entirely possible that a particular jurisdiction wants to promote its
arbitral neutrality with commensurate financial gain and would not want the
same solution applying to each.

7.15.8. Other Practical Issues

7.15.8.1. Insolvency and the Entity Relevant to the Arbitration

Declaration of insolvency has varying consequences for the insolvent entity and
the legal characterisation of its connection with its assets. The trustee of those
page "598" assets may continue to deal with them as if it stood in the shoes of
the insolvent, or the estate may take an assignment or transfer of those assets.
This may raise a question of which entity is party to the arbitration agreement.
For example, where an estate takes a transfer or assignment of the insolvent's
assets, the assignee may argue that it is not bound by arbitration agreements
entered by its predecessor. However, it is generally considered that an
insolvency trustee may be a party to an arbitration in just the same way as the
now-insolvent entity. More troubling is whether the insolvency trustee has an
election as to whether to be bound or not. As noted above, this can vary between
jurisdictions.

In addition, the appointment of a trustee should prompt a review of the arbitral

tribunal (where one has already been established) to ensure that the arbitrators
have no relationship or association with the trustee that could constitute a
conflict of interest. Similarly, it should be confirmed that counsel for the
insolvent party may continue to act. (326) A further administrative point to
consider is whether the insolvency precipitates any change in notice details or
requirements as a failure to give proper notice may leave the award vulnerable
to challenge. Staying proceedings to notify the trustee is also highly relevant
from a due process perspective if that person becomes the one who will now
speak for the arbitral party. Proceedings may be stayed to allow an
administrator to form a view where the latter has a discretion as to which cases
to settle and which to proceed with. Arbitral proceedings might also need to be
stayed while the creditor registers claims under domestic insolvency norms. (327)
Registration of claims may be required as a matter of public policy to ensure
that the administrator's rights are not undermined through ignorance. (328)

7.15.8.2. Ability to Pay Costs

Generally, costs of arbitration will be equally shared between the parties until
such time as final costs orders are made. However, where one of the parties is or
becomes insolvent, it may be unable to pay such costs, either because it simply
does not have sufficient funds or because it is prevented from making payments
that prejudice other creditors. In these cases, the tribunal may invite the other
(solvent) party to pay the insolvent party's costs in order to keep the arbitration
on foot. The insolvent party may then be considered to be indebted to the
solvent party in the amount of the costs, but again, the solvent party will need to
join the other creditors in order to have this debt satisfied.

page "599"

If the solvent party refuses to pay the insolvent party's costs, then the tribunal
may withdraw from the arbitration and the arbitration agreement will be
considered unworkable. Cost determinations may be problematic from a due
process perspective where an insolvent respondent wishes to bring
counterclaims or have set-off defences considered. Where a company is solvent,
a tribunal will often be entitled to say that it will only adjudicate claims and
counterclaims by those funding the arbitration. It would be more problematic in
an insolvency scenario where the insolvent is simply unable to do so. To hear
the claims but not the cross-claims in such circumstances might lead to a
general due process challenge or challenge on public policy grounds, although
views may differ as to whether this ought to succeed.

A solvent party may wish to apply for security for costs as against an insolvent
party that brings claims. As with an order to pay under an award, this may
amount to a requirement for an unfair preferential payment at the expense of
other creditors and may therefore be vulnerable to challenge. Where security is
sought against a claimant, that should not be problematic as decisions by the
relevant administrator are generally protected under the relevant national law.
A difficulty would arise if an insolvent respondent sought to bring counterclaims
or set-off entitlements that were subject to a distinct security for cost
application. A tribunal would again have to consider due process implications in
such circumstances. As noted above, many may feel reluctant to allow the claims
but not the counterclaims in such circumstances, regardless of whether the
respondent has the ability to pay or not.

7.15.8.3. General Due Process and Timing Issues

Depending on timing there may also be problems in terms of nomination of an

arbitrator by the insolvent. Insolvency proceedings also aim at transparency,
(329) which may pose problems for confidential arbitration. Other problems will
arise if an insolvent simply refuses to take part in proceedings. Here a tribunal
will simply treat the party in the same way as it would treat any other non-
appearing respondent.

Complicated timing issues may also apply particularly where the arbitral
proceedings are likely to take longer than consolidated insolvency proceedings
in the governing jurisdiction. It would be problematic to delay the latter while
arbitral proceedings are finalised. However, unduly constricting the timeframe
for the arbitral proceedings could offend against due process norms. The
procedural scenario may be different if the administrator is merely joined as a
party rather than taking over the rights of the debtor entity. A tribunal might
also need to consider the issue in the context of parallel proceedings where it is
requested to recognise the foreign insolvency actions.

page "600"

7.15.8.4. Enforcement

In determining the extent to which it heeds domestic requirements in relation to

insolvent entities, the tribunal may wish to consider the effect of its decisions on
the likelihood that its award will be enforced in the countries where the
insolvent entity has assets. In particular, aspects of insolvency law may be
considered to be part of a given jurisdiction's public policy or mandatory law.
Also, on a practical level, the insolvent party may not have sufficient assets to
satisfy the eventual arbitral award. As is argued throughout, however, the duty
to render an enforceable award should not cause a tribunal to act differently on
questions of arbitral validity and capacity where they arise from time to time.

Where the claimant wishes to continue with the arbitration regardless of the
possibility that the insolvent entity might be unable to satisfy the award, or the
possibility that the award will be blocked in its country of enforcement due to
relevant laws considered mandatory in that country, it is suggested that the
tribunal should accept this, as ‘no one knows best what suits the party's
interests than the party itself’. (330) Effectively, the claimant ought to be allowed
to accept the risk that the arbitral award will not be enforceable (whether from
a practical or a legal standpoint). Furthermore, a tribunal may simply choose to
make a determination on the merits without doing so in an immediately
enforceable manner. It might simply make an order or direction as to
entitlement or could specifically indicate that enforcement is not to take priority
over the rights of creditors. This would also be contentious as the pro-
arbitration proponents could naturally argue that the agreement was to have a
final and binding award and a tribunal should not refrain from producing this
outcome without clear legal justification for doing so. (331)

7.15.8.5. Executory Contracts

A tribunal might also need to consider what attitude it will take to the
underlying rights and obligations of the parties where there are still aspects of
the contract to be performed. While the insolvency issues typically apply to
jurisdiction and procedural questions, they can also apply to the merits. For
example, insolvency laws may indicate what is to occur with executory contracts
or may consider the underlying contract in its entirety and improper dealing
where it arises after there is de facto insolvency. It may also apply if the
insolvency laws granted an administrator the right to cancel contracts in
appropriate circumstances.

page "601"

7.15.8.6. Determination in Relation to Assets Subject to the Insolvency

Proceedings

The final scenario is where neither party is insolvent but insolvency proceedings
purport to claim rights over assets subject to the arbitration itself. In many
cases, a tribunal is simply identifying the inter-party rights. A tribunal certainly
cannot make determinations binding on third parties although this will be the
practical implication if an inter-party determination allows one to deal with an
asset in a way contrary to the desires of an insolvency administrator.

7.16. Amicus Curiae

An ‘amicus curiae’, or literally ‘friend of the court’, is an entity which is not a

party to a proceeding but which requests an opportunity to provide (or is
invited to provide) information or perspective to the adjudicator in order to
round out the adjudicator's understanding of the dispute and its broader
context. The concept originates in common law courts, applying to a ‘“neutral
bystander” engaged by a court to provide impartial assistance’, (332) but it is not
uncommon for third parties to now seek to make submissions to international
arbitration tribunals, in particular in investor–State investment disputes. In
such circumstances the tribunal will need to consider whether to allow the third
party's involvement, and in doing so will need to consider how this may affect
the parties (at least one of which may object to the involvement of the third
party).

7.16.1. What is an Amicus Curiae?

The amicus curiae, as noted above, is not a party to the arbitral proceeding. (333)
The amicus curiae is an entity (e.g., non-governmental organisations, academic
institutions, not-for-profit organisations (334) or governmental organs such as
competition or antitrust regulators, (335) as well as indigenous groups and
business associations) (336) that has an interest in the outcome of the dispute,
and which also has page "602" expertise in an aspect of the subject matter
related to the dispute. The amicus' interest and expertise may involve aspects of
public policy, regulation or environmental protection, and in any case has
tended to revolve around questions of the public interest on which the arbitral
proceedings have some bearing and which the actual parties to the dispute may
not bring to the tribunal's attention.

7.16.2. Apparent Tension between Acceptance of Submissions from Amici

Curiae and the Goals of Arbitration

The potential involvement of amicus curiae airing public interest considerations

in international commercial arbitration may seem to be at odds with the
generally private nature of such proceedings. In particular, it will generally be
the case that at least one of the parties to the arbitration will object to the
tribunal's acceptance of submissions or evidence from third parties. (337) The
parties are likely to find themselves arguing a broader range of points than
would have been necessary if there was no third-party intervention, the
arbitration will take longer as the submissions of the amicus curiae are
considered and argued and the tribunal may need to exert some additional
effort in order to reflect its consideration of the amicus' submissions in its
award. These consequences of amicus curiae involvement will increase costs for
all parties involved, and may raise points of fact or law that are disadvantageous
for one or more parties to the dispute. Costs orders cannot be made against an
amicus and it is not necessarily clear whether the parties would bear the
additional costs associated with the amicus' participation equally or whether
the losing party would bear them. Furthermore, the acceptance of amicus curiae
submissions is not generally explicitly countenanced within rules of arbitral
procedure, and accordingly it could be argued that acceptance of such
submissions goes beyond the tribunal's mandate as implied by the parties'
consent.

However, in certain circumstances tribunals may feel significant pressure, or

consider that there is significant benefit, in at least considering the acceptance of
amicus submissions. Specifically, amicus submissions have been accepted in
disputes which have a strong public interest dimension, in particular investor-
State and competition law disputes, as tribunals may feel that allowing
involvement of third parties who will advocate broader public interest
considerations may mitigate a public perception that private arbitration is not a
desirable method for resolving disputes involving sovereign States.

Amicus submissions had their arbitration debut in the 2001 Methanex case. This
was an investor-State dispute brought by a Canadian producer of ethanol
(Methanex Corporation) against the US in respect of a ban in California on an
ethanol-derived fuel additive. The ban was motivated by public health and
environmental considerations. Methanex considered that the ban was ‘arbitrary,
unfair page "603" and not based on credible scientific evidence’ (338) and that it
would cost Methanex close to USD 1 billion. An arbitral tribunal was established
pursuant to Chapter 11 of NAFTA using the UNCITRAL arbitration rules. The
International Institute for Sustainable Development (USD), an NGO, petitioned
the tribunal requesting the ability to submit an amicus curiae brief, and also to
be granted observer status and to make oral submissions. The USD wished to
raise ‘critical legal issues of public concern’, including environmental
considerations, and also suggested that the acceptance of its brief would help to
create a perception of transparency and stakeholder involvement in the
outcome of the dispute. (339) From a technical perspective, USD argued that the
tribunal had jurisdiction to accept its brief under what is now Article 17(1) of
the UNCITRAL rules (340) (at the time of the Methanex dispute the relevant article
was Article 15(1)). Broadly speaking, the tribunal found that it had jurisdiction
to accept USD's written amicus submission under the UNCITRAL rules. (341) The
tribunal also considered that any extra burden on the parties flowing from
acceptance of the amicus brief would not be ‘inevitably excessive’ and that its
acceptance did not appear to present any immediate risk of unfairness.

7.16.3. Considerations for Tribunals when Deciding whether to admit

amicus briefs

A petition from an aspiring amicus curiae raises a variety of issues for

consideration by the tribunal. First, the tribunal will need to determine whether
it has jurisdiction to accept the brief (as well as other requests that the amicus
might have, such as a request for observer status, or for access to key arbitral
documents) under its procedural rules. Most arbitral rules will probably not
include specific provisions allowing a tribunal to accept submissions from third
parties. In such cases, a tribunal might rely upon a general power to conduct the
proceedings in a manner that does justice between the parties. Indeed, as noted
above, in Methanex the tribunal considered that the general power to preserve
equality and fairness under what is now Article 17(1) of the UNCITRAL rules
provided sufficient jurisdiction for the acceptance of written submissions in the
absence of any page "604" contrary rule. However, following the advent of
amicus briefs in various investor–State arbitrations, the International Centre for
the Settlement of Investment Disputes (ICSID) changed its arbitration rules in
2006 to provide specifically for the possibility of admitting amicus submissions,
and also set out various criteria for the tribunal to consider when deciding
whether to accept the submission. Rule 37(2) of the ICSID Arbitration Rules
provides:

(2) After consulting both parties, the Tribunal may allow a person or entity that
is not a party to the dispute (in this Rule called the ‘non-disputing party’) to
file a written submission with the Tribunal regarding a matter within the
scope of the dispute. In determining whether to allow such a filing, the
Tribunal shall consider, among other things, the extent to which:
(a) the non-disputing party submission would assist the Tribunal in the
determination of a factual or legal issue related to the proceeding by
bringing a perspective, particular knowledge or insight that is different
from that of the disputing parties;
(b) the non-disputing party submission would address a matter within the
scope of the dispute;
(c) the non-disputing party has a significant interest in the proceeding.

The Tribunal shall ensure that the non-disputing party submission does not
disrupt the proceeding or unduly burden or unfairly prejudice either party, and
that both parties are given an opportunity to present their observations on the
non-disputing party submission.

Also of interest is Article 32(2) of the ICSID Rules, which relates to participation
in oral hearings:
(2) Unless either party objects, the Tribunal, after consultation with the
Secretary-General, may allow other persons, besides the parties, their
agents, counsel and advocates, witnesses and experts during their
testimony, and officers of the Tribunal, to attend or observe all or part of
the hearings, subject to appropriate logistical arrangements. The Tribunal
shall for such cases establish procedures for the protection of proprietary
or privileged information.

If a tribunal determines that the relevant procedural rules allow, in principle, for
the admission of third-party submissions or other third-party involvement, it
will then need to consider various practical aspects, most of which are included
in paragraphs (a) to (c) of Rule 37(2). Those considerations include an
evaluation of the likely value of any insight that the aspiring amicus will provide.
Part of this evaluation could include the apparent competence of the aspiring
amicus as part of consideration of whether the proposed entity is suitable for
inclusion as an amicus. In particular, the tribunal should consider the
‘background and accountability of potential amici curiae’. (342)

page "605"

A potential amicus may also request access to key arbitral documents. This
occurred in the Methanex dispute, where the tribunal declined such access on
the basis that it did not have jurisdiction to release those documents, and in
another dispute (the Biwater v. Tanzania (343) dispute) on confidentiality
grounds. The tribunal in Biwater further noted that it expected the amicus to
provide its perspective on ‘broad policy issues’ (344) and not on specific legal or
factual points. This should not be a blanket rule as the distinction is somewhat
hard to maintain. If a broad policy issue has no relevance to questions of law or
fact, it should not trouble a tribunal. More contentious is access to documents.
All other things being equal, the quality and relevance of an amicus submission
is likely to be enhanced if an aspiring amicus is allowed to see key arbitral
documents. It has been argued (345) that this is the preferred position and that
confidentiality concerns could be dealt with using appropriate and
commonplace measures for ensuring confidentiality, but apparently a tribunal is
yet to allow such access.

In order to limit the additional burden on the proceedings exerted by the

admission of an amicus brief, the tribunal may set out specific requirements in
relation to the brief and its format, including a page limit. The tribunal may set
out specific requirements for petitioners, as occurred in the Biwater v. Tanzania
dispute. In that dispute the tribunal required that the petitioners provide: (346)

(a) The identity and background of the petitioner, the nature of its membership
if it is an organization, and the nature of its relationships, if any, to the
parties in the dispute.
(b) The nature of the petitioner's interest in the case.
(c) Whether the petitioner has received financial or other material support
from any of the parties or from any person connected with the parties in
 this case.
(d) The reasons why the Tribunal should accept the petitioner's amicus curiae
brief.

7.17. Third-Party Funding in International Arbitration

7.17.1. Introduction

Recent times have shown a significant increase in third-party funding of

domestic litigation and international arbitration. There are strong policy
arguments either page "606" way that will not be explored to any degree in this
book. Those in favour suggest that it may promote access to justice by parties
without funds, allows risk to be spread and allows for expert appraisers to
ensure that claims are properly valued. Contrary arguments are that there may
be conflicts of interest, ignorant parties may be faced with unfair funding
contract terms and may lose control over their own adversarial process.
Whatever the arguments at the domestic level, international arbitration would
have less in the way of regulatory mechanisms to maximise the benefits and
minimise problems if such mechanisms are to be pursued, hence the debate is
more problematic again.

In any event, it is already a phenomenon and the aim in this section is simply to
highlight the procedural and evidentiary issues that may flow as a result. Section
15.12.4 deals separately with the question of whether a successful party seeking
costs recovery may have this evaluated in the context of its third-party funding
arrangements. A tribunal would not normally consider itself as having a power
to award costs as against a third-party funder, not a party to an arbitration
agreement.

7.17.2. Powers, Rights and Obligations of Third-Party Funders

This section simply seeks to highlight potential problems facing a tribunal, as it

is an area of some uncertainty. (347) The issues should be looked at from a range
of perspectives. A number of important questions arise as between the party
and a third-party funder that it enters into any relationship with. The first
question is whether such an arrangement is legal under the relevant domestic
law, whichever that may be. Other issues are whether the third-party funder will
have ultimate control over what claims can be brought or defended, over what
resources to allocate and whether and on what terms settlement may be
effected. Where resources are concerned, there is a question as to whether
advances to be provided by the third-party funder are to be conditional on any
demands made by them. In relation to all matters, funders may vary from those
who seek to exercise ultimate control, to those who demand consultation at each
stage, to those at the other extreme that wish to distance themselves from the
conduct of the proceedings once the investment decision has been taken.
Another question is whether the third-party funder will have access to
confidential information pertaining to an arbitration while first vetting the case
for investment purposes and during the currency of the proceedings.
In some cases, the contractual relationship between the third-party funder and
an investor in an investment arbitration may even lead to the funder arguing
that it itself has investor rights under the relevant BIT. Other than in relation to
this question, most issues are as between the party and the funder and would
not directly constrain a tribunal. For example, if a claimant is unable to provide
advances refused by a third-party funder, the normal responses should pertain
page "607" as the claimant would presumably have been in a worse position
absent the original funding decision. Where confidentiality is concerned, a
tribunal would presumably apply the appropriate confidentiality rules as
between the parties, although it would have little power to sanction any breach.
Where settlement is concerned, in the extreme, if the tribunal thought that a
settlement was unreasonable, it need not give it the imprimatur of an award.

Where obligations are concerned, once again there are questions of

confidentiality although the tribunal does not directly control the third-party
funder. Because a tribunal may be asked to deal with document production
requests, there may also be requested documents in the possession of the
funder. If these meet the normal tests of materiality then there is no reason why
an order should not be made, although again direct control may not be possible.
Adverse inferences would be potentially applicable where the party had
sufficient control over the third-party funder. Where the funder is acting in a
way that displays a clear conflict of interest, a tribunal would presumably
respond in the same way as if counsel acted in that manner.

7.17.3. Rights and Obligations of the Parties

The most important question is whether the parties would have a duty to
disclose the presence and/or terms of a third-party funding arrangement to
either other parties and/or the tribunal. One issue the parties need to consider
is whether the disclosure of confidential material to a third-party funder could
be said to be a waiver of privilege under the relevant law.

7.17.4. Rights and Obligations of the Tribunal

The most important issue from the tribunal's perspective is whether any links
between an arbitrator and the third-party funder could compromise the tribunal
in terms of impartiality and independence. (348) Given that the IBA Guidelines
deal with shareholder interest in parties, because third-party funding
arrangements are essentially investments, similar principles can readily apply.

The second key issue is in relation to the potentially negative impact on the
tribunal's control over parties and counsel if behind the scenes third-party
funders are making the key decisions. Once again a proactive tribunal can make
the necessary orders against parties and counsel and may not need to directly
involve themselves with third-party funders.

page "608"
* The title to this chapter uses the descriptor from Bernard Hanotiau, Complex
Arbitrations, Multi Party, Multi Contract, Multi-Issues and Class Actions (The
Hague: Kluwer Law International, 2005).
1 These questions have largely been drawn from Bernard Hanotiau,
‘Introduction’, in Multi-Party Arbitration, Dossier of the ICC Institute of World
Business Law (2010) 7.
2 Nathalie Voser, ‘Multi-Party Disputes and Joinder of Third Parties’, Fifty Years

of the New York Convention: ICCA International Arbitration Conference, Congress

Series No. 14, ed. Albert Jan van den Berg (Alphen aan den Rijn: Wolters Kluwer,
2009), 350.
3 This is the purpose behind the express mandate to bring set-off claims under

the Swiss Arbitration Rules 2012 Art. 21.5.

4 Parties can also always agree to modifications and tribunals sometimes will

invite agreement but cannot force it. An example of such an invitation was in the
Sofidis case. Interim Award No. 2 in ICC Case No. 5124 (unpublished) cited in
Klaus Peter Berger, ‘Setoff in International Economic Arbitration’, (1999) 15
Arbitration International 53, at 65 (fn. 88).
5 ICC Case No. 7453 of 1994 (1997) 124 Jnl du Droit Int'l (Clunet) 107.
6 For example, if an arbitration clause provides that claims are to be brought in

an arbitration where the respondent's country is to be the seat, should a

counterclaim also have that seat or was the intent to have any defender of a
claim to be at ‘home’ during proceedings? See Vladimir Pavić, ‘Counterclaim and
Set-Off in International Commercial Arbitration’, Annals International Edition,
2006, 105.
7 Nicholas Ulmer, ‘Winning the Opening Stages of an ICC-Arbitration’, Journal of

International Arbitration 8 (1991): 33, 42; Fletcher, ‘Unrealised Expectations –

The Root of Procedural Confusion of International Arbitrations’, Arbitration
International 4 (1988): 1, 40, 42.
8 Laurence Craig, William Park & Jan Paulsson, International Chamber of
Commerce Arbitration, 3rd edn (2000), 150.
9 Bernard Hanotiau, ‘Complex Arbitrations: Multi-Party, Multi-Contract, Multi-

Issue and Class Actions’, Arbitration International 14 (1998): 369, 371.

10 Philippe Leboulanger, ‘Multi-Contract Arbitration’, 13(4) J Int'l Arb 43 (1996),

43.
11 Jean-Francois Poudret & Sebastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), ¶ 312.

12 Laurence Craig, William Park & Jan Paulsson, International Chamber of
Commerce Arbitration, 3rd edn (2000), 150, 647.
13 ICC Case No. 12171, Award on Third Person Notice, 7 April 2004, B., Claimant

v. K., Respondent, ASA Bulletin 23 (2005), 271.

14 Martin Platte, ‘Multi-party Arbitration: Legal Issues Arising out of Joinder and

Consolidation’, in Enforcement of Arbitration Agreements and International

Arbitral Awards: The New York Convention in Practice, ed. Emmanuel Gaillard &
Domenico Di Pietro (London: Cameron May, 2008), 481.
15 Nathalie Voser, ‘Multi-Party Disputes and Joinder of Third Parties’, Fifty Years

of the New York Convention: ICCA International Arbitration Conference, Congress

Series No. 14, ed. Albert Jan van den Berg (Alphen aan den Rijn: Wolters Kluwer,
2009), 343.
16 Eric A. Schwartz, ‘Concluding Remarks’, in Multi-Party Arbitration, Dossier of
the ICC Institute of World Business Law (2010), 235, 236.
17 Bernard Hanotiau, Complex Arbitrations, Multi Party, Multi Contract, Multi-

Issues and Class Actions (The Hague: Kluwer Law International, 2005); Bernard
Hanotiau, ‘Non-signatories in International Arbitration: Lessons from Thirty
Years of Case Law’, in Albert Jan van den Berg (ed), International Arbitration
2006: Back to Basics, ICCA Congress Series No. 13, Alphen aan den Rijn: Kluwer
Law International, 2006, 341; see also Nathalie Voser, ‘Multi-Party Disputes and
Joinder of Third Parties’, Fifty Years of the New York Convention: ICCA
International Arbitration Conference, Congress Series No. 14, Albert Jan van den
Berg (Alphen aan den Rijn: Wolters Kluwer, 2009), 343.
18 See, e.g., l'Arbitrage et les tiers, in Revue de l'Arbitrage 3 (July–September

1988): 429–556. Multi-Party Arbitration: Dossiers of the Institute of International

Business Law and Practice, ICC Pub No. 480/1; Final Report of the ICC
Commission on Arbitration, Multi-Party Arbitrations (Chairman Jean-Louis
Delvolvé), ICC Court of Arbitration Bulletin 6 no. 1 (May 1995); Complex
Arbitrations: Perspectives on their Procedural Implications, ICC International
Court of Arbitration Bulletin, Special Supplement (2003).
19 See, e.g., the comprehensive analysis in Nathalie Voser, ‘Multi-Party Disputes

and Joinder of Third Parties’, Fifty Years of the New York Convention: ICCA
International Arbitration Conference, Congress Series No. 14, ed. Albert Jan van
den Berg (Alphen aan den Rijn: Wolters Kluwer, 2009), 343–410.
20 Nathalie Voser, ‘Multi-Party Disputes and Joinder of Third Parties’, Fifty Years

of the New York Convention: ICCA International Arbitration Conference, Congress

Series No. 14, ed. Albert Jan van den Berg (Alphen aan den Rijn: Wolters Kluwer,
2009), 344–345.
21 See, e.g., Karin Youssef, ‘The Limits of Consent: The Right or Obligation to

Arbitrate of Non-Signatories in Group of Company’, in Multi-Party Arbitration,

Dossier of the ICC Institute of World Business Law (2010) 71, 93; Bernard
Hanotiau, Complex Arbitrations, Multi Party, Multi Contract, Multi-Issues and
Class Actions (The Hague: Kluwer Law International, 2005), 8–9.
22 Nathalie Voser, ‘Multi-Party Disputes and Joinder of Third Parties’, Fifty Years

of the New York Convention: ICCA International Arbitration Conference, Congress

Series No. 14, ed. Albert Jan van den Berg (Alphen aan den Rijn: Wolters Kluwer,
2009), 347.
23 Nathalie Voser, ‘Multi-Party Disputes and Joinder of Third Parties’, Fifty Years

of the New York Convention: ICCA International Arbitration Conference, Congress

Series No. 14, ed. Albert Jan van den Berg (Alphen aan den Rijn: Wolters Kluwer,
2009), 348.
24 John M. Townsend, ‘Extending an Arbitration Clause to a Non-signatory

Claimant or Non-signatory Defendant: Does It Make a Difference?’, in Multi-Party

Arbitration, Dossier of the ICC Institute of World Business Law (2010) 111, 114.
25 Costs are considered in Chapter 15.
26 Fernando Mantilla-Serrano, ‘Multi Parties and Multiple Contracts: Divergent

or Comparable Issues?’, in Multi-Party Arbitration, Dossier of the ICC Institute of

World Business Law (2010) 25.
27 Yves Derains, ‘Is There a Group of Companies Doctrine?’, in Multi-Party

Arbitration, Dossier of the ICC Institute of World Business Law (2010), 138.
28 Stephen R. Bond, ‘Dépeçage or Consolidation of the Disputes Resulting from
Connected Agreements: The Role of the Arbitrator’, in Multi-Party Arbitration,
Dossier of the ICC Institute of World Business Law (2010), 35.
29 Noah Rubins, ‘Group of Companies Doctrine and the New York Convention’, in

Enforcement of Arbitration Agreements and International Arbitral Awards: The

New York Convention in Practice, ed. E. Gaillard & D. di Pietro (London: Cameron
May, 2008), 450.
30 Ibid., 459.
31 Ibid., 466.
32 See ibid., 466–468.
33 A number of these questions were raised by Eric A. Schwartz, ‘Concluding

Remarks’, in Multi-Party Arbitration, Dossier of the ICC Institute of World

Business Law (2010), 235, 238.
34 Cristián Comejero Roos, ‘Multi-Party Arbitration and Rule-Making: Same

Issues, Contrasting Approaches’, in Fifty Years of the New York Convention: ICCA
International Arbitration Conference, Congress Series No. 14, ed. Albert Jan van
den Berg (Alphen aan den Rijn: Wolters Kluwer, 2009), 411, 415.
35 Marc Blessing, ‘Extension of the Arbitration Clause to Non-signatories’, The

Arbitration Agreement – Its Multifold Critical Aspects (ASA Special Series No. 8,
1994), 151, 162.
36 Nathalie Voser, ‘Multi-Party Disputes and Joinder of Third Parties’, Fifty Years

of the New York Convention: ICCA International Arbitration Conference, Congress

Series No. 14, ed. Albert Jan van den Berg (Alphen aan den Rijn: Wolters Kluwer,
2009), 354–355.
37 Swiss Rules 2012 Art. 3 contemplates multiple parties but only as ‘claimants’

or ‘respondents’. A broader articulation is found in Art. 8.1 of the LCIA Rules.

38 Nathalie Voser, ‘Multi-Party Disputes and Joinder of Third Parties’, Fifty Years

of the New York Convention: ICCA International Arbitration Conference, Congress

Series No. 14, ed. Albert Jan van den Berg (Alphen aan den Rijn: Wolters Kluwer,
2009), 351.
39 Judgment of 7 January 1992, Société BKMI & Siemens v. Dutco, French Cour de

Cassation, Revue de l'Arbitrage (1992): 470. Platte suggests that where the
parties who must jointly appoint an arbitrator reach an agreement, equality of
treatment is maintained. In such cases, the right for each party to appoint an
arbitrator need not be treated as sacrosanct. Martin Platte, ‘Multi-Party
Arbitration: Legal Issues Arising out of Joinder and Consolidation’, in
Enforcement of Arbitration Agreements and International Arbitral Awards: The
New York Convention in Practice, E. ed. Gaillard & D. di Pietro (London: Cameron
May, 2008), 493.
40 Anne Marie Whitesell & Eduardo Silva-Romero, ‘Multi-Party and Multi
Contract Arbitration: Recent ICC Experience’, in Complex Arbitrations,
Perspectives on their Procedural Implications, ICC Ct. Bull, Special Supplement
(2003), 7 at 12.
41 Nathalie Voser, ‘Multi-Party Disputes and Joinder of Third Parties’, Fifty Years

of the New York Convention: ICCA International Arbitration Conference, Congress

Series No. 14, ed. Albert Jan van den Berg (Alphen aan den Rijn: Wolters Kluwer,
2009), 363.
42 Jean-Francois Poudret & Sebastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 202.

43 In addition to the ICC Rules, other rules that follow a similar post-Dutco
approach include Art. 13 DIS Rules; Art. 8 LCIA Rules; and Art. 8.5 Swiss Rules
2012. Art. 10 of the UNCITRAL Rules 2010 allows for whatever method the
parties have agreed and absent specific agreement, multiple parties as claimant
or respondent shall appoint an arbitrator. Where there is a failure to constitute
the tribunal the appointing authority shall, at the request of any party,
constitute the tribunal and may revoke any appointment already made and
appoint or re-appoint each of the arbitrators and designate one of them as
presiding arbitrator.
44 Nathalie Voser, ‘Multi-Party Disputes and Joinder of Third Parties’, Fifty Years

of the New York Convention: ICCA International Arbitration Conference, Congress

Series No. 14, ed. Albert Jan van den Berg (Alphen aan den Rijn: Wolters Kluwer,
2009), 363–364.
45 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2102.

46 Anne Marie Whitesell & Eduardo Silva-Romero, ‘Multi-Party and Multi
Contract Arbitration: Recent ICC Experience’, in Complex Arbitrations,
Perspectives on their Procedural Implications, ICC Ct. Bull., Special Supplement
2003, 7 at 12.
47 Nathalie Voser, ‘Multi-Party Disputes and Joinder of Third Parties’, Fifty Years

of the New York Convention: ICCA International Arbitration Conference, Congress

Series No. 14, ed. Albert Jan van den Berg (Alphen aan den Rijn: Wolters Kluwer,
2009), 365.
48 For the proposition that they might be treated as one and would normally be

expected to agree upon an arbitrator, see Yves Derains, ‘Is There a Group of
Companies Doctrine?’, in Multi-Party Arbitration, Dossier of the ICC Institute of
World Business Law (2010) 143.
49 Nathalie Voser, ‘Multi-Party Disputes and Joinder of Third Parties’, Fifty Years

of the New York Convention: ICCA International Arbitration Conference, Congress

Series No. 14, ed. Albert Jan van den Berg (Alphen aan den Rijn: Wolters Kluwer,
2009), 356; as to conflicts enquiries, see section 5.10.5.
50 Bernard Hanotiau, Complex Arbitrations, Multi Party, Multi Contract, Multi-

Issues and Class Actions (The Hague: Kluwer Law International, 2005), 218.
51 The problems that may arise with parallel arbitrations and some potential

solutions are examined by Hanotiau. Bernard Hanotiau, Complex Arbitrations,

Multi Party, Multi Contract, Multi-Issues and Class Actions (The Hague: Kluwer
Law International, 2005), 218–225.
52 Bernard Hanotiau, ‘Non-signatories in International Arbitration: Lessons

from Thirty Years of Case Law’, ICCA Congress Series No. 13, 341, 343. Born also
notes that extension is an inappropriate expression as the question is to
determine who has truly consented to an identified arbitration agreement. Gary
B. Born, International Commercial Arbitration (The Hague: Kluwer Law
International, 2009), 1139.
53 See, e.g., Motorola Credit Corp. v. Uzan 388 F. 3d 39 (2d Cir 2004).
54 Nathalie Voser, ‘Multi-Party Disputes and Joinder of Third Parties’, Fifty Years

of the New York Convention: ICCA International Arbitration Conference, Congress

Series No. 14, ed. Albert Jan van den Berg (Alphen aan den Rijn: Wolters Kluwer,
2009), 348.
55 James M. Hosking, ‘The Third Party Non-signatory's Ability to Compel
International Commercial Arbitration: Doing Justice without Destroying
Consent’, Pepp. Disp. Resol. L.J. 4 (2003): 483–485.
56 See, e.g., Dr Otto Sandrock, ‘Extending the Scope of Arbitration Agreements to

Non-signatories’, in The Arbitration Agreement – Its Multifold Critical Aspects,

A.S.A. Special Series No. 8 (December 1994): 169. The US case of Thomson-CSF
S.A. v.American Arbitration Association has sought to articulate the various
theories that flowed from traditional notions of contract, privity, agency and
corporate law. 64 F. 3d 773 (2d Cir 1995).
57 There are some cases particularly in the US where the corporate officer has

been accepted as party to the arbitration clause largely to provide it with

personal protection. Gary B. Born, International Commercial Arbitration (The
Hague: Kluwer Law International, 2009), 1146–1147. It could also be sought to
be extended where the officer is also the controlling ultimate owner.
58 Born notes but criticises the view that the law applicable is that of the alleged

agency agreement. Gary B. Born, International Commercial Arbitration (The

Hague: Kluwer Law International, 2009), 1147.
59 This is the approach adopted under the Hague Convention of 14 March 1978

on the Law Applicable to Agency, Art. 11.

60 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1150.

61 Ibid., 1148.
62 This approach to interpretation was discussed in section 3.2.4.
63 In the absence of other evidence, that person's intent might even be best

demonstrated by the law of agency they believe they are subject to, if that could
be known with confidence.
64 Pritzker v. Merrill Lynch, Pierce, Fenner & Smith 7 F. 3d 1110 (3d Cir 1993),

1112.
65 337 F. 3d 125 (2d Cir 2003).
66 Emmanuel Gaillard & John Savage (eds), Fouchard Gaillard Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 344, para. 716.
67 Daniel Girsberger & Christian Hausmaninger, ‘Assignment of Rights and

Agreement to Arbitrate’, Arbitration International 8 (1992): 121. The authors

note the complex choice of law problems where assignment arises.
68 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1192.

69 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 147.
70 Anthony C. Sinclair, ‘The Assignment of Arbitration Agreements’, in
Enforcement of Arbitration Agreements and International Arbitral Awards: The
New York Convention in Practice, ed. E. Gaillard & D. di Pietro (London: Cameron
May, 2008), 413.
71 Ibid.
72 Kaufman v. William Iselin & Co. Inc. 272 A.D. 578, 74 N.Y.S. 2d 23 (1947);

Lachmar v. Trunkline LNG Co., 753 F. 2d 8, 9–10 (2d Cir 1985).

73 Anthony C. Sinclair, ‘The Assignment of Arbitration Agreements’, in
Enforcement of Arbitration Agreements and International Arbitral Awards: The
New York Convention in Practice, ed. E. Gaillard & D. di Pietro (London: Cameron
May, 2008), 394 and cases cited in fn. 61.
74 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law
International, 2009), 1191.
75 See the English Contracts (Rights of Third Parties) Act 1999.
76 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1180.

77 Sea-Land Service, Inc. v. The Islamic Republic of Iran, Award No. 135-33-1 of

20 June 1984, 6 Iran–US Claims Tribunal Reports 149, 160–161.

78 Different results were obtained in the Swiss Federal Tribunal and an English

court in Tracomin S.A. v. Sudan Oil Feeds Co Ltd [1983] 1 WLR 1026 and Swiss
Tribunal Fédéral ATF 111 1b253 (1982) cited in Chartered Institute of
Arbitrators: Guidelines for Arbitrators Dealing with Jurisdictional Problems in
International Cases 119.
79 See, e.g., Emmanuel Gaillard & John Savage (eds), Fouchard Gaillard Goldman

on International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 344, para. 719.
80 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1198.

81 Ibid., 1209.
82 Ibid.; Nigel Blackaby et al., Redfern and Hunter on International Arbitration,

5th edn (Oxford: Oxford University Press, 2009), 105; Jean-François Poudret &
Sébastien Besson, Comparative law of International Arbitration, 2nd edn
(London: Sweet & Maxwell, 2007), 290.
83 Ibid., 1202.
84 Dow Chemical v. Isover Saint Gavain, Interim Award of 23 September 1982,

ICC Case No. 4131, Yearbook IX (1984), 131; confirmed by the Paris Court of
Appeal, Judgment of 21 October 1983, Rev Arb (1984), 98.
85 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1175.

86 Derains explores the Swiss response in that regard. Yves Derains, ‘Is There a

Group of Companies Doctrine?’, in Multi-Party Arbitration, Dossier of the ICC

Institute of World Business Law (2010), 131, 136–137.
87 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1167.

88 While there had been some debate, the better view is that there is no distinct

group of companies doctrine. An English court considered that the group of

companies doctrine forms no part of English law in Peterson Farms Inc. v. C & M
Faming Ltd [2004] 1 Lloyd's Rep 603 (QB).
89 Stephan Wilske, Laurence Shore & Jan-Michael Ahrens, ‘The Group of
Companies Doctrine – Where Is It Heading?’, Am Rev Int'l Arb 17 (2006): 73.
90 Bernard Hanotiau, Complex Arbitrations, Multi Party, Multi Contract, Multi-

Issues and Class Actions (The Hague: Kluwer Law International, 2005), 50.
91 Fernando Mantilla-Serrano, ‘Multi Parties and Multiple Contracts: Divergent

or Comparable Issues?’, in Multi-Party Arbitration, Dossier of the ICC Institute of

World Business Law (2010), 13.
92 Yves Derains, ‘Is There a Group of Companies Doctrine?’, in Multi-Party

Arbitration, Dossier of the ICC Institute of World Business Law (2010), 142–143.
93 Ibid., 140.
94 Noah Rubins, ‘Group of Companies Doctrine and the New York Convention’, in

Enforcement of Arbitration Agreements and International Arbitral Awards: The

New York Convention in Practice, ed. E. Gaillard & D. di Pietro (London: Cameron
May, 2008), 459.
95 Ibid., citing Society of Maritime Arbitrators Partial Final Award No. 1510 of 28

November 1980, VII Yearbook of Commercial Arbitration 151 (1982).

96 Laurence Craig, William Park & Jan Paulsson, International Chamber of
Commerce Arbitration, 3rd edn (2000), 150, para. 11.05.
97 Fernando Mantilla-Serrano, ‘Multi Parties and Multiple Contracts: Divergent

or Comparable Issues?’, in Multi-Party Arbitration, Dossier of the ICC Institute of

World Business Law (2010), 13.
98 As to the latter see Award in ICC Case No. 5103, 115 J.D.I. (Clunet) 1206

(1988) and criticised in Gary B. Born, International Commercial Arbitration (The

Hague: Kluwer Law International, 2009), 1169.
99 J.J. Ryan & Sons v. Rhone Poulenc Textile, SA, 863 F. 2d 315 (4th Cir 1988).
100 J.J. Ryan & Sons v. Rhone Poulenc Textile, SA, 863 F. 2d 315, 321 (4th Cir

1988).
101 J.J. Ryan & Sons v. Rhone Poulenc Textile, SA, 863 F. 2d 315, 320 (4th Cir

1988).
102 Société V2000 v. Société Project XJ220ITD et Autie, Paris Cour d'appel, 7

December 1994, Review de 1'Arbitrage (1996), 250, as translated in Fernando

Mantilla-Serrano, ‘Multi Parties and Multiple Contracts: Divergent or
Comparable Issues?’, in Multi-Party Arbitration, Dossier of the ICC Institute of
World Business Law (2010), 16–17.
103 Société Alcatel Business Systems (AbS) et al. v. Société Ankor Technologies et

al., Cour de Cassation, 27 March 2007, Pourvoi no. 04-20842, Revue de

1'Arbitrage (2007): 785 with a commentary by Jalal el-Ahdab, cited in Fernando
Mantilla-Serrano, ‘Multi Parties and Multiple Contracts: Divergent or
Comparable Issues?’, in Multi-Party Arbitration, Dossier of the ICC Institute of
World Business Law (2010), 17.
104 129 S. Ct. 1896 (2009).
105 ICC International Court of Arbitration Bulletin 2 no. 2 (1991), 31, 34.
106 Société V2000 v. Société Project XJ220ITD et Autie, Paris Cour d'appel, 7

December 1994, Review de l'Arbitrage (1996), 245, Commentary by Charles

Jarrosson, cited in Fernando Mantilla-Serrano, ‘Multi Parties and Multiple
Contracts: Divergent or Comparable Issues?’, in Multi-Party Arbitration, Dossier
of the ICC Institute of World Business Law (2010), 16.
107 Partial Award in ICC Case No. 60, ICC Ct Bull 2 No. 2 (1991) 31, 34.
108 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1194.

109 Deloitte Noraudit A/S v. Deloitte Haskins and Sells, US 9 F. 3d at 1060 cited in

James M. Hosking, ‘The Third Party Non-Signatory's Ability to Compel

International Commercial Arbitration: Doing Justice without Destroying
Consent’, Pepp. Disp. Resol. L.J. 4 (2003–2004): 469, 531.
110 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1198.

111 659 F. 2d 836 (7th Cir 1981).
112 Bernard Hanotiau, Complex Arbitrations, Multi Party, Multi Contract, Multi-

Issues and Class Actions (The Hague: Kluwer Law International, 2005), 24.
113 64 F. 3d 773 (2d Cir 1995).
114 See in the same vein E.I. Dupont de Nemours v. Rhone Poulenc Fiber & Resin
Intermediates, S.A.S., 269 F. 3d 187 (3d Cir 2001).
115 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1196-1197.

116 Corporate veil scenarios should not be confused with investment treaties

that expressly grant shareholders status as investors. Here claims by the

shareholder would relate to a distinct investment in law from that of the
corporation, although the commercial essence may be identical.
117 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1154.

118 Case Concerning the Barcelona Traction, Light & Power Co. [1970] ICJ Rep 3,

38–39.
119 Daniel Girsberger & Christian Hausmaninger, ‘Assignment of Rights and

Agreement to Arbitrate’, Arbitration International 8 (1992): 145–146.

120 Chartered Institute of Arbitrators: Guidelines for Arbitrators Dealing with

Jurisdictional Problems in International Cases 119.

121 Sébastien Besson, ‘Piercing the Corporate Veil: Back on the Right Track’, in

Multi-Party Arbitration, Dossier of the ICC Institute of World Business Law

(2010), 147, 149.
122 Ibid.
123 Should national law apply or should it be based on some transnational

standards? Sébastien Besson, ‘Piercing the Corporate Veil: Back on the Right
Track’, in Multi-Party Arbitration, Dossier of the ICC Institute of World Business
Law (2010) 147, fn. 15 and cases cited which take differing approaches.
124 England and Switzerland have been less inclined to look behind the
corporate veil. As to Switzerland see ICC Case No. 4402/1983 1 Collection of ICC
Arbitral Awards 153; ICSID Award on Jurisdiction in Takios Takelés v. Ukraine
(Case No. ARB/02/18 available at
<http://www.worldbank.org/icsid/cases/awards/htm>).
125 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1164–1165.

126 Sébastien Besson, ‘Piercing the Corporate Veil: Back on the Right Track’, in

Multi-Party Arbitration, Dossier of the ICC Institute of World Business Law

(2010), 147, 154.
127 For a discussion of the inconsistent Bridas cases see Derains and Schwartz, A

Guide to the ICC Rules of Arbitration, 2nd edn (The Hague: Kluwer Law
International, 2005), 86.
128 Bridas SAPIC, 345 F. 3d at 360; Bridas SAPIC, 447 F. 3d at 418.
129 Section 8.6 looks at pre-arbitral relief but this is consent based and is not

relevant to this concern.

130 Georgios Petrochilos, ‘Extension of the Arbitration Clause to Non-Signatory

States or State Entities: Does it Raise a Difference?’, in Multi-Party Arbitration,

Dossier of the ICC Institute of World Business Law (2010), 119, 122; see also
Svenska Petroleum Exploration AB v. Lithuania [2006] EWCA Civ 755.
131 Georgios Petrochilos, ‘Extension of the Arbitration Clause to Non-signatory

States or State Entities: Does It Raise a Difference?’, in Multi-Party Arbitration,

Dossier of the ICC Institute of World Business Law (2010), 119, 120.
132 Petrochilos criticises the case of Zeevi Holdings v. Bulgaria and the
Privatisation Agency of Bulgaria, Final Award, 25 October 2006, available at
<http://www.investmentclaims.com>, Georgios Petrochilos, ‘Extension of the
Arbitration Clause to Non-signatory States or State Entities: Does it Raise a
Difference?’, in Multi-Party Arbitration, Dossier of the ICC Institute of World
Business Law (2010) 119, 121–122, where the Republic of Bulgaria was
accepted as a party alongside the privatisation agency.
133 Dissolution of the signatory State entity was seen as the key factor in ICC

Case No. 7245. Interim Award of 28 January 1994 and cited in Karin Youssef,
‘The Limits of Consent: The Right or Obligation to Arbitrate of Non-signatories
in Group of Company’, in Multi-Party Arbitration, Dossier of the ICC Institute of
World Business Law (2010), 71, 95.
134 Georgios Petrochilos, ‘Extension of the Arbitration Clause to Non-signatory

States or State Entities: Does It Raise a Difference?’, in Multi-Party Arbitration,

Dossier of the ICC Institute of World Business Law (2010) 119, 124.
135 Court Appeal of Paris, 12 July 1984, République Arabe d'Egypte v.Southern

Pacific Properties (SPP) [1986] Rev Arb 75.

136 Award in ICC Case No. 8035, 124 J.D.I, (Clunet) 1040 (1997).
137 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1204.

138 Dallah Real Estate and Tourism Holding Co. v. The Ministry of Religious Affairs,

Government of Pakistan [2009] EWCA Civ 755.

139 Georgios Petrochilos, ‘Extension of the Arbitration Clause to Non-signatory

States or State Entities: Does It Raise a Difference?’, in Multi-Party Arbitration,

Dossier of the ICC Institute of World Business Law (2010), 119, 126.
140 Svenska Petroleum Exploration AB v. Lithuania [2006] EWCA Civ 755.
141 Final Award in ICC Case No. 9762, Yearbook of Commercial Arbitration 29

(2004), 38. The reasoning is criticised by Petrochilos on the basis of mixing

private law and international law concepts. Georgios Petrochilos, ‘Extension of
the Arbitration Clause to Non-signatory States or State Entities: Does It Raise a
Difference?’, in Multi-Party Arbitration, Dossier of the ICC Institute of World
Business Law (2010), 119, 123.
142 But see the contrary outcome in Bridas SAPIC et al. v. Turkmenistan et al.

(Bridas II) 447 F. 3d 411 (5th Cir 2006) paras 26–20.

143 Cristián Comejero Roos, ‘Multi-Party Arbitration and Rule-Making: Same

Issues, Contrasting Approaches’, in Fifty Years of the New York Convention: ICCA
International Arbitration Conference, Congress Series No. 14, ed. Albert Jan van
den Berg (Alphen aan den Rijn: Wolters Kluwer 2009), 411, 415.
144 Richard Bamforth & Katerina Maidment, ‘“All Join ln” or Not? How Well Does

International Arbitration Cater for Disputes Involving Multiple Parties or

Related Claims?’, ASA Bulletin 27, no. 1 (2009), 3:20.
145 James M. Hosking, ‘The Third Party Non-signatory's Ability to Compel
International Commercial Arbitration: Doing Justice without Destroying
Consent’, Pepp. Disp. Resol. L.J. 4 (2003–2004): 469, 569.
146 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1152.

147 Cour de Cassation, 26 Int'l Legal Material 1004 (1987). (Commonly referred

to as the ‘Pyramids’ case.)

148 Nathalie Voser, ‘Multi-Party Disputes and Joinder of Third Parties’, Fifty

Years of the New York Convention: ICCA International Arbitration Conference,

Congress Series No. 14, ed. Albert Jan van den Berg (Alphen aan den Rijn:
Wolters Kluwer, 2009), 405.
149 Nathalie Voser, ‘Multi-Party Disputes and Joinder of Third Parties’, Fifty
Years of the New York Convention: ICCA International Arbitration Conference,
Congress Series No. 14, ed. Albert Jan van den Berg (Alphen aan den Rijn:
Wolters Kluwer, 2009), 357.
150 William W. Park, ‘Arbitrators and Accuracy’, Journal of International Dispute

Settlement 1, no. 1 (2010): 25, 46–47.

151 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1164.

152 This is discussed in section 3.2.3.
153 Interim Award in ICC Case No. 9517, cited in Bernard Hanotiau, Complex

Arbitrations, Multi Party, Multi Contract, Multi-Issues and Class Actions (The
Hague: Kluwer Law International, 2005), 97.
154 John M. Townsend, ‘Extending an Arbitration Clause to a Non-signatory

Claimant or Non-signatory Defendant: Does It Make a Difference?’, in Multi-Party

Arbitration, Dossier of the ICC Institute of World Business Law (2010), 111, 117.
155 Karin Youssef, ‘The Limits of Consent: The Right or Obligation to Arbitrate of

Non-signatories in Group of Company’, in Multi-Party Arbitration, Dossier of the

ICC Institute of World Business Law (2010), 71, 93; Bernard Hanotiau, Complex
Arbitrations, Multi Party, Multi Contract, Multi-Issues and Class Actions (The
Hague: Kluwer Law International, 2005), 8–9.
156 See generally Karin Youssef, Consent in Context: Fulfilling the Promise of

International Arbitration (Minneapolis: West, 2009).

157 Noah Rubins, ‘Group of Companies Doctrine and the New York Convention’,

in Enforcement of Arbitration Agreements and International Arbitral Awards: The

New York Convention in Practice, ed. E. Gaillard & D. di Pietro (London: Cameron
May, 2008), 449, 457.
158 The Orri case cited in Yves Derains, ‘Is There a Group of Companies
Doctrine?’, in Multi-Party Arbitration, Dossier of the ICC Institute of World
Business Law (2010), 139, Cour de Cassation, 11 June 1991, Review de
1'Arbitrage (1992), 73 with a note by D Cohen.
159 Bernard Hanotiau, ‘Non-signatories in International Arbitration: Lessons

from Thirty Years of Case Law’, in International Arbitration – Back to Basics?,

ICCA Congress Series No. 13, ed. Albert Jan van den Berg (Alphen aan den Rijn,
Kluwer Law International, 2007), 341, 353.
160 Karin Youssef, ‘The Limits of Consent: The Right or Obligation to Arbitrate of

Non-signatories in Group of Company’, in Multi-Party Arbitration, Dossier of the

ICC Institute of World Business Law (2010), 71, 81.
161 Nathalie Voser, ‘Multi-Party Disputes and Joinder of Third Parties’, Fifty

Years of the New York Convention: ICCA International Arbitration Conference,

Congress Series No. 14, ed. Albert Jan van den Berg (Alphen aan den Rijn:
Wolters Kluwer, 2009), 352; Kristina M. Siig, ‘Multi-Party Arbitration in
International Trade: Problems and Solutions’, International J Liability and
Scientific Inquiry 1, no. 1/2 (2007): 72, 76.
162 Georgios Petrochilos, ‘Extension of the Arbitration Clause to Non-signatory

States or State Entities: Does It Raise a Difference?’, in Multi-Party Arbitration,

Dossier of the ICC Institute of World Business Law (2010), 119, 121; Sébastien
Besson, ‘Piercing the Corporate Veil: Back on the Right Track’, in Multi-Party
Arbitration, Dossier of the ICC Institute of World Business Law (2010), 147, 154.
163 Sébastien Besson, ‘Piercing the Corporate Veil: Back on the Right Track’, in
Multi-Party Arbitration, Dossier of the ICC Institute of World Business Law
(2010), 147, 154; Georgios Petrochilos, ‘Extension of the Arbitration Clause to
Non-signatory States or State Entities: Does it Raise a Difference?’, in Multi-Party
Arbitration, Dossier of the ICC Institute of World Business Law (2010), 119, 121.
164 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1142.

165 Nathalie Voser, ‘Multi-Party Disputes and Joinder of Third Parties’, Fifty

Years of the New York Convention: ICCA International Arbitration Conference,

Congress Series No. 14, ed. Albert Jan van den Berg (Alphen aan den Rijn:
Wolters Kluwer, 2009), 355.
166 Poudret and Besson support the more restrictive argument 221 criticising a

more liberal interpretation by the Swiss Federal Supreme Court, 16 October

2003, Av Z Sarl and Arbitral Tribunal ATF 129 III 727, in 22 ASA Bull (2004,
no.2): 364.
167 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1210–1211.

168 Articles II and V(1)(a) of the New York Convention.
169 Article V(1)(c) of the New York Convention.
170 Article V(1)(d) of the New York Convention.
171 Even an enforcement court, considering public policy questions may be

found to prefer its own domestic notions of good faith, abuse of rights and veil
piercing than those of the jurisdictions of the parties concerned. Separability
may also mean that a different law applies to the extension of the arbitration
agreement. See, e.g., Eurosteel Ltd v. Spinnes AG [1999] All ER 1394 (Con Ct).
172 Altain Khuder LLC v. IMC Mining Inc and Anor [2001] VSC 1.
173 Article 190(3) Swiss PILA.
174 Anne Marie Whitesell & Eduardo Silva-Romero, ‘Multi-Party and Multi
Contract Arbitration: Recent ICC Experience’, in Complex Arbitrations,
Perspectives on their Procedural Implications, ICC Ct. Bull., Special Supplement
2003, 7 at 12–14.
175 The following sections include material from Michael Pryles & Jeff
Waincymer, ‘Multiple Claims in Arbitrations Between the Same Parties’, in
Albert Jan van den Berg (ed.), International Council for Commercial Arbitration
Congress Series 14, Fifty Years of the New York Convention: ICCA International
Arbitration Conference (Alphen aan den Rijn: Kluwer Law International, 2009),
437–499 with the consent of the co-author.
176 Fernando Mantilla-Serrano, ‘Multi Parties and Multiple Contracts: Divergent

or Comparable Issues?’, in Multi-Party Arbitration, Dossier of the ICC Institute of

World Business Law (2010), 14.
177 Mantilla-Serrano sought to analyse the similarities and differences between

group of contracts and group of companies theory from the perspective of

doctrinal questions, economic issues and procedural scenarios. Fernando
Mantilla-Serrano, ‘Multi Parties and Multiple Contracts: Divergent or
Comparable Issues?’, in Multi-Party Arbitration, Dossier of the ICC Institute of
World Business Law (2010), 11.
178 Some commentators describe claims flowing in both directions between

claimant and respondent as ‘cross-claims’. This chapter refrains from using this
term as others limit it to the quite discrete question of whether one respondent
is able to bring separate claims as against other existing respondents or third
parties. See, e.g., Eduardo Silva-Romero, ‘Brief Report on Counterclaims and
Cross-Claims: The ICC Perspective’, in Arbitral Procedure at the Dawn of the New
Millennium: Reports of the International Colloquium of CEPANI, 15 October 2004,
73.
179 That may raise semantic debates about the difference between a contract

that ‘relates’ to another or which ‘results’ from another. There may still be
problems if later contracts have incompatible clauses.
180 Jean-Francois Poudret & Sebastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 268.

181 Pavić implies that this at first glance goes against the will of the parties but it

depends whether the express agreement to arbitrate under the Swiss Rules
provides the necessary consent per medium of the Article itself: Vladimir Pavić,
Counterclaim and Set-Off in International Commercial Arbitration, Annals
International Edition, (2006), 108.
182 Wolfgang Peter, ‘Some Observations on the New Swiss Rules of International

Arbitration’, ASA Special Series 22 (2004): 1, 9.

183 Phillipe Leboulanger, ‘Multi-contract Arbitration’, Journal of International

Arbitration 13 (1996): 43.

184 Bernard Hanotiau, ‘Problems Raised by Complex Arbitrations Involving

Multiple Contracts, Parties and Issues: An Analysis’ Journal of International

Arbitration 18 (2001): 253; Bernard Hanotiau, Complex Arbitrations: Multiparty,
Multicontract, Multi-issue and Class Actions (The Hague: Kluwer Law
International, 2005).
185 Award of 21 October 1983 (1986) 2 ICSID Reports 9.
186 In another arbitration an arbitral tribunal, applying the law of Luxembourg,

came to a different conclusion on the facts and did not consider two agreements
as a single legal relationship because ‘both the intentions of the parties and the
language of the relevant legal instruments do not permit such an application’.
ICC Award No 6829 of 1992.
187 Phillipe Leboulanger, ‘Multi-contract Arbitration’, Journal of International

Arbitration 13 (1996): 46–47, 52–53.

188 Martin Platte, ‘Multi-Party Arbitration: Legal Issues Arising out of Joinder

and Consolidation’, in Enforcement of Arbitration Agreements and International

Arbitral Awards: The New York Convention in Practice, ed. Emmanuel Gaillard &
Domenico Di Pietro (London: Cameron May, 2008), 489.
189 Bernard Hanotiau, Complex Arbitrations: Multiparty, Multicontract, Multi-

issue and Class Actions (The Hague: Kluwer Law International, 2005), 132.
190 An expert on banking law may take much more time to understand how a

Hudson formula works in a construction damages dispute. A non-legally trained

engineer may have more difficulty in considering the exceptions to immediate
operation of a demand guarantee.
191 These issues can also apply to consolidation applications discussed in
section 7.11.
192 Emmanuel Gaillard & John Savage (eds), Fouchard Gaillard Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 303–304 (fn. 262).
193 Emmanuel Gaillard & John Savage (eds), Fouchard Gaillard Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 301–306 (fns 264–267). hover v. Dow Chemical Co. (1984) Rev de l'Arb
137. An example is the award of 1 October 1980. In this case a set-off under a
loan agreement was accepted in a claim under a sales contract that alone had an
arbitration clause.
194 Jean-Francois Poudret & Sebastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), ¶ 309.

195 Emmanuel Gaillard & John Savage (eds), Fouchard Gaillard Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 302–304 (fn. 255).
196 [1990] Yearbook of Commercial Arbitration 74.
197 (1983) Rev L'Arb 119. See also Paulsson and Veeder's comments on ‘The

Vimiera’ in (1986) 2 Arbitration International 310; Hanotiau, supra n. 3, 274–

275 (fns 22, 23).
198 198 Phillipe Leboulanger, ‘Multi-contract Arbitration’, Journal of
International Arbitration 13 (1996): 77.
199 Anne Marie Whitesell & Eduardo Silva-Romero, ‘Multiparty and Multi-
contract Arbitration: Recent ICC Experience’, in ICC International Commercial
Arbitration Bulletin Special Supplement (2003) 7.
200 Bernard Hanotiau, Complex Arbitrations: Multiparty, Multicontract, Multi-

issue and Class Actions (The Hague: Kluwer Law International, 2005), 375.
201 For example, in ICC Case No. 5971, 13 ASA Bulletin 4 (1995), 728 a tribunal

concluded that three separate agreements with differing arbitration clauses

nevertheless all referred to the same purpose of the construction and operation
of a new facility to be operated as a joint venture. They therefore formed a unité
économique. As such, set-off claims arising in relation to the separate
agreements could nevertheless come to be directly covered under the broad
scope of the joint venture agreement. The only challenging conceptual question
was in relation to set-off claims that exclusively arose from the agreements. The
tribunal considered that because of the close interrelatedness of the three
agreements, set-off claims arising under either one must be heard and
considered by the tribunal under the principle le juge de l'action est le juge de
l'exception. The tribunal did not pass on the conclusion it would have drawn if
the set-off claims originated from a more ‘distant’ contract. Because of the
closeness, the tribunal felt it would have needed a clear indication that the
parties had the real intention to keep the three agreements totally separate from
each other if it was to rule against set-off claims.
202 For example, ICC Case No. 4392, 110 J.D.I. (Clunet) 907 (1983) refused to

extend an arbitration clause to a related agreement that had a jurisdiction

clause.
203 Dutch Code of Civil Procedure Art. 1046. In Australia, the different
Commercial Arbitration Acts of each State grant arbitral tribunals the power to
consolidate arbitrations where the arbitrations share an arbitrator. Where the
International Arbitration Act applies, consolidation may also be ordered under
the UNCITRAL Model Law.
204 Article 22.1(h) LCIA Rules; Art. 4.2 Swiss Rules. At the other extreme, the

rules might allow for the broadest category of reverse claims. An example 2012.
205 Price Waterhouse Report, ‘international Arbitration: Corporate Attitudes

and Practices’, (2006):7.

206 Nathalie Voser, ‘Multi-Party Disputes and Joinder of Third Parties’, Fifty

Years of the New York Convention: ICCA International Arbitration Conference,

Congress Series No. 14, ed. Albert Jan van den Berg (Alphen aan den Rijn:
Wolters Kluwer, 2009), 346–347.
207 See, e.g., Art. 12 Cepani Rules.
208 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2068 fn. 4.

209 See, e.g., Art. 41 NAI Arbitration Rules.
210 Nathalie Voser, ‘Multi-Party Disputes and Joinder of Third Parties’, Fifty

Years of the New York Convention: ICCA International Arbitration Conference,

Congress Series No. 14, ed. Albert Jan van den Berg (Alphen aan den Rijn:
Wolters Kluwer, 2009), 389.
211 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2070.

212 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 388.
213 Nathalie Voser, ‘Multi-Party Disputes and Joinder of Third Parties’, Fifty

Years of the New York Convention: ICCA International Arbitration Conference,

Congress Series No. 14, ed. Albert Jan van den Berg (Alphen aan den Rijn:
Wolters Kluwer, 2009), 359. For a contrary observation see Bernard Hanotiau,
Complex Arbitrations, Multi-Party, Multi-Contract, Multi-Issue and Class Actions
(Kluwer Law International, 2005), 165.
214 Martin Platte, ‘Multi-Party Arbitration: Legal Issues Arising out of Joinder

and Consolidation’, in Enforcement of Arbitration Agreements and International

Arbitral Awards: The New York Convention in Practice, ed. Emmanuel Gaillard &
Domenico Di Pietro (London: Cameron May, 2008), 487.
215 Gary B. Born, International Commercial Arbitration (The Hague:Kluwer Law

International, 2009), 2084.

216 Ibid., 2076.
217 Nathalie Voser, ‘Multi-Party Disputes and Joinder of Third Parties’, Fifty

Years of the New York Convention: ICCA International Arbitration Conference,

Congress Series No. 14, ed. Albert Jan van den Berg (Alphen aan den Rijn:
Wolters Kluwer, 2009), 398.
218 Bernard Hanotiau, Complex Arbitrations, Multi-Party, Multi-Contract, Multi-

Issue and Class Actions (The Hague: Kluwer Law International, 2005), 168.
219 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2087.

220 Ibid., 2077.
221 See, e.g., Art. 1045 Dutch Code of Civil Procedure; Art. 1696bis Belgian

Judicial Code.
222 Nathalie Voser, ‘Multi-Party Disputes and Joinder of Third Parties’, Fifty

Years of the New York Convention: ICCA International Arbitration Conference,

Congress Series No. 14, ed. Albert Jan van den Berg (Alphen aan den Rijn:
Wolters Kluwer, 2009): 387–388.
223 Anne Marie Whitesell & Eduardo Silva-Romero, ‘Multi-Party and Multi
Contract Arbitration: Recent ICC Experience’, in Complex Arbitrations,
Perspectives on Their Procedural Implications, ICC Ct. Bull., Special Supplement
2003, 7 at 11.
224 Richard Bamforth & Katerina Maidment, ‘“All Join ln” or Not? How Well Does

International Arbitration Cater for Disputes Involving Multiple Parties or

Related Claims?’, ASA Bulletin 21, no. 1 (2009): 3:13.
225 Nathalie Voser, ‘Multi-Party Disputes and Joinder of Third Parties’, Fifty

Years of the New York Convention: ICCA International Arbitration Conference,

Congress Series No. 14, Albert Jan van den Berg (Alphen aan den Rijn: Wolters
Kluwer, 2009), 400.
226 New Zealand Arbitration Act Schedule 2 Art. 2; Hong Kong Arbitration

Ordinance Art. 6B(1).

227 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2087.

228 Bernard Hanotiau, Complex Arbitrations, Multi-Party, Multi-Contract, Multi-

Issue and Class Actions (The Hague: Kluwer Law International, 2005), 166–168.
229 Report of the Working Group on Arbitration and Conciliation on the work of

its forth-sixth session, UN Doc. A/CN.9/619.

230 Secretariat Note A/CN.9/WG.II/NP.147/Add.1.
231 Working Group Report 619 para. 122; Markus Wirth, ‘The Current Revision

of the UNCITRAL Arbitration Rules’, in New Developments in International

Commercial Arbitration, ed. Christoph Müller (Zurich: Schulthess, 2007), 12 but
contrast J. Paulsson & G. Petrochilos, para. 138 who argue to the contrary.
232 Markus Wirth, ‘The Current Revision of the UNCITRAL Arbitration Rules’, in

New Developments in International Commercial Arbitration, ed. Christoph Müller

(Zurich: Schulthess, 2007), 12; Nathalie Voser, ‘Multi-Party Disputes and Joinder
of Third Parties’, Fifty Years of the New York Convention: ICCA International
Arbitration Conference, Congress Series No. 14, ed. Albert Jan van den Berg
(Alphen aan den Rijn: Wolters Kluwer, 2009), 408–409.
233 Secretariat Note A/CN.9/WG.II/NP.147/Add.1 para. 5.
234 Nathalie Voser, ‘Multi-Party Disputes and Joinder of Third Parties’, Fifty

Years of the New York Convention: ICCA International Arbitration Conference,

Congress Series No. 14, ed. Albert Jan van den Berg (Alphen aan den Rijn:
Wolters Kluwer 2009):409.
235 This is the view taken by Yves Derains, ‘The Limits of the Arbitration

Agreement in Contracts involving more than Two Parties’, in Complex

Arbitrations, Perspectives on Their Procedural Implications, ICC Ct. Bull. Special
Supplement 2003: 25, 33.
236 Nathalie Voser, ‘Multi-Party Disputes and Joinder of Third Parties’, Fifty

Years of the New York Convention: ICCA International Arbitration Conference,

Congress Series No. 14, ed. Albert Jan van den Berg (Alphen aan den Rijn:
Wolters Kluwer, 2009), 391.
237 Ibid., 394.
238 This section again extracts and draws from Michael Pryles & Jeff Waincymer,

‘Multiple Claims in Arbitrations Between the Same Parties’, Fifty Years of the
New York Convention: ICCA International Arbitration Conference, Congress Series
No. 14, ed. Albert Jan van den Berg (Alphen aan den Rijn: Wolters Kluwer, 2009),
and is reproduced with permission of the co-author.
239 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2077.

240 See, e.g., Art. 1046 Netherlands Code of Civil Procedure; Hong Kong
Arbitration Ordinance Schedule 2, s. 2.
241 The courts in New York had, until recently, allowed the possibility of court-

ordered consolidation of separate arbitral proceedings where they raised the

same issues of law or fact. However in 1993 the position changed and consent is
now necessary for consolidation. In Government of the United Kingdom of Great
Britain v. The Boeing Company, 998 F. 2d 68 (1993), the Court of Appeals for the
Second Circuit held that a District Court cannot order consolidation of
arbitration proceedings arising from separate arbitration agreements, even
where the proceedings involve the same questions of fact and law, unless the
parties have consented to such consolidation.
242 Ibid., 2090.
243 Martin Platte, ‘Multi-Party Arbitration: Legal Issues Arising out of Joinder

and Consolidation’, in Enforcement of Arbitration Agreements and International

Arbitral Awards: The New York Convention in Practice, ed. Emmanuel Gaillard &
Domenico Di Pietro (London: Cameron May, 2008), 490; Julian D.M. Lew, Loukas
A. Mistelis & Stefan M. Kröll, Comparative International Commercial Arbitration
(The Hague: Kluwer Law International, 2003), paras 16–66.
244 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2073.

245 Stephen R Bond, ‘Dépeçage or Consolidation of the Disputes Resulting from

Connected Agreements: The Role of the Arbitrator’, in Multi-Party Arbitration,

Dossier of the ICC Institute of World Business Law (2010), 37.
246 Section 24(7).
247 Article 4(6) of the ICC Rules of Arbitration 1976 previously provided for

consolidation as follows: ‘When a party submits a Request in connection with a

legal relationship in respect of which arbitration proceedings between the same
parties are already pending under these Rules, the Court may, at the request of a
party, decide to include the claims contained in the Request in the pending
proceedings provided that the Terms of Reference have not been signed or
approved by the Court. Once the Terms of Reference have been signed or
approved by the Court, claims may only be included in the pending proceedings
subject to the provisions of Article 19.’ Schäfer, Verbist and Imhoos suggested
that the discretion to consolidate goes against the principle of party autonomy
but argue nevertheless that where there is a ‘genuine connection’ between the
cases, consolidation will result in more effective proceedings and avoid the risk
divergent decisions. Erik Schäfer, Maître Verbist & Christophe Imhoos, Die ICC
Schiedsgerichtsordnung in der Praxis (Bonn: Economica Verlag, 2000), 34. If the
parties have expressly selected the earlier version of the ICC Rules and if one
concentrates on consent at the outset, one cannot necessarily view a decision by
the ICC Court to consolidate as going against party autonomy. There was also a
debate as to whether the court had the sole power to rule on consolidation or
whether in cases where the terms of reference have already been signed or
approved, indirect joinder can be effected by a tribunal under former Article 19,
dealing with acceptance of late claims. On its plain meaning, Article 19 allowed
for such a discretion. A converse argument would be that such an application is
in essence a consolidation application which should have been dealt with solely
under Art. 4(6). Schäfer, Verbist and Imhoos suggested that the tribunal might
make a decision under Art. 19 and then convey the decision to the parties and
the court for the latter to rule on the matter. Ibid. It is not clear how that
procedure can easily be derived from the express rules. If the tribunal does not
have the power under former Art. 19, then all the court would be doing is
determining that the tribunal has improperly applied Art. 19. If it does have that
power, there is no express jurisdiction for the court to have supervisory
jurisdiction over the tribunal’s determination. Where there are multiple
claimants, the ICC practice moved from an earlier strict position which held that
common proceedings could not be conducted if only one of the claimants could
not validly be incorporated with the rest, to a more relaxed position allowing
some to be consolidated. Anne Marie Whitesell & Eduardo Silva-Romero, ‘Multi-
Party and Multi Contract Arbitration: Recent ICC Experience’, in Complex
Arbitrations, Perspectives on Their Procedural Implications, ICC Ct. Bull., Special
Supplement 2003, 7 at 10. Voser rightly criticises the previous ICC practice in
demanding that the third party be a signatory when considering respondent’s
position as this puts it in a less favourable position than claimant where
extension arguments are presented. Nathalie Voser, ‘Multi-Party Disputes and
Joinder of Third Parties’, Fifty Years of the New York Convention: ICCA
International Arbitration Conference, Congress Series No. 14, ed. Albert Jan van
den Berg (Alphen aan den Rijn: Wolters Kluwer, 2009), 394.
248 ‘Where a Notice of Arbitration is submitted between parties already
involved in other arbitral proceedings pending under these Rules, the Court may
decide, after consulting with the parties and any confirmed arbitrator in all
proceedings, that the new case shall be consolidated with the pending arbitral
proceedings. The Court may proceed in the same way where a Notice of
Arbitration is submitted between parties that are not identical to the parties in
the pending arbitral proceedings. When rendering its decision, the Court shall
take into account all relevant circumstances, including the links between the
cases and the progress already made in the pending arbitral proceedings. Where
the Court decides to consolidate the new case with the pending arbitral
proceedings, the parties to all proceedings shall be deemed to have waived their
right to designate an arbitrator, and the Court may revoke the appointment and
confirmation of arbitrators and apply the provisions of Section II (Composition
of the Arbitral Tribunal).’
249 Cepani Arbitration Rules 2005 Art. 12.
250 Bernard Hanotiau, Complex Arbitrations, Multi-party, Multi-contract, Multi-

issue and Class Actions (The Hague: Kluwer Law International 2005) p 188.
251 If a Tribunal established under Art. 1126 assumes jurisdiction then other

Tribunals previously established under Art. 1120 cease to have jurisdiction with
respect to the claim or part of the claim over which the Art. 1126 Tribunal has
established jurisdiction. Art. 1120 Tribunal will ordinarily adjourn its
proceedings or they can be stayed by order of the Art. 1126 Tribunal.
252 Kaj Hobér, ‘Parallel Arbitration Proceedings – Duties of the Arbitrators’, in

Parallel State and Arbitral Procedures in International Arbitration, ed. Bernardo

N. Cremades & Julian D.M. Lew (Paris: ICC Publishing, 2005), 259.
253 Audley Sheppard, ‘The Scope and Res Judicata Effect of Arbitral Awards’, in

Arbitral Procedure at the Dawn of the New Millennium: Reports of the

International Colloquium of Cepani, 15 October 2004 (Brussels: Bruylant, 2005),
274.
254 254 Phillipe Leboulanger, ‘Multi-contract Arbitration’, Journal of
International Arbitration 13 (1996): 60.
255 [1982] 2 Lloyds Law Reports 425.
256 256 Phillipe Leboulanger, ‘Multi-contract Arbitration’, Journal of
International Arbitration 13 (1996) citing the former Arbitration Ordinance
1982 of Hong Kong.
257 Kaj Hobér, ‘Parallel Arbitration Proceedings – Duties of the Arbitrators’, in

Parallel State and Arbitral Procedures in International Arbitration: Dossiers of the

ICC Institute of World Business Law, ed. Bernardo N. Cremades & Julian D.M. Lew
(Paris: ICC Publishing, 2005), 258.
258 Whitesell and Silva-Romero,, ‘Multi-Party and Multi Contract Arbitration:

Recent ICC Experience’, in Complex Arbitrations, Perspectives on their Procedural

Implications, ICC Ct. Bull, Special Supplement 2003.
259 Phillipe Leboulanger, ‘Multi-contract Arbitration’, Journal of International

Arbitration 13 (1996): 90–91.

260 Associated Electric and Gas Insurance Services Ltd (AEGIS) v. European
Reinsurance Co of Zurich (European Re) [2003] 1 WLR 1041.
261 Hanotiau, Complex Arbitrations, Multi-Party, Multi-Contract, Multi-Issue and

Class Actions (The Hague: Kluwer Law International, 2005), 350.

262 Martin Platte, ‘Multi-Party Arbitration: Legal Issues Arising out of Joinder

and Consolidation’, in Enforcement of Arbitration Agreements and International

Arbitral Awards: The New York Convention in Practice, ed. Emmanuel Gaillard &
Domenico Di Pietro (London: Cameron May, 2008), 498.
263 Kaj Hobér, ‘Parallel Arbitration Proceedings – Duties of the Arbitrators’, in

Parallel State and Arbitral Procedures in International Arbitration, ed. Bernardo

N. Cremades & Julian D.M. Lew (Paris: ICC Publishing, 2005), 250.
264 Ibid.
265 Ibid., 243, 247.
266 Ibid., 258.
267 See, e.g., EU Directive 2009/22/EC, which requires EU Member States to

recognise an injunction ‘aimed at the protection of the collective interests of

consumers’ issued by an EU Member State.
268 See generally Bernard Hanotiau, ‘A New Development in Complex
Multiparty-Multicontract Proceedings: Classwide Arbitration’, Arbitration
International 20, no. 1 (2004): 39.
269 The Supreme Court noted in AT&T Mobility LLC v. Concepcion 131 S. Ct. 1740

(2011) that, of the approximately 300 class actions on the AAA's class
arbitration docket as at 2011 (which is made publicly available under the AAA
rules) not one had yet resulted in a final award ‘on the merits’ (although some
had been settled).
270 In the US, the AAA and JAMS procedural rules for class arbitrations mirror

Rule 23 of the Federal Code of Civil Procedure.

271 AT&T Mobility LLC v. Concepcion 131 S. Ct. 1740 (2011).
272 Gary B. Born, describing the court's position in Concepcion, “The U.S. Supreme

Court and Class Arbitration: A Tragedy of Errors' on the Kluwer Arbitration Blog,
1 July 2011.
273 Gary B. Born, “The U.S. Supreme Court and Class Arbitration: A Tragedy of

Errors' on the Kluwer Arbitration Blog, 1 July 2011.

274 For examples of high-value arbitrations in the past, see Gary B. Born, ‘The

U.S. Supreme Court and Class Arbitration: A Tragedy of Errors’ on the Kluwer
Arbitration Blog, 1 July 2011.
275 For example, in Concepcion the amount claimed by the Concepcions was

around USD 30.

276 For example, see Vasquez-Lopez v. Beneficial Oregon, Inc. 210 Or. App 553
(2007), referred to in Philip Lacovara, ‘Class Action Arbitrations – The Challenge
for the Business Community’, Arbitration International 24, no. 4 (2008): 554.
277 Philip Lacovara, ‘Class Action Arbitrations – The Challenge for the Business

Community’, Arbitration International 24, no. 4 (2008): 558–559.

278 Green Tree Fin. Corp. v. Bazzle 123 S. Ct. 2402 (2003).
279 The AAA's amicus brief to the Supreme Court in Concepcion listed 283 class

arbitrations opened with the AAA.

280 Stolt-Nielsen S.A. v. Animalfeeds International Corp. No. 08-1198, U.S. Sup

(2010).
281 AT&T Mobility LLC v. Concepcion 131 S. Ct. 1740, 9.
282 Gerald Aksen, ‘Class Action in Arbitration and Enforcement Issues: and

Arbitrator's Point of View’, in Dossier of the ICC Institute of World Business Law:
Multiparty Arbitration (Paris: ICC Publishing, 2010), 215–216.
283 Bernard Hanotiau, ‘A New Development in Complex Multiparty-
Multicontract Proceedings: Classwide Arbitration’, Arbitration International 20,
no. 1 (2004): 39, 54.
284 Stacie Strong, ‘Class Arbitration Outside the United States: Reading the Tea

Leaves’, in ‘Dossier VII – Multiparty Arbitration’ 183 (ICC Institute of World

Business Law, 2010): 191.
285 Abaclat and Others v. The Argentine Republic (ICSID Case No. ARB/07/5).
286 Andrew Newcombe, ‘Mass Claims and the Distinction between Jurisdiction

and Admissibility’, Kluwer Arbitration Blog, 24 October 2011.

287 Abaclat and Others v. The Argentine Republic, para. 551.
288 Stacie Strong, ‘Guest-Post: ICSID Accepts First-Ever Class-Type Arbitration’,

Karl Bayer ‘Disputing’ blog, 29 August 2011,

<www.karlbayer.com/blog/?p=15468>.
289 Domitille Baizeau, ‘Arbitration and Insolvency: Issues of Applicable Law’, in

New Developments in International Commercial Arbitration, ed. Christoph Müller

& Antonio Rigozzi (Université de Neuchâtel, Schultheff editions Romands, 2009),
98. This book is not concerned with proceedings before an insolvency court
where a creditor seeks to bar proceedings on the basis of the arbitration clause.
290 A US court has described the relation of arbitration and insolvency law as ‘a

conflict of near polar extremes …’. U.S. Lines, Inc. v. An. S.S. Oweners Nut. Prot. &
Indem. Ass'n Inc (In re United States Lines, Inc) 197 F. 3d 631 (640) (2d Cir
1999).
291 See generally Gabrielle Kaufmann-Kohler & Laurent Levy, ‘Insolvency and

International Arbitration’, in The Challenges of Insolvency Law Reform in the

Twenty-First Century, ed. H. Peter, N. Jeandin & J. Kilborn (2006), 257; Mantilla-
Serrano, ‘International Arbitration and Insolvency Proceedings’, Arbitration
International 11 (1995): 51; Vesna Lazic, Insolvency Proceedings in Commercial
Arbitration (The Hague: Kluwer Law International, 1998); Doug Jones,
‘International Dispute Resolution in the Global Financial Crisis’, The Arbitrator
and Mediator (October 2009): Doug Jones, ‘Insolvency and Arbitration: An
Arbitral Tribunal's Perspective’, presented at the INSOL Asia Pacific Rim Annual
Conference, Singapore, 13–15 March 2011; Darius Chan, Singapore Apex Court
Lays Down Clear Framework for Arbitrability of Insolvency Related Claims
(Kluwer Arbitration Blog, 23 May 2011)
<http://kluwerarbitrationblog.com/blog/2011/05/23/s...>.
292 Domitille Baizeau, ‘Arbitration and Insolvency: Issues of Applicable Law’, in
New Developments in International Commercial Arbitration, ed. Christoph Müller
& Antonio Rigozzi (Université de Neuchâtel, Schultheff editions Romands, 2009),
106 referring to ICC Award No. 5954 of 1991 and an unpublished ICC Award of
2009 where tribunals refused trustees the right to intervene and even refused to
recognise the trustees' power to represent the insolvent party.
293 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 809.

294 EC Regulation 1346/2000 of 29 May 2000 on Insolvency Proceedings, OJ L

160, 30 June 2000.

295 Stefan Kröll, ‘Arbitration and lnsolvency – Selected Conflict of Laws
Problems’, in Conflict of Laws in International Arbitration, ed. Franco Ferrari &
Stefan Kröll (Germany: Sellier, 2010), 243.
296 From an insolvency regulator's perspective, they might want to bring these

claims on behalf of respondent unimpeded, but still block claims against the
respondent that would preference an arbitral creditor.
297 One could of course hypothesise that general arbitration agreements could

impliedly be limited to cases of solvency with an understanding that applicable

insolvency laws will override party autonomy.
298 English Arbitration Act 1996 ss 107 and 349A English Insolvency Act 1986.
299 Eco Suisse v. Benetton [1999] ECR I 3055.
300 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 808.

301 Laurent Levy, ‘Insolvency in Arbitration (Swiss Law)’, International
Arbitration Law Review 8, no. 1 (2005): 28.
302 Spanish Insolvency Act Art. 52(1); Alejandro Lopez Ortiz, ‘Legislative
Comments, Spain: The New Insolvency Act and Arbitration’, International
Arbitration Law Rev 8, no. 2 (2005): 22.
303 Latvian Civil Procedure Law Art. 478(8). Stefan Kröll, ‘Arbitration and
Insolvency – Selected Conflict of Laws Problems’, in Conflict of Laws in
International Arbitration, ed. F. Ferrari & S. Kröll (Germany: Sellier, 2010), 216.
304 Netherlands Bankruptcy Act Art. 122.
305 As set out in Art. 142 of the Polish Bankruptcy and Reorganisation Law and

explored in the Elektrim dispute discussed below.

306 Domitille Baizeau, ‘Arbitration and Insolvency: Issues of Applicable Law’, in

New Developments in International Commercial Arbitration, ed. Christoph Müller

& Antonio Rigozzi (Université de Neuchâtel, Schultheff editions Romands, 2009),
102.
307 For a comparative summary of positions taken by various selected countries,

see J. Sutcliffe & J. Rogers, ‘Effect of Party Insolvency on Arbitration Proceedings:

Pause for Thought in Testing Times’, Arbitration 76 no. 2 (2010): 277–290.
308 Emmanuel Gaillard & John Savage (eds), Fouchard Gaillard Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 344, para. 577.
309 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th

edn (Oxford: Oxford University Press, 2009), 130.

310 Laurent Levy, ‘Insolvency in Arbitration (Swiss Law)’, International
Arbitration Law Review 8, no. 1 (2005): 23.
311 In Re United States Wines Inc. 197 F. 3d 631, 640 (2d Cir 1999).
312 This is not typically the result in arbitration but there are exceptions. See,
e.g., Award in ICC Case No. 9163, 2003 Rev Arb 227.
313 As has been considered in France. Gary B. Born, International Commercial
Arbitration (The Hague: Kluwer Law International, 2009), 814, fn. 1266.
314 Laurent Levy, ‘Insolvency in Arbitration (Swiss Law)’, International
Arbitration Law Review 8, no. 1 (2005): 30.
315 Ibid., 32, suggesting that a stay will be compulsory where a failure to impose

a stay would ‘prevent the trustee from putting his case (for lack of time to
prepare the files). More specifically, a failure to stay proceedings in France, for
example, will be considered a violation of France's public law and accordingly
leaves the award vulnerable to challenge’ (see at 30, and Domitille Baizeau,
‘Arbitration and Insolvency: Issues of Applicable Law’, in New Developments in
International Commercial Arbitration, ed. Christoph Müller & Antonio Rigozzi
(Université de Neuchâtel, Schultheff editions Romands, 2009), 102).
316 Domitille Baizeau, ‘Arbitration and Insolvency: Issues of Applicable Law’, in

New Developments in International Commercial Arbitration, ed. Christoph Müller

& Antonio Rigozzi (Université de Neuchâtel, Schultheff editions Romands, 2009),
101.
317 For a more detailed review of the Elektrim case, see I. Fletcher, ‘Josef Syska,

as Administrator of Elektrim SA v. Vivendi Universal SA: the EU Insolvency

Regulation and pending arbitration proceedings – The Court of Appeal Ruling on
Article 15’, in Insolvency International 22, no. 10 (2009): 155–157, and in
relation to choice of laws, see M. Robertson, ‘Cross-Border Insolvency and
International Commercial Arbitration: Characterisation and Choice of Law
Issues in Light of Elektrim SA. and Vivendi S.A. and Analysis of the European
Insolvency Regulation’, International Arbitration Law Review 12, no. 6 (2009):
125–135.
318 Domitille Baizeau, ‘Arbitration and Insolvency: Issues of Applicable Law’, in

New Developments in International Commercial Arbitration, ed. Christoph Müller

& Antonio Rigozzi (Université de Neuchâtel, Schultheff editions Romands, 2009),
114; Arts 154 and 155(c) SPILA.
319 Article 4.1 EU Insolvency Regulation.
320 Stefan Kröll, ‘Arbitration and Insolvency – Selected Conflict of Laws
Problems’, in Conflict of Laws in International Arbitration, ed. Franco Ferrari &
Stefan Kröll (Germany: Sellier, 2010), 253. For other comments that support the
outcome see Philipp K. Wagner, ‘When International Insolvency Law Meets
International Arbitration’, Disp Res Int'l 3 (2009): 56, 62.
321 Stefan Kröll, ibid., 244. For a thorough exploration of characterisation in this

context, see Kröll, 244–251.

322 Lars Markert, ‘Arbitrating in the Financial Crisis: Insolvency and Public

Policy versus Arbitration and Party Autonomy – Which Law Governs?’,

Contemporary Asia Arbitration Journal 2, no. 2 (2009): 217, 234.
323 Lars Markert, ‘Arbitrating in the Financial Crisis: Insolvency and Public

Policy versus Arbitration and Party Autonomy – Which Law Governs?’, in

Contemporary Asia Arbitration Journal 2, no. 2 (2009): 217, 236–237. While it
may not be appropriate to consider it as a question of capacity, there was some
expert opinion that this was the approach under Polish law. Domitille Baizeau,
‘Arbitration and Insolvency: Issues of Applicable Law’, in New Developments in
International Commercial Arbitration, ed. Christoph Müller & Antonio Rigozzi
(Université de Neuchâtel, Schultheff editions Romands, 2009), 115, fn. 73. A
capacity-based analysis might also limit itself to capacity at the time of
commencement of the proceedings. See, e.g., Pierre Karrer, ‘Views on the
Decision by the Swiss Supreme Court of March 31, 2009, in Re Vivendi et al v.
Deutsche Telekom et al’, ASA Bulletin 28 (2010): 111.
324 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 815.

325 Available at <www.uncitral.org/uncitral/en/uncitral_texts/insolv...>;
UNCITRAL Practice Guide on Cross-Border Insolvency Co-operation,
<www.uncitral.org/uncitral/en/uncitral_texts/insolv...>. The Model Law on
Cross-Border Insolvency has not been adopted by many States to date. Even
those adopting the law may apply a different rule as to applicable law. Philipp K.
Wagner, ‘When International Insolvency Law Meets International Arbitration’,
Disp Res Int'l 3 (2009): 56, 64.
326 Laurent Levy, ‘Insolvency in Arbitration (Swiss Law)’, International
Arbitration Law Review, 8, no. 1 (2005): 32.
327 Stefan Kröll, ‘Arbitration and Insolvency - Selected Conflict of Laws
Problems’, in Conflict of Laws in International Arbitration, ed. Franco Ferrari &
Stefan Kröll (Germany: Sellier, 2010), 217.
328 See, e.g., French Cour de Cassation decision in Société Almira Files v. Pierrel

(Almira Films) 5 February 1991, Rev Arb 37 (1991): 625 with note by Idot cited
in Stefan Kröll, ‘Arbitration and Insolvency – Selected Conflict of Laws
Problems’, in Conflict of Laws in International Arbitration, ed. Franco Ferrari &
Stefan Kröll (Germany: Sellier, 2010), 222.
329 Doug Jones, ‘International Dispute Resolution in the Global Financial Crisis’,

The Arbitrator and Mediator (October 2009), 3,

330 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 814.

331 Jonathan Sutcliffe & James Rogers, ‘Effect of Party Insolvency on Arbitration

Proceedings: Pause for Thought in Testing Times’, Arbitration 76, no. 2 (2010):
283.
332 Amokura Kawharu, ‘Part III Chapter 11: Participation of Non-Governmental

Organisations in Investment Arbitration as Amici Curiae’, in The Backlash

against Investment Arbitration, ed. Michael Waibel et al. (The Hague: Kluwer
Law International, 2010), 282.
333 Loukas Mistelis, ‘Confidentiality and Third Party Participation’, Arbitration

International 21, no. 2, (The Hague: Kluwer Law International, 2005), 223.
334 Ibid., 218.
335 Renato Nazzini, ‘A Principled Approach to Arbitration of Competition Law

Disputes: Competition Authorities as Amici Curiae and the Status of the Their
Decisions in Arbitral Proceedings’, in European Business Law Review Special
Edition – Arbitrating Competition Law Issues 19, no. 1, ed. Gordon Blanke (The
Hague: Kluwer Law International, 2008): 105.
336 Eugenia Levine, ‘Amicus Curiae in International Investment Arbitration: The

Implications of an Increase in Third-Party Participation’, Harv. E.U.L.A.W.P.S.

(2010): 15.
337 If the parties are in agreement, there is no problem.
338 Patrick Dumberry, ‘The Admissibility of Amicus Curiae Briefs in the
Methanex Case: a Precedent Likely to be Followed by other NAFTA Chapter 11
Arbitral Tribunals’, ASA Bulletin 19, no. 1 (2001).
339 Patrick Dumberry, ‘The Admissibility of Amicus Curiae Briefs in the
Methanex Case: A Precedent Likely to Be Followed by other NAFTA Chapter 11
Arbitral Tribunals’, ASA Bulletin 19, no. 1 (2001).
340 Article 17(1) of the UNCITRAL Arbitration Rules 2010 provides as follows:

‘1. Subject to these Rules, the arbitral tribunal may conduct the arbitration in
such manner as it considers appropriate, provided that the parties are treated
with equality and that at an appropriate stage of the proceedings each party is
given a reasonable opportunity of presenting its case. The arbitral tribunal, in
exercising its discretion, shall conduct the proceedings so as to avoid
unnecessary delay and expense and to provide a fair and efficient process for
resolving the parties' dispute.’
341 However, the tribunal did not consider that it had jurisdiction to grant USD

observer status or to make oral submissions. Also, it should be noted that the US
supported the USD petition.
342 Amokura Kawharu, ‘Part III Chapter 11: Participation of Non-Governmental

Organisations in Investment Arbitration as Amici Curiae’, in The Backlash

Against Investment Arbitration, ed. Michael Waibel et al. (The Hague: Kluwer
Law International, 2010): 286, 293.
343 Biwater Gauff (Tanzania) Ltd v. United Republic of Tanzania (ICSID Case No.

ARB/05/22).
344 Nathalie Benasconi-Osterwalder, ‘Chapter 9: Transparency and Amicus
Curiae in ICSID Arbitrations’, in Sustainable Development in World Investment
Law, Global Trade Law Series, vol. 30, ed. Marie-Claire Cordonier Segger et al.
(The Hague: Kluwer Law International 2011): 204.
345 Ibid., 205.
346 Ibid., 201.
347 See Transnational Dispute Management 8, no. 4 (2011) which devoted an

entire issue to the question of third-party funding.

348 Mark Kantor, ‘Third-Party Funding in International Arbitration: An Essay

about New Developments’, ICSID Review – Foreign Investment Law Journal 24,
no. 1 (2009): 44 and S. Khouri, K. Hurford & C. Bowman, ‘Third Party Funding in
International Commercial and Treaty Arbitration – A Panacea or a Plague? A
Discussion of the Risks and Benefits of Third Party Finding’, Transnational
Dispute Management 8, no. 4 (2011): 1.
Part II: The Process of an Arbitration,
Chapter 8: Preliminary, Interim and
Dispositive Determinations

Jeff Waincymer,

8.1. Jurisdictional Challenges

This section deals with jurisdictional challenges in the context of identifying an

arbitrator's procedural approaches to such disputes. There are a range of
potential jurisdictional challenges. These might include claims that there was no
consummated arbitration agreement, that one of the parties was not bound by it,
that the agreement is not in writing, that the subject matter is not arbitrable, that
the claims as brought are outside of the scope of the arbitration agreement, or
that the contract is not to be supported because of fraud or other violations of
transnational public policy. Jurisdictional challenges are generally made by a
respondent but a claimant may also challenge the jurisdictional basis for a
counterclaim or set-off.

The Chartered Institute of Arbitrators Guidelines has identified a list of scenarios

where disputes may be described as jurisdictional. These relate to disputes as to:

(1) Identity of the parties.

(2) Identify of the contracts governed by the arbitration clause.
(3) Formal validity of the agreement to arbitrate.
(4) The effect of a problem with the main contract on the validity of the
arbitration clause (separability).
(5) Whether a particular subject-matter can be arbitrated.
(6) Time limits for commencing proceedings.
(7) Failure of a condition precedent to the right to arbitrate contained perhaps
in an agreement to mediate in advance. page "609"
(8) Whether an arbitrator may apply a law other than the one expressly chosen
by the parties.
(9) Whether a particular remedy can be awarded. (1)

The Guidelines further consider that jurisdictional issues could be divided albeit
uneasily, into three categories, namely whether there was a valid agreement to
arbitrate at all; whether the scope of the agreement covers the dispute; and
whether the arbitrator has certain powers.
While the presence of the power of a tribunal to rule on its jurisdiction under the
competence-competence principle is clear, there are differences in approach as
to the attitude and procedure to be taken. These are discussed in sub-sections
below.

8.1.1. The Timing of Jurisdictional Challenges and Determinations

Where challenges to jurisdiction are brought, the lex arbitri or rules will typically
provide time limits. For example, the English Arbitration Act 1996 requires an
objection to jurisdiction to be taken on or before the first step in the proceedings
to contest the merits. (2) The ICC Rules 2012 imply that a challenge would be
made in the Answer to the Notice. The UNCITRAL Rules 2010 are also permissive
in that regard, (3) but require such notice no later than at the time of the defence.
(4)

Many rule systems will allow subsequent challenges by way of leave. If that is not
the case, a question then arises as to whether a valid jurisdictional challenge is
thereby lost if not brought within the appropriate time period. (5) This mixes a
number of concepts and issues. First, there is a need to consider the intent
behind time limits that do not articulate ramifications on non-compliance, the
implications of intent when such provisions are selected, the role of waiver
within the lex arbitri, arbitral rules or general principles, the need for consent in
any event and the ability of a tribunal to consider jurisdictional questions sua
sponte. In the context of arbitration theory, it would be somewhat troubling to
find arbitral jurisdiction in a case based simply on a mere inadvertent failure to
raise an objection within a specified time period where the objection is
unassailably valid. Even where lex arbitri expressly indicate that principles of
waiver may apply, they rarely articulate the criteria. The better view might then
be that waiver situations should be limited to cases where the relevant party has
behaved inconsistently with the position it puts as part of a challenge. (6) Waiver
may also not apply where fundamental principles of page "610" public policy are
concerned, which may encompass jurisdictional issues. An example is Article 4 of
the UNCITRAL Model Law dealing with waiver. Redfern and Hunter observe that
because the provision deals only with matters from which the parties ‘may
derogate’, they may not waive mandatory provisions of the applicable law. (7) In
some legal systems the lex arbitri will provide an answer to this question. (8) In
other instances there may be a need to consider whether the particular
annulment or enforcement court would apply principles of waiver or estoppel or
hear arguments as to a lack of good faith in terms of the lack of timeliness of the
challenge.

8.1.2. Who Should Make the First and/or Final Determination on Jurisdiction?

From a procedural perspective the first question is who is to make such

determinations. While all systems support competence-competence, they also
support a supervisory jurisdiction of a national court and allow for an indirect
supervisory role for enforcement courts where the award arguably exceeds the
tribunal's jurisdiction.
While different theoretical perspectives vary between advocating anational
arbitration and strong jurisdictional control, most would now agree that
ultimately, jurisdictional questions will inevitably be resolved by the courts,
either through immediate jurisdictional challenges to arbitral decisions, attempts
to have courts rule under provisions such as Article 8 of the Model Law or
applications to block enforcement of awards. At the same time, the principle of
competence-competence is firmly entrenched, allowing an arbitrator to rule on
jurisdictional matters, including the power to make negative findings. In such
circumstances, an important question is how the role of the tribunal is best
integrated with the role of supervisory courts.

If both the arbitrator and some supervisory courts can each consider the
question, the issue is then who ought to have priority and what is the
appropriate timeline for consideration by each. Historically, the norm was to
provide a clear right for the arbitrator to make a jurisdictional determination
and only allow challenges before supervisory courts within strict timeframes and
in relation to positive determinations only. From a policy perspective, this was
thought to best support the independence of arbitrators and the concept of
competence-competence itself. Over time, however, countervailing efficiency
considerations have led some to question this approach. It is typically the case
that the supervisory court in the Seat will have the last word. That is the
approach in Article 16(3) of the UNCITRAL Model Law which allows for
challenges to positive jurisdictional determination before the supervisory court
in the Seat. There is no separate provision for challenges to negative
determinations. Even where the tribunal is given primacy, such as under Article
16 of the Model Law, if a challenge can then be made to a supervisory court, the
latter has ultimate priority. If that is accepted, some have page "611" queried
whether it ought to be preferable to bring the matter immediately to the highest
level adjudicator to avoid a two-step process. In some cases, the situation may be
reversed and a court ruling can be sought at the outset. (9) An example of a rule
system that uses this process is the German Arbitration Act which allows for
court challenge before a tribunal is established. (10) The English Arbitration Act
1996 section 32 also allows challenges to be brought to the court before
consideration by the tribunal. (11) Some jurisdictions formerly sought to provide
‘right of way’ for arbitration. In France, Article 1448 of the New Code of Civil
Procedure indicates that once a matter has been referred to an arbitral tribunal,
French courts must wait until an award is rendered for any review application.
Prior to constitution of the tribunal, court proceedings can only ensue if the
arbitration agreement is manifestly void. Under section 32(2)(b) of the English
Arbitration Act 1996, a tribunal may consent to a party bringing a jurisdictional
application to the court. The parties themselves may so agree under section
32(1) and 32(2)(a). At the other extreme are systems such as the ICSID
Convention which allow no opportunity for challenges in local courts. Instead it
provides an internal annulment process which can only be invoked after a final
award is obtained. (12)

Section 3.14 noted the difference between various institutions as to the role they
may play in determining validity of the arbitration agreement. At one extreme,
institutions such as CIETAC take it upon themselves to make such
determinations. (13) Most other institutions leave it for the tribunal to determine
questions of validity. Others such as the ICC involve a preliminary screening
mechanism but ultimate determinations are made by the tribunal. In ICC
arbitrations, challenges to jurisdiction operate in two stages. If there is a
challenge, the ICC Rules require the ICC Court to first determine if there is a
prima facie existence of an arbitration agreement. (14) Effectively the court,
supported by the Secretariat, will look to see whether there is any possible basis
for the determination of jurisdiction. If that is the case, the decision on
jurisdiction is remitted to the tribunal itself. (15)

8.1.3. Jurisdictional Questions Not Raised by the Parties

As noted above in the context of late challenges to jurisdiction, determining how

to deal with waiver scenarios is in part dependent on the view one takes as to the
page "612" entitlement of a tribunal or indeed an obligation of a tribunal to
consider jurisdictional questions sua sponte. If a tribunal must consider
jurisdiction from the outset, it may be harder to justify a waiver analysis. The
Chartered Institute's Guidelines deal with this in very circumspect language. The
Guidelines suggest that ‘the arbitrator should not normally raise issues on his
own …’ but notes that ‘in some rare public policy situations, it can be argued that
the arbitrator has an active duty not to deal with a case for fear of offending local
norms. Only then, and if he is confident of the point's validity, should he perhaps
raise the jurisdictional objection’. (16) Some would decline jurisdiction in this way
while others would, where appropriate, consider the contract to be illegal. (17)

8.1.4. Bifurcation and Jurisdiction Decisions

Another question is whether a jurisdictional challenge should be decided as a

preliminary matter or whether it is ever appropriate to leave the determination
until a final award, including an award on the merits if jurisdiction is found. In
deciding such timing questions, an arbitrator ought to consider the same issues
of fairness and efficiency that should circumscribe all procedural discretions. In
most cases it would be reasonable to expect such a determination at the outset. If
the conclusion is that there is no jurisdiction, then there is no wasted time and
expense in running full proceedings on the merits. Nor is there an opportunity
for identifying conflicts between that evidentiary base and an evidentiary base
that might then be presented to a court that has residual jurisdiction. The Swiss
arbitral legislation indicates that normally jurisdictional matters should be
considered separately. (18)

However, in some cases it might not be possible to determine a jurisdictional

question absent a consideration of the merits argument. An example might be a
case where a non-signatory is sought to be brought into the arbitration under an
argument that it is contractually bound to the commercial relationship. In those
circumstances it may be more efficient to hear all of the evidence that impacts
both on substantive and jurisdictional rights. There are problems with
bifurcation where jurisdictional challenges are made in bad faith or where delays
in the supervisory legal system flowing from a challenge to a jurisdictional
determination could delay the proceedings. (19)
page "613"

8.1.5. The Role of a Supervisory Court

After considerations of timing and order of review, the next key question in
resolving the relationship between arbitrators and supervisory courts is the
standard of review that a supervisory court will undertake. Some will take the
view that they have an independent obligation to determine jurisdiction. Others
may take the view that they will only interfere with a jurisdictional decision
where they see that this conclusion was not reasonably open to the arbitrator.

A 1995 US Supreme Court decision considered that the standard of review may
vary as to whether the parties agreed to submit the arbitrability of their dispute
to arbitration. (20) The Supreme Court considered that in such circumstances the
tribunal was the primary judge and a cautious standard of review should be
employed by the court. The case concerned an application for joinder of
corporate owners to an arbitration also brought against the company in relation
to debts owed. The owners had not signed the arbitration agreement. The matter
went to the US Supreme Court. It considered that the court's supervisory
function allowed it to consider the scope of the agreement de novo and it was not
required to show deference to the finding of the arbitrator. The appellate court
held that there was no jurisdiction in relation to the owners. Nevertheless in
obiter dicta, the court suggested that in some circumstances the arbitrability
question itself might be submitted to arbitration and that in such cases the
courts must give considerable leeway to the arbitrator's decision as to their
jurisdictional ambit. In other cases the court felt that a full de novo hearing was
appropriate. Redfern and Hunter rightly criticise the distinction on the basis that
arbitral rules which grant competence-competence could be said to always
confer a power to determine arbitrability. (21) William Park criticises ‘shallow
judicial examination’ of jurisdictional decisions by arbitrators. (22) He asserts that
parties' intent should be the basis for determining the allocation of tasks
between judges and arbitrators. He poses the question ‘did (or should) the
parties expect that the particular issue would be decided by a court or by the
arbitral tribunal?’ In some cases, reasonable people may differ as to the way such
a question would be answered.

A supervisory court that is also a court in which a party claims jurisdiction on the
merits must decide the question one way or another as it must make a ruling as
to whether to stay the litigation and force arbitration or not. This will be so
regardless of how deferential it chooses to be in terms of any reasoning already
page "614" promulgated by the tribunal. The situation will vary in those
jurisdictions that allow a direct jurisdictional challenge to be brought to the
supervisory court. Here there is a different question as to the standard to be
applied. While the supervisory court is the first to make a determination, and
hence is not reviewing the reasoning of the tribunal, it still must take a view as to
whether it wishes to look for prima facie jurisdiction, leaving the ultimate
determination to the tribunal or whether it seeks to impose what it believes to be
the best view from the outset. That may depend in part on the wording of the lex
arbitri. From a policy perspective at least, there ought to be consistency in
standards of deference regardless of the order in which challenges are brought.

Regrettably, such consistency is difficult to achieve and is not always present

across national courts. The key recent development in this area is the Dallah
case, in which the UK Supreme Court and the Paris Cour d'appel came to
different conclusions as to whether an ICC tribunal was correct to assume
jurisdiction over the Government of Pakistan, a non-signatory to the arbitration
agreement. (23) The UK Supreme Court held that the tribunal was wrong to
assume jurisdiction. A few months later, the Cour d'appel, applying the same
legal test, upheld the tribunal's finding on jurisdiction. (24)

8.1.6. Jurisdictional Challenges, Procedure and Due Process

Each party must be given a reasonable opportunity to present its case on

jurisdictional challenges. That can be particularly important when the challenge
is as to the extension of an arbitration clause to a third party. A tribunal must
also be sensitive to the fact that a challenge to jurisdiction is often a challenge to
its own authority and should avoid seeing this as a personal attack or giving the
impression that it takes issue with the party making the challenge. (25)

Where a party seeks to challenge jurisdiction, it may be desirable to have an

agreement about the ambit of such challenges, perhaps through the use of terms
of reference to reduce the scope for future challenges beyond the grounds
articulated. (26)

page "615"

A party who attends an arbitral hearing to raise challenges to jurisdiction should

make it clear that it does so under protest to avoid any argument that it has
waived its challenge through participation. Such a protest will also naturally
support its entitlement to seek annulment of the award or challenge
enforcement in the face of an adverse determination.

8.1.7. What Standards Tribunals Should Apply

The next issue is what attitude a tribunal should take to challenges to

jurisdiction. Should there be any presumption either for or against jurisdiction?
A presumption in favour might, for example, suggest that if there is any evidence
that the parties intended arbitration, then a generous interpretation of their
documentary or other agreement should occur to best give effect to their true
intent. A contrary argument is that parties' rights to litigate are significant and
should only be found to have been waived where there is clear evidence to this
effect. Questions of interpretation and applicable law as to an arbitration
agreement were considered in detail in sections 3.2.2–4. It was noted that this
would be particularly relevant where there were jurisdictional challenges as to
the validity of the arbitration agreement. Only some additional comments are
made in this section. In part, some of the key observations are repeated.

Much again depends on whether one adopts a consent or a jurisdictional

paradigm. That may also depend on whether one prefers to look at all
circumstances, including presumptions and inferences from those circumstances
to best identify intent, or whether one adopts a rigorous conflicts methodology
and analyses an agreement under the applicable law and its traditional
presumptions as to interpretation, only looking at surrounding circumstances
when the applicable law directs this to occur. Where a consent approach is
adopted, there is even a division between those who seek to limit themselves to
evidence that allows confident conclusions about subjective intent and those
who are prepared to infer that from an a priori pre-dispute perspective, business
persons can be presumed to have intended efficient solutions to their dispute
resolution.

At times this can be impacted upon by directions in the lex arbitri. For example,
the Swiss PIL Article 178(2) indicates that:

(a)s regards its substance, the arbitration agreement shall be valid if it conforms
either to the law chosen by the parties, or to the law governing the subject
matter of the dispute, in particular the law governing the main contract, or if it
conforms to Swiss law.

Thus the Swiss law seeks to promote maximum validity. To similar effect
Fouchard, Gaillard and Goldman have dealt with approaches to interpretation,
arguing in favour of effete utile but not the more extreme pro-validity approach
of in favorem validitatus. These approaches were discussed in section 3.2.4.

At times an arbitrator will be faced with a defect in the gateway to arbitral

validity. Such a defect arises where there is no clear way to identify a procedural
page "616" framework. In most cases, the gateway to a procedural framework is
through direct or indirect identification of a Seat or lex arbitri. If the parties
identify a Seat and no more, then they subject themselves to the lex arbitri of that
Seat. If the Seat is a recognised venue for international arbitration, the lex arbitri
will provide a sufficient procedural model and will typically support
enforcement under the New York Convention. Another approach is to select a set
of procedural rules, either ad hoc or via selection of institutional support with
that institution's own rules. Here again a minimum framework is typically
guaranteed. In some cases there may be no need to even identify a Seat, given
that the procedural model in a recognised set of rules will again be reasonably
comprehensive. Furthermore, most such rules provide that the tribunal or the
institution can select the Seat. One particular challenge that might arise is in a
situation where a tribunal has a discretion to select a Seat where that selection
may impact upon the validity of the agreement itself. It seems more consistent
with notions of consent to make a selection that gives effect to the arbitration
agreement rather than one which undermines it, unless such a choice would be
wholly removed from any likely intent of the parties.

One author has asserted that where a tribunal exceeds its jurisdiction in relation
to pathological clauses or joinder of non-signatories, it may in some
circumstances be breaching individual rights to alternative tribunals guaranteed
under human rights norms. (27) The argument may be somewhat circular. If the
jurisdictional determination of the arbitral tribunal is the preferable one, then it
is based on a determination of original consent, notwithstanding differences of
view between the parties at the time of the dispute. Furthermore, international
arbitration should be seen as of sufficient stature to be a true alternative to
litigation and hence not a violation of human rights entitlements per se. Human
rights considerations also seem unnecessary where an inappropriate
determination of arbitral jurisdiction is made. In such circumstances, the award
or enforcement might be challenged in any event and a consequential
infringement of entitlement to court jurisdiction is simply one of a number of
problems of an erroneous conclusion.

8.1.8. The Form of Rulings on Jurisdiction

If a tribunal intends to make a separate decision as to jurisdiction, there is a

question as to whether this should be rendered as an award or a mere order or
via some other preliminary procedural determination. If it is an award, it is open
to recognition and enforcement and annulment processes. If it is a mere
procedural order, it is capable of review in due course. At times it may be that
the parties indicate what their preferences are in that regard.

A negative determination on jurisdiction should be made by way of an award so

that it has res judicata effect, and in order to settle tribunal's fees and, if page
"617" appropriate, costs. Where a tribunal makes a negative decision on
jurisdiction, the question arises as to what consequential orders may also be
made. There is a conceptual problem in this regard. If the tribunal has decided
that there is in fact no valid arbitration agreement and/or no properly
constituted tribunal, from where would the power be derived to grant such
consequential orders? (28) From a practical perspective there are many such
orders that a tribunal might wish to make. These include costs in favour of the
successful respondent, orders as to arbitrator fees, interim measures and
confidentiality determinations. (29) If the terms of reference can be seen as a
distinct agreement, that might overcome any conceptual challenges to the
tribunal making such consequential orders. (30)

Where positive decisions are concerned, there is some question as to whether

these should be by way of an award or merely an order. Legal systems vary as to
whether they are willing to respect a tribunal's nomenclature, with most
jurisdictions considering that the essence of the decision is important and not
the title it is given. It is particularly important with institutions such as the ICC
and SIAC which require scrutiny of awards. If a decision is truly considered to be
an award and has not been scrutinised, then it would have no validity. (31)

8.2. Interim Measures

8.2.1. Introduction

At times a party may require temporary steps or measures to be taken by the

tribunal or a national court in order for the arbitral proceeding to function
effectively. These measures are usually known as ‘interim’, ‘provisional’ or
‘conservatory’ measures. There are a number of possible measures of interim
protection that an arbitral tribunal may choose to exercise. These include: anti-
suit injunctions, security for costs, interim payments, sale and holding and
escrow accounts, freezing and injunctive remedies, preserving remedies,
handing over of property or information, inspection, administration of
companies, and orders as to confidentiality. (32)

page "618"

From the applicant's perspective, there may be a commercial imperative for such
relief. For example, if a dispute is in relation to the ownership of a particular
asset about to be sold, the ultimately successful party might get nothing if the
sale proceeds have been dissipated. While the benefit to an applicant may be
obvious in many circumstances, there are significant problems in utilising such
measures within an arbitral forum. These include conceptual questions as to the
relationship to party autonomy, due process and standards of proof and
enforcement.

Urgent applications for interim measures will typically arise contemporaneously

with the dispute. This poses a particular problem with arbitration as there is
then no tribunal before whom such an application can be made. As noted, some
rules allow for pre-arbitral referees for such purposes. Pre-arbitral referee
processes are discussed in section 8.6. In other cases an approach may be to have
an expedited empowerment of a tribunal. In some cases the issue is not really
about interim measures but instead the need for expedited proceedings,
although often these may go hand-in-hand in circumstances where an urgent
result is needed with the status quo maintained in the meantime. Expedited
proceedings are discussed in section 6.4. The following sections look at policy
issues before analysing laws and rules, and procedural and evidentiary issues.

8.2.2. Should Arbitrators Utilise Interim Measures?

Many commentators suggest that arbitration cannot adequately replace litigation

as a fair and efficient mechanism if arbitrators cannot respond to problematic
scenarios through the use of interim measures aimed at preserving the status
quo while a final resolution is pending. Hence much work has been done to
amend institutional rules to establish at times relatively elaborate processes for
considering such measures. Interim measures have been said to play a particular
role in assisting the tribunal ‘preserve a factual or legal situation so as to
safeguard rights the recognition of which is sought from the [tribunal] having
jurisdiction as to the substance of the case’. (33) Despite this traditional definition
of such measures, their functionality can extend beyond that of just maintaining
the status quo or preserving a factual or legal situation as they may also allow
the tribunal to restore a particular situation or order additional new actions. (34)
In some cases an application is effectively an interim order for specific
performance of the underlying contract requirement. In the extreme, there might
be an order sought whose effect is contrary to a purported termination of the
very contract by one of the parties. (35) For example, a page "619" tribunal might
direct that a joint venture continue to operate with profits being paid into an
escrow account and invested as directed pending the outcome of the final
dispute.

Some respected arbitrators are more sanguine about arbitrators engaging in

injunctive relief generally. (36) Some commentators suggest that it is not
appropriate to try and have arbitrators simulate the sanction-backed power of
domestic judges as this could never effectively occur. Some critics are most
concerned with the possibility of ex parte relief. This is discussed separately
below. Others are concerned with the uncertainty as to the nature of the criteria.
Others are concerned with issues of enforceability. Some may be concerned
more conceptually with the relevance of such measures to a consent-based
dispute resolution system. Obviously some critics may combine a number of the
above concerns. While the debate about the use of interim measures in
arbitration is contentious, much may depend on the elements of any proposed
system and other circumstances peculiar to any individual case. For example, an
analysis of the advantages and disadvantages of interim measure powers for
arbitrators must consider the alternative. This will depend on the particular
jurisdiction where measures would otherwise be sought and the effectiveness of
that jurisdiction's laws. The less there are suitable court alternatives, the more
there can be a policy justification for tribunal measures. Even if there are
adequate court processes, it can still be desirable to have one adjudicator deal
with all issues and not force parties to go to State courts for measures of
protection. A countervailing factor is that some would suggest that it is better to
have a different person dealing with the interim measure application than the
case itself as attitudes from the first may prejudice or appear to prejudice the
adjudicator in the latter, particularly where some preliminary view as to the
merits or bona fides is a relevant criterion. For example, in some cases, a party
may simply wish to dispose of an asset based on a bona fide view as to its
entitlement, albeit where that entitlement is disputed. In other cases a party
seeking to avoid payment where it is clearly due, may seek to dissipate assets or
remove them from the jurisdictional ambit of any enforcement authority.

The latter point suggests that much may depend on the criteria used and the
enforceability or otherwise of such measures. Where criteria are concerned,
there is a policy tension. To be meaningful on a practical level it may be that the
powers afforded to arbitrators need to be particularly broad and it should not be
too onerous to seek measures where fairness would suggest that this occur.
However, the more far-reaching such interim powers, the more it could be
argued to go against the consent basis of arbitration and lead to effective
prejudgment on the merits without parties having had the fullest opportunity of
presenting their respective cases. While much depends on the particular options
being considered, interim measures powers will always call for some rule of
reason where a tribunal will consider certain allegations and determine whether
to exercise a discretion in favour of or against granting the measures. It is
particularly difficult to employ such powers in a page "620" way that would be
beyond debate. The difficulty is that one cannot know at the outset which party
is likely to succeed and hence exactly what the fairness and efficiency
ramifications are at the stage of the interim measures. Is it fair to make binding,
albeit temporary orders over assets and actions without a review of the claim on
the merits? What ramifications should flow if the interim measure ultimately
proves to be unnecessary, either because the person proposing it loses the case
or does not obtain a remedy consistent with the conservatory measure? In
common law systems, adjudicators will in part consider the merits in
ascertaining the balance of convenience between the parties in terms of the
interim measure sought. In some legal systems, the applicant for the measure
must also give an undertaking as to damages if the measure is ultimately shown
to be inappropriate. This can raise further problems when considered in an
arbitral framework, particularly if a party does not have the funds to provide any
meaningful security. Should such a party be denied the remedy for that reason
alone? Conversely, do their effective due process rights entitle them to such
remedies notwithstanding an inability to provide security? Such questions do not
necessarily militate against the use of interim measures but show that whatever
design is employed, there are tenable arguments that this impacts upon due
process.

Another concern is that interim measures applications can be used in a manner

that is counter to their intended purpose. (37) A party may initiate a claim for an
interim measure from the arbitral tribunal as a tactic in circumstances where the
granting of such a device will operate to apply additional, unnecessary and
undue pressure on the opposing party. Another contentious aspect is where
interim measures are sought against governments in investment disputes. Such
measures can have a significant impact on governmental behaviour and hence
the rights of other citizens in the host State. (38) A further complication is that in
extreme cases, to be effective, the initial application for an interim measure may
need to be taken on an ex parte basis. This occurs in cases where proper notice
would allow a disreputable party to speed up its removal of the assets in issue.
Obviously any ex parte measure in arbitration is even more contentious again. A
further practical concern is that given the need for some consent to any interim
measure regime whether ex parte or not, the more a contractual party would be
likely to seek to abuse and frustrate the process, the less likely they will agree on
rules and processes allowing for interim measures. Hence litigation systems are
to that extent more suitable, at least insofar as they require no particular consent
at the outset from such strategically minded individuals.

The debate can never be fully resolved, as the arguments in favour tend to
concentrate on a very real potential mischief that implies the need for a solution,
page "621" while the arguments against tend to concentrate on problems for the
arbitral process where it seeks to provide any form of solution. Each contention
is valid but not reconcilable. In any event, most arbitral laws and rules allow for
the use of such measures and differ primarily in relation to whether they allow
for ex parte applications, whether they allow for concurrent applications to a
court and whether they seek to elaborate on criteria for the application of such
measures.

8.2.3. Pre-arbitral Interim Remedies

Unlike a court system, a tribunal is, for obvious reasons, not able to apply
provisional remedies prior to appointment. This is a particular problem for
arbitration as concerns as to treatment of assets or evidence typically arise
immediately upon a dispute arising. In such circumstances, resort may be made
to the courts. Some institutions have also introduced a number of specialised
mechanisms in order to deal with the issue of pre-arbitral interim remedies.
These must also be consent based, either directly or indirectly. This is discussed
in section 8.6 Others also allow for expedited tribunal appointment to advance
the time when a tribunal could consider such applications. (39)

8.2.4. The Jurisdictional Basis of Interim Measures Powers: Consent, Lex

Arbitri and Rules

Because arbitration is fundamentally consent based, the parties could expressly

agree to empower a tribunal to entertain applications for interim measures. In so
doing they could establish criteria and the parameters of circumstances where
this may apply. While that is possible, it is unlikely to be undertaken in the
negotiations traversing the commercial transaction. More typically, consent to
interim measures is indirectly given through the selection of arbitral rules
expressly allowing for such powers and/or a lex arbitri that does so. Where laws
and rules are concerned, there is also the question of whether general
discretionary powers are broad enough to encompass interim measures and
whether a power to entertain such applications could be said to be inherent.
Each of these questions is considered in turn.

There is no express basis for arbitral tribunals to exercise a power to grant

provisional or interim measures contained within any of the major arbitration-
based conventions, such as the New York Convention. (40) This represented the
view at the time of creation since many jurisdictions limited, or exclusively gave
the power to order interim measures to their courts. Nevertheless, the majority
of institutional arbitration rules, and the UNCITRAL Arbitration Rules today,
page "622" instil the arbitral tribunal with the power to grant interim measures.
(41) This principle is mirrored in many arbitration laws around the world

including the UNCITRAL Model Law, (42) although some countries still do not
allow tribunals to order interim measures. (43) Key variations are as to the point
in time when an interim measure may be sought, the time it takes for such an
application to be resolved, the criteria or otherwise articulated and the form of
relief that may be granted. There is also the question as to whether ex parte
relief may be sought.

The UNCITRAL Model Law now has one of the most detailed sets of provisions
and is an appropriate starting place for analysis. Article 26 of the UNCITRAL
Rules 2010 largely follows this model save as to ex parte applications. Article
17(2) of the Model Law indicates that ‘(a)n interim measure is any temporary
measure, whether in the form of an award or in another form …’ Article 17(2) of
the Model Law provides that orders may be made as to a party to:

(a) maintain or restore the status quo pending determination of the dispute;
(b) take action that would prevent, or refrain from taking action that is likely to
cause, current or imminent harm or prejudice to the arbitral process itself;
(c) provide a means of preserving assets out of which a subsequent award may
be satisfied; or
(d) preserve evidence that may be relevant and material to the resolution of the
dispute.'

The UNCITRAL Model law, and now the UNCITRAL Arbitration Rules, state that
the granting of interim measures is only possible by the arbitral tribunal when
certain criteria are met. First, the party requiring the measure must demonstrate
that a harm, which is not reparable by an award of damages, will be dealt to it
should the measure not be granted. Additionally, this harm must not be
substantially outweighed by the harm the opposing party will be exposed to in
the event of the measure being fixed. (44) A further requirement that the tribunal
is bound to take into account is that there is a reasonable opportunity that the
party claiming the interim relief will win the substantial claim. (45) These criteria
are considered further in section 8.2.5 below. UNCITRAL Model Law Article
17H(1) requires the party granted the measure to inform the court of any
modification, allows the court to page "623" request the party to provide security
and sets out grounds for refusing recognition or enforcement. The provision
specifically indicates that the court is not to review the substance of the interim
measure determination. (46)

The laws and rules also need to be looked at to see to what extent applications
can be made to the courts. Most lex arbitri leave it for the parties to be able to opt
out of any interim measure provisions. (47) Others allow for separate applications
to a court. Some rules allow for concurrent applications to a court. (48) Some rules
make it clear that a court application is not a waiver of arbitration rights. Court
assistance is discussed in section 8.2.13 below. Generally speaking a tribunal
cannot itself empower enforcement of any interim measure. Nor can a tribunal
generally impose penalties for failure to comply, although there are exceptions.
(49) Some lex arbitri allow a tribunal to seek judicial assistance where provisional

and protective measures have not been voluntarily complied with. (50) Adverse
inferences would not seem appropriate except in cases where there is no other
logical reason for failure to comply. An example where an adverse inference
would be appropriate is where an interim measure sought preservation of
evidence and the order was not complied with. Even then, in some
circumstances, a party might be able to argue that it had to dispose of the
evidence as part of the ordinary course of business, for example, where the
evidence concerned was the subject goods. Even if the party refusing to comply
was wrong in law to do so, where the notion of an adverse inference is
concerned, the more it has commercial arguments in support of a failure to
respond, the less one can conclude legitimately that the failure shows the
likelihood that facts would be adverse.

The Model Law does not seek to empower tribunals or supervisory courts to take
interim measures against third parties not subject to the arbitration agreement.
There is still the problem of potentially adverse impact on third parties where
the measure only operates between the parties to the arbitration. An example
would be an injunction against sale of intellectual property to a third party. At
the very least this would be a factor in any discretionary determination. More
problematic would be cases where a party seeks to enjoin another from
following separate rights that have broader societal impact such as applications
for winding up or as to bankruptcy status.

Some rule systems merely speak of interim, intermediary or partial awards,

which could be interpreted broadly enough to encompass such interim
measures, although that itself is contentious. Chapter 16 argues that the notion
of an interim award is ambiguous and is not an ideal expression to use in
articulating whether page "624" there is an entitlement to interim measures or
not. If the intent is to have a partial award, being final as to that part as all
awards are required to be, then it is inappropriate to use the phrase partial
award for interim measures. Where the latter are concerned, there is a separate
question as to whether they can be rendered as an award and made enforceable
for that reason. This is discussed further below.

In some cases where the rules and laws are silent on the issue of interim
measures, the tribunal might assume those powers based on an implied principle
of necessity. (51) That view is impacted upon by the view one has about the
essential nature of arbitration and arbitrators. The more one sees an arbitrator
being akin to a judge, the more such an inherent power may seem tenable. The
logic would also run parallel to the requirements of an arbitral tribunal to
protect the parties essential rights and decide the dispute in an effective manner.

8.2.5. Criteria Underlying Interim Measures

Before looking at the criteria used in various laws and rules, it is first
appropriate to address an aspect of the perennial policy debate between
flexibility and certainty as this is the key reason why the articulation of
principles varies significantly. Some would argue that there is a need for clear
criteria before such measures could be imposed. Otherwise too much will be left
to the individual preferences of arbitrators with little certainty and consistency.
Furthermore, the more the criteria are clearly expressed, the more their
application can truly be linked to party consent. Conversely, others argue that it
is impossible to objectively clarify or rank the many and potentially conflicting
factors and it should best be left to general norms allowing for such measures
when it would be seen as reasonable in all of the circumstances. As always,
neither view is inherently correct. This is just another example of the problems
of optimising the conflicting norms of flexibility and certainty.

Whether one seeks to opt for clear criteria or not depends on one's general view
about legal drafting and regulation and also on whether there can be said to be
harmonised norms on these factors in differing legal systems, although it is
possible to still argue both ways in either event. The more there are identifiable
consistent norms, the more it makes sense to clearly articulate these in the rules
to ensure conformity with uniform standards. Even where there is a significant
difference between legal families, it may still be desirable to find common
features to articulate and consider whether best practice would opt for one
model over another. If that is not the case, the inherent uncertainty of
unconstrained discretions remains, particularly as it has been suggested that the
utilisation of these interim devices is highly dependent on the subjective
thoughts of arbitrators as influenced page "625" by their own domestic systems
of law. (52) Domestic predilections can also impact on choice of arbitrator where
the provision of, or barring of an interim measure is crucial.

Typically domestic requirements mirror those stated in the UNCITRAL Model

law and UNCITRAL Arbitration Rules in any event. The UNCITRAL Rules now
mirror the elaborate criteria in the UNCITRAL Model Law. The Commentary on
the UNCITRAL Rules and the Secretariat Note draws attention to the strengths of
the merits of the applicant's claim, the urgency of the need for a provisional
remedy, the practical burdens that may result from granting the remedy and
reasonableness and proportionality. Where the UNCITRAL regime is concerned,
the harm to the applicant needs to be ‘not adequately reparable by an award of
damages (being) likely to result if the measure is not ordered’. (53) Such harm
must also substantially outweigh the harm likely to result to the party against
whom the measure is directed if the measure is granted. These principles apply
to measures seeking to maintain or restore the status quo, prevent action
negatively affecting the arbitral process and preserving assets. The tribunal has
discretion not to apply the balancing test where the application is simply in
relation to preservation of evidence. (54) How one identifies when harm is
irreparable depends in part on one's views about remedies generally. For legal
systems that believe damages are adequate compensation, there is less
likelihood that harm would be seen as irreparable. Conversely, systems that
support specific performance might see damages as inadequate in such
circumstances and be more inclined towards interim measures that preserve the
status quo.

As to the merits, there must be a ‘reasonable possibility’ that the requesting

party will succeed. That is a threshold requirement and otherwise the degree of
possibility should not affect the discretion. (55) Kaj Hodér considers that
identification of the possibility of success requires the party to establish a prima
facie case. (56) Abascal has argued that it is impossible to articulate a standard in
terms of the merits that is ‘anything but a practical, or commonsense, assessment
of the possibility of success’. (57) These questions of criteria and their relevant
standards of proof are considered further in section 8.2.7 below dealing with
procedural and evidentiary issues. When dealing with the possible merits, there
is the added page "626" problem in making such a determination without in any
way prejudging the final determination or being perceived to do so by either
party. While some suggest that this is a particular and even insurmountable
problem with the use of interim measures in arbitration, this is not unassailably
so and there are other situations where some inevitable preliminary
consideration of the merits naturally applies. For example, any arbitrator taking
a proactive approach in designating matters of importance or ruling on
materiality in contested document production applications is to some degree
taking a preliminary view on the merits and on the means to best achieve a fair
and reasonable outcome. (58)

Given that the tribunal's jurisdiction is related to particular disputes, the interim
measure power has to have a sufficient link. Article 17 of the Model Law limits
measures to ‘the subject-matter of the dispute’. Importantly, attention must also
be given to the linking words, originally being ‘in respect of such subject matter
as per former Article 26(1) of the UNCITRAL Arbitration Rules. Based on the
wording, some sought to argue that the measure had to be in respect of
something that could justifiably be described as the subject matter, which was
contrary to the intent behind the provision. (59)

The Model Law allows a tribunal a discretion to request provision of security as a

condition of the grant of the measure. The Model Law provisions use differing
language between interim measures and preliminary orders discussed in the
following section in relation to ex parte applications. Where interim measures
are concerned, it simply indicates that a tribunal may require appropriate
security. In the case of preliminary orders it provides that these shall be required
unless the tribunal considers it inappropriate or unnecessary to do so. (60)

While most institutional rules now allow for interim measures, they vary in
approach. A number do not see the value in trying to articulate criteria. Many
rules simply give the tribunal power to grant any interim measures it deems
appropriate. (61) Article 28 of the ACICA Rules allows for an order where
irreparable harm is ‘likely to result if the measure is not ordered’. The harm must
‘substantially’ outweigh the harm to the party affected if the measure is granted
and there is ‘a reasonable possibility’ of success on the merits. (62) The ICDR Rules
allow for relief when a decision-maker deems it ‘necessary’. (63) Some argue that
elaborate provisions should not be contained in rules as these are quite different
to lex arbitri. In the latter circumstance, a country seeking to adopt the
UNCITRAL Model Law page "627" with its elaborate provisions will simply
decide whether to include the interim measures provisions or not. Where ad hoc
rules such as the UNCITRAL Rules are concerned, parties may not turn their
mind to this issue and may be surprised at the content of elaborate provisions.

8.2.6. Ex Parte Applications

A particularly challenging aspect of interim measures in the field of arbitration is

whether they can be granted on an ex parte basis. This is generally permissible in
domestic legal systems where there is a demonstrable risk that an informed
respondent might dissipate the subject matter of the application prior to any
court order. While dishonest people could be expected to behave the same way
when faced with an arbitral forum, the key problem is that of consent and the
mandatory obligation to give each party the opportunity to fully present its case.
The latter presumes adequate notice as a gateway requirement. Ex parte
applications would offend against this principle. It is important to remember that
the policy issue in relation to such measures is concerned only with cases where
notice of an application would itself frustrate the outcome. Hence it concerns
situations where there is a concern that the target of the application would
behave inappropriately where proper notice is provided. While such situations
can be envisaged, the problems of burdens and standards of proof are significant.
There would not be a problem with consent where lex arbitri or institutional
rules selected expressly provide for such powers but due process concerns
remain.

Hans Van Houtte has provided ten key arguments against incorporation of ex
parte interim measure provisions. He suggests that it was inappropriate to utilise
the Model Law which could in turn impact on its utility; such measures are
unlikely to be needed as orders would be ineffective, they are incompatible with
the consensual nature of arbitration, they are incompatible with sufficient
respect for the right to a defence, they would be difficult to enforce, such
determinations would lead to prejudiced arbitrators or the appearance of
prejudice, they would be particularly problematic in the context of party-
appointed arbitrators, in many cases they would be irreversible in commercial
effect, there would be problems in terms of full and frank disclosure absent
appropriate international bar ethics and tribunal control and they could impact
on arbitrator liability or immunity. (64) Nael Bunni adds further reasons against,
namely, that a court is better able to control such an application, ensure proper
safeguards and sanction abuse and is best able to be available on short notice. (65)
Another key concern behind the effectiveness of page "628" arbitration-based ex
parte measures is the problem with enforceability where one party has not been
properly heard. (66) Such orders would not meet the requirements under the New
York Convention in any event given the requirement for notice and an adequate
opportunity to present a case as per Article V(1)(b). Derains makes the point that
an ex parte order by a tribunal cannot be immediately enforceable without an ex
parte order of a relevant court. (67) He makes the further point that if problems at
the court level are the justification for affording tribunals the power of making ex
parte orders, those very same court jurisdictions would have the last word in any
event. (68) Gary Born suggests that ex parte relief ‘virtually never makes any sense
or accomplishes any serious purpose’ given the lack of effective coercive powers.
(69) Ex parte procedures may also make little sense in the absence of pre-tribunal

relief given that the most serious examples of inappropriate behaviour are likely
to be found to occur contemporaneously with the dispute itself. (70)

The counter argument often opposed to these concerns is that due to the nature
of interim measures, they can be constantly reviewed and updated by the
tribunal, and therefore the tribunal may remove or alter the device at its
discretion. In such cases, the obstacles preventing a party from being heard can
be removed as the party can simply approach the tribunal and assert their case
in opposition to the interim measure. (71) The ex parte measure will also be of the
shortest possible duration, seeking to preserve the status quo while an
opportunity is provided for a more fulsome argument. Policy issues also depend
on whether there is a suitable alternative court forum. In some cases there may
be no court with a suitable power. In others, such a court may simply not wish to
grant orders in favour of foreign parties or in relation to foreign arbitrations.
There are also problems in terms of confidentiality in going to a court. (72) The
strongest argument in favour is simply that the commercial problem exists but as
with the general policy discussion of interim measures, there is no inherently
correct trade-off as proponents concentrate more on the mischief while critics
concentrate more on the admittedly numerous theoretical and practical
problems with implementation. At the very least, the criticisms explain why
some laws and arbitral rules do not allow for ex parte measures.
page "629"

Most arbitral rules do not permit ex parte applications. (73) In addition, some
arbitration laws expressly forbid a tribunal from granting such measures. (74) The
UNCITRAL Model Law provisions differentiate between interim measures
generally and preliminary orders, only the latter may operate via ex parte
applications and can only be provided on a very short basis. (75) This ‘preliminary
order’ may be granted on an ex parte basis by the arbitral tribunal within a
twenty-day time limit. (76) An application can be made where there is a risk that
the other party will frustrate the purpose of the measure if notice is provided. (77)
No indication is given as to the evidentiary standards required. Article 17C(5)
states that whilst binding on the parties, the order is not enforceable by a court.
The revised UNCITRAL Rules 2010 chose not to incorporate such a provision
after some significant debate.

Where such applications are allowed, particular care is needed on the part of the
tribunal. Any ex parte processes can provide an opportunity for inappropriate
submissions that cannot readily be challenged on an immediate basis. A further
problem is that considering merits on an ex parte basis is only considering
submissions of one party alone. (78) Even if carried out impeccably, an ex parte
application creates perceptions of undesirable behaviour in the minds of the
non-appearing party. A tribunal can deal with this to some degree by ensuring
that a proper record is kept and that the representations are communicated to
the other party at the appropriate time. That is naturally required under ex parte
rules.

8.2.7. Procedural and Evidentiary Issues in Considering Interim Measure

Requests

There are specific procedural and evidentiary requirements that need to be met
in order for an arbitral tribunal to grant an interim measure. Determining
comparable harm and damage, identifying damage that is not adequately
compensateable and assessing reasonable possibility of successful claims, all
require appropriate procedural and evidentiary bases. Because there is so much
discretion as to timing, access to evidence, standard of proof and the relevance of
inferences both positive and negative, the answers on these questions will have a
significant impact on the way the rules are applied from time to time.

Some procedural and evidentiary requirements are expressly included in the

relevant laws and rules. Others are developed through general discretionary
powers in the context of due process, fairness and efficiency. The three key
procedural page "630" requirements are said to be as follows: a party must
request the relief; there must be prima facie jurisdiction over the parties; and the
tribunal must always ensure that fairness between the parties (their right to be
heard) is maintained. (79) As to a request, the majority of arbitration rules require
a party to initiate a request for interim relief before the tribunal can consider
granting such relief. (80) However, the ICSID Rules are somewhat unique in that
they afford the tribunal the ability to recommend on its own initiative. (81)
There is some suggestion that in order to be able to rule on an interim issue and
therefore issue an interim measure, the arbitration tribunal needs to first
determine that it has jurisdiction, (82) or at least make a prima facie
determination to that effect. (83) That can be complicated by whether the parties
have agreed to some pre-arbitral referee power that itself allows for interim
measures. This is discussed separately in section 8.6. In more general
circumstances, despite this contention, arbitral tribunals have at times ordered
provisional measures where there is an unresolved issue on a jurisdictional
basis. (84) This seems inevitable as circumstances must arise where a status quo
needs to be maintained while a dispute is being resolved and it is as likely as not
that a respondent party seeking to subvert the arbitral process will, amongst
other things, take any possible jurisdictional challenges whether meritorious or
not. While a tribunal or pre-arbitral referee must therefore be able to consider an
interim measure application before jurisdiction is determined, the fact that the
claimant must show an ability to succeed on the merits makes jurisdictional
merits an essential part of that analysis. Furthermore, if there is a ready ability to
make the jurisdictional determination concurrent with the interim measure
application, there seems very good reason to do so. This scenario is even more
problematic when the interim measure is for a stay of parallel court proceedings,
given that the interim measure is in effect asking the tribunal to view itself as
having jurisdiction. Of course a tribunal does not need to make a final decision in
that event but simply needs to consider the order of priority of different
adjudicatory bodies where jurisdictional challenges are brought. (85) If, however,
the page "631" tribunal decides that it lacks jurisdiction, then any provisional
measures previously ordered by it will cease to have any logical effect. (86)

The final procedural requirement is based on the overarching principle of

natural justice in arbitration, that is, each party has the right to put forth their
case for deliberation. More problematic is a consideration of the evidentiary
standards needed to meet the test for the award of interim measures. This is also
impacted upon by attitudes to disclosure. The party claiming the requirement of
an interim measure holds the evidentiary burden of proving a number of key
aspects. (87) The initial prerequisite for a tribunal under most rule systems
considering the grant of an interim measure is that it must be at least
‘reasonable’ or even ‘necessary’ (or a term of similar application). The party
must then proceed to show an urgent need due to impending irreparable harm,
something akin to a prima facie successful or at least tenable case on the merits,
and finally, a balancing of interests between the parties. (88) The wording of each
rule system must be looked at individually as there are variances between those
that require ‘substantial’ or disproportionate harm, those that refer to concepts
which themselves connote some probabilistic element such as ‘may’ or ‘likely’ or
‘risk’ where ex parte orders are concerned. As is inevitably the case with such
criteria, rules give inadequate guidance as to the standards of proof required in
each of these circumstances. This is made doubly problematic when there are no
clear rules as to the way each party is to present conflicting evidence in relation
to interim measures. This is discussed in the following section.

There may also be greater problems with imbalance between the parties where
interim measures are concerned given that a claimant has had a lot of time to
prepare before bringing a case. It is one thing to impose tight timeframes on a
respondent to prepare defences and counterclaims but may often be more
problematic if it must urgently respond to an interim measure request the
outcome of which will have significant impact upon commercial viability in many
cases.

8.2.8. Document Production, Hearings and Witnesses to Assist Interim

Measure Applications

The broad discretion to control procedure extends to any interim measure

applications conducted during, or before, the arbitration. (89) The tribunal can
determine a page "632" number of key procedural issues such as whether there
will be a hearing, and/or whether there will be written submissions, and/or
whether to hear witnesses including experts on the interim measure topic. (90)

Where final decisions on the ultimate merits are concerned, there is some debate
as to whether each disputant is required to naturally disclose adverse material.
(91) For interim measures and preliminary orders, there is a greater justification

for this to occur in order to ensure proper disclosure in all circumstances. This is
particularly so with preliminary orders made on an ex parte basis. Full
disclosure is called for under the Model Law. (92) In addition, the tribunal may
require a party to promptly disclose any material change in circumstances on the
basis of which the measure was requested or granted. (93) Even that is a
conservative formulation, leaving it to the tribunal to make such a direction
rather than imposing it as an ongoing obligation as a matter of course.

One difficulty with disclosure requirements is that bar ethics would vary
between different countries and there is no international standard applying to
arbitration. Hence merely stating strong duties of disclosure at the time of
application and in relation to changed circumstances cannot be presumed to be
effective measures in all circumstances.

8.2.9. Discretionary Considerations

Discretions must always be exercised on a case-by-case basis but some examples

are illustrative. In some cases preservation of evidence may be problematic
where it relates to the subject matter of the dispute. An example might be
disputed quality of perishable goods. In such circumstances a tribunal may
prefer to order an expedited examination and not simply consider a freezing
order. (94) While the typical case is where there is some alleged misuse of assets
or evidence that is said to be imminent, in some cases at least a tribunal might
wish to impose an interim measure as a condition for exercising another
discretion such as an application for extension of timelines. In some cases a
tribunal might wish to fashion a set of procedures that will be in everyone's
interests regardless of the outcome of the dispute. This might lead to orders as to
conduct of a joint venture business or research and development of contested
intellectual property. A tribunal might also need to consider an application for an
interim measure that may be motivated more by a wish to have practically
enforceable final awards. An example would be a request that profits from a joint
venture be deposited in an escrow account under page "633" the control of the
tribunal, the profits being paid over to the party who is ultimately held to be
justified in continuing the business after termination.

Attitudes will also vary depending on other aspects of the case. Where
investment arbitration is concerned, key issues would be whether the State is
engaging in activities that may seek to undermine the arbitral process and the
degree to which pecuniary awards are likely to be honoured in any event. An
interim measure should certainly not be imposed if it would effectively conclude
the case on the merits. (95) There may be a problem where differing tribunals, all
with valid jurisdiction, purport to impose interim measures about the same
essential transaction where they would inevitably conflict. An example may be a
construction dispute with a number of parties engaged through separate
contracts with discrete dispute settlement provisions. There may also be
confidentiality issues in considering the merits of each application. If a tribunal is
faced with a court order of an interim measure that would significantly affect the
merits of the case before it, it will need to consider what interim measure powers
it has itself. (96) A third possibility is that a national court purports to impose an
interim measure that interferes with the way the tribunal would wish the
proceedings to run.

8.2.10. Timing and Form of Remedies

As noted, there are a number of different forms that interim measures can take.
Traditionally, many have special requirements that need to be met in order for a
tribunal to issue the interim relief.

8.2.10.1. Preservation and Inspection Remedies

There are fundamental requirements that the tribunal will take into account
when determining whether to grant an injunctive remedy for the preservation or
inspection of property. Due to the nature of the measure, the tribunal may assess
whether the property in question is actually relevant to the case in dispute.
Tribunals might direct that goods that are central to the dispute not be disposed
of. Where goods may need to be disposed of in the ordinary course of business,
perhaps because they are perishable, a tribunal might order inspection, might
order proceeds kept in a special account or require samples to be preserved, if
possible, for later adjudication. In a construction dispute there may be a
particular need for an urgent inspection of a stage of the works if there is to be
further work that would make inspection problematic at a later stage. In some
cases there may be a need to halt such further page "634" work if the dispute
revolves around whether some foundational aspect was improperly completed
and needs to be redone or revisited.

8.2.10.2. Abstention Decisions

Tribunals might direct that parties abstain from using intellectual property or
manufacturing and selling certain goods. A tribunal might also direct that claims
not be made against letters of credit, insurance policies or guarantees.
8.2.10.3. Positive Orders

In some cases an interim measure application will seek to require a party to do

certain things. This may include continuing with the contract while the dispute is
being resolved, providing access to intellectual property or Internet resources,
allocating profits or revenue to an escrow account, allowing access to experts or
inspection or testing, providing signatures for joint account payments where this
is required for ongoing business purposes and commencing actions where
statute of limitations may shortly expire. (97) A tribunal might also order the
appointment of an expert. In a construction scenario there may also be an order
for payment of moneys directed by an engineer or a dispute adjudication board.

8.2.10.4. Confidentiality Orders

Due to the nature of confidentiality breaches, damages are rarely sufficient to

cover the relevant loss that can arise from such a breach. Tribunals will often
order parties to adhere to strict confidentiality requirements about the arbitral
process. (98) This will depend in large part on the principles of confidentiality
otherwise adhering to the particular arbitration. This may vary depending on the
Seat and the rules selected. In some cases, confidentiality is a substantive
entitlement under the contract and an interim measure is simply seeking to
maintain the status quo of such an entitlement until a final determination is
made.

8.2.10.5. Anti-suit Injunctions

Anti-suit injunctions are devices designed to restrict a party from pursuing the
dispute outside of any contractual agreement the parties have made. Essentially
a tribunal will forbid a party from engaging domestic courts or other tribunals.
This is discussed separately in section 8.4.

page "635"

8.2.10.6. Security for Costs

The most common form of interim measure sought is that of security for costs. A
tribunal may order that a party provide security for costs in certain
circumstances. This is discussed separately in section 8.3. One reason why it
should be treated separately is that to some, it is antithetical to arbitral consent.
If a party has agreed to allow for arbitration in an arbitration clause, why can it
then turn around and say that the bringing of a claim should be conditional on
the provision of security? A second reason is that arbitral laws and rules typically
have express provisions dealing with advances for fees and costs of the
arbitration which has been argued to be the proper exposition of parties' intent
on that issue. (99)

8.2.11. The Form of the Measure and Enforceability

As noted at the outset, one reason why some oppose the use of interim measures
by tribunals is the problem with enforceability. That in turn can depend on
whether interim measures can be seen as awards and enforceable as such or
whether there is some other provision allowing for enforcement. Interim
measure rules ought to expressly indicate how and on what basis they may be
enforced rather than leaving it to the broader debate as to whether they can
constitute an award, enforceable under the New York Convention. Such
articulation is rarely the case, although it would not be easy to articulate a
regime as laws and rules in one place cannot guarantee enforcement outcomes in
a different jurisdiction and under an enforcement convention which itself has
preconditions. Some countries will not enforce interim measures regardless of
their form. (100)

There are three key forms of provisional measures that the tribunal can provide.
(101) The first is the use of an order or direction as opposed to an award. A

provisional order can traditionally be completed in a much faster time scale. This
provides for a lesser requirement of formality and limited scrutiny by an arbitral
institution where that is required for awards. (102) However, it typically lacks the
extra enforceability of an award. (103) The second approach is to make the
direction as an award, in the hope that this aids enforceability. There are
advantages and disadvantages. Some would still argue that an interim measure is
never enforceable as an award as it does not have the requisite finality. Simply
calling it an award will page "636" not make it such and will not guarantee that
an enforcement court will treat it accordingly. The important question is
whether the determination has sufficient finality to constitute an award.
Different views are possible. Some would argue that by its very nature an interim
measure is only a holding pattern and can never have finality as to an issue in
dispute. Others have taken the view that such measures are final in the context of
the distinct objective that they have. (104) Neither view is inherently illogical. The
JAMS Rules appear to aim at increasing enforceability by deeming awards as to
interim relief as complying with the New York Convention. (105) That should only
be effective if such a determination could constitute a waiver of a right that may
be so waived. Neither institutions nor private individuals cannot readily deem a
different meaning to an international convention to that which was intended.

Whether a determination may be enforceable can also depend on the subject

matter in the instant case. In some circumstances at least, an interim measure is
in fact a final ruling on a contractual issue. An example would be a ruling that the
parties must continue to perform their contractual obligations pending the
determination of the dispute. That would not simply be an interim measure in
aid of the analysis of the contract's rights and obligations but instead, is a
determination of a core contractual requirement. (106)

Deciding whether to make a determination in the form of an award or not can

impact on other timing issues such as a requirement of scrutiny before an
institution such as the ICC or SIAC. (107) Another problem is that awards lack the
ability to be easily reversed. (108) In some cases it is possible with judicious
drafting to convert what would otherwise be an interim measure issue into a
true award issue. For example, a licensor might not wish a licensee to use
technology while it seeks to adjudicate any dispute. The licensor could make this
an express contractual obligation so that if breached, a binding award is possible.
That itself may not be enough as an award could not be granted until such time
as breach is determined. To overcome this, the contract could say that
immediately upon a claim being lodged the licensee must not use the technology
and must return necessary materials. A decision on this is still a decision on the
merits, but one which could hopefully be reached in expedited form if not
voluntarily complied with.

The final method for a tribunal is to combine the two approaches and make a
provisional order and then enshrine the details for the measure in an interim
and reasoned award. (109) The logic behind this method is that it preserves the
best aspects page "637" from both methods, i.e., the speed of an order combined
with the enforceability of an award, particularly where institutional scrutiny is
required. (110) Of course where enforceability is also required it is the speed of the
latter that matters. The actual form that may be utilised can depend on the
wording in the relevant rules. For example, the LCIA Rules refer to an ‘order’ and
indicate that such orders are subject to final determination in the form of an
award. (111) Thus it is not clear whether a tribunal may render an interim award
in the context of interim measures under such rules. (112)

The tribunal can also offer a mere recommendation to the parties involved
absent any express reference in the rules. However, this has no binding effect.
(113) The ICSID Rules are different. While Rule 39 refers to a tribunal

‘recommendation’ as to interim relief, nevertheless, tribunals have seen this

power as allowing for binding determinations. (114) In other cases a tribunal
might accept a declaration by a party that it does not intend to act in a particular
way as a substitute for an interim measure. (115)

8.2.12. Review and Liability for Costs and Damages

Due to the fundamental nature of an interim or provisional order, the arbitral

tribunal has the opportunity at any time to lift the measure where it feels the
factual basis for granting it no longer applies or exists. (116) In order to maintain
supervision over the circumstances surrounding the granting of an interim
order, the arbitral tribunal should exercise an ongoing process of review. This
process will allow the tribunal to determine when and if the interim measure is
no longer necessary. While it makes sense to allow a tribunal to modify measures
given their temporary nature, doing so based on some of the criteria on which
the measure was imposed would be problematic. For example, it would be
difficult to remove a measure on the basis that subsequent steps have shown that
what was thought to be a reasonable possibility of success is no longer likely to
be the case.

page "638"

Another aspect of review relates to situations where there is some pre-arbitral

referee process that has already entertained an interim measure request. The
question is then whether a tribunal once constituted may reconsider the interim
measure decision. As a matter of principle, if a tribunal has a broad power to
itself award interim measures and to review them for changed circumstances, it
ought to be able to review a decision made by a pre-arbitral referee. That will
often be expressly provided for where there are pre-arbitral rules. (117) For
example, the ICDR Rules allow for modification of an emergency arbitrator
decision by the tribunal. (118) That ought to be a de novo consideration of the
desirability or otherwise of the measure. An arbitral tribunal is not empowered
as a mere review body over the pre-arbitral referee's determinations.

Under the UNCITRAL Arbitration Rules, if the measure granted ‘turns out to have
been an undue measure’, (119) the arbitral tribunal has the ability to award not
only costs but damages against the claiming party as well. It must also be noted
that an order of this nature can be made by the tribunal at any point in the
proceedings. (120) The Model Law provision indicates that the party who obtained
the measure may be held liable ‘for any costs and damages caused by the
measure or order to any party’ where the tribunal ‘later determines that, in the
circumstances, the measure or the order should not have been granted’. (121) It is
not clear what that actually entails. Is it simply a case where the party with the
benefit of the measure ultimately loses on the merits or is it a case where
subsequent evidence shows that a different outcome of the application on the
relevant criteria should have been determined (which could be argued to have
been a failing of the tribunal itself) or is it limited to cases where subsequent
evidence shows that the applicant improperly obtained the measure by reason of
misleading information and submissions? The Secretariat Note indicated that the
decision on the merits is not intended to be determinative on this issue but the
language is again ambiguous and does not highlight the other factors. The
Secretariat Note indicated that the ‘final decision on the merits should not be an
essential element in determining whether the interim measure was justified or
not’. (122) The phrase ‘essential element’ is awkward as any relevant factor is an
essential element but not necessarily determinatively so.

There is no equivalent section in the ICC Arbitration Rules, however, the broad
discretion for the determination of costs no doubt includes the ability to
apportion costs for such interim proceedings. (123) As noted, most rules also make
clear that security can be required of the requesting party. This suggests that,
should excessive damage be caused for little reason, this security will either be
page "639" sacrificed or that damages will be paid. There may also be questions
of causation as to costs and damages, levels of proof and how to deal with
scenarios where there are multiple causes. Causation under such a procedural
test would not necessarily mirror causation of damages provisions under
substantive contract law.

8.2.13. Court Involvement

National courts can play a number of roles in the interim measure procedure.
Depending on the lex arbitri, a court can provide an alternative source of interim
relief for a party. (124) Often described as involving a concurrent jurisdiction,
when so empowered courts and arbitral tribunals share the ability to provide
interim measures based on an arbitration. Court application can be made under
UNCITRAL Rules, (125) ICC Rules, (126) LCIA Rules, (127) ICDR Rules, (128) Swiss Rules
2012, (129) SCC Rules, (130) WIPO Rules, (131) JAMS Rules, (132) and the ICSID Rules.
(133) The ICSID Rules only allow for applications to a court if expressly provided

in the agreement. (134)

Ideally, the relevant laws would clarify the role of a court in such circumstances
and whether it has concurrent jurisdiction to that of a tribunal. It is important to
expressly allow for this under the rules to prevent an argument that such an
application to a court contravenes the promise to resolve disputes solely by
arbitration. Rules will typically expressly deem such behaviour to not conflict
with the arbitration agreement itself. Where the rules are unclear, some courts
have concerned themselves with whether jurisdiction would offend against
Article II(3) of the New York Convention requiring matters pursuant to an
arbitration agreement to be referred to a tribunal on the request of a party to do
so. The better view is that this should not be a barrier, particularly where the
applicant acknowledges that the dispute is subject to arbitral jurisdiction.

Where concurrent jurisdiction is possible there is a problem where there may be

conflicting applications and determining which should take priority and how
conflicts between rulings are to be resolved. Court applications are allowed for
under the Swiss Rules 2012 to national courts prior to tribunal appointment. (135)
page "640" The LCIA Rules only allow for application to court in ‘exceptional
circumstances’ once the tribunal has been constituted. (136) There is no indication
of what would constitute exceptional circumstances, who would make such a
determination and whether a blocking application to a tribunal would be a
meaningful process. Section 44 of the English Arbitration Act 1996 allows for
such application. That is not limited to cases where the Seat of arbitration is
inside the UK. The English Court of Appeal in Dabouran Group Int Inc v. Sims &
Ors set out the factors to be used in considering such applications. The relevant
criteria were that:

1. It should be ‘just and convenient’, ensuring effectiveness without being

‘oppressive’.
2. All relevant circumstances and options must be considered including
questions of proportionality. Third-party rights must be considered in this
context.
3. There should be a balancing of the interests of the applicant as against the
other parties to the proceedings.
4. The applicant should not obtain relief in the foreign proceedings superior to
the relief in the freezing order.
5. The applicant's evidence must be as complete as is reasonably possible.
6. As to standard of proof, there must be a ‘real prospect’ that the assets exist
and are located as asserted.
7. There must be evidence of a risk of dissipation of the assets in question.
8. Notice is not required in cases of urgency but otherwise should be given as
soon as possible. (137)

There are no provisions as to interim measures in the US Federal Arbitration Act.

Article 75 of the New York Civil Practice Law and Rules deals with circumstances
where a New York court may provide for provisional measures in support of
arbitration. It provides for such a power where any resultant award ‘may be
rendered ineffectual without such provisional relief’. (138) The US courts have
taken differing views as to whether this is the sole test or whether it
encompasses the traditional criteria alluded to above, dealing with matters such
as irreparable harm, balance of convenience and possibility of success on the
merits. (139)

Where local courts are empowered to provide interim measures, these are not
limited to those known under domestic law. Thus it is theoretically possible to
seek a Mareva injunction in a civil law Seat. (140) Nevertheless, a Singapore Court
of Appeal refused a Mareva injunction to protect money with regard to
prospective damages, when the Seat was not Singapore.

page "641"

The second key role national courts have within the interim measure process is
the enforcement of the device. As noted above, attitudes of national laws and
courts differs on this issue. Court support is provided under section 1041(2) of
the German Code of Civil Procedure and Article 183(2) Swiss PILA Article 183(2)
Conversely, Swedish law does not consider interim measures to be enforceable.
(141)

8.3. Security for Costs

8.3.1. Introduction

Section 8.2 above looked at interim relief generally. It considered the practical
demand for interim relief powers as against theoretical concerns that these may
be antithetical to the notion of party autonomy. It also considered the utility of
affording relief powers to arbitrators that may be difficult or impossible to
enforce. One specific form of interim is that of security for costs. A security for
costs request is brought by a party responding to a claim and seeks to force the
party bringing the claim to provide sufficient security to cover the costs of any
fees and expenses that may be awarded against it by the arbitral tribunal. (142)
When referring to the party responding, this does not mean the respondent per
se, but rather any person answering a particular claim or counterclaim. Hence
the respondent may seek security as against claims made by claimant, claimant
may seek security as against counterclaims by respondent and third parties may
seek security when they are the subject of joinder applications. Security for costs
applications might also be made against a party seeking another form of interim
measure. That would simply be one form of security as dealt with in provisions
such as Article 17(E) of the UNCITRAL Model Law. The suggested need to
provide security for costs arises out of a policy goal to prevent frivolous or
nuisance litigation. It is designed to ‘make the prevailing party page "642" in a
dispute truly whole by compelling his opponent to pay legal fees, court costs, and
other reasonable expenses the winner incurred in pursuing its claim or defense’.
(143)
The issue is contentious and deserves separate treatment for a range of reasons.
The first challenging hypothesis is that security for costs orders undermine the
very nature of arbitration itself. If parties have agreed to arbitrate without any
qualifications or preconditions, then that is what must ensue. Some would also
argue that such powers offend against notions of neutrality and equal treatment
as they only apply against the party bringing the claim. (144) The neutrality
argument also contrasts security as against that party only, with the normal
approach to advances on costs where these are to be contributed equally with
the default position being contribution by any party who wishes the matter to
proceed. An arbitrator accepting a mandate would need to find clear justification
for barring a person from pursuing their claim, absent the provision of a
designated security. An added complication is that at the domestic litigation level
security for costs has tended to be a common law concept, particularly in relation
to Commonwealth jurisdictions. Nevertheless, it is not unknown in civilian
jurisdictions where it is described as cautio judicatum solvi. (145) To some, security
for costs should never apply. To others this should at least mean that where the
laws and rules permit such a discretionary decision, higher standards should be
required before an order is granted.

From an efficiency perspective, there are arguments either way that may depend
on the particular circumstances. Perhaps an extreme example underlying the
demand for security for costs entitlements would be the case of a claimant with
no funds, using a law firm working on a contingency basis to bring arbitral
claims. If the claimant succeeds, it will gain relief and be able to pay its costs as
well. If it loses, it pays nothing to its own lawyer and has no funds to compensate
a deserving respondent. Even where contingency fees are not available, a
claimant might only have the funds to cover its own legal fees, hence the same
policy concerns may arise. In such a scenario, it might be argued that only the
claimant is truly afforded a meaningful arbitration right. At the other extreme,
one would not wish to see security for costs powers being a mechanism to
provide an insurmountable impediment to a deserving claimant who cannot
afford the security. That would be doubly so if the claimant's impecunious
position was caused by the respondent's actions that are the very subject matter
of the dispute. If a party was excluded from the arbitration because of an adverse
security for costs decision, there is also a possibility that it would seek to file
court proceedings. (146)

page "643"

A further practical concern is that to the extent that these variable factual issues
may properly be considered in any discretion, it can be problematic to ask an
arbitrator to make such determinations during the currency of a proceeding.
Consideration of the likely merits could be argued to be prejudgment.
Considerations of financial circumstances can involve complex analysis of a
company's true position and involve contests about highly confidential material.
Any degree of arbitrariness in making the order would be concerning, because as
with all interim measures, a security for costs order entails certain immediate
damage to the party subject to it, including lost interest and missed investment
opportunities. The imposition of such an order may at times impose considerable
practical constraints on the ability of a claimant to proceed with its legitimate
claims and is sometimes therefore seen as a limitation on the right to a fair
hearing. A converse argument in favour is that a good faith agreement to
arbitrate involves a good faith agreement to accept that awards are binding and
are to be honoured. A claiming party who knows that it cannot honour a costs
award, could be argued to violate this principle.

A final policy element is that the view a tribunal may take on these issues can be
impacted upon in part by the view it takes on costs recovery generally. This may
be dependent on party agreement in that regard, or arbitral laws and rules. For
example, in an arbitration where there will be no ultimate costs recovery for the
successful party, there is little justification for interim security in that regard.

To the extent that recovery of costs is analogous to damages, security for costs
may need to be looked at alongside security for ultimate damages awards. The
latter are only allowed in extreme circumstances in arbitration. One difference
where costs are concerned is that a successful respondent will often have no
separate damages relief and costs recovery is the only outcome they are looking
for. In an extreme case a security application may traverse similar policy issues
as arise where multi-party, corporate veil and extension issues are concerned.
This would arise where the claiming party is impecunious and is being funded by
a related entity who will benefit if the claims succeed but be beyond exposure if
they are lost. Third-party involvement may also be relevant where there is an
assignment of arbitral rights.

8.3.2. Party Autonomy and Security for Costs

Parties can generally include specific provision for security for costs within their
arbitration agreement or conversely exclude the possibility of such a measure
altogether. (147) The only significant restriction on party autonomy to provide or
deny security for costs arises out of mandatory procedural rules that may be
applicable to the contract or dispute or public policy issues applicable under
page "644" enforcement norms. (148) One question alluded to above is whether a
security award which cannot be met offends against due process norms such as
giving a party an opportunity to be heard as, for example, under Article 18 of the
UNCITRAL Model Law. That should not be so as due process must be seen within
the context of the arbitration as agreed. If it is a conditional agreement where a
party will only arbitrate on the provision of security, those are the parameters
within which due process must be afforded. (149) In the context of party
agreement in favour of security, the more important concern would be whether
there are any rules in the Seat that prevent such measures. If so, the tribunal may
be operating beyond valid power even when it is responding to a request of the
parties themselves.

In most cases parties will not address procedures for security orders in an
arbitration agreement. This is probably due to the fact that drafting such
provisions would require protracted and potentially costly negotiation for what
appears to be a minor procedural detail. Arbitrators are as a result generally free
to exercise complete discretion in deciding whether to award security for costs.
(150)

8.3.3. Enforcement of Security for Costs Orders

Parties and arbitrators alike must be attuned to the effects that the various
domestic legal rules relevant to the proceedings may have on an order for
security for costs in arbitral procedure. The determination itself will not be
enforceable as an award as it would not meet the conditions of a final
determination of an issue in dispute although the contrary view is considered in
section 16.3.3. The situation may be different if there is an express agreement for
arbitration to be conditional on security and the tribunal is making a partial
award on that very issue. Nevertheless, and as noted, enforceability concerns
may be raised if a security determination unduly limits a party's ability to
present its case which it then loses.

8.3.4. Lex Arbitri and Rules and Security for Costs

Lex arbitri and rules tend to fall into three categories. Some provide specific
express powers for a tribunal to award security for costs. (151) Where the parties
page "645" have selected a Seat with such laws or have selected such rules, this is
part of party autonomy in any event. A tribunal can simply apply the laws and
rules as designated. If the power or prescription is in the lex arbitri, the next
question is whether it is mandatory or whether the parties could contract out of
its provisions.

A second category is where a tribunal relies on broad interim measure powers,

seeing security as simply a subset of these. Here there needs to be a concern as to
whether a security application can truly be seen as an interim measure as
defined in the express norms. If so, the tribunal will need to apply the designated
criteria for such measures. Fouchard, Gaillard and Goldman generally see interim
measures powers as being broad enough to encompass security for costs. (152)
The 2006 UNCITRAL Model Law amendments broadening the interim measure
powers were intended to encompass security for costs through wording which
stated that interim measures include ‘a means of preserving assets out of which a
subsequent award may be satisfied.’ (153) The ICC Rules are also said to permit
security for costs orders under the broad interim measures power. (154) A third
possibility is to simply see the power to grant security for costs as falling within
the general discretionary power of a tribunal. (155) Lew, Mistelis and Kröll note
ICC Case no 7489, (156) where a tribunal considered the power to be part of the
inherent powers under the earlier ICC rules. (157)

In the US, neither the Federal Arbitration Act nor most state statutes clearly
allocate power to order security for costs between courts and arbitrators in the
absence of previous agreement between the parties. (158) However the widely
accepted position in the US is that ‘arbitrators have the power to fashion relief
that a court might not properly grant’, (159) particularly interim and conservatory
measures such as security for costs. Nevertheless, in legal systems like that in the
US in which parties are expected to cover their own costs even in the case of an
adverse final award, judges and arbitrators may not be particularly receptive to
requests for security for costs.

Although Swiss law historically did not provide a means for ordering security for
costs, it is now accepted that Article 183 (and possibly 182) of the Swiss Private
page "646" International Law Act provides arbitral tribunals and courts with
authority to order security for costs. (160)

8.3.5. Criteria

While security for costs applications are a subset of interim measures generally,
they raise discrete issues that require separate elaboration. As noted above, a
distinct policy question is whether a claiming party can ever be barred from
bringing a claim in arbitration simply because it may not have the funds to pay
an adverse costs award. A second key difference is that the standard criteria for
considering interim measure applications do not normally apply to security for
costs. A typical interim measure application such as preserving evidence or
assets or enjoining a party against taking a proposed course of action looks for
urgency, serious or irreparable harm and a tenable case on the merits. Where
security for costs applications are concerned, urgency and injunctive relief is not
relevant. The irreparable harm is simply the allegation that a losing claimant will
not be able to pay a costs award. As to consideration on the merits, that can
remain relevant and is effectively looked at in reverse. The less likely the merit in
the claims, the more there might be concern to protect a potential cost award in
favour of the responding party.

For these reasons it might be thought to be relatively easy to state the criteria as
a serious risk of inability to pay an adverse costs award coupled with a serious
risk that the claim will fail on its merits. However, the policy concerns as noted
above to the effect that an unconditional agreement to arbitrate is antithetical to
a security for costs award, at least in circumstances where it would bar the claim
and particularly where the responding party's behaviour is alleged to have led to
the impecuniosity of the claiming party, means that a more nuanced analysis
may be necessary. Arbitrators need to be careful in calling on their own
background experience to this question, given that many civilian arbitrators
would see this as a peculiar feature of English law. (161) It has been suggested that,
given the importance of balancing the competing goals, each request for security
for costs should be assessed on a case-by-case basis and that there are no ‘hard
rules’. (162) However, the tribunal should apply ‘equitable principles’ including
‘whether or not the interests of justice lie in favor of granting or denying the
application’. (163) Yet these unassailably valid observations provide little concrete
guidance.

page "647"

Further guidance is available for tribunals and parties that sets out both what the
party seeking the order should aim to establish, and how the other party should
seek to defend itself against an order for security for costs. Bunni cites the
following extract from an ICC arbitration:
[I]t would be appropriate for the tribunal to exercise its discretion to make an
order for security for costs[:]

(i) if the Respondent, which has requested that such an order be addressed to
the Claimants, can show:
(a) that the factual situation at the present time is substantially different
from that which existed at the time the parties entered into their
arbitration convention, and
(b) that the present situation is of such a nature as to render it highly unfair
to require it to conduct the arbitration proceedings without the benefit
of such security;
(ii) unless the Claimants, which oppose the making of an order for security for
costs, can show:
(a) that the making of such order for security for costs would in effect deny
their right of access to arbitration for reasons not attributable to them,
and
(b) that, after having weighed the parties respective interests considering
both the subject matter of the dispute and the circumstances giving rise
to the request for an order for security for costs, the making of such
order would appear to be highly unfair to Claimants. (164)

Guidelines have also been prepared by the Chartered Institute of Arbitrators. (165)
The Chartered Institute Guidelines 1997 call for attention to six key factors
namely, whether a claimant has reachable assets in the jurisdiction, whether the
application is used as a weapon to block a legitimate claim, the extent to which
respondent has contributed to claimant's lack of means, the timing of the
application, the appropriateness of a security determination for the specific
nature of the arbitration concerned and equitable concerns in light of all
circumstances. Other factors may apply which should only be considered in
exceptional circumstances, namely the merits of the claim and chances of
success, settlement offers that may have been made and the likelihood that a
claiming party will page "648" abscond. (166) The following discussion further
canvasses a range of such and similar factors that have been mentioned in cases
or literature or which might otherwise be argued to be relevant as there are
considerable problems even with those that are conceptually relevant.

Costs criteria

What is the likely approach that the tribunal will take to an award of costs in due
course? The less likely a tribunal to order costs in favour of the winning party,
the less justification there is for a successful security for costs application. In
some cases a tribunal may not wish to make such a determination as to costs
principles at an early stage although it may feel more comfortable in doing so in
an area such as investment arbitration where there may be a greater reluctance
to award such costs or where the parties come from legal systems where the
reasonable expectations would be against such costs awards. (167)
Nationality or residence of the party

Some rules specifically prohibit the fact that the proposed subject of an order for
costs is of foreign nationality or residence as a basis for making an order for
costs. Even where this is not the case, the concept of nationality should not be
relevant in the international arbitration setting. (168) Any view to the contrary
would undermine international arbitration itself. While nationality and
residence per se should not be held against an applicant, a relevant criterion is
the degree of connection to the Seat, a factor considered relevant in Ken-Ren (169)
and Bank Mellat. (170)

Prospect of success of the requesting party's arbitral claim

This is a problematic criterion regardless of the way it is treated. On the one

hand, tribunals generally ought not to engage in an assessment of the apparent
merits of the parties' claims in order to decide whether to make an order for
security for costs against a party. This is because the tribunal should avoid giving
an indication of its position in this regard before the case has ended. On the other
hand, express interim measures powers consider this a relevant factor. From a
policy perspective, the higher the chance of success of the claiming party, the less
justification there is for security. Even if the criterion is thought necessary on
logical grounds, it is exceedingly difficult to employ in practice. This is simply
because it would be hard to ever assert that there is no tenable chance of failure
and hence no possibility that an adverse costs award could be unmet. To so
conclude would be equivalent to concluding that there was an entitlement to
summary judgment. Hence it would almost certainly be the case that a tribunal
would conclude that the claims might fail. This would certainly be so where there
is conflicting evidence and no opportunity as yet to hear and evaluate conflicting
witnesses both expert and factual.

page "649"

To the extent that the merits of the claims are a relevant factor, one possibility to
shelter a tribunal from claims of prejudice is to have a discrete person deal with
security applications. That would obviously add to the costs and in any event, if
that person imposed security which was not then paid, the tribunal would have
delegated the power to control its proceedings to that other party. (171)

Financial circumstances

This is an important consideration for the tribunal. Is the proposed subject of the
order for security for costs in financial difficulty? If so, this increases the
likelihood that the responding party will be left out of pocket. There are both
practical, procedural and policy concerns where this criterion is being evaluated.
The first difficulty is that the key data is in the possession of the claiming party.
The responding party has to make an allegation of financial difficulty without
naturally having access to its opponent's confidential material. At times it will be
forced to rely on commentary and rumour in the financial press. A second
problem is to engage in projections over the likely time period of the arbitration.
What presumptions can legitimately flow from a downward trend? A further
problem is that in some cases, a relevant factor might be the claiming party's
attitude to paying its debts voluntarily. In assessing the financial position of the
claiming party, a tribunal might also need to consider enforceability and the
location of assets.

There are arguments in favour of the proposition that the tribunal ought to
concentrate on whether the financial condition of the proposed subject is
materially different to its condition at the time that the parties entered into the
arbitration agreement, and if so, why this is the case. The reason for this
consideration is that, if the proposed subject's financial condition, however
perilous, is substantially the same as at the time that the parties concluded the
arbitration agreement, or has deteriorated in a manner that is not commercially
unusual, the responding party could be taken to have accepted the risk of
transacting with a financially unstable entity. It may be that only where some
unusual or unforeseeable circumstance has precipitated the company's poor
financial condition that the tribunal should order security for costs. For example,
on this view, currency fluctuations should not be a relevant factor unless they
were so extreme as to have been beyond the reasonable contemplation of the
parties. Even when the circumstances are fundamentally different, this should
not itself support an application as it is still necessary to show why it would be
unfair to allow the claiming party to proceed without posting security.

An example where circumstances might have changed is where a third-party

funder is supporting a claimant to bring an expensive action where this could not
have been anticipated from the outset. (172) A further relevant factor but which
might be contentious to consider at an interim stage, is whether the claiming
party's impecuniosity has been caused by the inappropriate behaviour of the
responding party. If such facts were clear then this would be a strong argument
against awarding security for costs. However, it may be problematic to engage in
that analysis page "650" given problems of proof. Even where it can readily be
shown that the responding party's behaviour caused the impecuniosity, that very
behaviour may be the substance of the dispute as to whether it was valid
contractual behaviour or otherwise, for example, non-payment for goods alleged
to be faulty. Care would need to be taken to ensure that a determination in that
regard is not somehow prejudging the dispute. For example, there may even be a
dispute as to whether required payments have actually been made. This could
not easily be factored into a security application without a full opportunity to
resolve that disagreement as to the facts.

Enforceability and other execution problems

Where there is a need for a fundamental change in circumstances to ground a

security application, the fact that awards may not be enforceable in the claiming
party's country should not be a relevant factor as this would have been known at
the time of entering the arbitration agreement. The same is the case where
exchange controls would limit the ability to transfer funds paid on an award. (173)

Bad faith
The potentially oppressive nature of security for costs means that such orders
may be used as a weapon to stave off consideration of the dispute. The tribunal
might then consider whether the party requesting the order is acting in good
faith. Where good faith is concerned, the party seeking security should also have
made its own contribution to advances. It should not be favoured as a result of its
application. (174) There is a problem in considering whether a security application
is simply motivated by a wish to frustrate the claiming party. It is difficult for a
tribunal to pass judgment on such intent. In many cases such a determination
would not be needed as bad faith considerations would only need to arise where
there is otherwise some basis for the application. Once again, if the reasonable
expectations were that the responding party took the risk in a commercial
transaction in dealing with an impecunious party, it should not be able to call for
security at this stage of the dispute.

Timing

Another relevant factor is the timing of the application. As with most such
applications, claims should be made in a timely manner. A failure to make a
request for security at the earliest opportunity could also be seen to be a waiver
of the right. However, in some circumstances, the change in financial
circumstances or knowledge about them might only occur well into the tribunal
process.

8.3.6. Security where the Tribunal has yet to Establish Jurisdiction

A question addressed above is whether tribunals ought to order interim

measures prior to determining whether they have jurisdiction where the latter is
contested. It was suggested that while there should be some caution, the power
to do so is page "651" justifiable in appropriate circumstances. The same should
be so where security for costs applications are concerned. A responding party
challenging jurisdiction ought to be able to assert that the claim has no
jurisdictional merit, the claiming party has no funds to pay an adverse costs
award and hence security should be possible. (175)

8.3.7. Form and Amount of Security

The tribunal will need to determine the form of security where the rules allow
for some discretion. This can typically take the form of cash or a bank guarantee
(176) or standby letter of credit. (177) Security can also involve payments into an

escrow account or some kind of security or lien over property. The tribunal
should set out in the order various details such as dates and accounts by which
or into which the security is to be provided. Security for costs should correspond
as closely as possible to the security needs of the responding party and minimise
as much as possible any oppression to the claiming party. (178) If orders are to be
made there is a difference between segmenting into stages or making one order
for the entirety of the proceedings. It may be appropriate to consider security in
stages and not in a lump sum. (179)

Another challenge is to set an appropriate amount as this would entail an

analysis of the fee estimates of the party seeking security. While parties are not
normally obliged to indicate their fee arrangements to the other side, this may be
relevant to a security for costs application. There may also be an analysis of
whether these are reasonable both as to rate and time and a determination at
that stage as to whether the tribunal might limit cost awards to only reasonable
amounts. The latter determination should not be of concern as the parties ought
to be aware of this from the outset. (180) Another question is whether an estimate
should be discounted or conversely whether a leeway should be built in as a
margin for error. A further question is whether the security should attract
interest that can then be compensated if the claiming party is successful.

page "652"

8.3.8. Release of the Security

In its final award the tribunal should address the release of the security for costs.
(181) In order to avoid the subject party paying the costs twice, the value of the

security should be offset against any amount payable by that party under the
award, or it should otherwise be released. (182)

8.3.9. Enforcement of Security

Under the English legislation, if the order for security is not complied with then
under section 41(5) the tribunal can make a peremptory order to the same effect
stating a date for compliance. If that is not complied with then pursuant to
section 41(6) the tribunal may make an award dismissing the claim. There may
be a question as to whether such an award would have res judicata effects even
though the merits were not actually determined or whether the claim can
subsequently be brought when security is available.

Obviously a tribunal's power can only be as to obligations of a party. Where

security is provided by way of a third-party guarantee, if payment is not
forthcoming, enforcement would not be within the jurisdiction of the tribunal.

8.3.10. Procedural and Evidentiary Issues on Security Determinations

While it will naturally be the case that the party seeking the order has the
burden, (183) there are challenging procedural and evidentiary elements of such a
process. This is particularly so as the party seeking the order is making
allegations as to the financial position and intentions of its opponent where the
latter would not willingly afford access to financial records. A tribunal might
even be faced with an application for document production in support of the
security application. This is unlikely to be granted, the better view being that the
applicant should have some meaningful evidence before making the claim. Any
contrary view would invite a fishing expedition as a matter of course in support
of security applications. Exceptional circumstances might lead to a contrary
conclusion. Where contested discovery applications are allowed there will then
be a problem as to whether adverse inferences can appropriately be drawn after
a claiming party's refusal to provide confidential documents.
page "653"

A difficult evidentiary question is as to the standard of proof and the means by

which a tribunal will make a prediction as to the likely ability of the claiming
party to pay costs at some future point in time. As noted above, some would
argue that there should be higher standards in arbitration given that security
discretions should not interfere with the parties' prior agreement to arbitrate.
(184) If there were to be higher standards in arbitration there would need to be

some logic behind that determination. It is easier to argue against a security

power in its entirety on the basis of the unconditional agreement to arbitrate
without qualification. Conversely, an agreement to arbitrate under rules such as
that of the LCIA which expressly afford a power, should see that as an
unconstrained discretion.

In other circumstances, one possible criterion for identifying higher standards

argued for above would be to consider the financial circumstances at the time of
the dispute and the reasonable expectations of the parties when entering their
arbitration agreement. If the current claiming party's financial circumstances are
significantly worse than the reasonable expectations of the parties at the outset,
this may be a relevant factor. An extreme example would be a claimant that has
stripped itself of assets, arguably to shelter itself from a negative determination.
As to the standard, there certainly must be more than a possibility of non-
payment but there would then be a debate as to whether a tribunal would expect
mere evidence of a probability or would require convincing evidence that on
balance there will not be payment. (185)

It may be easier to present evidence against a publicly listed company where

annual accounts and statutory returns are publicly available. Public documents
might also show that there are primary securities over its entire assets. Proof
will obviously be easier if the claiming party is in receivership or otherwise
under some arrangement with creditors. Assessment will be far more difficult
when it is based on analysis of likely future cash flows in uncertain market
conditions. Other problems involve analysis of likely borrowing power where the
claiming party's business generally relies on credit.

8.4. Anti-Suit Injunctions by Arbitral Tribunals

8.4.1. Introduction

A party to an arbitration proceeding or prospective proceeding might initiate or

threaten another separate arbitration proceeding or a domestic court proceeding
in apparent violation of the arbitration agreement. The question is then what can
the page "654" other party do? Should the court or other tribunal be relied upon
to refer the parties back to the original arbitration? Should the other party be left
to contest the jurisdiction of the other forum? Should the tribunal in the original
arbitration proceeding purport to restrain any other proceedings? One response
is that an arbitral tribunal might issue an anti-suit injunction. An anti-suit
injunction is an order to restrain a party from pursuing such other proceedings.
8.4.2. Comparative and Policy Perspectives

There is no consensus view as to the merits of such an approach. This is

unsurprising given the differences in view between legal families. The
comparison between common law and civil law legal traditions in the context of
anti-suit injunctions is stark. The anti-suit injunction is historically a uniquely
common law remedy (an ‘oddity of the Anglo-American legal system’). (186) The
anti-suit injunction is generally not a remedy that is available in civil law
jurisdictions. (187) Nevertheless, it is said that some civil law jurisdictions
recognise equivalent remedies. (188)

Policy considerations will always be more problematic where there is such

divergence between legal systems. Nevertheless, there are some arguments in
favour of such a remedy although these raise other contentious issues.
Arguments in favour include the wasted cost of multiple proceedings, the
potential for conflicting determinations, aggravation of the dispute and the
potential for cross-injunctions between the differing fora. (189) A further
suggested argument in favour of the power is the general duty to refrain from
conduct that may aggravate the dispute. (190) It is difficult to rely on such a
principle as a basis for the supposed page "655" power. It cannot trump any
rights the parties may have, including rights to assert jurisdiction of a different
forum. The essential notion should be to refrain from aggravating the underlying
dispute while an adjudicatory solution is sought. Concern for retaliatory
injunctions between different fora may be a real problem which has been raised
by proponents but can hardly be an argument in favour of an anti-suit order as
the action by one adjudicator may well inspire retaliatory action by another.

There is a particular concern that parties not be denied the fundamental right of
access to courts. (191) An anti-suit injunction necessarily involves the arbitral
tribunal affirming its jurisdiction to the exclusion of the jurisdiction of competing
fora. (192) This is in apparent conflict with the fundamental principle whereby
each court or tribunal can decide upon its own jurisdiction. (193) The argument is
that an arbitral tribunal should not purport to affirm its jurisdiction to the
exclusion of the jurisdiction of another court or tribunal. There is also a problem
in seeing an anti-suit order as simply a form of interim measure as in many cases
the aim is to be a permanent denial of an alternative forum's jurisdiction. A
further argument against anti-suit orders is that it is entirely possible that in
some cases, poor drafting and planning has led to different fora having arguable
jurisdiction. The fact that one tribunal finds jurisdiction does not prove ipso facto
that all other fora do not have similar rights.

The counterargument is that the parties have, by making an arbitration

agreement, agreed to submit their dispute to arbitration, and an anti-suit
injunction is, in page "656" that sense, only a mechanism to enforce the
expectations of the parties. (194) Another counterargument is that an anti-suit
injunction is addressed to one of the parties and not to the forum where the
other proceedings are instituted. (195) There are also policy arguments in relation
to comity. (196) However, comity is said to be irrelevant to an arbitral tribunal and
that, in any case, other considerations are predominant. (197) Furthermore, comity
entails one adjudicator deferring to another, rather than one making demands of
the other.

8.4.3. Laws and Rules

The bases of the power of an arbitral tribunal to issue an anti-suit injunction are
suggested as being via the inherent jurisdiction of the tribunal; (198) via the lex
arbitri (e.g., Article 17 of the UNCITRAL Model Law); (199) via procedural page
"657" rules; (200) and via the arbitration agreement itself. (201) The power of an
arbitral tribunal is explicitly provided for in the UNCITRAL Model Law Article 17
which provides in relevant part that: ‘[t]he arbitral tribunal may, at the request
of a party, grant interim measures … by which … the arbitral tribunal orders a
party to … [t]ake action that would prevent, or refrain from action that is likely to
cause, current or imminent harm or prejudice to the arbitral process itself. It has
been suggested that Article 17 was redrafted specifically to encompass anti-suit
injunctions. (202) The Iran–US Claims Tribunal considered that it had an inherent
power to issue such orders to conserve the respective rights of the parties and
ensure that the tribunal's jurisdiction and authority are made fully effective. (203)

Gaillard surveyed reported decisions of arbitral tribunals in relation to anti-suit

injunctions and concluded that anti-suit injunctions had been awarded in a large
number of proceedings over a long period of time. (204) Furthermore, in
jurisdictions where courts themselves are not allowed to grant anti-suit
injunctions in favour of arbitration, parties will more naturally turn to tribunals
for such remedies. (205)

8.4.4. Discretionary Considerations and Anti-suit Applications

While there are strongly differing views as to whether anti-suit orders are
advisable or not, all would agree that where they apply it is still a discretionary
matter. There are a number of factors that a tribunal ought to take into account
where it accepts that such a power is possible. Here it is important to consider
the exercise of the power itself and alternative measures such as damages for
breach of the arbitration agreement. The latter may depend on whether an
application is actually made to this effect and if not, whether it would be
appropriate for a tribunal to raise the possibility. That itself would be
contentious. Relevant factors include the reasons why the relevant party is
seeking jurisdiction before an alternative forum, the likelihood that the party will
comply with an order either immediately or because page "658" of some
enforceability power, the possibility that the other forum will respect the order
or conversely the possibility that it might engage in retaliatory injunctive action,
the extent to which the application to the other forum aims to directly hamper
the arbitral proceedings and the adverse commercial impact on the party seeking
the anti-suit order.

If reliance is made on an interim measure power there is then a question of the

criteria to be employed. As noted in section 8.2.5 there are differing views about
this and differing laws and rules express criteria either generally or more
specifically. One limitation under traditional criteria is whether there could be
said to be irreparable harm. In Boeing Co, Judge Holtzmann considered that the
loss of the right to be free of litigation in another forum could itself be
irreparable. (206) While the argument is tenable, it certainly takes interim
measures beyond their traditional purview.

Most commentators suggest that anti-suit injunctions should be approached with

extreme caution. (207) The cautious approach restricts the use of anti-suit
injunctions to extreme circumstances. Lévy states that ‘[a]rbitrators will have to
ensure that the requested measures are urgent, aimed at preventing irreparable
harm or necessary to facilitate the enforcement of the upcoming award’ (208) and
further that ‘arbitrators should only issue anti-suit injunctions when it comes to
their attention that one of the parties has committed fraud or otherwise engaged
in abusive behavior in order to revoke the arbitration agreement’. (209) Gaillard
suggests that it is necessary to consider factors including whether or not the
anti-suit injunction ‘is necessary or urgent, or if a party would suffer an
irreparable harm’. (210) Moloo suggests a more expansive approach: ‘Once it is
found that an exclusive arbitration agreement exists, an antisuit order should be
granted to remedy a breach of the agreement in all circumstances. Parties should
be held to their bargain.’ (211) Moloo suggests that it is within the jurisdiction of a
tribunal to remedy a breach of the arbitration agreement in the same way as any
other breach of contract. (212) The restrictive approach treats an anti-suit
injunction as a form of interim measure. The more expansive approach is
concerned with the substantive obligations under the arbitration agreement and
sees the injunction as a form of negative specific page "659" performance
obligation. The latter contention does not address the conceptual challenge as to
the right of one adjudicator to make negative jurisdictional rulings under criteria
where they are not empowered adjudicators. There is a problem in considering
anti-suit orders as simply specific performance directives in relation to breach of
arbitration agreements. If one is speaking of specific performance, it is important
to consider exactly which performance is being promoted. The natural
performance under an arbitration agreement is to arbitrate. Yet we do not
consider it appropriate to make orders for specific performance of the
arbitration promise itself where a respondent simply does not turn up. Instead,
arbitration calls for a claimant to proceed to make its case, and usually requires
it to advance all the fees and costs prior to a final determination. Hence an anti-
suit order is not truly specific performance of the arbitration agreement but an
order not to do anything that is contradictory to it. (213) Furthermore, an anti-suit
injunction in favour of arbitration is essentially making a final determination on
a key issue, either that the tribunal itself has jurisdiction or at least has priority
in determining that very question. Once again it is not a typical interim measure.
For better or worse, many systems will nonetheless treat anti-suit injunctions by
tribunals as a form of interim measure.

One question is whether a tribunal must determine that it has jurisdiction before
it would promulgate an anti-suit injunction against parallel proceedings. The
contrary view is suggested by Gaillard arguing that parties should not undermine
a tribunal's prima facie jurisdiction. (214) If that was the case, it would truly be an
interim measure of a temporary nature simply seeking to avoid contradictory
conduct while a jurisdictional determination is being made. As argued previously
while open to debate, the better view is that in some circumstances at least, a
tribunal ought to be able to make such an order. (215) In such circumstances a
tribunal is simply likely to order a stay of the other proceedings while a
jurisdictional decision is to be made. (216) A tribunal in SGS v. Pakistan granted a
temporary anti-suit and anti-arbitration order pending its determination on
jurisdiction. (217) Once again, such decisions are conceptually different to more
general interim page "660" measures being based on views as to the essential
approach to determining jurisdictional questions.

The party seeking the injunction should carefully consider whether the
injunction would likely be enforceable and any potential adverse effects, in
particular in the case that the injunction is not voluntarily complied with. The
party should consider alternatives such as an application for monetary
compensation in relation to any parallel proceedings instituted by the other
party in breach of the arbitration agreement. This is discussed below. Another
possibility is to only consider exercising the power where the alternative action
would frustrate the arbitration process directly. This might arise where a party
seeks an anti-arbitral injunction from a court and hence is itself trying to obtain
orders barring the arbitrator from continuing. The ICSID tribunal in Plama
concerned itself with whether the contested actions would make resolution of
the dispute by the tribunal more difficult. (218)

8.4.5. The Form and Consequences of an Anti-suit Injunction

Another question is whether any determination should be made by way of a

procedural order or should be as an award. Where the decision is truly a final
decision on a matter, for example, damages for breach of an arbitration
agreement, it ought to be by way of an award to allow for enforceability. Where it
is purely a temporary measure, such as a holding injunction while the tribunal
determines its own jurisdiction, the better view is that this does not meet the
essential nature of an award. Where a tribunal has made a jurisdictional
determination that an arbitration agreement is exclusive and countenances no
possible recourse to alternative fora, then there seems no reason in principle
why this should not be rendered as an award with whatever enforceability may
then ensue. Because an anti-suit injunction, unless voluntarily complied with,
must be enforced in the jurisdiction in which the competing court or arbitration
proceedings are instituted the likelihood of it being enforced in that context is
slight. (219) There also is a risk that the final award may be unenforceable or
vacated on the ground of lack of jurisdiction, impartiality or public policy. (220)
However, even if such an order is not strictly enforceable, it page "661" still may
be useful. (221) It may have indirect effects, such as putting a ‘recalcitrant’ party on
notice. (222)

8.4.6. Alternatives to Anti-suit Injunctions

An order for monetary compensation for breach of the arbitration agreement has
been suggested as an alternative to an anti-suit injunction. The argument in
favour is that a good faith acceptance of an arbitration agreement is an
agreement to not have inappropriate recourse to a court and where the latter
occurs, damages caused to the other party naturally flow from breach of the
arbitration agreement. (223) It is suggested that monetary relief addresses the
same wrong and offers a compensatory remedy that anti-suit injunctions do not,
and is usually a simpler and more effective way of affirming jurisdiction. (224) A
tribunal will need to consider whether the arbitration agreement is broad
enough to allow for claims as to breach of its own terms. (225) If that is not the
case, it still might be possible to consider a temporary order under a general
interim measure power. However, the damage resulting from a party bringing
additional proceedings may be difficult to compensate. It might be difficult to
quantify such damage beyond the direct costs of challenging such a proceeding.
(226) If damages are to be assessed in relation to the costs of the other

proceedings, care will need to be taken to consider any existing or prospective

costs rulings within those proceedings themselves. Otherwise there is the
possibility of double compensation, although it may not matter whether the
other forum might consider costs in the future as whichever forum goes second
ought to take into account payments already made in compensation for legal and
related expenses.

page "662"

Where damages are considered, this may raise the added contentious question of
whether punitive damages may be possible. If that is not so, a damages order
limited to the extra legal costs is unlikely to be much of a deterrent, although that
should not be a reason against a remedy that naturally flows from breach of the
arbitration agreement itself. One problem in considering damages as an
alternative would be if there was a claim of failure to mitigate by failing to seek
an anti-suit order. (227) Landau states: ‘[T]he impact of the local court process will
rarely be quantifiable in damages in any event. Procedural safeguards, such as
confidentiality, may have been lost; sensitive documents may have been ordered
to be produced; individuals may have been subjected to local pressures; and the
politics (and so economics) of the project may have been forever changed.’ (228)
An order for compensation might not be any more enforceable than an anti-suit
injunction. In addition, monetary compensation does nothing directly to prevent
a party from pursuing parallel proceedings. (229)

In some cases parties may have tried to promote their rights to force arbitration
or enhance the powers of the tribunal to make appropriate orders via other
forms of contractual drafting. Examples include undertakings not to bring
contradictory proceedings, supported by liquidated damages remedies for
breach. (230) Liquidated damages rights can themselves be contentious under
some applicable laws.

8.5. Anti-Arbitration Injunctions by Courts

8.5.1. Introduction

A court may issue an injunction purporting to restrain arbitration proceedings (a

so-called anti-arbitration injunction). Anti-arbitration injunctions are ‘relatively
rare’ but they present an extremely serious problem. (231) In what circumstances,
if any, should an arbitral tribunal respect an anti-arbitration injunction? (232)

page "663"

8.5.2. Power to Issue an Anti-arbitration Injunction

An anti-arbitration injunction, as a type of anti-suit injunction, is, generally

speaking, available as a remedy in common law jurisdictions, but not in civil law
jurisdictions. (233) A party might seek an anti-arbitration injunction where it
contests the jurisdiction of an arbitral tribunal. ‘In theory at least, it may be
entirely justified for a party contesting the existence of a valid arbitration
agreement to commence court actions before the court it considers to have
jurisdiction …’. (234) Section 72(1) of the English Arbitration Act 1996 allows a
person alleged to be a party to arbitral proceedings who has not taken any part
in those proceedings, to challenge substantive jurisdiction before a court. The
provision expressly allows for injunctive or other appropriate relief.

Absent express laws to that effect in the Seat, an anti-arbitration injunction does
not sit easily with the New York Convention or typical lex arbitri such as the
UNCITRAL Model Law, (235) whether issued by a court in the arbitral seat or
elsewhere. (236) Such an injunction would also be contrary to the Model Law
system establishing judicial control only after an award has been rendered. (237)
Anti-arbitration injunctions also go against the fundamental principle whereby
an arbitral tribunal has the power to decide upon its own jurisdiction. (238)

8.5.3. In What Circumstances Should an Arbitral Tribunal Respect an Anti-

arbitration Injunction?

Born has suggested that ‘… there may be instances in which an arbitral tribunal
concludes that it would be efficient and sensible for arbitral consideration or
resolution of a jurisdictional objection to be suspended or delayed on the basis of
an page "664" earlier-filed, and much-advanced, national court proceeding,
addressing the same jurisdictional objection’. (239) This should depend on the
relative jurisdictional powers of the fora. If it is a supervisory court to which
deference is owed, this may be the most efficient and practical solution.

An arbitral tribunal that believes it has jurisdiction may be justified in flaunting

an anti-arbitration injunction in most circumstances. Several arbitral tribunals
have ignored and proceeded despite an anti-arbitration injunction. (240) The
jurisdiction of the tribunal is derived from the agreement of the parties and a
domestic court does not have the power in the first instance to decide upon the
jurisdiction of the tribunal. ‘The arbitral tribunal's primary duty is owed to the
parties. If it finds that the parties have agreed to resolve their dispute by
arbitration, it must respect this agreement and make an award.’ (241) Even a
negative jurisdictional decision by courts in the arbitral seat should not bind the
arbitral tribunal. (242) The general principle should be so unless the lex arbitri
expressly gives a court temporal priority on jurisdictional questions.
Nevertheless, in other cases it should be understood that an award will probably
be unenforceable where the injunction was granted. (243)
There may still be an issue where the supervisory court is the higher authority.
Because enforceability is possible in relation to annulled awards, a tribunal with
a duty to complete a mandate and operate with due care should at least be
prepared to countenance a circumstance where it concludes that no reasonable
supervisory court could have come to the view that was taken. (244) An extreme
example would be if the tribunal knows that the supervisory court made an
adverse ruling following a bribe. The right to take such a position is an extreme
one and should not be adopted without the clearest of reasons for doing so. An
illustrative case as to problems that may arise is ICC Arbitration No 10623. (245)
The case related to a European contractor in a dispute with an Ethiopian public
authority. The arbitration clause provided for arbitration in Addis Ababa and
rules of arbitration under the Civil Code of Ethiopia. The clause called for ICC
arbitration unless otherwise page "665" specified in the contract. The
respondent argued that the reference to the Civil Code of Ethiopia constituted an
agreement to substitute ad hoc arbitration for ICC arbitration. The ICC
constituted a tribunal under the ICC Rules which chose to conduct hearings in
Paris, a place convenient for all but the Ethiopian party and which in practical
terms prevented the attendance of one of the Ethiopian witnesses. The
respondent challenged the arbitrators. It argued that they had abused their
discretion and failed to act fairly and impartially by making a determination that
clearly preferred one of the parties. The ICC Court rejected the challenge. The
respondent then appealed this decision to an Ethiopian court. It also sought an
order from the court that ICC Arbitration was not agreed. It sought injunctions
enjoining the continuation of the arbitration which were granted. The tribunal
then invited the parties to Paris to discuss the implications of the injunctions.
The respondent refused to attend. The tribunal, in a partial award, decided that
the arbitration should not be suspended and that it had jurisdiction. The parties
then settled the dispute.

The important aspect of the award is that the tribunal considered whether it
must defer to a judicial order to halt the proceedings when that emanated from a
country where the parties had agreed to hold their arbitration. It concluded that
it had a discretion to do so. It concluded that because its mandate comes from
the authority of the parties, and because it has a duty to complete the mandate,
in appropriate situations ‘it may be necessary to decline to comply with an order
issued by the court of the Seat, in the fulfilment of the Tribunal's larger duty to
the parties’. (246) The tribunal also opined that notwithstanding a contrary view of
a supervisory court, a tribunal ‘owes a duty to the parties to ensure that their
agreement to submit disputes to international arbitration is rendered effective
even where that creates a conflict with the courts of the Seat of the arbitration’. A
failure to do so would be ‘a denial of justice and fairness to the parties …’. (247)
Perhaps unfortunately, the tribunal also opined that these principles ‘carry even
more force in an arbitration between a State or State entity and a private party,
where the Seat of the arbitration is in the country of the State involved in the
dispute and it is the State party that is attempting to frustrate the arbitration
agreement’. (248) This is an unfortunate comment given that it acknowledges the
agreement of the parties to make the Seat of the arbitration the country of the
State party. If the challenge is valid then it should succeed. If it is in fact a
challenge ‘attempting to frustrate the arbitration’, then it should fail regardless of
whether it is a private party or a State that brings it.

The tribunal then referred to the accepted view that a State party cannot rely on
its own law to contest the validity of an arbitration agreement. Yet it was not its
own law but the contractual agreement of the parties that was in issue. In a
commentary on his own award, the Chair argued that a court decision of the
place of arbitration should not be accorded more importance than an
enforcement page "666" jurisdiction and that a conclusion to the contrary would
encourage State parties to commit abuses. (249) Schwartz rightly criticises this line
of reasoning on the basis that there needed to be a direct articulated reason for
disregarding an injunction from a supervisory court. (250) As Schwartz notes, a
tribunal cannot derive the power to ignore courts from an arbitration agreement
where that agreement refers to a supervisory court code. (251) Express selection
of a Seat must surely imply an agreement by the parties to be bound by the
supervisory oversight of courts in that jurisdiction. (252) It is problematic
therefore to justify rejection of the view of the supervisory court based on party
autonomy.

In Saipen v. Petrobangla a State-owned entity obtained a court order in the Seat

revoking a tribunal's authority when various procedural matters were decided
against it including in relation to allowing a witness statement from a witness
prevented from attending, determinations that witnesses should be present and
admissions of evidence at various stages. Notwithstanding that the court issued
an anti-suit injunction against the continuance of the arbitration, the tribunal
rendered an award (253) which held Bangladesh liable for an abuse of rights
through the disruption of the arbitration via the court actions.

The key question is then when a tribunal may depart from an order of the
supervisory court. Schwartz argues that arbitration agreements must be
performed by parties in good faith. State parties cannot therefore rely on their
own law to obviate jurisdiction formerly granted. (254) In that sense the award in
ICC Arbitration No 10623 suggests perhaps that a tribunal must consider
whether ‘to follow a decision of the court would conflict fundamentally with the
tribunal's understanding of its duty to the parties’. (255) Again, however, Schwartz
rightly observes that while the principle may be valid, there needs to be an
indication of the criteria as to when it should apply. It cannot simply be a matter
for the tribunal to determine what in its own view is fair and reasonable. At least
where State parties are concerned, Schwartz suggests that the relevant standard
should be determined from customary international law. One such obligation is
to provide page "667" foreigners with ‘a fair and effective system of justice’. (256)
Different authorities use different terminology but Schwartz concludes that the
denial of justice must be manifest, flagrant, clearly improper and discreditable or
arbitrary. (257) Ultimately it is not the blocking of an arbitration per se but doing
so in a manner which is manifestly arbitrary and improper that matters. (258)

The distinction ought to be between good faith and abusive determinations by a

supervisory court. If the decision is in good faith and the supervisory court is
given a higher authority under the lex arbitri, then its view should be respected
simply because of the hierarchy of adjudicatory power. The difficult question is
then when and how an inferior adjudicator is entitled to rule that the behaviour
of a superior body is abusive and can be ignored.

8.6. Pre-Arbitral Referee Procedures

8.6.1. Introduction

Section 8.2 above looked at the general issue of interim measures in arbitration.
One problem area is where urgent measures are needed before a tribunal has
been formed. To simply wait for tribunal formation in the normal course of
events would often be problematic. The events calling for urgent measures will
typically arise concurrently with the dispute. Even where parties are acting in
good faith, many rules set time limits that mean it can take a few months before a
tribunal is appointed. That can be further frustrated where there are challenges
to nominations or refusals to pay a share of advances. (259) It will often be too late
to provide for an effective measure after this time period has elapsed. One
response is to provide for pre-arbitral referee procedures. An alternative is to
provide for expedited tribunal appointment in emergency situations. It would
also be possible to combine the two approaches. (260)

page "668"

While the demand for interim measures prior to appointment or expedited

appointment of a tribunal could be expected, there are a number of problems in
utilising such procedures in an arbitral context. The first requirement is that
there be consent. This means that policy analysis tends to be bound up in
debates as to whether a particular institution should establish discrete pre-
arbitral referee procedures. From a policy perspective, pre-arbitral referee
procedures raise all of the challenges as to the relevance of interim measures to
arbitration with the added concern that it is someone other than the tribunal
making the determination. Some approach this debate from a practical level,
looking at the need in appropriate circumstances; some from a conceptual level
as to the compatibility with the core norms of arbitration, while others simply
decry what they see as the slow but inexorable development of court-like
procedures in arbitration to the latter's suggested detriment.

A further consideration is the criteria by which such decisions ought to be made

and when and why such determinations may be revisited. These issues all need
to be considered against the alternative scenario where a party simply seeks
court assistance for interim measures in support of effective arbitration. There
can be no a priori presumption that one approach will be better than the other in
all circumstances. In a comparative utility sense pre-arbitral processes, no
matter how quick, are unlikely to be heavily used simply because it would take
longer to establish a referee than to seek an immediate urgent injunction from a
relevant court. While there are difficulties in selecting an appropriate court it
also will typically have more coercive powers than a referee. The next issue
relates to enforcement. The important questions are whether pre-arbitral
referee processes are arbitrations themselves, whether orders could constitute
enforceable awards, if so, how orders are to be enforced and what annulment or
enforcement challenges would be permissible.

8.6.2. Elements of Pre-arbitral Procedures

This section outlines some of the key elements of pre-arbitral procedures before
looking at the particular models adopted by various institutions.

8.6.2.1. Consent

There is a design question whether these should be opt out or opt in procedures
notwithstanding that consent is required in all cases. There is a problem either
way. If opt in is required, few people will agree to the provisions as it will
typically be in one party's interest rather than another's. If it is opt out, one
would imagine that it will be utilised in many cases where there was no
subjective understanding that the provisions would apply. It is somewhat
interesting to see a debate between those arguing for opt in or opt out, with the
latter camp at times suggesting that actual consent would be too hard to achieve.
Yet the power must be based on consent through the selection of the rules.
Hence the argument is that expediency would page "669" justify accepting
unintended application of the procedures simply through selection of a broad
system of rules. This is particularly problematic if new rules purport to be
retroactive.

Another legal question is whether the parties could establish a pre-arbitral

referee process by themselves simply through agreement or whether there
needs to be some statutory support for such a process. (261) There seems little
reason why the latter should be so.

8.6.2.2. Criteria

Pre-arbitral rules may need to indicate the selection procedure for the referee,
challenge, replacement, place and language of the hearing and allow for
determinations on process. There will also be issues as to confidentiality, and
relationship with courts and the subsequent tribunal. It is also necessary to
determine the type of measures that may be ordered and their duration.

An important question is what guidance is provided in the rules as to the

relevant standards. These can vary from concern for impending damage to the
applicant to concern for alleged improper behaviour by the respondent. Both
might of course be relevant but here it may be that key factors point in different
directions and are supported by different degrees of proof. A further policy issue
applying generally to interim measures is whether to have very detailed
regulations along the lines of the UNCITRAL Model Law or a more minimalist
approach relying on the discretion of the referee or arbitrator. These challenging
aspects of standards and evidence for interim measures were addressed in
sections 8.2.5 and 8.2.7 and are not repeated here save for one observation as to
the merits analysis. While a contentious aspect in determining interim measures
is whether the applicant may succeed on the merits, such a decision as a
preliminary matter or as a relevant factor should be less contentious with a pre-
arbitral referee where that is a distinct person that is not ultimately going to
make the final merits determination.

A further important question is what should ultimately happen in terms of

damages and costs in the tribunal hearing if the party who obtained an interim
measure was not ultimately successful. The rules may also indicate whether
failure to comply leads to the potential for further damages awards for the
refusal. It is not clear whether a referee would have an astreinte power where
there is noncompliance.

8.6.2.3. Tribunal Review

Another issue is whether requests may be modified during the course of the
proceeding. Some pre-arbitral rules may impose time limits while others will
allow the measure to remain until overturned by a subsequent tribunal. There
will also be page "670" variations as to whether they require a party to disclose
material changes in circumstances, although this could be imposed by a referee
in any event. There are problems with each model. Setting an optimal lapse
period is problematic. If the period is too short, the measure has little meaning. If
the period is too long, it provides the applicant with too much of an advantage
absent subsequent tribunal determination. (262) Conversely, leaving it to a
tribunal to change can lead to delays. There is still a need to consider a lapse
period if and when the applicant does not proceed to the arbitration itself. If it is
too short there may be problems while waiting for the establishment of the
tribunal as the order may lapse when it is still needed. If the period is too long,
this may provide too much power to the applicant and allow it to, in some cases,
force an unsatisfactory settlement on the responding party without the need for
arbitration.

8.6.2.4. Due Process

Due process issues would involve ensuring that there was actual notice to the
responding party and not simply deemed notice under arbitral rules, particularly
where requests can be made before arbitral proceedings are commenced in
some circumstances. There is also the need to give each party a proper
opportunity to be heard, considerations of appropriate evidence, proper
standards of proof where allegations of impropriety or improper tactics are
suggested, including the means by which subjective intent will be determined.
Where due process issues are concerned, it may need to be kept in mind that the
applicant might have had considerable time before bringing an application whilst
tight timeframes give the responding party little time to prepare a detailed
response. (263)

Problems may also arise where there is a challenge to the referee on the grounds
of impartiality, independence or expertise. Rules may have their own challenge
provisions or may be integrated into arbitral rules where the general norms
would apply. Another question is confidentiality vis-à-vis material presented to
the emergency referee and material submittable to the subsequent tribunal.
There may also be concerns about advocates trying to make too much out of
slight differences in language of submissions or testimony between the two.

8.6.2.5. Enforceability

Enforceability is always a key aspect of utility where a measure presumes a

responding party is acting improperly. No matter what is said in individual rules,
this cannot be determinative as to whether a direction is enforceable under
national statutes or international conventions. (264) A Paris Court of Appeal has
considered that a pre-arbitral referee is not acting as an arbitrator and hence the
page "671" decision cannot be subject to annulment. (265) The Court saw it as
contractual in nature. While that might be debatable, the fact that the decision is
temporary or provisional means that it ought not to be challenged as an award in
any event. (266) A contrary view is presented by van den Berg. (267) This issue is
discussed further in section 16.3.3. If it is not an award, and is not made by a
tribunal empowered under the relevant lex arbitri, then enforcement is not
possible and the normal approach to annulment of the decision is also precluded.
A particular lex arbitri can of course seek to set up supervisory court powers if it
saw fit. Enforceability may also depend on whether the decision is taken before
commencement of arbitral proceedings or after. It may also depend on whether
the procedures are integrated into arbitral rules such as is the case with the SCC
or is a separate procedure such as was the case with the pre-2012 ICC Rules. (268)

A further policy issue is whether negative inferences may be drawn if an interim

determination is not honoured and is not otherwise enforceable. While that is
possible, Chapter 10 argues that an adverse inference should only legitimately
arise where the logical conclusion is that undisclosed facts are adverse to the
party's interest. That might logically flow from failure to produce a document but
may not be the case with the rejection of a freezing order over assets needed in
the responding party's day-to-day business. Where a separate referee is
involved, any subsequent tribunal seeking to draw an adverse inference would
also need to consider whether the order was properly made.

8.6.3. Pre-arbitral and Emergency Arbitrator Rules

This section elaborates on some of the longer-standing pre-arbitral regimes. The

first institution to introduce such a procedure was the ICC in 1990. Many other
institutions, including the NAI, the ICDR, the SCC, SIAC, the LCIA and ACICA now
have some form of emergency arbitrator or expedited tribunal establishment
procedure.

page "672"

8.6.3.1. International Chamber of Commerce (ICC)

The ICC brought in Rules for a Pre-Arbitral Referee Procedure (ICCPA Rules) in
1990. The ICCPA Rules as previously drafted, were separate to the general ICC
Rules. At that stage they were opt in, requiring specific consent by the parties to
be bound by them. While the ICC procedures were established as long ago as
1990, it appears that there are only two instances where they were applied in
the first ten years. (269) The ICC Rules 2012 have changed the structure
significantly. Article 29 of the ICC Rules 2012 now expressly refers to the
emergency arbitrator procedure. The Emergency Arbitrator Rules are contained
in Appendix V of the Rules themselves. An application must be made prior to the
Secretariat having transmitted the file to the arbitral tribunal. The emergency
arbitrator's decision is to take the form of an order and the parties undertake to
comply with it. (270) The emergency arbitrator's order does not bind the arbitral
tribunal that may then modify, terminate or annul the order or any modification
to it. (271) The tribunal can decide on any requests or claims in relation to the
emergency arbitrator proceedings including reallocation of costs and any other
related claims. (272) The revised provisions only apply to parties that are
signatories to the arbitration agreement under the rule that is relied on or who
are successors to such signatories. (273) Article 29(6) indicates that the emergency
arbitrator provisions shall not apply if the arbitration agreement was concluded
before the date on which the current ICC Rules came into force, (274) the parties
have agreed to opt out of the provisions, or the parties have agreed to another
pre-arbitral procedure providing for conservatory, interim or similar measures.
Finally, Article 29(7) indicates that the provisions are not intended to prevent a
party seeking urgent measures from a competent judicial authority.

Appendix V Article 1 sets out a number of form requirements including a

requirement for details of the circumstances giving rise to the application, a
statement of the measures sought and the reasons why the applicant needs
urgent interim or conservatory measures that cannot await the constitution of an
arbitral tribunal. The President of the Court is to consider whether the
emergency arbitrator provisions apply to the request in which case the
application is transmitted to the responding party. The President is to appoint an
emergency arbitrator within as short a time as possible which is normally within
two days. (275) Emergency arbitrators are required to be and remain impartial
and independent (276) and must sign a page "673" statement of availability,
impartiality and independence. An emergency arbitrator is barred from acting as
arbitrator in relation to the dispute. (277) Challenges may be made within three
days and will be decided by the Court after each relevant person is given an
opportunity to comment in writing. (278)

General due process norms are contained in Article 5(2). The emergency
arbitrator is required to consider whether he or she has jurisdiction and whether
the application is admissible pursuant to Article 29(1) of the ICC Rules 2012. (279)
An Order must be made no later than fifteen days from the date on which the file
was transmitted to the emergency arbitrator unless the time is extended by the
President. An order can be made subject to such conditions as the emergency
arbitrator thinks fit, including requiring the provision of appropriate security.
(280) There is also power to modify, terminate or annul the order upon a

reasonable request by a party prior to transmission of the file to the arbitral

tribunal. (281) A flat fee of USD 40,000 applied at the time of writing but this can
be increased by the President in appropriate circumstances. (282) The emergency
arbitrator's order is required to fix the costs of these proceedings. The President
is given residual powers and all administrators and the emergency arbitrator are
required to act in the spirit of the Rules and the Appendix. (283) Importantly, and
consistent with the general ICC approach to interim measures, the revised
procedures do not seek to articulate the type of measures that may be imposed
or the criteria for imposition. (284)

8.6.3.2. Stockholm Chamber of Commerce (SCC)

The SCC Arbitration Rules Appendix II provides for an Emergency Arbitrator.

These emergency arbitrator rules are restricted to application prior to the actual
arbitration tribunal being appointed. (285) No standards are set for the application
of the discretion. The rules allow for imposition of security in return for the
measure and also allow a party to seek interim relief from the court. (286) Upon
receiving the page "674" application the Board will seek to appoint an arbitrator
within twenty-four hours, but only if it feels that the SCC has jurisdiction. (287)
These proceedings then may commence in the Seat (as determined by the parties
or by the Board if no agreement is made) in a manner determined by the
arbitrator as he or she sees fit (with special allowance for urgency). (288) These
emergency decisions on interim measures should then be made within five days
of the emergency arbitrator receiving the file, which will be in writing, be dated
and signed by the emergency arbitrator. This is then delivered to each party and
the SCC. (289) Patricia Shaughnessy reports that on the first occasion the
procedures were used, an emergency arbitrator was appointed twelve hours
after the request was made and the decision was rendered within a further ten
days. (290)

These SCC rules contain a specific provision pertaining to the binding effect of
the decision of the emergency arbitrator. Under Appendix II, Article 9, whilst
expressly binding, the decision can be varied or removed where a party applies
to the emergency arbitrator for such course of action. The parties expressly
agree to the emergency decision without delay by accepting the SCC rules. This
decision is final and binding until the order is revoked, or no arbitration is
commenced within thirty days from the decision or the case is not referred to an
arbitral tribunal within ninety days from the decision. It is important to note that
this decision is not binding on the arbitrators in the full proceeding. Expressly
stipulating them to be binding but indicating that they may be revised and do not
bind a contrary decision of a Tribunal, would raise interesting questions as to
enforceability when ordered as an award.

A controversial aspect of the SCC Rules is that the latter purport to be retroactive
in application as the Emergency Arbitrator Rules purport to apply to all SCC
arbitrations commenced after 1 January 2010. An important corollary of the SCC
Rules being opt out and purportedly retroactive is that there will be some
uncertainty where parties have SCC Arbitration agreements pre-dating the new
rules. Here there will need to be a case-by-case determination as to what the
parties truly intended. If they are clear as to whether they wish for the rules in
force at the time of the agreement or the rules in force at the time of the dispute
this could be determinative, although it is arguable that the intent was not as to a
conceptually different and unknown element such as a pre-arbitral referee
process. Patricia Shaughnessy makes the point that pre-arbitral referee rules are
somewhat different in concept to other arbitral rules and could be argued to go
beyond the mere revision of arbitral rules where some would presume intent to
apply updated rules. (291) In other circumstances, the mere fact that the rules
themselves purport page "675" to be the ones applicable in ambiguous
circumstances also leads to differing views. (292)

8.7. Summary or Early Disposition

8.7.1. Introduction

Domestic litigation systems in common law countries, which work slowly

towards an elaborate single hearing, typically provide a mechanism for early
disposition of an unmeritorious claim or defence. In addition, they will typically
allow for immediate judgment in the face of non-appearance by the opposing
party. Because much of the court expenses are borne by the State, such principles
save on costs of the taxpayer as well as allowing for speedy relief. As always,
issues of efficiency have to be tempered with notions of fairness. Where
judgment in default of appearance is concerned, there need to be mechanisms to
allow for cases to be reopened in appropriate circumstances as there may be a
legitimate excuse for non-appearance. Where there is summary judgment of
contested claims before domestic courts, there still needs to be an appropriate
evidentiary basis on which an adjudicator can make such a determination and
still assert that minimum standards of due process were duly afforded.

Where arbitration is concerned, the value of expediency and cost reduction is no

less a relevant factor, but the notion of minimising taxpayer costs is not relevant
as the parties themselves by their arbitration agreement have agreed to bear all
necessary costs, subject to final costs awards by the tribunal. The key policy
challenge is conceptual and leads to the conclusion that arbitration has nothing
akin to judgment in default of appearance. Except where there was express
agreement to such a power, such a notion would be antithetical to arbitral
consent. Even if parties agreed on such a process, there may still be a debate as
to the validity and implications of such an agreement. On the one hand, one
might argue that this should be accepted because the parties have consented. On
the other hand, any purported award in such circumstances is not truly such, as
it was not a considered view on the merits. The view one takes on this issue
should perhaps be consistent with the view one takes about settlement
agreements as awards. (293) Here again a tribunal may not have made its own
independent determination on the facts, but is instead encapsulating the parties'
own agreement. In each case, some might take the view that consent should be
respected and enforceability should flow to give effect to the parties' a priori
good faith intent. Others would take a contrary view based on their conceptual
understanding of the nature of an independent page "676" adjudicated award.
Even in the former case, it makes sense to afford a tribunal a discretion not to
endorse a claim if thought inappropriate.

8.7.2. Early Disposition and Proactivity

Because of these concerns, it is preferable to consider the notion of early

disposition rather than summary disposition where arbitration is concerned.
Before looking at individual rules that seek to set up such processes some
general observations are made as to the way these issues might be dealt with
under broad discretions and proactivity powers. Early disposition could be
fostered through a range of mechanisms from express powers to seek an early
ruling on allegedly unmeritorious claims, to broader expedited proceedings
models, to efficiency norms under general discretionary powers. Even without
rules directing attention to early disposition of gateway issues, such a power is
available to tribunals through general notions of expediency and directions to
adopt procedures suitable to the circumstances of the case. (294) The latter is
simply a subset of proactive arbitration discussed at the outset of Chapter 6.
Whichever model is adopted, the tighter the timeframes and the more restricted
the evidentiary analysis, the more the model selected might be argued to
contradict the essence of an agreement to arbitrate and offend against
mandatory due process norms as to an adequate opportunity to have a fair and
equitable determination on the merits.

It ought to be possible for an arbitrator to use broad directions to concentrate

the parties attention on key issues to come close to such a position, without
offending against due process norms. While due process must always be
considered, it should be acknowledged that an adjudicator will tend to have an
emerging view of the merits of a case. As long as that view is not set in stone and
the tribunal understands how it may properly be challenged, such an
acknowledgment also leads to the corollary that proactive tailoring of gateway
issues should not inherently raise due process concerns. In some cases the claim
will appear so obviously valid without some excuse, that a tight timeframe seems
desirable to promote a speedy resolution. An example would be arbitration of a
claim under a demand guarantee where the defence shows simple ignorance of
such an obligation. Rules that direct the attention of the tribunal to notifying the
parties as to the key issues (295) may even be legitimately used to indicate to the
parties that claimant has presented a tenable case that calls for a timely
response. Early disposition can also overlap with decisions whether to hold a
hearing or to decide on documents alone. Thus Judith Gill suggests that rather
than widespread rule changes, it may be more important to increase focus on use
of existing powers to dispose of issues or claims page "677" at an early stage. (296)
In some cases a party might seek to use expedited procedures as an indirect
means of obtaining early disposition. This is entirely permissible although the
relevant criteria might not be the same. For example, the ‘exceptional urgency’
required for expedited tribunal formation under the LCIA Rules (297) may look
more to the impact on the claimant of delay rather than the potential lack of
merits in the defence.

As is argued throughout, a tribunal is entitled to be proactive and more

controversially, may adopt a proportional approach to the amount of time and
effort as a factor of the amount in dispute and the complexity of the issues. Even
then, a tribunal must allow for sufficient evidence and argument to be confident
on balance that the determination it makes is the correct one. (298)

The Swiss Supreme Court has held:

The arbitral tribunal can refuse to proceed to the taking of evidence without
violating the right to be heard if the evidence offered is not suitable to be the
basis for (the tribunal's) conviction, if the fact to prove is already established, if it
is irrelevant or if the tribunal, in proceeding to an anticipated evaluation of the
evidence, reaches the conclusion that its conviction is already formed and that
the result of the evidentiary measure sought cannot modify its conviction. (299)

8.7.3. Early Disposition Rule Models

Concern for growing expense and dilatory tactics have led some to question
whether litigation style summary proceedings ought to be explored as express
elements of arbitral rules. Leaving aside the case of judgment in default of
appearance, such proceedings still call for an adjudicatory determination but
seek to do so where it can be said that there is no real prospect of the claimant
succeeding or no genuine defence in response to an arguable claim. (300) They
may page "678" also help narrow the issues where there are multiple claims and
some seem farfetched. (301)

It is typically bad regulatory practice to simply look at a problem and presume it

needs a solution. There is certainly a legitimate concern that too many
respondents raise spurious defences and simply seek to delay inevitable
payment of debts or damages or afford other relief. At the same time, claimants
with deep pockets may conceivably engage in abusive tactics with multiple
excessive ambit claims. Nevertheless, any response must consider a range of
factual permutations and evaluate costs and negative effects as well as benefits.
One policy issue is whether early disposition powers might themselves be
tactically abused, hence adversely affecting the cost-benefit analysis. Early
disposition applications which might be brought to simply add to the costs of the
other party or to try and engender a psychological view that claims are
unmeritorious may encourage greater adversarial disposition and add to overall
transaction costs in some cases at least. As noted in the previous section, an
important question is whether summary disposition offends against the consent
basis of adjudicatory arbitration and/or against due process norms. As to the
first, there should be no problem with consent where parties have expressly
agreed to early disposition rules. Where due process is concerned, views may
legitimately differ. Reference may be made to the case of CDC Group where the
argument was rejected that summary disposition was barred by Article 6(1) of
the European Convention on Human Rights dealing with the right to a fair trial.
(302)

A number of arbitral rules have introduced such options without seeking to

undermine due process norms. Article 41(5) of the ICSID Arbitration Rules (as
amended in 2006) provides as follows:

Unless the parties have agreed to another expedited procedure for making
preliminary objections, a party may, not later that 30 days after the constitution
of the tribunal, and in any event before the first session of the tribunal, file an
objection that a claim is manifestly without legal merit. The parties shall specify
as precisely as possible the basis of the objection. The tribunal, after giving the
parties the opportunity to present their observations on the objection, shall, at
its first session or promptly thereafter, notify the parties of its decision on the
objection.

Article 41(6) provides that where a tribunal decides that ‘all claims are
manifestly without legal merit’ it shall render an award in that regard. (303)

page "679"

A more limited power is contained in the 2004 United States Model Bilateral
Investment Treaty. Article 28.4 provides:

Without prejudice to a tribunal's authority to address other objections as a

preliminary question, a tribunal shall address and decide as a preliminary
question any objection by the respondent that, as a matter of law, a claim
submitted is not a claim for which an award in favour of the claimant may be
made under Article 34. (304)

These powers are directed against spurious and ambit claims by investors that
themselves can be very disruptive to the functioning of States.

Commercial rules would be at least as concerned with claims that have no

legitimate defence, although few have included express powers. Rule 18 of the
JAMS Comprehensive Arbitration Rules and Procedures states: ‘an Arbitrator
may permit any party to file a motion for summary disposition of a particular
claim, either by agreement, or by request of one party, with reasonable notice
given to the other party.’ The JAMS International Arbitration Rules Article 20.3
does not repeat the express reference to summary disposition contained in Rule
18. Conversely, an ICC Task Force on Arbitrating Competition Issues concluded
that summary judgment should not become part of the ICC Rules. (305) No such
provision has been included in the 2012 Rules. The ICDR Rules Article 16.3
provides a more general power, but can be used to similar effect. The tribunal
has the power to order parties to focus their presentation on issues ‘the decision
of which could dispose of all or part of the case’.

The Dutch regime is the most elaborate in this regard. The Dutch Code of Civil
Procedure Article 1051(1) indicates that parties may empower a tribunal or a
chair person to render an award in summary proceedings. The law also allows
for an application to a court for a form of summary disposition under Article
1022(2). Such an application may be rejected where the arbitration agreement
allows for such determinations by a tribunal. (306) There have been cases where
the President of the District Court has accepted jurisdiction where it was thought
that referral to a tribunal would be a waste of time. (307) Article 37 of the NAI
Rules indicates that after appointment, a tribunal can at a party's request make
an award in summary arbitral proceedings at any stage. The NAI Rules are only
available where the Seat of Arbitration is in the Netherlands. The Rules indicate
that a provisional decision will not prejudice a final award. (308) This also applies
after proceedings on the merits page "680" have commenced. Where that is not
the case or where the tribunal has not been appointed, Section 4A and Article
42(a) to (o) of the NAI Rules apply. A request is made to the NAI who
subsequently appoints a sole arbitrator. The sole arbitrator may render an
award or determine that the case is not sufficiently urgent or too complex for
such a disposition. Article 42(m) also indicates that such an award does not
influence the proceedings on the merits, hence establishing the powers on a
potentially provisional basis alone.

In terms of enforcement, as long as due process was followed, such an award

should be protected although the provisional nature may mean that it is not
binding. A contrary view is that it is final and binding in the context of its
mandate, notwithstanding that a contradictory award may ensue in other
proceedings. (309) Where such a mechanism concentrates on urgency and is
provisional in nature, it becomes akin to expert determination where the parties
may accept the decision or proceed to a full hearing on the merits. Dutch courts
considering the Kort Ceding procedure on which the NAI Rules are based have
looked at three factors: whether the claimant has shown an urgent need for the
requested relief; whether it has made out a prima facie case on the merits; and
whether the balance in interests point in favour of granting the relief requested.
(310) Thus the principles are similar to interim measures, although they differ

from common law notions as there is no requirement of concurrent proceedings

and the courts are more willing to impose interim relief that may mirror a final
anticipated award. Due process issues may arise where a foreign party is
unfamiliar with the principles behind the Dutch procedures and calls for more
time to allow it to engage and be advised by local counsel. The fact that the relief
is only provisional provides benefits to a successful applicant but adds to overall
transaction cost. There may be more feelings of uncertainty given that the
parties did not select the arbitrator. (311) There are also problems in mixing
notions of interim relief and summary disposition in arbitration, particularly
where the lex arbitri, rules or party agreement has distinct procedures for
interim measures. Furthermore, there are also evidentiary problems if attention
is only given to a prima facie case. Does that mean that only claimant's evidence
and contentions will be considered? Will a generous view be made of the
claimant's facts? In either event, a respondent ought to be able to argue that
there was a denial of due process if it was not able to challenge the evidence or
present conflicting testimony.

Entitlement to summary judgment can also be dependent on wording in lex

arbitri. For example, the previous English Arbitration Act 1975, Section 1 called
for courts to stay actions in the face of arbitration agreements ‘unless satisfied
that… there is not in fact any dispute between the parties with regard to the
matter page "681" agreed to be referred…’. The use of this language to allow
early merits considerations had been supported in Nova (Jersey) Knit Ltd v.
Kanngam Spinnerei GmbH, (312) and was supported more circumspectly in
Channel Tunnel Group Ltd v. Balfour Beatty Construction Ltd? (313) The removal of
the words in the 1996 Act led the Court of Appeal in a majority decision to
consider that the power no longer existed. (314) US courts have held that
arbitrators have jurisdiction to entertain motions for summary disposition
notwithstanding the lack of an express rule to that effect under the Federal
Arbitration Act. (315) Express provisions are now contained in the revised
Uniform Arbitration Act 2000. (316)

Where early disposition rules apply, in each of these scenarios it is necessary to

consider exactly what the rules provide for. In addition, there are important
procedural and evidentiary questions as to the way an adjudicator should go
about dealing with applications that might be made from time to time. Beginning
with the ICSID Rules, Article 41(6) only provides for an award where all claims
are manifestly without legal merit. On one view, determinations that only some
claims are without merit would have to be given in another form other than as an
award. (317) A contrary view might be that there is no power in such
circumstances. (318) The amendment was seen as being introduced to allow for
early dismissal by tribunals ‘of patently unmeritorious claims’. (319) Attention also
needs to be given to what is meant by a claim that is ‘manifestly without legal
merit’ and the extent to which the suggested intent of the drafters in dealing with
patently unmeritorious claims can colour the analysis of the text of the Rules.
Something that is patently or manifestly unmeritorious implies that the tribunal
need not make a contentious finding of fact at the early disposition stage. Hence
the analysis might best be based on undisputed facts where these might
nevertheless show that there is no legal right. (320) It may also apply where there
is no claim even on the assumption that the facts asserted to underlie the
imputed claim are putatively true. If there is no valid page "682" claim even on
that assumption, rejection can confidently be made. The reference to a lack of
‘legal’ merit may also support this interpretation.

The narrower language in the United States Model Bilateral Investment Treaty
extracted above, expresses this approach more clearly. There is still a debate as
to whether the Treaty provision should be limited to determining as a matter of
law whether the type of relief claimed is contemplated in the treaty provisions or
whether it is possible to go further and consider an argument that no award can
be made because the claim must fail under law. (321) Judith Gill rightly observes
that regardless of the ambit of the test, it simply calls for a particular issue to be
resolved as a preliminary matter. Hence it calls for consideration of a matter that
might have been chosen as a gateway issue by a proactive arbitrator in any
event.

There has been some reporting of cases under these models. In December 2010,
two ICSID tribunals dismissed their respective cases under ICSID Arbitration
Rule 41(5). In Global, (322) the claim was for failure to pay for and take delivery of
poultry products. A claim was brought against Ukraine, asserting that officers of
the State were responsible for negotiating the relevant export contracts. The
preliminary challenge was on the basis that the contracts did not constitute an
‘investment’ under the relevant bilateral investment treaty. The tribunal was
concerned to elaborate carefully on the nature of the rule and the appropriate
procedures under it. The first issue was whether Rule 41(5) should be available
for jurisdictional challenges. The tribunal saw no reason why this should not be
so. A previous tribunal in Brandes Investment Partners, LP v. Bolivarian Republic
of Venezuela also concluded that preliminary objections as to jurisdiction were
permitted, although it rejected the application before it. (323) In terms of the
applicable procedure, the tribunal conceded that each party should be afforded a
proper opportunity to be heard both in writing and orally. There were two
rounds of written submissions and two rounds of oral argument held. In that
regard the tribunal also noted that before one could confidently conclude that a
claim was manifestly without legal merit, all relevant materials must be
considered by the tribunal before reaching a decision. In the instant case that
meant coming to a determination that it could not see what further relevant
materials would alter the scenario. In terms of the relevant standard required by
the word ‘manifestly’, the tribunal supported the comments of the tribunal in
Trans-Global Petroleum Inc page "683" that ‘the ordinary meaning of the word
requires the respondent to establish its objection clearly and obviously, with
relative ease and dispatch’. (324)

The tribunal concluded that under no reasonable process of interpretation could

the poultry contract or the monies expended be construed to be ‘investments’.
The language is interesting as it is not simply saying that on balance it is not an
investment, but instead, that it could find no reasonable argument to this effect.
This is more akin to a judicial review standard. A tribunal will always feel safer in
summarily dismissing a claim when it can say that there is no tenable argument
to support it. Where instead there is a tenable but non-preferred argument in
support of the claim, it would be more appropriate to simply order the
consideration of the issues in a timely fashion such that a preliminary
determination can be made under the appropriate standard of proof without
needing to separately prove that the claim is manifestly unmeritorious. While
rejecting the claim, the tribunal chose not to give a ruling on costs ‘given the
newness of the Rule 41(5) procedure and given the reasonable nature of the
arguments concisely presented to it by both parties …’. (325) It is not fully clear
why a failed argument opposing a proposition that claims are manifestly without
legal merit can be described as reasonable in the context of a costs award. The
conclusion might be better understood within the more reluctant cost tendencies
that have applied from time to time in investment arbitration.

A more far-reaching use of a summary disposition power under the more general
principles of the ICDR Rules Article 16.3 occurred in Matthew v. Papua New
Guinea. (326) In that case an arbitrator dismissed a claim partially based on
quantum meruit because the evidence submitted was wholly insufficient under
the governing law to allow for such a claim. A district court confirmed the award.
The difficulty with adopting this approach on a widespread basis is that it
connotes a ‘one-shot’ approach to the provision of evidence which is commonly
not the norm in arbitration. This is also affected by questions of proactivity and
document production. At times a tribunal will ask for elaboration on the
evidentiary basis of contentions. At other times the relevant evidence will arise
because of a document production request. To deal with such important
evidentiary questions on summary disposition would be particularly
contentious, although the facts of the case may have justified the particular
outcome that arose. (327)

page "684"

What is practically permissible may also be dependent on challenge rights. For

example, the broader rights to challenge for serious irregularity under Section 68
of the English Arbitration Act 1996 make decisions in that jurisdiction at least
arguably more susceptible. Nevertheless, it is unlikely to differ in practice as the
issues are due process matters which could either be described as substantial
injustice or more generally considered under New York Convention style tests.
Where challenges to enforcement are made, the issue will be whether the
arbitrator has failed to hear relevant argument or evidence. The issue will be as
to the comprehensive nature of the material considered and not its form, as there
is no inherent right to oral hearings in arbitration under the lex arbitri or rules so
provide. On the one hand an enforcement court should not interfere readily as
the nature of its powers is not to review evidentiary findings. On the other hand,
given that the evidentiary finding is that a full hearing on the merits is
unnecessary, it is an evidentiary finding as to the ambit of due process and hence
should naturally be within the purview of an enforcement court.

8.7.4. Summary Dismissal and Res Judicata

Claims for summary dismissal will in some cases be based on res judicata nations
rather than lack of intrinsic merit per se. This arose in RSM. (328) Just as the
tribunal may take a favourable view of the claimant's factual allegations in
deciding whether they are manifestly without legal merit, the tribunal in RSM
also believed that the Request for Arbitration should be construed liberally with
doubt or uncertainty as to scope resolved in the claimant's favour in dealing with
such questions of coverage. (329) The Tribunal relied on Article 53 of the ICSID
Convention and collateral estoppel to uphold the objections and find that each of
the claims was manifestly without legal merit. (330) The RSM Tribunal was
prepared to award costs because of its conclusion and because of its view that
the claimant should not attempt to revisit previous decisions by re-
characterising the claims. (331)

8.8. Lis Pendens, Res Judicata and Issue Estoppel

8.8.1. Introduction and Policy Issues

The intended exclusive effect of an agreement between parties to resolve a

dispute by way of arbitration does not always prevent a party from issuing
proceedings in page "685" another forum (whether in a court or before another
tribunal). This results in parallel proceedings, occurring between a national
court and a tribunal, or between two tribunals. A related problem is where a
party might seek to retry matters already determined in previous proceedings.
Section 7.11 looked at whether two arbitral proceedings may be consolidated.
Section 7.13 looked at issues of coordination where this is not possible. Section
8.4 looked at the related question of whether one adjudicator can render an anti-
suit injunction as against the other. This section is concerned with the potentially
limiting effects of one case on another. It is included in this chapter as it can be
another means by which a dispositive determination is made. There are two key
questions. First, can the mere existence of the first proceedings be a barrier to
the commencement of the second? Secondly, are any determinations from a prior
adjudication binding on the parties or on the second adjudicator?
Legal systems are naturally concerned to prevent unnecessary parallel
proceedings or a retrial of former determinations. (332) These can involve
unnecessary costs, potentially conflicting decisions and might be misused as
abusive adjudicatory tactics. (333) Where domestic courts are concerned, attitudes
to multiple proceedings are based on efficiency, seeking to minimise duplication,
bring finality to disputes and avoid inconsistent judgments. Fairness
considerations are more complex, simply because they require careful
limitations to the tests applied. Where there is an allegation that a new case is in
part dealing with matters already decided, fairness would suggest that once a
case is actually decided it should not be relitigated, whilst matters that are of
sufficient difference whether as to person or claims, should be allowed,
otherwise justice is denied. It is difficult to devise tests to differentiate between
actions properly considered as distinct, particularly at preliminary stages.

Policy considerations not only need to deal with actual duplication but also
potential duplication. In this sense a further complex fairness and efficiency
question is whether preclusion should also apply to matters which should have
been addressed in the first proceedings but were not. If relevant, there may then
be a need to consider whether the person who failed to do so was culpable or
instead, whether new evidence subsequently came to light that made the
argument viable.

Parallel proceedings between two courts will typically involve each having prima
facie jurisdiction under national rules, with the question then being which should
defer to the other. Where courts in different countries claim jurisdiction over
similar matters there is also the policy issue of intergovernmental relationships
and questions of deference and comity. Because litigation infrastructure costs
are largely borne by the State in most systems and because of the importance of
intergovernmental relations, many countries would see these principles being
matters of public policy where litigation is concerned. Legal systems have used a
page "686" number of principles to deal with these concerns, namely lis alibi
pendens, res judicata, and in some legal systems, issue estoppel. Some legal
systems also consider questions of former recovery and abuse of process. Lis
alibi pendens or more commonly, lis pendens, is a situation in which ‘parallel
proceedings, involving the same parties and the same cause of action, are
continuing in two different states at two different times’. (334) Not only are there
fairness and efficiency tensions in the problems arising but so too do we find
divergences in solutions between systems opting for a more certain solution and
systems purporting to opt for a fairer case-by-case analysis. Hence where lis
pendens is concerned, the civilian tradition was to give priority to the first case
in time, a preference for certainty, while many common law jurisdictions utilised
the more uncertain principle forum non conveniens, looking for the most
appropriate forum. (335)

Res judicata is the general doctrine that an earlier and final adjudication by a
court or arbitral tribunal is conclusive in subsequent proceedings involving the
same subject matter or relief, the same legal grounds and the same parties. (336)
As a result, it has the effect of preventing a claimant bringing the same claim or
seeking further relief in a matter that has already been litigated. (337) Issue
estoppel as utilised in the common law tradition prevents a party in subsequent
proceedings from contradicting an issue of fact or law that has already been
distinctly raised and finally decided in earlier proceedings between the same
parties (or their privies). (338) The ‘issue’, being an assertion, whether of fact or of
the legal consequences of facts, must be an essential element in the cause of
action or defence. Not all legal systems apply all of these concepts or do so in
identical ways. Nevertheless, they are all concerned with the policy issues
outlined above.

Lis pendens is a response to parallel contemporaneous proceedings while res

judicata and estoppel are about multiple proceedings on the same topic,
generally which do not overlap in time. Even when parallel proceedings deal
with differing subject matters, there may still be efficiency issues as to how the
two cases should run. Obviously they should not clash in their dates so that
parties cannot be in two places at once. Conversely, they may wish to coordinate
so that they are heard at similar times for greatest convenience to the parties and
their witnesses.

page "687"

Parties may also have contractually agreed to be bound by certain decisions from
other fora. This typically arises with string arbitrations and maritime disputes.
(339) An alternative to joinder or consolidation where subcontracts are concerned

is to separately provide contractual entitlements for the subcontractor to make

submissions on behalf of the contractor in the arbitration, be bound by the
outcome of the arbitration and provide it with access to materials and impose
obligations to comply with relevant time limits. (340) Another alternative to
joinder is to define a subcontractor's rights and obligations by reference to the
main contractor's rights and obligations as against the employer. (341)

Multiple and parallel proceedings are more prevalent where investment cases
are concerned given that there are a number of potential sources of arbitral
jurisdiction. Those sources might also lead to different standing. In some cases
there may be tensions between treaty-based jurisdiction and arbitral agreements
in contracts between the parties.

8.8.2. Policy Issues in Arbitration

Where arbitration is concerned, similar efficiency issues arise as were discussed

in the previous section, but these must be integrated with notions of jurisdiction
and consent. Hence where contemporaneous proceedings are concerned there is
a need to consider whether the second proceedings in time are prima facie
within arbitral jurisdiction and if so, by what legal principle or power can an
arbitrator decline jurisdiction or stay the second proceedings. The answer to this
question will vary depending on whether the first proceedings are in court or
before another tribunal and in either event, whether both proceedings were
validly constituted and if so, under what agreement. Where the first proceeding
is before a court, a tribunal must also be alert to the need to deal with
inappropriate decisions by courts in matters that should never have been before
them. (342) In some cases that can be problematic, simply because there may be
factual allegations as to whether parties waived their arbitral rights through
their actions before that court. It is difficult for the tribunal to pass judgment in
relation to those facts although it must do so if relevant to jurisdiction. Other
issues are whether the parties themselves must raise a question or whether a
tribunal may consider the matter of its own volition. This was considered in the
context of jurisdictional challenges in section 8.1. Complications in arbitration
arise where there are potentially overlapping arbitration clauses leading to the
parallel proceedings. Other complex scenarios are where the parties are not the
same but they are related or controlled entities.

page "688"

While it is important to consider how to resolve problems of duplicate

proceedings, ideally these would be prevented. While this can never be
guaranteed, parties can draft their arbitration agreements in such a way that it
becomes clear to a court that it should not entertain jurisdiction in a particular
matter, and/or there can be adequate provision for consolidation of multi-party
or multiple claim scenarios. Parties can also agree to elements such as issue
estoppel. This can typically apply with string contracts and proceedings where
alleged faults in goods are disputed through a long supply chain.

In most cases, parties have not clearly expressed their intent. There are
problems in finding harmonised solutions given the differing approaches
between legal families at the national level where litigation is concerned. If the
various principles are to apply to arbitration and their content is not seen as
identical between differing legal families, a question then arises as to which rules
should apply. Some would take a conflicts approach while others would simply
argue for application of the lex arbitri and principles of competence-competence.
It is appropriate to also consider questions of res judicata and lis pendens in the
context of the parties' intent and what they have intended by their various
agreements to arbitrate and what can be implied into notions that awards are
final and binding. (343)

The difficulties in this area led the International Law Association to invite its
Arbitration Committee to prepare reports on res judicata and lis pendens. After
these reports had been prepared and presented, the Executive Committee of the
ILA approved recommendations for arbitrators on each topic. (344) The two
concepts are discussed in the following sections where the discussion is framed
by the ILA analysis in order to evaluate it as an optimal model for arbitral
tribunals.

8.9. The Nature of Lis Pendens

8.9.1. Introduction

This section looks at situations where there are other proceedings on foot, but no
binding determination as yet from either. In order to avoid costly litigation and
duplication and/or conflicting judgments, there are several ways in which a
forum faced with a dispute, already filed in a foreign forum/arbitral tribunal can
deal with the parallel proceeding:

(a) decline jurisdiction or suspend (or stay) its own proceedings; or

(b) the forum could seek to restrain the foreign proceedings/arbitration; or
(c) both sets of proceedings could be allowed to proceed.
page "689"

Domestic systems will typically provide a rule as to which takes priority. The
civilian approach is to give priority to the first in time under the principle of lis
pendens. Issues of lis pendens might simply be seen as questions of case
management as they deal with the process and not the outcome. Nevertheless,
where stays of proceedings are granted, or where both proceed, it may lead to
res judicata effects when the first proceedings conclude. Hobér describes lis
pendens as ‘a fundamental principle of procedural fairness and justice which is
normally considered to form part of procedural public policy in most legal
systems’. (345) Some common law systems utilise a forum non conveniens test.
The principles underlying forum non conveniens have gained broader approval
in recent years. (346)

Nevertheless, lis pendens has traditionally been deemed inapplicable to

arbitration, since the effect of a valid arbitration clause is to confer exclusive
jurisdiction on an arbitral tribunal, ensuring that a dispute arising between the
parties is resolved by way of arbitration. If an agreement to arbitrate is present,
it is hard to deny the mandate based on other proceedings.

While most would see the principle as inappropriate for arbitration, there have
been exceptions. Various investment tribunals have considered the principle
without holding it to be a binding norm. (347) In addition, in Fomento de
Construcciónes y Contratas S.A. v. Colón Container Terminal S.A. (348) the Swiss
Supreme Court, using the concept of lis pendens, set aside an arbitral award on
jurisdiction, which was rendered while court proceedings were already pending
in an action in Panama. In that case Colon (a Panamanian company) employed
Fomento (a Spanish company) to construct a port terminal in Panama. The
contract provided for ICC arbitration in Geneva. A dispute arose and Fomento
issued court proceedings against Colon in Panama. Colon challenged the
jurisdiction of the court, on the basis of the arbitration agreement. The Swiss
Court of First Instance rejected the challenge as late. Colon appealed this
decision and filed a Request for Arbitration, without waiting for the decision on
appeal. While the arbitration proceeded, the Court of Appeal in Panama
considered the challenge to be timely and annulled the first instance decision.
Consequently, Fomento appealed to the Supreme Court of Panama. The arbitral
tribunal did not wait for the outcome of the appeal to the page "690" Supreme
Court but ruled in an interim award of 30 November 2000 that it had
jurisdiction. It held that the principle of lis pendens does not apply in the relation
between courts and arbitral tribunals and that neither the applicable rules
governing the arbitration proceedings nor the law applicable to the merits
provided for a mandatory stay. Subsequently, the Supreme Court of Panama
quashed the decision of the Court of Appeal. It found that Colon had not raised
the arbitration defence in time. In Switzerland, Fomento challenged the interim
arbitral award before the Swiss Federal Tribunal on the ground that the
arbitrators had no jurisdiction and had disregarded the principle of lis pendens.
The Federal Tribunal admitted the challenge. The Federal Tribunal held that the
lis pendens provision in Article 9 of the Swiss Code of Private International Law
applies between a foreign court and an arbitral tribunal sitting in Switzerland. As
the foreign court has been seized first (and as the judgment could be enforced in
Switzerland), the arbitral tribunal had to decline jurisdiction. However, because
it had not done so, the award was set aside.

The Swiss legislature has since clarified the position in relation to parallel
proceedings and Article 186 of the Swiss PIL provides that:

[A tribunal] shall decide on its jurisdiction notwithstanding an action on the

same matter between the same parties already pending before a state court or
another arbitral tribunal, unless there are serious reasons to stay the
proceedings. (349)

It appears therefore that Swiss legislators have restored the position that the
principle of kompetenz-kompetenz, prevails over lis pendens.

8.9.2. Lis Pendens and Prior Tribunal Proceedings

The ILA Recommendations opt for a presumption that arbitrators should

continue to decide on jurisdiction where they believe that jurisdiction prima
facie applies notwithstanding prior proceedings. The Recommendations set out
exceptions to this general rule. A tribunal may stay proceedings where the first
tribunal appears to have jurisdiction, and where the interests of the party
opposing stay or termination will not be significantly affected. In such
circumstances a tribunal may stay or terminate. Where it chooses to stay, it
should identify an appropriate period. In doing so, the Committee considered
this to be a question of case management and should be as short as possible.
Because of the inter-relationship with res judicata, a natural time would be until
the first award is rendered. Less certain is whether that should also include a
time period where that award may be open to challenge.

In some circumstances where a tribunal might wish to stay proceedings, a

question may arise as to whether this breaches a duty to complete the mandate.
For example, is there any entitlement to apply principles of comity, staying page
"691" proceedings until related proceedings are determined? In the Pyramids
case, the arbitrators concluded:

When the jurisdiction of two unrelated and independent tribunals extend to the
same dispute … in the interests of international judicial order, either of the
tribunals may, in its discretion and as a matter of comity, decide to stay the
exercise of its jurisdiction pending a decision by the other tribunal… Every court
has inherent powers to stay proceedings when justice so requires … (350)
Decisions to stay proceedings may ultimately be efficient but may be argued to
be problematic in terms of a tribunal's duty of expediency. That should not be a
barrier. The extra time is a cost but the minimisation of effort, and avoidance of
conflicting judgments would generally be benefits outweighing the cost, hence on
balance the tribunal's determination could be presumed reasonable. The
situation would be problematic if the parties are not identical in each
proceeding. (351)

There may also be questions as to which is the first in time given that different
arbitral rules define commencement at different stages. This can even raise
questions as to when service was effected where that is the relevant criterion.

8.9.3. Lis Pendens and Prior Court Proceedings

The dominant view of ILA Committee members was that the question of lis
pendens as between tribunals and courts is again simply a matter of
competence-competence to be resolved by the way arbitral laws and rules
determine which body speaks first or last on questions of jurisdiction. Here a
tribunal may again have to consider questions of waiver in terms of the parties
actions before the other forum and in addition, the likely attitude of its own
supervisory court.

8.10. The Nature of Res Judicata

An arbitral tribunal may be faced with res judicata problems where parties have
previously commenced proceedings in tribunals (352) or courts. While civilian
codes and common law principles will commonly recognise res judicata, this is
not usually the case with lex arbitri or arbitral rules. Little is said in arbitral laws
and rules as to the relevance of res judicata although its applicability can be
inferred and many would see it as a transnational norm of procedure or
procedural public policy. French NCCP Articles 1476 and 1500 codify res
judicata. (353) page "692" Res judicata was accepted as a principle of international
law and a general principle of law in Waste Management v. Mexico. (354) As to
inferral from the laws and rules, provisions such as Article 35(1) of the
UNCITRAL Model Law state that an arbitral award shall be recognised as binding
and therefore indirectly recognise the effect of res judicata if a binding nature is
to have any meaningful impact.

The ILA Interim Report noted the very differing definitions of res judicata under
national laws. It did not purport to provide a common definition. In common law
jurisdictions, the estoppel effect of res judicata broadly falls into two categories:
cause of action estoppel, which prevents either party from relitigating the same
action against the other; and issue estoppel, which prevents a party from
questioning or denying an issue already decided in previous proceedings
between the parties. The latter is treated separately below. Many civil law
jurisdictions only apply res judicata as a cause of action estoppel, and the
estoppel is said to attach only to the dispositive part of the judgment/award and
does not attach to the reasons. (355) Other civilian jurisdictions take a broader
approach. In civil law countries like Switzerland, Germany and Sweden, the
narrow approach to res judicata is applied but less so in France, Belgium, the
Netherlands and in Italy. (356)

Whatever the preferred view from a policy perspective, there will obviously be
problems as is commonly found in arbitration where there is no clear guidance
in the lex arbitri and some adopt a transnational approach while others adopt a
conflicts methodology articulated in the lex arbitri. A conflicts approach would
consider whether to apply the res judicata rules of the place of arbitration or the
place where the other decision was made. Under principles of autonomy, the
parties could themselves indicate what res judicata rules the tribunal should
apply, given that it is not properly seen as a matter of mandatory public policy.
Governing law could be affected by party agreement, including an agreement to
apply substantive law to questions of res judicata. Practical problems arise
where the approach of the Seat to res judicata would be narrower than under
such international norms, for example, not seeing the concepts going so far as to
deal with issue preclusion. (357) Tribunals taking the view that res judicata is a
matter of procedure may look to the law of the place of arbitration. (358) In
looking at both the law of the emanating country and that of the Seat of the later
tribunal, limitations page "693" under the former might be applicable in the
latter, for example, principles denying an issue estoppel-type effect. (359) Another
question is whether res judicata should be seen as a question of admissibility or
one of jurisdiction. Jurisdictional questions are typically subject to certain time
limits. Sheppard suggests that the consensus is that res judicata and estoppel are
rules of evidence and hence simply questions of admissibility. (360) The ILA
Recommendation 2 provides that:

the conclusive and preclusive effects of arbitral awards in further arbitral

proceedings set forth below need not necessarily be governed by national law
and may be governed by transnational rules applicable to international
commercial arbitration.

The ILA Report considered that there was widespread acceptance that there are
both positive and negative effects of the res judicata principle. The positive effect
arises where a party raises a prior decision as against another party in
subsequent proceedings. The negative effect is where the prior decision acts as a
defence to prevent the same matter being heard again. While it is generally
accepted that res judicata has both positive and negative effects, it is the latter
effects that become problematic given that they will be raised before a second
adjudicator where there may otherwise be prima facie jurisdiction. The Final
Report considered that the conclusive or positive effect of res judicata was a
matter of substance while the preclusive or negative effect was a matter of
procedure. The practical consequence is that procedural defences generally must
be raised at a time stipulated in the lex arbitri or arbitral rules while raising new
substantive arguments and claims can vary, for example, whether there are
terms of reference as in ICC arbitration and a need for a tribunal to exercise a
discretion to allow new claims.

There will also be differences in view as to how strictly any tests are to be
applied. Typically, tests of res judicata call for an analysis of whether the
previous determination involved the same claims, relief and parties. Some would
take a more technical approach to such a test, while others would look towards
the economic realities. (361) Such questions may also overlap with cases where a
party has taken over previous rights through assignment, subrogation or merger.
Differing identities can typically apply in investment cases as a corporation will
typically own an investment and the shareholdings themselves will typically be
within the broader definitions of investment by reason of the shareholding. In
the CME case, the claims were also brought under different treaties. Courts and
tribunals tend to apply the three-part test quite strictly in terms of same parties,
same subject matter and same relief. The burden of proof should be on the
person relying on such principles as they are properly categorised as defences
and not elements to be proven by a person asserting jurisdiction.

In some cases it may be necessary to consider whether the first decision met
minimum standards of due process as the essence of the res judicata concept
should page "694" not apply where that is not so, although it would be
particularly difficult for a second tribunal to pass judgment in any but the most
extreme cases. Even then there may be questions as to waiver of a right to
challenge when a previous court case was not appealed or a previous arbitration
was not sought to be annulled. Res judicata will not generally apply to an
interpretation of law where a later case decides different issues. This would also
be a limitation of issue estoppel in that regard. A new tribunal is bound to
interpret the law as it sees it. (362) When considering court determinations as to
applicable law, these should not be looked at as precedents per se but instead as
evidence of what the law is.

8.10.1. Res Judicata and Prior Tribunal Decisions

Res judicata might arise between different arbitral tribunals for a range of
reasons. Sheppard outlines some typical scenarios including proceedings under
different alleged agreements in respect of the same relationship, for example,
where there is a battle of the forms; different agreements relating to the same
project; allegations that a prior award did not exhaust all of the differences;
amendments to claims and counterclaims not accepted in the first dispute; and
where the terms of the agreement require separate proceedings. (363)

Because the essential nature of an award is that it is binding on the parties, it

logically follows that the parties promise not to attempt to have the same matter
adjudicated once again. The natural logic is that consent to arbitration should be
interpreted to be consent up until there is a final and binding award on a matter.
Once that occurs, the relevant claims are merged in the decision and no longer
exist as outstanding claims. Where one party is unhappy with the earlier award,
it should respond through the corrective or annulment provisions in the relevant
lex arbitri.

The ILA Recommendations suggest that a tribunal should give res judicata effect
to awards where they are final and binding where rendered, are capable of
recognition at the place where the second arbitration is being conducted, the
award is based on the same cause of action, the award deals with the same
parties, and the award deals with the same claim as before the current tribunal.
The Recommendations require tribunals to consider the likely behaviour of
courts in the current Seat and those in the Seat of the first arbitration. (364) The
ILA Recommendations suggest that where arbitration is concerned, res judicata
is not a matter of public policy and hence both positive and negative effects can
be waived by the parties.

page "695"

These recommendations are the only sensible ones, but they raise complex
problems of one tribunal analysing the jurisdiction of another and bring into
consideration the application of national principles as to validity and recognition.
(365) When considering notions of finality and recognition, the ILA

Recommendations take the view that if the first award is set aside at the place of
arbitration it is no longer valid and capable of recognition or of producing
conclusive and preclusive effects. Thus the Recommendations opt against the
enforceability views of French and US courts in cases such as Norsolor, (366)
Hilmarton (367) and Chromalloy. (368) There is not only a vexed question of what
attitude one should take to the enforceability of annulled awards, but there may
also be a mismatch between jurisdictions that would allow for enforcement and
those which would not. It would make sense to read the Recommendations to
allow the second tribunal to take this into account and ensure that there are not
conflicting enforceable judgments. The need to integrate questions of
applicability of doctrine and likely enforceability are shown by the Hilmarton
dispute where two awards related to a consultancy fee payable by a French
entity to an English one in relation to an Algerian contract. The first award held
that the payment was not permissible under Algerian law even where that was
not chosen by the parties to apply. The award was challenged in Switzerland but
recognised by the Cour de Cassation. A second award was rendered which the
Cour de Cassation in due course held to violate res judicata. The second award
was however enforced in England. (369)

The ILA Recommendations also suggest that claims which could have been
raised in the earlier proceedings but were not, should also be precluded by res
judicata. This becomes a form of estoppel. This is consistent with the common
law abuse of process principle as to matters which should have been raised in
other proceedings. (370) The recommendations allow for exceptions where there
is ‘procedural injustice or abuse’, matters to be left to determination by tribunals
and which were not sought to be further defined. There may be a conceptual
difference between a previous broad claim where particular matters were not
submitted within its parameters and instead where a narrower claim had been
made. As with any estoppel-type arguments, there may be differences in view on
this question.

A number of complex questions were left open in the Recommendations. The

authors raised the question as to whether there should be ‘identity of legal page
"696" orders’ before awards should be given res judicata effect. There are also
uncertain questions as to whether decisions in a commercial arbitration would
have preclusive effect in a case under a bilateral investment treaty. There ought
not be any presumption against. The Interim Report also noted the general
principle of international public law that decisions of national courts do not have
res judicata effect before international adjudicators.

The Recommendations do not purport to define what is an ‘award’ and the

notion of identity of ‘parties’, the latter obviously an issue in multi-party and
related-party scenarios. The identity tests were applied strictly by a tribunal in
the CME case. This arose where actions were brought under different bilateral
investment treaties, in one instance by the investor and in another by its
controlling shareholder. The Svea Court of Appeal concluded that res judicata did
not apply in such circumstances. (371) A number of commentators question
whether a strict application should be justified on policy and practical grounds,
arguing instead for a more commercially realistic and substantive assessment.
(372) When considering the questions of identity of claims, relief and persons, a

substance over form approach should be adopted. A party should not be entitled
to casuistically recast its relief or reasons to overcome the prior determination.
An example would be a party who filed a claim of misrepresentation under
contract law, then sought to bring new proceedings on the same basis in tort.
There would be similar differences in view between those who would adopt a
conflicts approach, looking to the most appropriate res judicata law, or those
who opt for international and commercially efficient solutions. As noted, the
Final Report considered that the conclusive or positive effect of res judicata was
a matter of substance, while the preclusive or negative effect was a matter of
procedure which would obviously impact upon a conflicts approach.

Res judicata also applies where a partial award has been rendered by a tribunal.
Because all true awards are final on a particular issue, such a decision should be
final and binding on all concerned, including the tribunal. There would need to
be some justification for revisiting that determination. An example where
tribunals have considered that they may reopen final determinations as awards
is where there is corruption or fraud. (373) Another example of an exception to res
judicata may be where actual bias of an arbitrator was subsequently shown. (374)
If such an earlier page "697" decision was only provisional, it would have no
preclusive effect in any event. (375) Res judicata does not cover procedural
matters although there may be separate arguments as to when it would be
improper for a tribunal to revise earlier and fundamental procedural
determinations.

Res judicata generally has effect at the time the judgment or award is rendered.
(376) Some jurisdictions have different requirements, such as registration. As to

the technical question whether there is a prior award that may invigorate res
judicata, attention may need to be given to the date of effectiveness of that award
such as when it is signed and in some cases when registered and served. Even if
that is not as yet the case, that may lead to a lis pendens scenario with a stay
potentially being granted until such time as the award is binding. Even then
there is also a need to consider the law where the first decision was made to
determine whether it is indeed final and binding and made by an appropriate
tribunal. (377) In some cases an award might be said to be final and binding where
only non-controversial mathematical computation is required to make it
completely effective. (378)

The Model Law does not expressly address whether an award can be set aside, or
its enforcement refused if it is inconsistent with an earlier decision in the forum
that is res judicata. To the extent that the grounds for challenge are exhaustive,
this argument would need to fit within one of the categories of challenge. At the
very least, public policy could be argued. Similarly, Article III of the New York
Convention only provides that each Contracting State must recognise and
enforce arbitral awards as binding. While this also impliedly recognises the effect
of res judicata, the Convention does not address what a court facing enforcement
of an arbitral award should do if it is faced with more than one decision
concerning the same dispute. (379) Provisions such as Article III of the New York
Convention requiring recognition of arbitral awards as binding simply deal with
the positive effects. (380) Once a dispute has been decided in arbitration, the valid
arbitral page "698" award will give rise to res judicata and render any
subsequent judgment/award invalid. In such circumstances, the first award will
be enforced and recognition of a judgment or the subsequent award ought to be
refused.

However, the commencement of setting aside proceedings by a party resisting

enforcement may suspend the res judicata status of the arbitral award. Once an
award is set aside, or refused recognition and enforcement on the grounds that
the arbitrator had ruled in the absence of an arbitration agreement or on the
basis of an agreement that was void or had expired, res judicata no longer
attaches to the award rendered.

8.10.2. Res Judicata and Prior Court Decisions

Where a court decision is involved, this would raise jurisdictional questions for
the subsequent arbitral tribunal. An important question is how a court came to
resolve a matter that was subject to an arbitration agreement. There is a
difference between court decisions simply dealing with arbitral or court
jurisdiction and those that purport to entertain the merits.

In some cases a court may sometimes declare itself competent to hear a dispute
in breach of an arbitration agreement. In those circumstances, a judgment on the
merits will be issued by the court. This may result in a party commencing
arbitration in another country because a tribunal will not be bound by a
judgment rendered outside the seat. In such circumstances, the tribunal will have
full competence to determine its jurisdiction, unaffected by res judicata. Issues
may arise when seeking to enforce the award in the country where judgment
from a court has been rendered.

While the behaviour of the parties in the litigation is relevant, it is not the
essential question as to res judicata. If the arbitral parties agreed to that court
action, then it is arguable that they have waived the arbitration agreement, hence
no determination of res judicata per se needs to be made. If the court
proceedings occurred in the face of objections by the party now opposing res
judicata impact, a tribunal's conclusion that it has valid jurisdiction suggests that
the court decision was in error and ought not to apply as res judicata if there is
no identity of dispositive rulings. Hence there seems no logical basis for
inconsistent but valid findings between a court and tribunal. Hence the issue
should simply be dealt with on a jurisdictional basis.

Where the court decision is simply as to jurisdiction, denying such to an arbitral

tribunal, it is again not a question of res judicata but instead, whether the court
has competence to make such a ruling. That should only be so if it is the
supervisory court in the Seat and only then where it is empowered to make a
determination prior to and binding upon the tribunal. In some civilian
jurisdictions it may be that a decision by a foreign court or tribunal on a question
of jurisdiction could be binding where the courts in the Seat would recognise
that page "699" decision. (381) As noted above, however, Swiss legislators
overturned this position when a court had adopted this approach.

8.10.3. Issue Estoppel

This concept of issue estoppel was traditionally seen as a common law notion
which has not been adopted by civil law countries. As noted, this is because the
civil law doctrines of res judicata (from which issue estoppel flows) are more
restrictive in nature. The principle of issue estoppel has been recognised in
international arbitration, at least where related to common law systems. In
Fidelitas Shipping Co Ltd v. V/O Exportchleb, Diplock LJ held: (382)

Issue estoppel applies to arbitration as it does to litigation. The parties, having

chosen the tribunal to determine the disputes between them as to their legal
rights and duties, are bound by the determination of that tribunal on any issue
which is relevant to the decision of any dispute referred to that tribunal.

The English Courts have held that even a partial final award, if valid, can give rise
to issue estoppel. In such cases, the tribunal consequently becomes functus officio
in respect of the issues decided in the partial award. (383) As noted, another way
to describe this scenario is to say the final determination is res judicata. This
seems the better view as it is the entire determination that is sought to be
sheltered from subsequent review. While civilian systems do not recognise issue
estoppel, some consider res judicata more broadly and in ways that could
encompass similar outcomes.

The ILA Recommendations do not deal with issue estoppel discretely but instead
consider it within the context of the res judicata analysis. They opted for a more
expansive notion. The Recommendations endorse the principles underlying issue
estoppel and similar concepts as subsets of the res judicata notion. The
Recommendations consider that decisions on questions of fact and law contained
within the original award and which are essential to its reasoning or conclusions
should also have res judicata effect. This is sometimes described as issue
preclusion or collateral estoppel. The ALI/UNIDROIT Draft Principles of
Transnational Civil Procedure (not directly applicable to arbitration), consider
that ‘issue preclusion, as to an issue of fact or application of law to the facts,
should be applied only to prevent substantial injustice’. (384) It should make sense
to limit issue preclusion to more significant circumstances where behaviour
would challenge notions of good faith, given that it is entirely possible that
parties to a commercial transaction bound by a previous determination would
still wish to raise subsidiary page "700" issues afresh. This is particularly so
under modern notions of proactive and efficient disputes. A party in a long-term
commercial relationship might simply have lost a previous case because it failed
to satisfy its burden. On a future occasion it could naturally wish to learn from
that error and present appropriate evidence, arguing that the original and
reasonable expectations of the parties preferring correct outcomes could be
justified.

It has been argued that because awards are both binding and require reasons,
the reasons themselves are final and binding, effectively implementing principles
of issue estoppel. (385) A contrary argument would be to the effect that the
requirement of reasons is simply to afford the losing party an opportunity to
understand why it lost and consider whether there is a basis for challenge. It is
only the award that is expressly stated to be final and binding. Furthermore,
there may be many instances where it would be difficult to debate the ambit of
the reasons articulated, particularly where a range of alternatives were
considered and where the tribunal might have opined that certain things need
not have been determined but were considered in any event.

8.11. Mediation – Arbitration

8.11.1. Introduction

It is important to distinguish between what alternative dispute resolution (ADR)

specialists describe as a mediator/arbitrator process and the discrete question of
whether an already appointed arbitrator may assist in settlement. The mediator/
arbitrator process is an escalation process where there is first mediation, after
which resort may be had to arbitration if the mediation fails. Under such a
scenario, there is no reason to constrain a separate mediator from all of the
normal tools and processes. In most cases, if an arbitrator is then necessary, a
different person will be appointed. Even if the mediator is appointed as
arbitrator, the parties can waive any concerns with caucusing and other
potentially problematic processes. This book is concerned with a distinct issue,
namely whether an already appointed arbitrator may engage in a mediation
function and if so, on what terms and conditions. It does not address the overall
policy question, but instead looks at optimal processes where lex arbitri, arbitral
rules and/or party agreement allow an arbitrator to exercise a
mediation/conciliation function. It also touches on other means to promote
settlement.

page "701"

8.11.2. Rules and Lex Arbitri

Most rules do not expressly indicate their view on these issues. A positive duty is
contained in some arbitral statutes and institutional rules. (386) Rule systems that
expressly allow for arbitrator involvement in settlement vary in terms of the
directions they give to a tribunal in such circumstances. Most are general, merely
speaking in permissive language in terms of attempting to effectuate a
settlement, bringing the parties together for such purpose, considering the
possibility of settlement and the like. Such directions do not give a clear
indication to a tribunal as to what it may or may not do in such circumstances. A
number of civilian statutes expressly indicate that arbitrators may make
settlement proposals. (387) French NCCP Article 21 calls for an arbitrator ‘to assist
the parties towards a settlement between themselves of their dispute … in a way
similar to that of the judge in a French court’. Article 1043 of the Dutch
Arbitration Act 1986 allows a tribunal to order personal appearance of the
parties in order to attempt a settlement.

Rule 1(4) of the ICSID Rules takes a negative approach, indicating that a person
who has acted as a mediator is disqualified from acting as an arbitrator in the
same dispute. However, ICSID Arbitration Rule 21(2) indicates that a tribunal
may hold a conference at the request of the parties ‘to consider the issues in
dispute with a view to reaching an amicable settlement’. An indirect prohibition
as to certain forms of mediation is contained in Article 5.2 of the LCIA Rules
which prohibits an arbitrator whether before or after appointment advising a
party on the merits or outcome of the dispute. Many institutions provide
separate rules of conciliation. (388) WIPO Arbitration Rules Article 65(a) indicates
that a tribunal may suggest that the parties explore settlement. The WIPO
Mediation Rules Article 13(b)(4) indicates that a mediator may propose
arbitration and may act as sole arbitrator with the express consent of the parties
and may, in such circumstances, take into account information received during
the mediation. While the common law has generally had a greater reluctance to
allow a mixed adjudicator/ mediator role, that is not uniform. That is not
surprising as there are different cultural attitudes to dispute settlement that vary
otherwise than in relation to traditional legal norms. Thus a number of common
law jurisdictions also expressly allow for arbitrators to act as conciliators or
mediators. (389) While people speak of page "702" pro- mediation attitudes in
China, CIETAC Rules also include a requirement of consent and a requirement
not to use information from the conciliation in the arbitral proceedings. (390)

Other regimes leave it to broad proactivity discretions which are more open to
debate. Does the general power to conduct an arbitration by whatever procedure
deemed applicable give power to an arbitrator to engage in a mediation
function? That should not be so as provisions such as Article 19 of the Model Law
are designed to apply within the context of the arbitration itself. General
procedural discretions ought to be seen as broad enough to allow an arbitrator
to order a meeting to discuss possibility of settlement at least. The ICC
Commission on Arbitration Report on Techniques for Controlling Time and Costs
in Arbitration suggests that tribunals may remind parties that they are free to
settle at any stage either via negotiations or through use of ADR techniques. (391)
Section 47 of the UNCITRAL Notes on Organising Arbitral Proceedings, 1996
acknowledges that attitudes differ as to whether a tribunal should bring up the
possibility of settlement. It goes on to state:
Given the divergence of practices in this regard, the arbitral tribunal should only
suggest settlement negotiations with caution. However, it may be opportune for
the arbitral tribunal to schedule the proceedings in a way that might facilitate the
continuation or initiation of settlement negotiations.

The position shifted somewhat by the time of the IBA Guidelines. General
Standard 4(d) states:

An arbitrator may assist the parties in reaching a settlement of the dispute at any
stage of the proceedings. However, before doing so, the arbitrator should receive
an express agreement by the parties that acting in such manner shall not
disqualify the arbitrator from continuing to serve as arbitrator. Such express
agreement should be considered to be an effective waiver of any potential
conflict of interests that may arise from the arbitrator's participation in such
process or from information that the arbitrator may learn in the process. If the
assistance by the arbitrator does not lead to final settlement of the case, the
parties remain bound by their waiver. However, consistent with General
Standard 2(a) and notwithstanding such agreement, the arbitrator shall resign if,
as a consequence of his or her involvement in the settlement process, the
arbitrator develops doubts as to his or her ability to remain impartial or
independent in the future course of the arbitration proceedings.

page "703"

Less clear is what should occur if the arbitrator does not consider that there are
doubts but the parties might well do so.

The Working Party considered that informed consent is the best way to deal with
differences in attitude amongst differing legal systems. (392) A discretion in that
regard is acknowledged more generally by Rule 8 of the IBA Ethics Rules:

Where the parties have so requested, or consented to a suggestion to this effect

by the arbitral tribunal, the tribunal as a whole (or the presiding arbitrator
where appropriate), may make proposals for settlement to both parties
simultaneously, and preferably in the presence of each other.

Rule 8 further states that while any procedure is possible by agreement, ‘the
arbitral tribunal should point out to the parties that it is undesirable that any
arbitrator should discuss settlement terms with a party in the absence of the
other party since this will normally have the result that any arbitrator involved
in such discussions will become disqualified from any future participation in the
arbitration.’

8.11.3. Party Autonomy and Mediation Functions

In any situations of uncertainty, an arbitrator should be guided by the parties'

consent. Settlement discussions may not be begun without the informed consent
of all parties. Autonomy is also relevant in that it is normally the case that
prohibitions against an arbitrator/mediator function can be waived by the
parties. Any such waiver should be based on informed and un-coerced consent.
(393) Informed consent implies a waiver of a challenge to the award if settlement

fails. Informed consent is particularly difficult for a party who has never
experienced mediation. The parties' freedom should be respected and settlement
should not be forced. The notion of coercion can even arise if there is strong
pressure from the arbitrator to allow a mediation function where a party would
not wish to offend someone who may ultimately impose a binding decision upon
them. (394)

While consent is crucial, there are also potential problems if parties seek to
combine functions in their arbitration agreement. Escalation clauses raise
complex interpretation difficulties where there are potential problems or
disagreements in fact as to the conduct of the process ordained as preliminary
matters to arbitration. In drafting an escalation clause there is a need to be clear
as to whether the mediation stage is a necessary precondition to arbitration and
if so whether there are any specific elements that must be conducted by each
party prior to arbitral consent. In a worst-case scenario, defects in drafting may
make the whole arbitration agreement pathological. In some cases, multi-tiered
or escalation clauses might even page "704" indicate that the parties do not wish
the arbitrator to engage in the settlement function. This is when mediation and
negotiation are expressed as preconditions to an arbitrator function. Such
clauses are considered further in section 3.2.1.2.

8.11.4. Protocols and Procedures

There is a difference between whether it is permissible to have the same neutral

act in both capacities and whether it is desirable in an instant case. If it is felt to
be desirable, the next question is what protocols and protections should be
utilised to optimise the dual processes.

The following are a list of suggestions for desired behaviour by an arbitrator

exercising mediation functions. Here it is vital to understand that there will
inevitably be trade-offs between the procedures that would apply in an optimal
independent mediation and the procedures necessary to preserve a future
arbitration if it ensues. A protocol that sought to only concern itself with the
interests of one dispute settlement process would inevitably be significantly
suboptimal in relation to the other. Constraining the neutral to protect
challenges to a subsequent arbitral phase may be unfortunate constraints on
broad mediation practices. While trade-offs per se are not problematic in
complex areas, it is only where a trade-off still leads to a net position where a
common neutral is preferable to discrete processes, that such a situation should
be recommended.

Abramson has suggested a number of protocols as requirements for an

arbitrator/mediator function. These are:

1. The neutral be trained in both processes and strive to maintain the integrity
of each.
2. The neutral should consent to both roles, ensuring adequate expertise, that
the matter is appropriate for settlement and that impartiality would not be
impaired.
3. The neutral must respect party self-determination as to whether to settle
and in particular not take initiatives that may appear coercive.
4. Involve parties with settlement authority, not just counsel although that
may vary depending on the different cultural practices in various
jurisdictions.
5. The parties and the neutral should execute a confidentiality agreement
prohibiting introduction of settlement information not independently
discoverable. (395)
6. The neutral as settlor will not evaluate merits, evidence or reasonableness
of positions although even-handed questions can be used to help parties
understand strengths and weaknesses.
7. Neutrals will not caucus unless the parties agree to an exception. page "705"
8. The parties agree to reconfigure the arbitration panel to suit the settlement
process. A range of permutations are possible such as having the co-
arbitrators act as a mediation team with the presiding arbitrator then being
given sole arbitral power. The converse could also be utilised, with the
presiding arbitrator exercising a casting arbitral vote when necessary to
break an impasse between decisions of the co-arbitrators.
9. The arbitrator is not to be influenced by information revealed in the
settlement process. While a person cannot blank their memory, a proper
approach to a reasoned articulation of legal principles applicable to facts as
found should allow a capable arbitrator to follow this protocol. It is no
different to judges and jurors ignoring testimony struck from the record.
10. Parties agree not to challenge the arbitrator or the award based on such a
combined role. While that is desirable, it would be particularly difficult to
ensure that such a waiver prevented challenges alleging breach of
mandatory norms.
11. Settlement initiatives should not unduly delay the arbitration process.
12. Parties consent to the combined processes.

Collins has also suggested guidelines. He also agrees that there should be consent
and this should be in writing. He also suggests that there be written consent to
the neutral engaging in an arbitral function if necessary in due course. The
mediation procedure should be discussed at the outset. There needs to be a clear
agreement as to whether there may be caucusing, whether confidential
information may be subsequently used, how long the process will take, when it
should take place, what impact on a subsequent arbitral timetable is necessary,
and the charges to be imposed and how the costs are to be borne. (396) If the
arbitration is under the auspices of a particular institution, it may be desirable to
consider the institution's conciliation rules if they exist. At the very least they are
more likely to have a thought out integration mechanism between the two
dispute settlement processes.
8.11.5. Mediation and Multi-member Tribunals

The next question is who should engage in such a mediation function where
there are multi-member tribunals. A range of permutations is possible in such a
scenario. Lalive has pointed to a complex methodology where each party
conferred with a nominated arbitrator after which the party-appointed
nominators reported to the chairperson, then each party conferred with the
opposing nominated arbitrator after which the full tribunal met and then the
presiding arbitrator met with each party. (397) Discussions of a settlement
function might also consider the respective roles of page "706" party-appointed
and neutrally appointed arbitrators. One normally considers the role of a party-
appointed arbitrator in terms of the extent to which they can advocate for their
nominator. Where settlement is concerned, a quite distinct function might be
admirable. The party-appointed arbitrator might help their nominator see the
potential weaknesses in their case to ensure that they give adequate
consideration to a negotiated solution. Such a process might even be less likely to
raise justifiable doubts as to their impartiality if an adjudicated outcome is
required. The party-appointed arbitrator could simply suggest that while they
will certainly deal with the matter on its merits, they wish to warn the party as to
how other arbitrators might take an adverse view. Understanding what adverse
views might be formed is simply a corollary of any obligation of the party-
appointed arbitrator to ensure that their nominator's contentions in that regard
are fully understood. This is acknowledged as contentious, however.

8.11.6. Mediation Methodology and Arbitration

A tribunal can avoid many potential conflicts by limiting mediation to what is

known as ‘interest only’ mediation or simply try and indicate how costs will
outweigh the amount in dispute. Interest-based mediation is where the parties
are told not to concentrate on the legal and factual elements of the dispute as
they are simply the cause of the differing positions that have led to a dispute in
the first place. Interest-based mediation suggests that a better way to adopt a
mediated outcome is to get the parties to concentrate on their long-term
interests and see if a mutually agreeable solution can be achieved in that regard.
An example where this is ideal and where there may be no conflict between the
mediated and arbitrated processes would be where there is an acrimonious
partnership dispute over relatively minor actions but which show an
irretrievable breakdown in an otherwise viable business. Absent empowerment
as amiable compositeur, an arbitral tribunal will concentrate on the alleged
breaches in the context of the rights and duties of the parties but will find it
difficult to fashion a remedy that best allows the parties to move forward in an
optimal manner. In particular, many key issues may not be able to be addressed,
for example, the future conduct of affairs if the breaches are not fundamental. In
a mediation, a sensible mediator will try and have the parties understand that
the historical breaches are relatively insignificant, the costs of adjudication will
far outweigh the potential damages and in any event a legal adjudication will not
be able to resolve the key question as to who, if anyone, will conduct the business
and on what commercial basis the parties will transfer their entitlements
between themselves if they are unwilling to continue to collaborate.
Similarly, in some cases settlement might be aided by an arbitrator proposing a
formula that does not prejudge the outcome but simply offers a more cost
effective methodology. For example, in a complex damages assessment of the
impact on the value of shares in a subsidiary based on expropriation by a host
government, each party might have expert accounting witnesses and the tribunal
might appoint its own expert. A settlement could be on the basis that the parties
agree to be bound by page "707" the quantum assessment of an agreed expert,
obviating the need for a much lengthier and expensive process. An arbitrator
should be careful to ensure that any such settlement formula proposed is not in
itself a prejudgment of the merits.

While it is correct to say that a purely interest-based mediation could alleviate

many of the problems, this is not necessarily giving effect to the arbitrator's
ethical and due care obligations as a mediator. While some mediators would see
interest-based approaches as the only proper form of mutual assistance, others
would invite a mediator to consider a range of techniques as they seem merited
from time to time. While current mediation theory typically urges adoption of
interest-based mediation over a positional approach, at times business people
would actually wish an arbitrator/mediator to give some indication of the
possible merits or at least the parameters within which settlement negotiations
ought reasonably to occur. While that can often have a very beneficial impact on
the likelihood of settlement, the downside is serious as this would most often
compromise the person from continuing in an arbitrator role. To state particular
parameters of damages when there is still further evidence and submissions to
be heard would lead to accusations of prejudging the outcome. Impartiality could
also be challenged on the basis that an arbitrator might feel predisposed to make
a final award within the pre-announced parameters otherwise the parties may
be critical in hindsight of that earlier process. Conversely, for an arbitrator to
avoid such techniques simply to preserve their own potential position as a future
arbitrator can raise a conflict of interest. (398) Even in a purely interest-based
mediation, the parties will vent their views on the merits for some time at least,
although a good mediator will quickly invite them to consider a different
orientation. Nevertheless, it is hard to prevent parties from making statements in
a mediation that could be seen as problematic if the arbitration continues. Even
opening statements of this nature could be said to compromise a
mediator/arbitrator. (399) Interest-based mediation might also invite a neutral to
explore other issues that might be seen to colour an ultimate adjudicatory
process. For example, in an arbitration about a joint venture dispute, an interest-
based mediation might show that one party is particularly concerned at the lack
of respect they receive from the other. Allowing such issues to be explored could
provide irrelevant and potentially prejudicial background information if an
adjudicated outcome was ultimately required. (400)

8.11.7. Caucusing and Reality Testing

There is no sense in encouraging arbitrators to adopt the mediation function if it

cannot be justified on efficiency grounds. The more an arbitrator/mediator acts
in a page "708" way which compromises them if a settlement is not achieved, the
less efficient their involvement proves to be. Hence the presumption should
always be that an arbitrator does nothing during a mediation that would
compromise them if called upon to render an award in due course. Secondly, no
matter how much that might limit an arbitrator as compared to a fully
independent mediator, there is nothing to stop the parties utilising the services
of such a person either instead of the arbitrator or in conjunction with the
arbitrator/mediator function. One of the most contentious aspects of an
arbitrator/mediator function is whether the person should be entitled to caucus
with individual parties. That is the norm in pure mediations and has obvious
benefits. The parties can be candid as can the arbitrator. They can vent their
emotions without upsetting the other party, helping to calm them down before
seeking agreement. The mediator can discuss pros and cons of settlement offers,
again in confidence. Subject to the comments above about interest-based
mediation, a mediator will often consider confidential material during caucusing
and will at times try and help parties understand the weaknesses in their case to
try and shift their thinking about the parameters of settlement.

Nevertheless, such caucusing is antithetical to the norms of arbitration and ought

not to be encouraged absent clear consent from the parties, even if empowered
by the lex arbitri. It would go against the duty to communicate openly and
equally. This is the preferable view for a range of reasons. Giving someone an
indication of the weaknesses of the case in a mediation could be argued to be a
prejudicial determination by an arbitrator. There is another problem in
arbitrator/ mediators engaging in reality testing. If an arbitrator over-
emphasises weaknesses to make the point, the party may undervalue its own
case and wrongly believe that the arbitrator has a set view. (401) In many
instances, reality testing on an ex parte basis would also be giving that party
advice on how to bolster their case if an arbitration ensues. For example, helping
them to understand that they may not have as strong a set of expert opinions as
the opponent, invites them to seek more.

In other cases reality testing will not appear to be problematic but could still be
so. An example would be a case that depends entirely on one witness's
recollection as to a pre-contractual conversation. Reality testing that does no
more nor less than warn the party that the case depends on how the witness
gives evidence in due course, responds to cross-examination, and is ultimately
believed or not by the tribunal, helps the party understand the difference
between the case as they prepare it and the case as is ultimately perceived by an
adjudicator. Even in that circumstance, some questions become problematic. An
independent mediator might ask a series of questions such as, how articulate is
the witness; is the witness likely to be able to withstand cross-examination; are
they intelligent enough to understand where the questions are leading to via
counsel cross-examining them; will the person cross-examining them be able to
make them angry etc. Answers to these questions in advance of hearing the
witness might appear to colour the arbitrator's page "709" perceptions before
the witness begins to perform their function. If the view was expressed that the
witness is honest but easily flustered when not speaking in their native tongue, a
tribunal might feel compromised whatever it does in the face of very vigorous
and antagonistic cross-examination. If the tribunal tries to tone down the tenor
of the cross-examination, that counsel might feel disadvantaged because the
discussions in the mediation helped predispose the tribunal to protect the
witness. Conversely, if the tribunal does not interfere, the side presenting that
witness may feel aggrieved that a forewarned tribunal did not respond to
relevant information about the witness's attributes.

8.11.8. Evaluative Mediation

Some leading arbitrators suggest that it is not improper for an arbitrator to give
some preliminary thoughts or suggestions on the merits. Such preliminary
thoughts are inevitable as an arbitrator reviews and synthesises documents and
witness statements. Such advice would obviously not indicate who is likely to
win but would instead make comments about the sufficiency of evidence already
presented and the importance to the outcome of which of two conflicting
witnesses are to be preferred. (402) Such behaviour is not conceptually different to
a civilian inquisitorial judge indicating where the key focus should be directed
during the forthcoming stages. Rivkin goes further and suggests that where an
arbitrator would typically have an idea where the decision is headed, a Town
Elder may have provided such suggestions at an early stage. (403) This would be
controversial in a mediation context as the aim is to have an immediate
settlement, whereas in the context of proactive arbitration, the aim is primarily
to concentrate future efforts.

Another controversial question is whether the arbitrator is entitled to make a

settlement proposal? A settlement proposal in the form of a formula or
methodology would be controversial in many instances. Suggesting a figure or
even a range within which settlement is recommended, is a clear indication of
what on that occasion, a tribunal would apply if asked to arbitrate. If discounts
are made for saved legal fees and uncertainty, one could work back to a likely
award figure. If an arbitrator/mediator is entitled to provide such guidance,
there seems no reason to also not allow an express discussion of their
preliminary views. An informed discussion might even then involve the parties
discussing the likelihood or otherwise of the views being changed, what that
would require and the likely costs of such processes. Rivkin suggests that there
are a number of circumstances where assisting settlement would be far less
problematic. Examples include merely indicating where each party has its own
problems in proving its case; where there is page "710" a middle ground that
would be of benefit to both sides and where a written award may be problematic
for both parties. (404)

8.11.9. Confidentiality Protocols in Mediation

Another question is whether there is a conflict between duties of confidentiality

in mediation and the tribunal's duty of due process in arbitration? An arbitrator
viewing confidential information while acting as mediator could have a difficulty
if that information was not provided during the arbitral hearing. Using the
standards of justifiable doubts as to impartiality, an arbitrator who received
confidential information during a mediation process could not alleviate all
possible concerns that the information might influence their later decision. Even
if the parties do not reveal sensitive information, they are likely to be generally
reticent, which will decrease the chances of an effective settlement. At the
extreme, a combined mediation/arbitration function may be abused as a pre-
trial fishing and posturing process. Such reticence or attempts at fishing
expeditions simply do not apply in normal mediations where the mediator has a
distinct identity to that of an ultimate adjudicator.

Where confidential information is concerned, there are two approaches taken in

the rules. One invites the arbitrator to ignore such information. The alternative
requires disclosure to the other party. (405) For example, the Singapore
International Arbitration Act allows for conciliation by an arbitrator by consent.
Where confidential information is obtained and settlement is not achieved,
section 17(3) requires such confidential information to be disclosed insofar as
the information is considered material for the arbitral proceedings. Such a
provision fails to indicate how information is to be dealt with in subsequent
arbitral proceedings. The Hong Kong and Singapore models accept that
mandatory disclosure may discourage frankness but prefer this to trying to have
an arbitrator ignore information and do so in a way which maintains an
appearance of impartiality. (406) At the very least, the parties need to be warned
from the outset as to the disclosure obligations as to material presented during
mediation.

8.11.10. Timing of Mediation or Settlement Processes

Even if an arbitrator is entitled to engage in settlement promotion, questions of

timing may also be important both as to the likelihood of promoting settlement
and also the fairness to each of the parties. For example, if one party has
provided more substantial arguments and evidence than the other, a settlement
process undertaken page "711" at that stage might be adverse to the interests of
the party whose counsel has not engaged as fulsomely in the process. If
recommended too early, the case is not well framed. Furthermore the parties
may worry as to the arbitrator's own motives and willingness to complete the
mandate. Leaving it to a later stage will also have the parties understanding how
the timeframe is likely to occur if arbitration is needed and how delaying tactics
will be dealt with. (407) All of these matters will help place a concrete framework
on the settlement negotiations. Settlement may also be recommended at the
eleventh hour, particularly where it is felt that the remedies to be imposed by an
arbitrator might not satisfy either party.

Kaufmann-Kohler suggests that the ideal time is not so early as to occur when
the arbitrator is not on top of the issues but not so late as to occur when the
parties have spent too much time and resources on the adjudicatory process. (408)
Even this entirely sensible suggestion that an arbitrator might pick the optimal
moment to promote settlement raises a distinction with a function of a mediator.
It would be antithetical to mediation to suggest that if approached, the mediator
might defer the mediation until a later stage. Thus an arbitrator identifying an
optimal moment is really trying to identify the best time within the
mediation/arbitration hybrid and is not promoting mediation alone. Even
timetabling of the mediation processes raises conflicts. On the one hand, an
arbitrator has a mandate to complete the arbitration in as short a time as
possible. On the other, an appropriate timeframe needs to be given if mediation
is to be successful, particularly where key personnel need to come from different
countries and then get instructions from a corporate board of directors.
Timelines are likely to be longer where States are involved.

8.11.11. Extension of Settlement beyond the Arbitration Agreement or Terms

of Reference

Where an adjudicatory function is concerned, an arbitrator is limited to deciding

under clear principles of law unless expressly empowered as amiable
compositeur. A question might arise as to whether an arbitrator not so
empowered may nevertheless suggest settlement terms other than those that
would apply in the adjudication. For example, in the joint venture dispute
referred to above dealing with alleged non-fundamental breaches and damages
claims, a mediated solution might look to the price that one party would pay to
buy out the interests of the other where it is clear that there are irreconcilable
breakdowns in the business relationship.

A further issue is whether an arbitrator as mediator is limited to the same

jurisdictional scope as the arbitration clause. If the arbitration agreement barred
page "712" set-off claims, can these be included in the mediation? Other
possibilities might include an arbitrator becoming a mediator in related
proceedings where joinder is not permissible. This should not be a problem for a
range of reasons. First, regardless of a mediator's authority, a mediated outcome
is an agreement between the parties. If they come to an agreement, this waives
any further right to adjudication. Furthermore, if parties acquiesce in such
discussions, they can be taken to have impliedly extended the
arbitrator's/mediator's powers in that regard. Nevertheless, in those
circumstances where an arbitrator is entitled to perform mediation functions, it
might be best to clarify this issue, and consent should be fully informed consent.

The situation is different if a mediator is asked to render a settlement agreement

as an award. This could only be effective if there is an appropriate agreement to
arbitrate. Once the dispute is settled, any documentation purporting to operate
as an agreement to arbitrate to ensure enforcement of the mediated solution
would run the risk of an argument that it is not in fact a submission to
arbitration. (409)

It is also desirable to consider how the conciliation process affects the ensuing
arbitration. It may have clarified or reduced certain issues. (410) If the tribunal
presented any parameters during the mediation, it should clarify whether its
award will remain within these boundaries. An arbitrator should not be bound
but equally should not mislead the parties in any way.

8.12. Aiding Settlement without Mediation

Most agree that arbitrators are at least entitled to encourage the parties to
consider settlement even if they may not themselves engage as mediators.
Settlement via agreement ‘is of the essence of the spirit of arbitration’. (411) An
arbitrator can aid settlement in a number of ways without consciously taking on
a mediator role. A sensible procedural timeframe can go a long way to
encouraging settlement. Having the parties present key arguments and
documents at an early stage allows each to understand the merits of the
opposing arguments. Setting a timeframe with enough ‘space’ for meaningful
settlement negotiations as the evidentiary picture emerges, can also aid in
settlement. Treating the parties with adequate respect, exploring contentious
aspects of the evidence with polite but probing questions, can all aid in the
ongoing analysis. Except in the most acrimonious disputes, all parties would be
happy to settle all cases as long as they see the terms as being favourable. An
assessment of the value of the case should be an ongoing one as the evidentiary
matrix and factual and legal arguments are better understood. page "713" An
arbitrator who helps the parties hone in on the key strengths and weaknesses of
their respective cases is therefore aiding them in that ongoing assessment. The
terms of reference process can also aid settlement, particularly when the parties
are forced to consider counterclaims and defences alongside their own claims.
(412) When both parties have a reasonably similar assessment, settlement is thus

likely.

Bifurcating proceedings may also assist in settlement. The parties are most
immediately opposed on questions of primary liability as these views will often
be based on fundamentally different views on facts and/or law. While parties of
course may disagree on damages as well, methods of calculation may be more
functional and ripe for settlement processes once liability is known. Arbitrators
can subtly aid settlement simply by ensuring that senior representatives of
corporate parties be present at appropriate stages. An arbitrator can indirectly
suggest the desirability of settlement by pointing out potential costs during
procedural directions and noting the limits of the adjudicatory mandate.

Another option is to have parallel mediation with an independent mediator. A

question then arises as to what the tribunal should do to support the effective
mediation process. Another possibility is to use a ‘shadow mediator’ who can
attend the arbitration hearings and seek to work with the parties in effecting a
settlement. (413) The use of complimentary processes are more problematic
where both parties do not wish to engage in them. Thus if the tribunal believes it
would be desirable to delay the arbitration to facilitate mediation and one party
is not agreeable, the tribunal ought to feel bound by its duty of expediency to
proceed with the arbitration without delay. (414) Reverting to the joint venture
dispute example alluded to above, there would not be a conflict if an arbitrator
sought to explain to the parties that resolution of the allegations of breach will
not help them move forward in terms of their irretrievable business breakdown
and a tribunal might suggest an adjournment to allow the parties to seek an
independent mediator to resolve their entire conflict in a more efficient manner.
Even if a tribunal might not be willing to defer proceedings pending settlement
negotiations or mediation, at the very least, procedural timelines should not be
drawn so that such steps are difficult if not impossible. (415)

page "714"
More contentious would be an indication of the key issues in the mind of the
tribunal although this has been advocated as a possible corollary of proactive
arbitration. For example, a tribunal might simply indicate that the case seems to
turn on which of two conflicting witnesses is to be believed as and when a
hearing is conducted. The parties are made aware of the risks that they can then
factor into their settlement negotiations. In such a fact scenario, a party who
wrongly inflates the value of its case because its legal arguments are strong, is
then reminded that the real case is about disputed facts on which a priori
predictions are more troublesome.

Another approach is to simply say to the parties that the matter ought to be
settled, the arbitration will continue, direct the parties to meet to attempt to
settle and make it clear what the timeframe for rendering an award is. Capable
and sensible business people may take the hint and ensure that an appropriate
settlement is obtained. (416) Another subtle observation is to suggest that while
the parties may wish to pursue their arbitration on a discrete matter, they might
well consider mediation of other broader inter-party issues. Similarly, the parties
might be invited to compartmentalise the broader dispute into those elements
more appropriate for arbitration and those more appropriate for a mediated
solution.

Another possibility is to withhold the award, giving the parties an opportunity

for a final negotiated solution. (417) A natural situation where this ought to arise is
where it is clear to an arbitrator that the parties have fundamental differences
that will not be fully resolved by the arbitration itself. While an arbitrator has a
duty to complete the mandate, the duty of commerciality would at least suggest
that the parties ought to be made aware of an experienced arbitrator's legitimate
concerns. If an arbitrator knew of these problems at the outset, perhaps the
appointment should not be accepted. In many cases, however, this will only
become obvious as the proceedings evolve.

page "715"

1 Chartered Institute of Arbitrators, ‘Guidelines for Arbitrators Dealing with

Jurisdictional Problems in International Cases’, 113 (available from
<www.ciarb.org>).
2 Section 31(1).
3 UNCITRAL Rules 2010 Art. 4.2(a).
4 UNCITRAL Rules 2010 Art. 23.2.
5 There is also a need to consider whether a failure to bring a timely challenge as

to jurisdiction prevents an objection being raised to enforcement or an

annulment challenge to the award itself. This is outside the scope of this book.
6 Chartered Institute of Arbitrators, ‘Guidelines for Arbitrators Dealing with

Jurisdictional Problems in International Cases’, 115 (available from

<www.ciarb.org>).
7 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn
(Oxford: Oxford University Press, 2009), 357, n. 177.
8 See, e.g., s. 73 English Arbitration Act 1996.
9 See Chartered Institute of Arbitrators, ‘Guidelines for Arbitrators Dealing with

Jurisdictional Problems in International Cases’, 116 (available from

<www.ciarb.org>) and ss 4, 16(2) and 206 of the US Federal Arbitration Act.
10 German Arbitration Act 1998 Book 10 ZPO Art. 1032(2).
11 Section 72(1) of the English Arbitration Act 1996 also allows a party to

arbitral proceedings who takes no part in them to challenge jurisdiction by

proceedings in court. For a discussion of this provision and its implications see
Peter Aeberli, ‘Jurisdictional Disputes under the Arbitration Act 1996: A
Procedural Route Map’, Arbitration International 21 no. 3 (2005): 253.
12 ICSID Convention Art. 52.
13 CIETAC Rules 2012 Art. 6.1. However CIETAC has the power to delegate such

a decision to the arbitral tribunal.

14 ICC Rules 2012 Art. 6(4).
15 Ibid.
16 Chartered Institute of Arbitrators, ‘Guidelines for Arbitrators Dealing with

Jurisdictional Problems in International Cases’, 115 (available from

<www.ciarb.org>), which refers to the Award of Judge Lagergren in the
Argentine Bribery case ICC Award No. 1110, Arbitration No. 410/1056.
17 See, e.g., ICC Award No. 2730, 3913 and 3916, Journal du Droit International

(1984): 914, 920 and 930. See Chartered Institute of Arbitrators, ‘Guidelines for
Arbitrators Dealing with Jurisdictional Problems in International Cases’, 115, n. 6
(available from <www.ciarb.org>).
18 Swiss Private International Law Art. 186(3).
19 Chartered Institute of Arbitrators, ‘Guidelines for Arbitrators Dealing with

Jurisdictional Problems in International Cases’, 117 (available at

<www.ciarb.org>).
20 Kaplan v. MK Investments Inc. (US) 115 Supreme Court 1920 (1995); First

Options of Chicago v. Kaplan 514 US 938 (1995).

21 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 350. The need for arbitrators to at times
defer to court judgments was considered in Switzerland in Judgment of 14 May
2001, Fomento de Constructións y Contratas SA v. Colón Container Terminal SA,
Swiss Federal Tribunal, ACF 127 III 279 [2001].
22 See William W. Park, ‘Determining Arbitral Jurisdiction: Allocation of Tasks

between Courts and Arbitrators’, American Review of International Arbitration 8

(1997): 142, n. 23. Park appears to criticise Apollo Computer v. Berg, 886 F. 2d
468 (1st Cir 1989).
23 Dallah Real Estate & Tourism Holding Co. v. Ministry of Religious Affairs,

Government of Pakistan [2010] 3 WLR 1472; Judgment of 17 February 2011,

Cour d'appel de Paris.
24 For commentary on the case, see Jacob Grierson & Mirielle Taok, ‘Conflicting

Judgments from the UK Supreme Court and the Paris Cour d'Appel’, Journal of
International Arbitration 28, no. 4 (2011): 407–422.
25 Chartered Institute of Arbitrators, ‘Guidelines for Arbitrators Dealing with

Jurisdictional Problems in International Cases’, 114 (available at

<www.ciarb.org>). See also Judgment of 19 July 1988, République Arabe d'Egypte
v. Westland Helicopters Ltd, Swiss Federal Supreme Court, ASA Bulletin 7, no. 1
(1989): 55, which dealt with an application to remove the tribunal for bias after a
jurisdictional award had been set aside.
26 Chartered Institute of Arbitrators, ‘Guidelines for Arbitrators Dealing with

Jurisdictional Problems in International Cases’, 115–116, available at

<www.ciarb.org>.
27 See Aleksandar Jaksic, ‘Procedural Guarantees of Human Rights in Arbitration

Proceedings: A Still Unsettled Problem’, Journal of International Arbitration 24,

no. 2 (2007): 165, referring to the European Convention on Human Rights.
28 Naturally, an award without jurisdiction can be subject to challenge. See, e.g.,

English Arbitration Act 1996 s. 67.

29 Simon Greenberg & Matthew Secomb, ‘Terms of Reference and Negative

Jurisdictional Decisions: A Lesson from Australia’, Arbitration International 18,

no. 1 (2002): 132–133.
30 Ibid., 135.
31 Brasoil v. GMRA, Cour d'appel de Paris, Revue de l'Arbitrage (1999): 834;

Publicis v. True North, 206 F. 3d 725.

32 See generally Ali Yesilirmak, Provisional Measures in International Commercial

Arbitration (The Hague: Kluwer Law International, 2005); Luis Enrique Graham,
‘Interim Measures: Ongoing Regulation and Practices (A View from the
UNCITRAL Arbitration Regime)’, in 50 Years of the New York Convention, ICCA
Congress Series No. 14, ed. Albert Jan van den Berg (The Hague: Kluwer Law
International, 2009), 539. James E. Castello, ‘Arbitral Ex Parte Interim Relief’,
Dispute Resolution Journal 58, no. 3 (2003): 60.
33 Van Uden Maritime BV, trading as Van Uden Africa Line v.
Kommanditgesellschaft in Firma Deco-Line [1998] ECR I 7091, 7133.
34 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1943.

35 See, e.g., the fact scenario in Judgment of 29 April 2003, Société Nationale de

Pétroles du Congo SNPC and the Republic of Congo v. Total Fina Elf E & P Congo
(PEP Congo), Cour d'appel de Paris, where a pre-arbitral referee had ordered the
Republic of Congo to continue to deliver oil pursuant to an agreement it was
seeking to terminate.
36 See, e.g., Arthur Marriott, ‘Less Is More: Directing Arbitration Proceedings’,

Arbitration International 16, no. 3 (2000): 356.

37 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 585.
38 See, e.g., reference to a freezing order in the amount of USD 12 billion against

Venezuela's National Oil Company's assets as noted in Edward Poulton & Matt
Totman, ‘Freezing Orders in Support of Arbitration Proceedings – The Case of
Mobil Cerro Negro Ltd v. Petroleos de Venezuela SA’, Mealey's International
Arbitration Report 23, no. 8 (2008).
39 LCIA Rules Art. 9. See also the ‘emergency arbitrator’ provisions in ICDR Rules

Art. 37; SIAC Rules, Schedule 1; ACICA Rules, Schedule 2; SCC Rules, Appendix II.
40 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1945–1946.

41 See ICC Rules 2012 Art. 28(1); LCIA Rules Art. 25; UNCITRAL Rules 2010 Art.

26; ICDR Rules Art. 21; HKIAC Rules Art. 24; SIAC Rules Art. 26; SCC Rules Art.
32; Swiss Rules 2012 Art. 26; ACICA Rules Art. 28.
42 UNCITRAL Model Law Art. 17; French New Code of Civil Procedure Art. 1468.
43 For example, Italy and Argentina. Argentina National Code of Civil Procedure

Art. 753; Italian Code of Civil Procedure Art. 818; Chinese Arbitration Law Art.
68; Quebec Code of Civil Procedure Art. 940(4). See Peter J.W. Sherwin & Douglas
C. Rennie, ‘Interim Relief under International Arbitration Rules and Guidelines: A
Comparative Analysis’, The American Review of International Arbitration 20, no. 3
(2009): 326-327. Born also refers to similar provision in China and Quebec. Gary
B. Born, International Commercial Arbitration (The Hague: Kluwer Law
International, 2009), 1951–1952.
44 UNCITRAL Model Law Art. 17A(1)(a); UNCITRAL Rules 2010 Art. 26.3(a).
45 UNCITRAL Model Law Art. 17A(1)(b); UNCITRAL Rules 2010 Art. 26.3(b).
46 Article 17I(2).
47 See, e.g., Hong Kong Arbitration Ordinance Art. 56.
48 See, e.g., ICC Rules 2012 Art. 28(2) which limits that to a time before the file is

transmitted to the arbitral tribunal although it also said ‘in appropriate

circumstances even thereafter’. It is not clear what that requires and who would
make a determination as to whether these are indeed appropriate circumstances.
49 Article 1056 of the Dutch Code of Civil Procedure allows for such penalties

where a court of law has such a power.

50 See, e.g., Swiss Private International Law Art. 183(2).
51 Final Award in ICC Case No. 7589, ICC International Court of Arbitration

Bulletin 11, no. 1 (2000): 60; Final Award in ICC Case No. 7210, ICC International
Court of Arbitration Bulletin 11, no. 1 (2000): 49. See also Ali Yesilirmak, ‘Interim
and Conservatory Measures in ICC Arbitral Practice’, ICC International Court of
Arbitration Bulletin 11, no. 1 (2000): 32.
52 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 585; Nigel Blackaby et al., Redfern and Hunter on International Arbitration,
5th edn (Oxford: Oxford University Press, 2009), 321-322. Gary B. Born,
International Commercial Arbitration (The Hague: Kluwer Law International,
2009), 1980.
53 UNCITRAL Model Law Art. 17A(1)(a); UNCITRAL Rules 2010 Art. 26.3(a).
54 UNCITRAL Model Law Art. 17A(2); UNCITRAL Rules 2010 Art. 26.4.
55 UNCITRAL Model Law Art. 17A(1)(b); UNCITRAL Rules 2010 Art. 26.3(b).
56 Kaj Hodér, ‘Interim Measures by Arbitrators’, in International Arbitration

2006: Back to Basics?, ICCA Congress Series No. 13, ed. Albert Jan van den Berg
(The Hague: Kluwer Law International, 2007), 735.
57 José Maria Abascal, ‘The Art of Interim Measures’, in International Arbitration

2006: Back to Basics?, ICCA Congress Series No. 13, ed. Albert Jan van den Berg
(The Hague: Kluwer Law International, 2007), 764.
58 Luis Enrique Graham, ‘Interim Measures – Ongoing Regulation and Practices

(A View from the UNCITRAL Arbitration Regime)’, in 50 Years of the New York
Convention, ICCA Congress Series No. 14, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 2009), 552.
59 Ibid., 542.
60 The latter formulation suggests that security ought to be the norm or at least

raises a rebuttable presumption to that effect.

61 ICC Rules 2012 Art. 28(1); SIAC Rules Art. 26; HKIAC Rules Art. 24.1; SCC
Rules Art. 32; LCIA Rules Art. 25. Note, however, that the LCIA Rules appear to
prohibit a party from going to a national court for interim measures after the
tribunal has been formed, unless there are ‘exceptional circumstances’.
62 ACICA Rules Art. 28.3.
63 ICDR Rules Art. 21.1.
64 Hans Van Houtte, ‘Ten Reasons against the Proposal for Ex Parte Interim

Measures of Protection in Arbitration’, Arbitration International 20, no. 1 (2004):

85.
65 Nael G. Bunni, ‘Interim Measures in International Commercial Arbitration: A

Commentary on the Report of Luis Enrique Graham’, in 50 Years of the New York
Convention, ICCA Congress Series No. 14, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 2009), 604–605.
66 See Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th

edn (Oxford: Oxford University Press, 2009), 322–323.

67 Yves Derains, ‘The View against Arbitral Ex Parte Interim Relief: Confirmation

that the Best May Be the Enemy of the Good’, Dispute Resolution Journal 58, no. 3
(2003): 62.
68 Ibid.
69 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2017.

70 Peter J.W. Sherwin & Douglas C. Rennie, ‘Interim Relief under International

Arbitration Rules and Guidelines: A Comparative Analysis’, The American Review

of International Arbitration 20, no. 3 (2009): 365.
71 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 606–607.
72 James E. Castello, ‘Arbitral Ex Parte Interim Relief’, Dispute Resolution Journal

58, no. 3 (2003): 66.

73 They indeed tend to forbid them by prohibiting ex parte communications with

the arbitrator or requiring all communications to be sent to the other parties.

See, e.g., SIAC Rules Arts 10.6 and 16.6; ICDR Rules Art. 7.2; SCC Rules Art. 8.
74 Australian International Arbitration Act s. 18B.
75 UNCITRAL Model Law Art. 17B.
76 UNCITRAL Model Law Art. 17C.
77 Article 17B.
78 Yves Derains, ‘The View against Arbitral Ex Parte Interim Relief: Confirmation

that the Best May Be the Enemy of the Good’, Dispute Resolution Journal 58, no. 3
(2003): 62.
79 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröil, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 585. The latter issue obviously does not apply with the immediate use of
ex parte rights.
80 See, e.g., UNCITRAL Model Law Art. 17; Swiss Private International Law Art.

183(1); HKIAC Rules Art. 24.1; SCC Rules Art. 32(1).

81 ICSID Arbitration Rules Art. 39(3) states: ‘the tribunal may also recommend

provisional measures on its own initiative or recommend measures other than

those specified in a request. It may at any time modify or revoke its
recommendations.’
82 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law
International, 2009), 1991–1992.
83 Stefan Bandel, Einstweiliger Rechtsschutz im Schiedsverfahren (München: C.H.

Beck, 2000), 86. See also Partial Award in ICC Case No. 8113, ICC International
Court of Arbitration Bulletin 11, no. 1 (2000): 69.
84 Ibrahim F.I. Shihata & Antonio R. Parra, ‘The Experience of the International

Centre for Settlement of Investment Disputes’, ICSID Review – Foreign Investment

Law Journal 14 (1999): 326.
85 This is discussed in section 8.12.
86 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1992.

87 Julian Lew, ‘Commentary on Interim and Conservatory Measures in ICC
Arbitration Cases’, ICC International Court of Arbitration Bulletin 11, no. 1 (2000):
23.
88 See Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 601–604. See also Gary B. Born, International Commercial Arbitration
(The Hague: Kluwer Law International, 2009), 1979–1992.
89 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 606–607.
90 Ali Yesilirmak, Provisional Measures in International Commercial Arbitration

(The Hague: Kluwer Law International, 2005), 52–53.

91 This is discussed in section 11.5.
92 UNCITRAL Model Law Art. 17F(2).
93 UNCITRAL Model Law Art. 17F(1).
94 Luis Enrique Graham, ‘Interim Measures – Ongoing Regulation and Practices

(A View from the UNCITRAL Arbitration Regime)’, in 50 Years of the New York
Convention, ICCA Congress Series No. 14, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 2009), 552.
95 José Maria Abascal, ‘The Art of Interim Measures’, in International Arbitration

2006: Back to Basics?, ICCA Congress Series No. 13, ed. Albert Jan van den Berg
(The Hague: Kluwer Law International, 2007), 766.
96 Luis Enrique Graham, ‘Interim Measures – Ongoing Regulation and Practices

(A View from the UNCITRAL Arbitration Regime)’, in 50 Years of the New York
Convention, ICCA Congress Series No. 14, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 2009), 546.
97 For more examples of interim measures that may be ordered, see Luis Enrique

Graham, ‘Interim Measures – Ongoing Regulation and Practices (A View from the
UNCITRAL Arbitration Regime)’, in 50 Years of the New York Convention, ICCA
Congress Series No. 14, ed. Albert Jan van den Berg (The Hague: Kluwer Law
International, 2009), 575.
98 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2007.

99 W. Laurence Craig, William W. Park & Jan Paulsson, International Chamber of

Commerce Arbitration, 3rd edn (Dobbs Ferry: Oceana, 2000), 468–469.

100 Murray Lee Eiland, ‘The Institutional Low in Arbitrating Patent Disputes’,

Dispute Resolution Journal 9 (2009): 315 referring to Sweden.

101 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 607–609.
102 The ICC Rules 2012 are an example. See ICC Rules 2012 Art. 28(1). See also

Julian D.M. Lew, ‘Commentary on Interim and Conservatory Measures in ICC

Arbitration Cases’, ICC International Court of Arbitration Bulletin 11, no. 1 (2000):
28.
103 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2013.

104 For a discussion of conflicting views in American decisions see Publicis

Communications v. True North Connc'ns Inc. 206 F. 3d 725 (7th Cir 2000).
105 JAMS Rules Art. 32.1 and 32.6.
106 See, e.g., Island Creek Coal Sales v. City of Gainesville 729 F. 2d 1046 at 1048.
107 Presumably an institution would ensure that it could scrutinise on an urgent

basis so as not to unduly hold up the process.

108 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 607.
109 Interim Award in ICC Case No. 8879, ICC International Court of Arbitration

Bulletin 11, no. 1 (2000): 88.

110 Georgios Petrochilos, ‘Interim Measures under the Revised UNCITRAL
Arbitration Rules’, ASA Bulletin 28, no. 4 (2010): 885.
111 LCIA Rules Art. 25.1.
112 Peter J.W. Sherwin & Douglas C. Rennie, ‘Interim Relief under International

Arbitration Rules and Guidelines: A Comparative Analysis’, The American Review

of International Arbitration 20, no. 3 (2009): 344.
113 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 607.
114 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1947.

115 Kaj Hodér, ‘Interim Measures by Arbitrators’, in International Arbitration

2006: Back to Basics?, ICCA Congress Series No. 13, ed. Albert Jan van den Berg
(The Hague: Kluwer Law International, 2007), 737.
116 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 606–607.
117 See, e.g., SCC Rules Appendix II Art. 9(4).
118 Article 37(f).
119 Georgios Petrochilos, ‘Interim Measures under the Revised UNCITRAL
Arbitration Rules’, ASA Bulletin 28, no. 4 (2010): 886.
120 UNCITRAL Rules 2010 Art. 26.8.
121 UNCITRAL Model Law Art. 17G.
122 Secretariat Note A/CN.9/WG.II/WP.127, para. 27.
123 ICC Rules 2012 Art. 37(4).
124 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2028.

125 UNCITRAL Rules 2010 Art. 26.9.
126 ICC Rules 2012 Art. 28(2).
127 Article 25.3.
128 Articles 21.3 and 37.8.
129 Article 26.5.
130 Article 32.5.
131 Article 46(d).
132 Article 26.3.
133 ICSID Rules Art. 39(6).
134 Ibid.
135 Article 26.5.
136 Article 25.3.
137 Dabourian Group Int Inc. v. Sims & Ors [2009] EWCA Civ 169.
138 CPLR s. 7502(c).
139 The better view is that it does not replace such criteria. See SJ Cowen

Securities Corp Co. v. Nessih 224 F. 3d 79 (2d Cir). For a recent case attaching US
assets in aid of a foreign arbitration see Sojitz Corp. v. Prithvi Information
Solutions Ltd 2011 N.Y.App.Div. Lexus 1709.
140 A Mareva injunction is an injunction to freeze assets.
141 Kaj Hodér, ‘Interim Measures by Arbitrators’, in International Arbitration

2006: Back to Basics?, ICCA Congress Series No. 13, ed. Albert Jan van den Berg
(The Hague: Kluwer Law International, 2007), 731. A number of US cases have
suggested that interim awards, even though not ‘final’, are to be treated as ‘final’
in nature and thus enforceable by national courts under the New York
Convention. See also Gary B. Born, International Commercial Arbitration (The
Hague: Kluwer Law International, 2009), 2023. English and French courts have
enforced protective measures in Judgment of 7 October 2004, SAOtor
Participations, Emballage 48, Yves Bacques et Michéle Bouvier v. SARI Claryle
(Luxembourg) Holdings 1 and SARL Claryle (Luxembourg) Holdings 2, Cour
d'appel de Paris and John Foster Emmott v. Michael Wilson & Partners [2009]
EWHC 1 (Comm) respectively.
142 Noah Rubins, ‘In God We Trust, All Others Pay Cash: Security for Costs in

International Commercial Arbitration’, American Review of International

Arbitration 11 (2000): 310; Wendy Miles & Duncan Speller, ‘Security for Costs in
International Arbitration – Emerging Consensus or Continuing Difference?’, The
European Arbitration Review (2007): 34; Pierre A Karrer and Marcus Desax,
‘Security for Costs in International Arbitration: Why, When and What If …’, in
Law of International Business and Dispute Settlement in the Twenty-First Century:
liber Amicorum Karl-Heinz Böckstiegel, ed. R. Briner et al. (Cologne: Carl
Heymanns Verlag KG, 2001), 339.
143 Noah Rubins, Ibid., 312; US Supreme Court Rules, 43 (in extraordinary

circumstances, the Court may adjudge double costs: rule 43.7). Other policy
justifications in litigation, such as deterring inadvisable claims, should not apply
to arbitration where consent is paramount.
144 Weixia Gu, ‘Security for Costs in International Commercial Arbitration’,
Journal of International Arbitration 22, no. 3 (2005): 177, referring to the
previous position in Switzerland.
145 Weixia Gu, Ibid., 167.
146 Noah Rubins, ‘In God We Trust, All others Pay Cash: Security for Costs in

International Commercial Arbitration’, American Review of International

Arbitration 11 (2000): 307.
147 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 366.

148 Noah Rubins, ‘In God We Trust, All others Pay Cash: Security for Costs in
International Commercial Arbitration’, American Review of International
Arbitration 11 (2000): 316.
149 In that context the tribunal must give the parties an opportunity to make

submissions in relation to the requested security.

150 Noah Rubins, ‘In God We Trust, All others Pay Cash: Security for Costs in

International Commercial Arbitration’, American Review of International

Arbitration 11 (2000): 320.
151 For national laws, see, e.g., English Arbitration Act 1996 s. 38; Australian

International Arbitration Act s. 23K. For rules, see LCIA Rules Art. 25.2; SIAC
Rules Art. 24(k); ACICA Rules Art. 28.2(e).
152 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 686–687.
153 Report of the 48th Session of the Working Group on Arbitration and
Conciliation, Doc. A/ CN.9/641, Vienna 10–14 September 2007, UNCITRAL.
154 Yves Derains & Eric A. Schwartz, A Guide to the ICC Rules of Arbitration, 2nd

edn (The Hague: Kluwer Law International, 2005), 297.

155 Craig, Park and Paulsson see that as justified by old Art. 23(1) of the ICC

Rules. W. Laurence Craig, William W. Park & Jan Paulsson, International Chamber
of Commerce Arbitration, 3rd edn (Dobbs Ferry: Oceana, 2000), para. 26–05.
156 ICC Case No. 7489, Journal du Droit International 120 (1993): 1078.
157 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 601.
158 Noah Rubins, ‘In God We Trust, All others Pay Cash: Security for Costs in

International Commercial Arbitration’, American Review of International

Arbitration 11 (2000): 327.
159 Sperry Int'l Trade, Inc. v. Government of Israel, 689 F. 2d 301, 306 (2d Cir

1982).
160 Bernhard Berger, ‘Arbitration Practice: Security for Costs: Trends and
Developments in Swiss Arbitral Case Law’, ASA Bulletin 28, no. 1 (2010): 8.
161 Weixia Gu, ‘Security for Costs in International Commercial Arbitration’,
Journal of International Arbitration 22, no. 3 (2005): 185.
162 Alastair Henderson, ‘Security for Costs in Arbitration in Singapore’, Asian

International Arbitration Journal 7, no. 1 (2011): 74.

163 Nael G. Bunni, ‘Interim Measures in International Commercial Arbitration: A

Commentary on the Report by Luis Enrique Graham’, in 50 Years of the New York
Convention, ICCA Congress Series No. 14, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 2009), 598.
164 Nael G. Bunni, ‘Interim Measures in International Commercial Arbitration: A

Commentary on the Report by Luis Enrique Graham’, in 50 Years of the New York
Convention, ICCA Congress Series No. 14, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 2009), 600–601, quoting ICC Case No. 10032.
165 ‘Guidelines for Arbitrators on How to Approach an Application for Security

for Costs, Issued by the Chartered Institute of Arbitrators’, Arbitration 63 (1997):

166; Gary Soo, ‘Securing Costs in Hong Kong Arbitration’, International
Arbitration Law Review 3 (2000): 25; Locknie Hsu, ‘Orders for Security for Costs
in International Arbitration in Singapore’, International Arbitration Law Review 3
(2000): 108; Peter Fitzpatrick, ‘Security for Costs under the Arbitration Act
1996’, International Arbitration Law Review 1 (1998): 139.
166 Alastair Henderson has also suggested some relevant considerations for the

tribunal. See Alastair Henderson, ‘Security for Costs in Arbitration in Singapore’,

Asian International Arbitration Journal 7, no. 1 (2011): 71–73, discussing criteria
with particular regard to Singaporean practice.
167 Costs are discussed further in Chapter 15.
168 Julian D.M. Lew, Loukas A. Mistelis and Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 601.
169 SA Coppee Lavalin NV v. Ken-Ren Chemicals and Fertilisers Ltd [1994] 2 All ER

449.
170 Bank Mellat v. Hellinki Techniki SA [1983] 3 All ER 428.
171 Weixia Gu, ‘Security for Costs in International Commercial Arbitration’,
Journal of International Arbitration 22, no. 3 (2005): 198.
172 Ibid., 169.
173 Ibid., 188.
174 See, e.g., ICC Case No. 7047 (1994), ASA Bulletin 13 (1995): 301. 13 ASA Bull

301 (1995) cited in Weixia Gu, ‘Security for Costs in International Commercial
Arbitration’, Journal of International Arbitration 22, no. 3 (2005): 185.
175 See Bernhard Berger, ‘Arbitration Practice: Security for Costs: Trends and

Developments in Swiss Arbitral Case Law’, ASA Bulletin 28, no. 1 (2010): 12.
176 Alastair Henderson, ‘Security for Costs in Arbitration in Singapore’, Asian

International Arbitration Journal 7, no. 1 (2011): 74.

177 See Bernhard Berger, ‘Arbitration Practice: Security for Costs: Trends and

Developments in Swiss Arbitral Case Law’, ASA Bulletin 28, no. 1 (2010): 12.
178 Noah Rubins, ‘In God We Trust, All others Pay Cash: Security for Costs in

International Commercial Arbitration’, American Review of International

Arbitration 11 (2000): 320.
179 Neil Kaplan QC, ‘Interim Measures – A Practical Experience’, in International

Arbitration 2006: Back to Basics?, ICCA Congress Series No. 13, ed. Albert Jan van
den Berg (The Hague: Kluwer Law International, 2007), 772.
180 Michael O'Reilly, ‘Orders for Security for Costs: From the Arbitrator's
Perspective’, Arbitration 61, no. 4 (1995): 249.
181 Bernhard Berger, ‘Arbitration Practice: Security for Costs: Trends and
Developments in Swiss Arbitral Case Law’, ASA Bulletin 28, no. 1 (2010): 12.
182 Ibid.
183 Nael G. Bunni, ‘Interim Measures in International Commercial Arbitration: A

Commentary on the Report by Luis Enrique Graham’, in 50 Years of the New York
Convention, ICCA Congress Series No. 14, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 2009), 597.
184 Lew, Mistelis and Kröll support such an approach without elaborating on

exactly what those standards should be. See Julian D.M. Lew, Loukas A. Mistelis &
Stefan M. Kröll, Comparative International Commercial Arbitration (The Hague:
Kluwer Law International, 2003), 601.
185 Weixia Gu, ‘Security for Costs in International Commercial Arbitration’,
Journal of International Arbitration 22, no. 3 (2005): 189.
186 Toby Landau, ‘“Arbitral Lifelines”: The Protection of Jurisdiction by
Arbitrators’, in International Arbitration 2006: Back to Basics?, ICCA Congress
Series No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer Law International,
2007), 282–284.
187 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1041–1042; Rahim Moloo, ‘Arbitrators Granting Antisuit

Orders: When Should They and on What Authority?’, Journal of International
Arbitration 26, no. 5 (2009): 675–700.
188 Toby Landau, ‘“Arbitral Lifelines”: The Protection of Jurisdiction by
Arbitrators’ in International Arbitration 2006: Back to Basics?, ICCA Congress
Series No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer Law International,
2007), 284. The European Court of Justice has recently ruled, notoriously, that
anti-suit injunctions issued by courts between European Union Member States in
support of arbitration proceedings are ineffective. It did so based on its
interpretation of a Community Regulation establishing general court jurisdiction.
189 Emmanuel Gaillard, ‘Anti-suit Injunctions Issued by Arbitrators’, in
International Arbitration 2006: Back to Basics?, ICCA Congress Series No. 13, ed.
Albert Jan van den Berg (The Hague: Kluwer Law International, 2007), 262;
Rahim Moloo, ‘Arbitrators Granting Antisuit Orders: When Should They and on
What Authority?’, Journal of International Arbitration 26, no. 5 (2009): 678.
190 Emmanuel Gaillard, ‘Anti-suit Injunctions Issued by Arbitrators’, in
International Arbitration 2006: Back to Basics?, ICCA Congress Series No. 13, ed.
Albert Jan van den Berg (The Hague: Kluwer Law International, 2007), 239
referring to cases such as ICC Award No. 3896 (1982), Journal du Droit
International 110 (1983): 914 and the commentary by Sigvard Jarvin. In Amco
Asia Corporation & Ors v. Republic of Indonesia, ICSID Case No. ARB/81/1 (9
December 1983), the tribunal composed of Pierre Lalive (President), Jacques
Robert and Berthold Goldman, recalled ‘the well established principle of
international arbitration law according to which the parties must abstain from
any action likely to have a prejudicial effect on the execution of the forthcoming
decision and, in general, to refrain from committing any act, whatever its nature,
likely to aggravate or to prolong the dispute.’
191 Toby Landau, ‘“Arbitral Lifelines”: The Protection of Jurisdiction by
Arbitrators’, in International Arbitration 2006: Back to Basics?, ICCA Congress
Series No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer Law International,
2007), 285; Paul Friedland & Kate Brown, ‘A Claim for Monetary Relief for
Breach of Agreement to Arbitrate as a Supplement or Substitute to an Anti-suit
Injunction’, in International Arbitration 2006: Back to Basics?, ICCA Congress
Series No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer Law International,
2007), 267–281; Emmanuel Gaillard, ‘Anti-suit Injunctions Issued by
Arbitrators’, in International Arbitration 2006: Back to Basics?, ICCA Congress
Series No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer Law International,
2007), 241.
192 Laurent Lévy, ‘Anti-suit Injunctions Issued by Arbitrators’, in Anti-Suit
Injunctions in International Arbitration, IAI International Arbitration Series No. 2,
ed. Emmanuel Gaillard (Huntington: JurisNet LLC, 2005), 117.
193 Ibid.; Toby Landau, ‘“Arbitral Lifelines”: The Protection of Jurisdiction by

Arbitrators’, in International Arbitration 2006: Back to Basics?, ICCA Congress

Series No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer Law International,
2007), 285; Emmanuel Gaillard, ‘Anti-suit Injunctions Issued by Arbitrators’, in
International Arbitration 2006: Back to Basics?, ICCA Congress Series No. 13, ed.
Albert Jan van den Berg (The Hague: Kluwer Law International, 2007), 243; Ali
Yesilirmak, Provisional Measures in International Commercial Arbitration (The
Hague: Kluwer Law International, 2005), 211.
194 Rahim Moloo, ‘Arbitrators Granting Antisuit Orders: When Should They and

on What Authority?’, Journal of International Arbitration 26, no. 5 (2009): 694;

Toby Landau, ‘“Arbitral Lifelines”: The Protection of Jurisdiction by Arbitrators’,
in International Arbitration 2006: Back to Basics?, ICCA Congress Series No. 13,
ed. Albert Jan van den Berg (The Hague: Kluwer Law International, 2007), 287-
288; Emmanuel Gaillard, ‘Anti-suit Injunctions Issued by Arbitrators’, in
International Arbitration 2006: Back to Basics?, ICCA Congress Series No. 13, ed.
Albert Jan van den Berg (The Hague: Kluwer Law International, 2007), 241; Ali
Yesilirmak, Provisional Measures in International Commercial Arbitration (The
Hague: Kluwer Law International, 2005), 211; Gary B. Born, International
Commercial Arbitration (The Hague: Kluwer Law International, 2009), 1024.
195 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2010; Rahim Moloo, ‘Arbitrators Granting Antisuit Orders:

When Should They and on What Authority?’, Journal of International Arbitration
26, no. 5 (2009): 694; Toby Landau, ‘“Arbitral Lifelines”: The Protection of
Jurisdiction by Arbitrators’, in International Arbitration 2006: Back to Basics?,
ICCA Congress Series No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer
Law International, 2007), 289.
196 Toby Landau, ‘“Arbitral Lifelines”: The Protection of Jurisdiction by
Arbitrators’, in International Arbitration 2006: Back to Basics?, ICCA Congress
Series No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer Law International,
2007), 285.
197 Rahim Moloo, ‘Arbitrators Granting Antisuit Orders: When Should They and

on What Authority?’, Journal of International Arbitration 26, no. 5 (2009): 695–

696; Toby Landau, ‘“Arbitral Lifelines”: The Protection of Jurisdiction by
Arbitrators’, in International Arbitration 2006: Back to Basics?, ICCA Congress
Series No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer Law International,
2007), 287.
198 Emmanuel Gaillard, ‘Anti-suit Injunctions Issued by Arbitrators’, in
International Arbitration 2006: Back to Basics?, ICCA Congress Series No. 13, ed.
Albert Jan van den Berg (The Hague: Kluwer Law International, 2007), 237;
Rahim Moloo, ‘Arbitrators Granting Antisuit Orders: When Should They and on
What Authority?’, Journal of International Arbitration 26, no. 5 (2009): 679–680.
199 Laurent Lévy, ‘Anti-suit Injunctions Issued by Arbitrators’, in Anti-suit
Injunctions in International Arbitration, IAI International Arbitration Series No. 2,
ed. Emmanuel Gaillard (Huntington: JurisNet LLC, 2005): 120–121; Toby Landau,
‘“Arbitral Lifelines”: The Protection of Jurisdiction by Arbitrators’, in
International Arbitration 2006: Back to Basics?, ICCA Congress Series No. 13, ed.
Albert Jan van den Berg (The Hague: Kluwer Law International, 2007), 289-290;
Emmanuel Gaillard, ‘Anti-suit Injunctions Issued by Arbitrators’, in International
Arbitration 2006: Back to Basics?, ICCA Congress Series No. 13, ed. Albert Jan van
den Berg (The Hague: Kluwer Law International, 2007), 259–261.
200 Rahim Moloo, ‘Arbitrators Granting Antisuit Orders: When Should They and

on What Authority?’, Journal of International Arbitration 26, no. 5 (2009): 682–

684.
201 Ibid., 676.
202 Emmanuel Gaillard, ‘Anti-suit Injunctions Issued by Arbitrators’, in
International Arbitration 2006: Back to Basics?, ICCA Congress Series No. 13, ed.
Albert Jan van den Berg (The Hague: Kluwer Law International, 2007), 259–261.
203 See Moloo's discussion of E-Systems Inc. v. Islamic Republic of Iran et al. in

Rahim Moloo, ‘Arbitrators Granting Antisuit Orders: When Should They and on
What Authority?’, Journal of International Arbitration 26, no. 5 (2009): 679.
204 Emmanuel Gaillard, ‘Anti-suit Injunctions Issued by Arbitrators’, in
International Arbitration 2006: Back to Basics?, ICCA Congress Series No. 13, ed.
Albert Jan van den Berg (The Hague: Kluwer Law International, 2007), 244–259.
205 Rahim Moloo, ‘Arbitrators Granting Antisuit Orders: When Should They and

on What Authority?’, Journal of International Arbitration 26, no. 5 (2009): 676.

206 See Moloo's discussion of Boeing Co. v. Islamic Republic of Iran in Rahim

Moloo, ‘Arbitrators Granting Antisuit Orders: When Should They and on What
Authority?’, Journal of International Arbitration 26, no. 5 (2009): 689.
207 Laurent Lévy, ‘Anti-suit Injunctions Issued by Arbitrators’, in Anti-suit
Injunctions in International Arbitration, IAI International Arbitration Series No. 2,
ed. Emmanuel Gaillard (Huntington: JurisNet LLC, 2005): 126; Gary B. Born,
International Commercial Arbitration (The Hague: Kluwer Law International,
2009), 2011.
208 Laurent Lévy, Ibid.
209 Ibid., 126.
210 Emmanuel Gaillard, ‘Anti-suit Injunctions Issued by Arbitrators’, in
International Arbitration 2006: Back to Basics?, ICCA Congress Series No. 13, ed.
Albert Jan van den Berg (The Hague: Kluwer Law International, 2007), 264–265.
211 Rahim Moloo, ‘Arbitrators Granting Antisuit Orders: When Should They and

on What Authority?’, Journal of International Arbitration 26, no. 5 (2009): 685.

212 Ibid., 680–681, 693.
213 For the view that it is tantamount to ordering specific performance see

Emmanuel Gaillard, ‘Anti-suit Injunctions Issued by Arbitrators’, in International

Arbitration 2006: Back to Basics?, ICCA Congress Series No. 13, ed. Albert Jan van
den Berg (The Hague: Kluwer Law International, 2007), 229.
214 Emmanuel Gaillard, ‘Anti-suit Injunctions Issued by Arbitrators’, in
International Arbitration 2006: Back to Basics?, ICCA Congress Series No. 13, ed.
Albert Jan van den Berg (The Hague: Kluwer Law International, 2007), 264.
215 For a similar view, see Emmanuel Gaillard, ‘Anti-suit Injunctions Issued by

Arbitrators’, in International Arbitration 2006: Back to Basics?, ICCA Congress

Series No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer Law International,
2007), 265–266.
216 See, e.g., Société Générate de Surveillance SA (SGS) v. Islamic Republic of

Pakistan, ICSID Case No. ARB/01/13, Procedural Order No. 2 dated 16 October
2002, ICSID Review-Foreign Investment Law Journal 18, no. 1 (2003): 293.
217 Société Générate de Surveillance SA (SGS) v. Islamic Republic of Pakistan, ICSID

Case No. ARB/01/13, Procedural Order No. 2 dated 16 October 2002, ICSID
Review-Foreign Investment Law Journal 18, no. 1 (2003): 305.
218 Plama Consortium Ltd v. Republic of Bulgaria, ICSID Case No. ARB/03/24

(2005) para. 45.

219 Paul Friedland & Kate Brown, ‘A Claim for Monetary Relief for Breach of

Agreement to Arbitrate as a Supplement or Substitute to an Anti-suit Injunction’,

in International Arbitration 2006: Back to Basics?, ICCA Congress Series No. 13,
ed. Albert Jan van den Berg (The Hague: Kluwer Law International, 2007), 268;
Laurent Lévy, ‘Anti-suit Injunctions Issued by Arbitrators’, in Anti-suit Injunctions
in International Arbitration, IAI International Arbitration Series No. 2, ed.
Emmanuel Gaillard (Huntington: JurisNet LLC, 2005), 126.
220 Laurent Lévy, ‘Anti-suit Injunctions Issued by Arbitrators’, in Anti-suit
Injunctions in International Arbitration, IAI International Arbitration Series No. 2,
ed. Emmanuel Gaillard (Huntington: JurisNet LLC, 2005), 129; Paul Friedland &
Kate Brown, ‘A Claim for Monetary Relief for Breach of Agreement to Arbitrate as
a Supplement or Substitute to an Anti-suit Injunction’, in International
Arbitration 2006: Back to Basics?, ICCA Congress Series No. 13, ed. Albert Jan van
den Berg (The Hague: Kluwer Law International, 2007), 269.
221 Toby Landau, ‘“Arbitral Lifelines”: The Protection of Jurisdiction by
Arbitrators’, in International Arbitration 2006: Back to Basics?, ICCA Congress
Series No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer Law International,
2007), 296.
222 Emmanuel Gaillard, ‘Anti-suit Injunctions Issued by Arbitrators’, in
International Arbitration 2006: Back to Basics?, ICCA Congress Series No. 13, ed.
Albert Jan van den Berg (The Hague: Kluwer Law International, 2007), 264.
223 Luis Enrique Graham, ‘Interim Measures – Ongoing Regulation and Practices

(A View from the UNCITRAL Arbitration Regime)’, in 50 Years of the New York
Convention, ICCA Congress Series No. 14, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 2009), 579.
224 Paul Friedland & Kate Brown, ‘A Claim for Monetary Relief for Breach of

Agreement to Arbitrate as a Supplement or Substitute to an Anti-suit Injunction’,

in International Arbitration 2006: Back to Basics?, ICCA Congress Series No. 13,
ed. Albert Jan van den Berg (The Hague: Kluwer Law International, 2007), 267.
225 Rahim Moloo, ‘Arbitrators Granting Antisuit Orders: When Should They and

on What Authority?’, Journal of International Arbitration 26, no. 5 (2009): 679.

226 Toby Landau, ‘“Arbitral Lifelines”: The Protection of Jurisdiction by
Arbitrators’, in International Arbitration 2006: Back to Basics?, ICCA Congress
Series No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer Law International,
2007), 294.
227 Rahim Moloo, ‘Arbitrators Granting Antisuit Orders: When Should They and

on What Authority?’, Journal of International Arbitration 26, no. 5 (2009): 698.

228 Toby Landau, ‘“Arbitral Lifelines”: The Protection of Jurisdiction by
Arbitrators’, in International Arbitration 2006: Back to Basics?, ICCA Congress
Series No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer Law International,
2007), 294.
229 Rahim Moloo, ‘Arbitrators Granting Antisuit Orders: When Should They and

on What Authority?’, Journal of International Arbitration 26, no. 5 (2009): 698.

230 Neville Byford & Aszalah Farwar, ‘Arbitration Clauses After West Tankers:

The Unanswerable Conundrum? Practical Solutions for Enforcing Arbitration

Clauses’, International Arbitration Law Review 12, no. 3 (2009): 29.
231 Julian D.M. Lew, ‘Control of Jurisdiction by Injunctions Issued by National

Courts’, in International Arbitration 2006: Back to Basics?, ICCA Congress Series

No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer Law International, 2007),
186–187.
232 There is also the question as to whether one court can render an anti-suit

injunction against another court in support of arbitration. This is beyond the

scope of this book although note should be taken of the controversial decision of
the European Court of Justice in Allianz SpA and Ors v. West Tankers Inc. [2009]
EUECJ C-185/07 holding that such an injunction violated Brussels I, European
Council (EC) Regulation No. 44/2001.
233 Julian D.M. Lew, ‘Control of Jurisdiction by Injunctions Issued by National

Courts’, in International Arbitration 2006: Back to Basics?, ICCA Congress Series

No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer Law International, 2007),
190–201; Gary B. Born, International Commercial Arbitration (The Hague:
Kluwer Law International, 2009), 1049–1050.
234 Julian D.M. Lew, ‘Control of Jurisdiction by Injunctions Issued by National

Courts’, in International Arbitration 2006: Back to Basics?, ICCA Congress Series

No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer Law International, 2007),
189.
235 Marco Stacher, ‘You Don't Want to Go There – Antisuit Injunctions in
International Commercial Arbitration’, ASA Bulletin 23, no. 4 (2005): 652–653.
236 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1053.

237 Julian D.M. Lew, ‘Control of Jurisdiction by Injunctions Issued by National

Courts’, in International Arbitration 2006: Back to Basics?, ICCA Congress Series

No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer Law International, 2007),
198.
238 Marco Stacher, ‘You Don't Want to Go There – Antisuit Injunctions in
International Commercial Arbitration’, ASA Bulletin 23, no. 4 (2005): 653; Julian
D.M. Lew, ‘Control of Jurisdiction by Injunctions Issued by National Courts’, in
International Arbitration 2006: Back to Basics?, ICCA Congress Series No. 13, ed.
Albert Jan van den Berg (The Hague: Kluwer Law International, 2007), 187.
239 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2946.

240 Julian D.M. Lew, ‘Control of Jurisdiction by Injunctions Issued by National

Courts', in International Arbitration 2006: Back to Basics?, ICCA Congress Series

No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer Law International, 2007),
206–210; Gary B. Born, International Commercial Arbitration (The Hague:
Kluwer Law International, 2009), 1055.
241 Julian D.M. Lew, ‘Control of Jurisdiction by Injunctions Issued by National

Courts’, in International Arbitration 2006: Back to Basics?, ICCA Congress Series

No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer Law International, 2007),
219.
242 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1056–1057 and 2944–2945.

243 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 363–364.
244 For example, Himpurna California Energy Ltd v. PT (Persero) Perusahaan

Listruik Negara (Republic of Indonesia) in Albert Jan van den Berg (ed.), Yearbook
of Commercial Arbitration Volume XXV (The Hague: Kluwer Law International,
2000), 11.
245 Award in ICC Case No. 10623 (2001), ASA Bulletin 21, no. 1 (2003): 82.
246 Ibid., 83.
247 Ibid., 86–87.
248 Ibid., 91.
249 See Eric A. Schwartz ‘Do International Arbitrators Have a Duty to Obey the
Orders of Courts at the Place of the Arbitration? Reflections on the Role of the Lex
Loci Arbitri in the Light of a Recent ICC Award’, in Global Reflections on
International Law, Commerce and Dispute Resolution: Liber Amicorum in Honour
of Robert Briner, ed. Gerald Aksen et al. (Paris: ICC Publishing, 2005), 801.
250 Ibid., 802.
251 Ibid., 806.
252 Ibid., 807.
253 ICC Case No. 7934, unpublished, cited in Emmanuel Gaillard, ‘The
Representations of International Arbitration’, Journal of International Dispute
Settlement (2010): 4. See also the Award rendered on 30 June 2009 in ICSID Case
No. ARB/05/07, Saipen SpA v. Bangladesh.
254 See cases extracted in Himpurna California Energy Ltd v. PT (Persero)
Perusahaan Listruik Negara (Republic of Indonesia) in Albert Jan van den Berg
(ed.), Yearbook of Commercial Arbitration Volume XXV (The Hague: Kluwer Law
International, 2000), 11.
255 Award in ICC Case No. 10623 (2001), ASA Bulletin 21, no. 1 (2003): 87.
256 Loewen Group Inc. v. United States of America, Award on Merits, ICSID Case

No. ARB (AF/ 98/3 para. 120) (26 June 2003).

257 Eric A. Schwartz, ‘Do International Arbitrators Have a Duty to Obey the

Orders of Courts at the Place of the Arbitration? Reflections on the Role of the Lex
Loci Arbitri in the Light of a Recent ICC Award’, in Global Reflections on
International Law, Commerce and Dispute Resolution: Liber Amicorum in Honour
of Robert Briner, ed. Gerald Aksen et al. (Paris: ICC Publishing, 2005), 811.
258 Ibid., 812.
259 P. Shaughnessy, ‘Pre-arbitral Urgent Relief the New SCC Emergency
Arbitrator Rules’, Journal of International Arbitration 27, no. 4 (2010): 337.
260 While the timeframes in expedited appointment regimes are a considerable

advance over normal tribunal constitution, they will never be as quick as

applications to court where there will always be standby judges for urgent
requests. Emmanuel Gaillard & Philippe Pinsolle, ‘The ICC Pre-arbitral Referee:
First Practical Experiences’, Arbitration International 20, no. 1 (2004): 14. See
also Bernard Hanotiau, ‘The ICC Rules for a Pre-arbitral Referee Procedure’,
International Arbitration Law Review (2003): 23.
261 P. Shaughnessy, ‘Pre-arbitral Urgent Relief the New SCC Emergency
Arbitrator Rules’, Journal of International Arbitration 27, no. 4 (2010): 355–356.
262 Ibid., 344.
263 Ibid., 339–340.
264 Ibid., 345.
265 Emmanuel Gaillard & Philippe Pinsolle, ‘The ICC Pre-arbitral Referee: First

Practical Experiences’, Arbitration International 20, no. 1 (2004): 14. See also
Bernard Hanotiau, ‘The ICC Rules for a Pre-arbitral Referee Procedure’,
International Arbitration Law Review 6, (2003): 421.
266 Judgment of 29 April 2003, Société Nationale de Pétroles du Congo SNPC and

the Republic of Congo v. Total Fina Elf E & P Congo, Cour d'appel de Paris, cited in
Emmanuel Gaillard & Philippe Pinsolle, ‘The ICC Pre-arbitral Referee: First
Practical Experiences’, Arbitration International 20, no. 1 (2004): 14. See also
Bernard Hanotiau, ‘The ICC Rules for a Pre-arbitral Referee Procedure’,
International Arbitration Law Review 6, (2003): 22.
267 Albert Jan van den Berg, ‘The Application of the New York Convention by the

Courts’, in Improving the Efficiency of Arbitration Agreements and Awards, ICCA

Congress Series No. 9, ed. Albert Jan van den Berg (The Hague: Kluwer Law
International, 1999), 29.
268 P. Shaughnessy, ‘Pre-arbitral Urgent Relief the New SCC Emergency
Arbitrator Rules’, Journal of International Arbitration 27, no. 4 (2010): 346.
269 Emmanuel Gaillard & Philippe Pinsolle, ‘The ICC Pre-arbitral Referee: First

Practical Experiences’, Arbitration International 20, no. 1 (2004): 14. See also
Bernard Hanotiau, ‘The ICC Rules for a Pre-arbitral Referee Procedure’,
International Arbitration Law Review 6, (2003): 75.
270 Article 29(2).
271 Article 29(3).
272 Article 29(4).
273 Article 29(5).
274 That would be so even if the agreement was to use rules in force at the time

of the dispute.
275 Appendix V Art. 2(1).
276 Appendix V Art. 2(4).
277 Appendix V Art. 2(6).
278 Appendix V Art. 3.
279 Article 6.
280 Article 6(7).
281 Article 6(8).
282 Article 7.
283 Article 8(3).
284 Previously, under Art. 2.1 of the ICC Rules for a Pre-Arbitral Referee
Procedure the referee could make four key types of orders: to order any
conservatory measures that are urgently necessary to prevent immediate
damage or irreparable loss and so as to safeguard any of the rights or property of
one of the parties; to order a party to make any payment which ought to be
made; to order a party to take any step which ought to be taken according to the
contract between the parties; to order any measures necessary to preserve or
establish evidence. It has been decided in two recorded cases that this provision
(Art. 2.1) has been drafted broadly and thus should not be interpreted in a
restrictive manner.
285 Appendix II, Art. 1.
286 Appendix II, Art. 32.
287 Appendix II, Art. 4.
288 Appendix II, Arts 5 and 7.
289 Appendix II, Art. 8.3.
290 P. Shaughnessy, ‘Pre-arbitral Urgent Relief the New SCC Emergency
Arbitrator Rules’, Journal of International Arbitration 27, no. 4 (2010): 338.
291 Ibid., 354 and 347. Shaughnessy's comments imply that she would see the

new rules generally as applicable.

292 See generally Simon Greenberg & Matthew Secomb, ‘Terms of Reference and

Negative Jurisdictional Decisions: A Lesson from Australia’, Arbitration

International 18, no. 1 (2002): 125–136.
293 Such awards are discussed in section 16.3.7.
294 For example Art. 14.1 LCIA Rules. See also Art. 16.3 AAA arbitration Rules
which directs that parties should ‘focus their presentations on issues the
decision of which could dispose of all or part of a case’.
295 LCIA Rules Art. 19.3; IBA Rules of Evidence Art. 2(3).
296 Judith Gill, ‘Applications for the Early Disposition of Claims in Arbitration

Proceedings’, in 50 Years of the New York Convention, ICCA Congress Series No.
14, ed. Albert Jan van den Berg (The Hague: Kluwer Law International, 2009),
525.
297 LCIA Rules Art. 9.
298 Ned Beale, Lisa Bench Nieuwvelv & Matthijs Nieuwvelv, ‘Summary
Arbitration Proceedings: A Comparison between the English and Dutch Regimes’,
Arbitration International 26, no. 1 (2010): 145. See also La Pine v. Kyocera Corp.,
2008 US Dist LEXIS 41172 (22 May 2008 NDCal) where a challenge to
enforcement under the New York Convention in relation to summary
adjudication was dismissed.
299 See the discussion of the Judgment of 21 September 2007, Swiss Federal

Supreme Court, Case No. 4 A 220/2007 in Teresa Giovanninni, ‘Comments on

Judith Gill's Report on Applications for the Early Disposition of Arbitration
Proceedings’, in 50 Years of the New York Convention, ICCA International
Arbitration Conference, ICCA Congress Series No, 14, ed. Albert Jan van den Berg
(The Hague: Kluwer Law International, 2009), 526, 529.
300 See, e.g., Rule 56 US Federal Rules of Civil Procedure; Rule 24 English Civil

Procedure Rules.
301 Michael M. Collins, ‘Summary Disposition in International Arbitration’, in 50

Years of the New York Convention, ICCA Congress Series No. 14, ed. Albert Jan van
den Berg (The Hague: Kluwer Law International, 2009), 536.
302 CDC Group PLC v. Republic of the Seychelles, ICSID Case No. ARB/02/14 (14

July 2004).
303 Similar provisions are contained in the ICSID Additional Facility Rules,
Schedule C, Chapter VIII, Art. 45(7).
304 See also Art. 10.20.4 of the Central America – Dominican Republic – United

States Free Trade Agreement to similar effect.

305 Judith Gill, ‘Applications for the Early Disposition of Claims in Arbitration

Proceedings’, in 50 Years of the New York Convention, ICCA Congress Series No.
14, ed. Albert Jan van den Berg (The Hague: Kluwer Law International, 2009),
525.
306 Article 1051(2).
307 Ned Beale, Lisa Bench Nieuwvelv & Matthijs Nieuwvelv, ‘Summary
Arbitration Proceedings: A Comparison between the English and Dutch Regimes’,
Arbitration International 26, no. 1 (2010): 153–154.
308 NAI Rules Art. 37.6.
309 Ned Beale, Lisa Bench Nieuwvelv & Matthijs Nieuwvelv, ‘Summary
Arbitration Proceedings: A Comparison between the English and Dutch Regimes’,
Arbitration International 26, no. 1 (2010): 157.
310 D. Brian King & Melanie van Leeuwen, ‘Summary Arbitral Proceedings: A

Powerful New Mechanism in NAI Arbitrations’, Mealey's International Arbitration

Report 15, no. 3 (2000): 61.
311 Ibid., 63.
312 [1977] 1 WLR 713.
313 [1993] 2 WLR 262; but note the dissent of Lord Salmon in Nova (Jersey) Knit
and Savillej in Hayter v. Nelson [1990] 2 Lloyds Rep 265.
314 Halki Shipping Corporation v. Soapex Oils Ltd [1998] 2 All ER 23.
315 Michael M. Collins, ‘Summary Disposition in International Arbitration’, in 50

Years of the New York Convention, ICCA Congress Series No. 14, ed. Albert Jan van
den Berg (The Hague: Kluwer Law International, 2009), 533.
316 The provision of summary judgment discretions in US Domestic Arbitration

clauses was noted by R Doak Bishop, ‘A Practical Guide for Drafting International
Arbitration Clauses’, International Energy Law and Taxation Review (2000): 32.
317 Judith Gill, ‘Applications for the Early Disposition of Claims in Arbitration

Proceedings’, in 50 Years of the New York Convention, ICCA Congress Series No.
14, ed. Albert Jan van den Berg (The Hague: Kluwer Law International, 2009),
517.
318 The contrary position might be inferred from comments of Aurélia Antonietti,

‘The 2006 Amendments to the ICSID Rules and Regulations and the Additional
Facility Rules’, ICSID Review-Foreign Investment Law Journal 21 (2006): 439.
319 Antonio Parra, ‘The Development of the Regulations and Rules of the
International Centre for Settlement of Investment Disputes’, International
Lawyer 41 (2007): 47.
320 For example a claim by a person who is clearly not an investor even under

the broadest definition in a bilateral investment treaty.

321 Judith Gill, ‘Applications for the Early Disposition of Claims in Arbitration

Proceedings’, in 50 Years of the New York Convention, ICCA Congress Series No.
14, ed. Albert Jan van den Berg (The Hague: Kluwer Law International, 2009),
518-519.
322 The above commentary draws on John Willems, ‘Summary Dismissal under

ICSID Arbitration Rule 40 1(5) – the Global and RSM Awards’

<http://kluwerarbtirationblog.com/blog/ 2011/02/02/...> 2 February 2011.
Global Trading Resource Corp and Globex International, Inc. v. Ukraine, Award,
ICSID Case No. ARB/09/11 (23 November 2010) (‘Global’); RSM Production Corp.
v. Granada, Award, ICSID Case No. ARB/10/6 (10 December 2010).
323 ICSID Case No. ARB/08/3.
324 Global para. 35 citing Trans-Global Petroleum, Inc. v. The Hashemite Kingdom

of Jordan para. 88.

325 Global para. 59.
326 See discussion of Matthew v. Papua New Guinea in Nancy M. Thevenin & John

A. Basinger, ‘Increasing Efficiency in International Arbitration: Use of Common

Law Dispositive Motions’, IBA Arbitration News 15, no. 1 (2010): 197.
327 In domestic common law systems there may be applications for striking out

for want of evidence or an application by a defendant that there is no case to

answer. These processes can be justified where they are built into the rules and
the plaintiff knows that a sufficient evidentiary record must be presented as part
of a prima facie case, otherwise such an application is likely to be forthcoming.
These processes are not found in international arbitration, nor is the concept of a
prima facie case a natural element.
328 A Treaty claim was brought under the Grenada–US BIT after a previous ICSID

Tribunal had rejected a contractual claim brought in relation to the same

Petroleum Exploration Agreement.
329 Rachel S. Grynberg, Stephen M. Grynberg, Miriam Z. Grynberg and RSM
Production Company v. Grenada, Award of 10 December 2010, ICSID Case No.
ARB/10/6 (2010): para. 6.1.3.
330 Ibid., para. 7.2.1.
331 Ibid., para. 8.3.4.
332 This section is not dealing with proceedings in parallel where one is simply in

relation to interim measures.

333 International Law Association, Final Report on Res Judicata and Arbitration

(2006), available at <www.ila-hq.org>, 2 para. 1.4.

334 James Fawcett (ed.), ‘Declining Jurisdiction in Private International Law’,

Report to the XlVth Congress of the International Academy of Comparative Law,

Athens, 1994 (Oxford: Oxford University Press, 1995), at 27.
335 There was even divergence in view between English and Australian courts as

to the proper test to apply, English courts looking to see which was the more
convenient court (Spiliada Maritime Corporation v. Cansulex Ltd [1986] 3 WLR
972) while Australian courts would only defer jurisdiction if the other court was
clearly more appropriate (Voth v. Manildra Flour Mills (1990) 171 CLR 538).
336 International Law Association, ‘Interim Report: Res judicata and
Arbitrationr’, Berlin Conference 2004, 2 available at <www.ila-hq.org>.
337 In civil law countries the application of res judicata in arbitral awards has

often been codified. See, e.g., Art. 1476 of the French Code de Procedure Civile,
Art. 1703 of the Belgian Judicial Code, Art. 1055 of the German Code of Civil
Procedure.
338 Peter Barnett, Res Judicata, Estoppel and Foreign Judgments (Oxford: Oxford

University Press, 2001), at 134–182.

339 R. Chandra Mohan & Lim Wee Teck, ‘Some Contractual Approaches to the

Problem of Inconsistent Awards in Multi-party, Multi-contract Arbitration

Proceedings’, Asian International Arbitration Journal 1, no. 2 (2005): 161, 167.
340 Jane Jenkins & Simon Stebbings, International Construction Arbitration Law

(Kluwer Law International, 2006), 86.

341 Ibid., 156.
342 This book is not dealing with the reverse scenario where a court is asked to

deal with a matter that was previously before an arbitral tribunal.

343 Audley Sheppard, ‘The Scope and Res Judicata Effect of Arbitral Awards’, in

Arbitral Procedure at the Dawn of the New Millennium: Reports of the

International Colloquium of Cepani, 15 October 2004 (Brussels: Bruylant, 2005),
283–284.
344 International Law Association, Final Report on Res Judicata and Arbitration

(2006), available at <www.ila-hq.org>. The Recommendations are available at

<www.ila-hq.org/html/layout_ committee.htm>.
345 Kaj Hobér, ‘Parallel Arbitration Proceedings – Duties of the Arbitrators’, in

Parallel State and Arbitral Procedures in International Arbitration: Dossiers of the

ICC Institute of World Business Law, ed. Bernardo N. Cremades & Julian D.M. Lew
(Paris: ICC Publishing, 2005), 253.
346 Resolution of the Institut de Droit International on the Principles for
Determining When the Use of the Doctrine of Forum Non Conveniens and Anti-
Suit Injunctions is Appropriate. Session de Bruges 2003, cited in Francisco
Arrego Vicunña, ‘Lis Pendens Arbitralis’, in Parallel State and Arbitral Procedures
in International Arbitration: Dossiers of the ICC Institute of World Business Law,
ed. Bernardo N. Cremades & Julian D.M. Lew (Paris: ICC Publishing, 2005), 207–
208.
347 Flour Corp v. Iran, Iran–United States Claims Tribunal Reports, Vol. 11
(1986), 296 (Interim Award); SPP v. Egypt, ICSID Reports, Vol. 3 (1985), 129;
Benvenutti and Bonisant Srl v. Congo, ICSID Reports, Vol. 1 (1980), 330.
348 Official Court Reporter 127 III 279 (14 May 2001).
349 Swiss Private International Law Art. 186 (1bis) entered into force from 1

March 2007.
350 Southern Pacific Properties (Middle East) Ltd v. Egypt (1998) 3 ICSID Rep

129–130.
351 Kaj Hobér, ‘Parallel Arbitration Proceedings – Duties of the Arbitrators’, in

Parallel State and Arbitral Procedures in International Arbitration: Dossiers of the

ICC Institute of World Business Law, ed. Bernardo N. Cremades & Julian D.M. Lew
(Paris: ICC Publishing, 2005), 257.
352 See, e.g., ICC Case No. 3383 (1979), Collection of ICC Arbitral Awards 1974-

1985 (Paris: ICC Publishing, 1990), at 100 and 394.

353 See also German ZPO Art. 1055; Swiss PILA Art. 190.
354 Waste Management v. Mexico (Mexico's preliminary objection) ICSID,
reported in 41 ILM 1315 (2002).
355 Audley Sheppard, ‘The Scope and Res Judicata Effect of Arbitral Awards’, in

Arbitral Procedure at the Dawn of the New Millennium: Reports of the

International Colloquium of Cepani, 15 October 2004 (Brussels: Bruylant, 2005),
270.
356 ‘The Award’ in Nigel Blackaby et al., Redfern and Hunter on International

Arbitration (Oxford: Oxford University Press, 2009), 513–583.

357 Roger Alford, ‘The “Transnational Approach” of the ILA Recommendations on

Res Judicata and Arbitration’, Kluwer Arbitration blog 25 July 2009.

358 Audley Sheppard, ‘Res Judicata and Estoppel’, in Parallel State and Arbitral

Procedures in International Arbitration: Dossiers of the ICC Institute of World

Business Law, ed. Bernardo N. Cremades & Julian D.M. Lew (Paris: ICC Publishing,
2005), 229.
359 Ibid., 230.
360 Ibid., 231–232.
361 Ibid., 262, 263.
362 There is a related question as to whether a tribunal should follow decisions

by national courts or previous tribunals on matters of legal interpretation.

363 Audley Sheppard, ‘Res Judicata and Estoppel’, in Parallel State and Arbitral

Procedures in International Arbitration: Dossiers of the ICC Institute of World

Business Law, ed. Bernardo N. Cremades & Julian D.M. Lew (Paris: ICC Publishing,
2005), 219, 220.
364 Denis Bensaude, ‘The International Law Association’s Recommendations on

Res Judicata and Lis Pendens in International Commercial Arbitration’, Journal of

International Arbitration, 24, no. 4 (2007) 415, 416.
365 Ibid.
366 Norsolor S.A. v. Pabalk Ticaret Sirketi S.A. (Cour de Cassation, October 9,

1984) YB Comm Arb, Vol IX (1984), 159.

367 Société Hilmarton Ltd v. Société OTV (Cour de Cassation, 23 March 1994) YB

Comm Arb, Vol XX (1995), 663.

368 Chromalloy Aeroservices Inc v. The Arab Republic of Egypt (District Court of

District of Columbia, 31 July 1996) YB Comm Arb, Vol XXII (1997), 1001.
369 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 718.
370 Henderson v.Henderson (1843) 3 Hare 100, 67 ER 313; Johnson v. Gore Wood

& Co. [2001] 2 WLR 72 (HC).

371 Decision 15 May 2003, English translation in Mealey's International
Arbitration Report, 18, no. 6, Al-37 cited in Audley Sheppard, ‘The Scope and Res
Judicata Effect of Arbitral Awards’, in Arbitral Procedure at the Dawn of the New
Millennium: Reports of the International Colloquium of Cepani, 15 October 2004
(Brussels: Bruylant, 2005), 282.
372 See, e.g., Audley Sheppard, The Scope and Res Judicata Effect of Arbitral

Awards', in Arbitral Procedure at the Dawn of the New Millennium: Reports of the
International Colloquium of Cepani, 15 October 2004 (Brussels: Bruylant, 2005),
282; August Reinisch, ‘The Use and Limits of Res Judicata and Lis Pendens as
Procedural Tools to Avoid Conflicting Dispute Settlement Outcomes’, The Law
and Practice of International Courts and Tribunals 3 (2004): 37.
373 This is discussed further in section 16.19.2.
374 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 314.
375 See section 16.3.3 which discusses the problematically named provisional

awards. They are problematic in the sense that they are either an award with
final effect or are provisional in which case they are best described as something
other than an award.
376 For example, see Art. 1703 of the Belgian Judicial Code, Art. 1059 of the Dutch

Code of Civil Procedure.

377 Audley Sheppard, ‘Res Judicata and Estoppel’, in Parallel State and Arbitral

Procedures in International Arbitration: Dossiers of the ICC Institute of World

Business Law, ed. Bernardo N. Cremades & Julian D.M. Lew (Paris: ICC Publishing,
2005), 229.
378 Martin Domke, Gabriel Wilner & Garry Edmonson, Domke on Commercial

Arbitration (St Paul, MN: Thompson/West, 2003) § S36:2.

379 As the principle of stare decisis does not apply in international arbitration, it

would appear that an award will not be binding on any subsequent disputes that
arise between the same parties over different subject matter or a different cause
of action (even if related). However, an arbitral tribunal hearing a dispute before
the same parties may apply the principle of issue estoppel. ‘The Award’, in Nigel
Blackaby et al., Redfern and Hunter on International Arbitration (Oxford: Oxford
University Press, 2009), 513–583.
380 Audley Sheppard, ‘The Scope and Res Judicata Effect of Arbitral Awards’, in

Arbitral Procedure at the Dawn of the New Millennium: Reports of the

International Colloquium of Cepani, 15 October 2004 (Brussels: Bruylant, 2005),
277.
381 Ibid., 231.
382 [1965] 1 Lloyd's Rep 13 (CA).
383 Fidelitas Shipping Co. Ltd v. V/O Exportschleb [1965] 1 Lloyd's Rep 13 (CA).
384 Joint American Law Institute/UNIDROIT Work Group on Principles and Rules

of Transnational Civil Procedure (UNIDROIT, Rome 2004) para. 28.3 available at

<www.unidroit.org>.
385 V.V. Veeder QC, ‘Issue Estoppel, Reasons for Awards and Transnational
Arbitration’, in Complex Arbitrations: Perspectives on their Procedural
Implications (Special Supplement – ICC International Court of Arbitration
Bulletin, 2003), at 73.
386 See, e.g., § 31.1 DIS Rules which indicates that ‘(a)t every stage of the

proceedings the arbitral tribunal should seek to encourage an amicable

settlement of the dispute or of individual issues in dispute.’
387 See, e.g., Austrian Code of Civil Procedure Art. 204; Austrian ZPO § 204;

German ZPO § 278; Netherlands CCP Art. 1043; Japanese Arbitration Law Art.
38(4).
388 See ICC Rules of Conciliation, UNCITRAL Model Law on Conciliation.
389 See, e.g., Hong Kong Arbitration Ordinance s. 33 which allows this to occur

where parties consent and the Indian Arbitration and Conciliation Act Art. 30(1)
which allows the tribunal to encourage settlement as of right and allows the
tribunal to engage in mediation, conciliation or other procedures to encourage
settlement with the parties' consent. The Arbitration and Conciliation Act 1996
(India) s. 30(1) expressly indicates that it is not incompatible for arbitrators to
use mediation to encourage settlement.
390 CIETAC Rules Art. 45.1, 45.9.
391 Christopher Newmark, one of the co-chairmen of the Task Force behind that

report, noted that even this minimalist proposal received some criticism on the
basis that settlement was outside the remit of a tribunal whose mandate was to
produce a binding final award. Christopher Newmark, ‘Controlling Time and
Costs in Arbitration’, in The Leading Arbitrators' Guide to International
Arbitration, ed. Lawrence W. Newman & Richard D. Hill (Huntington: Juris Pub,
2008), 88. That was not the majority view of the Task Force.
392 Hilmar Raeschke-Kessler, ‘The Arbitrator as Settlement Facilitator’,
Arbitration International 21, no. 4 (2005): 525.
393 Harold I. Abramson, ‘Protocols for International Arbitrators Who Dare to

Settle Cases’, American Review of International Arbitration 10 (1999): 4.

394 Ibid.
395 Abramson notes rightly that this cannot shelter the knowledge of the
arbitrator. Harold I. Abramson, Ibid., 5.
396 Michael Collins, ‘Do International Arbitral Tribunals Have Any Obligations to

Encourage Settlement of the Disputes before Them?’, Arbitration International

19, no. 3 (2003): 341.
397 Pierre Lalive, ‘The Role of Arbitrators as Settlement Facilitators: A Swiss

View’, in New Horizons in International Commercial Arbitration and Beyond, ICCA

Congress Series No. 12, ed. Albert Jan van den Berg (The Hague: Kluwer Law
International, 2005): 562–563.
398 Harold I. Abramson, ‘Protocols for International Arbitrators Who Dare to

Settle Cases’, American Review of International Arbitration 10 (1999): 1.

399 Ibid., 2.
400 Christian Bühring-Uhle, Arbitration and Mediation in International Business:

Designing Procedures for Effective Conflict Management (The Hague: Kluwer Law
International, 1996), 366.
401 Michael E. Schneider, ‘Combining Arbitration with Conciliation’, in Planning

Efficient Arbitration Proceedings/The Law Applicable in International Arbitration,

ICCA Congress Series No. 7, ed. Albert Jan van den Berg (The Hague: Kluwer Law
International, 1996), 88.
402 David W. Rivkin, ‘Towards a New Paradigm in International Arbitration: The

Town Elder Model Revisited’, Arbitration International 24, no. 3 (2008): 382.
403 Ibid.
404 Ibid.
405 Hong Kong Arbitration Ordinance s. 33(4) requiring disclosure of material

information in such circumstances.

406 See, e.g., Advisory Sub-Committee to Hong Kong Law Reform Commission.
407 Michael E. Schneider, ‘Combining Arbitration with Conciliation’, in Planning

Efficient Arbitration Proceedings/The Law Applicable in International Arbitration,

ICCA Congress Series No. 7, ed. Albert Jan van den Berg (The Hague: Kluwer Law
International, 1996), 85.
408 Gabrielle Kaufmann-Kohler, ‘When Arbitrators Facilitate Settlement:
Towards a Transnational Standard: Clayton Utz/University of Sydney
International Arbitration Lecture’, Arbitration International 25, no. 2 (2009):
197.
409 Christopher Newmark & Richard Hill, ‘Can a Mediated Settlement Become an

Enforceable Arbitration Award?’, Arbitration International 16, no. 1 (2000): 82.

410 Michael E. Schneider, ‘Combining Arbitration with Conciliation’, in Planning

Efficient Arbitration Proceedings/The Law Applicable in International Arbitration,

ICCA Congress Series No. 7, ed. Albert Jan van den Berg (The Hague: Kluwer Law
International, 1996), 94.
411 Fali S. Nariman, ‘The Spirit of Arbitration: The Tenth Annual Goff Lecture’,

Arbitration International 16, no. 3 (2000): 261, 267.

412 Gerald Aksen, ‘Comments on Enforceablitiy of Awards on the Role of
Arbitrators as Settlement Facilitators’, in New Horizons in International
Commercial Arbitration and Beyond, ICCA Congress Series No. 12, ed. Albert Jan
van den Berg (The Hague: Kluwer Law International, 2005), 565–566.
413 Harold I. Abramson, ‘Protocols for International Arbitrators Who Dare to

Settle Cases’, American Review of International Arbitration 10 (1999): 11, fn. 5.

414 A contrary view is suggested by Michael Collins, ‘Do International Arbitral

Tribunals Have Any Obligations to Encourage Settlement of the Disputes Before

Them?’, Arbitration International 19, no. 3 (2003): 337 but it is hard to see how
an arbitrator who has accepted an arbitral mandate can ever deviate from that
where the parties are not in agreement as to a settlement alternative.
415 Piero Bernardini, ‘The Role of the International Arbitrator’, Arbitration
International 20, no. 2 (2004): 121.
416 Gerald Aksen, ‘Comments on Enforceablitiy of Awards on the Role of
Arbitrators as Settlement Facilitators’, in New Horizons in International
Commercial Arbitration and Beyond, ICCA Congress Series No. 12, ed. Albert Jan
van den Berg (The Hague: Kluwer Law International, 2005), 568.
417 Martin Hunter, ‘Commentary on Integrated Dispute Resolution Clauses’, in

New Horizons in International Commercial Arbitration and Beyond, ICCA

International Arbitration Congress Series No. 12, ed. Albert Jan van den Berg
(The Hague: Kluwer Law International, 2005), 446, 472.
Part II: The Process of an Arbitration,
Chapter 9: Hearings

Jeff Waincymer,

9.1. Preparation for Hearings

Chapter 12 looks at the oral aspect of a hearing in the context of the treatment of
witnesses. While hearings are not required in all cases, many rules give either a
party or the tribunal the right to call for a hearing. (1) This chapter considers the
procedural steps that ought to be taken prior to any hearing to ensure that it is
conducted fairly and efficiently. For example, the parties should know in
advance how witnesses are to be utilised, whether there is to be examination in
chief, what time allocation rules will apply, order of presentation and whether
post-hearing submissions will be permitted. Prior to this, to the extent that it
has a discretion a tribunal needs to consider whether hearings are even
appropriate to the case in issue. This is discussed in the following section.

9.2. Whether to Conduct Hearings or have Documents-Only Arbitration

The parties and the tribunal should give consideration to whether an oral
hearing is necessary or conversely, if the arbitration can be conducted by
documents alone. At the very least, the tribunal and counsel should not presume
that a hearing is required in every case. The very need for one should be raised
at a preliminary conference. At times, it may be appropriate to designate a case
as documents-only from the earliest articulation of claims and cross-claims. In
other circumstances, it may only be after all documents and witness statements
are produced that a tribunal can see that credibility is not in issue and oral
testimony will not advance the page "717" matter. Even if an oral hearing is
necessary, attention ought to be given to the appropriate mix between
documentary and oral submissions.

From a cost perspective, documentary submissions are generally less costly,

being prepared by individual parties under their own timeframe.
Contemporaneous documents simply need to be presented and not created.
Conversely, oral hearings collect all of the parties together with the tribunal. A
large group sits there while each presents material. Furthermore, it will often
take longer for witnesses to expound on evidence than it would to present this
in written form. Witnesses, counsel and arbitrators also have to travel. However,
it is difficult to determine the veracity of conflicting witnesses simply by looking
at their written statements, particularly if these have been drafted by
experienced advocates. It is generally accepted in arbitration that if witness
testimony is important, opportunities to cross-examine should be provided. One
hybrid tendency is to treat witness statements as evidence in chief and begin
where necessary with cross-examination.

Another approach is to invite the parties to consider the various issues to be

determined and designate those that might best be dealt with on a documents-
only basis. (2) As always, added problems arise if the tribunal seeks agreement on
such issues which is then not forthcoming and is then asked to make a ruling. An
arbitrator will not always have discretion in that regard. Because party
autonomy is paramount, parties may agree to a document-only arbitration.
Arbitral rules differ as to whether each party has a unilateral right to call for a
hearing or whether it must be by agreement. Most statutes and rules generally
indicate that a hearing must take place if a party so requests, or if designated by
the tribunal. In some cases, the rules will make it clear that the tribunal can call
for a hearing even if the parties agree to the contrary. In other situations there is
no express guidance in that regard. Absent express rights in the lex arbitri or
rules, a documents-only hearing does not offend against either a party's
fundamental right to present its case, principles of natural justice or the audi
alteram partem rule. (3)

At times the rules may be unclear and the tribunal and parties may differ as to
the desirability of a hearing. In those circumstances, considerations outlined in
section 6.2 will be determinative. It was suggested that if the parties wish to
have a documents-only hearing where a tribunal believes this to be
inappropriate and is correct in its view, this may be valid grounds for
resignation if the parties cannot be convinced to change their mind. This is
because a party may not waive due process rights and a tribunal cannot be
forced by subsequent agreement of the parties alone to adopt unreasonable
adjudicatory standards. (4) For example, Born queries whether an advance
agreement to documents-only arbitration would be valid if live page "718"
testimony would be crucial to the particular dispute that has arisen. (5) A difficult
scenario arises if one party subsequently argues for an oral hearing after the
earlier agreement to documents-only. Even where they are justified as a matter
of due process, a successful argument might undermine the arbitration
agreement itself if it was seen as being conditional on the documents-only
stipulation.

In any event, documents-only arbitrations are reasonably rare except in specific

fields. (6) Concern for efficiency should at least mean that arbitrators and parties
give careful consideration to whether an oral hearing is necessary or desirable
in each case. Too many people would presume that an oral hearing is required
as a matter of course, perhaps prompting Redfern and Hunter to observe ‘that
the only thing wrong with “documents only” arbitrations is that there are not
enough of them.’ (7)

9.3. Timing of Hearings

Key policy and procedural issues with timing are first, whether there should be
a single hearing, and secondly how long hearings should take and how time
should be allocated.
9.3.1. Single or Multiple Hearings

There is a policy question as to whether to conduct a number of short hearings

on an issue-by-issue basis or conduct one long hearing with all oral testimony
and submissions. In some cases, separate hearings might be held on any or all of
jurisdiction, applicable law, interim measures, liability and/or quantum.
Multiple hearings might also be desirable if there are multiple parties with
distinct claims against each, although that would not seem appropriate if there
are cross allegations of fault, for example, in a construction dispute. Logistical
constraints have tended to ensure that there is either a single hearing or a
bifurcated hearing between liability and quantum. This tends towards the
common law main hearing litigation norm and hence may be less familiar to
some civilian practitioners who are used to a series of short hearings and strong
reliance on written briefs. (8) The ALI/UNIDROIT Principles also provided a clear
preference for a main hearing rather than a staged page "719" civilian model, (9)
which may also be an emerging trend in many civilian jurisdictions in any event.

Where discretions are concerned, the key aim will be to promote efficiency
without impacting upon due process rights. From an efficiency perspective,
tribunals could consider bifurcating evidentiary stages where decisions on
certain facts will significantly impact upon the type and extent of evidence at
later stages. This is simply a logical corollary of the traditional bifurcation
between liability and quantum. Bifurcating proceedings may also assist in
settlement. Typically the parties are most opposed on questions of primary
liability as these views will often be based on fundamentally different views on
facts and/or law. While parties of course may disagree on damages as well,
particularly in the face of inflated claims, methods of calculation may be more
functional once liability is known. However, bifurcation may extend the time
within which an award can be rendered and may unfairly delay a successful
party's entitlement to compensation. In many instances, bifurcation will lead to
duplication of expenses and may not operate as a cost saver as intended. Often, a
tribunal will have to make predictions as to likely efficiency as it will be unclear
what savings are possible until the first hearing is held. It is impossible to
determine a priori whether bifurcation is more efficient or not. If the parties
settle after the first stage it is obviously efficient but if that does not occur the
contrary may be the case. In deciding whether to deal with certain issues
sequentially, a tribunal should consider the various permutations depending on
the determination it might make. In some cases such an approach will be highly
efficient if one party is to succeed but more problematic if a converse decision is
reached. Furthermore, if the tribunal ultimately decides that the issue cannot be
resolved at that stage and requires further evidence, there may be wasted costs
on both sides. (10)

Bifurcation need not follow the typical order of a legal claim. In some cases a
tribunal may consider that an issue further along the logic process is
nevertheless a sensible efficiency gateway. Tallerico and Behrendt (11) speak of
reverse bifurcation, for example, where damages might be considered first in a
case where lack of proof of damages may be dispositive. A simple example is
where a contract has an exclusion of damages clause where damages are
claimed nonetheless. It may be easier to first determine whether the clause is
valid before determining issues of liability and quantum calculation.

There can also be segmentation within issues. For example, if there is a

disagreement on quantum but acceptance that a certain amount at least is due,
the page "720" tribunal might render a partial award as to the undisputed
amount. (12) A tribunal might also do so where there are ongoing losses and a
quick determination may lead to a form of relief that would stem escalating
damages.

9.3.2. Timing and Time Periods

It is desirable to set hearing dates as early as possible to ensure that such times
are blocked out and parties understand the key dates that they are working
towards. (13) For example, Article 28(1) of the UNCITRAL Rules 2010 calls for
‘adequate advance notice.’ An adequate time needs to account for availability
and reasonable preparation. If there is a timing problem in terms of availability
of counsel, some sensitivity may be needed. While the common law tradition is
that delays are not an appropriate response simply in order to have the
barrister of one's choice, the situation is quite different where there is not a
divided profession and the relevant counsel has been doing all of the work from
the outset.

In setting total time periods for the hearing well in advance, one difficulty is that
this will only be a rough estimate well before the true issues are clarified in the
light of document exchange and witness statements. In such circumstances, a
generous allocation is to be preferred, with the recommended time period
identified but with some reserve days also articulated to account for problems
and future developments. If the tribunal simply limited the time period to an
optimal number of days with a view to promoting expediency, the opposite
could result if there then needed to be a lengthy adjournment for matters not
dealt with in the designated period. Given the logistical difficulties of finding
common times, flying in and out and organising accommodation, it is generally
better to schedule an extra day or even two rather than have to reschedule the
hearing if it goes over time. Even if the time is not needed, if a case finishes early,
a tribunal with nothing else to do with the parties can certainly turn its mind to
starting on the deliberations while all issues are fresh.

Timing issues are also affected by proactivity questions. David Rivkin has
suggested that virtually all cases could be heard within two weeks or less, even
though that is by no means the norm in complex matters. (14) As noted, a tribunal
can also try and ensure that as much as possible is done in writing beforehand
as this is typically less costly. For example, one aspect of the process that can
generally be dealt with in writing is a detailed opening submission. If the
arbitral process has incorporated more fulsome written submissions, then these
should have already page "721" articulated the case in a logical and
comprehensive format. There would then be less need for the advocate to
present a detailed oral outline at the beginning of the hearing as is the norm in
the common law system. This is rarely efficient if the tribunal is properly
prepared.

A tribunal might also call for further synthesis of materials and arguments. For
example, Scott Schedules are sometimes used in construction arbitrations to
summarise pleadings and at time evidence into a single document for ease of
reference. The ICC Construction Arbitration Report suggests that:

In the right hands it is a useful tool. It defines the positions of the parties and
ultimately it will or can be used by a Tribunal to record its views and decisions.
At the pre-hearing stage its main value is that, if properly compiled, it
establishes the position of each party where the existing submissions or
pleadings do not already do so adequately. (15)

A schedule will typically outline the claim, the particulars of the claim, the
defence to the claim, the particulars of the defence, the response of the defence,
the particulars of the response, the quantum claimed, and the particulars as to
quantum. If a Scott Schedule is utilised before evidence is submitted, it will help
the parties in identifying all relevant evidence. If it is filed at a later stage, it may
seek to cross-reference the evidence that has been provided.

Closing oral submissions are arguably more valuable than opening oral
submissions for two reasons. First, the advocate may seek to make submissions
along the lines of the conclusions of fact and law it has hoped the tribunal will
have adopted after hearing the witnesses. The tribunal might benefit from the
opportunity to explore these issues by way of discussion. Secondly, because the
parties are already there and the evidence is fresh in the tribunal's mind, it may
help if counsel seeks to tie it all together on the spot. However, in some cases, it
may be appropriate to allow a further round of written submissions to tie the
matters together. This may be more desirable in highly complex cases where
there has been a vast amount of documents and testimonies to consider and
which need concentrated synthesis. In cases where post-hearing written
submissions may be desirable, a tribunal could also give thought to limiting the
pre-hearing submissions to one round alone, (16) and/or give guidance as to the
desired focus of the post-hearing briefs.

9.4. Place of Hearings

While it is sensible to designate a Seat of arbitration, leading to identification of

an appropriate lex arbitri, virtually all rules allow hearings to be conducted in
other page "722" places without impacting on this procedural framework. Once
again, party autonomy should generally be paramount, although many rules give
the tribunal power to choose and only require consultation with the parties. (17)
Where a tribunal exercises a discretion, hearings may be held other than in the
Seat where this would save overall costs in terms of travel and other expenses
and also where it would allow for site visits or access to witnesses who
otherwise would not agree to or be allowed to travel. In extreme cases, the Seat
might temporarily be an unsafe venue for a range of reasons.

9.5. How Hearings are Conducted

Ideally, if the procedural framework has been sensibly managed, then the
hearing itself should be able to be conducted with utmost efficiency. This
includes minimal surprise, minimal attempts to add new procedural rulings or
items of evidence and a concern for efficiency. In the context of the debate as to
who ultimately controls the process, Jan Paulsson has observed that ‘(t)he
purpose of hearings is to help the tribunal resolve the dispute. Counsel are to be
given a reasonable opportunity to put in their evidence, and to persuade the
arbitrator; but it is for the tribunal, not counsel, to determine what is
reasonable.’ (18)

9.5.1. Pre-hearing Conferences and Hearing Logistics

It is appropriate to have a pre-hearing conference to finalise the logistics of the

hearing. One of the most important questions to ensure adequate timing is a
final schedule of witnesses to be called and an indication of which witnesses are
to be cross-examined. Ideally, the parties will identify the order in which the
witnesses are to be called, the days on which this is proposed and the
anticipated time for cross-examination. The tribunal might also give advance
notice of key issues it feels need to be carefully prepared. (19) The tribunal should
warn each of the parties that witnesses need to be available so as to
immediately give their evidence when the preceding witness' testimony is
concluded. Because one cannot know in advance the exact amount of time taken
in cross-examination, this means that efficient proceedings require overlapping
attendance of witnesses. Tribunals might give a general direction or might
specify which witnesses must be available at each session.

page "723"

The chair or sole arbitrator needs to designate who is to arrange the hearing
room and determine whether it is of an appropriate size and layout, whether it
is configured in a way that allows all participants to deal with materials and has
an appropriate ambience for witness examination and perhaps conferencing if
this is to be employed. The parties may wish to have separate rooms for their
own preparation and deliberations.

Even the seating arrangements may be important, and it may be desirable to

have the Claimant sit near the Respondent's selected Arbitrator and vice versa.
There is also a need to ensure the security of the venue during breaks and also
overnight. There needs to be an appropriate space for transcribers or
interpreters where utilised and somewhere separate for future witnesses to
congregate if they are not to be in the hearing room. Given that witnesses can
sometimes be allowed to stay after giving their evidence, there also needs to be
sufficient space within the hearing room for that purpose. If transcription
services are utilised, there needs to be a clear indication of when transcripts will
be provided. Where there are voluminous documents in a case, it is reasonable
for the tribunal to call for extra copies to be made for them at the place of the
hearing so that they are not required to transport numerous files. There should
be appropriate arrangements for water in the hearing room and refreshments
during breaks. There is a need to consider the use of technological tools such as
electronic documents, live transcripts, Power-Point and the like. Modern
technology might call for networked computers in a hearing room, possibly with
dual screens for arbitrators, one to view exhibits and the other to make their
own notes or view live transcript. Böckstiegel has suggested that in some cases
it may even be appropriate to videotape witness testimony to allow arbitrators
to recollect the quality and veracity at a later date alongside an examination of
transcript alone. (20)

In setting down the times at which the tribunal will sit on a daily basis, on the
one hand, there is a tension between keeping costs down by sitting longer hours
in the hope of shortening the number of days and on the other hand, ensuring
that everyone can concentrate to a sufficient level to do justice to the dispute.
Here it is important to understand that most courts do not expect to hear more
than four hours of core legal and factual issues in a day, with added time for
logistics and procedural discussions. With refreshment breaks and lunch, this
will typically take up to eight hours per day. Flexibility is appropriate and it is
invariably better to sit for longer on some days at least to try and fit within the
designated hearing schedule or to deal with unanticipated requirements.
Procedural challenges and evidentiary challenges might be left until the end of
the day's hearing where appropriate, so that these do not disrupt the timetable,
(21) although in many cases they will need to be dealt with when raised. In

scheduling breaks, it is appropriate page "724" for the tribunal to consider

cultural preferences such as when lunch is typically taken, whether a relevant
culture has a midday rejuvenation break and whether there are religious
observance rituals that should be respected. A typical example of the latter is
Friday prayer times for Muslims, although if a tribunal feels the need to sit over
a weekend, other religious observances become potentially relevant.

The chair or sole arbitrator will typically commence the hearing with an opening
address indicating the procedures and principles to apply. The Tribunal will
typically introduce themselves and ask each party to identify who is present.
Counsel may then be permitted to make opening statements although again
these are of questionable value with a properly prepared tribunal. The trend is
to restrict such statements, particularly when common law advocates in
complex matters would otherwise wish to spend many hours or even days on
their opening remarks. This is not appropriate for international arbitration
where detailed submissions combining facts, documents and legal reasoning
have already been exchanged. To be useful, an opening address should
concentrate on outlining exactly what is sought to be achieved at the hearing
and how it integrates into the overall case presentation. Some tribunals limit
opening statements and bar closing statements in favour of final written
submissions. At the very least, tribunals should carefully consider whether
closing oral submissions and post-hearing briefs are necessary and provide
early guidance. In other cases a tribunal will simply designate how much overall
time each party has and allow it to allocate as it sees fit.

The chair or sole arbitrator needs to consider how documents and witnesses
will be handled during a hearing to promote maximum fairness and efficiency.
There may be issues as to which documents should be addressed, how they
should be physically presented, how easily they can be called up and cross
referenced without wasting time and whether electronic tools can be effectively
used. These issues are discussed further in section 11.9 In relation to witnesses,
there are questions as to the order, whether examination in chief will be utilised
or whether witness statements will stand in lieu of evidence in chief and the
approach to cross-examination. There may also be questions as to witness
sequestration and the possible use of witness conferencing. These matters are
discussed further in Chapter 12. Where witnesses are involved, the tribunal
would explain directly or via counsel how their evidence in chief is submitted,
prepare them for the impending cross-examination and advise as to the
entitlement of their counsel to re-examine.

Section 12.8.7 deals with the issue of questioning by the tribunal. A contentious
issue is whether the tribunal should balance the analysis where it believes one
counsel has failed to adequately question the opposing witnesses. The tribunal
will need to consider cultural issues where cross-examination occurs. Those
from a common law background are used to aggressive cross-examination
seeking to undermine a witness's creditability or which aims to trap them into
particular adverse comments or concessions. Such an approach can be even
more contentious where expert witnesses are concerned. It also poses problems
in terms of timing of questions from the tribunal. When cross-examining,
common law counsel seek to avoid giving the witness an opportunity to expand
and qualify. Hence, quite narrow and directed questions are used. There is an
issue for a tribunal page "725" if it wishes to hear more on a point, but knows
that this could undermine counsel's cross-examination strategy. A tribunal
should be sensitive but should feel entitled to ask questions it believes are
important as a good faith agreement to arbitrate would imply such an
entitlement. It should however be noted that ‘some legal traditions may view
questioning or other active intervention by fact-finders as possible indicators of
prejudgment or lack of neutrality’. (22)

As noted previously, it has also been observed that ‘humour does not travel well
cross-culturally, nor does it read well in a transcript in the cold light of a distant
courtroom.’ (23) The recommended demeanour is friendly while at the same time
being serious and professional rather than informal. (24)

9.5.2. Closing the Hearings

When the hearing has been concluded it is appropriate for the tribunal to make
a declaration to that effect. This is required by a number of rules. (25) It is
customary for a tribunal to ask the parties prior to the close of the hearing
whether they have any comments to make as to the conduct of the proceedings.
This has two advantages. If there are any concerns, they can be considered by
the tribunal while the hearing is still on foot. If a concern is valid, it can generally
be immediately redressed. The second and related benefit is to minimise the
chance that a party will subsequently casuistically assert errors only when an
adverse finding is made against it.

It is also important to understand the psychological aspects of dispute

settlement. Parties will better respect the outcome if they respect the process.
Part of the latter is the respect with which they were treated by the tribunal.
Consequently, the best arbitrators will thank the parties for their contribution
and indicate genuinely that at the end of the hearing and submissions, they now
fully understand the nature of each party's contentions and have been aided in
that regard for the purposes of their ensuing award. Showing respect to each
party individually is also likely to induce greater mutual respect between them,
perhaps helping those with entrenched positions to better see that the other
side may have some merit in their argument at least. (26)

page "726"

9.6. Chess Clock Arbitration

Leading arbitrators have coined the phrase the ‘Böckstiegel Method’ to describe
the prevailing practice of Karl-Heinz Böckstiegel in many of the arbitrations he
has chaired. (27) It is otherwise described as the chess clock method. Böckstiegel
himself has made it clear that ‘(t)here is no one “best” way to conduct a hearing.’
(28)

Nevertheless, he describes the method as follows:

From the gross time available for the hearing one deducts estimated time for
coffee and lunch breaks, procedural discussions and questions by the
arbitrators, so as to arrive at net time available for the parties. This time will
normally be divided by two between the two parties and each party will be free
to use its time as it prefers for introduction and examination of witnesses
presented by itself or the other party. This method seeks, on the one hand, to
ensure equality between the parties and, on the other hand, to give each party
the freedom to use its time for what it considers most important in the hearing.
(29)

Jan Paulsson quotes a typical Böckstiegel method order as follows:

Each party will have a maximum of four hours for its first round presentation
and a maximum of one hour for its rebuttal presentation, after deduction of time
for breaks and other business. Each side is free to determine how much time it
will spend on the presentation of evidence, including witnesses and experts, and
on argument, respectively. Time not used by a party for its first round
presentation may be transferred to its rebuttal time. Time used by a party
examining witnesses or experts presented by the other party will be deducted
from the time allotted to the examining party … As the hearing has to end after
two days, the tribunal does not intend to grant any extensions of the above time
period. (30)

While such a method raises both conceptual and practical questions, it has
become a widely utilised and regarded approach. David Rivkin has opined that
wherever chess clock arbitration has been used, the time limits simply forced
parties to present only material and relevant evidence, avoided duplication and
never forced important evidence to be excluded. (31) Albert Monichino has also
suggested that the method emphasises the importance of written advocacy and
encourages surgical page "727" cross-examination and efficient document
assembly. (32) Nevertheless, as with any recommended methodology, it needs to
be considered on a case-by-case basis and at times varied, as Böckstiegel himself
notes. Jan Paulsson suggests that to be appropriately effective, the Böckstiegel
method ‘requires: (1) sufficient early familiarity with the case to make a serious
assessment of the amount of hearing time the case warrants; (2) consultation
with the parties; (3) notice to the parties of the purposes of the hearing; (4)
notice to the parties of time available to each side; (5) freedom of parties to use
time as they wish; and (6) suppression of any form of disruption by counsel,
witnesses, or indeed the arbitrators.’ (33)

It has been suggested that chess clock arbitration should only be used by
agreement of the parties. (34) While it is always preferable to have the agreement
of the parties to any methodology, there is no residual alternative to chess clock
arbitration that is inherently more reasonable. Hence there seems no reason
why a tribunal cannot direct such a method absent party agreement. It is simply
directing unconstrained equal time. As long as the total time is reasonable, so
too must be the method.

There are, however, some important factors to consider on a case-by-case basis

and some desirable actions on the part of the tribunal if a chess clock approach
is to be optimised. While it seems intuitively correct to give the parties the
freedom to use the time as they wish, their preferences do not subsist in a
vacuum, but relate instead to the anticipated importance of issues to the
tribunal. Hence, it may be advisable for the tribunal to give guidance as to the
matters of most concern and which should hence get the most time. A tribunal
may also properly indicate those areas where the written submissions and
witness statements have not fully resolved an issue and where elaboration
would be desirable. At other times a tribunal may indicate that certain forms of
cross-examination will be less helpful. An example would be in relation to cross-
examination of legal experts who have already provided detailed written
submissions on points of law. (35)

A tribunal might also give guidance when time may seem to be misused by
counsel. Time limits will often be set in the context of what will ideally occur in
terms of witness and counsel behaviour. While some leeway should always be
allowed, it is appropriate to expect that in some circumstances, it may appear
clear to the tribunal that a party is inappropriately allocating its time, perhaps
spending too long with peripheral witnesses or on peripheral points. A tribunal
can help by indicating when a witness is presenting inappropriate opinions,
irrelevant material or repeating a point already made or not in contention.
Another reason for the page "728" tribunal to deal with inappropriate allocation
of time is to give counsel every opportunity to meet the challenge and prevent
an argument at a later stage that a failure to grant an extension of time was a
violation of due process rights.

Some might challenge the chess clock presumption of equal time on the basis
that parties do not have an equal number of claims to present. (36) For example, a
claimant might need to have numerous witnesses to prove a factual breach. A
respondent might simply rely on an exemption clause in a contract, albeit
without first admitting fault. There are obviously many permutations but the
starting presumption should always be equal time for a range of reasons. Where
opening and closing submissions are concerned, there ought to be a natural
presumption of equality as each must address all of the live issues. This will be
so even if the parties have a disparate number of primary claims that they are
making. Not only must each party present its own claims but it must seek to
refute those of its opponent, given that these have not been conceded prior to
the hearing. This presumption will be so in relation to overall conclusions and
also in relation to submissions about the proper way to treat witness testimony,
interpret documents and consider questions of law.

The key area that is more problematic is cross-examination where the parties
have a differing number of witnesses and/or the witnesses deal with differing
levels of complexity or the importance of veracity varies between each party's
witnesses. The bulk of the time at a hearing is typically utilised for cross-
examination, although the number of witnesses per se may not indicate the
breadth and depth of issues to be covered. (37) While the cross-examination
challenges of one party tend to increase the more witnesses the other party
calls, the appropriate aim is to provide a sufficient amount of time that allows
each party an adequate opportunity to do all things necessary to make its case.

The same issues arise as to whether the tribunal should factor in differential
abilities in the language of the arbitration and differential experience and
competence of counsel in setting time allocations. These factors should be more
relevant to determination of reasonable total time and would not normally lead
to differential a priori allocation. Another related problem in calculating chess
clock time is where interpreters are used. If one party's witnesses require
translation and the others do not, extra time is inevitable, although this again
could be built into the original global estimate. Once the time allocation has been
announced, a tribunal will also need to be vigilant to ensure that parties do not
use interpretation to disrupt their opponent. This could arise by calling for an
interpreter when not truly required and/or asking witnesses to speak slowly or
calling for questions to be page "729" repeated. (38) Others problems arise where
there are disputes between counsel as to the accuracy of the translation.

If a tribunal awards differential time to account for such matters as translation

and cross-examination, this should not offend against equal treatment. (39) As
noted, if one side has many more witnesses than the other, some extension of
the time for cross-examination may be necessary. Similarly if some witnesses
need interpretation and others do not, allowances can again be made. (40) Having
said that, high quality interpretation should not take that much longer as the
interpreter is merely translating questions and answers and not formulating
thoughts, (41) although one timing issue is whether there should be simultaneous
or consecutive translation. This is discussed further in section 12.8.11.

The ultimate question is more about each having a reasonable opportunity and
not purely technical equality of time. A party can always complain about a less
than adequate opportunity regardless of the treatment of the opposing party.
Equal treatment is not a defence to a claim of inadequate treatment, although it
is relevant from an evidentiary perspective. Conversely, if each has clearly had
an adequate opportunity, it is less valid to complain that the other party had
more than an adequate time. (42) Because of these uncertainties, it is preferable
that a tribunal makes it clear from the outset that rigid equality of time is not the
intention. Instead, the aim is to give each an adequate opportunity and provide
equal treatment in the context of any differing circumstances as between the
parties. (43) If the time limits are established sensibly and the hearing is
conducted firmly but fairly, parties should be able to adequately present their
cases within the time limits and not run out of time. If matters addressed in the
hearing show that the time allocated has not been reasonable, a tribunal must be
prepared to make modifications. While all would acknowledge that some
flexibility needs to be built in, more problematic is when and how this is to
occur, particularly after parties may have prepared on the assumption of carte
blanche entitlements within equal time parameters.

page "730"

A key challenge is that to be useful as a time and cost saver, the time limits
should be as short as possible but still guarantee due process. A chess clock
timeframe will thus call for strategic decisions as to whether to have lengthy
openings and closings, oral evidence in chief (if allowed) or whether to reserve
maximum time for cross-examination. Ideally, tribunal time limits should follow
submissions by the parties as to the amount of time they wish to utilise and
exactly how they intend to utilise it. Counsel will need to consider what time to
allocate to weaker, albeit tenable arguments. In these circumstances it is
important to ensure that parties can never say that they felt obliged to discard
such tenable arguments as a reasonable opportunity to present a party's case
must allow all arguments on which it might ultimately win to be put. Tenable
arguments must be put somewhere. Hence it is important to ensure that written
submissions are not unduly constrained and allow presentation of arguments
that might not be expanded on in the oral hearing.

There is another reason to be slightly more generous in time allocation. If both

parties have been able to present their cases without fully utilising their
allocated time, this also obviates the potential for a challenge based on failure to
afford the right to be heard. (44)

Other implementation problems arise because some scenarios will be difficult to

allocate. An example would be a challenge by cross-examining counsel to the
effect that the witness is being unduly evasive and calling for appropriate
directions by the tribunal. Opposing counsel may argue to the contrary. The time
allocation is best seen as part of the procedural control by the tribunal and
should not simply be allocated to the counsel who is speaking, particularly as
one of them will be justified in their assertions on a matter not anticipated. (45) A
rigid rule can again be problematic, particularly if the objections are validly
made, a point acknowledged by Paulsson. Even a presumption commonly used
that successful objections should be debited against the opposing party is not
inherently logical because the fault may be that of the tribunal, at times caused
by gaps or ambiguities in its previous procedural orders. (46)

There are also important questions of measurement. If chess clock arbitration is

to be employed, some tribunals will measure the time on their own, use a
secretary where available and/or have a representative of each party confer as
to timing. The latter can be problematic if there is a dispute and the tribunal has
no independent way of identifying time usage. A tribunal might announce time
so that it appears in the transcript. Marriott and Tackaberry counsel against
excessive concern for precision of time measurements under chess clock
schemes rather than a substantive approach in aid of a fair hearing. (47) Others
argue that time allocation must be strictly enforced. (48)

page "731"

Whatever the methodology, it should be confirmed at the pre-hearing

conference. If a chess clock approach is employed, parties should have a clear
understanding as to the time elapsed and time remaining so that they can
continually reflect on the optimal way in which they wish to use their remaining
time. (49) That should also involve indications from the tribunal as to which
matters may have been covered sufficiently to enable a determination and
which matters still would be of particular benefit to the tribunal. As long as such
comments are merely recommendations as how to best use time, they are
wholly consistent with a chess clock laissez-faire paradigm. Counsel is free to
accept or reject the recommendation as the case may be. Such designation has to
be sensitively conducted where there is a multi-person tribunal and the views of
each arbitrator may not be identical on each central issue.

A chess clock arrangement must also think about how to deal with tribunal
intervention. The implications of tribunal interventions are potentially
exacerbated with multi-person tribunals, in particular where a party-appointed
arbitrator seems to be intervening more than is reasonable. Careful control may
need to be exercised by the Chair. There are also procedural matters that need
to be dealt with by the tribunal and some time delays that inevitably arise in
moving from one witness to another. The tribunal will naturally explain matters
to each witness, including the obligation to tell the truth, and confirm witness
statements. In setting the time for tribunal interventions and dealing with
procedural queries, a tribunal should also understand that there is some conflict
between the opportunity it seeks to reserve to itself and the impact this has on
the party's own presentations. While significant time does need to be reserved
for such matters, the more a tribunal has properly prepared the hearing
procedurally, the less scope there should be for legitimate procedural disputes
as to its own general managerial approach. Burr and Karrer suggest that one
hour per day is normally appropriate for the tribunal for such matters. (50)

Allocating tribunal intervention time against a party also has problems as a

substantive approach to allocation might need to look at which party the
intervention aids. For example, if a tribunal raises a concern it has with a
particular witness's testimony, there is little substantive difference whether the
tribunal raises this concern during evidence in chief or during cross-
examination. The same is true when a tribunal invites a witness to elaborate on
a complex notion so that the tribunal better understands what is being put.
Another aspect of tribunal questioning is that if there are too many
interruptions in the context of questions that occur to the arbitrator, this
interferes with the advocate's choice of order of presentation which is a logical
corollary of a chess clock predisposition. (51) However, alleviating a concern as
soon as it occurs to the arbitrator can only be in counsel's best interests.

page "732"

One problem with too tight a timeframe might be that tribunals are reluctant to
intervene lest they be seen to be interfering with the party's right to present its
case within the time designated. Here there is a difference between designating
a separate amount of time to cover such intervention or allocating it as against
one of the parties. (52) There are challenges either way. Designating a separate
time in advance for the tribunal can never be more than guesswork as the
tribunal at that stage will not know what points need clarification or challenge if
these only arise from comments in oral submissions and testimony. However, a
proper perusal of prehearing written submissions and witness statements may
alert a tribunal to a particular need for its own exploratory time at the hearing.

A chess clock approach must still involve the exercise of judgment and
discretion and ensure that unanticipated or disruptive actions can be
appropriately redressed. Monichino has suggested the following draft
procedural order.

Draft procedural order for‘stop clock’hearing

The Arbitrator directs as follows:

Stop clock Procedure

(1) The hearing of the arbitration shall commence on [complete date] and shall
conclude on [complete date] in [complete city] in accordance with the
following directions, and any further directions made by the Arbitrator.
(2) The sitting hours shall be 9:30am to 5:00pm each day with one hour for
lunch and a morning and afternoon break of 15 minutes each.
(3) The time fixed for the hearing, after allowing one hour each day for the
Tribunal's interventions and time for administrative and procedural
matters, will be apportioned equally between the parties such that:
(a) the claimant shall have a total of [complete] hours; and
(b) the respondent shall have a total of [complete] hours.
(4) Each party is responsible for the way it chooses to use the time available to
it.
(5) The following list illustrates categories of activity which will typically be
 charged against each party's allocation of time:
(a) oral examination of witnesses (irrespective of who presented the
witness, but subject to adjustment in the event of consistent
unresponsiveness) ;
(b) oral submissions;
(c) causing an unjustified interruption or prolonging a justified
interruption (for example, an unsuccessful objection will generally be
charged against the party that made it, and a successful objection
against the party that resisted it); page "733"
(d) setting up displays or presentations whilst the arbitral tribunal is
sitting;
(e) late arrival of counsel or witnesses;
(f) other unjustified delays, including matters which could reasonably
have been dealt with outside the hearing time fixed.
(6) Each party should designate one person to track time. In this task, the
designated person will be instructed by the Arbitrator and shall report
directly to him.
(7) A party is not bound by opposing evidence which it does not challenge but
is expected to cross-examine at least one opposing witness with respect to
any significant matter which the other party should be given the
opportunity to answer.

9.7. Representation and Control Over Counsel

9.7.1. Right to Representation

A logical corollary of the mandatory norm to have an adequate opportunity to

prepare a case is that a party who does not have the expertise to do so should be
allowed to use a representative for such purposes. There is no general
requirement that the representative be legally trained, although that is the
norm. Where legal representation is concerned, there is also no requirement for
a particular qualification of counsel. While that is the norm, a contrary position
may pertain under some lex arbitri and/or in some enforcement countries. In
the latter case, there would be an issue as to whether an enforcement court
would see this as a matter of public policy. That ought not to be so but such a
position cannot be guaranteed in all jurisdictions.

9.7.2. Oral Argument and Counsel Submissions

If counsel is involved, they will typically draft relevant documents, including

witness statements. As indicated in section 12.5.2, it is accepted in arbitration
that legal counsel is entitled to interview prospective witnesses and help draft
statements. Counsel also has significant involvement in hearings. Here the
common law style tends to dominate, with tribunals deferring to counsel as to
the mode and order under which they wish to present the case. That approach is
supported if a chess clock methodology is utilised as outlined in the previous
section, although general notions of proactivity and appropriate tribunal
directions might militate against.

The tribunal and parties should consider how best to deal with oral arguments
at the hearing, given that a conscientious tribunal will have read all the
materials as they emerge, will have considered pre-hearing written submissions,
and considered key documents and witness statements. Oral submissions
should simply hone in on page "734" the key strategic elements and aim to be as
persuasive as possible. If the tribunal has obviously not come appropriately
prepared, opening and closing submissions are the way to tie issues together,
educate the tribunal and present the entire case strategy in a succinct form. In
the normal course, where the tribunal is appropriately prepared, the closing
submissions can usefully be used to deal with those matters of importance that
may still be unresolved in the minds of the tribunal. If the tribunal is
forthcoming as to those matters, counsel should address them. If not, counsel
needs to predict what they are likely to be. It is salutary to note comments of
experts as to how rare it is for oral summations to make a difference. (53) Because
of this, counsel may wish to encourage the tribunal to identify its remaining
areas of uncertainty. (54)

There are differences in view as to the use of PowerPoint and other visual and
technological aids. Most people do not use such tools effectively. They are
particularly inappropriate when they force the listener to divide their attention
between the slide and the speaker, without any clear indication of how this
should occur. Other problems with technology such as PowerPoint is that there
may often be problems in setting it up, getting the technology to work on cue
and having different tribunal members able to read it with ease. (55) Their main
value is when the speaker wishes to talk to something while it is visualised by
the tribunal. They are often valuable when diagrams, pictures or graphs are
involved. Visual aids can also be particularly effective where counsel must
address the interpretation of documents, statutes or treaties. If instead, the
PowerPoint is merely summarising the points being made orally, it must have
been a poorly presented case from the outset if the key issues cannot be
understood by listening alone. This is particularly so if the advocate
concentrates attention on the matters that are likely to be still troubling the
tribunal. An additional problem with PowerPoint is that advocacy is the art of
persuasion and counsel may only be aware of the key matters on which the
tribunal still needs to be convinced when seeing the body language of the
arbitrators and hearing their observations during the hearing. In many cases, it
is impossible to optimally identify those matters in advance when trying to
generate PowerPoint slides. However, the more complex the matter and the
greater the number of claims and cross-claims, the more justification there may
be for hard copy or technological outlines of arguments.

As noted above, opening statements in arbitration are likely to be shorter than

in common law litigation as the tribunal has seen detailed written submissions,
often two rounds from each party. Best practice would also involve presenting a
sufficiently detailed written outline of arguments in advance, in part because of
page "735" familiarity of European arbitrators with this approach and also to
obviate the need for detailed note taking. (56)
Another question is the order of presentation of argument. While the claimant
would normally go first, legal systems differ as to who has the final word in
concluding submissions. Different legal systems vary as to whether the final
submission is made by a claimant in reply or via a respondent in response to
claimant's last submission. Under the common law, the claimant usually speaks
last. Under the civil law, it is usually the respondent that has the final say. Simple
equality would say that there should be one presentation each and if a second
round, each should have a second right. An alternative approach is to see why
there is ever a need for a second round. Claimant has brought the case and in all
legal systems ought to present first, based on all of the evidence and
submissions to date. Respondent is entitled to hear the final articulation of the
claim against it and then respond. To the extent that this response varies to any
degree from previous evidence and submissions, a claimant can then be
considered to be entitled to reply so that it has an equal opportunity to answer
contentions as did the respondent, even if that gives it a numerically higher
number of opportunities to speak to the arbitrator. Neither view is inherently
correct and it is perfectly permissible for an arbitrator to vary on a case-by-case
basis and also give directions as to the matters that may be addressed in trying
to limit the scope of reply and surrebuttal procedures.

9.7.3. Ethical Rules, Legal Representatives, Conflict of Interest and Control

over Counsel

National lawyers will generally be subject to certain ethical rules about

behaviour. For example, these might prevent counsel from contacting
prospective witnesses prior to a hearing. The question arises as to when such
ethical rules would apply in international commercial arbitration. (57) Generally
speaking, such ethical rules are drafted as part of litigation processes and are
not intended to apply to conduct by counsel in international commercial
arbitration. (58) Nevertheless, in some circumstances, matters that are
considered unethical under such domestic rules would be page "736" of such a
nature as to be similarly considered unethical under general principles
applicable to arbitration.

Arbitral rules generally do not provide specific directions as to the ethical

behaviour of counsel, although decisions in that regard naturally fall within the
broad ambit of a tribunal's discretionary procedural powers. It is possible to
argue to the contrary on the basis of a consent theory. The parties have agreed
to arbitrate and thus be subject to the decisions of an arbitrator but this is not so
with counsel. On this view, there is a question as to what control the tribunal can
exercise. To those who see the status of an arbitrator as akin to that of a judge,
one might argue that there is an inherent power to control counsel. An
alternative argument to the same end is to say that the tribunal certainly has
control over the parties and that counsel is only there by reason of the parties'
right to present their case through a representative rather than directly. Such a
representative cannot have broader rights than the parties themselves. Hence
any direction that could have validly been given to a party if it was self-
represented should similarly be able to be given to counsel. That should
certainly be so where counsel is engaging in guerrilla tactics and otherwise
acting in bad faith.

The issue remains contentious and a particular problem is that even if such a
power can be said to exist, it is more difficult to determine what standards
should be met by counsel and what ramifications should follow if standards are
breached. Here the difficulty is that there is such a difference in ethical and
professional norms between different legal families, that there is no uniform
model that could unassailably apply. If counsel in an arbitration come from
different legal families, there is also no easy way to even begin to establish
appropriate norms. The duty to treat parties equally would also apply to their
counsel. But what does that mean in this context? Is it equality to treat each
counsel in the context of its domestic ethical standards, even in the unlikely
event that these apply to arbitration? Conversely, should there be exact identity
of treatment where this may be closer to the traditional norms of one counsel
rather than the other? Once again there is no easy answer.

One problematic area is to consider how to deal with different cultural values
and experiences. Not only may parties come from different legal cultures but so
too may their counsel. At times a party and its own counsel may come from
differing traditions. That commonly arises where foreign parties choose counsel
working out of the seat of arbitration so as to reduce logistical costs, or where
they choose those who have particular expertise in arbitration. For example, if
two Islamic parties select English and French multi-national law firms
respectively for an arbitration in Beijing, it is more challenging to determine
which ethical model ought to prevail at the margin where answers are not
uniform between systems.

Despite these challenges, there must always be a need for ethical behaviour
notwithstanding the difficulty in defining it and such behaviour must be
contemplated under good faith a priori consent to arbitration. Some institutions
have sought to identify core standards that would generally be accepted page
"737" worldwide. For example, there has been a draft code of conduct for
arbitration proceedings taking place in Switzerland. (59) Matters covered include
that ‘counsel shall maintain high standards of professional conduct;… act
honestly, independently, fairly, diligently, and efficiently… has a duty of loyalty
to its clients which is consistent with its duty to the tribunal to act with
independence in the administration of justice; not advise or assist a client to
engage in conduct which counsel knows is criminal or fraudulent under the
applicable law or rules, act with competence, skill, care, honesty and loyalty;
exercise independent professional judgment and render open and honest
advice;… never permit its independence, integrity and standards to be
compromised by external pressures;… represent its client diligently and
promptly in order to protect its client's best interests; … preserve
confidentiality;… put the interests of the tribunal in terms of its requirement to
act with independence in the interests of justice before its own interests or that
of any other; exercise all care to ensure that no conflict of interest arises; not
knowingly make an incorrect statement of material fact or law to the tribunal or
knowingly offer inaccurate evidence and not pursue frivolous proceedings.’
Extreme cases of abuse will be easy to identify, but problems at the margin can
be envisaged. In addition, these will be impacted upon by tensions between
tribunal proactivity and the party's right to present its case as it sees fit. The
same issues apply as between tribunal and parties. There are a number of
typical problem areas where control of counsel might be seen as ideal but which
are simply about proactivity and not ethics. These include:

1. Counsel is spending precious time on irrelevant submissions.

2. Counsel is taking too long to make a point that is already understood.
3. Counsel is repeating a core contention already made.
4. Counsel is making excessive procedural challenges as to the conduct of the
hearing, the behaviour of witnesses or of opposing counsel.
5. Witnesses of fact are expressing opinions at the encouragement or with the
acquiescence of counsel.
6. Witnesses are taking too long, being allowed by counsel to deal with
irrelevant matters or to repeat themselves unduly.

If the Chair believes that a matter is irrelevant, the counsel might be asked why
he or she thinks the contrary pertains. If a point has already been made, the
Chair can point this out. In these situations, there can be no argument that due
process rights have been denied. Objections are more problematic as this is in
part cultural, with the common law tradition allowing for numerous objections
to the framing of questions to witnesses and often as to the admissibility of
answers and tendered documents.

page "738"

9.8. Language, Translation and Interpreters

9.8.1. The Language of the Arbitration

Most rules indicate that the parties may select the language and, absent such a
choice, the language is to be determined by the tribunal. (60) ICC Techniques
suggest that if the parties have not agreed on the language, the tribunal should
consider use of a procedural order after consulting the parties and prior to
establishing the terms of reference. (61) Some rules provide for a default language
absent agreement from the parties. (62) Others provide that the arbitration
should be in the language of the arbitration agreement unless the parties decide
otherwise. (63) Some rules will specifically direct a tribunal to consider all of the
circumstances, including the language of the contract, in exercising a discretion.
(64) These rules also typically indicate that if documents are in languages other

than the language of the arbitration, the tribunal may order the submitting party
to provide a translation. (65)

There is a need for a fair and efficient choice. Key questions are whether one
party will be disadvantaged in the choice of language and whether the language
chosen would lead to delays in hearings. Where a tribunal has a broad
discretion, it may be natural to consider the language of the contract. However,
this may not be the best as this may not be the same as the language in which
the parties communicated when consummating the deal. Another reason why
the language of the contract is not necessarily determinative is that contracts
are often drafted by legal representatives. Yet many disputes may be resolved
by oral testimony. Selection of language should not be a major issue as
regardless of the language selected, that does not mandate that each witness
present its testimony in that language.

Arbitration need not be limited to a single language. An arbitrator or the parties

could determine that the hearings will be in one language, with pleadings and
written submissions in another. Conversely, a decision might be taken to allow
each party to make all submissions in its native language and run a multi-
language arbitration. If proceedings involve two or more languages, ICC
Techniques suggest that steps be taken to avoid duplication, only translating
documents where necessary. The parties could also agree that only one version
is binding. Where one specific language of arbitration would not normally
appeal to both parties, page "739" it may be desirable to specify this in the
arbitration agreement, although if there is no likely preference at the time of the
dispute, that may also be a problem at the initial negotiation stage.

Language can also impact upon the choice of arbitrator and the choice of a Chair.
For example, the presiding arbitrator will need to have sufficient language skills
to draft the award in the designated language. The rules are often problematic
as many tend to refer to the tribunal's right to select the language but selection
of the tribunal comes first, which will in turn colour their language preference.
Importantly, while it is for the tribunal to make the determination, this is a
matter that needs to be considered by the relevant persons at the time of
appointment of the tribunal. In one sense, those persons are making decisions
about suitable arbitrators prior to knowing how decisions as to language will be
taken. (66) In many cases there will be a natural conflict between the language of
the commercial transaction and the language that would be most comfortable
for highly respected international arbitrators. Undesirable problems may arise if
the party-appointed arbitrators cannot even agree on the preferred language
attributes of the Chair. (67)

9.8.2. Translations and Interpreters

Regardless of the solution adopted as to the language of the arbitration,

attention needs to be given to the question of translations or interpreters.
Where tribunals allow parties to make submissions in their native language, it
would be common to ask that appropriate translations be provided, at least if
the other party so requests. (68) There is then the question as to who pays for the
cost of translation or interpretation. The tribunal could ask parties to share
expenses on an interim basis pending a final costs award or may call on the
party needing the service to pay for it in the first instance. Arbitration does not
normally involve a separate stage of inspection of an opposing party's
documents, although this might be appropriate in specific cases. In these
circumstances there would be similar questions as to who is obliged to pay for
any translations. The norm should be that each party is obliged to ensure
translation into the language of arbitration in the first instance, pending an
ultimate costs award. It is desirable that interpreters be fully independent as
well as competent. If an interpreter is selected by an individual party there may
be a greater tendency to editorialise, make suggestions or edit responses. (69)
While independence is page "740" desirable, it cannot be guaranteed and a party
may at times wish to utilise a member of its staff for translation purposes. As
noted above in dealing with control over timing, problems may arise where a
party inappropriately calls for an interpreter, seeks to disrupt proceedings by
calling for questions to be repeated or challenges the accuracy of translations.
Counsel should always ensure that each team has someone sufficiently skilled in
the language to object if necessary. Similarly, it would be ideal for the tribunal to
have the language skills to do so and to resolve disputes as to the accuracy of
translations. If that is not the case, the tribunal may need to make rulings, calling
for the parties to indicate where they disagree as to translation and utilise a
tribunal-appointed language expert to resolve that issue.

While simultaneous interpretation seems inherently faster, it may be harder to

identify and resolve disputes about translation and accuracy. Hence the norm is
for consecutive translation, perhaps leaving simultaneous interpretation for
cases where the interpretation skills are high enough and bilingual transcripts
allow for challenges as to accuracy. (70) Live transcripts may also help in some
cases where reading comprehension in the official language is better than oral
comprehension. (71) If a witness ‘testimony subject to translation is particularly
complex and lengthy, it may even be that it could be videotaped and then edited
to exclude the translation pauses, although that would be a problem if
demeanour and veracity still need to be assessed. A tribunal also needs to
ensure that the party having its witnesses’ testimony translated is not
disadvantaged simply because their answers might sound confused and perhaps
even contradictory simply because of comprehension difficulties and lack of
eloquence.

page "741"

1 See, e.g., UNCITRAL Rules 2010 Art. 17(3).

2 Jack J. Coe, ‘Pre-hearing Techniques to Promote Speed and Cost-Effectiveness:
Some Thoughts Concerning Arbitral Process Design’, Pepperdine Dispute
Resolution Law Journal 2, no. 1 (2002): 62.
3 See, e.g., the Supreme Court of Québec case of Silverberg v. Hooper [1990] J.E.

90-437 (S.C.).
4 See generally Michael Pryles, ‘Limits to Party Autonomy in Arbitral Procedure’,

Journal of International Arbitration 24, no. 3 (2007): 327-339.

5 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2747.

6 Documents-only arbitrations are commonly conducted under the rules of the

London Maritime Arbitrators Association.

7 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn
(Oxford: Oxford University Press, 2009), 413.
8 Paolo Michele Patocchi & Harold Frey-Brentano, ‘The Provisional Timetable in

International Arbitration’, in Global Reflections on International Law, Commerce

and Dispute Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald
Aksen et al. (Paris: ICC Publishing, 2005), 563.
9 Rolf Stürner, ‘The Principles of Transnational Civil Procedure: An Introduction

to Their Basic Conceptions’, RabelsZ 69, no. 2 (2005): 211.

10 Judith Gill, ‘Applications for the Early Disposition of Claims in Arbitration

Proceedings’, in Fifty Years of the New York Convention, ICCA International

Arbitration Conference, ICCA Congress Series No. 14, ed. Albert Jan van den Berg
(Alphen aan den Rijn: Kluwer Law International, 2009), 515.
11 Thomas J. Tallerico & J. Adam Behrendt, ‘The Use of Bifurcation and Direct

Testimony Witness Statements in International Commercial Arbitration

Proceedings’, Journal of International Arbitration 20, no. 3 (2003): 297.
12 Judith Gill, ‘Applications for the Early Disposition of Claims in Arbitration

Proceedings’, in Fifty Years of the New York Convention, ICCA International

Arbitration Conference, ICCA Congress Series No. 14, ed. Albert Jan van den Berg
(Alphen aan den Rijn: Kluwer Law International, 2009), 513, 515.
13 Karl-Heinz Böckstiegel, ‘Case Management by Arbitrators: Experiences and

Suggestions’, in Global Reflections on International Law, Commerce and Dispute

Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al.
(Paris: ICC Publishing, 2005), 120.
14 David W. Rivkin, ‘Towards a New Paradigm in International Arbitration: The

Town Elder Model Revisited’, Arbitration International 24, no. 3 (2008): 384.
15 ICC, ‘Final Report on Construction Industry Arbitrations’, ICC International

Court of Arbitration Bulletin 12, no. 2 (2001): 20.

16 Jack J. Coe, ‘Pre-hearing Techniques to Promote Speed and Cost-Effectiveness:

Some Thoughts Concerning Arbitral Process Design’, Pepperdine Dispute

Resolution Law Journal 2, no. 1 (2002): 64.
17 See SCC Rules Art. 20(2). See also HKIAC Rules Art. 15.
18 Jan Paulsson, ‘The Timely Arbitrator: Reflections on the Böckstiegel Method’,

Arbitration International 22, no. 1 (2006): 20.

19 Jack J. Coe, ‘Pre-hearing Techniques to Promote Speed and Cost-Effectiveness:

Some Thoughts Concerning Arbitral Process Design’, Pepperdine Dispute

Resolution Law Journal 2, no. 1 (2002): 66-67. See also section 6.2.8.6.
20 Karl-Heinz Böckstiegel, ‘Case Management by Arbitrators: Experiences and

Suggestions’, in Global Reflections on International Law, Commerce and Dispute

Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al.
(Paris: ICC Publishing, 2005), 124-125.
21 David Hacking, ‘The Effective Arbitrator’, International Arbitration Law
Review 1, no. 7 (1998): 240.
22 Mark Kantor, Valuation for Arbitration (Alphen aan den Rijn: Kluwer Law

International, 2008), 279, n. 791.

23 Neil Kaplan & Karen Mills, ‘The Role of the Chair in International Commercial

Arbitration’, in The Asian Leading Arbitrators' Guide to International Arbitration,

ed. Michael Pryles & Michael Moser (Huntington: JurisNet LLC, 2007), 146.
24 ASA, ‘ASA Special Series No. 31: International Arbitration Do's and Don'ts’,

6th edn, <www.arbitration-ch.org/publications/>, June 2009, 21.

25 See, e.g., ICC Rules 2012 Art, 27; HKIAC Rules Art. 27.1; SIAC Rules Art. 28.1;
Under other rules, it is expressed in a permissive manner. See UNCITRAL Rules
2010 Art. 31.1; ICDR Rules Art. 24.1; ACICA Rules Art. 30.1.
26 Hilmar Raeschke-Kessler, ‘The Arbitrator as Settlement Facilitator’,
Arbitration International 21, no. 4 (2005): 530.
27 See, e.g., Jan Paulsson, ‘The Timely Arbitrator: Reflections on the Böckstiegel

Method’, Arbitration International 22, no. 1 (2006): 19.

28 Karl-Heinz Böckstiegel, ‘Case Management by Arbitrators: Experiences and

Suggestions’, in Global Reflections on International Law, Commerce and Dispute

Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al.
(Paris: ICC Publishing, 2005), 123.
29 Ibid., 123.
30 Jan Paulsson, ‘The Timely Arbitrator: Reflections on the Böckstiegel Method’,

Arbitration International 22, no. 1 (2006): 25, n. 2.

31 David W. Rivkin, ‘Towards a New Paradigm in International Arbitration: The

Town Elder Model Revisited’, Arbitration International 24, no. 3 (2008): 378.
32 Albert Monichino, ‘Stop Clock Hearing Procedures in Arbitration’, Asian
Dispute Review (2009): 79.
33 Jan Paulsson, ‘The Timely Arbitrator: Reflections on the Böckstiegel Method’,

Arbitration International 22, no. 1 (2006): 26.

34 Louis Chang, ‘Keeping Arbitration Easy, Efficient, Economical and User
Friendly’, Dispute Resolution Journal 61, no. 2 (2006): 20.
35 Karl-Heinz Böckstiegel, ‘Case Management by Arbitrators: Experiences and

Suggestions’, in Global Reflections on International Law, Commerce and Dispute

Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al.
(Paris: ICC Publishing, 2005), 124.
36 There is also always a need to consider whether quantitative equality equates

to substantive equality. See Jan Paulsson, ‘The Timely Arbitrator: Reflections on

the Böckstiegel Method’, Arbitration International 22, no. 1 (2006): 24. As noted
in Chapter 1, this idea dates back to Aristotle. See Aristotle, Nichomachean
Ethics, trans. Martin Ostwald (Upper Saddle River: Prentice Hall, 1999),
1131a22-1131a25.
37 Andrew Burr & Pierre A. Karrer, ‘“Chess Clock” Arbitration and Time
Management – Techniques in International Commercial Arbitration: From the
Perspective of the Arbitrator and Counsel’, Construction Law Journal 26, no. 2
(2010): 72.
38 Ibid., 73.
39 Burr and Karrer refer to Brunswick Bowling & Billiards Corporation v.
Shanghai Zhonglu Industrial Co. Ltd, HCCP 66/2007 (unreported) where Hon
Lam J refused to set aside an award where extra time was given to one party for
cross-examination because the other party's witnesses all required translation.
See Andrew Burr & Pierre A. Karrer, ‘“Chess Clock” Arbitration and Time
Management – Techniques in International Commercial Arbitration: From the
Perspective of the Arbitrator and Counsel’, Construction Law Journal 26, no. 2
(2010): 53, n. 19.
40 Karl-Heinz Böckstiegel, ‘Case Management by Arbitrators: Experiences and

Suggestions’, in Global Reflections on International Law, Commerce and Dispute

Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al.
(Paris: ICC Publishing, 2005), 124.
41 Andrew Burr & Pierre A. Karrer, ‘“Chess Clock” Arbitration and Time
Management – Techniques in International Commercial Arbitration: From the
Perspective of the Arbitrator and Counsel’, Construction Law Journal 26, no. 2
(2010): 73.
42 Jan Paulsson, ‘The Timely Arbitrator: Reflections on the Böckstiegel Method’,

Arbitration International 22, no. 1 (2006): 24.

43 Andrew Burr & Pierre A. Karrer, ‘“Chess Clock” Arbitration and Time
Management – Techniques in International Commercial Arbitration: From the
Perspective of the Arbitrator and Counsel’, Construction Law Journal 26, no. 2
(2010): 74-75.
44 Ibid., 75.
45 Jan Paulsson notes that the Iran-US Claims Tribunal at times counted
significant objections against the party making them. See Jan Paulsson, ‘The
Timely Arbitrator: Reflections on the Böckstiegel Method’, Arbitration
International 22, no. 1 (2006): 25.
46 Ibid., 26.
47 Arthur Marriott QC & John Tackaberry QC, Bernstein's Handbook of
Arbitration and Dispute Resolution Practice (London: Sweet & Maxwell, 2003),
55.
48 Jan Paulsson, ‘The Timely Arbitrator: Reflections on the Böckstiegel Method’,

Arbitration International 22, no. 1 (2006): 25.

49 Andrew Burr & Pierre A. Karrer, ‘“Chess Clock” Arbitration and Time
Management – Techniques in International Commercial Arbitration: From the
Perspective of the Arbitrator and Counsel’, Construction Law Journal 26, no. 2
(2010): 76.
50 Ibid., 71.
51 Jan Paulsson, ‘The Timely Arbitrator: Reflections on the Böckstiegel Method’,

Arbitration International 22, no. 1 (2006): 22-23.

52 Marriott and Tacaberry suggest that it should be deducted from the time of

the party having the floor at the relevant moment. Arthur Marriott QC & John
Tackaberry QC, Bernstein's Handbook of Arbitration and Dispute Resolution
Practice (London: Sweet & Maxwell, 2003), 56.
53 Yves Fortier, ‘Oral argument: Report of the Session’, in International
Arbitration 2006: Back to Basics?, ICCA Congress Series No. 13, ed. Albert Jan van
den Berg (The Hague: Kluwer Law International, 2007), 830-831.
54 Ibid., 835.
55 Ibid., 838.
56 Siegfried H. Elsing & John M. Townsend, ‘Bridging the Common Law-Civil Law

Divide in Arbitration’, Arbitration International 18, no. 1 (2002): 59; Catherine A.

Rogers, ‘Regulating International Arbitrators: A Functional Approach to
Developing Standards of Conduct’, Stanford Journal of International Law 41, no.
1 (2005): 53; Catherine A. Rogers, ‘Fit and Function in Legal Ethics: Developing a
Code of Conduct for International Arbitration’, Michigan Journal of International
Law 23, no. 2 (2002): 341; Detlev Vagts, ‘Professional Responsibility in
Transborder Practice’, Georgetown Journal of Legal Ethics 13 (2000): 677.
57 See generally Carrie Menkel-Meadow, ‘Are Cross-Cultural Ethics Standards

Possible or Desirable in International Arbitration?’, in Mélanges en l'honneur de

Pierre Tercier, ed. Peter Gauch et al. (Zurich: Schulthess, 2008).
58 Jean-François Poudret & Sébastien Besson, Comparative Law of International
Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 591, n. 650.
59 François Dessemontet, ‘Draft Code of Conduct for Arbitration Proceedings’,

paper given at the Study Day of the Swiss Arbitration Association (Berne, 21
September 2007).
60 For rules, see UNCITRAL Rules 2010 Art. 19; ICC Rules 2012 Art. 20; SIAC

Rules Art. 19; HKIAC Rules Art. 16; Swiss Rules 2012 Art. 17; SCC Rules Art.
21(1); ACICA Rules Art. 16.1. See also UNCITRAL Model Law Art. 22.
61 ICC, ‘Techniques for Controlling Time and Costs in Arbitration’,
<www.iccwbo.org/uploaded-Files/TimeCost_E.pdf>, 2007, para. 18.
62 CIETAC Rules Art. 71.1. The ICSID Rules provide that if the parties do not

agree on a language, each may choose one of the official languages of ICSID
(English, French and Spanish) and the arbitration is to be conducted in both
languages. See ICSID Rules Art. 22.
63 LCIA Rules Art. 17. Other rules are similar but expressly give the tribunal

power to decide otherwise: ICDR Rules Art. 14; WIPO Rules Art. 40.
64 See ICC Rules 2012 Art. 20.
65 UNCITRAL Rules Art. 19.2; LCIA Rules Art. 17.4; HKIAC Rules Art. 16.2; SIAC

Rules Art. 19.2; ACICA Rules Art. 16.2; SCC Rules Art. 21(2).
66 Jean-Claude Goldsmith, ‘How to Draft Terms of Reference’, Arbitration
International 3, no. 4 (1987): 308.
67 Neil Kaplan, ‘Problems at Both Ends’, in Liber Amicorum Eric Bergsten:
International Arbitration and International Commercial Law: Synergy,
Convergence and Evolution, ed. Kröll et al. (Alphen aan den Rijn: Kluwer Law
International, 2011).
68 See ICC Case No. 4451, cited in Julian Lew, Contemporary Problems in
International Arbitration (London: Queen Mary College, 1986), 57.
69 Andrew Burr & Pierre A. Karrer, ‘“Chess Clock” Arbitration and Time
Management – Techniques in International Commercial Arbitration: From the
Perspective of the Arbitrator and Counsel’, Construction Law Journal 26, no. 2
(2010): 53, 73.
70 Ibid., 53, 74.
71 Ibid., 53, 76. See generally Tibor Várady, Language and Translation in
International Commercial Arbitration: From the Constitution of the Arbitral
Tribunal Through Recognition and Enforcement Proceedings (The Hague: TMC
Asser Press, 2006), 1.
Part II: The Process of an Arbitration,
Chapter 10: Approaches to Evidence
and Fact Finding

Jeff Waincymer,

10.1. Introduction

This chapter, and the following two chapters, look at issues of evidence, that is,
the way that facts are presented by parties and evaluated by a tribunal. These
issues are particularly important as the outcomes in most arbitrations are highly
dependent on factual determinations. (1) For example, claims as to faulty
construction or manufacture, misrepresentation, defences of force majeure, and
the like are fundamentally dependent on the factual findings that a tribunal will
make. This will also be partially so with some jurisdictional questions, such as
claims to extend an arbitral agreement to non-signatories via a group of
companies theory, agency or alleged assignment of arbitral rights. Choice of
applicable law can also be heavily dependent on facts, particularly if a closest
connection test is being employed to decide the applicable law. In most cases, a
tribunal will be asked to make factual determinations about past events. In some
cases it may also be asked to determine a future likelihood, such as a profit
estimate for damages purposes. (2)

There are two broadly related questions in relation to fact-finding. The first is
the nature and amount of material that may be considered by a tribunal. As to
nature, evidence can be categorised in a range of ways. Attention could be given
to the physical form, such as documents versus oral testimony. Chapters 11 and
12 page "743" deal respectively with documentary evidence and general witness
and expert evidence. Attention could be given to its essential nature, whether
objective or subjective. Attention could be given to its relevance to the case,
regardless of form. Attention could be given to its location and to whom control
can be attributed, if not admitted, as fact. There are related questions as to who
is to provide relevant evidence and the ramifications of a failure to do so. This
includes consideration of any independent fact-finding powers of the tribunal.
Where amount of evidence allowed is concerned, this is affected by general
discretions aimed at controlling costs and promoting efficiency, subject to the
right of a party to present its case and deal with that of its opponent. The second
broad issue is the evaluative methodology by which adjudicators come to make
factual determinations where there is conflicting evidence. This includes
consideration of the probative value of each party's evidence. This needs to be
considered in the context of burden, standard and onus of proof. Allied to this,
are the proper inferences, positive, negative and adverse, which a tribunal might
appropriately draw when that evidence does not present a complete picture. In
the context of arbitration, these questions all need to be considered in terms of
party autonomy and the provisions in the lex arbitri or arbitral rules. Rules and
principles may expressly address questions of evidence, or may be implied from
other general principles that indirectly impact on evidentiary matters. These
would include mandatory norms of due process and general procedural
discretions. Relevant mandatory due process norms include the entitlement to
an adequate opportunity to present the party's case, the right to equal treatment
and more contentiously, good faith disclosure obligations.

There may also be evidentiary issues involving persons not party to the arbitral
proceedings, such as prospective witnesses or other third parties who may have
relevant documentary material in their possession or who may be able to attest
to relevant facts. Here there are important questions as to the powers that a
tribunal might have to compel the involvement of such persons. Whenever any
duties or powers are considered, it is also appropriate to consider the
ramifications if parties do not comply with tribunal directives, as this will impact
upon the practical utility of the powers.

As is the case with most procedural topics being considered, there are important
and potentially conflicting policy questions that arise. For example, should there
be any limit on the amount of evidence that each party may wish to present,
given their due process entitlements? How should a tribunal deal with evidence
that may be relevant but which is suspected of being unreliable? What inferences
are appropriate, both positive and negative, as a result of material presented and
as a result of material the parties choose not to disclose?

There are a myriad of other similar questions in relation to which contentious

choices must be made, either in arbitral rules or under the application of arbitral
discretions. When legal systems determine how truth is to be assessed and how
these questions are to be answered, choices are inevitably made between
conflicting values. As always, questions of fairness and efficiency need to be
balanced. Differing conceptions of fairness must also be considered. To some, the
pursuit of truth as an element of fairness implies that a correct outcome is page
"744" the most just. This approach would see few, if any, limits on the amount of
evidence. Yet concerns with efficiency suggest that some reasonable limits
should be imposed. Excessive cost could itself be unfair. There are also
libertarian concerns as to whether a party, through disclosure, is required to
assist its opponent in developing its case. Some would rank truth higher than
liberty, while others would take the opposite perspective, arguing that a claimant
should be required to have sufficient evidence before interfering with the liberty
of a respondent.

All legal adjudicatory systems make choices as to these important policy

questions through evidentiary rules dealing with admissibility and weight of
evidence and burden and standard of proof and also through the controlling
discretions given to adjudicators. Once a system selects a set of evidentiary rules,
it thus identifies a methodology by which an adjudicator determines truth in the
face of conflicting factual assertions by the parties. (3) Because we must find
compromise policy positions in any set of rules, it is particularly important to
understand that differing rules about the nature of relevant evidence and the
inferences to be drawn will have a significant impact on the likely factual
conclusions. Identical scenarios in different legal systems can legitimately be
decided differently. Similarly, discretionary decisions by an adjudicator can
impact upon the body of evidence available for analysis and hence the likely
outcome. Once it is understood that any general rules and procedures are
compromises, it must also be acknowledged that their application may be unfair
to one or both parties in a particular case. (4) Hence no optimal model can readily
be discerned. It is important to understand that there is no absolute and
objective truth discerned by any legal fact-finding methodology. Thus, fact-
finding in most instances of legal adjudication is very different to fact-finding in
physical science experimentation. (5)

There are a number of important conclusions that should be drawn from these
observations. First, because there must always be choices between tenable
alternatives and trade-offs between fairness and efficiency, extreme criticisms of
any particular model or arbitral decision based on assessment of failings as to a
single goal, are usually misguided. This is the typical logical flaw in most
annulment or enforcement challenges that are based on evidentiary issues. A
second and related observation is that criticisms from the perspective of one
legal tradition addressed against another, can too easily hone in on a negative
aspect of the latter's trade-offs without a proper cost/benefit comparison of
each. The following section considers comparative perspectives in that light.

page "745"

10.2. Comparative Law Perspectives

Before considering the provisions in arbitral statutes and rules, it is valuable to

analyse the issues from a comparative perspective, particularly as differences
between legal families are probably greatest in relation to evidentiary matters. (6)
That poses problems for international arbitration where the parties come from
different legal systems. In such an environment, arbitration statutes and rules
are unlikely to be too prescriptive, as this would appear biased against certain
legal families. Tribunals must therefore routinely make factual decisions in the
absence of any clear guidance in the rules. Critics then raise issues about
inconsistency and the application of unconstrained discretions.

The major differences in approach to evidence between legal families arise in

relation to primary responsibility for evidence presentation, rules on
admissibility, obligations to produce evidence of benefit to the opposing party,
the type of witnesses that may present evidence, the role of experts and methods
and standards of proof. Where control over evidence is concerned, the key
difference between common law party-led or adversarial systems and civilian
judge-led or inquisitorial systems is that the former historically left it to the
parties to determine what evidence to present, while the latter involved the
adjudicator giving guidance as to the material thought most suitable. Common
law systems tend to oblige parties to present all relevant evidence, including that
which is adverse to their own interests, in part to deter misleading selectivity.
Such systems also provide rights to call for document production by the
opposing party so that all relevant material is before the tribunal. Civilian
systems, however, leave it to the parties to determine what information they
wish to rely upon and have that supplanted by evidence called for by the tribunal
and do not require adverse evidence to be produced as a matter of course. The
systems will also differ in relation to the type of evidence that may be presented.
Common law systems have complex exclusionary rules where the potential
prejudicial value of evidence is presumed to significantly outweigh probative
value, in part because of historical use of lay jurors as fact finders. Common law
systems have also relied heavily on oral testimony, tested by cross-examination,
while civilian systems tended to give greatest weight to contemporaneous
documents.

As is the case with all comparative law questions of significance, differences are
often oversimplified. Not all common law or civilian jurisdictions take the same
approach. For example, there are important differences between American and
other common law systems. Where the civil law is concerned, differences
typically arise between Germanic and Latin systems and even within such
systems. (7) In addition, differences on a particular issue often flow logically from
choices made on other aspects of evidentiary policy. For example, adversarial
page "746" systems, which rely heavily on the material presented by the parties,
will more typically impose ethical obligations on them to produce all evidence,
including adverse evidence. Similarly, they are more likely to allow for document
production requests so that one party imposes checks and balances on
potentially inadequate disclosure of the other. In other cases, supposed
differences in a particular area are counterbalanced in others. For example,
civilian legal systems have no equivalent of the parol evidence rule which
prevents oral evidence being used to alter the meaning of written agreements,
but distrust oral evidence per se, hence ending up in a not dissimilar position.
While civilians may be concerned with broad discovery rights, particularly as
against third parties, other processes may lead to similar results, for example,
where a court-appointed expert is invited to interview such persons and
consider their documentation.

At other times, suggested differences are misleadingly simplistic. A description of

the common law as being predominantly concerned with objective truth and
hence supporting the broadest range of discovery rights needs to be tempered
with the adversarial disposition which encourages the retention of key
evidentiary material until it can be presented with the most force and with the
least opportunity for rejoinder. The difference to the civilian approach is also
less meaningful if civilian judges are prepared to draw adverse inferences from
failure to produce documents requested by the judge, including adverse material.
(8) Similarly, the supposed difference between each system's preference for

either contemporaneous evidence or oral testimony is largely overstated. All

systems would see contemporaneous evidence as highly valuable. It will also
often be the case in complex commercial matters that a combination of oral and
written evidence would be the only way to truly understand the issues in
dispute. If the essential dispute is as to what occurred in an oral discussion, there
is no alternative but to call and test the witnesses as to their recollections. (9)
Consider the case of a joint venture dispute where one party alleges oral
misrepresentations prior to entering into the agreement. Each agree that
discussions were held, neither kept written records but disagree as to what was
said. Even if each can be viewed with suspicion, (perhaps an unfair presumption
as to human veracity and honesty), a priori, the concerns with each are
counterbalanced. An experienced arbitrator ought to be able to hear what they
each say about the conversation and form a view about it.

There are also arguable inconsistencies in the way each system preferences
probity as against prejudicial value in relation to various types of evidence. This
is not to say either or both are wrong in their range of choices but at least it can
be said that neither system is a likely optimal model for international arbitration
if there is no rational basis for internal inconsistencies. For example, it is
interesting to juxtapose the page "747" civil law exclusion of parties as witnesses
with the common law's exclusionary rules of evidence such as hearsay. In the
common law, free allowance of party witnesses does not sit well alongside
exclusionary rules against hearsay in cases where juries are not involved.
Civilian concerns for the reliability of party-appointed experts because of a fee
relationship, may be hard to reconcile with support for party-appointed
arbitrators, particularly in ad hoc arbitrations. Concerns for discovery of adverse
material need to be balanced against an inquisitorial judge's right to ask for this.

In reality, the systems started from more polarised positions but tend to
gravitate to more common approaches over time. This typically arises because
each system is dealing with the same complex questions but took a different
initial view of the appropriate way to reconcile conflicting policy values. Each
makes compromises that are in part based on other cultural values and
procedural features of the system. In seeking compromises, no system uniformly
and consistently prefers one value over others. Each is typically dissatisfied with
the resulting position and tries to find a more balanced resolution. For example,
while the common law typically left it to the parties to manage the evidence and
the process, the judiciary and legislature soon found that this was inefficient and
a drain on public resources. Hence, common law civil procedure reforms have
heightened the role of case management by the judiciary. (10) Such developments
at the domestic level can inspire arbitral rule reform or the exercise of arbitral
discretions, subject to the comparative law caution not to readily presume that
transplants of regulatory models will be desirable.

Finally there may also be merit in a comparative perspective that considers what
the parties from different jurisdictions would prefer and not simply what their
lawyers have been brought up to view with suspicion. That could well be
relevant in a consent-based system such as arbitration and could lead to quite
different attitudes of parties as opposed to counsel. For example, many common
law clients may be surprised and disappointed to hear about the hearsay and
parol evidence rules while civilian clients may be surprised and disappointed to
hear that they are not entitled to give evidence, being presumptively
untrustworthy. Arbitration seeks to find a compromise position that best fits
with the reasonable expectations of the parties themselves. Because it is at least
arguable that each system has some inconsistent responses to the inevitable
compromises, there is much scope for international arbitration to help develop a
more coherent pattern.

10.3. Rights and Powers of the Parties and Tribunal as to Evidentiary Issues

As with each area of arbitral process, there is a need to consider the rights, duties
and discretions available to each person involved and determine how conflicting
page "748" rights are to be resolved. As always, there is a need to consider party
autonomy and the law governing the arbitration. Where the latter is concerned,
there is an important gateway question of characterising evidentiary issues as
either substantive or procedural, as this will determine which provisions are
relevant. This will also have implications for annulment and enforcement rights.
This is addressed in the next section.

10.3.1. Are Evidentiary Questions Matters of Substance or Procedure?

Different legal systems do not all classify evidentiary issues the same way. The
common law tended to see them as procedural in nature, while some civilian
systems treated them as substantive questions. Most now acknowledge the
difficulties in classification, the possible dual nature of at least some evidentiary
questions and the impact of particular circumstances on the ultimate proper
classification. For example, the Iran–US Claims Tribunal in Teheran
Redevelopment Corp. commented that notwithstanding a governing law of the
contract, ‘it is arguable that the type of evidence admissible to establish a
contract is a procedural or evidentiary matter.' (11) The tribunal considered part
performance and estoppel by conduct as examples of the latter, although not all
would agree.

The procedure/substance distinction may also be impacted upon by the

presence or otherwise of certain substantive rights to information applicable
between the parties or between the parties and public institutions. For example,
each party in a joint venture will typically have a right to all accounts and
perhaps to have them audited. Direct substantive provisions of this nature
relating to evidentiary entitlements are quite distinct from general approaches to
evidence pertaining to other substantive obligations. The distinction is not
always clear-cut. Part of the complexity is that the substantive law will always
indicate which facts are in issue. This will also be the case with certain specific
contractual terms. (12)

Classification can affect the ambit of a tribunal's powers. To the extent that
evidentiary issues are a substantive matter, a tribunal might be under a more
constrained discretion than is the case with procedural matters. This is because
provisions dealing with choice of substantive law may require a conflicts
methodology, albeit allowing the tribunal to choose which conflicts rules to
apply. (13) page "749" Conversely, substantive decisions are generally less open to
challenge on annulment or enforcement applications than procedural
determinations. However, a substantive decision is more likely to meet the test
of what constitutes an award, hence being enforceable or challengeable as the
case may be. (14) The substance/procedure distinction is considered further
below in sub-topics where it has particular importance.

10.3.2. Evidentiary Principles under Lex Arbitri and Arbitral Rules

10.3.2.1. General Provisions

Specific rules will be addressed throughout this and the next two chapters on a
topic-by-topic basis. In this section, only general observations are made. Most lex
arbitri and arbitral rules say very little about matters of evidence. A typical
example is Article 25(1) of the ICC Rules 2012 which merely provides that ‘the
arbitral tribunal shall proceed… to establish the facts of the case by all
appropriate means.’ Some rules confer broad discretions on arbitral tribunals
subject to the contrary agreement of the parties. Article 19(2) of the UNCITRAL
Model Law provides that ‘(t)he power conferred upon the arbitral tribunal
includes the power to determine the admissibility, relevance, materiality and
weight of any evidence.’ (15) Article 34(1) of the ICSID Arbitration Rules indicates
that ‘(t)he Tribunal shall be the judge of the admissibility of any evidence
adduced and of its probative value.’ The IBA Rules of Evidence 2010 also provide
for a residual broad discretion where the rules are silent and the parties have not
agreed otherwise. (16) The LCIA Rules expressly indicate that the tribunal has a
power to decide whether or not to apply any strict rule of evidence. (17) Some
rules make clear that a tribunal may call for evidence and may undertake
inspections as it sees fit. (18)

It is commonly accepted that broad discretions cover all necessary evidentiary

powers regardless of whether the rules expressly refer to particular matters
such as documents, witnesses, experts or inspections. The International Court of
Justice has considered that a tribunal generally has a broad discretion as to the
way to approach the evaluation of evidence. It has stated that ‘(t)he appraisal of
the probative value of documents and evidence appertained to the discretionary
power of page "750" the arbitrator and is not open to question.’ (19) That would
not be so where the particular rules prevent a particular evidentiary approach.
While evidentiary discretions are interpreted broadly, even express provisions
need to be considered alongside broader discretions such as those contained in
Article 19 of the UNCITRAL Model Law, (20) and obligations under mandatory
norms discussed in the following section. The express rules also need to be
looked at alongside the general question as to whether a tribunal wishes to be
proactive in relation to evidentiary matters and the way tribunals will balance
competing policy goals when exercising discretions or making evidentiary
rulings. Party autonomy is also important as always. This is discussed in section
10.3.2.3 below.

10.3.2.2. Mandatory Norms

While tribunals have broad discretions subject to party choice, these are not
without constraints. Concern must always be had for mandatory procedural
rules such as Article 18 of the UNCITRAL Model Law requiring equal treatment
and an adequate opportunity for presentation of one's case. (21) Different lex
arbitri use differing terms to describe the degree of this entitlement. While some
commentators have raised concerns as to the difference in terminology between
a ‘full' opportunity to present a case (22) and expressions such as a ‘reasonable
opportunity’, (23) nothing should turn on this as the key word is ‘opportunity’. (24)
A full opportunity is not an open-ended one.

While such mandatory provisions are always paramount, they are difficult to
apply. One reason is that they can too easily be used casuistically by parties to
argue for and against any evidentiary ruling being sought. This is because where
evidentiary issues are concerned, the right to an opportunity to present a case
and the right to equal treatment can be argued to conflict when the two parties
believe that there should be fundamentally different ways in which to provide a
reasonable opportunity. An example would be if a tribunal is asked to consider
whether to order production of a document held by in-house counsel, where
legal professional page "751" privilege is claimed and the parties come from legal
systems that take different views on the issue. Whichever decision the tribunal
comes to, the unsuccessful party might assert a violation of one of these
fundamental norms. While a reasonable annulment or enforcement court would
not accept most such arguments, the mere availability of a tenable challenge can
influence proceedings.

10.3.2.3. Tribunal Discretion

Specific aspects of discretionary power will be discussed on a topic-by-topic

basis. Some more general observations are made at this stage. As noted above,
tribunals are given a broad discretion as to admissibility, relevance, materiality
and weight of evidence. A corollary is that a review court should not readily
interfere with the discretionary determinations made but will only limit itself to
claimed breaches of due process in relation to that activity. The most challenging
issue will be limits on evidence for efficiency reasons. Some efficiency-based
discretionary decisions will be non-contentious. For example, tribunals could
consider bifurcating evidentiary stages where decisions on certain facts will
significantly impact upon the type of evidence and extent of evidence necessary
at later stages. This is simply a logical corollary of the common bifurcation
between liability and quantum and does not diminish the right to be heard.
Where more contentious discretions are to be employed, a tribunal will naturally
consider due process, fairness, a sufficient opportunity of presenting a case,
emerging best practice and the reasonable expectations of the parties. The latter
could, for example, include considering the solutions in the national laws of
evidence common to both parties. (25) This could be justified on the basis of the
reasonable expectations of the parties based on their familiarity with
adjudicatory processes.

As always, discretions must be exercised within the parameters of the rules

granting the discretion and subject to mandatory due process norms. Because
party autonomy is paramount and parties can have very differing views on
evidentiary matters, it is desirable to seek the input of the parties and wherever
possible have agreement between them to obviate the need for a discretionary
determination. This is particularly important as the kinds of measures a tribunal
might employ to promote efficiency can invite disappointed parties to consider
challenges based on interference with their ability to fully present the case. Such
agreement can at the very least be an a priori agreement to accept a tribunal's
discretionary determination. While that is presumed in any event under general
rules, this could be made clearer in contentious areas. (26) However, in some
extreme cases, a tribunal may ignore an agreement of the parties if it offends
principles of justice and page "752" efficiency. (27) A tribunal should also warn the
parties of the ramifications of failing to meet evidentiary obligations. This will
hopefully promote greater compliance and also reduce the opportunity for
challenge.

10.3.3. Good Faith and Evidence

Preamble 3 to the IBA Rules of Evidence 2010 indicates that:

the taking of evidence shall be conducted on the principle that each Party shall
act in good faith…

Rule 34(3) ICSID Arbitration Rules provides that ‘(t)he parties shall cooperate
with the Tribunal in the production of the evidence…’ Article 15.7 of the Swiss
Rules 2012 provides that ‘(a)ll participants in the proceedings shall act in good
faith…’. Good faith can also be said to arise as an implied element of the
agreement to arbitrate. (28) Separate obligations may apply to legal counsel. (29)

10.3.4. Public International Law and Evidence

Another source of guidance might be from the practices of other international

adjudicatory bodies and the rules and practices under public international law.
The approach to evidence and fact-finding in public international law may be
directly relevant in investment arbitration. Evidentiary norms in public
international law are also generally consent based and derive from statute and
treaty. (30) It is also possible that court behaviour over time can lead to the
development of customary norms. Brower suggested that a lex evidentia may
thus develop. (31) In a detailed study some time ago, Sandifer suggested that there
has page "753" developed ‘what is tantamount to a customary law of evidence’,
(32) although he is also quoted below to the effect that ‘each tribunal tends to be a

law unto itself…’, (33) propositions that do not sit well together.

Even if public international law principles do not directly apply, cases from that
arena may provide useful guidance. (34) The influence is likely to be greatest
where at least one party is a State. General deference to a sovereign party will
often see adjudicators being reluctant to severely limit evidence provided. Public
international law cases can also give guidance as to the way to treat
confidentiality claims by sovereigns and the possibility of drawing adverse
inferences in such circumstances.

10.3.5. Party Autonomy

Regardless of whether evidentiary issues are procedural or substantive, party

autonomy is paramount, save where mandatory norms are concerned.
Autonomy is not unlimited however. A tribunal is entitled to draw its own
conclusions of fact notwithstanding any contrary agreement by the parties. (35)
Yet it is rare for parties to expressly agree on evidentiary provisions in their
arbitration agreements. Agreement is too often unlikely once a dispute has
arisen as some evidentiary issues will favour one party over another. Parties can
broadly agree on evidentiary principles or agree as to specifics, for example, as to
a documents-only adjudication. One option as noted below is to consider
whether to agree on the application of the IBA Rules of Evidence 2010.

Any agreement between the parties should be accepted unless it offends against
mandatory due process norms which cannot be waived. An agreement that
increases the tribunal's workload over that contemplated in the original contract
of engagement would also not bind the tribunal. A more challenging issue is
where the parties want to agree to something that the tribunal thinks is
unsuitable, for example, a documents-only adjudication when the key issue is
what was said at a meeting where minutes were not kept. These scenarios were
considered more fully in section 9.2 An arbitrator must be able to give contrary
directions in some cases at least.

Another contentious aspect of party autonomy is when it is not clear whether

there has been an agreement. One example might be a dispute about a
potentially pathological arbitration clause where it is not clear which
institutional or ad hoc rules have been selected. This will simply be a matter for
proper interpretation of the arbitration agreement. If there is a disagreement
about a separate alleged page "754" agreement on procedural matters, typically
by way of discussion between counsel after the dispute has commenced, it would
be difficult for a tribunal to make a determination. In most cases a tribunal would
feel bound to conclude that there is insufficient evidence of any agreement,
although in some cases, correspondence would allow for a contrary conclusion.

A more contentious question is whether a preference for a particular evidentiary

approach can be justified under an implied consent analysis as to the way
business people could be expected to want discretions to be exercised. For
example, does implied consent point to a preference for the fully objective
document production approach over the libertarian perspective? Such an
argument would be difficult to maintain in all situations, although there are some
modest arguments in favour of the objective model. First, as much international
business is between parties with ongoing business relationships, one might
expect that the business people involved would prefer the correct outcome and
would see any libertarian and gamesmanship perspectives as unduly zero sum in
nature. (36) Such an analytical perspective might seem to be supported by some
recent empirical data (37) which found that the correct outcome was by far and
away the most important objective of the business people canvassed, although
admittedly, there are some contradictory studies. (38) The argument would
proceed that most business people behind a veil of ignorance as to whether they
will ultimately succeed in the proceedings, would prefer truth to dominate. The
presumption is that a just resolution of business disputes is likely to be a win-
win outcome in the context of a long-term business relationship. This is
particularly so as the original business deal was intended to be win-win. Hence
any variation from the original intention via alleged breach should bring the
parties back into the original position. That original intent itself might be
supported under an implied consent paradigm. A dispute settlement outcome
that allows the ‘wrong' party to win, is a disincentive to business transactions or
adds costs through parties pricing the risk factor into their future commercial
dealings. The logic would depend on the tribunal being able to keep the cost in
proportion to the value of the claim at the same time as pursuing the truth.

10.3.6. Harmonisation Initiatives

Some argue that without standardised rules, ad hoc international tribunals will
too readily fall back on their own experiences and preferences. One earlier study
of the evidentiary approaches of international tribunals considered that ‘(e)ach
tribunal tends to be a law unto itself, the rules adopted and applied for the
occasion being to page "755" a considerable degree determined by the legal
background of the members of the tribunal.’ (39) Others suggest that there is
already convergence of practices. Fouchard considered that international
arbitrations tend to follow an increasingly uniform pattern, with the written
stage largely based on continental procedure and with the oral stage influenced
to a greater extent by common law techniques. (40) Dezalay and Garth take a
different view, arguing that Anglo-American lawyers have shaped arbitration
proceedings in the context of their legal cultures. (41) Others argue that there is an
undesirable obsession with creating complex rules that can never be drafted
with appropriate precision. The suggestion is that the real need is to employ
expert, practical and ethical arbitrators whose selection and approach will be
tailored to the needs of the dispute and the parties and not look to rules for the
promotion of optimal practice.

The very fact that there is an ongoing debate suggests that procedural and
evidentiary behaviour is not uniformly ideal and that further analysis should
occur. Issues such as this will typically be debated by practitioners from differing
legal families in symposia, (42) which may then lead to a more harmonised
perspective. Multi-person tribunals will often bring together arbitrators from
different legal families, requiring a compromise approach in any event. The
approach taken by other tribunals such as the Iran–US Claims Tribunal is also
typically influential. (43) The view of this author is that too often the debate
between those favouring rules and those favouring discretion gives insufficient
attention to the best way to respond to our natural wish for potentially
conflicting virtues, in this case certainty versus flexibility. Each has advantages
and disadvantages standing alone. An ideal compromise will be some form of
guided flexibility. This would entail enough flexibility to allow for the nuances of
each case to be dealt with fairly and efficiently, but with enough guidance to
promote predictability and consistency of tribunal behaviour and provide an
educative function for less experienced stakeholders. (44)

While the conceptual debate is still important, it is also necessary to consider the
utility of current harmonisation or standardisation initiatives. In addition to
directions in lex arbitri and arbitral rules, a number of institutions have
attempted page "756" to provide evidentiary models or guides. If agreed to by
the parties, they will be binding unless they offend mandatory norms. Even if not
agreed to, they may be highly influential as to the way arbitrators exercise
discretions if it is felt that they encapsulate best practice or converging practice.
Once again, the details in such instruments will be considered in specific sections
below.

10.3.6.1. The IBA Rules of Evidence 2010

The IBA Rules on the Taking of Evidence in International Arbitration (the ‘IBA
Rules of Evidence 2010’) (45) adopted on 29 May 2010 and the predecessor 1999
Rules have become particularly influential. (46) Because the IBA is a private body
that does not conduct institutional arbitration, the rules cannot have direct
binding force without any agreement to that effect by the parties, although they
might also be an informal guide to the exercise of broad discretions. If the parties
agree on the application of the IBA Rules, either in their arbitration agreement or
up to and including the time of the hearing, then a tribunal would be bound to
follow this. Agreement could be direct or indirect. As an example of the latter, the
ACICA Rules expressly provide that the arbitral tribunal shall have regard to but
not be bound by the IBA Rules of Evidence 2010 in their current version as at the
commencement of the arbitration. (47) The ICC publication Techniques for
Controlling Time and Cost in Arbitration also recommends consideration of the
IBA Rules of Evidence. (48)

Where the parties agree to the IBA Rules the rules themselves state that they are
deemed to have agreed to the version current on the date of such agreement,
absent a contrary intention. (49) The parties may prefer to expressly indicate that
they are agreeing to the rules in force at the commencement of the arbitration in
order to use the most up-to-date version. (50) If the parties proscribe their use, the
tribunal should respect that choice, although a broad discretion over evidence
might lead to similar solutions on a topic-by-topic basis unless these solutions
are also expressly or impliedly proscribed by the parties. The IBA Rules will also
not apply if they conflict with mandatory rules or specific processes expressly
selected by the parties. A more difficult question is what use can be made of the
IBA Rules where one party alone objects to their use. Here they can still be a
guide to discretionary determinations but do not apply as of right.

page "757"

If they apply, they will still be subject to mandatory norms that will shape the
exercise of discretions. They may also be impacted upon by the rules of any
jurisdiction where court assistance is sought and codes of ethics arguably
applicable to counsel. In some cases, national codes of ethics might expressly
exclude arbitration. In others they might expressly apply. In many cases they are
silent, leaving it open to debate as to their intent. If the Rules apply by agreement
but there is a conflict between them and arbitral rules chosen by the parties, it is
not clear which should prevail, as the parties have selected conflicting broad
ranging provisions. Typically this conflict will be unintentional. If the parties
cannot clarify the inconsistency, the tribunal will need to make a determination.
Because the Rules of Evidence are more specific, one might conclude that
selection by the parties evidences that they should prevail. Ideally, the parties
should seek to resolve this by agreement. The Commentary of the IBA Rules of
Evidence Review Sub-Committee suggests that absent an agreement ‘the arbitral
tribunal shall try to harmonise the two sets of rules to the greatest extent
possible.’ (51)

While the Rules are an important option, they leave many questions open as the
Rules do not seek to resolve many of the more contentious evidentiary issues
about which legal families have differed, although some key preferences are
evident. Even where they are utilised, they will commonly refer to conflicts
principles such as ‘applicable’ laws of privilege. Hence they contain some
contentious and uncertain elements which are also the subject of debate. The
Rules also rely heavily on some broad concepts rather than detailed
prescriptions. The overall flavour of the Rules is to promote openness,
collaboration, fair treatment and efficiency. Article 2.3 now indicates that the
tribunal ‘is encouraged’ to identify issues for the parties ‘as soon as it considers it
to be appropriate…’. While couched in diplomatic language, including this in an
Article rather than in the Preamble as was the case with the 1999 Rules suggests
a wish to further promote such behaviour. The Rules now expressly refer to the
tribunal's duty to consult with the parties so as to provide efficient, economical
and fair processes for the taking of evidence. (52) Consultation obligations are
mandatory and should occur at ‘the earliest appropriate time …’. (53) A greater
concern for efficiency is also supported by Article 2.2(e) which includes for
possible consideration, ‘the promotion of efficiency, economy and conservation
of resources in connection with the taking of evidence.’ Paragraph 3 of the
Preamble makes clear that the Rules are not intended to limit tribunal flexibility.
The Preamble also notes that the paramount principle is that each party shall be
‘entitled to know, reasonably in advance of any Evidentiary page "758" Hearing
or any fact or merits determination, the evidence on which the other Parties
rely.’ The Rules also expressly provide that costs discretions may apply if a party
has failed to conduct itself in good faith in the taking of evidence. (54)

Lew, Mistelis and Kröll suggest that if parties wish to adopt the IBA Rules or
similar, they should provide for this in their arbitration agreement. (55) Bühler
and Dorgan argue against adoption of the IBA Rules in pre-dispute arbitration
agreements as this might remove flexibility and discretion, (56) although that
comment was made prior to evaluation of the 2010 Rules. A contrary argument
is that there is sufficient flexibility and discretion within the current Rules, so
prior agreement would only be inappropriate if the Rules unduly favoured one
side, which should not be the case.

To some, the development of these rules by a task force comprising leading

practitioners from differing legal families is a clear example of emerging best
practice. (57) To others it reflects the approaches taken by skilled arbitrators. (58) It
has also been suggested that the vast majority of arbitrators will include the
rules in terms of reference or in subsequent directions. (59) Others are more
critical. One commentator has suggested that ‘the IBA Rules constitute a
misguided combination of various aspects of different traditions.’ (60) At the
extreme, some would see these rules as an example of common law domination
over long-standing principles of civilian evidence law in relation to matters such
as document production generally, the specific case of internal documents, the
case of cross-examination page "759" and the entitlement of parties as witnesses.
Conversely, it has been observed that the majority of members of the working
party for the 1999 IBA Rule revision were non-common lawyers. (61) It is true to
say that the express reference to a range of evidentiary matters makes greater
inroads into the traditional civilian approach. Where common law principles are
concerned, the rules simply remove the strict application of inadmissibility
criteria, although many of these are again articulated as grounds upon which
document production can be refused. The point to be taken from sections 1.8 and
10.2 above is that simple concern for the more questionable aspects of each legal
family or concern for which historical tradition comes to dominate is less
important than concern for finding compromises that lead to fairness and
efficiency in the context of the aspirations of the users of international
arbitration.

Even where no agreement is present, a significant number of tribunals will utilise

the rules simply because of their status. Over time, broad use will constitute a
consistent body of arbitral practice that can only be self-supportive. Counsel in
arbitrations will also naturally look to the IBA Rules as part of their evidentiary
submissions, so they may influence the matters raised before a tribunal even
before the arbitrator would naturally consider the question. Because the Rules
contain so many discretionary elements, it will always remain the case that
arbitrators even bound by the rules can take differing views about the trade-offs
between truth, liberty and efficiency on a case-by-case basis.

10.3.6.2. UNCITRAL Notes on Organising Arbitral Proceedings

These Notes were finalised by UNCITRAL at its 29th session in 1996. The Notes
aim to assist arbitration practitioners by listing and outlining questions on which
decisions may be useful. They are non-binding and the tribunal remains free to
use the Notes as it sees fit. In particular, they are not appropriate to be used as
arbitral rules. (62) There are also the UNCITRAL Draft Guidelines for Preparatory
Conferences in Arbitral Proceedings. (63)

10.3.6.3. Chartered Institute of Arbitrators Protocol for the Use of Party-

Appointed Expert Witnesses in International Arbitration

The Protocol was developed in 2007 and aims to be a comprehensive outline of

the issues to be dealt with for party-appointed experts. It is not intended to cover
page "760" tribunal-appointed experts or single joint experts. It is available to be
used in whole or in part.

10.3.6.4. ICC Techniques for Controlling Time and Cost in Arbitration

The ICC Commission on Arbitration has produced a report outlining

recommended techniques for controlling time and cost in arbitration. (64) The
report is designed to assist tribunals and counsel. It also allows for possible
agreement by the parties. The techniques are not intended to be prescriptive or
regarded as a code of best practice, but again are ideas to assist in arriving at
efficient procedures. However, key elements have now been incorporated into
Appendix IV of the ICC Rules 2012 and are hence applicable in ICC arbitration.

10.3.6.5. CPR/ICPR

Other models include those promulgated by CPR and ICPR. These include the
ICDR Guidelines for Arbitrators Concerning Exchanges of Information and the
CPR Protocol on Disclosure of Documents and Presentation of Witnesses in
Commercial Arbitration. (65)

10.3.6.6. ALI/UNIDROIT Principles and Rules of Transnational Civil

Procedure

There have also been transnational civil procedure initiatives not directly
applicable to arbitration that can provide guidance. The ALI/UNIDROIT
principles were prepared by a joint American Law Institute and UNIDROIT study
group and were adopted by the Governing Council of UNIDROIT in 2004. (66)

10.3.6.7. Further Initiatives

One problem if there are too many specific rules available and variances between
them, is that debate may ensue as to whether inferences of party intent can be
drawn by a consideration of the differences between rules selected and other
options. This would not be desirable. A simple approach to harmonisation aimed
at avoiding this problem is to try and use similar language in different
instruments. For example, the Chartered Institute of Arbitrators Protocol for the
page "761" Use of Party Appointed Expert Witnesses in International Arbitration
tried to follow the language in the IBA Rules of Evidence 2010 wherever
possible. (67)

10.4. Legal Approaches to Fact-Finding

The following sections look separately at questions of the burden and standard of
proof. These are key aspects of the methodology by which legal systems resolve
disputes on factual questions. These principles also need to be considered in
conjunction with other evidentiary policy choices, including the independent
fact-finding powers of the tribunal and the duties on the parties as to document
presentation and document production. (68) Problems of proof are compounded
when the tribunal has a discretion as to what kind and amount of evidence to
admit. A tribunal that restricts a party's right to tender particular evidence and
then concludes that it has not met its burden of proof, may be rightly or wrongly
accused of failing to afford the party an opportunity of reasonably presenting its
case. In addition, because the parties cannot know at the outset how proactive an
arbitrator will be and what balance he or she will seek to achieve between truth
and efficiency, they cannot know at the outset what evidentiary material they are
likely to need to succeed. This suggests that any tribunal should give as much
guidance as it can at the earliest stage as to the way it intends to approach
evidentiary issues.

10.4.1. Burden of Proof

There is commonality between legal families and within public international law
as to the application of the burden of proof in any case. It is widely accepted that
a party seeking to rely on a particular fact has the burden of establishing it. (69)
More broadly, a party requiring some determination from a tribunal has the
burden of establishing its entitlement, although Jarvin suggests that ‘(t)he strict
distribution of the burden of proof applied by national courts does not apply in
international arbitration.’ (70) Burden of proof relates to factual matters and not
questions of legal interpretation, although the party with the burden must be
able to identify a legal page "762" basis for the claim. (71) Nevertheless, if a
tribunal is asked to interpret a particular law, it will seek to do so as best it can
and will form a reasoned view without simply relying on questions of burden.

The burden of proof never shifts, (72) although the more evidence that the party
with the burden has in its favour, the more the opposing party must seek to
counter it. (73) This is discussed further below in relation to onus of proof. (74)

The commonly accepted view as to burden of proof is enshrined in various

Roman law maxims, namely, ei qui affirmat non ei qui negat incumbit probatio
(the onus of proof is on the person who affirms and not on the one who denies);
actori incumbit probatio (the burden of proof is carried by claimant); actore non
probante reus absolvitur (if the plaintiff cannot prove the case the defendant is
absolved); and reus excipiendo fit actor (the defendant, by raising an exception or
pleading, becomes a plaintiff). The references to claimant/plaintiff and
respondent/defendant page "763" can be misleading as it is clear that the burden
is on the party seeking to prove some fact. Thus claimants must prove claims, but
defendants then must prove defences and counterclaims or set-off rights.

Provisions in arbitral rules describing burden of proof are not intended to

override any provisions in the applicable substantive law. Provisions in arbitral
rules simply restate the general proposition but do not seek to vary specific rules
on matters such as mitigation, foreseeability and force majeure where
substantive rules will often indicate which element must be proven by which
party. The burden and standard of proof can also be impacted upon by the way
substantive rights are constructed. Sometimes the contract terms will
significantly alter what would otherwise need to be proven. For example, a
contract might give a buyer the complete right to reject goods for lack of quality
without being subject to challenge. A multinational supermarket chain may be
able to negotiate such an entitlement with suppliers. Here the buyer would
simply need to establish the fact of rejection and not a legal basis for doing so. At
most, a tribunal might interpret that provision as being subject to a duty of good
faith. At other times the applicable law will have a similar impact.

Arbitral rules rarely articulate the principles of burden and standard of proof. It
is considered part of a tribunal's inherent jurisdiction to consider such matters.
Some arbitral rules do make express reference to burden of proof. For example,
Article 27.1 of the UNCITRAL Rules 2010 state:

Each party shall have the burden of proving the facts relied on to support his
claim or defence …. (75)

The issue was not dealt with in the UNCITRAL Model Law, in part because of
differences in view between legal families as to whether burden of proof is
procedural or substantive. Civilian legal systems tended to consider it to be a
substantive issue that would hence be determined under Article 28. (76) While not
directly applicable to arbitration, the Rome Convention of 1980 on the Law
Applicable to Contractual Obligations stipulates that burden of proof is governed
by the lex causae (77) Andreas Reiner argues strongly that burden and standard
should be seen as substantive. (78) According to Reiner, the most important
reason why burden and standards of proof should be seen as substantive is that
they directly affect the page "764" claim itself. Reiner also argues that if the law
applicable to the substance of the dispute is also applied to burden and
standards of proof, it provides greater fore-seeability for the parties. Finally,
Reiner suggests that when parties decide on procedural rules they do not have
the intention to thereby determine the rules concerning the burden and the
standard(s) of proof.

It might also be argued that to separate burden of proof from substantive legal
rules could distort the latter as they are inextricably linked. (79) For these reasons,
Born suggests that burden of proof might present choice of law questions. He
concludes that ‘the tribunal should allocate the burden of proof in the light of its
assessment of the applicable substantive law and procedures adopted in the
arbitration. In so doing, the tribunal need not apply the burden of proof rules of
any specific jurisdiction, but can instead fashion specialised rules in light of the
particular substantive issues and procedures at issue in a specific instance.’ (80)
Such an approach would only be necessary if there truly are differences in
approach between legal families and adjudicatory bodies. The better view is to
see burden of proof as always being on a party seeking to establish a position,
but note that the substantive law will commonly impact upon this by indicating
what must be proven and by whom.

As noted at the outset, the more contentious variations in approach are likely to
arise because burden of proof is interrelated with document production
entitlements, (81) and the inferences that might properly be drawn from a failure
to produce. A tribunal still needs to decide how to behave if it is hard for the
party with the burden of proof to bring evidence, or if all key evidence is in the
hands of the respondent. This is discussed further below in section 10.4.15
dealing with integrating various legal approaches to fact finding and in sections
11.4 and 11.5 dealing with obligations to produce adverse evidence. Finally,
burden and standard of proof should not necessarily vary simply because a
tribunal has been appointed as amiable compositeur, although there are differing
views as to the proper methodology of such an adjudicator. This is discussed
further in section 13.14.
10.4.2. Burden as to Defences, Exceptions, Counterclaims and Set-Off

A respondent will have the burden of establishing the validity of any defences on
which it relies. The respondent also has the burden in relation to counterclaims
and set-off rights. Claimant would have the burden as to defences to such alleged
rights. In some cases it will be necessary to consider whether an issue is truly a
defence or page "765" whether the issue in dispute is an exception to a primary
obligation. In the latter event, the burden may remain on the claimant to show
that there is indeed an obligation notwithstanding the exception.

Burden of proof may also be problematic or at least unhelpful in cases where

there are claims and counterclaims about the same factual issue. For example, in
a construction dispute, a builder might render an account to the owner who then
refuses payment. The owner might seek a declaration from an arbitral tribunal
that the money is not owed, while the builder would naturally counterclaim as to
an entitlement. From a technical perspective, each has a burden of proof as to its
contention. Obviously, in such circumstances the tribunal will consider all the
evidence and simply decide which view is to be preferred.

10.4.3. Standard of Proof

It was stated above that the burden of proof remains on the party seeking to
establish a particular proposition. Burden of proof simply deals with
responsibility, but does not indicate the level of proof that is required. Standard
of proof deals with the degree of conviction that the adjudicator must have to be
satisfied that the burden has been met. As noted above, common law legal
systems treat it as procedural, while civilian systems see it as substantive.
Common law systems speak of the balance of probabilities, that is, that a
proposition is more likely than not. Civilian legal systems speak of the
satisfaction of, or the inner conviction of the adjudicator. (82) There is unlikely to
be any difference between the civil and common law standards as expressed,
notwithstanding some comments to the contrary. (83) Reiner has suggested that
continental law establishes a higher standard than common law, noting the use
in Austrian law of the term ‘full conviction’ (voile Überzeugeng) although he
concludes that the practical result seems the same in both systems. (84) He
concludes that the real test in each system ‘must be a test of preponderance of
evidence’. (85) Von Mehren and Saloman also speak of a higher civilian burden but
again argue that the ultimate test is a preponderance of evidence. (86)

page "766"

Arbitral statutes and rules rarely articulate the principles of standard of proof in
any detail. The Eritrea–Ethiopia Claims Commission noted that international
adjudicatory rules do not typically ‘articulate the quantum or degree of proof
that a party must present to meet this burden of proof.’ (87) Standard of proof is
seen as being more problematic as it not only may have issues of
characterisation in terms of procedural versus substantive law but might ‘also
reflect subjective standards (“inner conviction”) of arbitrators.’ (88) It may be that
different legal families are speaking at cross purposes or are describing only part
of a broader picture. To speak of the inner conviction of an adjudicator as
opposed to the balance of probabilities makes for a difficult comparison as each
formulation is incomplete. How is the balance of probabilities to be determined
other than by the inner conviction of the adjudicator as to the probability of each
view? To a civilian adjudicator, what is their inner conviction to be about if not
that one side's evidence is more believable or is preponderant to that of the
other? Thus it is overly simplistic to speak as some do of the civilian burden as
subjective, as if there is no subjective element in common law adjudication.
Another reason why there is unlikely to be any significant difference between the
articulations in different legal families, is that no system allows a party to
succeed if their opponent's contentions are preferred. In any contest, one side's
contentions will outweigh the other. If they are equal, all systems would
conclude that the party with the burden has failed, as it has failed to prove that
its contentions are properly preferred. Caron et al. suggest that the balance of
probability standard reflects arbitral practice. (89) The only remaining doubt is
whether some civilian systems in practice require a higher level of conviction
than mere on balance preponderance.

It is more important to consider the way that adjudicators make these

determinations and not simply consider the general phraseology used to
describe standard of proof. In a detailed study of public international law
adjudication, Kazazi argues that ‘standard of proof is a subjective and
discretionary matter subject to human judgement’, because of the need to
analyse and resolve the way to treat conflicting evidence. (90) Because of the
uncertainty of conflicting evidence, probabilistic analysis is commonly involved.
(91) The evidence of one party is not necessarily tested merely against that of the

other. One formulation accepted by an international arbitral tribunal was as

follows: ‘Does the evidence tested against human experience and common sense
persuade or does it not?’ (92) page "767" A practical scenario where differing
articulations may matter is whether claimant's case simply needs to be more
believable than respondent's or whether claimant's case must be likely to be true
in an absolute sense. It is entirely possible that each party's contentions are
unlikely, but one is still more likely than the other. Another scenario is where
proof of a causal link needs to be higher than on balance under civilian
conviction standards.

An expert arbitrator might also consider standards in the context of experience

as opposed to relying solely on the conflicting evidentiary material. For example,
a construction engineer might be swayed, albeit subconsciously, by what
experience suggests is the key cause of problems of this type on construction
sites. Where the allegation is of professional negligence and there is a need to
consider the appropriate standard of care, it would be hard for an expert
arbitrator to ignore what they think they would have done in the circumstances.
It is also important to consider standard of proof both in terms of the primary
obligation on the party with the burden of proof, and secondly, in terms of the
ultimate conclusion of the tribunal after each party has presented its evidence.
Individual items of evidence cannot be weighed in a vacuum. It is a relative
exercise to consider the relevance, materiality and probity of conflicting
evidence. A claimant with the burden of proof may present a prima facie case
that satisfies the standard of proof required at that stage, only to find that the
respondent subsequently presents a preponderant body of evidence leading to a
decision in favour of respondent. The notion of a prima facie case is discussed in
the following section. Onus of proof is discussed in section 10.4.7.

An adjudicator's approach to questions of standard of proof may also vary

depending on the issues and legal norms involved, although the concept itself
does not change. For example, where jurisdiction is concerned, some arbitrators
would look for strong evidence to ensure that consent is properly found to exist.
Others might take a contrary approach, looking to find validity wherever it is
reasonable to do so, on the basis that even potentially pathological references to
arbitration should be supported on an intent analysis. They may even have one
eye on the standard of proof likely to be applied by an annulment court or an
enforcement court where the place of enforcement can be anticipated. This could
be justified although not mandated under the duty to render an enforceable
award.

Approaches to standard of proof may also vary if the issue is a complex economic
or scientific question, a complex question of causation or a serious allegation of
impropriety such as lack of good faith or fraud. In all of these situations, the legal
articulation of the standard of proof remains constant but the body of evidence
that might be required can vary. Proving that a drug is safe as contractually
warranted may require clinical trials that must come close to certainty. Where an
allegation of fraud is concerned, a significant body of evidence might be required
to justify a conclusion that fraud is present and that the person under
consideration cannot have their behaviour readily explained on other bases.
page "768" This can also impact on the evidence that might be needed.
Documents evidencing fraud may need to be more compelling. Some seek to
articulate a higher standard in such circumstances. (93) The Iran-US Claims
Tribunal even went so far as to say that in the context of an allegation of bribery,
‘if reasonable doubts remain, such an allegation cannot be deemed to be
established.’ (94) In the context of an allegation of forgery, another Tribunal
considered that this ‘must be proved with a higher degree of probability than
other allegations.’ (95) Conversely, one leading arbitrator has suggested that the
burden of proof may need to shift when there are allegations of bribery simply
because of the difficulty of proving the proposition to the requisite degree. (96) In
ICC Award No 8891 (97) a tribunal considered that the burden of proof remains on
the party alleging bribery but that in some special circumstances if relevant
evidence of the allegations is brought, a tribunal might call for counter-evidence
from the other party in the absence of which adverse inferences might be drawn.
Other references to lower or higher standards of proof rather than a
consideration of a consistent standard in the context of available evidence have
been suggested by Caron et al (98) (the standard of proof… varies according to the
circumstances), and Born (99) ((i)n limited instances, a lower or higher standard
of proof arguably applies). The preferred view should be to apply uniform
standards in the context of the available evidence and the seriousness of the
allegations and not attempt to set up multiple standards of proof depending on
the circumstances. Because allegations of bad faith and illegality require
sufficient proof of a particular mental state, uniform standards rigorously
applied will protect against unmeritorious claims.

Serious allegations can also impact upon questions of timing and the need for the
person claiming such breaches to provide adequate particulars and evidence at
page "769" the earliest possible opportunity. This is not only because of the
seriousness of such allegations but because of the difficulty of disproving them.

Conversely, if a party simply needs to prove a share price on a particular day, a

commercial newspaper is likely to suffice unless contradicted. In some cases a
lesser evidentiary requirement may be appropriate in the context of likely best
evidence. For example, the Claims Resolution Tribunal for Dormant Accounts in
Switzerland utilised a ‘relaxed standard of proof’, also described as a ‘plausible’
standard when dealing with claims by victims of Nazi persecution or their heirs
as to their claims over dormant Swiss bank accounts many years after World
War II. (100) That standard was less contentious to apply in that context given that
it was not typically a matter of contested facts, but rather, deciding on adequate
proof for the banks purposes. While the solution seems eminently fair, it may
again be preferable in such circumstances for a tribunal to simply indicate that
applying broad standards of proof must be looked at in the context of the
documentary evidence that could reasonably be expected to be presented by a
party with a valid claim, rather than utilising phrases such as ‘relaxed’, that might
wrongly suggest that the claimant has been given a preferred position.

Some speak of a lower standard of proof where damages are concerned. (101)
Others instead criticise an alleged lax application of standards of proof by some
arbitrators where damages are concerned. (102) The issue is once again less about
the standard of proof and more about the evidence the tribunal will require as
compared to the best evidence available, and the time and effort it will wish to
apply to calculations. When a tribunal picks a ‘ballpark’ or rounded figure and
does so in good faith, it is providing its most likely estimate of the damages,
albeit on an evidentiary methodology that leaves it open to justifiable criticism.
Conversely, if it splits a figure to try and be fair and gives something to save face
to the losing party, absent express entitlement to do so as amiable compositeur, a
tribunal has simply not complied with its mandate. (103) In many cases the fault
partly lies with counsel if they put more effort into preparing the liability stage
than they do for the quantum stage.

The degree of exactitude that a particular substantive law expects in proof of

damages may also impact upon the way in which standards are satisfied. (104)
Even page "770" though some legal systems do not see standard as a matter of
substantive law, all would accept that substantive norms can impact upon the
degree of proof required. For example, a tortious claim based on common law
may have the principle of res ipsa loquitur applicable, that is, that the thing
speaks for itself in terms of negligence. (105) Another way to put this is to note that
the substantive law and contractual terms determine what must be proven. For
example, in ICC Case No 6653, the parties' contract indicated that if there is a
dispute about conformity of the goods, the purchaser would appoint an expert
whose decision would be binding on the parties. The tribunal accepted this
stipulation and did not revisit the issue of conformity. (106)

Legal systems that consider standard of proof in terms of preponderance of

evidence or other similar probabilistic notions, often have to deal with situations
where a party needs to prove a number of uncertain and related matters to
succeed. For example, if a claimant alleges that a pre-contractual representation
caused it to acquire a business which failed because the representation was
untrue, there are a number of steps in the logic. Was any representation made?
Whose version of the comments was most accurate? Was the language clear
enough to constitute a representation? Did the claimant rely on it? Was the
reliance reasonable? Was the representation untrue? Was it the cause of the
business failing? What was the amount of damage? Thus the party with the
burden may need to satisfy every required element of the claim by such a
preponderance of evidence. In these circumstances a tribunal makes an overall
assessment as to standard of proof and does not seek to technically apply the
standard to each step, although if the claimant fails on any of the steps, it ought
to fail overall.

10.4.4. Prima Facie Evidence

In common law legal systems there is the related principle that the party with
the burden of proof needs to at least establish a prima facie case, otherwise the
responding party can make an application that there is no case to answer before
being called on to present any of its own evidence. A prima facie case is one
where the party who has presented it should succeed in the absence of
refutation. The process of no case to answer submissions is unlikely to apply in
international arbitration in a formal manner. Kazazi suggested that ‘international
proceedings…. do (not) consist of two stages, with the court's decision to
proceed to the second phase depending on whether or not enough evidence has
been produced in the first phase…. The practice of international tribunals in this
regard page "771" follows the pattern of civil law countries…. in which it is only
at the end of the proceeding that the court decides whether or not the burden of
proof has been discharged.’ (107)

Nevertheless, the concept of a prima facie case may still impact upon the
practicalities of fact-finding. If the claimant cannot present even a prima facie
case, the respondent will ultimately succeed. Where a claimant has established a
prima facie case through its initial submissions, the responding party must
present a more compelling case or it will lose. While the burden of proof never
shifts, the burden of proceeding or onus of proof can be said to shift back and forth
depending upon whose case is relatively stronger at any stage of the
proceedings. (108) This is discussed further in section 10.4.7 below.

10.4.5. Proving a Negative

At times a party with the burden of proof must seek to prove that something did
not happen or some circumstance does not exist. While the legal principles of
burden and standard of proof are the same, it needs to be understood that it is
always more problematic to seek to prove a negative. An arbitrator must be alert
to this to deal fairly with both parties.

10.4.6. Standard of Proof Where Administrative Actions are Challenged

Investment disputes will often be concerned with the adequacy or otherwise of

domestic administrative practices. An example would be an expropriation or
national treatment claim based on the failure to renew a mining licence on the
grounds of alleged environmental breaches. Considerations of burden and
standard of proof within arbitration are complicated where the issue of concern
is the domestic bureaucratic decision which itself was subject to burden and
standard principles and domestic administrative review rights. In some cases
there might be arguments as to whether the bureaucrat correctly applied burden
and standard principles and whether sufficient evidence was sought and relied
on for such purposes. In an ordinary commercial dispute, conflicting evidence is
simply weighed up to determine which view is preponderant. In some
investment arbitration cases, it may be arguable that the key question is instead
whether the bureaucrat behaved reasonably, taking all relevant facts into
account and ignoring irrelevant criteria.

page "772"

Another issue arising where the challenge is to domestic administrative

behaviour, is whether there should be deference as is applied in some
administrative law systems in the context of standard of review. This issue might
arise in the investment arbitration example above. EU courts typically adopt a
margin of appreciation analysis in reviewing administrative action. (109) Applying
such a margin of appreciation might also mean that a tribunal is less inclined to
be concerned with expert evidence, at least insofar as it seeks to identify the best
outcome under a de novo reconsideration of the decision in issue. The better
view is that the tribunal can still avail itself of this form of evidence even when it
wishes to apply a margin of appreciation. The expert evidence could show that
the administrator's decision was certainly one of a range of reasonable options.

10.4.7. Burden of Proceeding and Onus of Proof

As argued above, the better view is to say that burden never shifts but whether
the standard of proof has been satisfied will depend on the material presented by
both parties. Because evidence is presented at different times, it is likely that the
balance of evidence may change back and forth. The concept of onus of proof
refers to the party who at any point in time is faced with a preponderance of
evidence against it or is the one otherwise needing to convince an as yet
unconvinced adjudicator. Some also describe this as the burden of proceeding but
this is confusing as burden of proof never shifts, so use of the term ‘burden’ in
the phrase is apt to mislead. In a practical sense, it is immaterial whether the
party with such onus has the burden of proof or not. At that point in time, its
opponent has better evidence. If it does not respond adequately, it should lose.
Hence it is appropriate to refer to that party as having an onus at that stage. If it
responds adequately so that its evidence is stronger, the onus shifts in a practical
sense. The opponent must respond adequately or should expect to lose. (110) Of
course, a tribunal is not making a running assessment, nor advising the parties in
most cases as to the relative position. It is more a matter for self assessment by
counsel. In some cases a tribunal might give guidance by indicating that one
party need not provide more evidence on a particular issue or by warning them
that they need to do so, although that itself is contentious in terms of how
proactive a tribunal should be. To do so, a tribunal has to be absolutely certain
that it has reached a positive decision and will not later change its mind when
drafting the award. This would be problematic with multi-person tribunals.

page "773"

10.4.8. Inferences and Presumptions

10.4.8.1. Positive Inferences and Presumptions

An important aspect of legal fact-finding is the use of inferences and

presumptions, both positive and negative. An example of a positive inference
would be that contemporaneous documentation is more likely to be reliable than
documents prepared for the purposes of the arbitration. Efficiency requires
many presumptions in adjudicatory processes, absent assertions to the contrary.
For example, a claimant does not need to prove that a request for arbitration was
served. It is for respondent to challenge adequate notice. The same will be so for
all commercial communications that underlie the substantive dispute.
Adjudicators will also presume that documents are authentic unless an allegation
is made to the contrary. Holtzmann suggests that a tribunal might direct that
unless challenged within a stipulated period of time, documents will be deemed
to be authentic. (111) A tribunal is always entitled to draw reasonable inferences.
(112) An inference ought to be as to an actual likely fact and not to a general

statistical likelihood.

While any inference and presumption cannot be accepted with certainty, under a
standard of proof simply looking for something akin to preponderant conviction,
these must be legitimate methodologies of inference. Nevertheless, an
adjudicator must understand that these are merely presumptions. Hence an
opposing party can attempt to rebut the presumption. In many cases, the
inferences flow from the behaviour of the parties in terms of the optimal
evidence that was available. The more a party can be excused for not providing
direct and cogent evidence, the more an adjudicator may be willing to operate by
way of positive inference. For example, in the Corfu Channel case, the ICJ
concluded that a party suffering a breach of international law which through lack
of power is unable to furnish direct proof of the facts giving rise to legal
responsibility ‘should be allowed a more liberal recourse to inferences of fact
and circumstantial evidence.’ (113) Conversely, adverse inferences may be drawn
when a party fails without reasonable excuse to present evidence that is
reasonably seen as being within its possession or control. This is discussed in the
next section.

The Iran-United States Claims Tribunal applied a number of presumptions on a

routine basis. These included: the failure to object in writing to a writing
received at or shortly after receipt is strong evidence of acceptance; statements
by a party contradicting a position taken in the proceedings is strong evidence
against the position; and the failure to produce evidence expected to exist and in
the control of page "774" a party, allows for a justifiable inference that the
evidence would be adverse to that party. (114)

10.4.8.2. Adverse Inferences

It is readily accepted that tribunals may draw adverse inferences from a party's
failure to provide information and documents where it would be reasonable for
them to do so. Adverse inferences can apply to witnesses who could easily be
called or who refuse to answer certain questions as well as to documents not
produced. Absent a valid excuse, it is logical to presume that withheld
information is adverse to the interests of the party controlling it. (115) An
appropriately drawn adverse inference is not a violation of a party's right to be
heard. (116)

An adverse inference is not a punitive action. It is an inference that can be

reasonably and logically drawn in appropriate circumstances. (117) This is a
crucial distinction as a punitive measure is simply a response to the clear fact of
non-production. An adverse inference requires a decision-making process as to
the unreasonableness of non-production and a logical inference that non-
production justifies a view as to the kind of information the document would
thus be likely to contain or that the non-appearing witness would have
presented. It then involves consideration of the way that material is to be
balanced from an evidentiary point of view with other evidence that is actually
before the tribunal, (which could in turn include other inferences). Less clear is
whether the adverse inference should go so page "775" far as to treat all of the
party's evidence as suspect unless corroborated. (118) At the very least, if the
tribunal draws an adverse inference about a document itself, the assumption
thereby drawn can be considered on balance to outweigh contrary views and
evidence on the same issue.

While the entitlement of an adjudicatory body to draw adverse inferences is

readily accepted, the issue is more complex when integrated with other
principles such as burden of proof and duties, if any, to produce relevant
information. One issue that may arise is whether an adverse inference can be the
predominant basis upon which a claimant satisfies its burden. Kazazi suggests
that this should not be the case in terms of a claimant's obligation to at least
present a prima facie case calling for refutation. (119) Sharpe recommends that
before such a conclusion could be reached, the party inviting the tribunal to draw
adverse inferences must produce all reasonably available evidence confirming it,
demonstrate that the requested evidence is reasonably accessible to the party
against whom production is sought, demonstrate that the inference is reasonable
both in terms of its consistency with other facts and the presumption from non-
production, the requesting party must produce prima facie evidence, and the
person against whom the inference is sought must understand its obligation to
produce evidence in rebuttal of the inference sought. It is not clear what prima
facie evidence would mean in such circumstances and why prima facie evidence
and all available corroborating evidence are gateway requirements. Where
corroborating evidence is concerned, its absence may lead to a countervailing
adverse inference and hence prevent an adverse conclusion against the party
which is the subject of a production request. That may be so on appropriate facts
but it should not be articulated as a gateway requirement. In any event, it makes
little sense from a policy perspective to limit the entitlement to draw adverse
inferences to situations where a claimant has made out a prima facie case. This is
because in such circumstances, the respondent needs to refute the case in any
event to be successful. In theory, there should be nothing to prevent adverse
inferences founding a claim but a tribunal would have to be particularly vigilant
in ensuring that the inferences were valid and appropriate. Having said that, the
issue is complicated by the fact that there are tenable arguments from the
civilian legal tradition to the effect that a party should never have to produce
evidence against interest and that it is for a claimant to have sufficient evidence
to support a case before bringing it. Nevertheless, this is not the dominant norm
in modern arbitral practice. This is discussed further in sections 11.4–11.6. It is
also important for a tribunal to remember that its discretionary behaviour can
impact significantly on these questions. The situation is likely to be quite
different if, on the one hand, the tribunal sits back and allows the claimant's case
to be tested for adequacy, or alternatively, demands production of documents
and demands page "776" answers to questions by the respondent. If it does not
ask, there is no possibility of refusal and hence less opportunity for adverse
inferences. This draws attention to the reason for refusal. The more a party has a
tenable reason for refusing production, the less legitimate an adverse inference
would be. It has also been suggested that a tribunal ought to warn parties about
the possibility that negative inferences may be drawn, (120) and give the relevant
party an opportunity to rebut the inference, otherwise due process challenges
may ensue. (121) Even if notification should not be seen as a legal gateway to the
drawing of adverse inferences, an appropriate warning may well be sensible and
adds to the evidentiary weight of the inference. If a party is invited to produce a
document, given a clear indication of the ramifications for failing to do so absent
good reasons for failure and provides neither the reasons nor the document, the
arbitrator can be more confident that the adverse inference is appropriate.

Another situation where a tribunal will have to assess the justification for
refusing production is where an order is made against a party on the basis that it
is presumed that it has control over the person in possession of the relevant
document. If the latter refuses to present the document, a tribunal ought to
reassess its reasons for concluding that there was sufficient control over the
document holder in order to be confident that adverse inferences can
legitimately be drawn. It may also be harder to draw adverse inferences from a
failure of a party to present a witness it has some control over than a document if
the individual concerned simply does not wish to travel and cannot be compelled
against their will. (122) Human beings can always say no, documents cannot.

Webster raises a number of problems with the notion of adverse inferences

under Article 9.5 of the IBA Rules of Evidence 2010 which allows for such
inferences where a party fails to make available evidence or testimony ‘without
satisfactory explanation…’. First there is a need to understand what the broad
expression ‘satisfactory explanation’ might entail. If a party has already objected
and the tribunal has ruled against it, may it repeat its objection or is it deemed
unsatisfactory by the prior ruling? If the party raises other matters, is it entitled
to do so, not having alluded to these at the time of the initial application? (123)
Another problem arises from the fact that Article 9.5 allows for an adverse
inference when an objection is not taken in due time. Yet if an objection is valid
but too late, in some cases it will be difficult to determine just what adverse
inference to justifiably draw about the material itself. Another situation where it
is difficult to determine just what inference to draw is where there is a page
"777" large volume of material that is not produced, when some material is likely
to be favourable and some unfavourable to the party's position. (124) The situation
is easy when the opponent has presented a prima facie case, as the lack of
documents simply means that the responding onus has not been satisfied, but
this is separate to the question as to when an adverse inference can itself form
part of the reasons for a decision.

A particular issue is whether there ought to be differences in approach where a

State is a party. This might arise because in various forms of international
adjudication, governments appear less inclined to respond fully to adjudicators'
requests for documents. (125) There should be no difference in such circumstances
as the State has agreed to adjudicate in good faith. The Iran–United States Claims
Tribunal stated that ‘it is an accepted principle that an adverse inference may be
drawn from a party's failure to submit evidence likely to be at its disposal.’ (126) In
a separate opinion in the Case Concerning the Barcelona Traction, Light and
Power Company Limited, Judge Jessup stated that ‘…if a party fails to produce on
demand a relevant document which is in its possession, there may be an
inference that the document if brought, would have exposed facts unfavourable
to the party….’ (127)

In legal systems where there are some, albeit limited rights of appeal on
questions of law against arbitral determinations, there is a related question as to
whether the view taken by a tribunal in this scenario is a legal determination or
simply a factual finding. (128) It has been suggested that an inappropriately drawn
adverse inference may lead to challenges. (129) This may depend on whether the
particular legal system would treat these matters as substantive or procedural. It
may also depend on the logic behind the adverse inference. For example, if an
adverse inference was drawn after a party was inadequately notified of a
production obligation, this would clearly be a procedural concern.

page "778"

10.4.9. Evidence of Intent where Relevant

At times, legal provisions will expressly or impliedly call for a determination of

the purpose, motive, object or intent of a party. A number of important questions
arise as to the degree of purpose when there are mixed purposes and whether
subjective or objective evidence of purpose should be relied upon. This would
generally be a question of substantive law but far too often, such laws fail to
indicate whether the purpose required is sole, dominant, significant or more
than de minimus. Furthermore, they will commonly fail to indicate the
importance of subjective as opposed to objective evidence. Questions of intent
will typically arise with some of the more serious allegations such as in relation
to bribery or fraud or where there are contractual exclusions for simple
negligence, but not for conscious recklessness or worse.

10.4.10. Best Evidence

There are three senses in which the notion of best evidence in adjudicatory
proceedings can be considered. The first simply identifies the entitlement of the
adjudicator to make a determination based on the evidence presented, even if
the totality is less than adequate. In the Protiva case, the Iran–US Claims Tribunal
considered that it was faced with evidentiary gaps, but after reviewing the
evidence presented, determined that claimant's evidence was preponderant. (130)
The second notion would imply some obligation on each party to present the
best possible evidence and the entitlement of the tribunal to draw adverse
inferences when it reasonably believes this has not occurred. A third aspect
involves consideration of which type of evidence is likely to be the most
influential. As noted above, civilian legal systems tended to prefer
contemporaneous documents as the most reliable, while common law systems
placed significant reliance on cross-examination to overcome the potential
problems of self-serving witness testimony after a dispute has arisen.

Best evidence must also be considered in a comparative sense. Because fact-

finding in law is about comparing contrary evidentiary material to find the
preferred view, all evidence must be viewed against appropriate frames of
reference. As noted above, the natural comparison appropriate for standard of
proof tests is to balance one party' s evidence against that of the other. Another
relevant comparator is to consider evidence presented as against evidence which
might reasonably have been expected to be presented. When discussing standard
of proof, it was noted that a positive application of best evidence standards
would occur where a tribunal understands that there are good reasons why
better evidence is not available. page "779" A negative conclusion arises where a
party fails to bring the clearest evidence where that is thought to be within its
control.

10.4.11. Fact-Finding Powers of a Tribunal to Justify Claims

As noted above in section 10.3.2.3, most rule systems either expressly or

impliedly allow a tribunal to engage in fact-finding exercises. (131) An important
question is whether a tribunal can exercise its independent fact-finding powers
to identify a factual basis for a conclusion different to the contentions of either
party. This should not be so. A tribunal should also not exercise broad powers to
establish a prima facie case where the claimant has failed to do so. Even this
statement is overly general in the context of a scenario where the respondent is
in sole possession of a key document without which claimant is unlikely to
succeed. A tribunal which simply orders production in such circumstances is
proactively supporting the making out of the claimant's case.
10.4.12. Evidence from the Tribunal's Own Knowledge

In some domestic litigation systems there is the concept of judicial notice of facts.
This arises where the facts are seen as being of sufficient common knowledge to
not require proof. While this would be rare in international arbitration, the
principle was applied in the Nicaragua case by the ICJ. (132) While the party with
the burden of proof must satisfy it, in some instances a tribunal will accept that
some propositions are so obvious that proof is not required. (133) Civilian judges
tend to feel more comfortable in making broad factual determinations from their
own knowledge. It has been suggested that ‘EC Courts… seem to enjoy a great
discretion as to the amount of personal experience or knowledge they can put
into their judgments in order to establish the factual background of a case.’ (134)

Where a tribunal member is an expert in a particular relevant profession, they

may be more inclined to use their own knowledge and not require formal proof.
An example would be standards of quality in commodity trade. The first issue is
whether the arbitrators are truly utilising their expert knowledge and are not
relying on distinct evidence to that presented by the parties. There is then a due
process page "780" issue as to whether and when a tribunal can take note of its
own personalised knowledge. On the one hand, arbitrators are often selected for
such specific knowledge, hence it naturally flows from their selection that they
may employ their expertise fully. On the other hand, due process obligations,
requiring each party a sufficient opportunity to present its case, mean that a
tribunal should make the parties aware of its inclinations in that regard and
allow them to make contrary submissions and present evidence accordingly. (135)
Due process obligations may depend on the factual conclusions under
consideration. Where the knowledge is common knowledge, even a failure to
notify will not have affected the outcome. If it is a complex matter pertaining to
the tribunal's own expertise, this is more likely to be part of the process of
dealing with expert testimony where full notification certainly should be given.

10.4.13. Agreed Facts

Where the parties agree on facts, these are not in dispute and the tribunal should
make its determination consistent with the facts as agreed. A tribunal anxious to
promote efficiency might consider inviting parties to agree on whatever facts
they can, so that the tribunal may concentrate its attention on the areas of factual
disagreement. It is always desirable to try and narrow the areas of contention
between the parties in any adjudicatory process. To this end, some rule systems
expressly call for parties to concede or refute contentions from the other side. An
example in the field of arbitration is Article 38 of the ICSID Additional Facility
Rules. A related issue is whether a tribunal can or ought to make costs orders
against a party it feels was inappropriately unwilling to agree on certain facts,
notwithstanding its ultimate success in proceedings. At the very least, a warning
in that regard should be given if that is the intention. Cost issues are discussed
further in Chapter 15.

10.4.14. Admissions
While adjudicatory tribunals will commonly accept admissions as valid evidence,
care needs to be taken to properly understand the circumstances of the relevant
utterance and whether it can properly be seen as an admission of fact. Tribunals
will commonly see particular value in admissions against interest as a person's
comments are more likely to be believable when there is no vested interest in
making them. An admission need not be an express statement to that effect but
could flow from inferences based on behaviour. Where contemporaneous
documents are concerned, a tribunal will commonly take the view that a failure
to object page "781" in writing to a writing received, is strong evidence of the
acceptance of its contents. (136)

10.4.15. Integrating Legal Approaches to Fact Finding

Previous sections have repeatedly noted that questions of burden and standard
of proof need to be looked at alongside other key evidentiary norms and the fact-
finding discretions that a tribunal has. It is appropriate to provide some
concluding remarks about the way tribunals might approach the integration of
these elements. One of the most important factors impacting upon burden and
standard is the question of whether there is a right to document production and
the duty, if any, to present adverse evidence. The practical implications for the
party bearing the burden of proof are significantly different where it must
present a winning case from its own evidentiary sources, or, conversely, where it
can call for its opponent to produce adverse documents and demand adverse
inferences where there is a refusal to do so. Some arbitrators would take the
view that it is for the parties to present the appropriate evidence on which they
intend to rely and for the tribunal to apply relevant burdens and standards of
proof to the material presented. To other arbitrators, document production
powers should be used to ensure that the best evidence is available so that
decisions are not made on technical questions of burden of proof.

There is also the degree to which the tribunal will engage in fact-finding or
questioning. For example, if there is a dispute as to what was said at a particular
meeting, with the claimant alleging a concession and the respondent denying it,
the respondent might be asked to explain what the meeting was about, rather
than simply submitting that the claimant has not met its burden. (137) The
evidentiary record will look different depending on whether the tribunal asks
such questions or not. A simple example demonstrates the contentious policy
challenges that may face a tribunal. Imagine an American claimant suing a
German company for breach of copyright. The American company asserts that
there is an internal memo in the German company's records that admits that
copying occurred and that it was intentional and knowingly illegal. Assume also
that the American claimant cannot page "782" win without production of a
memo to that effect. The tribunal orders production but the German company
refuses, arguing that it is against its view of justice to have to assist a claimant
that itself does not have any meaningful evidence of breach. The claimant argues
that the tribunal should draw adverse inferences from the refusal to produce and
that such adverse inferences should be enough to satisfy the standard of proof
required. There is no clear solution to this problem as each party is making
irreconcilable contentions that are each highly respected within their own
jurisdiction. The key point to draw from the example is that whatever view the
tribunal takes as to the request for document production and the drawing of
adverse inferences, this has a fundamental impact upon the way burden and
standard of proof will apply. (138)

As noted above, the practical application of the burden and standard of proof
may also be impacted upon by the wording of the legal basis of the claim. For
example, a contract for sale of goods might indicate that the buyer may return
faulty goods no matter how minimal the fault. Conversely, the contract might say
that the buyer cannot claim for faulty goods without proof that these were
knowingly provided. While the legal nature of burden and standard of proof are
identical in each scenario, the practical challenge facing the buyer is vastly
different in each case. Because due process norms and consent are paramount,
early proper communication with the parties is vital but even here there can be
pitfalls. A tribunal needs to be careful not to mislead the parties as to their
evidentiary requirements. The paradigm example of the problem occurred in
Avco Corp v Iran Aircraft Industries. (139) In that case, the claimant asked the
tribunal at a pre-hearing conference as to how it might establish amounts due on
a large number of invoices. One possibility was to submit all of the invoices and
invite the tribunal to spend significant time analysing them. An alternative
suggestion was to engage an independent auditor to certify as to the existence
and amounts contained in the invoices. The tribunal chair at that time indicated
that the tribunal was not ‘enthusiastic about getting kilos and kilos of invoices’,
(140) approved the alternative, but in the final award, a differently constituted

tribunal considered that this method was insufficient as compared to providing

the invoices themselves. (141) Enforcement was denied for that reason. (142)

page "783"

10.5. Interrogatories, Written Questions and Notices to Admit

Common law systems allow for interrogatories. These are lists of questions to be
posed by one party to the other where responses are required, subject to certain
exceptions. The aim of interrogatories is to refine the area of the dispute and
allow the hearing itself to be more efficiently conducted. The disadvantages of
interrogatories are that they may wrongly be used as part of a fishing expedition
and add their own transaction costs to the process. Those transaction costs are
increased where disputes arise as to the propriety of particular interrogatories
and the obligation to answer them. Interrogatories are not generally used in
international commercial arbitration. The parties are of course free to agree on
any such process.

While there are no such formal rules in international arbitration, a tribunal's

broad discretion might allow it to draft lists of questions for the parties. (143) A
tribunal might seek to do so where it believes that such questions will promote
fairness and efficiency in the process. (144) Questions are likely to be more useful
when they are clear-cut and help to hone in on the issues. (145) For example, if a
party is relying on pre-contractual representations, questions might inquire as to
the identity of the person said to have made the representations and the exact
wording of them. Conversely, most tribunals would take the view that is for the
party with the burden to present this evidence without guidance from the
tribunal. A tribunal might be more inclined to present written questions in a
documents-only arbitration to ensure that written submissions address the
matters of key concern. This is less important with oral hearings, where a
tribunal has an opportunity to question witnesses and/or counsel during the
latter's submissions. Written questions are also the essence of a tribunal's brief
to a tribunal-appointed expert.

10.6. Proof of Applicable Law

Domestic litigation systems utilise principles of their own private international

law by which courts identify the content and meaning of any applicable foreign
law. Legal families differ between those who see proof of foreign law as a matter
of evidence, to be demonstrated by expert witnesses and documents, and
conversely, those who consider that all legal questions should be determined by
the judges own knowledge or research. Common law systems tend to take the
former approach. This means that the party with the burden of proof must
provide sufficient evidence of that foreign law if it is to succeed, although in
many systems, this does not follow page "784" as a matter of course as there is
often a presumption that foreign law is identical to domestic law unless proven
otherwise. There is no such thing as ‘foreign’ law in an international arbitration,
but simply the need to determine the applicable law or laws based on party
autonomy or the tribunal's discretionary powers. Fouchard argues strongly that
the content of foreign laws should be treated as an issue of fact in international
arbitration. (146)

A number of procedural questions arise. The first is whether there should be a

preliminary determination of applicable law. Related to that is the question
whether evidence as to a range of options should be presented to the tribunal in
aid of its determination. For example, conflicts of law approaches include looking
to the place with the closest connection to the transaction or look to the domicile
of the person exercising characteristic performance under it.

Once the identity of the applicable law is known, there are then questions as to
its content and meaning. Documents seeking to establish the content of the
applicable law should typically be presented, notwithstanding the suggestion
from some civilian lawyers that the tribunal's duty to know the law would not
require this. (147) A range of secondary sources may be admissible to assist the
tribunal in understanding the meaning of the primary texts. Various legal
systems will also differ as to the importance of legislative history and past cases
as aids to understanding or as binding precedents in the case of the common law.
Even if the domestic legal system does not utilise the doctrine of precedent, the
tribunal ought to pay sufficient respect to municipal case law. For example, in the
Brazilian Loans case, the PCIJ stated that a tribunal of international law should:

pay the utmost regard to the decisions of the municipal courts of a country, for it
is with the aid of their jurisprudence that it will be enabled to decide what are
the rules which in actual fact, are applied in the country the law of which is
recognised as applicable in a given case. (148)

A related question is the extent to which experts may give opinions on questions
of applicable law. The general rule is that an expert should indicate to the
tribunal the nature and content of the law but should not seek to opine on the
way the law applies to the facts before the particular tribunal. That is the
function of the tribunal itself.

10.7. Evidence of Aids to Interpretation

In some cases a tribunal seeking to interpret a legal or contractual provision may

be willing to consider certain aids to interpretation for that purpose. Typically,
such aids are used to provide evidence of the likely intent or purpose of the legal
page "785" instruments. There are then evidentiary questions as to what kinds of
materials are properly admissible as aids and how this evidence should be
utilised.

Where contracts are concerned, it would be necessary to consider the applicable

law, and the view it takes as to the use of such aids to interpretation. There may
also be a provision in the contract that seeks to limit the relevance of extraneous
material, although such provisions can themselves be problematic. Where
domestic statutes are concerned, the tribunal would follow the approach to
statutory interpretation in that jurisdiction which would typically identify the
range of applicable aids and their legitimate uses. Where international treaties
are relevant in the field of investment arbitration, interpretation under the
Vienna Convention on the Law of Treaties also allows for consideration of
travaux preparatories. (149)

There can be problems with a number of the sources that a party might seek to
rely upon. Using the example of treaty interpretation, a statement of intent by a
negotiator is not ideal as evidence of the collegiate view of treaty signatories. The
person may be partial and may have failed to adequately distinguish between the
objectives they sought to achieve and the objectives that were actually reached.
Parliamentary debates within a State at the time it ratifies a treaty simply
indicate its own view and not the collegiate position. A comparison of the final
treaty text with some rejected draft provisions might be used to infer intent
behind changes that were made, but in many treaty negotiations there may have
been a myriad of rejected texts, and it will often not be readily apparent whether
the rejection was based on a concern for the content or the mere form of
expression.

10.8. Economic, Scientific, Statistical and Accounting Evidence

Where statistical evidence is concerned, it is important to identify any

assumptions used and the methodology employed. Challenges to the
assumptions and methodology would be the typical way that an opposing party
seeks to undermine the purported conclusions. A tribunal may need to be briefed
on important methodological issues, such as the adequacy of sampling, in order
to form its independent view as to the validity of the statistical analysis. Similar
concerns will often arise with causation analysis in damages assessments, for
example, where loss of profits or goodwill is being assessed in a complex
commercial and market environment.

Where scientific evidence is concerned, in some cases special attention may need
to be given to the appropriate way to address standard of proof or the
reasonable justification for a party's behaviour. As to standard of proof under
scientific disputes, it was noted above that if a dispute is whether a
pharmacological product is safe, this needs to be shown with a high degree of
exactitude to be accepted as true. The situation may be more complex when
science is a basis of an analysis of reasonable regulatory behaviour. For example,
in an investment page "786" arbitration case where a governmental
environmental measure is argued to constitute an indirect expropriation, must
the measure be based on the preponderant scientific opinion to be reasonable or
may it be based on a respected minority view, perhaps subject to a heightened
concern under a precautionary principle? This is primarily a question of
interpretation of the treaty norm itself, and not an evidentiary question per se,
but the interpretation selected will impact on the nature and degree of evidence
required to justify the governmental action. A number of permutations may arise
depending on whether the challenge is to a particular legislative enactment, or to
a bureaucratic determination under it, or to a domestic court's application of the
provisions.

Where highly technical documentary evidence is concerned, an important

question is whether the person presenting this to the tribunal has sufficient
expertise to be able to explain the nature, contents and relevance of the
documents. For example, an accountant may be the most appropriate person to
present evidence about stock valuations. An engineer may be the most
appropriate in dealing with construction drawings. Such issues need to be
considered in conjunction with the use of expert witnesses and the possible
appointment of an expert by the tribunal. This is discussed in sections 12.10 to
12.14 below. A related issue is for the tribunal to have the requisite expertise.

10.9. Objective Evidence and Subjective Evidence

In the majority of cases, commercial adjudication is concerned with objective

evidence, primarily the logical inferences that may legitimately be drawn from
documents and oral testimony. Subjective evidence looks at the state of mind of
relevant persons. In some cases, subjective evidence or at least subjective
conclusions are central, for example, where an allegation of bad faith is made. In
other cases it is not readily apparent how the two forms of evidence should be
utilised. An example where different outcomes might apply depending on which
form of evidence is paramount would be a case about an alleged pathological
arbitration clause. An objective approach looks at the clause itself, and perhaps
other documents to see what reasonable inferences can be drawn. A subjective
approach would analyse the beliefs of the signatories to the clause to determine
what they actually intended at the time. Some arbitrators would recommend a
hybrid approach in such circumstances. (150)
page "787"

10.10. Evidence from Prior ASR Processes between The Parties

There are two aspects of the use of evidence from prior processes. The first is the
general admissibility question which deals with potential confidentiality issues.
The second is what probative value there is with any material admitted. The
discussion below should also be integrated with section 10.16.7 dealing with
confidentiality generally. Some dispute resolution clauses call for the parties to
negotiate in good faith and/or utilise mediation prior to going to arbitration if
the dispute is not otherwise resolved. Mediation is essentially confidential and
without prejudice to adjudicatory proceedings. (151) The position will commonly
be expressed in well developed mediation rules. (152) A good mediator will ensure
that the parties have agreed to this at the outset. Hence material presented
during the mediation should not be admissible in subsequent arbitral
proceedings for that reason alone, but should be produced by a party in control
who seeks to rely on it and may be the subject of document production requests.
Offers of settlement are certainly not admissible, (153) although if the parties
disagree as to whether a settlement had actually been agreed upon, that goes to
the jurisdiction of the tribunal and would need to be resolved. (154) The privilege
should also apply to internal documents prepared specifically for the purposes of
the ADR processes. (155)

In some cases a dispute resolution clause has a preliminary adjudicatory process

prior to arbitration. For example, under FIDIC Conditions of Contract for Works
of Civil Engineering Construction, Part I General Conditions Clause 67, there is
first an engineer's decision after which an aggrieved party can commence
arbitration. The clause gives the arbitrator the power to review and revise any of
the engineer's decisions and opinions. The evidence before the engineer and the
engineer's decisions are admissible before the arbitration which is not limited to
that material. The engineer is not disqualified from being called as a witness to
page "788" give evidence before the arbitrator. In some cases there may be a
dispute as to whether arbitration is premature and whether the preliminary
steps have been taken or not. In such a jurisdictional dispute, evidence as to the
conduct of the earlier processes is central and must be admissible.

10.11. Evidence and Findings in other Arbitral Cases

At times there may be a number of arbitral proceedings between the same or

similar parties on similar issues. In such circumstances a question may arise as
to the admissibility and weight to be given to evidence from one forum, tendered
to another. The answers will vary depending on the type of evidence being relied
upon, but there are also a number of policy questions that arise. Would it be
unfair from a due process perspective to rely on such evidence? Conversely,
would it be a breach of a duty of efficiency to fail to at least consider saving time
in this manner?

Once again the gateway issues are confidentiality, if any, and the probative or
binding nature of evidence and findings from other cases. Whatever the view as
to admissibility, the second tribunal must still make an independent finding of
fact. A finding in an earlier case in favour of the person with the burden of proof
in the latter, should not operate to shift that burden in the latter instance,
although if admitted and if sufficiently probative, the onus can shift in a practical
sense. A tribunal might also need to consider whether an admission in an earlier
case between the same parties could constitute an estoppel in subsequent
proceedings. A related question is whether a finding of fact in the earlier case
could constitute an issue estoppel in subsequent proceedings. This is also
addressed in section 8.10.3 and the following section dealing with issue estoppel.
Evidence from other proceedings might also be tendered to show that the
opposing party has made contradictory assertions in different fora or has been
selective in evidence submission in one or both. Even if such evidence is
admitted, it does not readily show which version is to be believed, although it can
undermine the veracity of that party generally.

The situation as to confidentiality may depend on the rules applicable in the

prior case, although that is controversial as most would see arbitration as
presumptively confidential. Nevertheless, significant differences in view can be
discerned from a range of cases dealing with confidentiality in arbitration. (156)
Under some rules, confidentiality is indirectly enshrined. For example, Article
26(3) of the ICC Rules 2012 provides that persons not involved are not to be
admitted to the hearings without express agreement. (157) Article 27.4 of the ICDR
Rules states that page "789" ‘an award may be made public only with the consent
of all parties or as required by law.’ Article 30 of the LCIA Rules says that parties
‘undertake as a general principle to keep confidential all awards in their
arbitration, together with all materials in the proceedings created for the
purpose of the arbitration and all other documents produced by another party in
the proceedings not otherwise in the public domain – save and to the extent that
disclosure may be required of a party by legal duty, to protect or pursue a legal
right or to enforce of challenge an award in bona fide legal proceedings before a
state court or other judicial authority.’ (158) Laws of the latter kind may include
disclosure rules to stock exchanges or to the tax authorities. In the stock
exchange example, the information can be used by anyone as it is in the public
domain. In the tax example it is not. If information is already in the public
domain, it will not be rendered confidential simply because it has been
incorporated in an arbitration. (159) The enforcement stage also differs as to
confidentiality, given that it is implemented where necessary through national
courts, which are generally open.

10.12. Issue Estoppel, Collateral Estoppel and Preclusion

While most legal systems contemplate the preclusive effect of previous

judgments, domestic systems have varying approaches with quite technical
rules. This chapter is not concerned with general preclusion under principles
such as res judicata, but instead with evidentiary preclusion. (160) In the common
law this is described as issue estoppel, issue preclusion or collateral estoppel.
The policy is that if a particular issue has already been litigated between parties,
it should not be re-litigated by the loser. It can even apply by way of adverse
inference where a party is seen as abusing process if it raises a factual claim in a
subsequent matter that should have been asserted earlier. (161) It has been
suggested that issue estoppel does not apply in civil law jurisdictions, (162)
although the way res judicata can apply in such jurisdictions can have a similar
impact. (163) While arbitral awards and page "790" international tribunals have
consistently respected the principle of res judicata, the preclusive effect of issue
estoppel-type scenarios is more complex and often dependent on conflicts or
national law approaches. (164)

10.13. Evidence from Local Cases or International Cases

If a domestic or international case is conducted in an open forum, confidentiality

issues may be diminished, although there is no general access to all evidentiary
material in most legal systems. At times the question is not one of evidence from
those cases but an allegation of fact as to the behaviour in the case itself. For
example, in some investment arbitration cases, the issue in dispute could be the
behaviour of a domestic court and whether its actions on behalf of its
government constituted an indirect expropriation or otherwise breached fair
and equitable treatment. In such instances, the transcripts and judicial reasoning
would be relevant evidence. Given that the arbitral tribunal would not have
heard the actual evidence before the local court, care needs to be taken in
effectively sitting in judgment on that court's behaviour in such circumstances. In
other cases there may be an issue estoppel-type scenario as discussed in the
previous section. This would require commonality of parties.

10.14. Ex Parte Hearings

If a respondent fails to participate in the arbitration, the claimant must still meet
its burden of proof and prove its claims. Arbitral rules indicate that a tribunal
may proceed but direct that a failure to attend is not an admission of the claims
made against it. (165)

In a technical sense, this situation does not change questions of burden and
standard of proof. The tribunal makes findings on the material presented before
it. If the claimant does not produce enough evidence to support its claims, it
should lose. A tribunal might even draw adverse inferences against claimant if
documents presumed to exist are not presented and no adequate explanation is
provided.

The tribunal can have a significant impact on whether claimant adequately

proves its case. A tribunal could demand production of certain documents, or
could engage in questioning of witnesses that would seek to elicit responses that
a reasonable cross-examination might produce. This is a fairly well accepted
approach page "791" but places a greater burden on the tribunal than when both
parties are present. Default proceedings are considered more generally in
section 6.15.

10.15. Change of Tribunal Composition

Because a tribunal's mandate is to consider the evidence and make a decision, an

important evidentiary question is to determine what should happen if an
arbitrator dies, resigns or is replaced. It is important that the rules express
clearly that a truncated tribunal may continue with the case or that evidence
does not need to be reheard if a replacement arbitrator is introduced, otherwise
general principles of due process might require that the matter be reheard in its
entirety. Not only would this waste time and cost, but there would also be
problems in looking for variations between a witness' testimony at different
times. (166)

10.16. Admissibility and Inadmissible Evidence

Admissibility of evidence is generally seen as a procedural matter. (167) In the

arbitral context, it will usually be subject to broad discretionary rights of the
tribunal. Numerous rules indicate that the tribunal shall determine the
admissibility, relevance, materiality and weight of evidence. (168) The issues as to
admissibility and grounds for refusal of documents also apply to witness
testimony and reasons to refrain from answering questions or even appearing as
a witness. This section deals with admissibility issues generally and also includes
specific sub-categories where admissibility may be challenged.

International adjudicatory bodies generally consider questions of weight in

relation to all submitted evidence and do not wish to hear separate claims as to
admissibility. In the WTO context, a panel considered that ‘there is little to be
gained by expending our time and effort ruling on points of “admissibility” of
evidence vel non.’ (169) Redfern and Hunter nevertheless caution advocates from
particular legal systems that are presenting before tribunals whose members
page "792" come from different legal families, as the latter, if inexperienced in
arbitration, may nonetheless display their domestic predilections. (170)

It has been suggested that the liberal practice of international tribunals to admit
virtually any evidence subject to evaluation of relevance, credibility and weight
arises from the absence of appeal, the common problem of the unavailability of
‘best’ evidence and problems with deciding cases on technical evidentiary rules.
(171) The lack of appeals as a reason for liberal admissibility rules may not be so

powerful given that there are limited appeals on factual matters in most
jurisdictions. Concerns as to due process will typically be the main reason why
tribunals are reluctant to limit or exclude evidence. (172) Tribunals are concerned
to preclude arguments that exclusionary decisions have impacted on a party's
right to be heard, although the right to a full opportunity to present a case and to
adversarial proceedings does not presumptively override a tribunal's power to
determine admissibility or weight of evidence.

International arbitration would not be expected to follow common law

exclusionary rules of evidence. Those rules were largely developed because
jurors as decision-makers were not trusted to have the ability to deal with the
relevance and weight of material that could have a strong prejudicial value as
compared to its probative value. In doing so, the common law favoured liberty
over truth, being concerned to exclude relevant material of modest value where
the prejudicial effect is likely to be significant. While that is the base position,
there will from time to time be exceptions. Inadmissibility would usually be
based on general unreliability or because of a higher prejudicial to probative
value. At other times, public policy may be a reason for excluding otherwise
probative evidence, for example, privileged or otherwise confidential
information, or illegally obtained evidence. The following sections deal with
some of the more significant categories of contentious evidence.

10.16.1. Circumstantial Evidence

All adjudicatory systems are prepared to consider and draw inferences from
circumstantial evidence. Once again, a tribunal might vary in its leniency
depending on the ability or otherwise of the parties to obtain more direct forms
of evidence. Where international adjudication is concerned, it will commonly be
difficult for a party in one country to obtain evidence in another country,
particularly where the evidence is in the control of an adverse party. In the Corfu
Channel case, the ICJ said: page "793"

such a state should be allowed a more liberal recourse to inferences of fact and
circumstantial evidence. This indirect evidence is admitted in all systems of law,
and its use is recognised by international decisions. It must be regarded as of
special weight when it is based on a series of facts linked together and leading
logically to a single conclusion. (173)

In a dissenting opinion, Judge Badawi Pash described circumstantial evidence in

international law to mean ‘facts which, while not supplying immediate proof of
the charge, yet make the charge probable with the assistance of reasoning.’ (174)

10.16.2. Uncorroborated Evidence

While there is no general obligation to corroborate evidence, a failure to do so

when this could be expected to be reasonable, might go to the question of weight
and even allow for adverse inferences in some circumstances.

10.16.3. Hearsay

Arbitral tribunals will typically allow hearsay evidence to be submitted but are
not bound to do so. (175) Hearsay is ‘a statement, other than one made by the
declarant while testifying at the trial or hearing, offered in evidence to prove the
truth of the matter asserted.’ (176) It is an example of indirect evidence. While
common law jurisdictions typically exclude hearsay evidence, there are also
numerous exceptions to the rule. Some such jurisdictions have also relaxed the
admissibility of hearsay evidence in civil matters. (177) While such evidence may
be admitted, a tribunal must still consider what weight to give it. The ICJ in the
Nicaragua case considered that hearsay testimony is not ‘of much weight’. (178)
From time to time such evidence has been excluded by the Court. (179) On the one
hand it was suggested that hearsay evidence was customarily accepted by the
Iran–United page "794" States Claims Tribunal, (180) while Judge Aldrich
suggested that the Tribunal consistently refused to base decisions upon hearsay.
(181)

One particular form of hearsay evidence is press reports. The ICJ has suggested
that ‘although it is perfectly proper that press information should not be treated
in itself as evidence for judicial purposes, public knowledge of a fact may
nevertheless be established by means of these sources of information.’ (182) An
example may be newspaper reports of share prices or weather conditions.
Conversely, a tribunal is unlikely to see a newspaper report as clear enough
evidence of the reasons for contract repudiation or expropriation. A tribunal
ought to look at the nature of the testimony, consider what the best evidence
might have been and also consider the particular party's ability to present that
evidence in deciding what importance to place on the hearsay. Some of the
matters considered relevant by domestic courts can also be a useful guide. The
Civil Evidence Act 1995 (UK) invites regard to be had, inter alia, to:

(a) whether it would have been reasonable and practicable for the party by
whom the evidence was adduced to have produced the maker of the original
statement as a witness;
(b) whether the original statement was made contemporaneously with the
occurrence or existence of the matters stated;
(c) whether the evidence involves multiple hearsay;
(d) whether any person involved had any motive to conceal or misrepresent
matters; (183)
(e) whether the original statement was an edited account, or was made in
collaboration with another or for a particular purpose;
(f) whether the circumstances in which the evidence is adduced as hearsay are
such as to suggest an attempt to prevent a proper evaluation of its weight.
(184)

If hearsay is allowed, the tribunal should ensure that there is appropriate

exploration of the witness's knowledge and recollection. A witness of fact is only
to attest to the facts and if it is a comment made by another person, the accuracy
of page "795" recollection is what is important. (185) One potential way that
hearsay might be overvalued inappropriately is if one witness tries to cover
matters best able to be attested to by other persons, simply to save time and
expense.

Another response to hearsay evidence might be to ask the tribunal to use

whatever powers it has to compel the statement maker to attend as a witness.
(186)

10.16.4. Parol Evidence

As noted, some systems treat some evidentiary questions as substantive or

quasi-substantive. For example, if a contract dispute was subject to common law
of contract, a parol evidence rule may apply, restricting entitlement to look at
extraneous evidence to modify express terms of the contract. Lew, Mistelis and
Kröll suggest that treatment of parol evidence is ‘in the grey zone between
substance and procedure’. (187)

Even if a substantive approach is to apply, if the document to be interpreted is an

arbitration agreement in a contract subject to a common law choice of law
clause, it needs to be understood that without clear wording, such a choice of law
does not necessarily apply to the arbitration clause itself, although it may. (188)
Similar questions would arise with respect to other agreements that purport to
impact upon arbitral rights. (189) Absent an agreement on applicable law, Lew,
Mistelis and Kröll argue that modern lex arbitri do not constrain arbitrators to
follow any national approach. (190) They refer to three reasons why a broader
approach is to be preferred. First, rules of evidence in court proceedings in the
Seat of arbitration should not naturally apply to arbitration. Secondly, allowing
for flexible and incremental development of appropriate rules on a case-by-case
basis promotes equality and satisfies the parties' needs and expectations.
Thirdly, such an approach can help bridge cultural gaps. (191)

page "796"

10.16.5. Demonstrative Evidence

Born and Tschanz each refer to demonstrative evidence in the sense of

representations which depict and explain evidence otherwise submitted. (192)
This includes diagrams, charts, graphs, models and even computer simulations.
In some cases, such material is simply a way of arranging and synthesising
properly admissible evidence and would thus be non-contentious. In other
circumstances, the depiction aims to be persuasive in and of itself. Care needs to
be taken to ensure that the parties and tribunal all know the intended use. If it is
to be of evidentiary weight, then its probative value needs to be properly
assessed, including the validity of its underlying methodology.

10.16.6. Illegally Obtained Evidence

There is no to clear view in public international law as to whether illegally

obtained evidence should be accepted by a tribunal. (193) Such evidence was
allowed in the Corfu Channel case. (194) Where international arbitration is
concerned, it is likely to be a discretionary matter for a tribunal and may depend
on the circumstances, (195) in particular, who obtained it illegally. If the party
presenting that evidence can be said to be in breach of its good-faith obligations
under the arbitration agreement, then there may be more justification for
exclusion. Conversely, if in an investment arbitration, such evidence presented
by a State shows that it deserves to succeed on the merits but that normal police
procedures were not followed, an arbitral tribunal might be less concerned with
general domestic regulatory principles such as those demanding that police
officers follow strict procedures in obtaining evidence. A contrary view might
also be taken.

10.16.7. Confidentiality
The key aspect of confidentiality in the context of evidence is whether
confidential material can or should be produced to the tribunal. There are other
crucial page "797" confidentiality issues in arbitration, such as general
confidentiality over proceedings and outcomes. Related to that are the
mechanisms a tribunal should use to ensure that confidentiality is protected as
far as possible and ensure that the material is only used for the purposes of the
arbitration. For this and the following two chapters dealing with evidence, the
key questions are whether confidentiality can be a basis for refusal to produce
documents or answer questions. Such claims cannot be made vis-à-vis a party in
whose favour there is a contractual right to information. The following material
deals with cases where there are no such substantive rights.

As noted in the following chapter, one of the grounds for excluding evidence
under the IBA Rules of Evidence 2010 is commercial or technical confidentiality
that the arbitral tribunal determines to be compelling. (196) Section 10.17 below
deals with an important sub-category of privilege, which is also an express
ground under those Rules. Section 10.11 above looked at whether evidence and
outcomes from previous arbitrations are subject to confidentiality obligations
interfering with admissibility. If the parties are identical, there should be no
additional confidentiality concerns as between the two disputes. If the parties
differ, this raises broader issues as to confidentiality generally and also as to due
process, where one current party at least was unable to make submissions in the
process leading to the earlier factual findings.

At one extreme, arbitral scholars believe that it is an inherent element of

arbitration that there is a general duty of confidentiality. It might be seen as an
important aspiration of commercial parties to deal with their disputes in a way
which does not adversely affect their status in the market place through bad
publicity. Many disputes also deal with confidential aspects of technology and
business know-how. At the other extreme, some argue that an undue concern for
confidentiality comes at the expense of transparency and the ability to promote
consistency through the adoption of similar logic to other arbitral tribunals.
Arguments as to consistency are themselves contentious given that there is no
binding doctrine of precedent in arbitration. Furthermore, open recourse to
other awards will not necessarily ensure consistency or streamlining of
reasoning as this may simply give competing advocates more irreconcilable
cases to build ever more costly submissions around. Where broad notions of
confidentiality are concerned, there is a general but not uniform view that this is
inherent in arbitration. The Australian case of Esso v Plowman (197) was highly
controversial, arguing that there needs to be a positive agreement in relation to
confidentiality for it to apply. A related issue as noted above is that there will
also be issues as to admissibility of documents used in a prior arbitration in
terms of confidentiality obligations flowing from the earlier arbitration. The
common law does not consider that confidentiality duties to third parties are a
bar to discovery rights. (198) The policy page "798" preference in litigation may
not apply to arbitration as the former considers that a private promise of
confidentiality must yield to the general public interest in the administration of
justice. (199) The situation might be different if the prior arbitration involved the
same parties.

Where there is a claimant common to two arbitrations on the same subject

matter without a joinder provision, for example an engineering company seeking
damages on non-performance of capital plant both against the designer and the
manufacturer, there may arise a possibility of double recovery by the claimant.
Depending on the applicable law it is open to a tribunal to consider ordering
disclosure of the detail of the first arbitration on the basis that public policy
considerations on unjust enrichment would override a confidentiality objection
to disclosure. (200)

These arguments do not apply to the question whether inter-party disputes

should allow each to claim confidentiality over material. Where evidence is
concerned, there is an important difference between inter-party confidentiality
and confidentiality vis-à-vis the rest of the world based on what has occurred
between them. Inter-party behaviour should be based on obligations to avoid
surprise and to cooperate in good faith in the resolution of the dispute. Thus
there should only be limited grounds for withholding confidential evidence that
is material to the dispute. There are a range of permutations and potential
conflicts issues that may arise. Some documents and some forms of information
are considered confidential per se. Some are considered confidential from the
perspective of the party against whom a compulsion order is sought. In some
cases, that party will have a claimed duty of confidentiality vis-à-vis a third party.
This can be a matter for a tribunal's discretion. (201) Circumstances will vary
depending on whether the confidential information is that of a party or not.
There may also be situations where it is arguable that confidentiality has been
waived. Applicable rules of waiver may be a conflicts question and may be hard
to classify. While the tribunal is always entitled to draw adverse inferences from
a failure to provide information, the justification for doing so only arises in
circumstances where there are no legitimate grounds for the failure. Where
confidential information is involved, there would be a need to consider whether
this itself is a ground for valid refusal to provide information, before adverse
inferences can be drawn.

Confidentiality can simply be seen as one procedural element to be agreed upon

from time to time by the parties. Party agreement should normally be
paramount, although if a tribunal has a broad duty of confidentiality per medium
of the nature and status of arbitration, this could even mean that an arbitrator
may declare documents to be confidential regardless of the wishes of the parties.
Where page "799" a State party is involved in investment arbitration, it cannot
rely on its own domestic laws to avoid its broader obligations under an
international treaty. Nevertheless, Article 9.2(f) of the IBA Rules may apply in
terms of political sensitivity.

In appropriate circumstances, a tribunal could deal with confidentiality concerns

by directing that confidential information be redacted and/or that non-
confidential summaries be provided. This will only be satisfactory if the material
thus produced fulfils the same evidentiary function as the confidential material
excluded. Other approaches, such as limiting access to the tribunal or even to the
tribunal chair are technically possible but are not recommended absent party
agreement, as it could lead to challenges in terms of each party's right to be fully
heard. (202) Even party agreement may not suffice given that a party cannot waive
a mandatory entitlement. Because of the divergence of views, it makes sense for
a well-drafted lex arbitri to address the issue. Nevertheless, because of cultural
differences and varying policy preferences and the need to integrate
confidentiality obligations into broader disclosure obligations to third parties,
there has not been a single harmonised position adopted in various rules. A
tribunal may wish to deal with this in terms of reference or an early procedural
order.

An important aspect of the duty of confidentiality is how an arbitrator who is

appointed in a number of related matters is to deal with information obtained in
one case, when deciding another. This may depend on the identities of the
parties in each case. The more similar they are, the more the second set could be
taken to have impliedly consented to an arbitrator that cannot wholly shed his or
her mind of information previously obtained. (203)

10.17. Privilege and Professional Secrecy

10.17.1. Introduction

The previous section looked at confidentiality in evidentiary matters generally.

This section looks at one significant sub-topic, being that of privilege. The
concept of privilege covers a range of matters that may arise before, during and
after arbitral proceedings. This section seeks to provide an overview of the
concept, its relevance to evidentiary obligations, the way a tribunal identifies the
applicable law and the way a tribunal makes findings of fact as to entitlements to
claim privilege. The analysis is at some length owing to the significant divergence
of opinion on the topic and the lack of an emerging practice in the field. The key
page "800" category is that of legal professional privilege, sometimes described
as professional secrecy or confidentiality in the civilian tradition. (204) This is
given separate attention below in section 10.17.10. As well as legal privilege,
claims for confidentiality might be made in some jurisdictions with regard to
communications between spouses and other family members, medical
professionals and religious authorities, financial advisers and journalists. (205)
There may also be special rules in relation to self-incriminating statements. (206)
Some would argue that certain categories are sufficiently well respected to
constitute general principles of law or transnational public policy. (207) Article
9.2(f) of the IBA Rules of Evidence 2010 also refers to ‘grounds of special
political or institutional security’ which would cover privilege in relation to State
secrets. (208) While not directly applicable to arbitration, general principle Article
18 of the ALI/UNIDROIT Principles of Transnational Civil Procedure (2004)
provides that ‘effect should be given to privileges, immunities, and similar
protections of a party or non-party concerning disclosure of evidence or other
information.’

Even if privilege is to be supported, there are complex and unresolved questions

as to which law is applicable, whether conflicts or other methodologies should be
employed, how to deal with differing privilege laws of each party given the need
for equal treatment and the persons or communications over which privilege
applies. Arbitration has no established norms. The parties can agree as to the
nature and relevance of privileged communications but such agreement would
be rare. Arbitral statutes and rules rarely discuss evidentiary privileges. Nor do
they proscribe the right to claim such privileges. It is broadly accepted that
privileges may be claimed, although there is no consensus on the exact ambit of
each.

10.17.2. The Nature of Privilege

While common law speaks of privilege and civilian legal jurisdictions tend to
speak of confidentiality, the latter is the guiding policy criterion in all systems.
page "801" A privilege exception to disclosure is a policy choice of confidentiality
over probative value for some broader systemic objectives than the
identification of truth in the instant case. One policy reason to perhaps limit
privilege and other confidentiality claims in arbitration is that arbitration itself is
confidential and disclosure can thus be limited to persons for whom it is
necessary for the adjudicatory processes.

10.17.3. Tribunal Discretion and the Applicable Law

While privilege is generally respected, there is a need to consider its ambit and
any exceptions. In determining the relevant rules, party autonomy should first be
considered. To the extent that party autonomy is paramount, consideration could
be given to such issues in the arbitration agreement although that is highly
unlikely. Once a dispute arises, even early consideration is likely to lead to
polarised positions as at that stage each party knows the practical implications of
disclosure of documents held by its independent and in-house counsel. Hence it
will typically fall for determination by the tribunal. Agreement might be reached
by selection of broader principles that cover the issue, such as the IBA Rules of
Evidence 2010. Failing such agreement, it needs to be understood that arbitral
statutes and rules rarely discuss privilege expressly. Hence, it will commonly be
subject to tribunal discretion. The first question is whether a law of privilege
should be identified and if so, which. Historically, it appears that many
arbitrators approached privilege questions through conflict rules. Conflicts
issues may also arise in the context of mandatory laws, where for example, the
question is whether privilege may be claimed in respect of competition law or
antitrust alleged violations.

Article 9.2(b) IBA Rules of Evidence 2010 allows for objections based on
privilege ‘under the legal or ethical rules determined by the Arbitral Tribunal to
be applicable.’ This does not confine a tribunal to a pure conflicts approach.
While the Working Party for the 1999 IBA Rules considered that it was
important that privileges be recognised, it did not seek to articulate the
appropriate applicable law. (209) The IBA Rules of Evidence 2010 added further
criteria but did not opt for a pure conflicts approach, although the reference to
‘applicable’ rules remains. The IBA Rules as to privilege are addressed further in
section 10.17.9.

There are a number of problems in using strict conflict rules to deal with
privilege issues. (210) First, there is a classification question as to whether
privilege is a matter of substantive or procedural law. Civilian jurisdictions treat
it as procedural while it has been considered as substantive in some US cases.
(211) In England, legal privilege has been acknowledged as difficult to categorise

page "802" between procedure and substance, having features of both. (212) In
addition to the uncertainty of classification, there are problems with each
categorization. If privilege is procedural, the lex arbitri will typically leave it as
part of the broad discretion of the tribunal with no further guidance. The
domestic privilege laws of the Seat should not apply even as a mere presumption,
as rules suitable for litigation in the Seat would hardly be intended by parties to
an arbitration who looked for a neutral venue supportive of the arbitral process.
Where the parties do not select a Seat, another problem with this approach is
that at the time of the relevant communication, one cannot know what law of
privilege would apply and hence what protection is provided. A tribunal or
institution selecting a Seat would also have problems if the choice favoured one
party as to privilege laws. Conversely, treating it as substantive law would be
problematic in that parties rarely select an express law of privilege. Selecting a
law broadly applicable to a contract would not typically be intended as an
agreement covering privilege issues. Forcing parties to consider such questions
at the negotiation stage could undermine the negotiation process. To the extent
that privilege is a substantive matter, the tribunal might be under a more
constrained discretion than is the case with procedural matters. This is so
because lex arbitri provisions dealing with choice of substantive law may require
a conflicts methodology, albeit allowing the tribunal to choose which conflicts
rules to apply. (213)

If the parties have not made any selection, then the tribunal would still have
broad discretions as to applicable substantive law. Applying the law of the
domicile of the person exercising characteristic performance would hardly seem
appropriate if the question is the privilege applicable to another party. If a
tribunal sought to apply a closest connection test, commentators have pointed to
a range of potentially relevant factors. Sindler and Wüstemann (214) identify: the
law of the contract; the lex arbitri; the domicile of the relevant party; the
domicile of counsel; the place any document in issue was drafted; the place
where the information was provided; the place where the document is kept; the
law providing for the ethical and other duties of counsel; and the law of the State
where the award is likely to be enforced. Petrochilos considers additional
factors, being the law under which the document was produced or the law of the
party who received the document. (215) Berger points out that a closest
connection test may need to look at the events or the page "803" communication
in applying such a test. (216) The multiplicity of factors implies that a conflicts
approach may be highly problematic where parties come from different
jurisdictions as might their counsel, and appear before arbitrators from different
jurisdictions, in a different Seat and where relevant documents may have been
created in other places.
Other problems with a conflicts approach are that domestic rules may not
necessarily be intended for arbitration, being concerned with the balance
between truth and effective legal advice in a domestic litigation system. The
policy choices made at the domestic level are also highly influenced by the view
the country takes to issues of discovery. Generally speaking, the more extensive
the document production obligation, the more clearly defined are the exceptions
such as privilege. (217) Where arbitration is concerned, leaving document
production to a broad discretion of a tribunal but applying strict conflict
approaches to privilege questions would lead to a potential policy mismatch.
There is no consensus in international litigation either. The ALI/UNIDROIT
Principles of Transnational Civil Procedure (2004) drew attention to the law of
the place with the most significant relationship to the parties to the
communication. (218) The 1970 Hague Convention on the Taking of Evidence
Abroad in Civil and Commercial Matters allows for claims of privilege under the
law of the State of execution or the State of origin and allows for declarations to
respect the laws of a State with higher levels of protection. (219) EC Regulation
1206/2001 also allows for the application of the highest level of protection. (220)

10.17.4. Cumulative Approaches

Applying domestic rules may lead to parties being treated unequally if the
conflict rule leads to documents and communications of different parties being
treated differently. That could offend against due process norms of equal
treatment. The party denied access could also assert an interference with its
right to present its case. An approach that seeks to equalise the law applicable to
both parties still has to consider which regime of the two is applicable and
whether one party can page "804" waive its more preferential rights to deny
those to its opponent for strategic reasons. (221) Another approach is to attempt to
consider cumulatively various national laws that may have a connection. (222) It is
not clear whether this should be restricted to the parties' own laws or whether it
includes the law of the Seat or the law most closely connected to the evidence.
(223) If the laws are identical, a tribunal can feel comfortable. If not, the next

question would be whether the higher or lower level should be chosen.

10.17.5. Lowest Common Denominator

One approach is to look for the lowest common denominator in protective

standards. (224) Meyer argues for this as the starting point and then a
consideration of whether there are any reasons to apply a higher standard. To
that end, he advocates considering the rationale underlying the privilege and for
the tribunal to then determine whether in the arbitration context, that justifies
interference with best evidence of the truth. It is not clear how tribunals from
different legal backgrounds could be expected to uniformly and consistently
make such determinations, although the aspiration is sensible. As noted below,
however, such an analysis could even lead to a highest common denominator
presumption. (225)

10.17.6. Most Favourable Privilege

Conversely, Schlabrendorff and Sheppard advocate the application of the ‘closest
connection’ test and recommend an adjustment where the application of this test
would lead to different privileges for each of the parties or for different
categories of documents. (226) As to the adjustment, they advocate the application
of the most page "805" favourable privilege. This is the approach in the CPR
Protocol and the ICDR Guidelines, (227) as well as the Hague Convention and the
EC Regulation noted above. The authors rightly point out that the ambit of
privilege entitlements relates strongly to the ambit of presumptive production
obligations. As noted above, the broader the latter, the broader the former. As
international arbitration has gravitated towards stronger production obligations
than civilian practitioners were used to, the suggestion is that this supports a
similarly more favourable privilege approach on an equal treatment basis. (228)
Another policy reason to support this approach is that legal privilege does not
deal with direct evidence in relation to the dispute which should be produced
anyway, but rather, deals with information which is generally a matter of opinion
or admission. While it is true that privilege may sometimes apply to
unfavourable facts communicated to external counsel, it is at least arguable that
such facts should have been disclosed in their own right. At the very least,
communicating a fact to counsel should not shelter the fact from a primary
disclosure duty that otherwise exists. Berger also argues that such an approach
does justice to the reliance interests of parties who should feel that they will not
be required to produce information considered privileged under their own laws
and that such an approach avoids conflicts with public policy. (229)

Schlabrendorff and Sheppard rightly concede that in some cases it may be

difficult to determine just which set of privilege rules are most favourable when
there are a multiplicity of factors including ethical standards, waiver and
inference. (230) Excluding relevant evidence may also increase the chances of due
process challenges. (231)

page "806"

10.17.7. Equality and Reasonable Expectations

Some tribunals have felt unconstrained by conflicts methodology. (232) Other

approaches might look at equal treatment and reasonable expectations of the
parties. Broad notions of equality might mean that one party should not be
privileged in relation to the kind of evidence it must produce, otherwise the
elements of the contest are unequal. (233) Reasonable expectations as a test, seems
consistent with the consent basis of arbitration. While equal treatment and
reasonable expectations each have merit, they are difficult to define and
reconcile. Does equal treatment mean applying to each their domestic rules or
finding a common rule to apply to each? In the latter event, if it is the rule of one
party and not the other, is that true equality? An example would be the internal
notes by in-house counsel as a background to the drafting of the central contract
where the parties' legal systems have differing approaches to in-house counsel
privilege and where the applicable law of the contract allows for extraneous aids
to interpretation. Raeschke-Kessler suggests that under this approach a party
should only be able to call for production of internal documents to the extent
that it is itself required to disclose such documents. (234) Yet it might be fortuitous
that this party is the one who needs to obtain the document and comes from a
system that does not afford the privilege. That might not be true equality in
terms of the ability to present the case.

Similarly where reasonable expectations of the parties are concerned, would

each party's reasonable expectation be that they have the same protections that
they have domestically, or that the tribunal will equalise treatment, or ignore
domestic litigation rules in pursuit of the truth? Schlabrendorff and Sheppard
raise an interesting example of a US party being asked to disclose a
communication with a French in-house lawyer. (235) Would the US party expect
the same protection as at home or should it be expected to understand the
differing position in France? Schlabrendorff and Sheppard suggest that under
such a test, communications which are privileged when made would be expected
to remain privileged and hence privileges recognised by a domestic court will
likewise be recognised by international tribunals. Yet the more the client and
counsel are experienced in page "807" arbitration and understand the differing
views about legal privilege, the less such an expectation holds true. Legitimate
expectations may also be problematic if a party has chosen arbitration in part to
get away from domestic evidentiary norms and biases. To the extent that the
nationality of counsel is a relevant factor in determining expectations, this can be
problematic if it encourages parties to select counsel to promote their privilege
entitlements. It is also hard to apply a reasonable expectation test when parties
and counsel come from different legal families. Another reasonable expectation
is presumably that parties will be treated in a fair and reasonable manner. (236)
This may however be circular in terms of whether fairness equates to exact
equality of treatment or equates to equal respect for each party's domestic
tradition. Another possibility is that a legitimate expectation of a party is that by
choosing arbitration where broad discretions are granted to tribunals, they have
effectively waived their rights to insist on strict legal privilege, preferring a
practical application of fairness and efficiency tests on a case-by-case basis to a
rigid application of foreign conflict of law rules. Conversely, it could be tenably
asserted that certainty and consistency is at least of equal merit. (237)

The above analysis demonstrates why no single theory is ideal, although the
problem is real and requires a solution. The more reasonable expectation might
be that a party aware of the problem can seek an agreement with the opposing
party from the outset. Agreement can be indirect, through the choice of
guidelines or recommended principles. Absent express agreement, at least
having the parties agree to tribunal determinations would be preferable. Over
time, a dominant position is likely to emerge and parties who are not happy with
it can seek agreement to the contrary. The new IBA Rules of Evidence 2010 have
more detailed provisions and are at least a starting point for consideration.
These are discussed in section 10.17.9 below.

10.17.8. Waiver

If a conflicts approach is adopted, there may also be arguments as to whether

privilege has been waived by conduct. Such conduct could also be relevant to a
broad fairness and efficiency discretion unconstrained by conflicts approaches.
Mandatory laws might even apply if they purport to bar waiver, which might
even mean that under some conflicts approaches, agreement of the parties could
be ignored. The converse situation where similar issues may arise is where a
party has previously conceded that privilege applies and then tries to take a
different approach at a later stage.

Because privilege has classification issues as to substance/procedure, there will

be similar issues where waiver is concerned, although the relevant law of waiver
should be the same as the law of privilege. Where waiver is concerned it is
important to understand that because different legal families treat it as either a
page "808" party entitlement or a lawyer's ethical duty, they differ as to whether
and by whom the privilege may be waived. Another issue is whether inadvertent
disclosure will constitute waiver. Where privileged or confidential documents
are inadvertently produced, parties might have a pre-existing clawback
agreement providing for their return and indicating that they may not be relied
upon in the adjudicatory processes. (238)

10.17.9. IBA Rules of Evidence and Privilege

The IBA Rules of Evidence 2010 provide mechanisms for determining questions
of privilege. Article 9.2(b) of the 2010 Rules retains the prime provisions in the
1999 Rules and indicates that evidence shall be excluded where it is subject to a
‘legal impediment or privilege under the legal or ethical rules determined by the
Arbitral Tribunal to be applicable…’. When the IBA Rules of Evidence 2010 were
developed, leading experts were surveyed as to whether the rules should
provide further guidance, with those surveyed dividing roughly evenly both for
and against. (239) Similarly those surveyed divided roughly equally on whether a
choice of law approach should apply or whether there should be default rules.
(240) The final version is more open-ended in terms of relevant factors for

consideration. Under a new Article 9.3 the tribunal is to consider:

(a) any need to protect the confidentiality of a Document created or statement

or oral communication made in connection with and for the purpose of
providing or obtaining legal advice;
(b) any need to protect the confidentiality of a Document created or statement
or oral communication made in connection with and for the purpose of
settlement negotiations;
(c) the expectations of the Parties and their advisers at the time the legal
impediment or privilege is said to have arisen;
(d) any possible waiver of any applicable legal impediment or privilege by
virtue of consent, earlier disclosure, affirmative use of the Document,
statement, oral communication or advice contained therein, or otherwise;
and
(e) the need to maintain fairness and equality as between the Parties,
particularly if they are subject to different legal or ethical rules.

Because the rules speak of material ‘in connection with’ certain processes, it will
also cover background and internal documents for the purpose of legal advice
and page "809" settlement and not simply documents directly presented for
those activities. (241) The Commentary states that ‘the standard to be applied is
left to the discretion of the arbitral tribunal….’ Expectations of the parties and
their advisors are to be taken into account. The key provision is likely to be that
of fairness and equality as most commentators would not wish to see parties
subject to differing rules under a conflicts approach. (242) This criterion also
implies that those utilising the IBA Rules are most unlikely to adopt a pure
conflicts approach that would have different rules for each party. Nevertheless,
for reasons articulated above, any alternative conflicts approach is not easy to
identify under fairness and equality tests.

The protection can only be relied upon in good faith. Thus a party should not be
entitled to proffer a key document during settlement negotiations simply to
afford it protection. (243) In any event if it is a contemporaneous document it
would need to be disclosed. Carter suggests that the reference to ‘equality and
fairness’ may lead to application of the broadest privilege where the rules
applicable to each party differ as this is the approach that would both provide for
equal treatment and give credit to prior expectations. (244) He also suggests that
the rules need to maintain flexibility as parties may approach privileged claims
in differing ways within an arbitration and proper responses may be needed on a
case-by-case basis. (245)

10.17.10. Legal Privilege

General principles discussed above apply to all forms of privilege. This section
gives further attention to legal privilege as it is the single most significant
category. Contests as to legal privilege can combine complex legal questions in
terms of applicable law and coverage of in-house counsel and factual questions
in the context of the purpose of the communication. Where legal privilege is
concerned, the rational is that justice is served if there is an entitlement to seek
legal advice knowing that confidences will be maintained. Absent any privilege,
clients might withhold adverse information from their legal advisers or not seek
advice at all. page "810" A related policy justification is that counsel can provide
cautionary advice if the communication is protected. The more that a party is
forthcoming and receives accurate advice, the more likely that settlement will be
achieved and that the transaction costs of formal dispute settlement can be
avoided or at least minimised. Many communications between counsel and client
would also unfairly influence the decision maker if disclosed.

Legal privilege can typically cover pure advice and also communications in the
context of contemplated or actual adjudicatory processes. In the civilian
tradition, professional secrecy is typically considered as a matter of criminal law
and professional ethics rather than a rule of evidence and as such cannot be
waived by the client. (246) Civilian jurisdictions also differ as to whether
communications between lawyers are confidential and whether they may or may
not be disclosed to the client. (247) In Akzo Nobel, the ECJ confirmed confidentiality
of communications between counsel and client where the communications were
made for the purposes of the exercise of the client's rights of defence and
emanated from independent lawyers. (248) Confidential communications between
counsel and clients has been considered to be an element to the right to a fair
trial under Article 6 of the 1950 European Convention on Human Rights. (249)
Nevertheless, the principle is not absolute and is subject to proportionality tests
as applied generally under the Convention. (250)

The previous section outlined the relevant provisions of the IBA Rules of
Evidence 2010. ICDR Rules direct a tribunal to take into account ‘applicable
principles of legal privilege’. (251) The 2012 revision of the ICC Rules amended the
provision on confidentiality. The tribunal now has the power to make orders
regarding the confidentiality of proceedings and any other matters in connection
with the arbitration. (252) Others, such as the LCIA Rules simply do not mention
privilege issues. While the English Arbitration Act 1996 does not expressly
address page "811" legal privilege, section 43, which empowers the court to
support arbitration through orders to witnesses, provides that a person cannot
be compelled to produce documents ‘which he could not be compelled to
produce in legal proceedings’. The court at least is thus obliged to respect legal
privilege rights.

Litigation privilege should apply to communications between a party and its

experts. The same would relate to communications between counsel and
witnesses in terms of developing written statements and clarifying potential
evidence. Privilege may also apply to non-lawyers giving similar kinds of advice
in relation to intellectual property or customs law, although this might not be so
under some applicable laws. While much of the debate is about whether privilege
should ever apply and if so by what law, there are also important factual
questions as it is not every communication between a client and counsel that is
protected. There is a need to consider the nature of the advice or work and
analyse this in the context of the policy underlying the notion of privilege.

10.17.11. Privilege and In-House Counsel

Civilian and common law jurisdictions treat in-house counsel differently when
considering questions of privilege. The common law will tend to allow
communications with in-house counsel to be covered by legal professional
privilege. Civilian legal systems are less inclined to do so although there is no
common approach with France, Italy, Sweden and Switzerland denying such
privileges while these are allowed in Belgium, Denmark and Spain with the
position in Germany somewhat unsettled. (253) The ECJ has tended to deny
protection to in-house counsel. (254) The policy concern with in-house counsel is
that if the lawyer is subject to directions from senior corporate officers, there is
no need to be concerned that a lack of privilege will preclude the party seeking
legal advice. Furthermore, there is a greater risk that parties and their in-house
advisers could manipulate privilege entitlements to conceal key information. (255)
Even in the common law, the activities of in-house counsel may not actually be
legal advice but may simply be a normal executive function which would not be
subject to the privilege. (256)

page "812"
There might be problems if there is a disparity between the demands of an
arbitral tribunal and the ethical duties of counsel vis-à-vis allegedly privileged
communications. One variation is whether in-house counsel are expected to be
members of a local Bar. (257) This varies between jurisdictions. However, it would
be rare that a particular counsel’s national Bar sought to prevent members from
responding to directions of properly constituted foreign adjudicatory bodies. (258)
While there is as yet no emerging international standard, Meyer suggests that the
more expansive standard of protection of the common law is more appropriate
for arbitral purposes. (259)

10.17.12. Settlement Discussions and ‘Without Prejudice’ Communications

It is generally accepted that without prejudice or otherwise confidential

comments during settlement negotiations are not admissible before a tribunal.
(260) Privilege as to settlement also applies to discussions prior to commencement

of an action. (261) If these are tendered in any event, it may compromise the
tribunal and in extreme circumstances, lead to a need for reconstitution. (262) For
example, a generous settlement offer may simply be motivated by a desire to
extricate oneself from a dispute rather than being based on a belief as to liability.
While that is so, it is preferable for counsel to clearly and expressly indicate the
confidentiality of such communications rather than simply use terms of art such
as ‘without prejudice’ which may not be understood by lawyers from some
jurisdictions.

The situation will be different if there is a disputed assertion that the matter was
settled, in which case that alleged agreement goes to arbitral jurisdiction and
needs to be determined by the tribunal. The situation might be more complicated
if page "813" one party is negotiating with a non-party, perhaps because of an
indemnity or a separate cause of action. Such a situation may arise in a
construction dispute where there is a subcontractor or a negligence claim where
there is an insurance underwriter.

It is generally accepted that privilege applies to communications in mediation,

conciliation and similar ADR processes. (263)

10.17.13. Self-incrimination

It is not clear whether the privilege against self-incrimination applies in

arbitration. (264)

10.17.14. Resolving Contested Privilege Claims

A tribunal has a difficulty in deciding on admissibility without examining the

material. Yet if it does so and upholds the privilege, it cannot easily ignore what it
has seen. Hunter and Travaini see the advantage in the appointment of a neutral
expert to consider privilege/confidentiality arguments. (265) This is discussed
further in section 11.8.1 below as to the use of experts for document production
challenges.

10.17.15. Annulment and Enforcement

Whichever view is taken by a tribunal on privilege issues, there can be potential

enforcement challenges, although these should not be upheld in the vast majority
of cases. If a tribunal does not apply a privilege operating at the place of
enforcement, the enforcement court might consider this to be a breach of public
policy. Although legal privilege should only be considered in this way if it is truly
a transnational public policy, a failure to allow for equality of arms might be a
violation of international procedural public policy. (266) Conversely, if an
application for document production is refused on the grounds of privilege, the
applicant might argue page "814" that it has been denied a full opportunity to
present its case. (267) Brower and Sharpe suggest that a court is more likely to
intervene where privileges are not recognised. (268)

10.18. Inspections, Site Visits and Sampling

In some disputes it may be appropriate or even necessary for the tribunal to

inspect a place or some machinery or traded goods. A tribunal might also need to
inspect a system or process where its performance is in issue or is important
background.

In some cases, without such an inspection, a tribunal will simply be unable to

make the requisite determination. In other cases, a tribunal may best understand
conflicting evidence when it can view the relevant subject matter. An example
would be a construction dispute where there is a value in understanding the
layout and the elements that the witnesses will address. In some cases a tribunal
may wish to undertake experiments in aid of their determination. For example, a
claim that garments are faulty because the dye leaches on washing, could be
tested via experimentation. Site inspections might also be conducted by experts
for similar reasons.

If sampling is to occur, there needs to be proper attention to the degree of

sampling that would be statistically significant. It would be best if the parties
agreed to an appropriate amount. In case of disagreement, a tribunal might even
wish to obtain the advice of a statistics expert as to an appropriate sample size.

Site inspections are expressly referred to in many statutes and rules, including
the English Arbitration Act 1996 (269) LCIA Rules (270) and UNCITRAL Rules 2010.
(271) The ICSID Arbitration Rules also allow the tribunal to visit any place

connected with the dispute. (272) Nevertheless, most statutes and rules make no
express reference but would treat this as implied within general powers. Article
7 of the IBA Rules of Evidence 2010 provides a discretion for the tribunal at its
own motion or on the request of a party to allow for such inspections, subject to
the provisions of Article 9.2 dealing with grounds for refusal. Alternatives to
page "815" site visits might include computer simulations, models and video and
photo material. In all cases the tribunal needs to be clear in its own mind as to
the distinction between evidence and mere assistance to better understand other
evidentiary material.

There are also a number of important protocols to employ when engaging in

such inspections. If an inspection is to occur, it is sensible to make a procedural
direction to this effect that also gives an indication as to how it is to be organised,
who is to be present and what will occur during the process. Any site inspection
should be undertaken in the presence of representatives of each party. (273) A
tribunal might be empowered to engage in a site inspection without any
representatives of the parties, but only where this seems reasonable. The
process of inspection should also be distinguished from the process of oral
testimony. On a site visit, it will typically be the case that employees or other
representatives of one of the parties will assist the tribunal in its inspection. It is
important to clarify that any communications at such time are not to be taken as
evidence. If those employees are to give evidence they should do so as witnesses
in the usual manner. The UNCITRAL Notes make clear that explanations given at
a site visit are not evidence in the proceedings. (274) Arbitrators should not
normally question people as to matters of contested evidence during a site
inspection in ways where those answers would be relevant to their
determination. Instead the tribunal will merely ask questions that aid it in the
logistics and effectiveness of its physical inspection. Similarly, the parties'
representatives are not there to make submissions or present evidence on that
occasion. It has been suggested that best practice is to direct that there be no
transcript and that what is said should not form part of the record, otherwise the
inspection might become overly formalised. (275)

10.19. Evidence from Non-Parties

At times a party may wish to rely on documents in the possession of persons who
are not parties to the arbitration agreement or rely on their oral testimony. As a
general rule, the tribunal has no power of compulsion over such persons. In
some countries, an arbitrator is given power in the arbitration statute to
summon persons and to demand document production. (276) In some
jurisdictions, an application may be made to a court for such purposes. (277)
Articles 3.9 and 4.9 of the IBA Rules of Evidence 2010 deal with requests to a
tribunal to take whatever steps are legally available to obtain requested
documents or summon witnesses respectively. In some cases, a tribunal will
simply authorise a party to itself approach courts page "816" for their assistance.
It is not clear whether Article 3.9 of the IBA Rules of Evidence should be
interpreted so that it is mandatory to seek leave of the tribunal before seeking
court assistance. Different views have been taken in Methanex. (278) If the proper
view was that permission needed to be sought, there would then be a problem if
evidence was obtained without permission and then tendered to the tribunal.
There are no clear rules on how to treat improperly obtained evidence, with
conflicting policy arguments from due process, fairness and relevance
perspectives.

If a third party appears as a witness he or she may be asked to produce relevant

documents, although again compulsion is not generally possible. If the party
gives oral evidence and refuses to produce documents thought to be in its
possession, adverse inferences might well be drawn, but only where reasonable
to do so. (279) A person who has not consented to arbitrate may have many
reasons for not rendering assistance. The same is the case with refusals to
answer questions, although if a person is there anyway and simply does not
answer a key question, there might be less alternative hypotheses to an adverse
inference. If third-party evidence has been called for, their attitude to a voluntary
request may also impact upon the tribunal's decision as to whether to hold
evidentiary hearings in non-seat jurisdictions.

10.19.1. Unsolicited Evidence from Non-parties

A growing tendency in investment arbitration cases is for non-governmental

organisations to seek hearing rights, purport to present amicus curiae briefs and
submit relevant evidence. This section deals with submission of evidence. The
following section deals with amicus curiae briefs where arguments will be
presented. Subject to party autonomy and duties of confidentiality, a tribunal's
broad discretionary powers would allow for this in appropriate circumstances. A
tribunal would need to be circumspect, however, as the party in whose favour
the evidence is presented could easily have presented it directly. Hence the
attempted submission by a third party will commonly be contrary to party intent
which should be paramount.

There are also a range of practical problems with such material. On what basis
can a tribunal resolve to be selective with such material given duties of
efficiency? Should such material be provided to the parties with an opportunity
to respond and who should ultimately pay for that part of the process where a
global costs order is not thought appropriate? Should opportunities be provided
to other potentially interested parties to present similar material? Should a
tribunal make decisions as to admissibility before or after considering the
evidence sought to be page "817" submitted? (280) Because of these concerns, the
first thing a tribunal should do when faced with unsolicited evidence is to seek
the guidance of the parties as to their preferred process.

10.19.2. Amicus Curiae Briefs

Amicus curiae were considered in section 7.16 as to the basis of any appearance
right. This section gives further consideration to evidentiary issues. Some
authors consider that it is part of the inherent jurisdiction of any international
tribunal to accept amicus curiae briefs. (281) Whether this should apply to
international commercial or investment arbitration depends in part on the view
one takes as to its essential nature. Once again the more it is consent based, the
less one would justify the rights of audience of third parties. Duties of
confidentiality also impact upon the ability of amicus curiae to be aware of the
key issues that would need to be addressed in their submissions. In areas where
amicus curiae briefs might be beneficial, an arbitral tribunal can approach the
issue by expressly calling for relevant evidence from the parties. In most cases
where amicus curiae briefs are involved, these are not truly provided as friends
of the court but are effectively seeking to bolster the position of one of the
parties. Where the court wishes to have its own independent material, it may
appoint its own expert.

Some investment arbitration tribunals have accepted such submissions. The

NAFTA tribunal in Methanex accepted amicus submissions. (282) Two other
similar cases rejected applications. NAFTA tribunals have relied on broad
powers under the UNCITRAL Rules. The Statement of the NAFTA Free Trade
Commission on Non-Disputing Party Participation also expressly concludes that
tribunals have the power to accept and consider amicus curiae briefs, although
this is seen as recommendatory rather than a binding ruling. ICSID tribunals
have relied on Article 44 of the ICSID Convention. In addition, Rule 37(2) of the
ICSID Arbitration Rules now expressly indicates that tribunals have the power to
accept and consider written submissions from non-disputants. The 2004 US
Model BIT and recent US Free Trade Agreements typically include express
authority to accept and page "818" consider amicus submissions. (283) Article 15
of the Rules of Procedure and Evidence of the Iran-US Claims Tribunal also
allows for the receipt of third-party statements. While each regime asserts that
the power exists, each sees it as a discretionary matter for the tribunal.

NAFTA, ICSID and the WTO all call for applications for leave to file briefs so that
the tribunal can first consider whether to accept them. On the one hand, a two-
stage process may minimise the amount of material ultimately to be considered.
On the other hand, it is difficult to determine the value of a potential brief until
its contents are known. One important factor taken into account by NAFTA and
ICSID tribunals is whether there is a significant public interest component in the
particular dispute. This is a difficult test to employ as all investment disputes
dealing with a State as a party have a public interest element for that reason
alone. The more the investment centres on key services and facilities, the more a
tribunal is likely to see a particularly significant public interest. A tribunal will
also consider the suitability of the person seeking to tender amicus briefs. In
Aguas Provinciates, the tribunal indicated that ‘(i)t is not enough for a non-
governmental organisation to justify an amicus submission on general grounds
that it represents civil society or that it is devoted to humanitarian concerns. It
must show the Tribunal in specific terms how its background, experience,
expertise, or special perspectives will assist the Tribunal in the particular case…’.
(284) The tribunal also wished to judge the independence of the petitioner.

In considering whether to accept amicus submissions, there is also the question

of timing and delay and the added costs to the other parties of reviewing those
submissions. A tribunal would have no cost powers against the amicus if the
ultimate submission wasted people's time. In considering third-party
participation, a tribunal is also concerned with its general duty to promote
arbitration and respect for that form of dispute settlement. Where investment
arbitration is concerned, it is impossible to ignore criticisms from academia and
non-governmental organisations as to lack of transparency with respect to
matters of broad public interest. That does not mean that the criticisms need to
be accepted. Nevertheless, the ground-swell of criticism may have some
influence on decisions by certain governments to leave the investment treaty
system. It is in that context that investment treaty arbitrators may give thought
to broader concerns as to the sanctity of the system when making individual
procedural rulings.

page "819"

In addition to third-party rights to make submissions and present their own

documents, there is the related question as to whether they should be given
access to documents in the arbitration so that they can make meaningful
submissions on key matters. This question could arise more frequently in
investment arbitration. Basic principles of arbitration, consent and
confidentiality would not allow for this, although some regimes offer more in the
way of public access to documents. This is the case with NAFTA Chapter 11. (285)
Third parties would not have the right to bring witnesses or to cross-examine
witnesses of the parties. By definition, they are not needed as advocates for a
position being argued from the evidence before the tribunal. They are there to
provide further assistance that is not being provided by either of the parties.

10.20. Overseas Evidence

In some cases a party might request the tribunal to take evidence from persons
in countries outside the Seat who are not willing to attend voluntarily and to
consider documents in the position of third parties who again choose not to
submit them voluntarily for use in the hearing. Generally speaking, arbitral
tribunals will not have powers to subpoena documents or persons, although
there are exceptions in some jurisdictions. The Hague Convention on the Taking
of Evidence Abroad in Civil or Commercial Matters (286) does not apply to arbitral
proceedings, although Redfern and Hunter note that many signatory States do
lend judicial assistance to arbitral tribunals with a Seat in another contracting
State. (287)

Redfern and Hunter suggest that the most common way of compelling the
production of such evidence is indirectly, through the ability to draw adverse
inferences from unexcused failure to produce requested evidence. (288) However,
this would be problematic in many situations. The adverse inference ultimately
has to be against the claims of a party. Absent any controlling power of the
tribunal or a party to the proceedings over the third party, the latter has a
natural excuse. The non-party could argue that it has never consented to giving
an arbitrator a right to pass judgment on its basis for refusal. Adverse inferences
will be much easier where it is clear that a party to the proceedings has power
over the person whose evidence or documents are sought and chooses not to
exercise it.

page "820"

10.21. Preserving Evidence

A tribunal or court might be asked to make orders preserving evidence where it

is thought that there is a reasonable apprehension that it may be destroyed or
removed from jurisdictional control. Preserving evidence can also include
keeping equipment in its existing pre-repaired state so experts and the tribunal
can view it in assessing a damages claim. (289) This is discussed further in section
8.2.10.1 dealing with provisional measures.

10.22. Timing of Evidential Material

10.22.1. General Principles

Access to information is crucial in adjudications that are resolving factual

disputes. (290) There are important questions of timing in the exercise of a
tribunal's evidentiary discretions. The first question is what evidence should be
provided concurrently with arbitration requests and answers and subsequent
written submissions. This relates to the obligation on each party to produce the
documents on which it intends to rely. The next timing issue is when document
production should be ordered as a result of inter-party requests if that is to
occur. Related to that, is the power of a tribunal to direct production where a
party does not voluntarily wish to do so. As noted, this book uses the phrase
document production in relation to document requests between the parties and
tribunal rulings in that regard, rather than the common law term of discovery. It
refers to document presentation to deal with unilateral submission of documents
by the parties. A further timing issue relates to the amount of time that should be
allocated between production of witness statements and documentary evidence
and the hearing itself, so that the opposing party can take instructions and seek
to mount a contrary factual argument. Finally, there is a question as to whether
new evidence can be submitted after deadlines set down by the tribunal.

From a policy perspective, it is even wrong to consider the timing of evidence in

the context of these forms of document presentation alone. Domestic legal
systems include an array of mechanisms whereby one party might seek to elicit
facts and concessions from its opponent and which impact upon the need for and
utility of documents as evidence and affect timing of other stages. This may
include interrogatories, depositions, requests for admissions of fact (with
potential cost page "821" implications for refusals), physical inspections and
examination of witnesses. (291) Only the latter two are typically utilised in
international arbitration. These timing issues can be determined by agreement of
the parties or failing that, by tribunal discretion, in each case subject to
mandatory norms. Mandatory norms requiring due notice of the matters to
address should apply to the evidence on which the party intends to rely. (292)

There is also a question as to whether written submissions should be concurrent

or sequential in the context of document presentation. This poses challenges in
terms of the need to provide equal treatment. Where evidence is to be attached
to submissions, sequential submissions mean that the responding party is on
notice as to the body of evidence it is seeking to refute. Rightly or wrongly, a
claimant might argue that this favours respondent, although it would be rare for
such a challenge to be legitimately upheld as a respondent is entitled to first
understand the case against it before responding.

10.22.2. Deadlines for Evidentiary Submissions and Documentary Production

A tribunal will normally impose a ‘cut-off date after which further evidentiary
material may not be submitted. (293) Nevertheless, a tribunal needs to retain a
discretion in appropriate circumstances to allow new evidence. The principles
are considered in the following section.

10.22.3. Admissibility of New Evidence

There are two senses in which we can speak of new evidence. The first relates to
evidence that was not in existence when evidence was required to be produced.
The second relates to evidence that was always in existence but which was not
produced when an opportunity had previously been provided. In each case the
tribunal must consider whether to grant leave for the evidence to be presented.
(294)

There are a number of potentially conflicting principles to consider. A tribunal

must always be prepared to consider the reasons why an extension page "822" is
sought. Many rules require ‘exceptional circumstances’. (295) A tribunal might
consider the fault, if any, of the party seeking to submit late evidence. A party,
who for whatever reason has only gathered previously existing evidence after
deadlines have passed, might typically seek to rely on its right to an adequate or
full opportunity to present its case. The tribunal is not bound to accede to such
requests as the mandatory obligation is only to give each an ‘opportunity’. If the
previous procedural deadlines were adequate and a party simply did not comply,
it cannot say that it did not have an adequate opportunity. However, a tribunal is
not bound to reject such an application as other considerations are relevant. It
might allow the evidence if there is minimal inconvenience to the other party
and the tribunal, or if the inconvenience can be compensated with an
appropriate costs award. A tribunal will be more inclined to do so if the new
evidence is crucial and fills a gap in the prior evidentiary record. Conversely, in
many cases the inconvenience might be unacceptable, for example, if the other
party would have insufficient time to prepare and respond to the new material
or if the final award would be unduly delayed in order to give them sufficient
time. Another relevant factor is whether the material sought to be produced
should have been presented in any event by the other party. For example, one
party might gain last minute access to documents which should have been
disclosed by its opponent as relevant to the claim or in response to an order for
document production. If it was wrongly not produced, the other party could have
made an application for an order that full production take place, so producing
the newly acquired document itself seems similar in substance.

In Protiva, the Iran–US Claims Tribunal considered ‘fundamental requirements of

equality between, and fairness to, the Parties, and the possible prejudice to
either Party’. (296) It considered that ‘the orderly conduct of the proceedings also
requires that time limits be established and enforced.’ (297) The tribunal
considered that if there is no acceptable justification for lateness, it can be
rejected. (298) It allowed some documents where there would not have been
prejudice to the other party and another document where it would have been in
the possession of the other party in any event. (299)
In some cases a tribunal might need to consider whether the new material could
give rise to a new request for arbitration if its admissibility is rejected. That
might be a factor in favour of accepting it on efficiency grounds. Conversely,
some might take the view that the evidence can be adequately presented in the
new claim. page "823" Finally, in some cases new material can be rejected simply
on the basis that it is irrelevant to the issues in dispute. An example would be an
investment arbitration case challenging a bureaucratic determination on the
basis that it did not constitute fair and equitable treatment under treaty norms.
The bureaucrat should not be able to rely on subsequent evidence to justify a
decision that when taken, was inappropriate. Similarly, a decision to terminate a
construction contract in the face of building inadequacies needs to be analysed in
the context of the factors relied upon for termination and not subsequent defects
that have come to light during preparation for the arbitration proceedings.

It is also worth remembering that due process challenges can arise either way
depending on whether new evidence is accepted or rejected. Nevertheless, a
party would find it hard to challenge an award on the basis that the tribunal
improperly allowed late submissions unless it did so without giving it adequate
warning and an appropriate opportunity to respond. The most important thing is
for the tribunal to set deadlines with enough advance warning to the parties, and
ideally with their agreement, so that no due process arguments can be raised if
new material is rejected. It is also appropriate to indicate the consequences of
failure to meet deadlines at the outset and whether applications for late
submissions will even be entertained. This is sensibly recommended in both the
UNCITRAL Notes and ICC, Techniques for Controlling Time and Cost in
Arbitration. (300)

page "824"

1 Redfern and Hunter suggest that perhaps 60% to 70% of cases turn on facts
rather than the application of principles of law. Nigel Blackaby et al., Redfern and
Hunter on International Arbitration, 5th edn (Oxford: Oxford University Press,
2009), 384. The authors suggest that in the vast majority of other cases, the
outcome is based on a combination of factual and legal issues. Hence there are
only a very small number of cases where factual determinations are not
required.
2 Matti Kurkela & Hannes Snellman, Due Process in International Commercial

Arbitration (New York: Oceana Publications, 2005), 35.

3 Durward V. Sandifer, Evidence before International Tribunals, rev. edn
(Charlottesville: University Press of Virginia, 1975), 1.
4 Van Vechten Veeder, ‘Evidence: The Practitioner in International Commercial

Arbitration’, Forum du Droit International 1, no. 4 (1999): 228.

5 Nietzsche observed of historical analysis that ‘there are no facts, only
interpretations’. See Friedrich Nietzsche, The Will to Power, trans. Walter
Kaufmann & R.J. Hollingdale (New York: Random House, 1967), §481. Where
legal adjudication is concerned, the added challenge is that it is a battle between
contested interpretations. It has even been asked whether lawyers are really
interested in the truth or only what can be proven? Arthur L. Marriott, ‘Evidence
in International Arbitration’, Arbitration International 5, no. 3 (1989): 280.
6 For a discussion of these differences, see Yves Derains, ‘La pratique de
l'administration de la preuve dans l'arbitrage commercial international’, Revue
de l'Arbitrage (2004): 781–802.
7 Claude Reymond, ‘Civil Law and Common Law Procedures: Which Is the More

Inquisitorial? A Civil Lawyer's Response’, Arbitration International 5, no. 4

(1989): 358.
8 Peter Hafter, ‘The Provisions on the Discovery of Internal Documents in the IB

A Rules of 1999’, in Global Reflections on International Law, Commerce and

Dispute Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen
et al. (Paris: ICC Publishing, 2005), 355.
9 More meaningful differences may arise between those who might infer intent

from documentation and from hypotheses as to likely behaviour as opposed to

choosing between the more truthful of conflicting witnesses.
10 See, e.g., the Woolf reforms in England: Lord Woolf, ‘Access to Justice, Interim

Report to the Lord Chancellor on the Civil Justice System in England and Wales'
(1995); Lord Woolf, ‘Access to Justice, Final Report to the Lord Chancellor on the
Civil Justice System in England and Wales' (1996).
11 DIC of Delaware, Inc. & Underhill of Delaware, Inc. v. Tehran Redevelopment

Corp. & the Government of the Islamic Republic of Iran, in Albert Jan van den Berg
(ed.), Yearbook of Commercial Arbitration Volume XI (The Hague: Kluwer Law
International, 1986), 336.
12 It has been suggested that evidence, in the sense of actual facts pertaining to

such substantive law requirements, is substantive in nature, while evidence in

the sense of the dynamic process by which facts are presented and verified are
more procedural: Salvatore Satta, Diritto Processuale Civile (Padova: Cedam,
1981), 188–189 as described in Mauro Rubino-Sammartano, International
Arbitration: Law and Practice, 2nd edn (The Hague: Kluwer Law International,
2001), 657.
13 See, e.g., the UNCITRAL Model Law on International Commercial Arbitration

1985 Art. 28.

14 The nature of an award is discussed in section 16.2.
15 Article 27(4) of the UNCITRAL Rules 2010 is to similar effect, as is Art. 20.6 of

the ICDR Arbitration Rules 2009 Art. 9.1 of the IBA Rules on the Taking of
Evidence in International Arbitration 2010 and s. 1042(4) German Code of Civil
Procedure.
16 IBA Rules on the Taking of Evidence in International Arbitration 2010 Art. 1.5.
17 See LCIA Rules Art. 22.1(f).
18 See, e.g., UNCITRAL Rules 2010 Art. 27.3; ICC Rules 2012 Art. 25(5); ICDR

Rules Art. 19.3; LCIA Rules Art. 22.1(c)–(e); HKIAC Rules Art. 23.3; SCC Rules Art.
26(3); Swiss Rules 2012 Art. 24.3; IBA Rules on the Taking of Evidence in
International Arbitration 2010 Arts 3.9 and 4.9–10; CIETAC Rules Art. 41.1.
19 Judgment of 18 November 1960, Case Concerning the Arbitral Award Made by

the King of Spain on 23 December 1906 (Honduras v. Nicaragua), [1960] ICJ

Reports, 215–216.
20 Similarly, see the ICC Rules 2012 Art. 19; UNCITRAL Rules 2010 Art. 17.1;

LCIA Rules Art. 14.2; ICDR Rules Art. 16; HKIAC Rules Art. 14; SIAC Rules Art.
16.1; SCC Rules Art. 19(1); Swiss Rules 2012 Art. 15.1; ACICA Rules Art. 17.1.
21 Peter Hafter, ‘The Provisions on the Discovery of Internal Documents in the

IBA Rules of 1999’, in Global Reflections on International Law, Commerce and

Dispute Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen
et al. (Paris: ICC Publishing, 2005), 355; Hilmar Raeschke-Kessler, ‘The
Production of Documents in International Arbitration – A Commentary on Art. 3
of the New IBA Rules of Evidence’, Arbitration International 18, no. 4 (2002):
428.
22 See UNCITRAL Model Law Art. 18.
23 UNCITRAL Rules 2010 Art. 17(1).
24 Dadras International & Per-Am Construction Corporation v the Islamic Republic

of Iran & Tehran Redevelopment Company, in Albert Jan van den Berg (ed.),
Yearbook of Commercial Arbitration Volume XXII (The Hague: Kluwer Law
International, 1997), para. 61.
25 David D. Caron, Lee Caplan & Matti Pellonpää, The UNCITRAL Arbitration

Rules: A Commentary (Oxford: Oxford University Press, 2006), 621.

26 For example, the ICDR Guidelines for Arbitrators Concerning Exchanges of

Information (2008) para. 1(b) allow the parties to provide the tribunal with their
views on the appropriate level of information exchange but indicate that ‘the
tribunal retains final authority…’. Express agreement to these guidelines should
make it harder to mount a challenge to any tribunal determination.
27 This was discussed in sections 2.10.3, 2.12, 6.2.4 and 6.2.5.
28 Gary B. Born, International Commercial Arbitration (The Hague, Kluwer Law

International, 2009), 1009–1010.

29 See V.V. Veeder, ‘The Lawyer's Duty to Arbitrate in Good Faith’, Arbitration

International 18, no. 4 (2002): 431.

30 While not directly relevant to investment arbitration, Art. 36(2)(c) of the

Statute of the International Court of Justice allows parties to consent to the Court
determining the existence of any fact which, if established, would constitute a
breach of an international obligation. Art. 48 allows for arrangements as to the
taking of evidence. Art. 49 allows the Court to ask parties to produce evidence
although it cannot compel them to do so; Art. 50 allows the Court to establish
enquiries or call for expert opinions; Art. 43(5) allows for witnesses and experts
to be heard; while Art. 52 allows the Court to refuse untimely evidence.
31 Charles N. Brower, ‘The Anatomy of Fact-Finding before International
Tribunals: An Analysis and a Proposal Concerning the Evaluation of Evidence’, in
Fact-Finding before International Tribunals: Eleventh Sokol Colloquium, ed.
Richard B. Lillich (New York: Transnational Publishers, 1992), 150.
32 Durward V. Sandifer, Evidence before International Tribunals, rev. edn
(Charlottesville: University Press of Virginia, 1975), 458.
33 Ibid., 8.
34 For example, a case on expropriation might look at approaches to valuation by

the ICJ. See Factory at Chorzow (Germany v. Poland) (1928) PCD (ser. A) No. 17
(September 13), paras 127–143.
35 IAIGC v. DAII, Paris Court of Appeal, 23 October 1997, (1982) XXIII YB Com

Arb 644, 651.

36 A zero sum game is where any advantage to one party must come with an

equal disadvantage to the other.

37 Penny Martin et al., ‘2010 International Arbitration Survey: Choices in
International Arbitration',
<www.arbitrationonline.org/research/2010/index.html>, 2010.
38 See section 1.5.
39 Durward V. Sandifer, Evidence before International Tribunals, rev. edn
(Charlottesville: University Press of Virginia, 1975), 8.
40 Emmanuel Gaillard & John Savage (eds), Fouchard Gaillard Goldman on
International Commercial Arbitration (The Hague: Kluwer Law International,
1999), 690, para. 1260.
41 Yves Dezalay & Bryant G. Garth, Dealing in Virtue: International Commercial

Arbitration and the Construction of a Transnational Legal Order (Chicago: The

University of Chicago Press, 1996), 51–57.
42 See, e.g., Pieter Sanders (ed.), Comparative Arbitration Practice and Public

Policy in Arbitration, ICCA Congress Series No. 3 (The Hague: Kluwer

International, 1987), 19–22.
43 Howard Holtzmann, ‘Fact-Finding by the Iran-United States Claims Tribunal’,

in Fact-Finding before International Tribunals: Eleventh Sokol Colloquium, ed.

Richard B. Lillich (New York: Transnational Publishers, 1992), 101; Michael
Strauss, ‘The Practice of the Iran–United States Claims Tribunal in Receiving
Evidence from Parties and from Experts', Journal of International Arbitration 3,
no. 3 (1986): 58.
44 See further sections 1.2.13, 3.10 and 6.5.
45 First developed in 1983, revised in 1999 and again in 2010 and referred to

hereafter as the IBA Rules of Evidence 2010.

46 The revised Rules no longer refer to ‘commercial’ as this has become a term of

art in contradistinction to investment arbitration and the intent was to provide

rules for both forms of disputes.
47 ACICA Rules Art. 27.2.
48 ‘Techniques for Controlling Time and Costs in Arbitration: Report from the

ICC Commission on Arbitration’, ICC Publication No. 843, ICC International Court
of Arbitration Bulletin 18, no. 1 (2007): para. 54.
49 IBA Rules of Evidence 2010 Art. 1.2.
50 A recommended text is contained in para. 4 of the Foreword to the IBA Rules

of Evidence 2010.
51 IBA Rules of Evidence Review Sub-Committee, ‘Commentary on the Revised

Text of the 2010 IBA Rules on the Taking of Evidence in International

Arbitration’, 5, <www.ibanet.org/Document/Default.aspx?DocumentUid=D...>,
25 March 2011. The commentary on Art. 1 states that in the event of conflicts or
silence in the rules, the tribunal shall try to harmonise the rules or interpret
them based on their general principles, such as those set forth in the Preamble,
to the greatest extent possible.
52 Article 2.1; this brings para. 3 of the Preamble to the 1999 version of the IBA

Rules into a direct obligation. A similar sentiment is again expressed in the

Preamble to the 2010 version.
53 IBA Rules of Evidence 2010 Art. 2.1 and 2.2.
54 Article 9.7.
55 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 560.
56 Michael Bühler & Carroll Dorgan, ‘Witness Testimony Pursuant to the 1999
IBA Rules of Evidence in International Commercial Arbitration – Novel or Tested
Standards?’, Journal of International Arbitration 17, no. 1 (2000): 6.
57 Bernard Hanotiau, ‘Document Production in International Arbitration: A
Tentative Definition of “Best Practices’”, in ICC International Court of Arbitration
Bulletin, 2006 Special Supplement: Document Production in International
Arbitration (Paris: ICC Publishing, 2006), 114.
58 Jan Paulsson, ‘Cross-Enrichment of Public and Private Law Dispute Resolution

Mechanisms in the International Arena’, Journal of International Arbitration 9, no.

1 (1992): 63.
59 Simon Greenberg, Christopher Kee & J. Romesh Weeramantry, International

Commercial Arbitration: An Asia-Pacific Perspective (Cambridge: Cambridge

University Press, 2011), 312; footnote 34 gives details of a survey conducted by
the IBA Rules of Evidence Sub-Committee which noted 18% of participants
choosing the rules in most or all cases in their arbitration agreement and 31%
choosing them in some instances. Where the rules were not already selected by
the parties, these were included in terms of reference or subsequent directions
by 43% in most cases and by 42% in some cases. See also the reference to this
survey in Georg von Segesser, ‘The IBA Rules on the Taking of Evidence in
International Arbitration: Revised Version, Adopted by the International Bar
Association on 29 May 2010’, ASA Bulletin 28, no. 4 (2010): 736.
60 See Laurence Shore, ‘Three Evidentiary Problems in International Arbitration’,

SchiedsVZ (2004): 76–80 (although this is a reference to the 1999 Rules). For a
civilian critic see Klaus Lionnet, ‘Once Again: Is Discovery of Documents
Appropriate in International Arbitration?’, in Global Reflections on International
Law, Commerce and Dispute Resolution –Liber Amicorum in Honour of Robert
Briner, ed. Gerald Aksen & Robert Georg Briner (Paris: ICC Publishing, 2005),
499.
61 V.V. Veeder, ‘Evidence: The Practitioner in International Commercial
Arbitration’, International Law FORUM Du Droit International 1, no. 4 (1999):
229.
62 UNCITRAL, ‘UNCITRAL Notes on Organizing Arbitral Proceedings’,
<www.uncitral.org/pdf/english/texts/arbitration/arb...>, 1996. See further
section 6.3.1.
63 UNCITRAL, ‘Draft Guidelines for Preparatory Conferences in Arbitral
Proceedings’ <http://daccess-dds-ny.un.org/doc/UNDOC/GEN/V94/219...>,
1994.
64 ICC, ‘Techniques for Controlling Time and Costs in Arbitration’,
<www.iccwbo.org/uploaded-Files/TimeCost_E.pdf>, 2007. See further section
6.5.2.
65 ICDR, ‘ICDR Guidelines for Arbitrators Concerning Exchanges of Information’,

<www.adr.org/si.asp?id=5288>, May 2008.

66 CPR, ‘CPR Protocol on Disclosure of Documents and Presentation of Witnesses

in Commercial Arbitration’,
<http://cpradr.Org/Resources/ALLCPRArticles/tabid/2...>, 2009.
67 Doug Jones, ‘Party Appointed Expert Witnesses in International Arbitration: A

Protocol at Last’, Arbitration International 24, no. 1 (2008):141.

68 In this book, document presentation is used to describe an obligation on a

party to tender the material on which it intends to rely. Document production is

used to refer to the right to call for documents from the opposing party
(described as discovery in common law systems).
69 Shabtai Rosenne & Yaël Ronen, The Law and Practice of the International

Court, 1920–2005, 4th edn (Leiden: Martinus Nijhoff Publishers, 2006), 1040–
1042; Corfu Channel Case (United Kingdom v Alabania) [1949] ICJ Rep 4, 17–18;
Awards of 27 October 1989 and 30 June 1990, Albert Jan van den Berg (ed.),
Yearbook of Commercial Arbitration Volume XIX (The Hague: Kluwer Law
International, 1994), 24, para. 3; Antoine Biloune and Marine Drive Complex Ltd v.
Ghana Investment Centre and the Government of Ghana (1990) 95 ILR 184.
70 Sigvard Jarvin, ‘The Sources and Limits of the Arbitrator's Powers’, Arbitration

International 2, no. 2 (1986): 151.

71 Here there is a question as to whether the tribunal itself can identify a legal

basis even different to that presented by a party under the principle iura novit
curia. This is discussed further in section 13.19.
72 Much confusion occurs because different terms are used to display differing

concepts or are arguably used interchangeably by different authors and judges.

For example, Alan Redfern speaks of the onus of proof on the party making an
allegation, with the burden of proof being the evidential burden shifting from
one party to another: Alan Redfern et al., ‘The Standards and Burden of Proof in
International Arbitration’, Arbitration International 10, no. 3 (1994): 319.
Andreas Reiner discusses the burden of allegation in the same article. See Alan
Redfern et al., ‘The Standards and Burden of Proof in International Arbitration’,
Arbitration International 10, no. 3 (1994): 328. Reiner distinguishes between
notions of legal burden and evidential burden at 329. Kurkela and Snellman
suggest the burden of proof can shift when the respondent is the only one that
can be expected to have the relevant evidence: Matti Kurkela & Hannes Snellman,
Due Process in International Commercial Arbitration (New York: Oceana
Publications, 2005), 41. Sharpe speaks of the ‘burden of production or onus
proponendi’ and also speaks of the burden shifting: Jeremy K. Sharpe, ‘Drawing
Adverse Inferences from the Non-production of Evidence’, Arbitration
International 22, no. 4 (2006): 552. Von Mehren speaks of the burden of going
forward and the burden of persuasion: Robert B. von Mehren, ‘Burden of Proof in
International Arbitration’, in Planning Efficient Arbitration Proceedings: The Law
Applicable in International Arbitration, ICCA Congress Series No. 7 (Vienna 1994),
ed. Albert Jan van den Berg (The Hague: Kluwer Law International, 1996), 124.
Here the phrases ‘burden of going forward’ and ‘burden of persuasion’ are used
as synonyms for onus of proof.
73 Some tribunals also unfortunately speak of the burden of proof as shifting

once the claimant has provided sufficient evidence. See, e.g., Reza Said Malek v
the Government of the Islamic Republic of Iran, in Yearbook of Commercial
Arbitration Volume XVIII, ed. Albert Jan van den Berg (The Hague: Kluwer Law
International, 1993), 289. See also Edgar Protiva & Eric Protiva v. the Government
of the Islamic Republic of Iran, in Yearbook Commercial Arbitration Volume XXI,
ed. Albert Jan van den Berg (The Hague: Kluwer Law International, 1996), para.
61. Another troubling observation in a claim for damages for non-performance
was in ICC Case No. 1434, Journal du Droit International (1976): 982, where the
tribunal opined that the claimant ‘has the burden to prove the existence of the
duty of the other party to perform and its extent, while the defendant must prove
that such a duty has been discharged’. It must surely be the case that the
claimant must attest to non-performance, although the respondent would
succeed if it shows that this is not the case. The problem was perhaps alleviated
because the tribunal also commented that the parties have a duty to cooperate in
good faith in the taking of evidence, hence showing that it would not resolve the
matter via technical burdens of proof.
74 Some speak of burden of proceeding instead of onus but this is also likely to

confuse.
75 See also HKIAC Rules Art. 23.1; Swiss Rules 2012 Art. 24.1; CIETAC Rules Art.

39.1; ICDR Rules Art. 19.1; ACICA Rules Art. 27.1.

76 Poudret and Besson support the view that burden of proof and presumptions

which reverse the burden of proof in special cases are determined under the lex
causae: Jean-François Poudret & Sebastien Besson, Comparative Law of
International Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 550.
Arbitral awards adopting this view and cited in Poudret and Besson include ICC
Case No. 5294 of 22 February 1988, Albert Jan van den Berg (ed.), Yearbook of
Commercial Arbitration Volume XIV, (The Hague: Kluwer Law International,
1989), 145. Again, it is undesirable to speak of burdens shifting as opposed to
onus.
77 Convention on the Law Applicable to Contractual Obligations (Rome
Convention) 1980 Art. 14(1).
78 See Alan Redfern et al., ‘The Standards and Burden of Proof in International

Arbitration’, Arbitration International 10, no. 3 (1994): 331–332.

79 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1858.

80 Ibid., 1858.
81 Kurkela & Hannes Snellman, Due Process in International Commercial
Arbitration (New York: Oceana Publications, 2005), 41.
82 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 561. This concept is also known as L'intime conviction dujuge, il libero
convincimento del giudice, die richterliche uberzeugeng.
83 See the conflicting survey views in Michael J. Bond, ‘The Standard of Proof in

International Commercial Arbitration’, Arbitration 77, no. 3 (2011) 304.

84 See Alan Redfern et al., ‘The Standards and Burden of Proof in International

Arbitration’, Arbitration International 10, no. 3 (1994): 335.

85 Ibid., 326.
86 George M. von Mehren & Claudia T. Salomon, ‘Submitting Evidence in an

International Arbitration: The Common Lawyer's Guide’, Journal of International

Arbitration 20, no. 3 (2003): 291.
87 Partial Award Prisoners Of War –Eritrea's Claim 17, decision of 1 July 2003,

Reports of International Arbitral Awards, Volume XXVI, 41.

88 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 561.
89 David D. Caron, Lee Caplan & Matti Pellonpää, The UNCITRAL Arbitration

Rules: A Commentary, (Oxford: Oxford University Press, 2006), 572.

90 Mojtaba Kazazi, Burden of Proof and Related Issues: A Study on Evidence before

International Tribunals (The Hague: Kluwer Law International, 1996), 377.

91 R.D. Friedman, ‘Economic Analysis of Evidentiary Law: An Underused Tool, an

Underplowed Field’, Cardozo Law Review 19 (1998): 1535–1536.

92 Robert B. von Mehren, ‘Burden of Proof in International Arbitration’, in
Planning Efficient Arbitration Proceedings: The Law Applicable in International
Arbitration, ICCA Congress Series No. 7 (Vienna 1994), ed. Albert Jan van den
Berg (The Hague: Kluwer Law International, 1996), 128.
93 Where fraud is concerned ICC Award 6401 considered that this ‘must be

proved by clear and convincing evidence’: ICC Case No. 6401, Westinghouse
International Projects Company v. National Power Corporation, Mealey's
International Arbitration Report 7’, no. 1 (1992): 17–18.
94 Oil Field of Texas, Inc. v. the Government of the Islamic Republic of Iran &

National Iranian Oil Company, in Albert Jan van den Berg (ed.), Yearbook of
Commercial Arbitration Volume XII (The Hague: Kluwer Law International, 1987),
288.
95 Dadras International & Per-Am Construction Corporation v the Islamic Republic

of Iran & Tehran Redevelopment Company, in Albert Jan van den Berg (ed.),
Yearbook of Commercial Arbitration Volume XXII (The Hague: Kluwer Law
International, 1997), para. 124. See also Final Award in Case No. 6497 of 1994, in
Albert Jan van den Berg (ed.), Yearbook of Commercial Arbitration Volume XXIVa
(The Hague: Kluwer Law International, 1999), 73.
96 K. Mills, ‘Corruption and Other Illegality in the Formation and Performance of

Contracts and in the Conduct of Arbitration Relating Thereto’, in International

Commercial Arbitration: Important Contemporary Questions, ICCA Congress
Series No. 11, ed. Albert Jan van den Berg (The Hague: Kluwer Law International,
2003), 295.
97 Final Award in ICC Case No. 8891, Journal de Droit International (2000): 1076,

and cited in Anne Véronique Schlaepfer & Philippe Bartsch, ‘A Few Reflections on
the Assessment of Evidence by International Arbitrators’, International Business
Law Journal no. 3 (2010): 223.
98 David D. Caron, Lee Caplan & Matti Pellonpää, The UNCITRAL Arbitration

Rules: A Commentary (Oxford: Oxford University Press, 2006), 570.

99 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1902, 1858.

100 Howard M. Holtzmann, ‘Can the Plausibility Standard of Proof Developed in

Mass Claims Arbitrations Also Be Useful in Commercial Cases’, in Global

Reflections on International Law, Commerce and Dispute Resolution: Liber
Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al. (Paris: ICC
Publishing, 2005), 378.
101 George M. von Mehren & Claudia T. Salomon, ‘Submitting Evidence in an

International Arbitration: The Common Lawyer's Guide’, Journal of International

Arbitration 20, no. 3 (2003): 291.
102 Alan Redfern et al., ‘The Standards and Burden of Proof in International

Arbitration’, Arbitration International 10, no. 3 (1994): 337.

103 See section 13.14 in relation to the general rights and duties of an amiable

compositeur. At this stage it should be noted that there are differences in view as
to whether such a person can ignore legal rules and decide purely on questions
of fairness or whether legal principles should first be considered and then
mitigated where appropriate.
104 Alan Redfern et al., ‘The Standards and Burden of Proof in International

Arbitration’, Arbitration International 10, no. 3 (1994): 324–325.

105 This covers situations where the victim cannot directly attest to how they
were injured but the circumstances make it sufficient to conclude that they could
not have been injured without negligence by the person responsible.
106 ICC Case No. 6653 of 26 March 1993 (Steel bars case), cited in Mauro Rubino-

Sammartano, International Arbitration: Law and Practice, 2nd edn (The Hague:
Kluwer Law International, 2001), 702.
107 Mojtaba Kazazi, Burden of Proof and Related Issues: A Study on Evidence before

International Tribunals (The Hague: Kluwer Law International, 1996), 235.

108 Durward V. Sandifer, Evidence before International Tribunals, rev. edn
(Charlottesville: University Press of Virginia, 1975), 125.
109 See, e.g., Case T-13/99, Pfizer Animal Health SA v. Council of European Union

[2002] ECR II-3307 paras 168–169; Hauptzollamt München-Mitte v. Technische

Universitäs München [1991] ECR 1–5469.
110 See, e.g., Edgar Protiva & Eric Protiva v. the Government of the Islamic Republic

of Iran, in Albert Jan van den Berg (ed.), Yearbook of Commercial Arbitration
Volume XXI (The Hague: Kluwer Law International, 1996), para. 68. In that case,
the tribunal concluded that ‘the Respondent has not introduced any evidence
adequate to rebut the substance of (Claimant's witnesses) letters’.
111 Howard M. Holtzmann, ‘Procedural Aspects: Balancing the Need for Certainty

and Flexibility in International Arbitration Procedures’, in International

Arbitration in the Twenty-First Century: Towards ‘Judicialization’ and Uniformity?,
ed. Richard B. Lillich & Charles N. Brower (New York: Transnational Publishers,
1994), 21.
112 Bin Cheng, General Principles of Law as Applied by International Courts and

Tribunals, 3rd edn (Cambridge: Cambridge University Press, 2006), 34.

113 Corfu Channel case, United Kingdom of Great Britain v. People's Republic of

Albania (Merits), [1949] ICJ Rep 4, 18.

114 Charles N. Brower, ‘Evidence before International Tribunals: The Need for

Some Standard Rules’, International Lawyer 28, no. 1 (1994): 54.

115 ICC Award No. 8694 (1996) American Company v. Belgian Company and

observations by Yves Derains, Journal du Droit International (1997): 1056. For

another example of adverse inferences in an ICC award, see Yves Derains, ‘La
pratique de l'administration de la preuve dans l'arbitrage commercial
international’, Revue de l'Arbitrage (2004): 792. Adverse inferences were drawn
by a majority of the tribunal in ICSID Case No. AB/AF/99/1, Award and
Dissenting Opinion of 16 December 2002, Marvin Roy Feldman Karpa v. United
Mexican States, 7 ICSID Reports 407, 412–413, in relation to proof of a breach of
national treatment where the applicants contended that Mexican companies
were given rebates of excise taxes in a discriminatory way over the applicant.
Other examples of adverse inferences include INA Corporation v. the Government
of the Islamic Republic of Iran, in Albert Jan van den Berg (ed.), Yearbook of
Commercial Arbitration Volume XI (The Hague: Kluwer Law International, 1986),
312–324; Computer Sciences Corp. v. the Government of the Islamic Republic of
Iran, Award No. 221-65-1, 16 April 1986, 10 Iran–United States Claims Tribunal
Reports 269, 302 (1987); Sedco, Inc. v. Islamic Republic of Iran, Award No. 309–
129-3, 7 July 1987, 15 Iran–United States Claims Tribunal Reports 23, 55 (1987).
See also Ultrasystems Inc. v. Islamic Republic of Iran, Concurring Opinion of R
Mosk in Award No. 27-84-3, 4 March 1983, 2 Iran-United States Claims Tribunal
Reports 114, 115 (1983); Edgar Protiva & Eric Protiva v. the Government of the
Islamic Republic of Iran, in Albert Jan van den Berg (ed.), Yearbook Commercial
Arbitration Volume XXI (The Hague: Kluwer Law International 1996), para. 68.
116 Judgment of 28 March 2007, Swiss Federal Supreme Court ASA Bulletin 25,

no. 3 (2007): 610.

117 Ali Z. Marossi, ‘The Necessity for Discovery of Evidence in the Fact-Finding

Process of International Tribunals’, Journal of International Arbitration 26, no. 4

(2009): 529.
118 Ashford considers this permissible. See Peter Ashford, ‘Document Discovery

and International Commercial Arbitration’, American Review of International

Arbitration 17, (2006): 101.
119 Mojtaba Kazazi, Burden of Proof and Related Issues: A Study on Evidence before

International Tribunals (The Hague: Kluwer Law International, 1996), 321.

120 Pierre-Yves Tschanz, ‘Advocacy in International Commercial Arbitration:

Switzerland’, in The Art of Advocacy in International Arbitration, ed. R. Doak

Bishop (New York: Juris Publishing, 2004), 231.
121 Thomas H. Webster, ‘Obtaining Documents from Adverse Parties in
International Arbitration', Arbitration International 17, no. 1 (2001): 51.
122 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1902.

123 Arguments as to waiver would not help as they do not support a logical

inference that the document is in fact adverse.

124 Peter Ashford, ‘Document Discovery and International Commercial
Arbitration’, American Review of International Arbitration 17 (2006): 101.
125 John Ragosta, ‘Unmasking the WTO – Access to the DSB System: Can the WTO

DSB Live up to the Moniker World Trade Court’, Law and Policy in International
Business 31, no. 3 (2000): 762.
126 Arthur J Fritz & Co. v. Sherkate Tavonie Sherkathaye Sakhtemanie (Co-
operative Society of Construction Companies) 22 Iran–US Claims Tribunal Reports
170 (1989), 180.
127 Case Concerning the Barcelona Traction, Light and Power Company Limited

(Belgium v. Spain) ICJ Reports (1970), 215, para. 97.

128 In the WTO context, the Appellate Body, where it thought it might have

drawn an adverse inference if confronting the same issue as the panel,

nevertheless concluded that the panel did not err in law in refusing to do so.
Canada–Measures Affecting the Export of Civilian Aircraft Canada-Aircraft WTO
Doc WT/DS70/AB/R, (1999), Report of the Appellate Body, paras 199 and 205.
129 David D. Caron, Lee Caplan & Matti Pellonpää, The UNCITRAL Arbitration

Rules: A Commentary, (Oxford: Oxford University Press, 2006), 578–579. For a

contrary approach see Dongwoo Mann and Hummel Co. Ltd v. Mann and Hummel
GmbH [2008] SGHC 67, para. 70.
130 Edgar Protiva & Eric Protiva v. the Government of the Islamic Republic of Iran,

in Albert Jan van den Berg (ed.), Yearbook of Commercial Arbitration Volume XXI
(The Hague: Kluwer Law International 1996), para. 68.
131 See, e.g., UNCITRAL Rules 2010 Art. 27.3; ICC Rules 2012 Art. 25(5); ICDR

Rules Art. 19.3; LCIA Rules Art. 22.1(c)-(e); HKIAC Rules Art. 23.3; SCC Rules Art.
26(3); Swiss Rules 2012 Art. 24.3; IBA Rules on the Taking of Evidence in
International Arbitration 2010 Arts 3.9 and 4.9–10; CIETAC Rules Art. 41.
132 Military and Paramilitary Activities in and against Nicaragua (Nicaragua v.

USA) (Merits) judgment, [1986] ICJ Reports, 40–41.

133 Martin Hunter, ‘The Procedural Powers of Arbitrators under the English
1996 Act’, Arbitration International 13, no. 4 (1997): 352.
134 Eric Bardier de la Serre & Anne-Lise Sibony, ‘Expert Evidence before the EC

Courts’, Common Market Law Review 45, no. 4 (2008): 959.

135 Mauro Rubino-Sammartano, International Arbitration: Law and Practice, 2nd

edn (The Hague: Kluwer Law International, 2001), 704 citing Owen v. Nicholl
[1948] 1 All ER 707.
136 Charles N. Brower, ‘The Anatomy of Fact-Finding before International
Tribunals: An Analysis and a Proposal Concerning the Evaluation of Evidence’, in
Fact-Finding before International Tribunals: Eleventh Sokol Colloquium, ed.
Richard B. Lillich (New York: Transnational Publishers, 1992), 150–151; Behring
Int'l v. Islamic Republic of Iran (1991) 27 Iran-United States Claims Tribunal
Reports 218, 234. See also Edgar Protiva & Eric Protiva v. the Government of the
Islamic Republic of Iran, in Albert Jan van den Berg (ed.), Yearbook of Commercial
Arbitration Volume XXI (The Hague: Kluwer Law International 1996), para. 72.
137 Peter Hafter, ‘The Provisions on the Discovery of Internal Documents in the

IBA Rules of 1999’, in Global Reflections on International Law, Commerce and

Dispute Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen
et al. (Paris: ICC Publishing, 2005), 361.
138 The example also shows why document production is such an important

element in legal adjudicatory design and why there is such a need to find a
harmonised approach to this element in international arbitration, given the
disparity of approaches between legal families at the domestic level. Document
production is discussed further in section 11.6 below.
139 Avco Corp v. Iran Aircraft Industries et al., in Albert Jan van den Berg (ed.),

Yearbook of Commercial Arbitration Volume XIV (The Hague: Kluwer Law

International 1989), 377–382.
140 Ibid., 381.
141 Ibid., 379. See also Charles N. Brower, ‘Evidence before International
Tribunals: The Need for Some Standard Rules’, International Lawyer 28, no. 1
(1994): 53–54.
142 Iran Aircraft Industries et al. v. Avco Corporation, in Albert Jan van den Berg

(ed.), Yearbook of Commercial Arbitration Volume XVIII (The Hague: Kluwer Law
International 1993), 596–606.
143 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 563.

144 Martin Hunter, ‘The Procedural Powers of Arbitrators under the English

1996 Act’, Arbitration International 13, no. 4 (1997): 350.

145 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1902. This approach is commonly utilised in WTO dispute

settlement.
146 Emmanuel Gaillard & John Savage (eds), Fouchard Gaillard Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 692.
147 Ibid., 690. See further section 13.19 on this issue.
148 Payment in Gold of Brazilian Federal Loans Contracted in France (France v.

Brazil), 1929 PCIJ. (ser. A) No. 21 (12 July), 124, para. 81.
149 Vienna Convention on the Law of Treaties, United Nations, Treaty Series, vol.

1155, 331, Art. 32.

150 For example, Marc Blessing advocates a determination of what is subjectively
fair and objectively reasonable in interpreting arbitration agreements: Dr Marc
Blessing, ‘The Law Applicable to the Arbitration Clause’, in Improving the
Efficiency of Arbitration and Awards: 40 Years of Application of the New York
Convention, ICCA Congress Series No. 9, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 1999), 171; Marc Blessing ‘Choice of Substantive Law
in International Arbitration’, Journal of International Arbitration 14, no. 2 (1997):
42.
151 Klaus Peter Berger, ‘Evidentiary Privileges under the Revised IBA Rules on

the Taking of Evidence in International Arbitration’, International Arbitration

Law Review 13, no. 5 (2010): 174.
152 See, e.g., Art. 10 UNCITRAL Model Law on International Commercial
Conciliation; s. 10 DIS Mediation Rules; SM-12 ICDR International Mediation
Rules; s. 6, CEDR Code of Conduct for Mediators and Other Third Party Neutrals.
See also Mobil Oil Iran v. Islamic Republic of Iran Iran–US Claims Tribunal Reports
16, 55. See also ICC Award 6653 of 1993, Journal du Droit International (1993):
1040 as cited in Jason Fry, ‘Without Prejudice and Confidential Communications
in International Arbitration’, International Arbitration Review 1, no. 6 (1998):
209, 212.
153 See section 10.17 below dealing with privilege, including settlement
discussions.
154 This should be resolved by the tribunal and not by a court. In the
Singaporean case of Doshion Ltd v. Sembawang Engineers and Constructors Pte
Ltd [2011] SGHC 46, an application to the court on the basis that the tribunal was
functus officio was rightly rejected. Such an assertion would also be contrary to
the notion that a tribunal can render a settlement agreement as an award.
155 Klaus Peter Berger, ‘Evidentiary Privileges under the Revised IBA Rules on

the Taking of Evidence in International Arbitration’, International Arbitration

Law Review 13, no. 5 (2010): 171, 175.
156 See Esso Australia Resources Ltd v. Plowman [1995] 183 CLR 10. For
variations in the US, Sweden and England, see the discussion in Peter Ashford,
‘Document Discovery and International Commercial Arbitration', American
Review of International Arbitration 17, (2006): 128.
157 The ICC Court's Internal Rules also indicate that the Court's work is also

confidential: ICC Rules of Arbitration 2012, Appendix II, Art. 1.

158 One could question whether this only deals with materials or also includes

information.
159 Richard Garnett et al., A Practical Guide to International Commercial
Arbitration (New York: Oceana Publications Inc., 2000), 14.
160 Res judicata is discussed in section 8.10.
161 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2884.

162 ILA International Commercial Arbitration Committee, ‘Interim Report on Res

Judicata and Arbitration’, given at the ILA Seventy-First Conference (Berlin,

2004), 14, <www.ila-hq.org/download.cfm/docid/446043C4-9770-43...>, 25
March 2011; ILA International Commercial Arbitration Committee, ‘Final Report
on Lis Pendens and Arbitration’, adopted at the ILA Seventy-Second Conference
(Toronto, 2006), 36, para. 56, <www.ila-
hq.org/download.cfm/docid/C5443B2B-406F-4A...>, 25 March 2011.
163 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law
International, 2009), 2886–2887.
164 For a detailed analysis of the preclusive effects of international arbitral

awards see Gary B. Born, International Commercial Arbitration (The Hague:

Kluwer Law International, 2009), 2887-2970.
165 ICSID Arbitration Rules Art. 42(3); UNCITRAL Rules 2010 Art. 30.1(b); Swiss

Rules 2012 Art. 28; LCIA Rules Art. 15.8; HKIAC Rules Art. 26; CIETAC Rules Art.
14.4.
166 For a requirement as to rehearing, see Cia de Navegacion Omsil v. Hugo Neu

Corp 359 F.Supp. 898 (SDNY 1973).

167 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 561.
168 UNCITRAL Rules 2010 Art. 27(4); HKIAC Rules Art. 23.10; Swiss Rules 2012

Art. 24.2; SCC Rules Art. 26(1); IBA Rules on the Taking of Evidence in
International Arbitration 2010 Art. 9.1.
169 Panel Report, European Communities – Anti-dumping duties on Imports of

Cotton-Type Bed Linen from India, WT/DS141/R (30 October 2000), para. 6.33.
Similar problems arise when an arbitrator has previously undertaken a
mediation function: Jacob Rosoff, ‘Hybrid Efficiency in Arbitration: Waiving
Potential Conflicts for Dual Role Arbitrators in Med-Arb and Arb-Med
Proceedings’, Journal of International Arbitration 26, no. 1 (2009): 97.
170 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 387.

171 Charles N. Brower, ‘Evidence before International Tribunals: The Need for

Some Standard Rules’, International Lawyer 28, no. 1 (1994): 48.

172 George M. von Mehren & Claudia T. Salomon, ‘Submitting Evidence in an

International Arbitration: The Common Lawyer's Guide’, Journal of International

Arbitration 20, no. 3 (2003): 290.
173 Corfu Channel case, United Kingdom of Great Britain v People's Republic of

Albania (Merits), ICJ Reports (1949), 4, 18.

174 Ibid., 4, 59.
175 See American Steamship Company v. Thai Transportation Enterprise, in Pieter

Sanders (ed.), Yearbook of Commercial Arbitration Volume VIII (The Hague:

Kluwer Law International, 1983), 168 (allowing hearsay evidence).
176 US Federal Rules of Evidence (2011), Rule 801(c).
177 For admissibility of hearsay evidence in England see Civil Evidence Act 1995

(UK) s. 1(1).
178 Military and Paramilitary Activities in and Against Nicaragua (Nicaragua v.

USA) (Merits) [1986] ICJ Reports 14, 42 para. 68.

179 Shabtai Rosenne, The Law and Practice of the International Court, 1920–1996,

3rd edn (Leiden: Martinus Nijhoff Publishers, 1997), 1090. An American court
refused to interfere with an arbitral award based on hearsay evidence in
Petroleum Separating Company v. Inter-American Refining Corporation 296 F. 2d
124 (2d Cir 1961).
180 Charles N. Brower, ‘Evidence before International Tribunals: The Need for

Some Standard Rules’, International Lawyer 28, no. 1 (1994): 52. Hearsay
evidence was also admitted in American Steamship Company v. Thai
Transportation Enterprise, in Pieter Sanders (ed.), Yearbook of Commercial
Arbitration Volume VIII (The Hague: Kluwer Law International, 1983), 168.
181 Geroge H. Aldrich, The Jurisprudence of the Iran–United States Claims
Tribunal: An Analysis of the Decisions of the Tribunal (Oxford: Oxford University
Press, 1996), 352.
182 Military and Paramilitary Activities in and against Nicaragua (Nicaragua v.

USA) (Merits) [1986] ICJ Reports 14, 40 para. 63.

183 A related relevant factor is the vested interest of the speaker in terms of the

outcome of the dispute: Michael Bühler & Carroll Dorgan, ‘Witness Testimony
Pursuant to the 1999 IBA Rules of Evidence in International Commercial
Arbitration – Novel or Tested Standards?’, Journal of International Arbitration 17,
no. 1 (2000): 28.
184 Civil Evidence Act 1995 (UK) s. 5.
185 S.I. Strong & James J. Dries, ‘Witness Statements under the IBA Rules of

Evidence: What to Do about Hearsay?’, Arbitration International 21, no. 3 (2005):

307.
186 Ibid., 315–316. Other court assistance may be resorted to.
187 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 559.
188 See Marc Blessing, ‘The Law Applicable to the Arbitration Clause’, in
Improving the Efficiency of Arbitration and Awards: 40 Years of Application of the
New York Convention, ICCA Congress Series No. 9 (1998 Paris), ed. Albert Jan van
den Berg (The Hague: Kluwer Law International, 1999).
189 See Case No. 122, Judgment of 12 July 1991, Southern District of New York,

Oriental Commercial & Shipping Co Ltd., et al, v. Rosseel, NV, in Albert Jan van den
Berg (ed.), Yearbook Commercial Arbitration Volume XVII (The Hague: Kluwer
Law International, 1992), 696–704 (applying the parol evidence rule to
documents of parties that led to consent orders in domestic court proceedings).
190 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 559.
191 Ibid., 560.
192 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1860; Pierre-Yves Tschanz, ‘Advocacy in International

Commercial Arbitration: Switzerland’, in The Art of Advocacy in International
Arbitration, ed. R. Doak Bishop (New York: Juris Publishing, 2004), 218.
193 W. Michael Reisman & Eric E. Freedman, ‘The Plaintiff’s Dilemma: Illegally

Obtained Evidence and Admissibility in International Adjudication’, American

Journal of International Law 76 (1982): 737; Mojtaba Kazazi, Burden of Proof and
Related Issues: A Study on Evidence before International Tribunals (The Hague:
Kluwer Law International, 1996), 208.
194 Corfu Channel case, United Kingdom of Great Britain v People's Republic of

Albania (Merits), ICJ Reports (1949) 4, 34–36.

195 See, e.g., Final award (under NAFTA Chapter 11 and the UNCITRAL
Arbitration Rules) of 3 August 2005, Methanex Corporation v. United States of
America, Part II, Chp 1, 26, para. 54.
196 IBA Rules on the Taking of Evidence in International Arbitration 2010 Art.

9.2(e).
197 Esso Australia Resources Ltd v Plowman [1995] 183 CLR 10.
198 Charles N. Brower & Jeremy K. Sharpe, ‘Determining the Extent of Discovery
and Dealing with Requests for Discovery: Perspectives from the Common Law’,
in Leading Arbitrators' Guide to International Arbitration, ed. Lawrence W.
Newman & Richard D. Hill, 2nd edn (New York: Juris Publishing, Inc., 2008), 328.
199 D v. National Society for the Prevention of Cruelty to Children [1978] AC 171,

218 per Lord Diplock.

200 I am indebted to Tony Canham for this observation.
201 Science Research Council v. Nasse [1980] AC 1028, 1065.
202 Emmanuel Gaillard & John Savage (eds), Fouchard Gaillard Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 1265; Jean-François Poudret & Sébastien Besson, Comparative Law of
International Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 557.
203 Resort should also be made to the ILA International Commercial Arbitration

Committee report on Confidentiality in International Commercial Arbitration.

Available at <http://port.academia.edu/MunirManiruzzaman/Talks/4...>.
204 Martin Hunter & Gregory Travaini, ‘Electronically Stored Information and

Privilege in International Arbitration’, in Liber Amicorum Bernardo Cremades, ed.

Miguel Á. Fernández-Ballesteros & David Arias (Madrid: La Ley, 2010), 615.
205 Richard M. Mosk & Tom Ginsburg, ‘Evidentiary Privileges in International

Arbitrations’, International and Comparative Law Quarterly 50, no. 2 (2001): 345.
206 Fabian von Schlabrendorff & Audley Sheppard, ‘Conflict of Legal Privileges in

International Arbitration: An Attempt to Find a Holistic Solution’, in Global

Reflections on International Law, Commerce and Dispute Resolution: Liber
Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al. (Paris: ICC
Publishing, 2005), 745.
207 Charles N. Brower & Jeremy K. Sharpe, ‘Determining the Extent of Discovery

and Dealing with Requests for Discovery: Perspectives from the Common Law’,
in Leading Arbitrators' Guide to International Arbitration, ed. Lawrence W
Newman & Richard D. Hill, 2nd edn (New York: Juris Publishing, Inc., 2008), 378–
381.
208 Georgios Petrochilos, Procedural Law in International Arbitration (Oxford:

Oxford University Press, 2004), 221, n. 244.

209 IBA Working Party, ‘Commentary on the New IBA Rules of Evidence in

International Commercial Arbitration’, Business Law International 14, no. 2

(2000): 33.
210 Olaf Meyer, ‘Time to Take a Closer Look: Privilege in International
Arbitration’, Journal of International Arbitration 24, no. 4 (2007): 367–370.
211 Ibid., 367.
212 Three Rivers District Council & Ordv. Governor & Company of the Bank of

England (No 10) [2004] 3 WLR 1274 (HL), para. 26. Schlabrendorff and Sheppard
also see legal privileges as having both procedural and substantive characters:
Fabian von Schlabrendorff & Audley Sheppard, ‘Conflict of Legal Privileges in
International Arbitration: An Attempt to Find a Holistic Solution’, in Global
Reflections on International Law, Commerce and Dispute Resolution: Liber
Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al. (Paris: ICC
Publishing, 2005), 764.
213 See, e.g., UNCITRAL Model Law Art. 28.
214 Michelle Sindler & Tina Wustemann, ‘Privilege Across Borders in Arbitration:

Multi-jurisdictional Nightmare or a Storm in a Teacup?’, ASA Bulletin 23, no. 4

(2005): 619.
215 Georgios Petrochilos, Procedural Law in International Arbitration (Oxford:
Oxford University Press, 2004), 221.
216 Klaus Peter Berger, ‘Evidentiary Privileges under the Revised IBA Rules on

the Taking of Evidence in International Arbitration’, International Arbitration

Law Review 13, no. 5 (2010): 173.
217 Olaf Meyer, ‘Time to Take a Closer Look: Privilege in International
Arbitration’, Journal of International Arbitration 24, no. 4 (2007): 370.
218 Fabian von Schlabrendorff & Audley Sheppard, ‘Conflict of Legal Privileges in

International Arbitration: An Attempt to Find a Holistic Solution’, in Global

Reflections on International Law, Commerce and Dispute Resolution: Liber
Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al. (Paris: ICC
Publishing, 2005), 761.
219 Convention on the Taking of Evidence Abroad in Civil or Commercial Matters,

opened for signature 18 March 1970, Art. 11.

220 EC Regulation on Judicial Co-operation (Council Regulation (EC) No.
1206/2001 of 28 May 2001 on Cooperation between the Courts of the Member
States in the Taking of Evidence in Civil or Commercial Matters [2001] OJ L 174,
Art. 14, 17.
221 Olaf Meyer, ‘Time to Take a Closer Look: Privilege in International
Arbitration’, Journal of International Arbitration 24, no. 4 (2007): 270.
222 Convention on the Taking of Evidence Abroad in Civil or Commercial Matters,

opened for signature 18 March 1970, Art. 11; EC Regulation on Judicial Co-
operation (Council Regulation (EC) No. 1206/2001 of 28 May 2001 on Cooperation
between the Courts of the Member States in the Taking of Evidence in Civil or
Commercial Matters [2001] OJ L 174, Art. 14; Inter-American Convention on the
Taking of Evidence Abroad, opened for signature 30 January 1975, 1438 UNTS
385, Art. 12 (entered into force 16 January 1976).
223 Gabrielle Kaufmann-Kohler & Philippe Bärtsch, ‘Discovery in International

Arbitration: How Much is Too Much?’, SchiedsVZ 2 (2004): 19.

224 See, e.g., US Restatement (Second) Conflict of Laws (American Law Institute,

1971), § 139.
225 Meyer also strongly advocates the consideration of existing international

standards when making these determinations: Olaf Meyer, ‘Time to Take a Closer
Look: Privilege in International Arbitration’, Journal of International Arbitration
24, no. 4 (2007). In appropriate cases such standards might be drawn from the
laws and practices of UN covenants and conventions on human rights, the
European Convention on Human Rights and the Code of Conduct for Lawyers in
the European Union.
226 Fabian von Schlabrendorff & Audley Sheppard, ‘Conflict of Legal Privileges in

International Arbitration: An Attempt to Find a Holistic Solution’, in Global

Reflections on International Law, Commerce and Dispute Resolution: Liber
Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al. (Paris: ICC
Publishing, 2005), 768-769. A similar view is presented by Klaus Peter Berger,
‘Evidentiary Privileges under the Revised IBA Rules on the Taking of Evidence in
International Arbitration’, International Arbitration Law Review 13, no. 5 (2010):
177.
227 International Institute for Conflict Prevention and Resolution, ‘CPR Protocol

on Disclosure of Documents and Presentation of Witnesses in Commercial

Arbitration’ (New York: International Institute for Conflict Prevention and
Resolution), para. 1(b); International Centre for Dispute Resolution, ICDR
Guidelines for Arbitrators Concerning Exchanges of Information’, para. 7.
228 Fabian von Schlabrendorff & Audley Sheppard, ‘Conflict of Legal Privileges in

International Arbitration: An Attempt to Find a Holistic Solution’, in Global

Reflections on International Law, Commerce and Dispute Resolution: Liber
Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al. (Paris: ICC
Publishing, 2005), 772–773.
229 Klaus Peter Berger, ‘Evidentiary Privileges under the Revised IBA Rules on

the Taking of Evidence in International Arbitration’, International Arbitration

Law Review 13, no. 5 (2010): 177.
230 Fabian von Schlabrendorff & Audley Sheppard, ‘Conflict of Legal Privileges in

International Arbitration: An Attempt to Find a Holistic Solution’, in Global

Reflections on International Law, Commerce and Dispute Resolution: Liber
Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al. (Paris: ICC
Publishing, 2005), 773. There might also be gaming behaviour with a most
favoured privilege approach where a group of companies is involved where a
claimant might leave out a common law subsidiary simply to tactically reduce
the most favourable privilege.
231 Olaf Meyer, ‘Time to Take a Closer Look: Privilege in International
Arbitration’, Journal of International Arbitration 24, no. 4 (2007): 370.
232 An ICC Tribunal considered that the question of whether certain diaries

should be excluded was a matter within its general discretion and not subject to
any national rules of evidence. See ICC Case No. 7626 of 1995, in Albert Jan van
den Berg (ed.), Yearbook of Commercial Arbitration Volume XXII (The Hague:
Kluwer Law International, 1997), 134.
233 The European Court of Human Rights has considered this in the context of

‘equality of arms’. See, e.g., Dombo Beheer BV v. Netherlands, Application No.

14448/88, Case No. 37/1992/382/460, Judgment of 27 October 1993 (Merits
and Just Satisfaction), ECHR Series A, Volume 274; (1994) 18 EHRR 213.
234 Hilmar Raeschke-Kessler, ‘The Production of Documents in International

Arbitration – A Commentary on Art. 3 of the New IBA Rules of Evidence’,

Arbitration International 18, no. 4 (2002): 428.
235 Fabian von Schlabrendorff & Audley Sheppard, ‘Conflict of Legal Privileges in

International Arbitration: An Attempt to Find a Holistic Solution’, in Global

Reflections on International Law, Commerce and Dispute Resolution: Liber
Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al. (Paris: ICC
Publishing, 2005), 757.
236 Ibid., 766.
237 Ibid., 766.
238 Lukas F. Wyss, ‘Trends in Documentary Evidence and Consequences for Pre-

arbitration Document Management’, International Arbitration Law Review 13, no.

3 (2010): 124, n. 156.
239 Otto L.O. de Witt Wijnen, ‘Collection of Evidence in International Arbitration’,

in Liber Amicorum Bernardo Cremades, ed. M. Á. Fernández-Ballesteros & David

Arias (Madrid: La Ley, 2010), 357.
240 Ibid.
241 Klaus Peter Berger, ‘Evidentiary Privileges under the Revised IBA Rules on

the Taking of Evidence in International Arbitration’, International Arbitration

Law Review 13, no. 5 (2010): 175.
242 IBA Working Party & IBA Rules of Evidence Review Sub-Committee,
‘Commentary on the Revised Text of the 2010 IBA Rules on the Taking of
Evidence in International Arbitration’, 25. This follows on from the IBA Working
Party’s commentary in relation to a broad entitlement under the 1999 Rules to
exclude documents due to considerations of fairness or equality. It used as an
example documents privileged in one party's legal system but not the other's.
The commentary suggests that the tribunal may exclude production of the
technically non-privileged documents on this basis.
243 Klaus Peter Berger, ‘Evidentiary Privileges under the Revised IBA Rules on

the Taking of Evidence in International Arbitration’, International Arbitration

Law Review 13, no. 5 (2010): 171, 175.
244 James H. Carter, ‘Privilege Gets a New Framework’, International Arbitration

Law Review 13, no. 5 (2010): 179.

245 Ibid.
246 Martin Hunter & Gregory Travaini, ‘Electronically Stored Information and

Privilege in International Arbitration’, in Liber Amicorum Bernardo Cremades, ed.

Miguel Á. Fernández-Ballesteros & David Arias (Madrid: La Ley, 2010), 615;
Klaus Peter Berger, ‘Evidentiary Privileges under the Revised IBA Rules on the
Taking of Evidence in International Arbitration’, International Arbitration Law
Review 13, no. 5 (2010): 173.
247 Martin Hunter & Gregory Travaini, ‘Electronically Stored Information and

Privilege in International Arbitration’, in Liber Amicorum Bernardo Cremades, ed.

Miguel Á. Fernández-Ballesteros & David Arias (Madrid: La Ley, 2010), 619.
248 ECJ Case No. C-550/07 P of 14 September 2010, Akzo Nobel Chemicals Limited

and Akcros Chemicals Limited v. Commission of the European Communities, paras

40–44.
249 A submission to this effect was made but not ruled upon in Campbell v. UK,

Application no. 13590/88, Judgment of 25 March 1992 (Merits and Just

Satisfaction), Series A, Volume 233, para. 46; (1992) 15 EHRR 137, para. 46.
250 Fabian von Schlabrendorff & Audley Sheppard, ‘Conflict of Legal Privileges in

International Arbitration: An Attempt to Find a Holistic Solution’, in Global

Reflections on International Law, Commerce and Dispute Resolution: Liber
Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al. (Paris: ICC
Publishing, 2005), 755.
251 ICDR Rules Art. 20.6.
252 ICC Rules 2012 Art. 22(3). For the old rule, see ICC Rules 1998 Art. 20(7).
253 Klaus Peter Berger, ‘Evidentiary Privileges under the Revised IBA Rules on

the Taking of Evidence in International Arbitration’, International Arbitration

Law Review 13, no. 5 (2010): 174.
254 Case 155/79, AM&S Europe Ltd v. Commission of European Communities,

[1982] ECR 1575; Case T-30/89, Judgment of 12 December 1991, Hilti AG v.

Commission of the European Communities [1991] ECR 11-1439; See Joined Cases
T-125/2003 & T253/2003, Judgment of the Court of First Instance (First
Chamber) of 17 September 2007, Akzo Nobel Chemicals Ltd & Akcros Chemicals
Ltd v. Commission of the European Communities. Judgment of the Grand Chamber,
14 September 2010.
255 Olaf Meyer, ‘Time to Take a Closer Look: Privilege in International
Arbitration’, Journal of International Arbitration 24, no. 4 (2007): 276.
256 Balabel v. Air India [1988] Ch 317.
257 This is the norm in common law countries. In-house counsel are also
admitted to the Bar in Belgium, Denmark, The Netherlands, Germany and Spain:
Martin Hunter & Gregory Travaini, ‘Electronically Stored Information and
Privilege in International Arbitration’, in Liber Amicorum Bernardo Cremades, ed.
Miguel. Á. Fernàndez-Ballesteros & David Arias (Madrid: La Ley, 2010), 619.
258 Olaf Meyer, ‘Time to Take a Closer Look: Privilege in International
Arbitration’, Journal of International Arbitration 24, no. 4 (2007): 377.
259 Ibid.
260 The theoretical position is in fact more complex as common law sees this as a

question of admissibility, while the civilian perspective sees it as a question of

professional secrecy, in some cases allowing adjudicators to hear evidence but
attach whatever weight is considered fair. See Jason Fry, ‘Without Prejudice and
Confidential Communications in International Arbitration (When Does
Procedural Flexibility Erode Public Policy?)’, International Arbitration Law
Review 1, no. 6 (1998): 209. Fry argues persuasively against too ready
acceptance of such material and makes the key point that the parties should
clarify the position by agreement at the outset.
261 Bradford & Bingley PLC v. Rashid [2006] 1 WLR 2066 at 2091.
262 If that was the case, an aggrieved party might even consider arguing that the

tendering was in breach of an implied good faith term of the arbitration

agreement giving rise to damages, although such action might not be easy to
maintain.
263 Klaus Peter Berger, ‘Evidentiary Privileges under the Revised IBA Rules on

the Taking of Evidence in International Arbitration’, International Arbitration

Law Review 13, no. 5 (2010): 171, 174. See section 10.10 above dealing with
evidence from prior ADR processes.
264 Dolling Baker v. Merrett [1990] 1 WLR 1205 and see further Peter Ashford,

‘Document Discovery and International Commercial Arbitration’, American

Review of International Arbitration 17 (2006): 123–124.
265 Martin Hunter & Gregory Travaini, ‘Electronically Stored Information and

Privilege in International Arbitration’, in Liber Amicorum Bernardo Cremades, ed.

Miguel Á. Fernández-Ballesteros & David Arias (Madrid: La Ley, 2010), 624.
266 Ibid., 768.
267 Olaf Meyer, ‘Time to Take a Closer Look: Privilege in International
Arbitration’, Journal of International Arbitration 24, no. 4 (2007): 366; Fabian
von Schlabrendorff & Audley Sheppard, ‘Conflict of Legal Privileges in
International Arbitration: An Attempt to Find a Holistic Solution’, in Global
Reflections on International Law, Commerce and Dispute Resolution: Liber
Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al. (Paris: ICC
Publishing, 2005), 767.
268 Charles N. Brower & Jeremy K. Sharpe, ‘Determining the Extent of Discovery

and Dealing with Requests for Discovery: Perspectives from the Common Law’,
in Leading Arbitrators’ Guide to International Arbitration, ed. Lawrence W.
Newman & Richard D. Hill, 2nd edn (New York: Juris Publishing, Inc., 2008), 326
citing Richard M. Mosk & Tom Ginsburg, ‘Evidentiary Privileges in International
Arbitrations’, International and Comparative Law Quarterly 50, no. 2 (2001): 376.
269 Sections 38(4) and 44(2)(c).
270 Article 21.
271 Article 18.2.
272 ICSID Arbitration Rules Arts 34(2)(b) and 37(1).
273 This is now expressly provided in Art. 7 of the IBA Rules on the Taking of
Evidence in International Arbitration (2010).
274 UNCITRAL Notes on Organising Arbitral Proceedings (1996), para. 58.
275 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 412.

276 See, e.g., the US Federal Arbitration Act, §7.
277 English Arbitration Act 1996 (UK) s. 43.
278 Final Award, Pt II, Ch G para. 21, Methanex unreported 3 August 2005

NAFTA/UNCITRAL available at
<http://ita.law.uvic.ca/documents/MethanexFinalAwar...> and Caratube
International Oil Co. v. Republic of Kazakhstan (10-0285) (V.D.D.C. 11 August
2010).
279 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 399.

280 These concerns were raised in the WTO context in Donald M. McRae, ‘The

WTO in International Law: Tradition Continued or New Frontier?’, Journal of

International Economic Law 3, no. 1 (2000): 34.
281 Chester Brown, A Common Law of International Adjudication (Oxford: Oxford

University Press, 2007), 76; United States –Import Prohibition of Certain Shrimp
and Shrimp Products WTO Doc. WT/DS58/AB/R (1998), Report of the Appellate
Body, paras 99–110 (although the Appellate Body's approach was based in part
on a broad reading of principles in the Dispute Settlement Understanding).
282 Decision on Petitions from Third Persons to Intervene as 'Amicus curiae'

(under NAFTA Chapter 11 and the UNCITRAL Arbitration Rules), 15 January

2001, Methanex Corporation v. United States of America. See also Decision of the
Tribunal on Petitions for Intervention and Participation as Amici Curiae (under
NAFTA Chapter 11 and the UNCITRAL Arbitration Rules), 17 October 2001,
United Parcel Services of America Inc. v. Government of Canada, para. 61.
283 US Model Bilateral Investment Treaty 2004 Art. 28(3).
284 ICSID Case No. ARB/03/17, Order in Response to a Petition for Participation

as Amicus curiae of 17 March 2006, Aguas Provinciates de Santa Fe SA, Suez,

Sociedad General de Aguas de Barcelona SA and Interagua Servicios Integrates de
Agua SA v. the Argentine Republic, para. 33; see also ICSID Order No. ARB/03/19,
Order in Response to a Petition for Transparency and Participation as Amicus
Curiae of 19 May 2005, Aguas Argentinas SA, Suez, Sociedad General de Aguas de
Barcelona, SA and Vivendi Universidal SA v. the Argentine Republic. See further
Christina Knahr, ‘Transparency, Third Party Participation and Access to
Documents in International Investment Arbitration’, Arbitration International 23,
no. 2 (2007): 327.
285 See ‘Notes of Interpretation of Certain Chapter 11 Provisions’ (NAFTA Free

Trade Commission, 31 July 2001); see also OECD Investment Committee,

‘Transparency and Third Party Participation in Investor-State Dispute
Settlement Procedures: Statement by the Investment Committee’ (June 2005).
286 Convention on the Taking of Evidence Abroad in Civil or Commercial Matters,

opened for signature 18 March 1970.

287 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 405.

288 Ibid.
289 Deiulemar Compagnia di Navigazione S.p.A. v. M/V Allegra, United States
Court of Appeals (Fourth Circuit), in Albert Jan van den Berg (ed.), Yearbook of
Commercial Arbitration Volume XXV (The Hague: Kluwer Law International,
2000). A writ of certiorari was denied in Pacific Eternity, SA v. Deiulemar
Compagnia di Navigazione S.p.A, 529 US 1109.
290 Olaf Meyer, ‘Time to Take a Closer Look: Privilege in International
Arbitration’, Journal of International Arbitration 24, no. 4 (2007): 365.
291 Giorgio Bernini, ‘The Civil Law Approach to Discovery: A Comparative
Overview of the Taking of Evidence in the Anglo-American and Continental
Arbitration Systems’, in The Leading Arbitrators’ Guide to International
Arbitration, ed. Laurence W. Newman & Richard D. Hill (New York: Juris
Publishing, Inc., 2008), 269.
292 See, e.g., ICDR Rules Arts 19 and 20; LCIA Rules Arts 14 and 20–22; DIS Rules

Arts 26–27; WIPO Arbitration Rules Art. 48; SCC Rules Art. 26; CIETAC Rules Arts
12.1 and 15.2.
293 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1827.

294 See ICC Rules 2012 Art. 27; UNCITRAL Rules 2010 Art. 31.2; SCC Rules Art.

34; ACICA Rules Art. 30; HKIAC Rules Art. 27.2; Swiss Rules 2012 Art. 29.
295 See UNCITRAL Rules 2010 Art. 31.2; SCC Rules Art. 34; ACICA Rules Art. 30;

HKIAC Rules Art. 27.2; Swiss Rules 2012 Art. 29.

296 Harris International Telecommunications, Inc. v. Islamic Republic of Iran, et al.,

in Albert Jan van den Berg (ed.), Yearbook of Commercial Arbitration Volume XIV
(The Hague: Kluwer Law International, 1989), 408.
297 Ibid., 407.
298 Edgar Protiva & Eric Protiva v. the Government of the Islamic Republic of Iran,

in Albert Jan van den Bèrg (ed.), Yearbook of Commercial Arbitration Volume XXI
(The Hague: Kluwer Law International, 1996), para. 36.
299 Ibid.
300 UNCITRAL Notes on Organising Arbitral Proceedings (1996), paras 49 and

51; ‘Techniques for Controlling Time and Costs in Arbitration: Report from the
ICC Commission on Arbitration’, ICC Publication No. 843, ICC International Court
of Arbitration Bulletin 18, no. 1 (2007): para. 76.
Part II: The Process of an Arbitration,
Chapter 11: Documentary Evidence

Jeff Waincymer,

11.1. Introduction

A number of policy and practical issues arise in relation to the role of

documentary evidence in arbitration proceedings. The first is the relative
importance of documentary material as opposed to oral testimony. Common law
systems are said to rely heavily on the latter, while civilian systems are seen as
giving much more weight to contemporaneous key documents. Islamic principles
of dispute resolution are inclined to a contrary view on the basis that ‘the words
of an upright citizen were worthier than an abstract piece of paper or a piece of
information subject to doubt and falsification’. (1) Modern arbitration seeks to
combine the best features of both forms of evidence.

Where documents are concerned, there are still a number of important policy
questions. An important issue is who must produce relevant evidence. Is it
simply the party with the burden of proof on that issue? Conversely, must each
party produce all relevant evidence both favourable and adverse? Should the
primary responsibility be on the party in possession of a document or can that
wait until there is a specific request from its opponent? To what extent should
the tribunal itself call for relevant material, particularly where it believes the
parties themselves have not been comprehensive as to both production and
requests for production?

A second policy question relates to document production and whether a party is

entitled to demand production of documents by its opponent, even those adverse
to the latter's interests. (2) This is more contentious. There is a policy conflict
between those advocating full mutual knowledge of all relevant information
versus page "825" those suggesting the sanctity of one's possessions and the
libertarian right to demand that a claimant have an adequate case without calling
for evidentiary assistance from the respondent. Failure to comply with document
production orders may thus raise problematic issues as to when adverse
inferences should be drawn. Because of this policy debate, which is divided along
common law versus civilian traditions, (3) there is no automatic right to document
production in international arbitration, although this is becoming the norm,
albeit in a carefully qualified and controlled manner. A third policy question is
whether certain kinds of documents should be inadmissible on the basis that
they may be unreliable. This was discussed in section 10.6 and is discussed
further in section 11.7 below dealing with the grounds for refusing production
requests. A fourth policy question relates to timing and the appropriate stages at
which documents should be provided to the tribunal and opposing parties. This
was discussed in section 10.22 above.

These questions need to be considered in the context of party autonomy, arbitral

rules, mandatory norms, broad evidentiary discretions and even the
discretionary considerations in cost determinations that might shape procedural
behaviour throughout the process. (4) A number of technical questions also arise,
such as the proper way to treat electronic material, the relevance of
confidentiality and privilege, (5) proof of authenticity, accessing documents in the
possession of nonparties and translations where necessary.

11.2. Contemporaneous Documents

As noted in section 10.2, civilian legal systems tend to prefer contemporaneous

documentary evidence over oral evidence at hearings. The logic is that the latter
evidence, compiled after the dispute is known, is more likely to be self-serving
and hence less reliable. The same logic would support the value of
contemporaneous documents as compared to documents generated once the
dispute is known. (6) While that section questioned any rigid a priori preference
and argued that in many cases an amalgam of documentary and oral testimony
will be optimal, it is clear that all legal systems value contemporaneous
documents highly. This should be particularly important in commercial matters
where the disputes relate in large part to documents setting out and evidencing
rights and obligations and actions taken in respect of these.

page "826"

Within the arbitration field, many leading practitioners, even those with a
common law background, have argued for the relative importance of
contemporaneous documents. Charles Brower has suggested that the Iran-US
Claims Tribunal routinely saw contemporaneous written exchanges of the
parties pre-dating the dispute as the most reliable source of evidence and that
the actual course of conduct between the parties prior to the dispute arising was
the best evidence of the proper interpretation of any relevant contract. (7) A
similar preference for documentary evidence has been suggested in relation to
public international law generally. (8) The value relative to oral testimony may be
even more important where advocates without sufficient experience of cross-
examination techniques are representing one or both parties so that written
witness statements are not as well tested as they should be. (9) This is particularly
so where witness testimony is heavily rehearsed by opposing counsel (10) or the
written statements are clearly the work of counsel.

However, all evidence needs to be evaluated without any starting assumptions as

to probity. It would also be wrong to presume that documents are inherently
accurate, honest and without ambiguities. In particular, a tribunal should be
aware that a sophisticated party might draft its contemporaneous documents
with one eye on a potential dispute. A party could have created a self-serving
document such as minutes of a directors' meeting to falsify what actually
occurred. The words on paper may not fully reflect the reality at the time. Astute
business people will commonly make file notes of conversations they believe
might ultimately be central to any dispute. There can be no presumption that
people doing so will always adopt a balanced, accurate perspective rather than a
self-serving one. Psychology teaches us that even honest people will tend to sift
things they see and hear and give a higher priority to matters that meet their
preconceptions or suit their interests. A prospective customer might hear all of
the glowing adverbs about a product and fail to recall a waiver of liability or a
caution about certain kinds of misuse when listening to a sales pitch. Documents
can also be problematic in that they may be contradictory. (11)

Thought must also be given to cases where contemporaneous documents have

not been provided and the proper evidentiary inferences that might then be
drawn. page "827" The specific question is when adverse inferences might
legitimately be drawn that assume unfavourable content of such documents. This
must be looked at in relation to the circumstances and the ability of parties to
present such evidence. For example, where the Iran-US Claims Tribunal was
concerned, it was generally understood that because of the Iranian Revolution
and the swift departure of many individuals and corporations, many would not
have been able to provide such documents. (12)

In deciding how much to rely on contemporaneous documents, a cost benefit

analysis would also look at the costs involved in searching for relevant
documents, making requests, ruling on contested applications and having
counsel, the parties and the tribunal consider a vast array of documentary
material. Very often a tribunal is inundated with a plethora of potentially
relevant documents when only a small number are truly material to the outcome.
In many cases, even relevant documents are only material because a party
inappropriately refuses to concede what should have been uncontested facts.

11.3. Authenticity of Documents

International arbitration does not require that documents be admitted into

evidence by witnesses who attest to their nature and authenticity. This differs
from the common law tradition which calls for documents to be presented by
witnesses as part of their oral evidence and styled as exhibits. The witness
testimony attests to the authenticity of the document and helps explain its
context in terms of the broader evidentiary record. A civilian judge would
normally wait for challenges to authenticity before considering that as a separate
issue, although in some cases, notarisation is also employed. It is desirable to
operate on a presumption that documents are authentic and leave it to the other
party to raise any objections. Such a presumption saves time and only makes
validity an issue where it is truly a contested aspect of the case. The Iran-US
Claims Tribunal generally accepted documents without common law style
demands as to authenticity and presentation. (13)

A number of other presumptions would also be desirable. The first is that a

document does, in fact, emanate from the source as indicated on its face. The
second is that a document sent has, in fact, been received by the addressee. The
third is that a copy is an accurate representation of the original. There is no
blanket requirement to provide originals in international arbitration. (14) Only
copies need to be produced unless there is evidence suggesting inaccuracies or
page "828" incompleteness. It will generally be for the opposing party to make
any claims as to authenticity, provide reasons and seek an appropriate order
from the tribunal, either calling for proof or rejecting the documents. It will be a
matter for judgment on a case-by-case basis whether a challenge has presented
sufficient reasons for querying authenticity, what proof would be reasonable and
whether the tribunal should question authenticity on its own volition. In some
cases, a handwriting or similar expert might be called to give evidence in relation
to disputed documents where authorship is denied.

11.4. Document Presentation

Arbitral rules typically require parties to present all documents on which they
intend to rely. While the obligation is clear, the ambit is uncertain. All would
agree that parties must present the documents that they see as supporting their
case. More contentious is whether there is also a duty to present adverse
documents or documents which they intend to rely on to discredit opposing
witnesses. The general duty of disclosure is considered in section 11.5 below.

There are also timing issues. Documents to be produced by a party should either
be produced with their written submissions or shortly thereafter and in
sufficient time before the hearing to allow the other party to properly prepare its
responses. (15) This allows both parties to carefully evaluate the strengths and
weaknesses of opposing claims, defences and counterclaims. That is likely to aid
the parties in any possible settlement negotiations. It is particularly important to
have parties evaluate the strengths of each case at an early stage as one of the
biggest barriers to settlement is the amount of the legal costs in having a matter
ready for a hearing. By that stage, it is often the case that no reasonable
allocation of the primary claims can include enough contribution in relation to
legal fees to make a settlement viable for either party.

Presenting documentary evidence at the earliest stage also allows for

appropriate research on factual matters. That can be time consuming when a
range of witnesses need to be spoken to in different jurisdictions and/or where
scientific testing might be needed, for example to determine why a construction
project failed where a range of experts suggest different causes. There may also
be time needed for translation of documents in other languages. A tribunal's
procedural directions will be an important way of ensuring that these objectives
are met.

While each party must produce the documents in a timely fashion, there is also a
need to ensure equal treatment and that neither party gains an unfair advantage.
page "829" Sections 6.14.2 and 12.5.6 deal with timing of submissions and
witness statements respectively. Choosing between sequential or concurrent
submissions and statements will impact on the way documents are presented as
documentary evidence will typically be tendered at those times. If it is
sequential, the party presenting second can see what material its opponent is
relying on and then determine how much material needs to be presented. A
tribunal needs to take this into account when choosing between concurrent and
sequential processes. Even where concurrent submissions are concerned, there
is still a need to ensure that one party does not receive favourable treatment
simply by missing a deadline, receiving the other party's documents and then
being able to make revised submissions. Similar problems arise if only one party
is given an extension of a deadline. There is even the possibility of over-
production where one party aims to swamp its opponent with documents to
increase the latter's costs to unacceptable levels.

11.5. Duty of Disclosure

It was noted above that parties are required to present the documents on which
they intend to rely. The notion is ambiguous. Is it limited to favourable
documents or must each party present a full picture, including evidence adverse
to its interests? Must adverse documents be disclosed prior to a production
request or can a party wait to see if a request is made and granted by the
tribunal? (16) The following sections deal with requests for documents by one
party to another, grounds for refusal and tribunal approaches to such contested
applications. While this is separate to the presentation of documents that each
party intends to rely upon, the two are interlinked. Before a tribunal can
determine what attitude to take to a party's failure to produce documents, there
is a need to consider what primary obligation the party has. The more a party
should have presented a document, the more likely that a request for the
document is likely to be accepted. Hence, the analysis begins with a discussion of
the general duty of disclosure as to documentary evidence as there are
differences in view between legal families that polarise many practitioners and
scholars.

A number of cases and commentators have supported a general mutual duty.

Hanotiau considers that ‘the parties in dispute should co-operate in the
presentation of the truth by producing all the elements of evidence which they
have in their possession’. (17) As noted, the IBA Rules of Evidence 2010 declare
that each party page "830" shall act in good faith as to the taking of evidence. (18)
An ICC tribunal has also commented that the parties have a duty to cooperate in
good faith in the taking of evidence. (19) In William A Parker (USA) v. United
Mexican States, the Mexico-US Claims Commission said that:

it is the duty of the respective Agencies to cooperate in searching out and

presenting to this tribunal all facts throwing any light on the merits of the claim
presented. The Commission denies the ‘right’ of the respondent merely to wait in
silence in cases where it is reasonable that it should speak. (20)

The Commission also stated:

the parties before this Commission are sovereign Nations who are in honour
bound to make full disclosure of the facts in each case so far as such facts are
within their knowledge, or can reasonably be ascertained by them. (21)

Similarly, in the Sabotage case, the Claims Commission stated that:

It is well recognised that Governments who have agreed to arbitrate are under
obligation to enter in good faith to try to ascertain the real truth. (22)

A policy analysis might need to separately consider both claimant and

respondent's perspectives as the policy considerations might not be identical
even in the context of mandatory equal-treatment norms. From the claimant's
perspective, the obligation to arbitrate in good faith is said to extend to the
presentation of evidence. (23) A failure to provide relevant, albeit adverse
evidence may be a breach of a duty to engage in arbitration in good faith. (24) In
most cases, it would be highly misleading page "831" to present selective
favourable documents. In extreme cases it would be tantamount to fraud if the
selective documents suggested a contrary factual picture to that known by
claimant to be the reality. Legal counsel would commonly be in breach of ethical
standards in doing so, to the extent these apply in arbitration. From the
respondent's perspective, it too should not present a one-sided and misleading
picture but the complication is that it might not wish to make comprehensive
factual submissions at all and instead simply seek to test the claimant's evidence.
The libertarian perspective suggests that a respondent can challenge the
adequacy of the claim even if in possession of adverse evidence and need not
readily submit that evidence. Thus, a respondent might simply wish to provide
contrary evidence to that of claimant, to show that the latter's case was not
preponderant. A contrary view to the libertarian perspective is that the
respondent consented to arbitrate, which must also be seen as consent to
arbitrate in good faith. A failure to submit the adverse evidence might be bad
faith. This view may be harder to maintain as a finding of lack of good faith is a
very serious one. A party may refuse disclosure on the basis of a firm belief in the
libertarian position, confidentiality, lack of relevance, or similar grounds. Even if
the party's refusal is unreasonable, it may have taken the particular view in good
faith. While good faith obligations are, thus, at times troubling to apply in
practice, and are often of concern to common law practitioners in terms of the
vagueness of the concept, it seems preferable to stipulate that consensual
arbitration is indeed based on mutual principles of good faith. (25)

Another suggested basis of the duty for each party to disclose is that it flows
from mutual rights and responsibilities under due process considerations. (26)
When parties come from different countries and know that each has relevant
material, affording each an adequate opportunity to present its case may imply a
duty to cooperate as to evidence unavailable to the opponent. A related policy
consideration in favour of a general mutual duty to disclose is that in
international adjudication, a party in one geographical location has limited
ability to gather evidence in a foreign jurisdiction.

The parties may agree to rules which expressly provide an obligation to arbitrate
in good faith or at least act in good faith in the taking of evidence. (27) Where there
is an express obligation of this nature there would still be questions as to the
page "832" ambit of the duty. For example, if the duty only arises because of the
reference to the duty in the IBA Rules, that should be read in the context of other
obligations or express provisions in those Rules. A contextual reading would
suggest that the duty of good faith should not in and of itself require adverse
documents to be presented. (28) This is simply because the express provisions
only provide a duty to present the documents upon which a party intends to rely
and then allows for document production requests subject to specified criteria
and exceptions. It has also been suggested that the duty of good faith does not
give rise to an affirmative duty to preserve evidence or prevent destruction
simply because the IBA Rules do not impose such an express duty. (29) A contrary
argument is that there must be some circumstances where good faith would
require behaviour over and above that expressly indicated or implied
inferentially in other parts of the Rules. Relevant facts could vary from deliberate
destruction of evidence simply to mislead a tribunal to a simple failure to
interfere with a standard document destruction policy.

Duties of disclosure should be broad enough to encompass a duty to disclose the

existence if not the details of documents where production will be refused on
grounds such as confidentiality or State secrets. A good-faith approach to the
duty to arbitrate should involve bringing these matters to the attention of the
tribunal so that rulings can be made. The duty to disclose should be an ongoing
one, even after stipulated timeframes for presentation and production. (30) Less
clear is whether a duty to disclose documents on which a party intends to rely
covers documents to be used in cross-examination. The modern trend both
domestically and in international arbitration is to avoid surprises, hence
encouraging early disclosure of this information as well. (31)

Duties of disclosure and production rights may also be impacted upon by notice
obligations in the underlying commercial agreement. In some cases a party might
have a substantive contractual or other entitlement to documents. For example, a
joint-venture party may have a contractual entitlement to accounting records
and to have these audited by an expert of their choice. Shareholders may have
rights against the company. An insurer will have full disclosure rights against the
insured. Licensing and franchise agreements will provide for full disclosure.
Contractors are required to provide extensive information under FIDIC and
similar construction contracts. It is important to understand that these are
substantive rights that can be mixed up with broad procedural document page
"833" production requests. A tribunal will need to be careful how to deal with
claims that mix up substantive and procedural entitlements. A substantive claim
might allege a breach of that very entitlement, requiring a full deliberation and a
reasoned award by the tribunal. Conversely, a mere procedural determination as
to document disclosure is different in nature and may at times be made by a
chairperson alone. Where the entitlement is substantive, a party might seek an
interim measure as to preservation. It might also seek an interim award
providing a declaration that production be undertaken.

Duties of disclosure on parties may even be impacted on by lawyers' professional

duties to adjudicatory bodies. Under the common law, solicitors owe a duty to
the court to ensure that no relevant material is withheld. (32) Less clear is the
extent to which such duties apply to arbitration and the adverse inference
possibilities where the fault is that of counsel and not necessarily a party. While a
party has responsibility for counsel's behaviour, there still needs to be a
determination that the material is likely to be adverse and withheld for that
reason.

11.6. Document Production

11.6.1. Production Rights and Discretions

This section deals with production rights and discretions in relation to

evidentiary material. In some cases a party will have a substantive entitlement to
documents, such as the right to accounts in a joint venture. Such a right is a
substantive entitlement and is not dependent on procedural norms and
discretions. While there may be a difference between procedural and substantive
entitlements to documents, where substantive rights are concerned, procedural
issues can also be important. For example, a tribunal might not be in a position to
make a definitive ruling at the time of the request in relation to substantive
rights, not having heard key evidence as to the nature and meaning of the
relevant contractual terms. (33)

As noted above, there are significant differences at the domestic level between
and within legal families in terms of a party's obligation to present adverse
documents and to produce documents at the request of its opponent. These two
scenarios can be looked at alongside each other from a policy perspective as they
are likely to cover the same material. A valid document production request is
typically aiming to elicit adverse material that might have been presented in the
first instance. The difficulty as noted is that some page "834" civilian legal
systems adopt a libertarian approach and do not see it as appropriate for a
responding party to have to help the claimant make a case against it, although it
is inappropriate to speak of document production as a unitary concept in
relation to civilian systems. (34) The essential philosophy is that a claimant
bringing a case ought to have the appropriate material before doing so. A
respondent is not obliged to help a claimant find out if it has a case or help in
developing that case. A respondent might even suggest that a good-faith duty on
claimant requires it to produce adverse material so that its case can simply be
accurately and fairly assessed. Such views have to be tempered against the
general structure of civilian systems where the judge is concerned to find the
truth and takes the central role in directing the investigation and analysis.
Civilian judges generally do not feel constrained by the libertarian perspective. If
the judge believes a document is important, the party cannot ignore a production
request from a judge simply by saying it does not suit its interests to have it
presented. Civilian systems also allow parties to seek documents from the other
side, but this will generally be narrowly circumscribed and require that the judge
be convinced of the direct relevance of the documents sought. At the other
extreme, the US legal system often allows for expansive discovery of documents
even before trial, so the claimant can decide whether a case should be brought.
From a policy perspective, the key difference between civil and common law
jurisdictions is the trade-off they each make between truth on the one hand and
confidentiality, liberty and cost containment on the other. The right to privacy
and the libertarian right to demand that a claimant have a sufficient body of
evidence before bringing a claim justify the more restrictive approach in civilian
jurisdictions, (35) although this not uniform or static.
As with any extreme position, problems inevitably arise. Hence over time, both
civil and common law legal systems have gravitated to a more balanced position
where document disclosure is concerned. Common law legal systems now
typically involve judges determining just how appropriate discovery is per se
and how broad it should be, in the context of the nature and importance of the
case. (36) Some civilian legal systems have also broadened the scope for document
page "835" production. (37) Production rights can also be viewed from the
perspective that each party ought to have equal access to all relevant materials.
(38)

While the traditional view is thus that common law and civil law divide
drastically on document production, it is suggested that ‘(f)ull access to all means
of evidence under the control of the opponent or third parties is, in principle, the
present modern procedural standard of nearly all procedural cultures of
continental traditions’. (39) The core difference is not thought to be as to access
but rather as to the entitlement to ‘fishing expeditions’. (40) Civilian systems have
never accepted this entitlement and the English common law tradition has
gravitated to this position. The proper limits are not that of access but rather ‘the
sufficient specification and individualisation of facts and means of evidence as a
precondition for commencing fact finding procedures’. (41) Even where specificity
is concerned, the civilian tradition will allow more general requests when the
party seeking production has less means at its disposal to accurately identify the
documents in its opponent's possession. (42)

Arbitral statutes rarely provide express provisions in relation to document

production although commentators agree that this is implied in general
procedural powers. Where the UNCITRAL Model Law is concerned it has also
been suggested that Article 26(1)(b), which allows a tribunal to require a party
to give information, documents, goods and property to an expert, implies that the
same power should be seen as part of the broad Article 19 discretionary powers.
Similarly, the entitlement to seek court assistance under Article 27 supports this
broad interpretation. (43) The view is also supported by the drafting history. (44)
Production page "836" obligations can also arise under terms of reference of an
expert where the parties are directed to provide information upon which the
expert's report is to be written.

Where the parties have accepted the IBA Rules of Evidence 2010, Article 3
indicates that the taking of evidence shall be conducted on the principle that
each party shall act in good faith and be entitled to know reasonably in advance
of any hearing the evidence on which the other parties rely. (45) Knowing what is
to be relied upon does not expressly go so far as to cover adverse information,
although good faith is stipulated as an additional principle. Article 3.4 indicates
that a party to whom a request to produce is addressed ‘shall produce’ the
documents requested as to which it makes no objection. Limited objection rights
are available. Hence, agreement to these rules involves a presumption that
production will occur, absent proof of the application of the exempting
categories. Nevertheless, many of the exempting categories deal with questions
of degree, so the IBA Rules themselves lead to essentially discretionary
determinations, albeit ones constrained by defined criteria.
Objections can be raised as to the prima facie entitlement of the request and can
also be based on specified exempting categories. As to the gateway requirements,
the person requesting the document must make a sufficiently specific request,
indicate relevance and materiality to the outcome and demonstrate that it is
likely to be in the possession, custody or control of the other party and also not
in their own control or is too burdensome for them to obtain if it is. These issues
and the specific exempting criteria are discussed further in section 11.7 below.

In exercising discretions as to document production, an arbitrator may need to

consider the potential for challenges. A party may argue that absent documents
in the control of the other side, a full opportunity to present the case is
effectively denied. That argument may be strengthened if the issue is one on
which the party seeking discovery has the burden of proof. (46) There is greater
scope for challenge to negative determinations than to positive ones simply
because it is hard to allege a failure of due process if more relevant information
is provided rather than less. page "837" While challenges of this nature can never
be pre-empted, in the vast majority of cases they ought not to succeed. (47) Lord
Wilberforce in the House of Lords in overturning a decision granting broad
discovery at first instance on the basis that truth should be elicited whether
favourable or unfavourable to each party, noted that:

In a contest purely between one litigant and another… the task of the court is to
do and be seen to be doing justice between the parties… There is no higher
additional duty to ascertain some independent truth. (48)

The situation would be different if the parties had an agreement on document

production that was improperly ignored by the tribunal. Less clear is whether a
document production order by a tribunal could be rendered as an award and
hence made enforceable. Because an award must finally resolve an issue in
dispute between the parties, in most cases this would not be so. The situation is
different if a party is entitled to the documents under their commercial
arrangement and one of the claims is as to lack of documentary provision. (49)
Arbitrators are not normally seen as having a power equivalent to the civilian
astreinte order, compelling the payment of money for each day of non-
compliance with a production order, although parties could agree to grant such a
power, although there have been views expressed to the contrary. (50)

11.6.2. A Tribunal's General Approach to Production Discretions

It is important to separate out three issues. The first is whether relevant

documents in the possession of the other side ought presumptively to be ordered
by the tribunal when not voluntarily tendered unless a recognised exception
applies. The second issue is case management by an arbitrator to ensure that an
application is not a fishing expedition, and does not waste time and money by
calling for voluminous amounts of documents and lengthy debates about
exclusionary categories of confidentiality and the like. An overriding issue is
whether and why a tribunal might nevertheless have a broad discretion to refuse
an application even when limited to a discrete number of documents that would
appear to be of value, and under what page "838" criteria such a discretion
should be exercised? Another issue relates to timing. The IBA Rules of Evidence
indicate that an objection must be made within the time ordered by the tribunal.
Due process issues may arise if a party makes a late objection on grounds which
seem obviously valid.

Finally, a tribunal will have to take a view as to how to deal with relevant but
overly broad requests and whether the applicant should be given an opportunity
to refine such a request. Conversely, will the tribunal itself call for a narrow band
of material documents that seem crucial or will it simply deal with the presented
and produced material under principles of burden and standard of proof. As to
that alternative, Judge Holtzmann has made the following observation:

For me it is unsatisfactory to dispose of a claim or a counterclaim, on the stated

ground that a party has failed to bear its burden of proof, when an order to
produce specific documents would have permitted us to decide the issue on the
basis of evidence rather than lack of evidence - or more fairly to have drawn
adverse inferences against the party that failed to comply with the order.
Counsel may, through inadvertence or incompetence, not realise what facts the
Tribunal thinks are important, and we may never really know those facts unless
we ask for them. As I once wrote in a dissenting opinion, ‘(i)t seems unfair at the
late moment of writing the Award for the majority to indicate that (the claimant)
is somehow suspect for not answering questions (that) were never asked.’ (51)

Given that in each case the power to order production is a discretionary one,
granting a document production request involves balancing the obligations on
each party to prove their asserted facts, with the duty to cooperate in good faith
to assist the tribunal to find the truth, (52) and the general duty of efficiency both
as to time and cost.

There are a number of policy arguments either way as to whether orders should
be made as a matter of course, many of which have been alluded to above. If the
ultimate aim is to find the objective truth, then any relevant document ought to
be available for the adjudicator's benefit. Furthermore, if arbitration is to be
conducted in good faith, then an obligation to present evidence on which a party
intends to rely ought to be seen as an obligation to present a complete picture of
evidence and not simply documents in isolation that might appear more
favourable than ought to be the case. Hence, if either party presents an
incomplete picture, the other party should be able to demand the balance. As
noted above, the converse argument from respondent's perspective is that any
claim in an adjudicatory environment is an invasion of its liberty. The claimant
ought only to be able to do so when it has a sufficient evidentiary record to
support page "839" its case. Civilian attitudes also flow from a strict application
of the burden-of-proof principle. This leads many to consider that extracting
evidence from an opponent is contrary to the essential nature of the burden. Yet
to state that you have a burden does not indicate the method by which one may
satisfy it. Civilian systems do not require a prima facie case from claimant as a
formal first step towards an obligation to respond. Some also argue against the
practical value of forced production, particularly where arbitration relies heavily
on exchanges of written submissions that themselves require concurrent
documentation. (53) In turn, that might depend on how proactive the arbitrator is
in reading and analysing submissions when received and inquiring in relation to
evidentiary gaps. It also depends on the comprehensiveness of the unilateral
presentation and whether adverse material is presented.

Forcing a respondent to open up its own documentary material as a matter of

course could encourage fishing expeditions, looking for evidence to support a
claim. At the extreme, this might encourage premature cases in the hope that
document production can turn an unlikely claim into a valuable one. Early
contests about document production can also add to the costs, delay and
animosity between the parties, given that they are based on legal principles
drafted with great generality, covering situations where legal families have very
different views. Where a tribunal has to adjudicate on production requests
between parties with fundamentally different views, any decision taken might
lead the aggrieved party to feel that the arbitrator is not disposed towards them.
Any expensive and time-consuming process can also be improperly used as a
delaying tactic and/or a means to pressurise a disputant with less funds to
accept an unfavourable settlement offer. (54) Even the wish to assist the parties
with provisional timetables might unwisely encourage document requests that
might not otherwise have been made where this is listed as an agenda item as a
matter of course. (55) Less clear is whether comprehensive document production
aids settlement or not. The alleged benefits of forced production of adverse
material are said to include better perspectives on the strengths and weaknesses
of the case which should help refine issues and promote settlement and assist in
identifying misleading evidence from the other side. (56) The more that evidence
is all available, the easier the parties can identify the likely chances of success.
This should promote settlement as, if they are well advised, they should come to
similar conclusions as to the likely parameters of page "840" the adjudicator's
determination. However, the adversarial nature of contested production
applications might promote animosity and, hence, be less conducive to
settlement. Costly production also increases total legal costs, which itself is a
barrier to settlement.

Over time, international arbitration has attempted to find a compromise position

between the extremes and has concentrated on practical issues rather than the
libertarian debate. This has meant that the libertarian defence is not regularly
supported. It might be argued that in international arbitration, there is at least
majority agreement that each party, in presenting the evidence on which it
intends to rely, should volunteer adverse information in good faith as well as
positive information and second, that each party has a presumptive right to
demand production of such adverse evidence unless it falls within some
narrowly defined exceptions. If that is so, the common law position has come to
dominate international arbitration on this issue. There is at least one additional
policy argument in support of this outcome that at least addresses the libertarian
issue even if it does not do so unassailably. While domestic litigation systems
take different approaches to balancing truth versus privacy, arbitration differs
from litigation in that it is based on party consent and implied obligations of
good faith within that consent. This could imply that the parties intended that
whichever party deserves to win evidentially should indeed win, that victory
should not instead be based on tactical manoeuvring and that a tribunal's broad
discretions and parties' duties should work towards this outcome. Having said
that, a libertarian can argue tenably that the consent was not open ended and did
not waive their fundamental views on adjudicatory process.

At the very least, some of the more practical arguments for and against
prehearing document production, on the basis of timeliness, cost and efficiency,
are easily polarised and overly simplistic. In some cases, a sensibly managed
production process will save significant time at a later stage and may promote
settlement. In other cases, an excessive and unconstrained process may add
inordinate delay and cost and severely disadvantage a meritorious party that has
less in the way of financial resources. Some requests will be non-contentious.
Thus, the better conclusion from the practical side of the debate is that this
process, like many other discretionary processes in arbitration, needs to be
thought about carefully as to when it is appropriate, and designed on a case-by-
case basis to promote fairness and efficiency and control cost.

11.6.3. The Process of Document Requests

While there is no uniform rule, subject to party autonomy, most arbitral statutes
and rules provide directly or indirectly that arbitrators can exercise a broad
discretion on a case-by-case basis. While an express or implied power is
generally provided, most rules do not outline a process for requesting document
production. The IBA Rules of Evidence 2010 and their attempted compromise
have as a result become influential. The broad structure of the rules and typical
tribunal practice is page "841" that parties first present documents on which
they intend to rely. A tribunal can help smooth the process by indicating at the
outset that documents not disclosed cannot be relied on without tribunal
consent. (57)

Each party may then request further documents from the other. Document
production requests are made both to the tribunal and the other party so that the
tribunal need not intervene if the request is acceded to. (58) If there is an
objection, the tribunal may invite the parties to consult. (59) Under the IBA Rules,
there are two primary criteria and three key exceptions that apply to entitlement
to seek document production from other parties. As to the primary
requirements, the documents must be ‘relevant and material for the outcome of
the case’. Second, the request must be sufficiently specific, implying that it should
not constitute a ‘fishing’ expedition. The criteria are discussed further below in
the context of the grounds for validly refusing production requests in section
11.7. The tribunal then has the power to make determinations on contested
applications and otherwise provide fair and reasonable limits on the amount of
documentation sought. The tribunal might also give directions as to how
documents are actually to be handed over and who is to initially pay for the cost,
until such time as a later costs order is made in the arbitration itself.

Under the 2010 IBA Rules, the documents are not automatically provided to the
tribunal. (60) This allows the parties to first determine which produced
documents are worthy of submission for tribunal consideration. The tribunal
should not have to sift through voluminous amounts of documents to determine
what is, in fact, material. The tribunal also has its own power to call for
documents it considers material. If a tribunal intends to use these powers, it
should give a party the opportunity to raise objections before making a final
direction. (61) The general approach to exercising discretions in the context of
production rulings is discussed in section 11.8 below.

11.6.4. Timing of Document Production Requests

It is also important to consider the optimal timing for document production

requests. General evidentiary timing issues were considered in section 10.22 and
comments about timing of document presentation were made in section 11.4. As
to document production, the unique aspect is that contested production requests
require a determination by the tribunal against stipulated criteria. A tribunal
should ensure that this is undertaken at a point in time where meaningful
decisions may be made about relevance, materiality and reasonableness of
burden. page "842" There are conflicting arguments as to the timing of document
requests. One argument is that all documents should be available to each party
before it needs to clarify its submissions. A contrary argument is that the parties
should frame their submissions first and the tribunal can then circumscribe
production requests so as to be limited to the submissions as made. (62) The
predominant view is that the ideal time will typically be after detailed written
submissions have been tendered that include the documents upon which each
party seeks to rely, but sufficiently before the hearing so that requests can be
made, rulings made and documents analysed prior to the hearing. If there is to be
more than one round of written submissions, one may wish to time the request
for document production between the two stages so that the second round can
address the other party's evidence. In cases where the parties have each had all
or most key documents from the outset, an earlier production request may be
appropriate. An example might be a joint-venture dispute where one party
simply alleges that there is an evidentiary gap in respect of a document solely in
possession of the opposing party. Another situation where early production may
be called for is where a claimant investor has been evicted from a host State and
has no access to necessary documentation to bring its claim.

In some cases a tribunal may wish to deal with document production in stages,
perhaps ordering production of clearly relevant material or documents on which
each party intends to rely, after which further applications might be made to
supplant the material which has been produced. That may be expressly included
in terms of reference or in an early procedural order. Article 3.14 of the IBA
Rules of Evidence 2010 now contemplates that the tribunal may separate
document production into various phases such as in relation to jurisdiction,
preliminary determinations, liability or damages. Where an arbitration is being
conducted under tight timeframes, particularly by agreement of the parties,
there may be a need to organise document production processes that alleviate
onerous production obligations while at the same time allowing the other party
to properly prepare its case. Tavender alludes to an example where he asked the
party seeking production to prioritise its requests so that the other party could
deal with them in that order, with revisions made on a weekly basis. (63) Similarly,
a tribunal might wish to delay final decisions on contested production
applications until various submissions and witness statements have been
received so the parties can hone in on the gaps in the evidentiary record
presented. This might also follow an invitation to the parties to attempt to agree
on a common set of facts and documents. In extreme cases, it may even be that
further document production could be called for after the merits hearing where
there are crucial gaps in the evidence that the tribunal wishes to have filled. (64)

page "843"

In each case a tribunal will need to consider the request on its merits. In terms of
timing alone, the more that a late request flows from the lack of adequate
primary preparation by the party making it, the more easily it can be rejected.
Conversely, if the request comes soon after the opposing party has raised an
issue, no matter how late in the proceedings, then the request would seem more
reasonable. A further relevant factor is whether the material requested should
have been disclosed in any event by the party in possession of it. In such
circumstances, adverse inferences may be sufficient but in some cases a tribunal
would prefer to see the material before making its final determination. It is also
important to consider timing issues in the context of proactivity generally. The
more proactive an arbitrator, the more reason there may be to allow production
to occur in stages as and when the tribunal hones in on the key issues.

11.6.5. Pre-arbitral Discovery

International arbitration cannot provide the same extreme version of pre-trial

document production as occurs in the US, regardless of one's policy preferences,
simply because before constitution, there is no tribunal to make such an order. In
such circumstances it would be for a local court to award such an entitlement.
This would depend on whether a court has jurisdiction to make such an order in
situations that are clearly covered by an arbitration agreement. Even if there was
jurisdiction, a court might be reluctant to exercise such powers.

A policy analysis of pre-arbitral discovery options could bring together some

complex and potentially conflicting arguments. To some, such a request should
simply be dealt with under the applicable court rules and court discretions. To
others, it would be contrary to the arbitration agreement to use mechanisms
designed for litigation. Opponents of that view would say that the courts are
merely being utilised in aid of arbitration, albeit one not as yet commenced. In a
practical sense the question could be more complicated if the application for pre-
arbitral discovery sought to identify claims not within the arbitration agreement.
(65)

In Navigator Investment Services Ltd v. Acclaim Insurance Brokers Pte Ltd, (66) a
Singapore Court of Appeal considered that pre-arbitral discovery and
interrogatory requests were permissible and were not within section 6 of the
Singaporean International Arbitration Act which calls for courts to stay
proceedings pending arbitration. The decision in Navigator followed the earlier
Court of Appeal decision in Who Hup (Pte) Ltd & Ors v Lian Peck Construction Pte
Ltd (67) To be entitled to such an order it would appear that there needs to be
page "844" valid circumstances where some future court action may be
permitted. An example would be claims against parties not subject to the
arbitration agreement. In a practical sense it may be easier to claim pre-arbitral
discovery in group of company or agency situations, simply because at that stage
the court will not know whether an applicant intends to take a broad view as to
the coverage of the arbitration clause. An advocate may need to articulate one
potential cause of action that would fall outside of the arbitration clause. (68) In
Equinox, a Singapore High Court denied an application for pre-arbitral discovery.
(69) The plaintiff relied on the contractual right to inspect records and the

presence of an arbitration clause to seek such an order. The Court declined,

considering that it had neither express legislative power nor any inherent
jurisdiction to grant such a discovery order. The Court also considered that to
grant the order would go against the desirable policy of minimal interference
with arbitral proceedings. In Equinox, the affidavit filed did not make claims
against non-arbitral parties. (70)

In some cases, parties might allow a pre-arbitral referee to exercise such powers,
but this would be unlikely. (71)

11.6.6. Should Parties be able to Seek Production before Presenting the

Documents on Which They Intend to Rely?

A tribunal might face a request for production rights prior to an obligation to

submit a statement of claim. It would not be the norm in international arbitration
to grant such a request but there should be no hard and fast rule to that effect.
While a claimant may tenably argue that some situations could benefit from
access to documents before a clear statement of claim could be provided,
arbitration is not constrained by restrictive pleadings rules and timeframes.
Hence, even in such circumstances, it would usually make good sense for an
arbitrator to still require a statement of claim, noting that an application to
amend or provide supplementary material is likely to be forthcoming. However,
in some cases a statement of claim will be unhelpful without being able to refer
to documents that are known to only be in respondent's possession. An example
would be a claim against a non-signatory to an arbitration clause under
assignment, agency or group of company theories. Another situation might be an
expropriation dispute where a licence was revoked without reasons. The
claimant is aggrieved but has no paperwork on which to construct a meaningful
statement of claim.

In such circumstances there is a particular need to prevent fishing expeditions.

In practice this will be harder for the tribunal to resolve at the outset as opposed
to page "845" after submissions, as materiality tests have to be considered
against the frame of reference of the claims as made. If the request or notice
commencing the arbitration has sufficient articulation of the claims, this should
not be a problem. If not, that might be a factor to consider in deciding on timing.
There will also be a need to consider cases where terms of reference are
required and when new claims can be made following these. Claimant rights and
the incentive for fishing expeditions may vary significantly between requests
that pre-date and post-date finalisation of such terms. For these reasons, it
would be preferable in most cases to wait until claims are sufficiently articulated
either in the request, terms or statement of claim before allowing for the
document production stage. Due process will also need to consider equality in
terms of respondent's rights. There seems no reason to allow respondent to
access documents before the statement of claim unless claimant is granted such
access. After that time, similar considerations would arise as above.

11.6.7. Internal Documents

One contentious category is that of a party's internal documents and whether

these can be the subject of production orders. The problem is the difference in
view between legal families. Common lawyers would see all documents being
presumptively appropriate for production unless falling within defined
exempting categories. Many civilian attorneys consider that forcing production
of internal documents is tantamount to improperly forcing a party to testify
against its self interest. (72) From a policy perspective there are problems
whatever the treatment of internal documents. If they are privileged on that
basis, unscrupulous parties are likely to over-utilise this defence. If there is no
such privilege, it is a disincentive to fair and accurate recording of events in a
commercial environment.

While certain documents can be withheld on the basis of clearly defined

exceptions discussed below, there is no blanket category that is automatically
excluded. Importantly, internal documents now regularly fall under document
discovery processes as is contemplated by the 2010 and 1999 versions of the IBA
Rules of Evidence. This is to be contrasted with the 1983 Rules which limited the
request to documents which had ‘passed to or from… (a) party or to a third party
who is not a party to the arbitration’ . (73) While the IBA Rules of Evidence 2010
page "846" have broadened the category of documents to include internal
documents, unless these Rules are expressly agreed to by the parties they cannot
be determinative. Absent such agreement, there may thus be a residual debate
where civilian parties are concerned.

11.6.8. E-Documents

A significant amount of commercial information is now stored in electronic form.

It is now appropriate to consider electronically recorded information as part of
evidentiary obligations. Prima facie, this would mean that they should be treated
in the same way as hard copy documents. The same principles should apply in
respect of relevance and materiality to the outcome although their essential
nature adds practical problems that can impact upon discretionary
determinations. There are, however, special practical issues in relation to
storage, retrieval, ease of modification and ability to retrieve deleted data which
can all give rise to particular challenges to rule design and arbitral discretions.

The IBA Rules of Evidence 2010 now expressly refer to production of e-

documents. The Rules define Document to mean ‘a writing, communication,
picture, drawing, program or data of any kind, whether recorded or maintained
on paper or by electronic, audio, visual or any other means…’ The new Rules, in
dealing with the Request to Produce now also allow for inclusion of search
terms, files and other methods for identifying electronic documents.
Nevertheless, the Commentary indicates that the Rules are ‘neutral’ as to
whether e-documents should be produced and simply aim to provide a
framework for doing so where this is agreed or directed by the tribunal. (74) One
question is whether e-documents should be provided in electronic format or
should be produced in paper form. Article 3.12 of the IBA Rules indicates that
electronic documents are to be submitted or produced in the form most
convenient or economical to the provider that is reasonably usable by the
recipients, unless otherwise agreed, or the tribunal otherwise directs.

In considering discretionary decisions on production, there are some important

fairness, efficiency and jurisdictional challenges where e-documents are
concerned. Electronic storage means that there can be vast amounts of data
potentially available and readily transferable. While storage and transfer has
never been easier, the larger the volume, the more time and effort there will be
in evaluating the material and hence the cost of the arbitration. Second,
electronic documents can be stored in a multiplicity of places, impacting upon
jurisdictional power and the page "847" potential approach of conflicts rules.
Such multiplicity also gives rise to duplication. There may be numerous copies of
the same file on a range of computers and servers. One of the difficulties is that
electronic material is now typically stored on internal and external hard drives,
back-up devices, servers, DVDs and CDs, smart phones, portable flash memories
and cloud storage programs. Some of this material is commonly in use and
readily accessible, while others are simply back-ups or archives and may
themselves be hard to locate. It is now thought that some 90% of information is
at least generated in this form. (75)

There is also a problem in the fact that most users do not have consistent and
reliable processes for separately storing discrete documents. For example, key
documents might be stored in a contract specific folder, a folder of directors'
minutes or a folder for tax purposes. There will also be problems in determining
whether multiple files merely duplicate each other or whether there are relevant
and material variations between each. Emails may be stored and saved in a
number of places, at times in contravention of corporate directions. Databases
may be particularly problematic when constantly updated and when there is a
need for historical perspectives. (76) There are, however, programs that can check
for duplication that thus reduce this problem although not all subjects of
production requests would have access to such programs. Other problems arise
if there is no suitable way to track data when an employee leaves a company. (77)
Access to automatic backup tapes may also be an issue. Another problem with
electronic production is duplication with hard copy documents. Some companies
have elaborate systems to adequately manage electronic documents while most
do not. Whether the documents are printed or copied electronically may also
impact upon whether the request is seen as unduly burdensome. If documents
are only provided electronically, there is then a need to ensure that they have not
been tampered with. Attitudes may vary depending on whether it is a party or
third party who is disclosing e-documents. Given a party's obligation to make
complete disclosure, the burdens of electronic searches may be very high.
Conversely, where third parties are concerned who may simply wish to comply
with a request as quickly as possible, knowing that there are generally no
sanctions against them for gaps in disclosure, an electronic search may be highly
desirable.

Section 11.7 below deals with the grounds for refusal, one being a lack of
specificity. It is important to consider what specificity would mean in the context
of electronic data. One of the difficulties is that electronic data is stored in a
range of ways and there is no common process akin to chronological filing as
occurs with page "848" hard copy correspondence. Some email systems archive
chronologically, some link ‘conversations’ together. Earlier drafts of a contract,
which might constitute aids to interpretation, can be stored under different
names in different folders in a range of computers. There might be tens or even
hundreds of versions with modest variations. Electronic documents may be
stored with ambiguous titles, making it even harder to properly identify all
relevant material. While these are all problems, modern software allows for very
effective search methods based on key words. (78) Article 3.3(a)(ii) of the IBA
Rules of Evidence 2010 indicates that the requesting party may, or the tribunal
may require, identification of ‘specific files, search terms, individuals or other
means of searching for such documents in an efficient and economical manner’. It
is desirable to agree on key word lists wherever possible. One reason why
agreement is desirable is that while the typical obligation to specify the material
is on the person requesting it, it will commonly be the recipient of the request
who understands their own system and the ideal naming protocol to best
identify files. (79) In some cases a tribunal might even wish to use the services of
an expert to identify the appropriate key words and relationship criteria. If key
words are to be used, there ought to be an agreement that this replaces manual
searching. The difficulty would be if manual analysis of all e-documents was
required given the sheer volume, multiplicity of facts and variations in
descriptors.

However, it is important to understand that a key word search is essentially

different to an evaluative search. The latter involves a consideration of a
document and the exercise of human judgment as to whether it is likely to be
relevant and material to an issue. A key word search is simply a computer
generated investigation as to whether the key word exists or not that aims to be
an accurate alternative to such human judgment. Yet it can never fully do so.
Search terms are hard to get just right, capturing all relevant documents and
excluding irrelevant ones. Importantly, they cannot deal with key gateway issues
of true relevance, materiality, duplication, custody and exclusionary categories.
Furthermore, undue reliance on search terms would make it harder for the
tribunal to decide at the outset whether the request is an element of a ‘fishing
expedition’. For these reasons, search terms should only be seen as sifting
devices prior to the exercise of real human judgment and hence should err on the
side of prima facie inclusion. The important issue is then to give appropriate
consideration to the real tests of relevance and materiality.

Accepting the use of search terms too readily could also involve a shift in the
nature of the test of narrowness and specificity. Specificity should not simply be
page "849" about the key word, but should be about the prior identification of
the document. It is important to note the reference to search terms in relation to
‘such’ documents in the IBA Rules of Evidence, which should mean documents
sufficiently specified. A specific search term should not be a means to avoid the
need to separately articulate a particular document or a narrow and specific
category of documents. A search term is simply a necessary way to gain access to
e-documents where there is an entitlement to production at the request of a
party. Stated another way, if the tribunal has to determine whether a request is
sufficiently specific and material, the applicant must do more than simply
indicate the search term but should explain why that is the best way to identify
specific, relevant and material items.

Section 11.7.7 below also deals with a tribunal's discretion to reject a production
request on the basis that it poses an unreasonable burden on the party receiving
the request. The burden may vary depending on the software currently available
to the party against whom the request is made. Should a party without such tools
be expected to buy them or hire consultants for such purposes? There is no
reason why this should not be so in appropriate circumstances. In some cases the
data may be on outmoded hardware requiring obsolete software to be able to
read it. (80) The party requesting the data might even offer to cover the expenses
if the burdensome nature of the request is a potential barrier to a successful
application. A tribunal might also make entitlements conditional on paying all or
part of the marginal costs. This suggests that in appropriate circumstances, a
tribunal should not simply look at the burden on a particular party but should
also consider its cost-shifting powers and look at the overall cost/benefit to the
arbitration as a whole in either providing for production or rejecting such a
request. Another question is if a party utilises such an electronic search, is this
enough or should concurrent ad hoc physical searches also be undertaken?
Parties and tribunals will need to consider sensible processes that provide for
confidence as to coverage. Limits might be imposed by identifying particular
custodians, limiting time frames, limiting locations, and use of sophisticated
search terms. (81) A tribunal might also provide for sampling techniques to review
a portion of produced documents to determine whether broader production is
desirable on a cost/benefit basis. (82) Early consideration of these issues might
allow for a more limited primary search, with the possibility of follow-up
requests if gaps are shown to be likely.

Other issues such as confidentiality and privilege will generally apply equally to
e-documents as to paper-based documents. In some cases, particular issues are
more likely to arise. For example, computer-based search and production is
more likely to lead to inadvertent presentation of privileged materials, raising
questions page "850" as to whether this constitutes waiver under applicable
laws. Different legal systems have different views in relation to the nature of
privilege and the implications of inadvertent disclosure. Because arbitration
supports party autonomy, an ideal is to have an agreement allowing for clawback
of accidentally divulged material. (83)

Electronic means are also typically used for submitting documents. This is
encouraged under Article 2.2(e) of the IBA Rules of Evidence 2010. Production in
paper form may be appropriate where there are a small number of documents
retrieved. Conversely, if the arbitration is to be managed through paper
documents converted to electronic form, it makes no sense to convert an e-
document to paper, only to have to convert it back again. (84) In some cases,
tribunals are utilising web-based platforms to allow access to electronic
documents by all parties. (85) Finally, tribunals should accept that their broad
powers over the taking of evidence would allow for appointment of an
independent IT expert that could assist in managing these determinations by
commenting on search terms, identifying more readily accessible sources and
the like. (86)

11.6.8.1. Sedona Principles and E-Document Protocols

Developments in common law litigation have been influenced by the Sedona

Principles which in turn have inspired protocols by leading arbitral institutions
as referred to above. (87) The proposals were developed by the Sedona
Conference, a private organisation aiming to develop policy-based legal
principles. The Sedona Principles acknowledge that human review of every
potential document is not required and electronic search techniques may be
relied upon. The principles connote a proportionality standard, balancing the
difficulty against the importance of the requested information. (88) The party
requesting production should be invited to identify search terms so that the
outcome is agreed or a ruling can be sought from the tribunal as to the adequacy
of the terms. The Sedona Principles and other arbitration protocols are not
binding, absent agreement by the parties, but tribunals can expect to hear
argument that the proposed balance identified in the Principles page "851" ought
to apply in international arbitration under broad tribunal discretions.
Conversely, those opposing production may need to consider why they believe
such principles ought not to apply.

The Chartered Institute of Arbitrators has issued a Protocol for E-Disclosure in

Arbitration. (89) Under the Protocol, the parties are urged to confer at the earliest
opportunity regarding preservation and disclosure of electronically stored
documents and seek agreement on the extent of production. The tribunal is
asked to address this with the parties at the earliest opportunity and no later
than at the preliminary meeting. Parties are asked to consider whether either is
likely to call for such documents, the type of documents, preservation steps, rules
governing disclosure, whether there should be agreed limits, tools and
techniques, arrangements for privacy and privilege protection, including
clawback provisions and professional guidance that may be needed. Production
obligations may be limited to ‘reasonably accessible’ material which should
normally be produced in the format in which it is ordinarily maintained or in a
reasonably usable form. Comments as to e-disclosure are also included in a CPR
Protocol. (90) The ICDR Guidelines for Information Disclosure and Exchanges in
International Arbitration Proceedings also deal with e-discovery. (91) The ICC has
appointed a task force to make recommendations. (92)

11.6.8.2. Metadata and Deleted Documents

In some instances, computer information may be highly relevant but would not
normally be viewable when the document is printed. The obvious example is
deleted files, which can nevertheless be retrieved from a computer's hard disk.
Another potentially important source is metadata, being the additional
information stored and associated with electronic documents. For example, the
latter can show what amendments were made to a document and when and by
whom, which in some cases might help show the intent and reasonable
expectations of the parties, or even tampering with evidence. (93) A strict
application of legal rules dealing with ‘documents’ may not necessarily cover
metadata which is simply descriptive data page "852" about the documentary
data itself. (94) The broader definition of ‘document’ in the IBA Rules of Evidence
2010 covering ‘data… maintained…by electric… means’ is suggested to be broad
enough to encompass metadata, subject to overriding discretions of the tribunal.
(95) The Sedona Principles support access to metadata that is reasonably

accessible (96) but suggest that deleted files would require the requesting party to
demonstrate the need and relevance as compared to the cost and burden of
retrieval. (97) The CIArb Protocol suggests that the party seeking metadata should
demonstrate that relevance and materiality outweigh the costs and
administrative burden of production unless metadata would otherwise fall
within the produced format. (98) Concerns for good faith issues and a wish to limit
fishing expeditions perhaps led the CPR Protocol to suggest that ‘(r)equests for
back-up tapes, or fragmented or deleted files should only be granted if the
requesting party can demonstrate a reasonable likelihood that files were
deliberately destroyed or altered by a party in anticipation of litigation or
arbitration and outside of that party's document-retention policies operated in
good faith’. (99)

11.6.8.3. Document Destruction and Retention

A further problem relates to preservation. It will be more easily arguable that

electronic information was inadvertently deleted, either through the behaviour
of employees such as clearing email or through auto delete programs and the
like. (100) However, it is harder to fully erase electronic documents than it is to
destroy paper documents. (101)

Principle 5 of the Sedona Principles ‘requires reasonable and good faith efforts to
retain information that may be relevant to pending or threatened litigation’. This
would be relevant if breached. The balancing exercise also needs to be
considered in the context of a party's general document retention obligations for
such purposes as taxation and corporate law.

page "853"

11.6.9. Production Rights as against Third Parties

In many cases, evidence central to a dispute could be held by third parties. An

example would be material in possession of a subcontractor in a building dispute
between the owner and the contractor where the subcontractor is not a party to
an arbitration agreement. Another example may be a shipping dispute where
only two out of the shipowner, carrier, shipper and insurer are involved in
arbitration. In most cases the information adds to the factual matrix in the
dispute between the parties. In some cases the documents sought against a third
party are aimed at undermining that party's evidence as a witness. For example,
in London & Leeds Estates Ltd, a subpoena was upheld seeking an expert's
reports on similar but unrelated matters to support cross-examination as to the
report tendered in the instant case. (102) A tribunal's broad discretionary powers
will not lead to document production rights as against third parties as the latter
have not consented to arbitral power in that regard. They have not consented to
the economic burdens that flow from document production.

The Model Law and other Acts do not expressly provide this power, but are said
to be capable of interpretation in that regard. (103) In some cases, the arbitration
legislation provides such a direct power or allows the tribunal to seek judicial
assistance to that effect, although they tend neither to expressly mention third
parties nor limit reference to parties. (104) In the US, section 1782 of the United
States Code authorises US courts to provide assistance in relation to matters
brought before foreign tribunals. The ambit and application of this provision is
contentious. (105) This issue will typically raise questions as to the particular
legislative mandate of the relevant court, any inherent jurisdiction the judges
believe that they have, their views on timing as to whether application should
first be made to the tribunal and finally whether the application is a discrete one
or is an application for court assistance in respect of a positive determination of
the tribunal itself.

Article 3.8 of the IBA Rules of Evidence 2010 indicates that a party may ask the
tribunal to take whatever steps are legally available to obtain requested
documents that are relevant and material. A tribunal may also request
production on its own volition pursuant to Article 3.9. The party so requested
may again object under any of the factors enunciated in Article 9.2. It only seems
reasonable to afford the third party an opportunity to challenge the request on
similar grounds to a party. (106)

page "854"

While compulsion powers over third parties may be difficult, at times the very
existence of directive powers will assist voluntary compliance. (107) In some
cultures, a mere request from an arbitral institution may be successful in
obtaining documents from a third party simply because of the respect for the
institution. (108) However, if the third party refuses to produce the documents, the
tribunal will then need to consider whether adverse inferences are appropriate.
It has been suggested that adverse inferences cannot be drawn if a third party
refuses to produce documents. (109) This may depend on whether a direct power
exists. If not, the third party can have many reasons for not divulging the
information and it would be rare if the circumstances supported an adverse
inference. Even if there is a direct power, there is still a logical need to connect
the third party's failure to produce, with the proper treatment of the party
against whom the adverse inference is sought to be drawn. It is not the third
party being sued or bringing an action so it is not adverse inferences against that
person that matter. One suggestion is that a relevant factor might be the extent
to which the third party would share economic benefits or costs depending on
the outcome of the arbitral case. (110) The logic of an adverse inference in such a
scenario is that the third party would have been expected to divulge the
document if it would favour its commercial interests. Refusing to do so allows for
the adverse inference. While the logic is tenable, other aspects such as control
and quantum of documents may be determinative as to the reasons for refusal. If
there is no control over the third party, there is no primary obligation in relation
to document production. The third party could simply say that it wishes to have
no involvement where it is not subject to the powers of an arbitral party. Once
again, as soon as a tenable alternative reason for non-production exists, the
harder it is to justify an adverse inference against a separate person with no
control over the third party. Nevertheless, there should not be a negative blanket
rule either. An evasive third party witness can legitimately justify a tribunal in
forming an adverse view about the case the witness is brought in support of.
There must at least be a possibility that in some circumstances where documents
are not produced, there could be sufficient confidence that an adverse inference
against a party is merited.

page "855"

11.7. Grounds for Refusing Production

Section 10.16 above dealt with admissibility generally. It was noted that an
international arbitration will not apply common law exclusionary rules of
evidence for such matters as hearsay. Hence all documents are presumed to be
admissible and it is left for the tribunal to determine what weight if any to give to
documents whose veracity, relevance or persuasiveness may be questionable for
a range of reasons. Nevertheless, there will be circumstances where a party is
entitled to refuse production. The IBA Rules of Evidence 2010 have sought to
clarify and constrain broad discretionary powers to demand production of
documents. Article 9.2 outlines the grounds upon which an objection may be
made to documentary or other evidence. The same principles may logically apply
to blanket refusals to testify or refusals to answer specific questions when
testifying. One ICSID tribunal previously enumerated the considerations it had in
mind when deciding whether to order production of evidence. These included:
‘the necessity of the requests made to the point the requesting party wishes to
support, the relevance and likely merit of the point the requesting party seeks to
support, the cost and burden of the request on the (party) and the question of
how the request may be specified so as to both fulfil legitimate requests by a
party while not allowing inquiries that are an abuse of process’. (111)

Article 9.1 of the IBA Rules gives the tribunal a broad discretion as to
admissibility of evidence. Article 9.2 begins by indicating that the tribunal ‘shall,
at the request of a Party or on its own motion, exclude from evidence or
production any Document, Statement, oral testimony or inspection’ based on any
one of seven stipulated reasons. A party against whom an application is made
may also argue that the documents are not in its possession, custody or control.
If the parties cannot agree on a document production request, the tribunal will
make a ruling on the propriety of the request and the validity of any objections.
(112) The Article 9.2 grounds are:

(a) lack of sufficient relevance to the case or materiality to its outcome;

(b) legal impediment or privilege under the legal or ethical rules determined by
the Arbitral Tribunal to be applicable;
(c) unreasonable burden to produce the requested evidence;
(d) loss or destruction of the Document that has been shown with reasonable
likelihood to have occurred;
(e) grounds of commercial or technical confidentiality that the Arbitral Tribunal
determines to be compelling; page "856"
(f) grounds of special political or institutional sensitivity (including evidence
that has been classified as secret by a government or a public international
institution) that the Arbitral Tribunal determines to be compelling; or
(g) considerations of procedural economy, proportionality, fairness or equality
of the Parties that the Arbitral Tribunal determines to be compelling. (113)

As noted above, if the right to document production is contractual, or supported

by another express disclosure obligation, then the exempting elements of the IBA
Rules would not apply. (114) The party refusing production would have to attack
or seek to qualify the contractual or similar right itself. (115) In these cases,
production is not a matter for the unconstrained discretion of the tribunal, but is
a matter for the resolution by the tribunal of this disputed substantive
entitlement. At times there might be a procedural document production request
about ancillary documents relevant to such a substantive claim.

The principles under these exempting categories are considered in the sections
below. Some general observations are first made. On plain meaning, the
tribunal's obligations under Article 9.2 of the IBA Rules would be mandatory,
although of course the IBA Rules are themselves not binding unless expressly
selected by the page "857" parties. In any event, even if mandatory, the
articulated reasons include degree-based criteria where judgment must
inevitably be exercised. For example, the first ground is ‘lack of sufficient
relevance or materiality…’ (emphasis added). Second, there is some overlap
between the categories, particularly 9.2(c) and (g). These require overall
assessments that also build upon considerations of the degree of materiality. The
following sections examine the recognised categories of exceptions to production
entitlements.

11.7.1. Lack of Sufficient Relevance or Materiality

Article 3.3(b) of the IBA Rules of Evidence 2010 requires a statement as to how
the requested documents are relevant to the case and material to its outcome.
The structure of the IBA Rules thus implies that the burden is on the requesting
party to show both relevance and materiality. On some issues at least, the
standard of proof might not be very high at a point in time when the tribunal is
not fully abreast of all evidence to be submitted, including witness statements, as
the tribunal would not wish to exclude material that might logically be material.
The onus to successfully challenge might then shift to the responding party.

11.7.1.1. Relevance

It has been suggested that a tribunal looks for prima facie relevance and not
merely possible relevance, (116) although it may not always be easy to distinguish
between these concepts. In some cases whether a document is relevant or not
will depend on a legal determination yet to be made. For example, one party
might seek background documentation to a contract that contains an entire
agreement clause. If the party seeking the background documents has claimed
that context should be considered, then the documents are relevant to that claim.
In some cases a tribunal might make a preliminary ruling on the legal question as
to the impact of the entire agreement clause if that will significantly reduce the
cost of the document production process.

Relevance can relate to the claims the party is putting and also to the challenges
it knows it must meet from its opponent. This may also cover documents sought
in order to impeach the testimony of an opponent's witness in cross-
examination, although there is a separate question as to whether these must be
produced. (117)

page "858"

11.7.1.2. Materiality

Importantly, materiality is added as a further qualification to relevance, to avoid

wasteful duplication and the provision of unnecessary material. One articulation
is that materiality infers ‘that the arbitral tribunal must deem it necessary that
the document is needed as an element to allow a complete consideration as to
whether a factual allegation is true or not’. (118) To be ‘necessary’ to the case being
made does not mean that the case cannot be won without it, but that the case
cannot be presented optimally without it. (119) It must be related to the claims and
issues and consideration would need to be given to the matters sought to be
established by the document. (120) A material document has also been described
as one which ‘would have a tendency to influence the tribunal's determination of
issues in dispute’. (121) The UNCITRAL Draft Guidelines suggested that ‘the
document must be such that it likely contributes to the clarification of the case…’
(122) The CPR Protocol requires the applicant to show ‘a substantial,

demonstrable need in order to present its position’. (123) In a domestic discovery

contest under domestic litigation, an English court stated that ‘the document or
class of documents must be shown by the applicant to offer a real probability of
evidential materiality in the sense that… (it) can be expected to yield information
of substantial evidential materiality to the pleaded claim or as a defence to it… ’
(124) As with relevance, a tribunal has to err on the side of inclusion when it

cannot know the extent of the evidentiary record of both parties. Hence, in page
"859" assessing materiality, a tribunal ought to consider whether the assertion is
reasonable or prima facie valid and not require the contention to be on balance
correct.
It has also been suggested that a document needs to be material to an issue as to
which the requesting party carries the burden of proof. (125) There would be
problems in applying this as a blanket rule. It is certainly the case that in many
instances, a tribunal can rely on the party with the burden of proof having to
produce documents, otherwise they will fail. But to deny the opposing party the
opportunity to make targeted requests, removes an ability to easily identify
selective presentation by the party with the burden and makes it harder for
adverse inferences to be drawn where this has occurred. For example, if a
company director produced a personal file note but did not produce minutes of
directors meetings, the latter could be highly relevant to either corroborate or
contradict the notes. The important issue is that while burden of proof remains
on the party making a claim or defence, if that party presents enough
uncontradicted evidence, the onus of proof shifts to the other party. (126) That
party ought to be able to use limited production powers for that purpose, albeit
under strict tribunal supervision. In addition, a blanket rule would put undue
pressure on the tribunal itself if it is not clear whether the other party has
presented sufficient evidence to succeed under its burden. The opponent should
not have to rely on the tribunal holding that the burden was not satisfied. This
will commonly be a problem when incomplete evidence is presented on
quantum or when quantum claims are based on hypotheses or tabulations that
are potentially open to challenge.

The need for both relevance and materiality calls for appropriate case
management by the tribunal. The tribunal needs to be as reasonably familiar as
possible with the case as then presented to make decisions on such issues. A
tribunal will need to consider how much it asks the applicant to explain in
demonstrating materiality. To the extent that the tribunal itself considers this
question, the aim is not to prejudge the ultimate issue but only to deal fairly with
the production request. (127) A material document supporting what appears to be
a weak claim should still be produced.

11.7.2. Specificity

Generally speaking, a request must be sufficiently specific. Specificity can relate

to the identity of the document, its subject matter, its location, and/or the
identity of page "860" the person in possession or control. (128) It is ultimately
about identity of the document and its nature so all of these elements may come
into play. Article 3.3(a)(ii) of the IBA Rules of Evidence 2010 indicates that the
request shall contain a description of each document sufficient to identify it or ‘a
description in sufficient detail (including subject matter) of a narrow and specific
requested category of Documents that are reasonably believed to exist…’

Specificity is the key means to ensure that document production in arbitration

does not equate to broad-ranging common law style discovery. When allied to
materiality requirements, it should ensure that only targeted requests are made
and that the transaction costs of compliance are minimised. Specificity also
ensures that the party against whom the request has been made has a
meaningful basis to determine whether to comply or refuse to produce on the
basis of specific exempting grounds. Specificity also means that the process is not
dependent on how well counsel explains broad production discovery obligations
to clients and how well the clients are able to understand and apply them.
Conversely, if too much specificity is required this can encourage behaviour in
bad faith. For example if a request is for documents flowing directly between a
party and a third party it would not technically cover indirect communications.
However, a request may be rejected if it is too broad and seeks to identify a class
or category of documents without making sufficient effort to specifically identify
particular documents. (129)

The situation is usually easy with individual documents although the description
is more challenging when seeking a document whose existence is hypothesised
but not known. The greater challenge is in relation to categories. The aim is to
provide for a meaningful way to limit broad discovery. It would be unfortunate if
the only result was to encourage more elegant drafting by counsel to achieve the
same outcome. A request for all communications between the parties in relation
to the contract is obviously too broad. But is a request for ‘all letters and emails
passing between any officer or employee of either party that is part of the
process of negotiation and description of the contract’ sufficiently specific and is
there much in the way of meaningful difference? An answer cannot be given in a
vacuum and the important distinction is that in international arbitration, the
request will usually follow the tendering of the documents the other party
intends to rely upon. The more elaborate the other side's document presentation,
the easier it will be to circumscribe the gaps in a request. Conversely, the less
forthcoming the other party, the more a tribunal is likely to be supportive of a
broader categorisation. A tribunal should thus have at least a basic familiarity
with the kinds of documents presented when considering production contests.

page "861"

It has been suggested that where there is a suspicion that another party intends
to withhold documents to be used in cross-examination, this could be flushed out
with a request to produce any documents intended for such purposes. (130) Such a
request is broad and speculative if it is done as a blanket precaution but the first
question is whether the material should have been produced in any event. As
noted previously, the duty to produce documents upon which one intends to rely
should be interpreted broadly to encompass documents to be relied upon in
cross-examination, although some certainly take a contrary view. Even if the
duty to produce is accepted, a request for such material can easily look
speculative. Nevertheless there are practical reasons why it might be accepted
nonetheless. If there is no such material, it is easy to say so. If there is, it ought to
have been presented and the request is more justifiable, notwithstanding its
speculative nature.

As noted in section 11.6.8 above, specificity may be very different with regard to
e-documents as the specificity will often be first articulated as to a search term
which can be far more broad ranging and may need careful tribunal control. As
suggested in that section, the request for e-documents should also be specific as
to the particular document sought and not simply as to the search term. As noted
throughout, specificity would not be a gateway requirement if a party was
entitled under the contract to a broad array of documents, such as for audit
purposes. (131)

11.7.2.1. ‘Fishing’ Expeditions

While most would agree that the tenor of the IBA Rules is to prevent ‘fishing
expeditions’, less clear is the proper attitude to a request ‘aimed at obtaining
factual information as a basis for further allegations’ . (132) Document production
is often undertaken before there is a full understanding of exactly which
allegations will be contested. The requirement that the document be ‘relevant
and material to the outcome of the case’ has to be a test capable of application at
the time of the request. It must thus be in the context of the case as then
presented by the parties. As noted, some speak of the mere need to show prima
facie or likely relevance. (133) If the request is clearly within the ambit of an
allegation made, then it is not a fishing expedition, even if a party only has a
suspicion as to a document's existence and importance. Conversely, if a
document is requested with a view to page "862" considering amending the
claim to add new matters, then the relevant document is not material to the
outcome of the case in the form it has been put at the time the request is made.

While the distinction is important, if a tribunal is too concerned to limit

document production to existing allegations, it may simply encourage claimants
to draft multiple claims in the alternative. Hafter also points out that while no
arbitrator should allow fishing expeditions, an arbitrator subject to the IBA Rules
of Evidence 2010 would need to be careful that in rejecting such a request, he or
she has not violated the express production entitlements under Article 3.4. (134)
Skilful counsel can also frame their requests in a way that will appear sufficiently
specific to issues raised, but nevertheless are more motivated by the wish to
explore other possible but unknown issues.

Whether something is a fishing expedition can also depend on timing and other
evidence. For example, in an investment arbitration case, an investor might
suspect that the host State has treated domestic parties more favourably. They
may seek records of such treatment. Absent any other evidence, this is no more
than a suspicion or a hypothesis and could be alleged to be a fishing expedition.
Conversely, if there is evidence that the competitors are selling at prices that
could not be possible without some favourable treatment, document production
may be appropriate to help quantify it, given that competitors will obviously not
directly assist. The difficulty for a tribunal is that it will not wish to make a
presumptive decision about this logic as to competitive prices at a preliminary
hearing on document production, but it would also not wish to discount it.

11.7.3. Proof of Existence

The applicant does not need to know for a fact of the existence of the document,
but should have a reasonable belief as to its existence. In some cases this would
be based on reasonable presumptions, for example, that boards of directors have
met regularly and have kept minutes. Tribunals will not generally expect proof of
existence but will instead rely on specificity and leave it to the responding party
to argue as to existence or otherwise.

11.7.4. Possession, Custody and Control

A document production order must be made against a party. It would only be

reasonable to do so if that party had possession, custody or control over the page
"863" document. (135) Article 3.3(c) of the IBA Rules of Evidence 2010 also
requires a statement with the request that the documents are not in the
possession, custody or control of the requesting party or if they are, a statement
of the reasons why it would be unreasonably burdensome for that party to
produce the documents. This must be coupled with a statement of the reasons
why the requesting party assumes the documents are in the possession, custody
or control of another party.

The notion of control needs to be further considered for two reasons, first as to
when orders ought to be made and second, whether it is reasonable to draw
adverse inferences if a request is refused. The less that a party to the proceedings
truly has control over a third party who has possession of a document, the less
an adverse inference would be justifiable. The concept of control should be
interpreted in a practical and realistic fashion. Is the relationship between the
party against whom the order is sought and the person in possession of the
document such that the former could reasonably be expected to gain access from
the latter? That should include situations where the party against whom the
application is made could have power over a third party, either through a
corporate group or some contractual or other right to gain access to the
document. (136) An ICC tribunal has considered that this would be satisfied if a
party controls another entity or is controlled by another entity or both are part
of the same group. (137) It certainly makes sense to consider that there is effective
control where the arbitral party controls the entity in possession of the
document. The justification is less clear when the arbitral party is a subsidiary of
another entity as the power relationship is reversed, although the logic could be
that a simple request would normally be responded to favourably within a
corporate group.

At times a tribunal will be faced with burden and standard of proof issues where
a party against whom a request is made alleges that it cannot find the relevant
document. A tribunal might need to be confident that the documents are in the
party's custody or were deliberately destroyed before drawing adverse
inferences. It is also important to understand that if there is to be any debate
about the existence of a document, a tribunal would need actual evidence and
not simply submissions from counsel. (138) A tribunal would generally expect a
party to have taken reasonable or even best efforts to obtain documents from
those over whom it has some control. (139)

page "864"

One problem will be if a party wishes to seek documents that are in both parties'
possession, simply to see if there were any annotations on the opponent's
version. If there is no reasonable basis for this being so, this would appear to be a
fishing expedition and would also fall foul of the lack of possession requirement.
(140) Such a document should have been produced by the other party in any event.

If it was not produced, a specific request might be made on this basis. Another
scenario is whether the other party has access to a document. An example would
be an acceptance of an offer that the offer or alleges it never received. A
production request of this nature ought to be permissible, although it is unlikely
that it will be fruitful if the document was not tendered with the defence.

11.7.5. Non-existent Documents and Duties to Create

Generally, duties of disclosure relate to existing documents. However, in some

cases, a tribunal might wish to call on a party to create a document encapsulating
key information. This is not document production per se but instead could be
considered under a tribunal's broad discretionary powers. An example would be
a spreadsheet incorporating computer data that would not itself be easily
evaluated by the tribunal. The tribunal's discretionary powers ought to be seen
as broad enough to allow for such a request, although countervailing
considerations might apply if compliance would be extremely expensive and
time-consuming. The same power would not arise as a result of an individual
party's production rights. A party would also be able to rely on similar grounds
for rejecting a tribunal request to create a document as it would have if the
document was already in existence.

11.7.6. Loss or Destruction of the Document that has been Reasonably Shown
to have Occurred

As noted in section 10.4.5 above, it is impossible to conclusively prove a negative.

Loss or destruction must be shown with reasonable likelihood. Where
documents are no longer in existence, the tribunal will need to consider whether
there were valid grounds for their destruction. If that is not the case but the
documents are no longer available, adverse inferences may be appropriate. Such
inferences can only arise if there is no more compelling explanation of their loss
or destruction than a conscious attempt to destroy adverse evidence. For
example, if the tribunal accepts that the corporation accidentally threw out
archived records for the entire business during a relocation exercise in violation
of a statutory retention obligation, that says nothing about the contents of the
material. Section 11.6.8.2 above also made page "865" the observation that lost
or destroyed electronic information can often be retrieved and can itself be the
subject of a valid production request.

11.7.7. Unreasonable Burden to Produce the Requested Evidence

A tribunal will need to balance relevance and materiality on the one hand with
the possible burdensome nature of requests on the other. This criterion on its
own implies that relevant and material evidence can be validly rejected if a
requirement to produce would constitute an unreasonable burden on the party
against whom the order is sought. In the easy cases where this criterion is
applied, it is effectively otiose as the documents could be rejected on other
grounds. An example would be documents which are not material because other
documents are sufficient to prove the point. Article 3.12(c) of the IBA Rules of
Evidence 2010 now indicates that a party is not obliged to produce multiple
copies of documents which are essentially identical unless the tribunal decides
otherwise. If an applicant is merely seeking unduly repetitive documents, that
should fail the materiality test in any event. Thus, the difficulty with the notion of
an unreasonable burden is to determine how it might apply to exclude truly
material evidence. It has been said that ‘(t)he efforts required… should… be
proportionate to the evidentiary value of the document’. (141) The CPR Protocol
indicates that ‘(a)rbitrators should carefully balance the likely value of
documents requested against the costs and burdens, both financial and temporal,
involved in producing the documents or information requested. Where the costs
and burdens of disclosure requests are likely to be substantial in comparison to
the amount in dispute or the need for the information to aid in resolving the
dispute, the tribunal should ordinarily deny such request’. (142) But this does not
indicate to a tribunal exactly why material information should not be ordered in
some cases at least. In addition, how is this to be measured? How does a tribunal
make a determination without seeing the evidence? At that stage of the
proceedings, the costs may be clear but the relative materiality may not.

The more challenging situation will be where a party might concede that the
evidence is material but still argues it is unreasonably burdensome. A balancing
exercise of this nature would in theory need to consider the degree of materiality
but this would be extremely difficult for the tribunal to do without seeing the
material and thinking carefully about the issues in the case. In some instances,
materiality could be addressed in relation to the type of evidence involved. For
example, a party might argue that the relevant document is merely hearsay and
is of no evidentiary value.

page "866"

The degree of document production expected might take into consideration the
amount in dispute, given that time to collect documents together with time to
analyse, synthesise and deal with these at an oral hearing, all add significantly to
the cost of arbitration. Indeed in many cases, the total costs relating to
documentary material can constitute the major single cost item. Yet it would be
difficult to restrict documents on this basis that are otherwise material. At the
very least, in such circumstances a tribunal can warn the parties of the potential
expense problem and seek agreement about the most appropriate narrow class
of documents that may reasonably resolve the dispute.

If the key claim as to the unreasonableness of the burden is the cost, a number of
permutations may impact upon a tribunal's determination. First, the requesting
party may offer to pay the cost as a condition of its entitlement. Some
institutional rules or guides expressly indicate that a tribunal might condition
the request on the payment of part or all of the costs. (143) General discretionary
powers might be argued to cover this in any event, although a more restrictive
costs model in the lex arbitri or rules might suggest the contrary. (144) Less clear is
whether a tribunal might contemplate as a relevant factor that if the party
against whom the request is made succeeds, it may recover the costs in any
event. If it does not succeed, the concern for its costs may be less compelling. One
problem is that full costs indemnity does not uniformly apply.

In some cases electronic production could be far easier, particularly where the
provider is not required to discretely present electronic documents but can
instead provide an overly comprehensive disc with an effective search
mechanism. If it is not clear how relevant such material might be, a tribunal
might call for samples to be evaluated before a more time-consuming and
comprehensive production order is made. (145)

One aspect of an unreasonable burden would be if the applicant could access the
documents in other ways if it was reasonable for them to do so. (146) This would
be particularly so if the documents were in the public domain and equally
available to both parties, (147) although this scenario could be dealt with on
custody and control grounds as well. A tribunal would also concern itself with
the relative ease of obtaining public documents. It would generally be easier for a
State party to an page "867" arbitration to produce its own documents than a
private individual. That may not be so in countries with broad-ranging freedom
of information laws.

A test as to the unreasonably burdensome nature of the request might look only
at the difficulty of obtaining material or might also consider the impact on the
later proceedings. A tribunal could leave it to the later stage to determine what
amount of material should be submitted and reviewed by experts and tribunal
members. In some cases a tribunal might defer a final ruling on materiality on
the basis of efficiency, for example where it wishes to first hear argument as to
the legal basis of the claim and where it will only call for documentary evidence if
a potential legal entitlement is made out.

Another possible aspect of the reasonableness or otherwise of a request is the

legitimate expectations of the parties. For example, if both parties come from
civilian backgrounds, could the requestee argue that neither expected a broad-
ranging request for document production? Conversely, could the applicant argue
that consent to arbitration is consent to the emerging compromise practice of
generally allowing for production but limiting the extent? Once again there might
also be different attitudes to requests against claimants or respondents
notwithstanding the obligation as to equal treatment. A claimant who knows that
the only way the respondent can test its claim is through such a request, might
legitimately expect such an order flowing from their choice to bring an action. A
respondent might assert that consent to arbitration did not imply consent to
processes whose cost outweighs the amount in dispute and aim to assist
claimants interfering with the liberty of respondents. Another aspect of
reasonable expectation would relate to the type of dispute. In some
circumstances, the parties could only expect that a reasonably extensive
documentary analysis would be the only way to deal with the very kind of
dispute they were contemplating when selecting arbitration as the dispute
settlement method.

Where investment arbitration is concerned and a State is a party, other

arguments may come to the fore. On the one hand the State may successfully or
otherwise seek to argue national security or other sovereign immunities while
on the other hand, principles of freedom of information and transparency might
argue for an even broader duty to respond to production requests. (148)

page "868"

11.7.8. Legal Impediment and Grounds of Commercial or Technical

Confidentiality that the Arbitral Tribunal Determines to be Compelling

Broader issues of confidentiality were discussed in section 10.16.7 above and

need not be repeated. The arguments there apply equally to the issues with this
criterion as applied in the context of document production contests. Another
important consideration is to determine what is meant by the requirements that
it be ‘compelling’. At the very least, this suggests a presumption in favour of
disclosure, absent strong reasons to the contrary. Section 11.8.11 below deals
with the special problem facing a tribunal in seeking to make decisions on
confidentiality claims. The significant problem is that if a tribunal views a
document to determine whether it is truly confidential and then agrees that it is,
it has already seen the document.

11.7.9. Privilege under the Legal or Ethical Rules Determined by the Arbitral
Tribunal to be Applicable

Section 10.17 above dealt with privilege generally. The arguments there also
apply equally to the issues where document production contests are concerned
and do not need to be repeated.

11.7.10. Grounds of Special Political or Institutional Sensitivity (Including

Evidence that has been Classified as Secret by a Government or a Public
International Institution) that the Arbitral Tribunal Determines to be
Compelling

This will typically arise with investment arbitration and may call for tribunals to
make sensitive determinations as to governmental claims as to exemption. This
category was in part included because international political organisations did
not consider that the concept of commercial and technical confidentiality would
necessarily apply to them, (149) not being commercial entities.

A claim of public interest immunity was rejected as not being a general principle
of law or an identified matter of public international law in Biwater Gauff
(Tanzania)Ltd v United Republic of Tanzania. (150) The ICJ accepted an assertion of
naval secrecy as a reason for refusing to produce naval orders in the Corfu
Channel case. (151) No orders were made in relation to redacted material on the
basis page "869" of alleged military secrets in the Genocide case. (152) Importantly,
it is for the tribunal to make the final determination, although in many cases it
will simply be refused access to the relevant material. In appropriate cases,
adverse inferences may be drawn.
11.7.11. Considerations of Procedural Economy, Proportionality, Fairness or
Equality of the Parties that the Arbitral Tribunal Determines to be
Compelling

Article 9.2(g) of the IBA Rules of Evidence 2010 allows for exclusion of evidence
on the basis of procedural economy and proportionality as well as fairness
and/or equality of the parties where these considerations are considered to be
compelling. This reaffirms the overall need to promote fairness and efficiency
and contemplates that trade-offs will at times need to be made. It also overlaps
heavily with the criterion of an unreasonable burden and contemplates that the
balancing exercise will take into account the circumstances of both parties. By
bolstering the entitlement of a tribunal to promote efficiency of evidentiary
material, this should be an added barrier to claims of lack of due process where
restrictions are imposed, at least where the parties have selected these Rules.
Having said that, parties are always entitled to make such claims and the
requirement that the consideration be ‘compelling’ does not give much
assistance where clearly material evidence is excluded by a tribunal not wishing
to review voluminous documentation. Furthermore, the criterion also calls for
fairness and equality, which in many cases will be the basis of a due process type
determination. An example is equalising rules of privilege where parties come
from legal systems with very differing perspectives. (153) The concept of
‘procedural economy’ may also be uncertain and lead to inconsistent
applications. (154)

A number of other factors would also be relevant to the exercise of the discretion
to direct production of documents. Gabrielle Kaufmann-Kohler suggests
consideration of the origin and expectations of the parties, which would include
their familiarity with discovery rights, whether they would have expected
disclosure when they entered into the arbitration agreement or conversely,
whether this would be a shock to them. (155) Another approach to the exercise of
discretions is to seek to ensure that the time and expense of production is
proportional to the anticipated usefulness, although as William Park points out,
such a cost/benefit analysis will in part depend on what one perceives discovery
to normally be used page "870" for. (156) As noted above, it is also hard to apply as
a test before all other evidence is heard and considered.

11.8. Rulings on Contested Requests and Document Management by

Tribunals

If a party against whom a request has been made refuses the request and gives
reasons for doing so, it may be necessary for the tribunal to rule on the validity of
each party's position. Procedural determinations on document request
applications are not normally awards and hence are not able to be challenged.
This may be varied by the lex arbitri. A tribunal may seek to make a
determination as an interim award which would make it arguably enforceable
under the New York Convention, (157) although there could be a question whether
it is truly an award in the sense of being a final determination on an aspect of the
claim. (158) It was noted above that if the entitlement to information is under
substantive rights, such an award is possible as it is an award on the merits of a
claim as to that right.

Because document production requests will tend to be undertaken a long time

before the final hearing, arbitrators need to take care in being too prescriptive
where they might not be fully familiar with all key issues in the case and the
conflicting evidence that will be presented on each factual question. For this
reason, it is particularly important to involve the parties as much as possible in
the process with a view to finding a consensus position if at all possible. The IBA
Rules of Evidence 2010 now allow for the possibility of a tribunal ordering
consultations between the parties on objections to document production before
any ruling by a tribunal. (159) The tribunal might utilise a management meeting
with counsel to try and agree on limits to document requests. Counsel might
have side meetings on such occasions to see if they can agree among themselves.
(160)

One technique to streamline such management conferences and prepare for any
necessary rulings is to use a process described as the Redfern Schedule. (161)
Redfern and Hunter describe the process established by one of the authors to be
utilised after the parties have conferred on the question of document production.
The suggested schedule is a document comprising four columns to be completed
as page "871" quickly as possible. The first column is for the requesting party to
set out a brief description of the document sought in sufficient detail so that it
may be identified, or a sufficiently detailed description of a narrow and specific
category of documents reasonably believed to exist. In the second column, that
party explains why the requested documents are relevant and material to the
outcome of the arbitration. In the third column, the party whose documents are
requested indicates its attitude to the request and the grounds for objection
where it does not wish to produce any document. The fourth column is left blank
and is for the tribunal to indicate its decision in relation to that document or
category of documents as the case may be. Some parties call for a fifth column in
the Redfern Schedule, allowing the requesting party to comment on objections. If
a variant of the Redfern table is utilised, the tribunal should indicate who
prepares this and whether the preceding document requests passing between
the parties and the documents submitted in response should be copied to the
tribunal. (162) A tribunal can call for more information before making a decision
and in some cases may call a meeting of the parties, although the aim if possible,
is to manage the process without the need for a further meeting or formal
hearing. (163) A meeting could either aim to reach consensus or elaborate on the
issues so that the tribunal is better informed before making a final
determination. If after a sufficient attempt to find consensus, there are still
disputed categories of documents, the tribunal might move to a hearing format,
receive submissions from the parties and make a procedural ruling as to
document production. If a tribunal ruling is required, there may then be
questions of burden of proof on the party requesting the document to show its
existence, relevance and materiality and that party's attempt to obtain the
documents by other means, and on the other hand, a burden on the party against
whom discovery is sought, to show that one of the grounds for refusal applies.

The next issue is the weight of evidence in support of such determinations. While
it is natural to invite a tribunal to limit documentary excess by a careful
consideration of relevance, materiality and specificity and to do so at the earliest
possible stage, leading arbitrators often do not have the time to examine all
documents carefully to make the most informed decisions in relation to what will
ultimately be the key issues. In such circumstances they are likely to tend
towards leniency. Even if an arbitrator has the time and willingness to view the
documentation, at an early stage these are merely likely to be requests in
support of general contentions in preliminary submissions. An arbitrator cannot
confidently determine how much documentary evidence might be reasonable to
corroborate or refute witness testimony not then known. Too rigorous
restrictions might significantly impact upon later decisions as to the
preponderance of evidence.

page "872"

A particularly unsatisfactory situation is if parties make overly broad requests

that are then successfully challenged, with no indication as to whether there will
be a further opportunity to refine the application. At that point in time, a tribunal
could either take the view that a proper opportunity was afforded but not well
taken, denying any subsequent applications outside of the procedural timetable,
or conversely, could revise the procedural timetable and allow refined requests,
perhaps with appropriate costs orders. Neither approach is inherently optimal
and it would be most undesirable for the parties to have no prior indication of
the likely approach. It would be far preferable for the tribunal to indicate at the
outset at least a presumption as to which of these two approaches will be
employed. Parties who are forewarned that they have only one chance at
identifying sufficiently specific and material documents are more likely to
construct their requests accordingly. Similarly, parties who know that there will
be costs implications of refinement applications may be similarly encouraged. A
typical problem will be unduly broad requests that nevertheless seem material.
For example, to ask for all documents in the contract negotiations must be
inappropriate as most will have passed between the parties and be in the
requesting party's possession already. A tribunal can ask the party to revise the
request in writing or at a hearing, perhaps with an order that costs go against
them for the earlier inadequate request, or could reject the request entirely.
Arbitrators may differ on this issue. Some would say that it is still desirable to
decide on commercial information and not burdens and standards of proof, so a
follow up opportunity should be allowed. Others might argue that parties
consented to an impliedly efficient process which entails parties doing things
properly at the first opportunity and suffering the consequences if they do not.
One argument against the latter view is that until such time as more
practitioners become familiar with provisions such as the IBA Rules on Evidence,
too many problematic requests can be expected. Civilian lawyers may simply be
unfamiliar with the processes. Common lawyers may wrongly use broad
language from litigation precedents.

Another issue to consider is the potential for challenges to the award on due
process grounds to the effect that denying a right to revise may be a denial of due
process when there is plenty of time before a hearing and when costs orders can
deal with adverse effects on the other party. Even if respected jurisdictions
would not take this view, the less there is an opportunity for such arguments, the
better. At the very least, if a tribunal intends to deny a right to amend overly
broad requests, this should be made clear at the outset. Such a warning may also
help counsel devote enough time to the request to have it in appropriate form in
the first place. A related issue is the willingness of a tribunal to review a
production request prior to receiving objections from the other party. In many
cases, particularly with inexperienced common law counsel, production requests
are likely to be overly broad. It may be far more efficient for a tribunal to indicate
this and demand a revised form, rather than making the responding party go
through the laborious process of objecting to each item. There is no reason why a
tribunal could not give recommendations as to how to better specify the
document categories, although there should be no obligation to do so. Once again
this can depend on attitudes to page "873" proactivity. If the tribunal knows
which documents it will require, it can specify this in any event, so indicating this
to counsel is not ipso facto undue assistance. In addition, if the responding party
and perhaps the tribunal will ultimately need to make decisions as to relevance
and materiality, it seems reasonable to demand an explanation of these factors in
the request itself.

A tribunal direction framed around the principles enunciated in the IBA Rules of
Evidence 2010 might include observations along the following lines:

1. Requests for document production may be made in written form by… … … …

(insert date).
2. Where individual documents are requested the description must be
sufficiently specific to identify it.
3. If a category of documents is being requested this must also be narrow and
specific and must utilise a description in sufficient detail, including as to
subject matter, of the category requested.
4. In each case the request should indicate why the requested documents are
material to the case of the requesting party. That party should relate the
request to the particular claims, defences and/or counterclaims where
materiality is asserted.
5. In considering relevance and materiality, the request should ensure that it
does not cover documents already produced in submissions and explains how
the request relates to such documents.
6. The request must indicate the reasons why the requestor believes that the
relevant document or documents are in the possession, custody or control of
the party against whom the request is made. In addition the requesting party
must attest to those documents not being in its own possession, custody or
control or if they are, give reasons why it would be more appropriate for the
other party or parties to produce the documents.
7. Where e-documents are concerned the request must address all of the issues
of specificity, materiality and possession outlined above and in addition must
indicate appropriate search terms or file names or the like by which an
appropriate search ought to be made.
Hanotiau suggests the following as a possible text for the tribunal's first
procedural order:

Production of documents

1) The parties may request documents from each other at any time during the
proceedings. Correspondence and documents exchanged in the course of this
process should not be sent to the Arbitral Tribunal.
2) If any of the requests referred to in section 1) are not satisfied, the parties
may file requests for document production with the Arbitral Tribunal. These
requests, which may be made only after the parties have exchanged their
initial briefs as provided in [the relevant provision of the procedural
calendar], shall be presented as a joint submission in the form of a table
(commonly referred to as a ‘Redfern Schedule’) comprising two sections:
page "874"
i) the Claimant's request(s) for the production of documents; and
ii) the Respondent's request(s) for the production of documents.

The joint submission shall be presented in four columns, as follows:

– first column: identification of the document(s) or the category or

categories of document(s) that have been requested;
– second column: short presentation of the reasons for each request;
– third column: a summary of the objections by the other party to the
production of the requested document(s);
– fourth column: left blank for the decision of the Arbitral Tribunal.
3) When making its decision, the Arbitral Tribunal shall be guided by Articles 3
and 9 of the IBA Rules on the Taking of Evidence in International Commercial
Arbitration. On this basis, the Tribunal considers that the following standards
should guide its reasoning:
(i) the request for production must identify each document or specific
category of documents sought with precision;
(ii) the request must establish the relevance and materiality of each
document or each specific category of documents sought in such a way
that the other party and the Arbitral Tribunal are able to refer to factual
allegations in the submissions filed by the parties to date. This shall not
prevent a party from referring to upcoming factual allegations
(subsequent memorials) provided such factual allegations are made or
at least summarized in the document production request. In other
words, the requesting party must make it clear with reasonable
particularity what facts/allegations each document (or category of
documents) sought is intended to establish;
(iii) the Arbitral Tribunal will only order the production of documents or
categories of documents if they exist and are within the possession,
power, custody or control of the other party. If contested, the requesting
 party must show that it is likely that the document is indeed within the
possession, power, custody or control of the other party;
(iv) If necessary, the Arbitral Tribunal shall also balance the request for
document production against the legitimate interests of the other party,
including any applicable privileges, unreasonable burden and the need
to safeguard confidentiality, while having regard to all the surrounding
circumstances.
4) If, subsequent to the document production requests referred to above,
additional documents are needed by a party, leave should first be requested
from the Arbitral Tribunal. If leave is granted, the procedure detailed above
in sections 1 to 3 will be applicable.
5) The Arbitral Tribunal may also, at any time, on its own initiative, direct the
parties to file any documentary evidence in their possession or under their
control, which the Arbitral Tribunal considers to be relevant and material.
6) If any documentary evidence which the Arbitral Tribunal orders a party to
produce or file contains privileged and proprietary information or trade page
"875" secrets, that party shall so indicate to the Arbitral Tribunal and to the
other party. In that case, after consultation with the parties, the Arbitral
Tribunal shall determine what measures should be taken to respect the
proprietary nature of the information while allowing as much evidence as
possible to be produced for the purpose of the arbitration proceedings. (164)

Ashford has presented a number of models for document production

determinations by a tribunal. (165) Some of the matters he suggests as worthy of
inclusion for consideration include:

1. A time limit for production.

2. An explanation of the notion of control, which he suggests as including
‘physical possession, right to possession, and right to inspect or take copies’.
3. An indication whether documents thought to be not admissible should be
identified.
4. An indication whether documents adverse to interests should be produced.
5. An indication whether it is to cover internal documents, including those of
internal counsel.
6. An indication whether it is to cover documents used solely for cross-
examination.
7. An indication of how multiple copies of the same document are to be treated.
8. An indication of the use which can be made of documents not produced,
(typically that they cannot be relied upon without permission of the tribunal).
9. An indication as to the duties of confidentiality over documents produced by
the other party in the context of use in arbitration and collateral use.

Directions could also be given in relation to e-documents, methods of search and

duplication with hard copies. A tribunal may also wish to consider whether there
should be a document inspection stage so that the requester can determine
which documents actually need to be copied and tendered. Normally this would
not be necessary as the requester has already had to define a narrow class.
Nevertheless, in large matters, parties working collaboratively and in good faith
could use such a process to try and concentrate on a core set of documents to be
used in the proceedings. A tribunal will also need to consider supplementary
requests for document production where a party believes that inadequate
disclosure has been made. Here the important question will be the reasonable
basis for that belief and how a tribunal will make such a determination.

page "876"

11.8.1. Decisions as to Confidentiality

Section 11.7.8 above noted that a contentious issue is the way to deal with
confidentiality exceptions. At times a tribunal may need to view the documents
to pass judgment. This may prove problematic in cases where one party claims
confidentiality and asserts that the tribunal would be unable to treat the parties
fairly and with equality after viewing and then rejecting a document on this
basis. In some cases the concern would not be viewing by the tribunal, but
instead allowing the document to be viewed by the other party in a contested
application for production. Production to the tribunal or a tribunal chair alone
might be one option although some might argue that this is a breach of due
process and a ground for setting aside. (166) It might not be possible to waive due
process by even agreeing that the tribunal alone sees the document. (167)
Fouchard suggests however that if consent is given in full knowledge of the facts
this should constitute a waiver of their right to contest the award on those
grounds. (168)

One device commonly used in American litigation is to permit confidential

material to be seen by legal counsel on the condition that it is not disclosed to
any other person including counsel's client. There are advantages and
disadvantages of such an approach. The advantage is that relevant material can
still be included in the development of a case strategy notwithstanding its
confidentiality. An experienced legal practitioner even without instructions
should be able to determine how to best utilise the material. A contrary
argument is that in many cases, without being able to seek such instructions, that
evidence cannot be usefully challenged. In addition, such systems provide an
incentive to use lawyers rather than in-house counsel or self-representation in
some circumstances at least, although this may not be a realistic concern in
significant sized arbitrations. It is arguable that the parties' agreement to
arbitrate could be said to imply an entitlement of a tribunal to consider all
relevant information between parties who have chosen to do business together,
including that which is claimed to be confidential. In extreme cases a tribunal
might order production to itself and counsel alone and not other parties or
witnesses in return for counsel giving undertakings as to confidentiality. There
would be problems in international arbitration with lack of enforceability of such
undertakings.

The IBA Rules of Evidence indicate that in exceptional circumstances, where the
tribunal determines that it should not review the document it may appoint an
independent and impartial expert who is to be bound by confidentiality to
review page "877" the document and report on the objection. If time and cost
factors and the like are compelling, the tribunal may review the document itself.
The Commentary of the Sub-Committee indicates that the expert ‘need not
necessarily’ be appointed under Article 6 of the IBA Rules of Evidence,
presumably so that all Article 6 requirements need not be met, (169) such as an
entitlement to cross-examination of the expert. It would be impractical to allow a
document production expert to be cross-examined by the parties. The uncertain
comments suggest that opinions may differ on this issue. The expert would
provide a report but the tribunal is to make the final ruling as to validity. (170) An
independent expert was used to review confidentiality claims in Guyana v
Suriname. (171)

Hans van Houtte has postulated delegating document production challenges to ‘a

Document Production Master’ (DPM) in appropriate cases. A DPM would be an
impartial and independent person with appropriate expertise. (172) If a tribunal
uses an independent expert to determine confidentiality questions, it should
always be possible for the tribunal to ultimately decide that a document is
confidential even if the independent expert held to the contrary. Van Houtte
suggests that if the tribunal is to review the decision, it would make sense for a
deferential standard to apply, otherwise the DPM process would largely be a
waste. (173) It is not readily clear what form such an expert report would take in
order to allow the tribunal to make a truly independent decision without
separately viewing the document. If the tribunal looks at the document, it is
likely to make a de novo determination if it does not agree with the expert.

Without an express authorisation to do so, a tribunal might be alleged to have

delegated an adjudicatory power, although a respected supervisory court should
not support this approach. Nevertheless, agreement should be reached. Van
Houtte advocates that terms of reference should confirm confidentiality
undertakings by the parties and also be underwritten by the DPM. It would also
need to be clearly articulated that the expert can review the documents without
showing them to the other party. (174) Confidentiality agreements with such
experts may be problematic in terms of proof as to who divulged and as to proof
of loss and damage as a result. A tribunal would also need to be careful to ensure
that it does not cause an page "878" inadequate confidentiality document to be
drafted, leaving itself open to some future claims. Most importantly, the terms
should indicate whether the DPM's decisions are final or can be overridden by
the tribunal at the request of one of the parties. (175)

In some cases, a tribunal may order production but also give commensurate
orders as to confidentiality obligations. (176) Similarly, confidentiality issues can
lead to requests for interim measures. Article 3.13 of the IBA Rules of Evidence
2010 requires submitted or produced documents not otherwise in the public
domain to be kept confidential by the tribunal and the parties and only used in
connection with the arbitration. An exception is provided to the extent that
disclosure is required of a party to fulfil a legal duty, protect or pursue a legal
right, or enforce or challenge an award in bona fide legal proceedings before a
State court or other judicial authority. The requirement is without prejudice to
other confidentiality obligations in the arbitration.

11.8.2. Tribunal Response Where a Party Refuses to Produce or Conceals

Documents

An important aspect of document production rights and duties is a consideration

of the sanctions if any for failure to comply. Where there are no effective
sanctions, this might exacerbate the potential for guerrilla tactics and in turn
undermine the authority of the tribunal. Court powers of contempt, fines or
possible dismissal of action are not generally available to arbitrators. Arbitrators
do not generally have anything equivalent to the civilian astreinte order,
compelling the payment of money for each day of non-compliance with a
production order. Some national systems allow arbitrators to impose penalties.
(177) The Swedish Arbitration Act 1999 prohibits this. (178) French law permits the

arbitrators to issue penalties if necessary. (179)

Court assistance may be possible but depends on jurisdictional issues and

discretionary powers which will vary from country to country and may be
impacted on by the choice of seat and location of the documents and the
documents owners. For example, the Hong Kong Ordinance applies to awards,
orders and directions page "879" whether in or outside Hong Kong. (180) An
application may also be made under Hong Kong's Evidence Ordinance. (181)
Kaplan J in Astel-Peiniger held that Article 27 of the Model Law allowed for the
grant of subpoenas to produce documents. (182) In Transfield Philippines Inc a
Victorian Supreme Court judge allowed the subpoena powers under domestic
arbitration legislation to be used by an ICC arbitral tribunal with the seat in
Singapore but which sat in Melbourne, Australia. (183) Even if available, court
assistance might be impractical in terms of delay and added cost, particularly if
the delay in seeking court assistance suits the strategic interests of the party
refusing production. There are also problems as to whether a particular court
has practical control of people and documents and knowing in advance the likely
attitude of judges of that court.

Costs orders by a tribunal are possible but would in most cases, simply be
limited to the added cost to the other party of challenging the inadequacy of
document production. These are not likely to be high and hence would provide
little in the way of deterrence. Penal costs sanctions are not applied in
international arbitration. (184) However, the wording of Article 9.7 of the IBA
Rules of Evidence 2010 is broad enough to allow for wider costs implications
than simply in relation to the undesirable evidentiary behaviour. (185) A tribunal
also does not have the normal sanctions over counsel for failing to properly
advise parties as to their disclosure obligations. Control over counsel is
discussed in section 9.7. In any event, it would be difficult to draw adverse
inferences against a party simply because of inadequacies of counsel.

The IBA Rules of Evidence 2010 provide that if a party fails to produce requested
documents without satisfactory explanation, the tribunal may infer that such
documents would be adverse to the interests of that party. (186) Once again,
before adverse inferences may be drawn, there would need to be a sufficient
evidentiary basis for concluding that the document was available and hence
adverse, rather than some other reason, such as having been negligently mislaid.
In the latter event, the failure to produce is still unacceptable but the adverse
page "880" inference as to content is not valid. (187) Adverse inferences may also
be impractical where the other party and the tribunal are unaware of the
existence of a document. It makes sense for a tribunal to ask for a specific
explanation for a failure to follow a production order so that it has sufficient
material on which to determine whether adverse inferences would be
appropriate. (188) Born suggests that international tribunals are often ‘overly
hesitant’ in terms of drawing adverse inferences where production orders are
ignored or key evidence is otherwise not produced. (189)

In some circumstances, a failure to produce such a document might be grounds

for annulment of the award on the basis that the other party was denied due
process. For example, the US Federal Arbitration Act states that ‘refusing to hear
evidence pertinent and material to the controversy’ is grounds for a refusal to
enforce the award. (190) An arbitrator dealing with a contest as to the ambit of
production of documents knowing that enforcement would need to be sought in
the US could not ignore such a provision. That might be so even if both parties
come from civil law jurisdictions with the respondent having significant assets in
the US. The key word is ‘refusal’, which would suggest that the challenge would
be about rejection by a tribunal and not a failure to produce by a party, although
even that is unclear on that drafting. Where the tribunal is concerned, challenges
are potentially available either way. A challenge is possible on due process or
public policy grounds if a tribunal does not apply an exemption category
appropriately. Equity issues may also arise. In an extreme case, if one party is
confident that the other will not reciprocate in terms of document production,
the material may be directed to be given to the tribunal and only distributed
simultaneously. (191)

11.9. Document Management

11.9.1 . Logistical Management

The more efficiently documents can be presented, the more that there will be
time and cost savings. One issue is to avoid unnecessary duplication of relevant
documents which would otherwise need to be submitted by all parties. (192) One
page "881" suggestion is to have the parties agree on a common core set of
documents to be filed prior to individual submissions. (193) The parties might also
give careful consideration to the core documents that it would expect the
tribunal to be familiar with prior to any hearing. The more there is voluminous
material presented without any guidance, the less likely that busy arbitrators
will be able to be adequately prepared, no matter how well intentioned. (194)

A tribunal might wish to indicate how documents should best be collated and
perhaps bound. In cases with large amounts of documentary material, there
might need to be numbering or colour-coding systems to allow the tribunal,
counsel and witnesses to move back and forth efficiently among a large number
of folders. For example, colour coding could indicate the party which has
submitted the document. The tribunal can also give directions as to how
documents are to be referenced so there is commonality amongst all parties. (195)
It would be desirable to identify a numbering system as soon as possible so that
both sets of submissions and witness statements could all utilise the same
methodology. (196) In some cases it will be particularly useful to have documents
in appropriate chronological order and/or collected as to issue, rather than
having separate claimant and respondent lists. Where they are collated issue by
issue, there is still a need to differentiate between documents proffered by each
party. That could be by prefix or colour coded tabs. Where a numbering system is
utilised, care needs to be taken if documents are produced in different stages.
Whatever numbering system is originally utilised, it is appropriate to allow for
additions at a later stage. Charts and tables might also be used to systematise a
large volume of documents. A tribunal might give directions as to who is to
compile indices and lists. With large indexes, pages should have headers or other
ready means to identify which volume contains the actual document. In certain
kinds of disputes documents would be collated into particular kinds of working
files. An example would be a construction dispute where there may be separate
sets of site minutes or instructions. (197)

In some cases there will be significant savings of time and expense if the
documents are scanned and all persons are able to refer to documents on
computer screen. Other IT management processes may be utilised. (198) For
example, documents can be stored on secure websites accessible to all approved
persons. Another possibility is to highlight key parts of lengthy documents to
assist reference by the page "882" Tribunal and Counsel. (199) While there are
some impressive tools, a tribunal needs to be careful that suggested cost savings
for its own benefit do not occasion a higher amount of costs by counsel in
preparatory work.

A tribunal may need to intervene if it feels that one or both parties is seeking to
present a vastly larger body of documentary evidence than is necessary. A rough
rule of thumb may be to question why documents are included if they are not
likely to be addressed directly in written or oral submissions. It would
unnecessarily extend the time and inflate the cost if reams of technically relevant
but largely repetitive or ancillary materials are required to be read in detail by
the tribunal. As has been noted, it is not ‘the arbitrators' responsibility to find
their way through tons of paper but the obligation of the parties to indicate
exactly for which purpose and for which proof a specific document has been
submitted’. (200) Where a large number of documents are repetitive but still
necessary to show patterns, it might be important to try to agree on samples for
the tribunal's consideration. Where only parts of lengthy documents are
relevant, it is important to try and agree on appropriate extracts. The UNCITRAL
Notes suggest that in the case of voluminous and complicated documentary
evidence there may be value in it being presented in a report by a competent
person ‘in the form of summaries, tabulations, charts, extracts or samples. Such
presentation of evidence should be combined with arrangement to give the
interested party the opportunity to review the underlying data and the
methodology of preparing the report’. (201) It is important to be clear whether
such person is an expert whose representations are evidence or whether they
are merely tabulating demonstrative evidence. (202) In each case the tribunal
needs to ensure that it will be able to adequately assess the material in the face
of any shortcuts in their tabulation.

11.9.2. Translations

In most cases, key documents will need to be translated into the language of
arbitration if the languages are different. This will of course depend on the
language skills of the tribunal and counsel. Multi-lingual arbitrators need to be
careful in ensuring that their skills are sufficient to have a full understanding of a
foreign document or transcript as nuances can be vital. While parties might
separately organise translations of documents on which they intend to rely, it
may be preferable where this can be anticipated, to have them agree on a single
translator to promote consistency. In some cases it may not be necessary to
translate the entire page "883" document, although the tribunal would need to
ensure that other parts do not provide important context. There will also be
important issues of timing as translated documents need to be produced in
sufficient time to allow other parties to prepare. The IBA Rules of Evidence 2010
indicate that translations shall be submitted together with the originals and
marked as translations with the original language identified. This is to occur at
the document production stage as well. (203)

It is desirable to invite the parties to agree on the translations to be presented to

the tribunal. If that is not possible, the parties might be asked to present their
own translations of those parts where they cannot agree, perhaps with an
indication of the basis of the disagreement. This will assist the tribunal in
ultimately determining the proper meaning of the document in issue. Redfern
and Hunter suggest that even if there is no specific dispute about the accuracy of
a translation but there was no agreement either, it is advisable for the parties to
have it notarially certified. (204)

page "884"

1 Nudrat Najeed, ‘Good Faith and Due Process: Lessons from the Shari'ah’,
Arbitration International 20, no. 1 (2004): 97.
2 Some commentators speak of document ‘disclosure’. See Gary B. Born,
International Commercial Arbitration (The Hague: Kluwer Law International,
2009), 1821.
3 See the discussion in Yves Derains, ‘La pratique de l'administration de la

prevue dans l'arbitrage commercial international’, Revue de l'Arbitrage (2004):

789–790.
4 For example, disparities in presentation of documentary material can lead to

disparity in costs between the parties. This can ultimately pose policy challenges
for arbitrators in making final cost determinations where the indemnity rule
applies and in deciding whether loser pays should apply in full even in cases
where there is procedural inefficiency occasioned by the prevailing party.
5 These were discussed in sections 10.16.7 and 10.17.
6 Contemporaneous documents are to be distinguished from documents
prepared in contemplation of litigation.
7 Charles N. Brower, ‘Evidence before International Tribunals: The Need for

Some Standard Rules’, International Lawyer 28, no. 1 (1994): 54. See, e.g.,
Woodward-Clyde Consultants v. The Government of the Islamic Republic of Iran, 3
Iran-United States Claims Tribunal Reports 239, 249 (1983); Charles N. Brower,
‘The Anatomy of Fact-Finding Before International Tribunals: An Analysis and a
Proposal Concerning the Evaluation of Evidence’, in Fact-Finding before
International Tribunals: Eleventh Sokol Colloquium, ed. Richard B. Lillich
(Ardsley-on-Hudson: Transnational Publishers Inc., 1992), 150–151.
8 Bin Cheng, General Principles of Law as Applied by International Courts and

Tribunals, 3rd edn (Cambridge: Cambridge University Press, 2006), 318–319.

9 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 389.

10 Ibid.
11 Anne Veronique Schlaepfer & Philippe Bartsch, ‘A Few Reflections on the

Assessment of Evidence by International Arbitrators’, International Business Law

Journal 3 (2010): 210, n. 22.
12 Charles N. Brower, ‘Evidence before International Tribunals: The Need for

Some Standard Rules’, International Lawyer 28, no. 1 (1994): 55; Leonard &
Mavis Daley v. The Islamic Republic of Iran, 18 Iran-US Claims Tribunal Reports
232, 242 (1988).
13 Charles N. Brower, ‘Evidence Before International Tribunals: The Need for

Some Standard Rules’, International Lawyer 28, no. 1 (1994): 52.

14 IBA Rules on the Taking of Evidence in International Arbitration 2010 Art.

3.12(a). This Article leaves it for the tribunal to request originals to be presented
for inspection.
15 Rules requiring written statements to be accompanied by copies or at least

lists of all essential documents include the ICC rules, LCIA Rules and ICDR Rules.
UNCITRAL Rules are merely permissive in this regard. Document presentation is
also supported by the IBA Rules which designates production within the time
ordered by the tribunal. See IBA Rules on the Taking of Evidence in International
Arbitration 2010 Art. 3.
16 Von Mehren and Salomon make the point that even if this is permissible, if

disclosure is required in due course and it does not occur voluntarily at the
outset, the strategy may backfire and the party involved may be seen as evasive.
See George M. von Mehren & Claudia T. Salomon, ‘Submitting Evidence in an
International Arbitration: The Common Lawyer's Guide’, Journal of International
Arbitration 20, no. 3 (2003): 287.
17 Bernard Hanotiau, ‘Document Production in International Arbitration: A
Tentative Definition of “Best Practices”’, in Document Production in International
Arbitration, ICC International Court of Arbitration Bulletin: 2006 Special
Supplement, ed. ICC (Paris: ICC Publishing, 2006): 114; Alan Redfern et al., ‘The
Standards and Burden of Proof in International Arbitration’, Arbitration
International 10, no. 3 (1994): 344.
18 The ICDR Guidelines for Arbitrators Concerning Exchanges of Information

provide that the aim should be to balance ‘the goals of avoiding surprise,
promoting equality of treatment, and safeguarding each party's opportunity to
present its claims and defences fairly’. Concern for the other party's fair
opportunity adds to a disclosure obligation. Art. 1704.3(c) of the Belgian Judicial
Code allows for an arbitral award to be set aside if, after being made, ‘a
document or other piece of evidence is discovered which would have had a
decisive influence on the award and which was withheld through the act of the
opposing party’.
19 ICC Award No. 1434, Journal du droit international 103, (1976): 982.
20 William A Parker (USA) v. United Mexican States, Reports of International

Arbitral Awards 4 (1926): 39. See also Ali Z. Marossi, ‘The Necessity for Discovery
of Evidence in the Fact-Finding Process of International Tribunals’, Journal of
International Arbitration 26, no. 4 (2009): 515.
21 William A Parker (USA) v. United Mexican States, Reports of International

Arbitral Awards 4 (1926): 39.

22 See Stephen M. Schwebel, International Arbitration: Three Salient Problems

(Cambridge: Grotius, 1987), 218, citing Mixed Claims Commission, United States
and Germany, Opinions and Decisions in Sabotage Cases handed down 15 June
1939 and 30 October 1939 at 20.
23 ICC Case No. 1434, Journal du droit international 103 (1976), 982.
24 Kuijper makes this suggestion in the context of WTO litigation. See Pieter Jan

Kuijper, ‘The Appellate Body and the Facts’, in New Directions in International
Trade Law: Essays in Honour of John H. Jackson, ed. M. Bronckers & R. Quick (The
Hague: Kluwer Law International, 2000), 321. The conclusion does not
necessarily translate to arbitration, as Art. 13.1 of the WTO Dispute Settlement
Understanding expressly provides that ‘a Member should respond promptly and
fully…’ to Panel requests for information and documentation. The use of the term
‘should’ is problematic as it is less forceful than the mandatory ‘shall’.
25 See Georg von Segesser, ‘The IBA Rules on the Taking of Evidence in
International Arbitration: Revised version, adopted by the International Bar
Association on 29 May 2010’, ASA Bulletin 28, no. 4 (2010): 741.
26 In the WTO context, the Appellate Body has suggested that: ‘All parties

engaged in dispute settlement under the DSU must be fully forthcoming from the
very beginning both as to the claims involved in the dispute and as to the facts
relating to those claims. Claims must be stated clearly. Facts must be disclosed
freely.’ See India – Patent Protection for Pharmaceutical and Agricultural
Chemical Products WTO Doc. WT/DS50/AB/R, (1997), Report of the Appellate
Body, para. 94.
27 As to the latter see Preamble 3 of the IBA Rules of Evidence 2010. As to a

general duty of good faith see Art. 15.6 Swiss Rules.

28 Amy Cohen Klasener, ‘The Duty of Good Faith in the 2010 IBA Rules on the

Taking of Evidence in International Arbitration’, International Arbitration Law

Review 13, no. 5 (2010): 162.
29 Ibid.
30 IBA Rules on the Taking of Evidence in International Arbitration (1999) Art.

3.10 allows for possible later submission, but does not express this as a duty.
31 See, e.g., Khan v Armaguard [1994] 1 WLR 1204. It has even been suggested

that this could be an example of bad faith violation of the IBA Rules to withhold
documents so as to surprise witnesses. See Amy Cohen Klasener, ‘The Duty of
Good Faith in the 2010 IBA Rules on the Taking of Evidence in International
Arbitration’, International Arbitration Law Review 13, no. 5 (2010): 163.
32 Woods v. Martins Bank [1959] 1 QB 55, 60.
33 See, e.g., the reference to a 2010 UNCITRAL arbitration cited in Matthias

Scherer, ‘The Limits of the IBA Rules on the Taking of Evidence in International
Arbitration: Document Production Based on Contractual or Statutory Rights’,
International Arbitration Law Review 13, no. 5 (2010): 199.
34 Giorgio Bernini, ‘The Civil Law Approach to Discovery: A Comparative
Overview of the Taking of Evidence in the Anglo-American and Continental
Arbitration Systems’, in The Leading Arbitrators' Guide to International
Arbitration, 2nd edn, ed. L. Newman & R. Hill (Huntington: Juris Publishing,
2008), 270.
35 It was natural that the Iran-United States Claims Tribunal did not heavily

utilise discovery processes as this would have appeared as a systemic bias in

favour of claimants and their legal culture. See Klaus Lionnet, ‘Once Again: Is
Discovery of Documents Appropriate in International Arbitration?’, in Global
Reflections on International Law, Commerce and Dispute Resolution: Liber
Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al. (Paris: ICC
Publishing, 2005), 498, noting remarks of Robert Briner.
36 See, e.g., UK Ministry of Justice's Practice Directions Supplementing Part 31 of

the New Civil Procedure Rules in England and Wales – Practice Direction 31A
(Disclosure and Inspection) para. 5.4.
37 Kaufmann-Kohler notes broader discovery rights granted by § 142 of the

German Code of Civil Procedure. See Gabrielle Kaufmann-Kohler, ‘Globalization

of Arbitral Procedure’, Vanderbilt Journal of Transnational Law 36, no. 4 (2003):
1326. However, Trittmann and Kasolowsky, suggest that this is rarely used. See
Rolf Trittmann & Boris Kasolowsky, ‘Taking Evidence in Arbitration Proceedings
Between Common Law and Civil Law Traditions - The Development of a
European Hybrid Standard of Arbitration Proceedings’, University of New South
Wales Law Journal 31, no. 1 (2008): 336. The German Code of Civil Procedure,
§427 even allows for adverse inferences when documents are not produced.
Nevertheless, the power of civil adjudicators to call for documents is not the
same as common law discovery. A civilian judge will order a document that he or
she believes to be important. Common law discovery seeks to identify all
relevant information for counsel to review.
38 Bernard Hanotiau, ‘Document Production in International Arbitration: A
Tentative Definition of “Best Practices”’, in Document Production in International
Arbitration, ICC International Court of Arbitration Bulletin: 2006 Special
Supplement, ed. ICC (Paris: ICC Publishing, 2006), 114.
39 Rolf Stürner, ‘The Principles of Transnational Civil Procedure: An Into to the

Basic Conceptions’, RabelsZ 69, no. 2 (2005): 234.

40 Ibid.
41 Ibid.
42 Ibid.
43 See Gary B. Born, International Commercial Arbitration (The Hague: Kluwer

Law International, 2009), 1880.

44 UNCITRAL, ‘Report of the Secretary-General on the Analytical Commentary on

Draft Text of a Model Law on International Commercial Arbitration’, 25 March

1985, UN Doc. A/CN.9/264, Art. 19 para. 6. Some civil law commentators assert
that absent express agreement of the parties either directly or via selection of
appropriate rules, orders for discovery may be a violation of due process or a
breach of public policy. See Frank-Bernd Weigand, ‘Introduction’, in
Practitioner's Handbook on International Arbitration, ed. Frank-Bern Weigand
(Munich: Verlag C.H. Beck, 2002), 80-88; Mauro Rubino-Sammartano, ‘Italy’, in
Practitioner's Handbook on International Arbitration, ed. Frank-Bern Weigand
(Munich: Verlag C.H. Beck, 2002), 860 suggests that, in an Italian court, ordering
discovery is likely to be held to violate transnational public policy. It is not clear
that he thinks that this is correct or should be universal. Any suggestion to that
effect is rightly criticised by Gary B. Born, International Commercial Arbitration
(The Hague: Kluwer Law International, 2009), 1882. It has been further
criticised by numerous other civilian scholars and practitioners. Broad document
production could not be said to offend against transnational public policy or
transnational due process norms. While some national systems may preference
liberty over truth, it is certainly not a transnational norm.
45 Article 9.7 allows for adverse costs orders where a party has failed to conduct

itself in good faith in the taking of evidence. Similarly, ICDR Guideline 8(b) allows
for failures to comply with an order for information exchange to be taken into
account in allocating costs in addition to supporting adverse inferences.
46 Gabrielle Kaufmann-Kohler, ‘Globalization of Arbitral Procedure’, Vanderbilt

Journal of Transnational Law 36, no. 4 (2003): 1327, n. 66.

47 The contrary views of Weigand and Rubino-Sammartano were noted above.

Cases where challenges were rejected when a tribunal refused to order

document production include: Judgment of 21 January 1997, Société Nu Swift plc
v. Société White Knight, Cour d'Appel de Paris (Paris Court of Appeal) in Revue de
l'Arbitrage (1997), no. 3: 428; Judgment of 30 September 1999, Hanseatisches
Oberlandesgericht (Hanseatic Court of Appeal), Bremen in Albert Jan van den
Berg (ed.), Yearbook of Commercial Arbitration Volume XXXI (The Hague: Kluwer
Law International, 2006), 640.
48 Air Canada v. Secretary of State for Trade [1983] 2 AC 394 at 438.
49 The distinction may not have been fully apparent in Resort Condominiums

International Inc. v. Bolwell & Anor (1993) 118 ALR 655. An award on this basis
was upheld in Publicis Communication and Publicis SA v. True North
Communications Inc 2006 F. 3d 725 (UA CA 7th Cir 14 March 2000).
50 This is discussed further in section 14.21.
51 Howard M. Holtzmann, ‘Fact-Finding by the Iran-United States Claims
Tribunal’, in Fact-Finding before International Tribunals: Eleventh Sokol
Colloquium, ed. R. Lillich (Ardsley-on-Hudson: Transnational Publishers, 1992),
109.
52 Ali Z. Marossi, ‘The Necessity for Discovery of Evidence in the Fact-Finding

Process of International Tribunals’, Journal of International Arbitration 26, no. 4

(2009): 517.
53 Klaus Lionnet, ‘Once Again: Is Discovery of Documents Appropriate in
International Arbitration?’, in Global Reflections on International Law, Commerce
and Dispute Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald
Aksen et al. (Paris: ICC Publishing, 2005), 497.
54 Peter Hafter, ‘The Provisions on the Discovery of Internal Documents in the

IBA Rules of 1999’, in Global Reflections on International Law, Commerce and

Dispute Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen
et al. (Paris: ICC Publishing, 2005), 352.
55 Ibid., 362.
56 William W. Park, ‘Arbitration's Discontents: Of Elephants and Pornography’,
Arbitration International 17, no. 3 (2001): 265, n. 9.
57 Peter Ashford, ‘Document Discovery and International Commercial
Arbitration’, American Review of International Arbitration 17 (2006): 101.
58 IBA Rules on the Taking of Evidence in International Arbitration 2010 Art. 3.2

and 3.4.
59 Ibid., Art. 3.6.
60 Ibid., Art. 3.4 and 3.7.
61 Hilmar Raeschke-Kessler, ‘The Production of Documents in International

Arbitration: A Commentary on Article 3 of the New IBA Rules of Evidence’,

Arbitration International 18, no. 4 (2002): 426.
62 The latter is recommended by David W. Rivkin, ‘Towards a New Paradigm in

International Arbitration: The Town Elder Model Revisited’, Arbitration

International 24, no. 3 (2008): 384.
63 E. David B. Tavender, ‘To Restore Speed and Cost Effectiveness, Arbitrators

Must Focus on Time Allocation’, Alternatives to the High Cost of Litigation 28, no.
8 (2010): 163.
64 Charles N. Brower & Jeremy K. Sharpe, ‘Determining the Extent of Discovery

and Dealing with Requests for Discovery: Perspectives from the Common Law’,
in The Leading Arbitrators' Guide to International Arbitration, 2nd edn, ed. L.
Newman & R. Hill (Huntington: Juris Publishing, 2008), 324.
65 Jeffrey Pinsler, ‘Is Discovery Available Prior to the Commencement of
Arbitration Proceedings?’, Singapore Journal of Legal Studies (June 2005): 64–75.
66 [2010] 1 SLR 25.
67 [2005] SGCA 26.
68 Matthew Oommen & Tan Charis, ‘The Use of Pre-action Discovery to Aid

Arbitration? – A Practitioner's Perspective’, Asian International Arbitration

Journal 6, no. 2 (2010): 191.
69 Equinox Offshore Accommodation Ltd v. Richshore Marine Supplies Pte Ltd

[2010] SGHC 122.

70 Oommen and Charis criticise the distinction. See Matthew Oommen & Tan

Charis, ‘The Use of Pre-action Discovery to Aid Arbitration? – A Practitioner's

Perspective’, Asian International Arbitration Journal 6, no. 2 (2010): 193.
71 See section 8.6 in relation to pre-arbitral referee provisions.
72 Charles N. Brower & Jeremy K. Sharpe, ‘Determining the Extent of Discovery

and Dealing with Requests for Discovery: Perspectives from the Common Law’,
in The Leading Arbitrators' Guide to International Arbitration, 2nd edn, ed. L.
Newman & R. Hill (Huntington: Juris Publishing, 2008), 327. For a view that
internal documents cannot be compelled, see Jan Paulsson, ‘Overview of Methods
of Presenting Evidence in Different Legal Systems’, in Planning Efficient
Arbitration Proceedings/The Law Applicable in International Arbitration, ICCA
Congress Series No. 7, ed. Albert Jan van den Berg (The Hague: Kluwer Law
International, 1996), 118.
73 IBA Supplementary Rules of Evidence (1983) Rule 4.4. Lionnet suggests that

those from a civil law tradition not used to having to produce internal documents
could expressly include such a restriction when agreeing on evidentiary rules.
See Klaus Lionnet, ‘Once Again: Is Discovery of Documents Appropriate in
International Arbitration?’, in Global Reflections on International Law, Commerce
and Dispute Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald
Aksen et al. (Paris: ICC Publishing, 2005), 500.
74 1999 IBA Working Party & 2010 IBA Rules of Evidence Subcommittee,
‘Commentary on the revised text of the 2010 IBA Rules on the Taking of Evidence
in International Arbitration’ (2010), 9.
75 The Sedona Conference Working Group on Electronic Document Retention &

Production, The Sedona Principles: Best Practices Recommendations and

Principles for Addressing Electronic Document Production, 2nd edn (2007), 1.
76 John J. Range & Jonathan M. Wilan, ‘Techniques for Obtaining Efficient and

Economical E-Disclosure Despite Arbitral Resistance to U.S.-Style Discovery’, in

e-Discovery in Arbitration, Leading Lawyers on Recovering Electronic Evidence,
Meeting New Disclosure Guidelines and Implementing Measures to Streamline the
Process, ed. John J. Range et al. (USA: Aspatore, 2010), 46.
77 Ibid.
78 See generally Nicholas Fletcher, ‘The Use of Technology in the Production of

Documents’, in Document Production in International Arbitration, ICC

International Court of Arbitration Bulletin: 2006 Special Supplement, ed. ICC
(Paris: ICC Publishing, 2006); Bernard Hanotiau, ‘Document Production in
International Arbitration: A Tentative Definition of “Best Practices” ’, in
Document Production in International Arbitration, ICC International Court of
Arbitration Bulletin: 2006 Special Supplement, ed. ICC (Paris: ICC Publishing,
2006): 102.
79 Robert H. Smit, ‘E-Disclosure under the Revised IBA Rules on the Taking of

Evidence in International Arbitration’, International Arbitration Law Review 13,

no. 5 (2010): 205.
80 Ibid., 202.
81 John J. Range & Jonathan M. Wilan, ‘Techniques for Obtaining Efficient and

Economical e-Disclosure Despite Arbitral Resistance to U.S.-Style Discovery’, in e-

Discovery in Arbitration, Leading Lawyers on Recovering Electronic Evidence,
Meeting New Disclosure Guidelines and Implementing Measures to Streamline the
Process, ed. John J. Range et al. (USA: Aspatore, 2010), 17.
82 Robert H. Smit, ‘E-Disclosure under the Revised IBA Rules on the Taking of

Evidence in International Arbitration’, International Arbitration Law Review 13,

no. 5 (2010): 205.
83 See, e.g., the discussion in NAFTA Chapter 11 Arbitration, Vito G Gallo and

Government of Canada (Procedural Order No 4) unreported, 27 December 2009,

cited in Amal Bouchenaki, ‘The IBA Rules Lay the Grounds for Solutions to
Address Electronic Document Production Disputes’, International Arbitration
Law Review 13, no. 5 (2010): 185–186.
84 Robert H. Smit, ‘E-Disclosure under the Revised IBA Rules on the Taking of

Evidence in International Arbitration’, International Arbitration Law Review 13,

no. 5 (2010): 206. An example is Net Case as used by the ICC.
85 Ibid.
86 Ibid., 207.
87 The Sedona Conference Working Group on Electronic Document Retention &

Production, The Sedona Principles: Best Practices Recommendations and

Principles for Addressing Electronic Document Production, 2nd edn (2007). See
also ‘Electronic Disclosure – A Report of a Working Party Chaired by the
Honourable Mr Justice Cresswell’ (London: Her Majesty's Court Service, 2004).
88 Richard D. Hill, ‘The New Reality of Electronic Document Production in
International Arbitration: A Catalyst for Convergence?’, Arbitration International
25, no. 1 (2009): 91.
89 Chartered Institute of Arbitrators, ‘Protocol for E-Disclosure in Arbitration’,

available at <www.ciarb.org>.
90 The International Institute for Conflict Prevention and Resolution (CPR),

Protocol on Disclosure of Documents and Presentation of Witnesses in

Commercial Arbitration, <www.cpradr. org>. See also the CPR Global Rules for
Accelerated Commercial Arbitration, 20 August 2009.
91 John Beechey, ‘The ICDR Guidelines for Information Exchanges in
International Arbitration: An Important Addition to the Arbitral Toolkit’, Dispute
Resolution Journal 63, no. 3 (2008): 84–89.
92 See Loretta Malintoppi, ‘The ICC Task Force on the Production of Electronic

Documents in Arbitration – An Overview’, in Written Evidence and Discovery in

International Arbitration, ed. Theresa Giovanini & Alexis Mourre (Paris: ICC
Publishing, 2009), 415–421. John Beechey, ‘The ICDR Guidelines for Information
Exchanges in International Arbitration: An Important Addition to the Arbitral
Toolkit’, Dispute Resolution Journal 63, no. 3 (2008): 84–89.
93 Richard D. Hill, ‘The New Reality of Electronic Document Production in
International Arbitration: A Catalyst for Convergence?’, Arbitration International
25, no. 1 (2009): 97–98.
94 Robert H. Smit, ‘E-Disclosure under the Revised IBA Rules on the Taking of

Evidence in International Arbitration’, International Arbitration Law Review 13,

no. 5 (2010): 203.
95 Ibid., 204.
96 The Sedona Conference Working Group on Electronic Document Retention &

Production, The Sedona Principles: Best Practices Recommendations and

Principles for Addressing Electronic Document Production, 2nd edn (2007),
Principle 12.
97 Ibid., Principle 8.
98 CIArb Protocol para. 9.
99 International Institute for Conflict Prevention and Resolution, ‘CPR Protocol

on Disclosure of Documents and Presentation of Witnesses in Commercial

Arbitration’ (New York: International Institute for Conflict Prevention and
Resolution, 2009), para. 1(d).
100 John J. Range & Jonathan M. Wilan, ‘Techniques for Obtaining Efficient and

Economical e-Disclosure Despite Arbitral Resistance to U.S.-Style Discovery’, in e-

Discovery in Arbitration, Leading Lawyers on Recovering Electronic Evidence,
Meeting New Disclosure Guidelines and Implementing Measures to Streamline the
Process, ed. John J. Range et al. (USA: Aspatore, 2010), 20. They suggest that
counsel might send notices to their own clients and opposing parties to preserve
documents.
101 Robert H. Smit, ‘E-Disclosure under the Revised IBA Rules on the Taking of

Evidence in International Arbitration’, International Arbitration Law Review 13,

no. 5 (2010): 202.
102 London & Leeds Estates Ltd v. Paribas Ltd (No 2) [1995] 02 EG 134, [1995] 1

EGLR 102.
103 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1923 et seq.

104 See, e.g., UNCITRAL Model Law Art. 27; Swiss Private International Law Art.
184(2).
105 See, e.g., the discussion in Gary B. Born, International Commercial Arbitration

(The Hague: Kluwer Law International, 2009), 1933–1936. See also Charles
Owen Verrill Jr, ‘Discovery from Non-parties (Third-Party Discovery) in
International Arbitration’, Arbitration 76, no. 1 (2010): 113.
106 Thomas H. Webster, ‘Obtaining Evidence from Third Parties in International

Arbitration’, Arbitration International 17, no. 2 (2001): 161.

107 Peter Megens, Paul Starr & Peter Chow, ‘Compulsion of Evidence in
International Commercial Arbitration: An Asia-Pacific Perspective’, Asian
International Arbitration Journal 2, no. 1 (2006): 40.
108 See Bryant Yuan Fu Yang & Diane Chen Dai, ‘Tipping the Scale to Bring a

Balanced Approach: Evidence Disclosure in Chinese International Arbitration’,

Pacific Rim Law & Policy Journal 17, no. 1 (2008): 63–64, dealing with requests
by CIETAC.
109 Hilmar Raeschke-Kessler, ‘The Production of Documents in International

Arbitration: A Commentary on Article 3 of the New IBA Rules of Evidence’,

Arbitration International 18, no. 4 (2002): 411–430.
110 Seung Wha Chang, ‘Document Production under the Asian Civil Law System’,

in The Asian Leading Arbitrators' Guide to International Arbitration, ed. Michael

Pryles & Michael J. Moser (Huntington: JurisNet, 2007), 275.
111 Aguas del Tunari S.A. v Republic of Bolivia, Decision on Respondent's
Objection to Jurisdiction of October 21, 2005, ICSID Review – Foreign Investment
Law Journal 20 (2005): 461. See also Gary B. Born, International Commercial
Arbitration (The Hague: Kluwer Law International, 2009), 1907, n. 150.
112 IBA Rules on the Taking of Evidence in International Arbitration 2010 Art.

3.7.
113 Prior to the development of the IBA guidelines, UNCITRAL sought to develop

Draft Guidelines for Preparatory Conferences in Arbitral Proceedings which in

part had proposed a similar model for production requests. This was not
accepted at the time because of the then significant differences between the
views of common law and civil law countries. The draft guidelines suggested
that: ‘One possible set of conditions for requesting a document from the
opponent may be formulated along the following lines: the document must be
described with reasonable particularity; the document must be such that it
would likely contribute to the clarification of the case; the document must be
within the control of the party from whom production is sought; and the seeking
party must have made reasonable but unsuccessful efforts to obtain the
document. A further condition that might be included, either unqualified or
subject to discretion by the arbitral tribunal is that the document must have
passed between the requested party and a third party who is not a party to the
arbitration, a condition that would exclude requests for purely internal
documents. It might be appropriate to clarify that, if the requested party refuses
to comply with a proper request, the question as to whether the refusal is
justified is to be decided by the arbitral tribunal’. UNCITRAL's Draft Guidelines
listed a number of matters on which parties may refuse requests. The Draft
Guidelines indicated that ‘(g)rounds for refusal may concern, for instance,
national defence, diplomatic relations between countries, certain governmental
actions, certain communications between a client and its legal counsel, or the
right of a person to refuse to take a self incriminating action’. UNCITRAL, Draft
Guidelines for Preparatory Conferences in Arbitral Proceedings: Report of the
Secretary-General, 30 March 2004, UN Doc. A/CN.9/396/Add.1, 17. See also
Roberto Ceccon, ‘UNCITRAL Notes on Organizing Arbitral Proceedings and the
Conduct of Evidence: A New Approach to International Arbitration’, Journal of
International Arbitration 14, no. 2 (1997): 73.
114 Thomas H. Webster, ‘Obtaining Documents from Adverse Parties in
International Arbitration’, Arbitration International 17, no. 1 (2001): 56;
Matthias Scherer, ‘The Limits of the IBA Rules on the Taking of Evidence in
International Arbitration: Document Production Based on Contractual or
Statutory Rights’, International Arbitration Law Review 13, no. 5 (2010): 196.
115 It might, for example, seek to show that the exempting right is an implied

provision or a separate legal obligation that takes precedence.

116 Charles N. Brower & Jeremy K. Sharpe, ‘Determining the Extent of Discovery

and Dealing with Requests for Discovery: Perspectives from the Common Law’,
in The Leading Arbitrators' Guide to International Arbitration, 2nd edn, ed. L.
Newman & R. Hill (Huntington: Juris Publishing, 2008), 320. A similar conclusion
was reached in BNP Paribas & Ors v. Deloitte and Touche LLP [2003] EWHC 2874
(Comm), holding that under s. 43 of the English Arbitration Act 1996 a party was
not entitled to request documents simply for a general review as to their
potential relevance.
117 See section 11.4.
118 Hilmar Raeschke-Kessler, ‘The Production of Documents in International

Arbitration: A Commentary on Article 3 of the New IBA Rules of Evidence’,

Arbitration International 18, no. 4 (2002): 427. One ICSID tribunal unfortunately
used a phrase which suggested that necessity and relevance are synonyms when
it stated that the analysis ‘relates to a substantive inquiry into whether the
documents requested are relevant to, and in that sense necessary for, the
purposes of the proceedings where the documents are to be used.’ ADF Group
Inc. v. United States of America, Award of 9 January 2003, ICSID Review – Foreign
Investment Law Journal 18 (2003): 210.
119 Charles N. Brower & Jeremy K. Sharpe, ‘Determining the Extent of Discovery

and Dealing with Requests for Discovery: Perspectives from the Common Law’,
in The Leading Arbitrators' Guide to International Arbitration, 2nd edn, ed. L.
Newman & R. Hill (Huntington: Juris Publishing, 2008), 319.
120 Virginia Hamilton, ‘Document Production in ICC Arbitration’, in Document

Production in International Arbitration, ICC International Court of Arbitration

Bulletin: 2006 Special Supplement, ed. ICC (Paris: ICC Publishing, 2006), 70.
121 Byung Chol Yoon & Joel Richardson, ‘Discovery in Investment Arbitration

Involving Republic of Korea’, Asian International Arbitration Journal 4, no. 2

(2008): 139.
122 UNCITRAL, Draft Guidelines for Preparatory Conferences in Arbitral
Proceedings: Report of the Secretary-General, 30 March 2004, UN Doc.
A/CN.9/396/Add.1, 17. See also Roberto Ceccon, ‘UNCITRAL Notes on
Organizing Arbitral Proceedings and the Conduct of Evidence: A New Approach
to International Arbitration’, Journal of International Arbitration 14, no. 2 (1997):
73.
123 International Institute for Conflict Prevention and Resolution, CPR Protocol

on Disclosure of Documents and Presentation of Witnesses in Commercial

Arbitration (New York: International Institute for Conflict Prevention and
Resolution, 2009), para. 1(a).
124 O Co v. M Co [1996] 2 Lloyd's Rep 347 at 351.
125 Yves Derains, ‘Towards Greater Efficiency in Document Production before

Arbitral Tribunals – A Continental Viewpoint’, in Document Production in

International Arbitration, ICC International Court of Arbitration Bulletin: 2006
Special Supplement, ed. ICC (Paris: ICC Publishing, 2006), 87.
126 See sections 10.4.1 and 10.4.7 for the preferred distinction between burden

and onus of proof.

127 Charles N. Brower & Jeremy K. Sharpe, ‘Determining the Extent of Discovery

and Dealing with Requests for Discovery: Perspectives from the Common Law’,
in The Leading Arbitrators' Guide to International Arbitration, 2nd edn, ed. L.
Newman & R. Hill (Huntington: Juris Publishing, 2008), 320, n. 54.
128 The Iran-United States Claims Tribunal rejected an application for lack of

specificity as to location in Iran-US Order of 27 November 2001, Case No. A11

Islamic Republic of Iran and United States of America, cited in Charles N. Brower
& Jeremy K. Sharpe, ‘Determining the Extent of Discovery and Dealing with
Requests for Discovery: Perspectives from the Common Law’, in The Leading
Arbitrators' Guide to International Arbitration, 2nd edn, ed. L. Newman & R. Hill
(Huntington: Juris Publishing, 2008), 318.
129 BNP Paribas & Ors v. Deloitte and Touche LLP [2003] EWHC 2874 (Comm).
130 Peter Ashford, ‘Document Discovery and International Commercial
Arbitration’, American Review of International Arbitration 17 (2006): 107.
131 Julian D.M. Lew QC, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 568.
132 Peter Hafter, ‘The Provisions on the Discovery of Internal Documents in the

IBA Rules of 1999’, in Global Reflections on International Law, Commerce and

Dispute Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen
et al. (Paris: ICC Publishing, 2005), 349.
133 Gary B. Born, International Commercial Arbitration (The Hague, Kluwer Law

International, 2009), 1909; Virginia Hamilton, ‘Document Production in ICC

Arbitration’, in Document Production in International Arbitration, ICC
International Court of Arbitration Bulletin: 2006 Special Supplement, ed. ICC
(Paris: ICC Publishing, 2006), 69.
134 Peter Hafter, ‘The Provisions on the Discovery of Internal Documents in the

IBA Rules of 1999’, in Global Reflections on International Law, Commerce and

Dispute Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen
et al. (Paris: ICC Publishing, 2005), 361.
135 IBA Rules on the Taking of Evidence in International Arbitration 2010 Art.

3.4; David D. Caron, Lee M. Caplan & Matti Pellonpää, The UNCITRAL Arbitration
Rules: A Commentary (Oxford: Oxford University Press), 593, extracting MCA Inc.
v. Islamic Republic of Iran, Order in Case No. 768, 6 October 1983.
136 Virginia Hamilton, ‘Document Production in ICC Arbitration’, in Document

Production in International Arbitration, ICC International Court of Arbitration

Bulletin: 2006 Special Supplement, ed. ICC (Paris: ICC Publishing, 2006), 74.
137 See ibid.
138 See, e.g., Protiva v. The Government of the Islamic Republic of Iran Award in

Case No. 316 (566-316-2) of 14 July 1995 in Yearbook of Commercial Arbitration

Volume XXI, ed. Albert Jan van den Berg (The Hague: Kluwer Law International,
1996), para. 65.
139 Procedural Order 8(1), William Ralph Clayton v. Government of Canada (PCA

Case 2009-04) Unreported 25 November 2009 available at <www.pca-

cpa.org/upload/files/Bilcon-ProceduralOrde...>.
140 Even that is not necessarily so as the request could be for ‘documents

annotated by persons other than the requesting party’, which would make them
technically distinct.
141 Virginia Hamilton, ‘Document Production in ICC Arbitration’, in Document

Production in International Arbitration, ICC International Court of Arbitration

Bulletin: 2006 Special Supplement, ed. ICC (Paris: ICC Publishing, 2006), 73.
142 International Institute for Conflict Prevention and Resolution, ‘CPR Protocol

on Disclosure of Documents and Presentation of Witnesses in Commercial

Arbitration’ (New York: International Institute for Conflict Prevention and
Resolution, 2009), para. 1(e)(2).
143 International Centre for Dispute Resolution, ‘ICDR Guidelines for Arbitrators

Concerning Exchanges of Information’, para. 8(a).

144 For example, Art. 42 of the UNCITRAL Rules 2010 only mentions cost
allocation powers in relation to ‘awards’.
145 Loukas F. Wyss, ‘Trends in Documentary Evidence and Consequences for Pre-

arbitration Document Management’, International Arbitration Law Review 13, no.

3 (2010): 111, 117.
146 Charles N. Brower & Jeremy K. Sharpe, ‘Determining the Extent of Discovery

and Dealing with Requests for Discovery: Perspectives from the Common Law’,
in The Leading Arbitrators' Guide to International Arbitration, 2nd edn, ed. L.
Newman & R. Hill (Huntington: Juris Publishing, 2008), 321.
147 Different views were taken on this issue in ADF Group Inc. v. United States of

America, Award of January 9, 2003, ICSID Review – Foreign Investment Law

Journal 18 (2003): 210 and Pope & Talbot Inc v Government of Canada,
NAFTA/UNCITRAL, available at <http://www.international.gc.ca/trade-
agreements-ac...>.
148 Charles N. Brower & Jeremy K. Sharpe, ‘Determining the Extent of Discovery

and Dealing with Requests for Discovery: Perspectives from the Common Law’,
in The Leading Arbitrators' Guide to International Arbitration, 2nd edn, ed. L.
Newman & R. Hill (Huntington: Juris Publishing, 2008), 333, citing comments of
the United States-Mexico General Claims Commission in William A Parker (USA)
v. United Mexican States, Reports of International Arbitral Awards 4 (1926): 39, to
the effect that States ‘are in honour bound to make full disclosures of the facts in
each case so far as such facts are within their knowledge, or can reasonably be
ascertained by them’, although this view is not suggested as a uniform position.
149 1999 IBA Working Party & 2010 IBA Rules of Evidence Subcommittee,

‘Commentary on the revised text of the 2010 IBA Rules on the Taking of Evidence
in International Arbitration’ (2010), 26.
150 Procedural Order No. 2 of May 2006 in Biwater Gauff (Tanzania) Ltd v. United

Republic of Tanzania, ICSID Case No. ARB/05/22, 8.

151 Corfu Channel (United Kingdom v Albania) [1949] ICJ 4, 32.
152 Application of the Convention on the Prevention and Punishment of the

Crime of Genocide (Bosnia and Herzegovina v. Serbia and Montenegro) [2007] ICJ
75, para 204.
153 This is discussed in section 10.17.7 above.
154 Christopher Newmark, ‘Efficient, Economical and Fair: The Mantra of the
New IBA Rules’, International Arbitration Law Review 13, no. 5 (2010): 165, 168.
155 Gabrielle Kaufmann-Kohler, ‘Globalization of Arbitral Procedure’, Vanderbilt

Journal of Transnational Law 36, no. 4 (2003): 1328.

156 William W. Park, ‘Arbitration's Discontents: Of Elephants and Pornography’,

Arbitration International 17, no. 3 (2001): 266.

157 See, e.g., ICDR Arbitration Rules 2009 Art. 21; WIPO Arbitration Rules 2009

Art. 46.
158 The nature of an award is discussed in section 16.2 below.
159 IBA Rules on the Taking of Evidence in International Arbitration 2010 Art.

3.6.
160 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 395.

161 Ibid. This is similar in form to the Scott Schedule used to set out complex

contested technical issues, typically in building disputes. See Peter Ashford,

‘Document Discovery and International Commercial Arbitration’, American
Review of International Arbitration 17 (2006): 99. The Redfern Schedule is
recommended in ICC, ‘Techniques for Controlling Time and Costs in Arbitration’,
<http://www.iccwbo.org/uploadedFiles/TimeCost_E.pdf...>, 2007, para. 55.
162 Seung Wha Chang, ‘Document Production Under the Asian Civil Law System’,

in Asian Leading Arbitrators' Guide to International Arbitration, ed. Michael

Pryles & Michael J. Moser (Huntington: JurisNet, 2008), 267, 274.
163 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 396.

164 Article originally published in English and French in Document Production in

International Arbitration, ICC International Court of Arbitration Bulletin, 2006

Special Supplement, © International Chamber of Commerce (ICC). Reproduced
with permission of the ICC. The Bulletin is available from the ICC Business
Bookstore <http://www.iccbooks.com > and in the ICC Dispute Resolution
Library <http://www.iccdrl.com >. Bernard Hanotiau, ‘Document Production in
International Arbitration: A Tentative Definition of “Best Practices”’, in Document
Production in International Arbitration, ICC International Court of Arbitration
Bulletin: 2006 Special Supplement, ed. ICC (Paris: ICC Publishing, 2006), 119.
165 Peter Ashford, ‘Document Discovery and International Commercial
Arbitration’, American Review of International Arbitration 17 (2006): 135.
166 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 693.
167 Ibid. ;citing Judgment of 14 June 1985, Sarl Anciens Ets Harognan Comptoir

Euro-Turc v. Turkish Airlines ‘Turk Hava Yollari AO’, Cour d'Appel de Paris (Paris
Court of Appeal) in Revue de l'Arbitrage (1987), no. 3: 395 and observations by J
Pellerin. The decision itself relates to the late submission of documents to the
tribunal after the closing of oral arguments. It does talk about due process rights,
but only in the context of late submission of documents.
168 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 693.
169 The IBA Working Party that drafted the 1999 rules indicated that the
document production expert would not necessarily need to be appointed as such.
IBA Working Party, ‘Commentary on the New IBA Rules of Evidence in
International Commercial Arbitration’, Business Law International 14, no. 2
(2000): 21.
170 Ibid., 11.
171 Guyana v. Suriname, Permanent Court of Arbitration, Award of the Arbitral

Tribunal dated 17 September 2007.

172 Hans van Houtte, ‘The Document Production Master and the Experts'
Facilitator: Two Possible Aides for an Efficient Arbitration’, in Liber Amicorum
Bernardo Cremades, ed. M. Á Fernández-Ballesteros & David Arias (Madrid: La
Ley, 2010), 1147.
173 Ibid., 1154.
174 Hans van Houtte, ‘The Use of an Expert to Handle Document Production: IBA

Rules on the Taking of Evidence (Art. 3(7))’, in International Arbitration 2006:

Back to Basics? ICCA Congress Series 2006, ed. Albert Jan van den Berg (The
Hague: Kluwer Law International, 2007), 639.
175 Hans van Houtte, ‘The Document Production Master and the Experts'
Facilitator: Two Possible Aides for an Efficient Arbitration’, in Liber Amicorum
Bernardo Cremades, ed. M. Á Fernández-Ballesteros & David Arias (Madrid: La
Ley, 2010), 1154.
176 A useful order as to confidentiality is extracted in Virginia Hamilton,
‘Document Production in ICC Arbitration’, in Bernard Hanotiau, ‘Document
Production in International Arbitration: A Tentative Definition of “Best
Practices”’, in Document Production in International Arbitration, ICC International
Court of Arbitration Bulletin: 2006 Special Supplement, ed. ICC (Paris: ICC
Publishing, 2006), 75.
177 Dutch Code of Civil Procedure Art. 1056; Belgian Judicial Code Art. 1709bis.
178 Swedish Arbitration Act, 25.
179 French New Code of Civil Procedure Art. 1467. See also Emmanuel Gaillard &

John Savage (eds), Fouchard, Gaillard, Goldman on International Commercial

Arbitration (The Hague: Kluwer Law International, 1999), 697.
180 See Hong Kong Arbitration Ordinance s. 61.
181 Hong Kong Evidence Ordinance s. 75.
182 Astel-Peiniger Joint Venture v. Argos Engineering & Heavy Industries Co. Ltd

Clout Case 78.

183 Transfield Philippines Inc. v. Luzon Hydro Corporation [2002] VSC 215.
184 See generally Chapter 15.
185 The ICC suggests that the allocation of costs can be a useful tool to encourage

efficient and reasonable behaviour. See ICC, ‘Techniques for Controlling Time
and Costs in Arbitration’,
<http://www.iccwbo.org/uploadedFiles/TimeCost_E.pdf...>, 2007, para. 85.
Christopher Newmark, ‘Efficient, Economical and Fair: The Mantra of the New
IBA Rules’, International Arbitration Law Review 13, no. 5 (2010): 165, 167,
makes the observation that this among other things takes the IBA Rules beyond
mere issues of evidence.
186 IBA Rules on the Taking of Evidence in International Arbitration 2010 Art.

9.6. The Iran-US Claims Tribunal declined to draw adverse inferences from a
failure to provide certain direct reports in Arthur J Fritz & Co. v. Sherkate Tavonie
Sherkathaye Sakhtemanie (Co-operative Society of Construction Companies) 22
Iran-US Claims Tribunal Reports 170 (1989). Note, however, that Judge Allison
dissented on this point.
187 See Peter Hafter, ‘The Provisions on the Discovery of Internal Documents in

the IBA Rules of 1999’, in Global Reflections on International Law, Commerce and
Dispute Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen
et al. (Paris: ICC Publishing, 2005), 354.
188 David D. Caron, Lee M. Caplan & Matti Pellonpää, The UNCITRAL Arbitration

Rules: A Commentary (Oxford: Oxford University Press), 578.

189 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1920.

190 US Federal Arbitration Act, § 10(a)(3).
191 Charles N. Brower & Jeremy K. Sharpe, ‘Determining the Extent of Discovery

and Dealing with Requests for Discovery: Perspectives from the Common Law’,
in The Leading Arbitrators' Guide to International Arbitration, 2nd edn, ed. L.
Newman & R. Hill (Huntington: Juris Publishing, 2008), 332.
192 Without guidance from the tribunal, duplication will typically be inevitable,

simply because of each party's duty to disclose material on which it relies. Each
would have to disclose the same contract and inter-party correspondence.
193 Jack J. Coe, ‘Pre-hearing Techniques to Promote Speed and Cost-
Effectiveness: Some Thoughts Concerning Arbitral Process Design’, Pepperdine
Dispute Resolution Law Journal 2, no. 1 (2002): 69.
194 This suggestion is made in ICC, ‘Techniques for Controlling Time and Costs in

Arbitration’, <http://www.iccwbo.org/uploadedFiles/TimeCost_E.pdf...>, 2007,

para. 77.
195 UNCITRAL, ‘UNCITRAL Notes on Organizing Arbitral Proceedings’, para. 42.
196 ICC, ‘Techniques for Controlling Time and Costs in Arbitration’,
<http://www.iccwbo.org/uploadedFiles/TimeCost_E.pdf...>, 2007, para. 52.
197 See ICC, ‘Rapport final sur l'arbitrage dans le domaine de la construction’,

Bulletin de la Cour Internationale d'arbitrage de la CCI 12, no. 2 (2001): 27–28.

198 See ICC, Using Technology to Resolve Business Disputes, ICC International Court

of Arbitration Bulletin 2004 Special Supplement.

199 See ICC, ‘Rapport final sur l'arbitrage dans le domaine de la construction’,

Bulletin de la Cour Internationale d'arbitrage de la CCI 12, no. 2 (2001): 27–28.

200 Werner Melis, ‘Presentation of Documentary Evidence and Witnesses’, in

Preventing Delay and Disruption of Arbitration/Effective Proceedings in

Construction Cases, ICCA Congress Series 5, 1990 Stockholm, ed. Albert Jan van
den Berg (The Hague: Kluwer Law International), 515.
201 UNCITRAL, ‘UNCITRAL Notes on Organizing Arbitral Proceedings’, para. 54.
202 See section 12.14.5 as to expert reports and section 10.16.5 as to
demonstrative evidence.
203 IBA Rules on the Taking of Evidence in International Arbitration 2010 Art.

3.12(d).
204 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 401.

Part II: The Process of an Arbitration,
Chapter 12: General Witness and
Expert Evidence

Jeff Waincymer,
12.1. Introduction

The previous chapter considered the value of contemporaneous documents to

legal fact finding. Section 10.2 above looked at the supposed tendencies of
different legal families to prefer either contemporaneous documents or oral
testimony. As has been suggested throughout, there are advantages and
disadvantages to each. The advantage of oral evidence is that questions can be
asked and themes explored at the request of opposing counsel or the
adjudicator. Disadvantages include the expense, given that all parties sit and
listen to examination and cross-examination; individuals who are better
educated and more eloquent might have their testimony preferred even though
they are not more truthful; different people have different ability to withstand
the tricks of cross-examination techniques; and the evidence may be less
probative given that it is presented after the dispute is known and at times by
those with a vested interest in the proceeding's outcome. Despite these
concerns, in many instances, oral evidence will be the key to resolving
contentious factual questions. In other cases oral evidence will be needed to
explain the background, nature and even interpretation of key documents. As a
result, most arbitrations will typically involve oral evidence.

Lévy and Reed point to various benefits oral testimony may provide over and
above written statements. The authors note that this will obviously depend on
the circumstances. The questions to be considered in terms of utility are the
extent of the credibility of each witness; whether witnesses may clarify and
elaborate on facts in the statements or other relevant facts; whether they will
assist in authenticating or interpreting documents, normally those in the record
but at times other documents; and whether examination of witnesses will help
the counsel identify page "885" the key issues for arbitrators, which includes
helping them properly direct their attention to the important matters. (1)

There are two broad categories of oral evidence, general witnesses and experts.
In the common law tradition, experts are party appointed and are simply a
specialised form of witness. In the civilian tradition, experts are tribunal
appointed and are not witnesses in the normal sense. A number of policy and
procedural questions arise in relation to each category of testimony. The first is
who may give such testimony. The second is the use of written statements and
expert reports as well as, or in lieu of oral evidence. A related question is how
much assistance is it proper for counsel to give in the preparation of
statements? Where oral evidence is concerned, will evidence in chief be given or
will a written statement stand in its place? Is there entitlement to cross-
examination and by whom? Will the tribunal ask questions or leave it for
counsel? What assistance may counsel give in preparing witnesses for
questioning? In what language is testimony presented and what role will there
be for interpreters? Will there be a right of reexamination and yet another
opportunity for a second round of cross-examination? What evidentiary record
is to be made of oral testimony? Once again, the answers to these and other
questions depend upon an amalgam of party choice, mandatory norms, arbitral
statutes and rules and policy factors that should guide the application of
discretions held by the tribunal.

12.2. Arbitral Statutes and Rules and the Right to be Heard

Superimposed on any specific statutory provisions or arbitral rules is the

mandatory right to have an adequate opportunity to present the case, which
includes the right to be heard. It does not seem that the right to be heard on its
own includes an implied right to have oral hearings and examination and cross-
examination of witnesses, absent agreement to the contrary. Nevertheless, the
emerging norm expressed in arbitral statutes and rules is to expressly allow for
oral hearings when one party so requests and to allow witnesses to present
orally as well as in writing and to be questioned by counsel and the tribunal. (2)
Typically they will also indicate that each witness who has submitted a witness
statement shall appear for testimony if the opposing party or the tribunal so
requests. (3) If a witness refuses to submit for testimony without the agreement
of the parties, the tribunal is generally directed to disregard the statement,
unless due to exceptional page "886" circumstances, the tribunal determines
otherwise. (4) Some national laws allow arbitral tribunals to designate one
member to examine witnesses. (5) This is not the preferred view in international
arbitration. (6) In most instances, such a delegation would run the risk of
contradicting the tribunal's mandate.

The next question is how a tribunal should approach the exercise of a discretion
where there is no unilateral right of a party to call for a hearing and the parties
are in disagreement. National systems vary on the issue, but the better view is
that there is no automatic right. (7) The English Court of Appeal has rejected an
argument that a refusal to hear oral evidence was contrary to natural justice. (8)
The Swiss Federal Supreme Court considered that the right to be heard does not
afford a guaranteed right to have witnesses attend and be questioned. (9) A
contrary view might seek to rely on the jurisprudence applying Article 6 of the
European Convention on Human Rights where relevant. (10)

If the parties agree on a documents-only arbitration, then the tribunal will

normally accept this unless the facts in dispute cannot be adequately resolved in
this manner. (11) Even party agreement may not always resolve the issue, as
parties cannot generally waive mandatory due process rights. A number of
factual permutations can be envisaged where discretions and potential
challenges must both be considered. Least likely is an agreement by the parties
to have oral hearings where the tribunal believes that a documents-only
arbitration would be far more efficient. In such circumstances, most arbitrators
would allow a hearing but might severely limit its length. The more likely
scenario where a tribunal's discretionary determination may be challenged is
where parties are not in agreement as to whether there should be an oral
hearing. If the tribunal thinks that an oral hearing is appropriate, there is little
difficulty. The tribunal is following its mandate to come to the most appropriate
conclusion which it has determined requires oral testimony. If the tribunal
decides against a hearing, a challenge may be more likely but should not succeed
in most instances.
page "887"

12.3. Discretionary Control over Witness Testimony

The previous section looked at the general right of parties to have witnesses
give evidence. This section looks at the discretionary control over witness
testimony by the tribunal where witnesses are to be involved. The discussion
covers both general witnesses and experts as witnesses. A tribunal has a duty to
promote fairness and efficiency and cannot allow parties to have an open-ended
right to have as many witnesses as possible over an extended hearing period.
However, a tribunal will need to be careful to ensure that legitimate due process
challenges are not encouraged. These might be made on the grounds of failure to
allow an adequate presentation of case and/or unequal treatment. Judgment
will be needed on a case-by-case basis. The judgment needs to be made in the
context of some uncertainty. At the early procedural stages where procedural
directions are made about the length of hearings and number of witnesses, the
tribunal may not have a full picture as to the key issues and certainly cannot
have a firm view about the preponderance of non-witness evidence.

12.3.1. Choice of Witnesses

The first principle is that the parties can designate the witnesses that they wish
to rely upon. It would not be the norm for a tribunal to allow a preliminary
debate about whether a particular witness may be called although this is
entirely possible under proactive arbitration as a means to exclude superfluous
and irrelevant witnesses. Böckstiegel suggests, however, that every witness and
expert proposed should be invited to be heard unless the tribunal is sure that
the testimony is irrelevant to the outcome of the case. (12)

While parties generally have the right to select their own witnesses, in
exceptional circumstances a tribunal might require a particular witness that was
not proposed to be called. (13) While few arbitral statutes expressly provide a
tribunal with a power to summon witnesses, (14) the better view is that general
discretionary powers are broad enough to justify this. (15) Furthermore, the
intent behind the reference to ‘other evidence’ in the initial version of Article
24.3 of the UNCITRAL Rules was intended to include witness testimony. (16) The
powers would page "888" typically be limited to the parties, their officers and
employees and perhaps other persons under their clear control. Where such
persons are concerned, a tribunal might warn that failure to present a particular
witness may lead to adverse inferences. While a tribunal has a power to call for
the attendance of a particular witness, it may choose not to do so, relying
instead on the application of burden and standards of proof to the material
presented. It has been suggested that if a tribunal does direct that a witness be
heard and the witness does not appear, this might cause problems as to the way
the tribunal will then proceed, (17) although there seems no reason why this
should be different to any other non-attending key person.

12.3.2. Timing and Number of Witnesses

A tribunal might also separate oral hearings into different stages. It may hear a
first round of witnesses and then determine what further evidence is needed in
later rounds. This may be particularly important when a gateway factual finding
on one issue will significantly impact upon the utility of follow up evidence. In
some cases a tribunal might invite the parties to consider presenting further
evidence in support of areas of uncertainty.

It is also perfectly acceptable for a tribunal to impose reasonable limits if an

excessive number of witnesses are proposed. Yet even then it will need to decide
whether it wishes to limit the number of witnesses or the total time frame for
oral testimony or both and if so, to what degree. Obviously the tribunal should
not make procedural rulings that impact upon the ultimate conclusion as to
which party has the preponderance of evidence. Proper preparation and caution
should be the norm. In such circumstances it may be desirable for the tribunal to
indicate its belief as to the key issues. For example, the tribunal can indicate that
it wishes to know about a certain matter and may ask the parties whether one
witness could establish the relevant facts. If so, the tribunal can let the parties
choose who would be the best witness. If the issue instead is the weight of
opinion, there is a need to determine how many witnesses would be necessary
to attest to a widespread belief. A common view among experts can still be
shown by a select number of witnesses supported by an appropriate search of
scholarly literature. Furthermore, if a written witness statement suggests to the
tribunal that the witness would be irrelevant to the outcome, a question might
be raised as to the reason for the witness being called.

There are also policy challenges if there are no limits on number but instead,
overall time limits imposed for efficiency reasons. One influential approach is
chess-clock arbitration, giving the parties a total amount of time to use as they
see fit. This is discussed further in section 9.6. Such approaches raise unique
page "889" challenges where witness testimony is concerned. First, a decision
needs to be made by the tribunal as to whether time taken in cross-examination
counts against the party for whom the cross-examination is being conducted or
counts against the party who called the witness being cross-examined. There
would be a problem in the latter event as counsel could cross-examine at length
to take away available time from the opposing party. It also goes against the
substantive policy basis of a time limit for presenting one's case if factors
beyond counsel's control are used against it. Hence, the norm is to count cross-
examination time against the party conducting the cross-examination.

There are still other difficulties with cross-examination under chess-clock

arbitration even on this approach. Counsel is forced to choose between time
taken to primarily present a case and time taken to challenge the opponent's
case. Time allocated to oral evidence in chief with a view to settling the witness,
giving an indication of veracity and keeping a busy tribunal alert to the issues,
also comes at the expense of other activities. Problems also arise because the
opposing party's witness has the ability to affect the allocated time. Expansive
answers that take up time can simply frustrate cross-examining counsel. A
proactive tribunal will have to consider whether to intervene and demand more
succinct answers. In some cases the allotted time might run out with an earlier
witness and an application might then be made to extend the time of the hearing
to allow other witnesses to be called. There will then be challenges facing a
tribunal asked to revise its chess-clock demands in view of the behaviour of
earlier witnesses. For this reason, a tribunal should confirm at the outset that
chess-clock times may need to be altered as the need arises.

It has also been observed that the method of payment of arbitrators can impact
upon the amount of time made available for oral testimony. An arbitrator on a
lump sum or subject to a more modest ad valorem institutional scale might be
disinclined to a lengthy oral hearing, (18) although this should not be a
determining factor. (19)

page "890"

Finally, should a witness be permitted to give evidence if they have not

submitted a written statement within the time period as required? The general
presumption would be no, but no blanket rule can be suggested in terms of the
exercise of a tribunal's discretion even where rules raise a presumption of
exclusion. The tribunal would need to look at the circumstances as to why a
written statement was not presented and what implications allowance would
have for the due process rights of the other party.

These are just some examples. The overriding point is to ensure that efficiency-
based decisions do not pre-empt due process rights and are taken at a stage
when the tribunal has a sufficient grasp of the key issues. These matters should
all be dealt with at a pre-hearing conference. Such a procedural discussion can
be a useful way ‘to win over parties' counsel as partners in a joint venture with
the tribunal’. (20) Such a discussion ought to cover a range of issues, from
contentious aspects such as cross-examination and time limits, to practicalities
of availability and translation. Considering logistical issues as to witness
presentation also helps to minimise the disruption to the witnesses themselves,
where the tribunal can make it clear which witnesses must be on standby,
depending on the time of conclusion of previous testimony.

12.3.3. Pre-hearing Protocols for Witness Evidence

Lévy and Reed recommend such a special procedural pre-hearing conference

with counsel in appropriate cases, to be conducted shortly before the
evidentiary hearing. They also suggest that efficiency would be maximised if the
tribunal sent an advance list of questions to counsel. They provide an example
draft as follows:

Proposed draft: ‘Notes on Organising Fact Testimony in Advance of Evidentiary

Hearing in International Commercial Arbitration’.

A. IBA Rules.
Do you agree with the basic principles in Art 8 of the IBA Rules (arbitrator
control, normal order of witnesses and questions, affirmation, scope of
 tribunal questions)? If so, scope of conference can be abbreviated. (Note:
This focuses counsel new to international arbitration on the IBA Rules.)
B. Why? Underlying purposes for fact witnesses. page "891"
1. What is your perception as to the main purpose of fact testimony? To test
the credibility of witness statements? Or to maximise fact finding? Or
both?
2. Would you welcome the tribunal indicating in advance what fact
testimony it most wants to hear? If so, by witness or by issue? If not, why
not?
C. Who? Details as to the witnesses.
1. Subject to the time allowed (see below), what witnesses do you intend to
present?
2. Are there witness statements filed for all? Critical amendments to be
made?
3. Order of priority? Comfortable that they are not duplicative?
4. Are there witnesses from the other side you require?
5. Do you have witnesses you wish to call even if the other side indicates it
does not want to cross-examine them?
6. Does any witness require a formal order or subpoena?
7. Do you anticipate bringing witnesses for possible rebuttal? If so, why do
you anticipate a need for rebuttal?
D. When? Scheduling and order issues.
1. In what order do you intend to present the witnesses? All of claimant's
witnesses followed by all of respondent's? Or claimant-respondent by
issue? Or a combination?
2. Are all your witnesses available?
3. Any special scheduling requests?
4. Do you have an estimate, however rough, of how much time each witness
should take?
E. How and what? Method and scope of testimony.
1. Do you agree that the tribunal, specifically the chairman, shall be the
umpire for all questioning?
2. Any special requirements/expectations for affirmation/oaths? Any
Islamic witnesses?
3. Witness preparation: What are your expectations/intentions in relation
to preparing your witnesses to testify? How much time do you intend to
spend in cross-examination exploring the scope of preparation? How
much preparation is likely to lead you to lodge an objection?
4. Sequestration of witnesses? Both before and after testimony? Special
rules for parties' representatives?
5. Progression: Do you anticipate each witness going through direct, cross,
redirect? Re-cross?
6. Progression: Do you anticipate limiting opening direct to 30 minutes, to
 affirm the witness statement and ‘relax’ the witness?
7. Timing: Do you prefer the tribunal to use the ‘chess clock’ or ‘guillotine’
timing system?
page "892"
8. Timing: Does time spent on cross come out of the time ‘account’ of the
sponsoring side or the crossing side? How about time spent on tribunal
questions (asking and answering)?
9. Style: Do you anticipate using leading questions on direct as well as
cross?
10. Style: On cross, do you intend to use a relatively aggressive US approach
or a more conversational approach (subject to the chairman's control?
11. Scope: Do you intend to limit direct and cross-questions to subjects
covered in the witness statement? To object if the other side goes
beyond those subjects? What are your expectations as to the scope of
the arbitrator's questions?
12. Privilege: What are your expectations as to privilege/confidentiality for
witness testimony?
13. Objection: Do you anticipate making objections on relevance? Hearsay?
Privilege/confidentiality? Repetitiveness? Delaying tactics?
14. Objections: How do you envisage making objections? How do you
envision the tribunal should respond and rule on objections?
15. Use of documents: When you question a witness about a document, do
you intend to refer him/her to an agreed hearing bundle or will you use
loose copies (with copies for all)? Do you anticipate spending
substantial witness time on documents, to focus the tribunal on the
record?
16. Visual aids: Do you intend to use new charts/maps/etc with witnesses?
If so, have copies available, as no surprises will be allowed.
17. Are you open to witness conferencing? If so, for which issues or
witnesses? (21)

The authors suggest this only as a starting point, perhaps especially useful for
new arbitrators and subject to modification by experienced arbitrators and on a
case-by-case basis. Whether sent to counsel or not, the list is also an excellent
and comprehensive guide to matters that should be thought about by a tribunal.
The clear advantage of providing these questions prior to the hearing is that
counsel is educated as to procedural matters that may arise, and is invited to
think in advance of how to deal with certain situations in a more relaxed
environment and time frame than the hearings and try and agree upon
procedures. However, a list of potential areas of agreement can also be turned
into potential areas of disagreement, which might encourage some of the
problems to arise where they otherwise would not. Questions in advance can
play into the hands of guerrilla tactics, with counsel taking particularly
intransigent positions on a range of matters and threatening lack of due process
rights if the tribunal makes determinations to the contrary. Counsel will tend to
find it harder to take issue with oral directions of experienced arbitrators in the
middle of a hearing than they might in their own time via documentary
communications on such preliminary questions, although that page "893" would
not be uniformly so. While such forms of abuse of sensible practical suggestions
must always be a concern, these will often be issues facing a tribunal in any
event. In the above example, a tribunal has to decide whether efficiency
demands some indication of the issues of most concern to it. The time set down
for the evidentiary hearing will rarely be long enough to contemplate lengthy
debates about procedural issues. Hence, it is always better to seek to resolve
these in advance and the Lévy and Reed proposal is to be preferred.

A tribunal utilising such questions might also give consideration to the matters
on which it is happy to defer to the parties' preferences and conversely, those
matters that it believes must be presented in a particular way in the interests of
fairness and efficiency. The latter are matters for early notice and not requests
for party agreement. For example, if a tribunal is wholly against leading
questions on direct examination, it would not wish a question list to lead to an
agreement by counsel that this should be allowed. A more contentious example
would be if a tribunal believes that expert conferencing is needed to help it
understand and resolve technical issues. Some assert that this should only occur
with consent of the parties in any event, (22) although this is not the view taken
below.

12.3.4. Notification of Witnesses and Tribunal Directions

There needs to be advance notice as to which witnesses are intended to give

evidence. Article 8.1 of the IBA Rules of Evidence 2010 requires notification
within the time ordered of the witnesses whose appearance a party requests.
This is important as a party may not wish to have every witness who has
submitted a statement attend a hearing. Other rules seek to specify a time
period before the hearing where such notice is required. (23) The UNCITRAL
Notes also suggest that the notification indicates the language in which the
witnesses intend to testify; the relationship with any of the parties; the
qualification and experience of the witnesses and the means by which the
witnesses learnt about the facts on which they intend to testify.

Hwang and Chin provide an example of a direction as to witness testimony

commonly used by one of the authors:

(a) Parties are to prepare statements of evidence in chief (in numbered

paragraphs) containing the full evidence in chief of all witnesses of fact upon
whom they propose to rely. Photographs of the witnesses should be
attached to their respective witness statements if possible. All documents
intended to be referred to in the evidence in chief of the witnesses must be
attached to the statements of evidence in chief and copies provided with the
statements of evidence in chief if not previously provided to the Tribunal.
Statements of evidence in chief are to be filed and exchanged by page "894"
[insert date]. Parties are at liberty to file further statements of evidence in
chief (either of the same witnesses or of new witnesses) only in response to
the original statements. Responsive statements are to be exchanged by
 [insert date],
(b) All witnesses who are giving statements of evidence in chief are to attend
for cross-examination, if requested by the other Party. If a witness so
requested does not attend then, on good cause shown, the Tribunal may
accept the statement and decide what weight, if any, to attach to it. Each
Party is to give the other Party notice whether any of the other Party's
witnesses are not required to appear for cross-examination not later than
[insert date]. If any witness requested to attend cannot attend, notice of
non-attendance must be given at the earliest possible opportunity to the
other Party. (24)

More detailed guidance can be provided where experts are concerned in the
form of a detailed brief. This is discussed in section 12.14.3 below.

12.4. General Witnesses

12.4.1. The Function of General Fact Witnesses

The following sections deal with general fact witnesses. Sections 12.10 to 12.14
below deal separately with the question of experts. Some issues discussed in
this section, such as counsel's involvement in preparing witnesses, apply to both
forms of oral testimony. Nevertheless, the bulk of the discussion in the following
sections is focused on the treatment of general witnesses. The first issue is to
properly understand the role of a general witness. A general witness is there to
provide evidence of facts. It is not appropriate for a general witness to provide
an opinion on a matter to be determined by the tribunal. As has been said by the
International Court of Justice in the Nicaragua case:

The Court has not treated as evidence any part of the testimony given which was
not a statement of fact, but a mere expression of opinion as to the probability or
otherwise of the existence of such facts, not directly known to the witness.
Testimony of this kind, which may be highly subjective, cannot take the place of
evidence. An opinion expressed by a witness is a mere personal and subjective
evaluation a possibility, which has yet to be shown to correspond to a fact; it
may, in conjunction with other material assist the court in determining a
question of fact, that is not proof itself. (25)

page "895"

At times it will be appropriate for a fact witness to state an opinion if it shows

why they acted in a particular way. For example, a buyer of faulty machinery
might terminate a contract after the seller has attempted repairs on a number of
occasions. The buyer might ultimately form an opinion that the seller's repair
staff will never be able to adequately fix the machine in a reasonable period of
time. If the opinion is reasonable, it supports a conclusion that the breach
justifies termination under most applicable substantive laws.

The modern approach with witnesses of fact is to allow the opposing party to
call for cross-examination of a witness if it is not prepared to allow the written
statement to stand unchallenged. A witness who is called for cross-examination
must be available. If they do not attend for cross-examination, the written
statement will be disregarded unless there are exceptional circumstances that
lead to the tribunal directing otherwise. (26) Excluding the evidence of witnesses
who refuse to testify is considered common practice (27) unless the witness has a
compelling excuse. Even then, if the statement is allowed, the lack of cross-
examination may go to weight.

12.4.2. Parties as Witnesses

While some civilian systems do not allow parties to appear as witnesses, (28) no
such limitations apply in international arbitration. (29) While arbitral rules are
generally silent on this question, the norm in arbitration is to allow all parties to
be witnesses. This is confirmed by the IBA Rules of Evidence 2010 where they
apply. (30) It would page "896" be undesirable to exclude a party's testimony per
se. They may be the only person in possession of information that is important
and which in some cases will not even be contested by the opposing party.
Preventing a party from being a witness would lead to other problems where
the parties are in any event entitled to make submissions. It is better to allow
the material and have it tested by the other side. A rule against parties as
witnesses would also be difficult to apply in modern commercial environments
with multiple inter-related companies where there could be gateway questions
as to who in fact is a party.

While party testimony is habitually accepted, its credibility can of course be

tested. Credibility can be affected by a range of factors. For example, at least one
tribunal has noted the lack of perjury provisions which would be an inducement
to veracity in certain domestic litigation. (31) At times the Iran-US Claims
Tribunal distinguished between non-party witnesses and evidence provided by
persons having an interest in the proceedings. The approach was to consider
that witnesses gave ‘testimony’ while interested parties provided ‘information’.
(32) The Tribunal treated the latter as ‘party representatives’ or ‘party witnesses’.
(33) The information so provided was weighed against other evidence but parties

needed to consider the ‘ingrained prejudice’ that some arbitrators might have
against such evidence. (34) This approach was not applied uniformly. (35) If a
person was treated as an interested party rather than a witness, this may also
have impacted upon notice requirements as to witnesses who will testify, the
application of an oath and ability to sit in on proceedings other than when
providing comments to the tribunal. (36) Once again, this is not the better
approach in modern international arbitration.

page "897"

12.5. Witness Statements

It is common for witnesses to provide written statements as to their testimony

prior to the hearing, although this is not required as a matter of course and
there are domestic trends against in some jurisdictions. They have become
prevalent in international arbitration even though such statements are
commonly disregarded as evidence in civil law systems. (37) The IBA Rules of
Evidence 2010 allow a tribunal to call for witness statements to be submitted in
writing prior to the hearing. (38) This is a matter for the tribunal's discretion in
consultation with the parties. There may also be revised or additional
statements responding to matters in the other party's witness statements. (39)

There are a number of advantages and limitations to such statements. By

providing witness statements at an early point in time, the other party can
determine whether there is a need to cross-examine or whether it is possible to
agree on certain facts. In this way the written statements, together with the
written submissions themselves, will help narrow the points at issue that must
ultimately be resolved by the tribunal. Written witness statements also allow
the tribunal itself to determine whether it is necessary to hear witnesses orally.
In some cases, oral testimony may not be necessary because the statements are
irrelevant or unduly repetitious or the written statements from both sides show
that certain facts are not in dispute. Even if the witness is to be heard orally, a
well prepared tribunal can help direct the parties as to the more important
matters or at least impose overall chess-clock constraints properly informed by
the prior written statements. Having the written submissions in advance also
helps opposing counsel properly prepare for cross-examination. It also allows
the tribunal to prepare its own questions for the witnesses. (40) Written witness
statements as well as written submissions also help ensure that preparation is
undertaken early. Too often in domestic litigation in common law countries,
attention is given to minimal pleadings with a last minute flurry in preparing the
case. Such an approach will also commonly leave insufficient time for adequate
settlement negotiations.

page "898"

12.5.1. Statements in Lieu of Oral Evidence

More contentious is the question whether witness statements should be

accepted in lieu of oral testimony. There are likely to be significant cost savings
with written witness statements as opposed to oral testimony. Preparing a
statement in consultation with counsel involves only two people. Oral testimony
in the presence of counsel and parties from both sides, a multi-member tribunal
and transcribers, involves a significant number of people, together with travel
and accommodation costs. Written witness statements can also be read at
whatever speed is considered appropriate and in some circumstances at least,
might lead to a better understanding of what is being said than oral testimony,
although the availability of transcripts in the latter event will at times deal
adequately with that issue. However, the concern as to whether a written
statement was in fact drafted by a party, and the lack of an ability to assess the
veracity and memory of the witness concerned, suggest that efficiency gains
may come at the cost of reliability. (41) Written statements may also be unhelpful
where a more nuanced dialogue with the tribunal will be necessary. Too often
written statements make assertions that need to be explored orally in any event
and hence can be of little probative value even if believed.
For this reason, and as noted above, most arbitral statutes and rules give each
party a unilateral right to call for an oral hearing and an opportunity to
challenge opposing witnesses. Each party should be entitled to call for oral
presentation, for example if they want witnesses to comment on other
testimony (42) or if witness conferencing is called for. Because of this entitlement,
Lévy sees a witness statement as an offer to have a witness appear orally unless
the other party or the tribunal waives this. (43) The tribunal might also want to
hear the witnesses give oral evidence to determine veracity even if the parties
do not call for this. (44) As noted, some rules also expressly indicate that if the
witness is called for but does not give oral evidence, the statement can be
disregarded as evidence.

Where the parties agree that a witness need not appear, this is not taken to be
an admission of the correctness of the contents of the statement. (45) If that was
the presumption, parties would simply call for oral evidence in all cases,
undermining the efficiency value of written submissions. The witness statement
could be page "899" challenged on the basis that it is dealing with opinion not
fact, that it has insufficient particulars to meet the party's burden of proof, that it
is irrelevant, that it has internal inconsistencies or that it is contradicted by
other more probative evidence. (46) Nevertheless, counsel need to understand
that such evidence, if material, needs to be refuted or outweighed one way or
another, so a decision not to test it at a hearing needs to be based on a sensible
case strategy. It should also be understood that the IBA Rules only express the
position where both parties agree that the witness need not be called. In other
circumstances, choosing not to call and challenge an opposing witness will
inevitably have implications as to the weight of that evidence. (47) In ICC Case No.
9333, (48) a witness became ill and unavailable for oral examination. The
Tribunal collected questions from the other party and its own and sent a
questionnaire to the witness and allowed a second witness statement.

12.5.2. Preparation of Witness Statements

Typically, general witness statements are drafted by counsel. If statements were

only drafted by the witnesses themselves, they would often not be written with
sufficient focus, order and depth to assist the tribunal with its assessment.
Problems with witnesses drafting their own statements include conflicting
language and style between different statements when drafted by inexperienced
persons and the difficulty of having the witness understand and address each of
the key issues. Laypersons are also less likely to understand the difference
between attesting to facts on the one hand and rendering opinions or making
submissions on the other. While assistance from counsel is thus understandable,
the value of the statement will be significantly undermined if the tribunal has no
confidence that it is actually the witness's testimony. At the extreme, if there is
too much involvement of counsel the statement may stray too far from the
witness' exact belief. That should not be the case where counsel is concerned to
maintain ethical standards and only assist the witness in accurately and
eloquently setting out their true testimony. This may not always be the case. (49)
There are also no uniform ethical standards to apply to the page "900"
behaviour of counsel from a myriad of countries when involved in international
arbitration. (50)

12.5.3. The Ambit of Assistance of Counsel

An important ethical question is whether counsel is entitled to ask the witness

to consider strengthening the language in a statement in support of the case
strategy. Once again there is a difference between inviting a witness to consider
whether he or she is able to assert a particular proposition, which ought to be
acceptable and conversely, urging a witness to use stronger language than they
were naturally disposed to use. This is discussed further in the context of
witness preparation generally in the following section. It is often difficult to
state a principle which clearly articulates the dividing line between acceptable
and non-acceptable behaviour, although extreme versions of the latter will
usually be readily characterised as such.

Another reason why counsel should ensure that a witness statement reflects the
witness's own views and even modes of expression is that too much disparity
between the written statement and oral testimony can undermine the
persuasive value of the witness if the tribunal forms the view that the witness
must have been happy to sign anything presented by counsel. Legalistic phrases
in a statement from a lay person can raise doubt as to the true author of the
statement. (51) Another problem with not using the witness's own words is that
this can typically create uncertainty and embarrassment during cross-
examination when opposing counsel confronts the witness with a phrase from
the written statement that the witness may not understand or recollect. (52) A
valuable recommendation is to ask the witness to prepare the first draft. (53) That
is desirable although there may then be ethical issues if counsel wishes the
witness to present things quite differently following the draft. There may also be
privilege issues arising if there is a document production request as to the
drafts. (54)

12.5.4. Contents of Statements

It is important to ensure that a witness as to fact limits the statement to facts

alone and does not delve into submissions, which can too easily occur when
drafted by page "901" inexperienced or overzealous counsel. The written
witness statement should not be too long, should not be repetitive and should
not replicate pleadings. The witness should not speculate and should identify
the basis of knowledge. (55) If a witness statement is in fact mere opinion or
pleading, it may be appropriate for the Tribunal to note this at the earliest
opportunity. (56)

A comprehensive witness statement that also deals with the matters likely to be
subject to cross-examination can help with the assessment of veracity. It can
also prevent the feeling that the cross-examination has exposed a hidden
position. It can even undermine the impact of the cross-examination itself.
However, it should not always need to raise matters that the cross-examiner
might be likely to overlook. Judgment must of course be exercised.
If a witness statement is not in the person's natural language, then the counsel
will obviously help in ensuring that it is appropriately expressed. On the one
hand, such assistance aids the tribunal in understanding the witness's actual
evidence when oral presentation through an interpreter can sometimes be
difficult to follow. Such oral testimony is also very dependent on the skills of the
interpreter. On the other hand, a wish to alter the witness' natural expression
mandates greater counsel involvement and raises questions as to the extent to
which it is truly the witness's testimony. There will also be a difference between
drafting a statement in the natural language and having it translated and
conversely, giving instructions for direct articulation in the language of
arbitration. Whichever approach is chosen, the ideal is to present the statement
in a way which will help the tribunal but still keep the essential flavour of the
witness' articulation in a way that will harmonise with the oral testimony.

There is no general form requirement for witness statements and no

requirement that statements be presented on oath by way of affidavit, although
this is utilised in some legal systems. It is more common that witness statements
are merely signed by the witness. Lévy (57) suggests that a witness statement
should:

1. include personal information

2. perhaps include a photograph
3. state that the witness knows the use to which the statement is to be put
4. contain an affirmation and provide knowledge of the consequences of a
misrepresentation.
page "902"

The suggestion as to a photograph is that it would help busy arbitrators to

recollect their impressions as to the veracity and quality of the testimony when
deliberations only occur quite some time after the hearings. For similar reasons,
it may even be desirable to video testimony in large cases where the expense
would not be inappropriate. (58)

A tribunal might direct that a witness statement should contain an indication of

the nature and extent of any assistance provided in preparation. (59) The IBA
Rules of Evidence 2010 require the witness statement to contain, in addition to
the above, a statement regarding present or past relationship if any with any of
the parties; background qualification, training and experience if such a
description may be relevant to the dispute or the contents of the statement; and
a full and detailed description of the facts and the source of the witness'
information as to those facts sufficient to serve as that witness' evidence in the
matter in dispute. There is also to be a statement as to the language in which the
statement was originally prepared and the language in which the witness
anticipates giving testimony. Finally, the witness is to sign with a note as to the
date and place of signature. (60)

A decision needs to be made whether to include documents as part of the

witness statements or to separately produce documentary evidence. Note was
taken of suggestions to require inclusion of such documents in the draft
direction of Hwang and Chin. (61) While the norm would be to attach documents
to a witness statement, this may be affected by procedural orders as to
document production. If documents are attached to a witness statement and
were not produced by a party, this raises questions as to how a Tribunal should
treat this. Lévy suggests that attachment of documents should not be allowed to
add new factual allegations or bypass discovery time limits, (62) although this
may not always be practical and there are many cases where both parties are
still tendering documents close to the hearing. Nevertheless, this is an important
warning as the plain meaning of Article 4.5(b) of the IBA Rules of Evidence 2010
simply states that ‘(d)ocuments on which the witness relies that have not
already been submitted shall be provided…’. This should not be problematic if
the tribunal ensures that the timing page "903" of witness statements is in
harmony with cut-off dates for document presentation and production.

Regardless of whether relevant documents are appended, key parts might also
be extracted in the body of the statement so that the gist of the statement is
easily understood by the reader. It will be less time-consuming for counsel to
ensure an appropriate amount of referencing to documents rather than have
three tribunal members sift through bundles of documents and collate these
with the statements. If one witness statement is responding to another,
appropriate cross-referencing to paragraphs will also assist the tribunal. This
can also occur if a number of statements are being drawn by counsel for one
party where they could cross-reference each other to support the testimony and
save time for the tribunal. (63) However, there will be problems if modern word
processing leads to various witness statements using identical language even
where they are attesting to the same matters, as this can also undermine the
belief that it is the witness' real statement. In a very complex and lengthy
statement, it may be useful to include an executive summary. (64) The danger in
drafting such a summary would be to accurately encapsulate the more detailed
testimony. There may be a danger of over-reliance on the summary as opposed
to the detailed statement. If an executive summary was used, it would be useful
to cross-reference it easily to the relevant paragraphs of the witness statement.

12.5.5. Witness Statement Checklist

Based on views of the eminent arbitrators referred to above and the suggestions
in the IBA Rules, the following is a possible checklist for arbitrators in
considering directions to general witnesses as to the content of their statements.
It is not a recommended list as such but rather, a broad checklist of items that
might be relevant in an instant case. Such a direction might include a
requirement that a witness statement contain some or all of the following:

1. Personal information.
2. Background qualification, training and experience where relevant to the
dispute or the contents of the statement.
3. A statement regarding present, past or proposed relationship if any with any
of the parties or with anyone closely connected with any of the parties.
4. A photograph (if considered desirable).
5. A statement that the witness knows the use to which the statement is to be
put. page "904"
6. An affirmation as to truth and an indication of knowledge of the consequences
of a misrepresentation.
7. Documents on which the statement is reliant and documents that will be
alluded to in oral testimony.
8. A full outline of the factual evidence in chief.
9. A description of the source of the witness' information as to those facts.
10. An indication of the language in which the statement was originally prepared
and the language in which the witness anticipates giving testimony.
Translation pagination should follow the original.
11. An attestation that it is the witness' own factual evidence. The statement
could confirm the extent of any assistance given in the preparation of the
statement by counsel or other persons and confirm that the witness has not
changed what it believes to be true at the request of a party, party's counsel
or any other person acting on behalf of a party.
12. Date of signature.
13. If appropriate under the relevant rules, certain form requirements must be
met, for example, whether the statement is presented by way of affidavit,
although this would be rare.
14. A direction that paragraphs should be numbered and that references to
documents and submissions should use a particular method of referencing as
designated by the tribunal. Of particular value is prefixing the documents and
submissions C and R respectively.
15. A direction as to an executive summary if required.
16. If the direction is as to a second round of statements, the direction may state
that the second round may only pertain to information contained in the other
party's previous statements.

12.5.6. Simultaneous or Sequential Exchange of Witness Statements?

Similar issues arise in relation to the timing of witness statements as arise in

relation to the timing of submissions themselves. The latter will typically
predate the witness statements. Section 6.14.2 deals with the question of
whether written submissions should be ordered sequentially or concurrently.
There is often a strong argument for sequential ordering of submissions if one
party's allegations need to be evaluated by the other before a response. The
same logic does not necessarily apply to witness statements. Once all claims,
defences and counterclaims are known, witness statements can simply deal with
the evidence that each wishes to present on each issue. Hence, there is a
stronger argument in favour of simultaneous exchange. This will reduce delay
and promote equality. However, simultaneous exchanges may make it harder
for the tribunal to integrate such material and the submissions may be ‘ships
that pass in the night’. Integration might be easier when one statement focuses
directly on the points made in another, which requires sequential ordering. In
some cases a tribunal might organise two rounds of page "905" simultaneous
exchanges to optimise the above competing considerations. (65) In the latter
event, the second round might be directed to only address information
contained in the other party's statements, although this would be hard to
enforce and most would not constrain counsel in this way.

If simultaneous lodging is ordered, there is a risk that one party may delay,
receive the other statements and then amend accordingly. A tribunal can ask for
statements to first be served on itself or an institution and only exchanged when
both are available.

12.6. Witness Preparation

12.6.1. Introduction

Domestic legal systems vary greatly as to the degree to which counsel may
interview and prepare witnesses. Some legal systems even stipulate that it is
unethical for counsel or parties to contact witnesses prior to them giving
evidence. This relates both to oral testimony and to written statements,
although assistance with written statements is now fully accepted, subject to the
cautions noted above as to the need for the statement to truly remain that of the
witness. The issues addressed below apply equally to party-appointed expert
witnesses as to general witnesses.

12.6.2. Interviews

It would be undesirable if counsel from differing legal families approached the

question of witness preparation based on their domestic litigation experiences,
without understanding that opposing counsel could at times be likely to take a
different approach. Equal treatment suggests that the tribunal should make
clear its attitude, either in the terms of reference, or in a pre-hearing conference.
(66) Notwithstanding that parties may have different legal traditions, to treat

them differently on this issue would offend against principles of equality of

treatment. (67) The norm in international arbitration is to allow witnesses to be
interviewed and prepared. (68) While interviews and preparation are allowed,
the more a witness page "906" appears to be coached or the more it appears
that the statement is not really theirs, the less weight will be given to their
testimony.

The IBA Rules of Evidence 2010 expressly indicate that it is not improper for a
party, its officers, employees, legal advisers or other representatives to
interview its witnesses or potential witnesses. (69) The Rules expressly allow
prospective testimony to be discussed. The drafters of the new IBA Rules chose
not to provide further guidance on permissible interaction between witnesses
and counsel. (70)

12.6.3. Interviewing Opposing Witnesses

There is no arbitral rule against counsel contacting the other party's witnesses
although there is no express rule in favour either. A tribunal might retain
discretion to prevent this if there was a valid reason for doing so, although there
is no obvious sanction to apply if the tribunal is ignored. The first issue is
whether counsel actually knows that a person is a witness for the other side. A
second issue is whether the witness is also a party (or an employee or officer of
the party). National bar codes could impact upon counsel's powers but will not
be determinative from the tribunal's perspective.

It has been suggested that a proscription against approaching the other party
‘behind its counsel's back’ would, by analogy, apply to that counsel's witnesses.
(71) This does not flow as a matter of plain meaning from rules and guides or

necessary policy. Refraining from interfering with a person represented by

another lawyer may simply be a protection of that lawyer/advisee relationship.
This is not necessarily the prime consideration for an arbitrator dealing with
broad powers under a lex arbitri and arbitral rules. Stated differently,
preserving lawyer/advisee relationships does not necessarily trump arbitral
rights. It has been suggested to the same end that Article 4.3 of the IBA Rules of
Evidence 2010, indicating that it shall not be improper for a party and advisers
to interview ‘its’ actual or potential witnesses, suggests a contrario that the
other party's witnesses cannot be approached. (72) It can page "907" certainly be
inferred that the Working Party did not wish to expressly sanction this, although
they did not expressly proscribe it either. Another uncertainty is the breadth of
the notion of a ‘potential’ witness and whether one party can have an exclusive
entitlement to a central witness simply by being the first to approach the
person. This should not be the case and it will inevitably be a matter for tribunal
control where there are disputes as to the entitlement to interview opposing or
prospective witnesses.

The same principles should not apply to a party-appointed expert. Opposing

experts should not be approached. They have entered into a contractual
relationship as part of the opposing team, albeit with duties of independence.
Denying opposing counsel access to such an expert prior to the hearing cannot
be a denial of due process as might be arguable if access to a key factual witness
was barred as each party can access other experts.

12.6.4. Preparing Witnesses for Oral Testimony

Those opposing witness preparation before oral hearings argue that lawyers
cannot refrain from coaching witnesses to a point where the testimony is not
really theirs. Those supporting preparation argue that the legal process is too
unfamiliar to laypersons. The system should not allow a non-expert witness to
come face-to-face with an expert cross-examiner without some legitimate
guidance as to the reasons that they are being called, the key points to get across
and the areas where they are likely to be challenged. As always, there are
advantages and disadvantages with each position. Neither approach can
inherently be seen to optimally promote truth with maximum efficiency.

Article 4.3 of the IBA Rules of Evidence 2010, in allowing witness interviews to
‘discuss their prospective testimony’, is broad enough to suggest the entitlement
now seems settled. Nevertheless, witness preparation does pose added ethical
challenges for counsel to ensure that they do not overstep the mark. Certain
forms of witness preparation which merely aim to educate them about the
process should be less problematic. For example, in preparing a witness for
cross-examination, it is reasonable for counsel to explain to witnesses that they
should carefully consider questions, seek clarifications when they are not
understood, correct factual errors in questions, limit answers to the questions
and maintain a calm demeanour. If an experienced witness understands how to
clarify ambiguous questions, limit answers and the like, there seems no reason
not to educate less experienced witnesses in this manner. There is obviously a
significant difference between, on the one hand, explaining the issues to a
witness, explaining the role that the witness will play in the proceedings, and
warning them about the kinds of matters that may be raised by the tribunal or in
cross-examination, all acceptable aspects of witness preparation, and on the
other hand, detailed coaching as to a script drafted by counsel to be used in
answers to key questions. The latter would be unethical on any view of that
notion. It would also be grossly unethical to invite a witness to lie or encourage
the witness to leave out references to adverse facts that page "908" are
nonetheless central to their intended testimony. If a witness seems too
rehearsed, their credibility is likely to be affected in any event. That might be
even more so where the arbitrators are from a civilian law background and are
less used to and more suspicious of witness preparation.

C Mark Baker has suggested that in preparing witnesses for cross-examination

‘the attorney should inform the witness about the issues and the facts that he or
she may be questioned about during the examination. Key documents and
exhibits should be reviewed with the witness, and the witness should be
prepared for the kinds of questions that opposing counsel may ask depending
on the seat of arbitration and the background of the tribunal’. (73) It would also
be appropriate for counsel to prepare expert witnesses for conferencing where
that is used and where they have no experience with this technique. Lawrence
W. Newman also suggests that the witness ‘should be prepared for the kinds of
questions he or she will be asked on cross. Care must be taken not to provide
“canned” answers to such questions – much less coach the witness into
providing devious or misleading answers to anticipated questions. But the
witness can and should be permitted to think back on the facts underlying his or
her testimony and recall the circumstances about which questions might be
asked’. (74) It has also been suggested that witnesses should be taken through
their witness statements on a large projector screen which better focuses their
attention on their words and prepares them for cross-examination. (75)

The above suggestions outline some clear and reasonable distinctions between
acceptable and non-acceptable practices. However, a more difficult question is
whether counsel should be able to hold a practice session aiming to direct the
witness to the kind of comments that would be more successful. There are a
range of other potentially challenging ethical dilemmas. There will always be
ethical issues at the margin. For example, in addition to differentiating between
introducing the witness and leading the witness, the more that counsel engages
in thorough preparation of witnesses, the more counsel might identify
inconsistencies between proposed testimony and prior statements. Ethical
issues again naturally arise in such circumstances. (76)

A tribunal may wish to give guidance to civilian practitioners who may be less
familiar with witness statements and who might have a natural reluctance to
page "909" engage in witness preparation. A tribunal can spell out what it
wishes to receive and what it feels is desirable for counsel. (77)

12.6.5. Codes of Conduct and Obligations of Counsel

National bar codes dealing with proscriptions against contact with prospective
witnesses will commonly exclude arbitration. In some cases ethical standards
may indicate that they apply to any proceedings within their geographical
location. They may also expressly apply to arbitration. Even then, a professional
code from private lawyers in one jurisdiction cannot necessarily be binding on
the behaviour of lawyers in other fora and on international arbitrators
conducting an arbitration in that Seat. The situation would be different if a
comprehensive code of conduct is promulgated by the national government of
the Seat, in which case its provisions must be integrated with those of the lex
arbitri. The provisions may indeed be seen as part of the lex arbitri.

Another uncertainty with national codes of conduct is whether they seek to have
extraterritorial effect and regulate the behaviour of counsel in foreign
jurisdictions. There may be good reasons for professional bodies to wish to do
so. However, it is not clear that this should apply to arbitration, so
extraterritorial effect cannot be presumed where the rules are silent. In any
event, if it is a mere professional code, it cannot apply as of right to constrain the
control of a foreign arbitrator. (78) Where international arbitration is concerned,
there is always the need to separately consider counsel's obligations and a
tribunal's rights and obligations as the tribunal cannot be said to be bound to
uphold differential national bar ethics rules over counsel who happen to have
been selected to appear before it. Some have suggested that there ought to be
rules of conduct applicable to arbitration, at least with similar status to the IBA
Rules of Evidence and the IBA Guidelines on Conflicts of Interest to assist
tribunals. (79) But regardless of the merits or otherwise of such a proposal, it is
not likely to eventuate in the short term. There are as yet no general codes of
conduct for counsel in arbitrations.

12.6.6. Tribunal Responses to Inappropriate Coaching

Before a tribunal needs to consider a response to inappropriate coaching, there

needs to be enough evidence that this has occurred. Such issues can arise for
page "910" determination by a tribunal if on cross-examination, the witness
admits to a contentious degree of coaching and an application is thus made to
reject the witness's testimony or even to bar opposing counsel on the basis of
unethical behaviour, although this would be rare.

If the witness has been improperly coached, a tribunal is entitled to discount the
weight of the evidence, but what if the only allegation is that counsel has
approached the witness contrary to any legal entitlement to do so? A mere
approach, without any presumption of influence, does not logically discount the
veracity of the evidence, although countervailing arguments may need to be
balanced on a case-by-case basis.

12.6.7. Witness Access to Key Documents

Another area of policy contention relates to whether witnesses ought to be

given access to all relevant documents from all parties well before giving
evidence. While the norm in arbitration is to ask parties to include copy
documents on which they intend to rely at the earliest opportunity, the
important question is what rights a party has to strategically use some
documentation not previously notified. Must a document that will only be relied
on to undermine an opposing witness be disclosed? Where cross-examination is
concerned, the arguments in favour of full disclosure are that a truthful witness
will, as a result, be best able to consider the documents carefully, think about
what are often historical circumstances and be best able to efficiently and
articulately present their testimony and respond to cross-examination. The
contrary concern is that a dishonest or strategic witness is best identified
through the ability to challenge them with contemporaneous documents on the
spot, without an opportunity for them to prepare strategic and untruthful
responses. As with all complex policy questions, there is no obvious answer as
the competing concerns are both real.

There is no norm on this issue and arbitrators may have opposing views,
although the notion of providing documents on which one intends to rely seems
broad enough to encompass documents which will be relied on in cross-
examination. Furthermore, parties must respond fully to tribunal directions as
to document production. Some arbitrators request that, at a specified time
before cross-examination, each counsel is to provide documents upon which
they intend to rely during cross-examination. (80) The approach a tribunal takes
may also vary depending on whether the documents are in the public domain.
(81)

page "911"

12.7. Depositions

Depositions by witnesses as used in the US are not common in international

arbitration. (82) Nor are they expressly referred to in the Federal Arbitration Act
(US). The ICDR Guidelines indicate that depositions, interrogatories and
requests to admit as developed in American court procedures are generally not
appropriate for international arbitration. (83) The IBA Rules of Evidence 2010 do
not expressly refer to depositions. Conversely, the CPR Protocol allows for such
an approach in the discretion of the tribunal. The parties are of course free to
agree on any process even if not standard to arbitration. It is arguable that a
tribunal's broad discretion would allow it to order depositions, (84) but it would
be rare for this to occur. As a precaution with parties from differing legal
systems it is prudent at the preliminary meeting to establish what the party
expectations are.

Section 2(c) of the CPR Protocol explains the nature of depositions:

Depositions are recorded sessions at which witnesses are questioned by the

parties outside the presence of the tribunal, enabling the parties to obtain
information from witnesses in advance of their testifying at the hearings.
Depositions should be permitted only where the testimony is expected to be
material to the outcome of the case and where one or more of the following
exigent circumstances apply: witness statements are not being used, the parties
agree to the taking of the deposition and/or the witness may not be available to
testify, in person or by telecommunication, before the tribunal. The tribunal
should impose strict limits on the number or length of any depositions allowed.
Deposition transcripts may, as the tribunal determines, be used at hearings or
otherwise be made part of the record before the tribunal.

A key disadvantage is the lack of control by the tribunal as it is not present

during the process. When forced on parties against their will, it may contradict
the spirit of arbitration. An advantage of depositions is reducing cost and
allowing testimony from parties who would find it difficult to come to the place
of arbitration. Another value in using depositions might be where there is no
alternative means to preserve the testimony of an important witness. (85) In
many cases, however, video or telephone evidence before the tribunal would be
a far preferable alternative.

page "912"

12.8. Hearing Witnesses

12.8.1. Order of Examination

Typically, witnesses of fact are examined first by counsel for the party
presenting the witness, then cross-examined by the other party, and then are re-
examined by their party's counsel. In some cases a further round of cross-
examination might be permitted in relation to discrete matters raised on re-
examination. This order is not required, but is emerging practice. The Iran-US
Claims Tribunal tended to follow the civil law inquisitorial approach to receiving
oral testimony rather than allowing for direct and cross-examination. (86) There
is also the possibility of witness conferencing, that is, hearing a group of
witnesses together. This is discussed further in sections 12.13.11–14 below.

The IBA Rules of Evidence 2010 suggest what would ordinarily happen at an
oral hearing. Article 8.3 indicates that claimant would ordinarily first present
the testimony of its general witnesses followed by the respondent doing
likewise. After that, any other party may question the witness in the order
determined by the tribunal. The party initially presenting the witness shall
subsequently have the opportunity to ask additional questions on the matters
raised in the other party's questioning. After that the claimant would ordinarily
present testimony of party-appointed experts followed by respondent. There is
also an issue as to how the tribunal will integrate its own questions, if any, with
those of counsel. The typical approach in modern arbitrations is to allow counsel
to primarily direct evidence in chief and cross-examination, with the tribunal
interjecting where appropriate and utilising follow-up questions as well. (87)

12.8.2. Examination in Chief

In many cases, the witness will begin by presenting the tribunal with their key
personal details, name, address, occupation and expertise; acknowledge that any
written witness statement is theirs and is true and correct, if necessary amend
any errors which may have been found on reflection; and acknowledge, whether
under oath, affirmation or otherwise, their understanding of the need for truth
and confirm that the evidence that they intend to give will be truthful. Some of
this may simply have been dealt with through a curriculum vitae attached to the
witness statement and would not need to be repeated.

page "913"

Witnesses may be permitted access to their statements, documentary evidence

and appropriate notes. (88) In some cases, such notes may need to be disclosed
for examination. (89)

12.8.3. Matters to Cover

Direct testimony should normally be limited to the matters contained in the

witness statement. Evidence in chief should also be allowed to deal with new
developments since the time of the witness statements, (90) subject to tribunal
approval, although there would be a concern if this allows for surprise elements
to be introduced to the detriment of the opposing party. If the tribunal had
previously warned that no new material can be presented other than in
exceptional circumstances and only with the tribunal's approval, and that any
such proposed new material be notified promptly, this may alleviate this
concern.

12.8.4. Forgoing Oral Presentation

One method of reducing costs has been to develop a practice whereby

examination in chief is not conducted orally. The witnesses' written statements
suffice for that purpose and the hearing is limited to questions from the other
party's counsel and from the tribunal and where appropriate re-examination to
cover matters that were raised afresh during cross-examination. (91) One
advantage is that written statements are a more efficient means of presenting
evidence in chief than oral testimony. By cutting down the time for direct oral
evidence, this frees up more time for cross-examination and submissions,
particularly where chess-clock arbitration is employed. The advantage in
reducing the length of oral presentation is to save time, cost and avoid
duplication. The disadvantage of entirely removing any direct oral presentation
of evidence is that an adjudicator forms a view about veracity and expertise
based on the way a witness presents. (92) If an adjudicator is only listening to the
more adversarial form of cross-examination, an unfairly negative view might be
formed. This is particularly so where the witness is speaking in a second page
"914" language and is at the mercy of highly skilled common law cross-
examiners. There may also be psychological implications if witnesses are only
subject to cross-examination, particularly where the parties or key officers of
the parties are involved. A key witness will often wish to be able to express the
story in an inter-personal way and not just in writing. Skilful cross-examination
would not allow for such a narrative. Another problem in only hearing cross-
examination rather than direct examination at the oral hearing is that it might
be given more weight than the written statements. (93) There are added
difficulties if the witness statement stands in lieu of evidence in chief and the
cross-examination simply fails to test key issues. Should that witness' views thus
be presumed correct? How will an accurate evaluation be made? Would this be
unfair to counsel from civilian backgrounds?

Another problem with no evidence in chief is that some busy arbitrators might
not have properly read the witness statements. A short oral outline can ensure
they better understand the case being made. A tribunal will thus often benefit
from hearing some of the key elements orally to allow the witness to relax
before cross-examination, become used to the process, and provide an
indication of veracity and expertise and the degree to which the witness
statement is truly theirs. The way statements are drawn will also impact upon
the need if any for direct testimony. If it is simply a brief summary of the
testimony to be given, direct oral evidence will be required. The corollary is that
if the statement is to stand in lieu of direct evidence, counsel will need to be very
careful to ensure that it is comprehensive and sufficient.

If the parties have already agreed that no witness will give oral evidence in chief
but will only be available for cross-examination, it seems unreasonable to allow
a party to demand oral presentation when cross-examination is not called for.
(94) Nevertheless, it is conceivable that in some cases that would be proper under

a right to fully present one's case. This is particularly so when a claim is a

difficult one and the veracity of the witness is crucial. Again this is a matter that
can be dealt with at the outset. One party could still ask for its witnesses to
present orally even if the tribunal has recommended that statements stand in
lieu of evidence in chief. A challenge should not usually be successful if such a
request is declined but a tribunal might still consider such a request favourably.
While the tribunal has a duty to treat the parties equally, the same offer could be
made to each, whether it is accepted or not. Tribunals still have control over
total time and parties can be given flexibility within total time parameters.

Other problems arise in terms of the ideal approach to examination in chief if

the witness statement is incomplete as alluded to above. This is worth further
page "915" exploration in terms of a tribunal's duties and discretions. There is a
significant policy difference between statements that deal with issues in an
incoherent way and statements which fail to address a key issue. Each poses
distinct problems. The latter raises a question as to whether a witness may deal
with a new matter not properly notified beforehand. A tribunal would be
entitled to reject such evidence in the event of an application made, although it
may be accepted if there is no unfairness or if that can be addressed with costs
orders and adjournment after considering submissions. Rejecting the evidence
could also lead to due process challenges if events in the proceedings made the
new evidence more material than previously thought. That is addressed further
in section 10.22.3 dealing with new evidence generally.

As to the use of written statements in lieu of evidence in chief, it is the other

situation mentioned above that is of concern, where the statement addresses a
matter but in an ambiguous, unclear or incomplete manner. An example would
be an allegation that a representation was made without identifying the person
alleged to have done so. If the written witness statement is inadequate it
certainly makes sense for this to be expanded upon in oral testimony. If it is the
tribunal that sees the inadequacy, some direction might be given to ensure that
the party has a full opportunity to present its case. However, too much guidance
puts the arbitrator in the position of counsel, helping to develop the case. A
tribunal thus has a difficult choice as to whether to give directions or not. The
line is a fine one, particularly when the inadequacy in a witness statement could
simply lead to the conclusion that the party relying on that witness has failed to
meet its burden of proof on a key issue. Situations may vary from those where
opposing counsel shows this to be the case through cross-examination or
submissions and conversely, where the cross-examination or submissions fail to
address possible inadequacies to any degree.

Use of PowerPoint by witnesses should be limited to situations where it will

help a tribunal understand complex matters and not where the PowerPoint
seeks to articulate most eloquently the witness's assertions. In the latter
context, they are more akin to written witness statements which can have too
much involvement of counsel. (95)

12.8.5. Cross-Examination

While some civil law jurisdictions do not allow cross-examination, this process
is now the norm in international arbitration. A question may even arise as to
whether a party has been denied a full opportunity to present a case if it has
been refused the right to cross-examine or refused the right to cross-examine as
extensively as it wishes. (96)

page "916"

While cross-examination is now common, the more aggressive approach to

cross-examination, seeking to challenge the veracity and expertise of a witness,
might be considered unseemly by some civilian arbitrators and hence be a
questionable strategy. (97) Obviously as cross-examination becomes the norm in
arbitration, that may be less of a concern. Extreme forms of cross-examination
gamesmanship that are more about confusing the witness than eliciting the
truth would be frowned on by competent arbitrators from any jurisdiction. (98)
One aim of cross-examination is to distinguish between true recollection and
mere speculation or submissions. It is not simply about differentiating between
truth and lies.
Tribunals also need to consider when to intervene if counsel undertaking cross-
examination is inexperienced in the process. A common fault is to fail to
challenge a witness' answers with probing follow-up questions. A tribunal also
has a difficulty if an experienced cross-examiner concentrates on less important
material in the hope of undermining credibility on that aspect with a view to
undermining the overall testimony. A tribunal also needs to think about whether
it will interpose its own questions or allow counsel time and space for the cross-
examination strategy to be developed. A tribunal may also need to carefully
manage cross-examination where interpreters are being involved. (99)

12.8.6. Re-examination

A tribunal should ensure that re-examination only deals with matters raised in
cross-examination and does not deal with other matters. Nor should it merely
repeat evidence in chief. However, if the cross-examination has led to a distorted
body of evidence, a sensible re-examination can restore coherence.

12.8.7. Order of Witnesses

As with much of arbitral procedure, the parties are generally able to agree on
the order of witnesses. Even if all parties do not agree, it is also the norm that an
page "917" individual party might naturally select the order of its own
witnesses. This flows from their entitlement to have an adequate opportunity to
present their case. It is for them to determine how to optimise the case. The
corollary may be that it is for them to decide on the most effective order of
witnesses. However, the tribunal is the ultimate determiner of facts. The
tribunal also has a duty to promote fairness and efficiency and ensure that the
procedures aid it in achieving that outcome. Hence, in some cases, a tribunal
might prefer a different order to that of the parties collectively or individually
and may direct the parties accordingly. (100) Pro-active guidance as to the matters
of greatest importance can also help the parties select the most useful order.

One situation where a tribunal might wish to vary the preferences of the parties
is if it wishes to have all witnesses on a particular issue, whether from claimant
or respondent. The norm would be for claimant to provide all of its evidence
after which respondent replies. In a complex matter, however, the tribunal
might feel better able to understand and address the key issues if witnesses are
grouped on a topic basis rather than on a party basis.

12.8.8. Recalling Witnesses

In some cases a party might apply to recall a witness previously heard to deal
with a matter addressed in later testimony. This should be a matter for tribunal
discretion under normal principles of fairness and efficiency. The more that the
material being responded to is surprising, the more justification for allowing
this, as the surprise material should have been notified earlier in any event.

12.8.9. Conferencing and Fact Witnesses

Witness conferencing involves hearing groups of witnesses together and allows
for a dialogue between them, co-ordinated and controlled by the tribunal.
Instead of the tribunal hearing conflicting evidence separately, a tribunal might
feel better able to resolve conflicts if the witnesses have a controlled dialogue
about their differences. While the advantages and disadvantages of witness
conferencing apply both to fact and expert witnesses, the more detailed
discussion is left to the latter in sections 12.14.11–13.

Raeschke-Kessler supports the use of witness conferencing for fact witnesses.

(101) Similar views are presented by Wolfgang Peter and Clifford page "918"

Hendel. (102) Nevertheless, tribunals do not generally utilise conferencing

techniques for fact witnesses as opposed to experts, absent party agreement.
Some counsel might assert that this unduly deprives counsel of the ability to
present the case as they see fit and hence is a denial of due process. That should
not be so if counsel is given a separate and appropriate opportunity to raise
matters from their own witnesses and challenge the evidence of opposing
witnesses. There are a number of cases where conferencing of factual witnesses
could be particularly beneficial. A not atypical example arises where there is a
breakdown in some joint-venture or partnership arrangements, where the
commercial documentation is inadequate or at best ambiguous, and where the
key issues are the true intent of the parties to be discerned over a range of pre
and post-contractual meetings. Having a number of personnel present evidence
sequentially may be far less insightful than a discussion between the witnesses
present at the meeting that attempts to reconstruct the discussions and discern
the cause of differing perceptions under the careful control of the tribunal.

A problem with conferencing and factual witnesses is that the recollections of

one may be influenced, even unintentionally, by hearing the evidence of other
witnesses. (103) Another issue is that it is harder to keep each witness within the
confines of their written statements if conferencing occurs. A further problem as
compared to expert conferencing is that in theory if not in practice, party-
appointed experts may eventually come to a point of agreement at least as to a
methodology for further dispute resolution. Conversely, witnesses of fact,
particularly when they are the parties or parties' employees, are likely to remain
opposed and a conferencing process could, therefore, have an adversarial
tendency that requires particularly sensitive management by the tribunal.

Importantly, all suggestions for dealing with some of the more difficult aspects
of fact finding in international commercial arbitration need to be looked at
alongside the alternatives. If there was one optimal model, all sophisticated legal
systems would have adopted it. Instead, there are costs and benefits of each
model, in terms of both fairness and efficiency. Importantly, many of the
perceived problems of conferencing of fact witnesses are inherent problems in
party-nominated witnesses who know that they are there to advance the
interests of the appointing party. Redfern and Hunter suggest that, at the very
least, it should generally not be used as an alternative to cross-examination. (104)

page "919"
12.8.10. Oaths, Affidavits and Affirmations

Some legal systems compel the use of oaths when witnesses testify. Some
merely empower an arbitrator to do so and provide discretion whether to do so
or not. (105) Some laws prevent an arbitrator from administering oaths (106) as
some countries would find it improper for this to be undertaken by someone
other than a judge or notary. (107) In some jurisdictions which prevent an
arbitrator obtaining oaths, court assistance might be sought. (108) The lack of a
uniform rule is partly cultural but also arises because the practical utility of
oaths is thought to be greater where perjury under oath comes with significant
criminal sanctions. Perjury statutes rarely apply to arbitration, other that in
England, where the tribunal is expressly provided with power to administer any
necessary oath or take any necessary affirmation. (109) Others may be more
sanguine and believe that liars are undeterred by sanctions.

The ICSID Arbitration Rules provide for a specific declaration:

I solemnly declare upon my honour and conscience that my statement will be in

accordance with my sincere belief. (110)

The IBA Rules of Evidence simply call for a witness of fact to affirm that they
commit to tell the truth. (111) Some arbitrators will follow this approach. Others
may advise the witness of criminal sanctions for false testimony that may apply
where the arbitration physically takes place. Care should be taken not to make
legal assertions that may be debatable as to the application of such laws. Some
will page "920" ask the witness whether they have any religious beliefs and seek
to administer an oath if permitted where the answer is affirmative.

A failure to administer an oath where required may have different implications

in differing jurisdictions. Poznanski suggests that such a failure would not allow
for an award being set aside under the US Federal Arbitration Act but may do so
under English law, (112) although this would be rare in more recent times.

12.8.11. Interpreters

Witnesses will generally be heard in their native language unless they are fluent
in the language of the arbitration. If interpreters are required, a tribunal needs
to control how these are to be selected. An interpreter should be both
competent and independent, although the latter is not always possible. The
parties are typically asked to confer and also indicate how the fees are to be paid
and whether such fees are intended to come under an ultimate costs award.

It is desirable that both parties and ideally the tribunal each have at least one
person who understands the foreign language to ensure that accurate
interpretation is occurring. Less experienced interpreters often fail to
understand that their role is merely to translate and not to engage in side
discussions so as to explain questions to the witness. A tribunal may wish to
think in advance as to the way it will respond if there are disputes about the
quality and accuracy of the interpretation. There may be a need for a sentence-
by-sentence analysis in extreme cases where the testimony is crucial.

12.8.12. Transcription

Transcripts of evidence are common in large matters although no rules require

this as a matter of course. If post-hearing submissions are allowed, the parties
would also typically wish to have transcripts sufficiently in advance so that final
submissions can integrate all relevant references. An alternative in smaller cases
would be to simply record the proceedings and only refer to the recording
where necessary. Legal costs of listening to tapes need to be compared to
transcript fees. The civilian court approach of having the adjudicator draft
summaries of testimony (113) is not the preferred mode in international
arbitration. Transcripts were not common before the Iran-US Claims Tribunal.
(114) If transcripts are utilised, parties might agree that page "921" witnesses be

given an opportunity to check the accuracy. The tribunal might itself direct that
transcripts are provided to the parties who are entitled to correct errors. A
witness may only correct an inaccurate transcription and must not aim to recast
the testimony. (115)

If the parties cannot agree to a transcript and the tribunal does not order one,
circumstances may arise where one party alone provides a transcript. There
may be problems with the evidentiary value of such a document. (116) It would be
appropriate to direct that such transcripts and their source tapes be provided to
all for comment. (117)

12.8.13. Video Evidence

Modern technology means that in many cases, it ought to be appropriate to hear

evidence of a witness who is not physically present, as long as the audio and
visual link is good enough to allow the arbitrator to fully understand and
evaluate the veracity and expertise of the witness. A related question is whether
a party should have a right to present its evidence in this way in order to reduce
costs, or whether the other side has an automatic right to demand physical
presence for the purposes of cross-examination. While there are no strict rules
in that regard, most arbitrators would expect physical presence in such
circumstances, but practices may change as technology improves. (118)

The IBA Rules of Evidence 2010 provide that a personal appearance shall be the
presumption unless the tribunal allows the use of video conference or similar
technology. (119) The Commentary indicates that an application for permission to
do so should indicate the reasons why the person is unable to appear and should
propose a protocol. The tribunal should seek to ensure fairness and equality and
have the technology ‘approximate live testimony’. The tribunal should ensure
that the technology is of sufficient quality and include a fallback plan in case
technological problems arise, typically a teleconference. Consideration will also
need to be given as to how exhibits would be shared. There would also be a need
to ensure that the person is giving evidence under the same conditions as they
would be if present, for example having no better access to preparatory notes or
advisers in the background. (120) It is not uncommon for one or both parties to
have a solicitor or a page "922" local representative at the witness location of
the video conference. The provision of an attendee of both parties avoids any
suspicion of the witness being assisted or prompted out of sight of the video
cameras. (121)

It is certainly the case that many witnesses in complex matters are not really
needed in person. One of the difficulties in organising this in advance is that this
is often not known until one sees the degree to which the opposing side seeks to
cross-examine the witness. In an appropriate case, an arbitrator might invite
each party to indicate which witnesses it wishes to have present for significant
cross-examination and then use cost orders if that right has been abused
without just cause.

12.8.14. Challenges to Questions Put to Witnesses

A tribunal will need to determine whether it will allow questions subject to

objections from the other side or will itself seek to disallow questions it thinks
are inappropriate. Tribunals should give consideration to this issue before a
hearing, particularly when there may be differences in view amongst a multi-
person tribunal and/or counsel who come from differing legal traditions. It is
particularly problematic to try and make on the run evidentiary rulings on
questions by counsel in such circumstances. (122) It is particularly important to
understand the problems with questions and memory on a general level. It is
important to understand that the way we question, impacts upon an answer. For
example, some people seek to oblige when answering questions. It is also
important to understand that memory has three stages, perception, storage and
retrieval, which can significantly impact upon testimony.

12.8.15. Style and Content of Questions

In domestic litigation, counsel can often object to the way questions are put to
witnesses. In common law systems, it is not permitted to lead one's own
witness. A leading question is one which hints at the answer that is being
sought. In extreme circumstances, a party's own witness can be declared to be
hostile and be treated in a more interventionist manner. It is also not usual to
permit counsel to follow up questions as answered by one's own witnesses to
force an incompetent witness to complete the intended testimony. There are no
similar proscriptions on cross-examining counsel as the latter will typically lead
the witness to a proposition that they are disposed to disagree with in any
event. Without being able to lead and follow up, cross-examining counsel would
be unable to hone in on and probe the key parts of the testimony that need
exploration.

page "923"

It is necessary to consider what principles an international arbitrator should

apply in controlling questions to witnesses. There is no lex mercatoria of
international arbitration evidence in that regard. Nor do the IBA Rules of
Evidence 2010 say much on this issue. Article 8.2 of the Rules states:

The Arbitral Tribunal may limit or exclude any question to, answer by or
appearance of a witness, if it considers such question, answer or appearance to
be irrelevant, immaterial, burdensome, duplicative or otherwise covered by a
reason for objection set forth in Article 9.2.

A witness should thus be entitled to refuse to answer questions in situations

where Article 9.2 of the IBA Rules of Evidence or similar principles apply.
Questions can also be challenged on the basis of lack of relevance. Cross-
examination can be challenged on the basis of being unduly rigorous. Challenges
can also be made as to the admissibility of the evidence provided. Bühler and
Webster, (123) suggest that it is generally not possible to object against leading
questions. This view is rightly criticised by Born, (124) and is contradicted by the
IBA Rules of Evidence 2010 Article 8.2, although the Rules do not set up a
blanket proscription. Instead they state that ‘questions to a witness during
direct and re-direct testimony may not be unreasonably leading’. Tribunals will
generally bar such questions or at least warn counsel that they may reduce the
probative value of the answers.

12.8.16. Questions Eliciting Material beyond the Witness Statements

The idea behind requiring written statements in advance is to allow the other
side to appropriately prepare. Only then can the party be afforded its full due
process rights to adequately present its case. This is the notion of an adversarial
hearing and the right of contradiction as utilised in civilian systems. The
corollary is that if counsel believes that new matters need to be addressed after
reviewing opposing witness statements, leave should be sought from the
tribunal to submit supplementary witness statements prior to the hearing. To
fail to do so and try and introduce new evidence at the hearing could violate due
process norms. A proper test of whether the material is problematic is whether
the opposing party might legitimately have wished to make investigations,
conduct research and/or bring contrary evidence. In addition, there will be
problems if opposing counsel would legitimately require extra time to prepare
for cross-examination, (125) although that should ordinarily not require too much
extra time.

page "924"

One practical question where challenges arise is whether it is truly new

material. Too often, witness statements make broad allegations and leave it to
the hearing for elaboration of particulars. That might even be done for
undesirable tactical reasons, to minimise the effectiveness of cross-examination
or even prevent it entirely if the material is reintroduced on re-examination. (126)
Hwang and Chin suggest that a witness should only be allowed to add to the
matters contained in the witness statement in the following four situations:

(a) where they wish to correct an error or ambiguity;

(b) where they wish to elaborate on a relatively small detail;
(c) where the witness wishes to respond to matters raised in opposing party's
witness statements not seen before preparing their own statement;
(d) where the witness wishes to give evidence about facts which have occurred
since the date of the statement. (127)

One problem with any limitations is that they might lead to an inequality
between what the witness can present as evidence in chief and what a cross-
examiner might deal with or what a tribunal might legitimately question. An
example referred to above is a claim based on a misrepresentation inducing a
contract where the witness statement alleges this but does not particularise the
person or conversation involved. How is the tribunal to treat this? Will it
naturally allow the material to be expanded upon in chief? Will it ask for this to
occur and ask its own questions? What if the cross-examiner does not ask any
questions, either for tactical reasons or inadvertence? How will the tribunal
apply burden and standard of proof in such circumstances? These questions
suggest that it will be difficult to establish hard and fast rules that could apply
optimally in all factual permutations.

Lévy argues that it is permissible to cross-examine beyond the witness

statement at least so as to invalidate the statement. The distinction is justifiable.
Invalidating a statement made is in that sense, still dealing with the statement.
Supplanting it with positive evidence is varying its essential nature. (128) It was
also noted above that the proscription against dealing with matters outside the
statement becomes more problematic with witness conferencing where a
relevant dialogue moves back and forth and does not limit itself to individual
statements, although the tribunal can manage the scope of the conference
session.

page "925"

12.8.17. Questions by the Tribunal

The UNCITRAL Notes invite the tribunal to indicate how witnesses will be heard
and in what order questions will be posed. One approach is for the tribunal to
first question the witness, after which the parties themselves can pose
questions. (129) A second alternative is for the parties to engage in questioning,
with the tribunal interrupting where it thinks appropriate. A third alternative is
for the tribunal to wait until the parties have completed questions, including
cross-examination, and then ask any remaining questions of the witnesses.
Some continental European arbitrators may still wish to have questions directed
through the tribunal, although this would be rare. The tribunal might also
consider posing questions to witnesses to be answered in written form where
that would be appropriate. As noted, in ICC Case No. 9333, (130) a witness became
ill and unavailable for oral examination. The Tribunal collected questions from
the other party and drafted its own and sent a questionnaire to the witness and
allowed a second witness statement.
While party priority for oral questions is the most common, some arbitrators
will tend to allow cross-examination to be completed before they ask their own
questions. This is not a preference for common law over civilian systems but
instead a concern not to be seen to unduly interfere with the party's own
strategic choices as to how to best present its case. That is particularly so where
the tribunal has imposed strict time limits for the hearing with an invitation for
the parties to present the material as they see fit. If a tribunal was too invasive
under such a procedure, a party might legitimately be able to argue that its
rights were unduly interfered with. Even here it is important to consider other
factors and not be overly concerned with form over substance. A failure to draw
a witness' attention to the key aspects as soon as they are of concern to an
arbitrator may in reality be a more meaningful interference with their ability to
win the case.

The tribunal's right to question the witness would of course encompass the right
to deal with matters contained in the written submissions. While tribunals will
typically seek clarification of information provided, not all arbitrators would use
questioning to test the credibility of a witness in the way that cross-examining
counsel might employ. Nevertheless, in some cases this may be appropriate. (131)
A tribunal might wish to do so when the respondent is not represented, on the
basis that only the tribunal can test the quality of the claimant's evidence. In an
appropriate case a tribunal might also provide written questions to the parties
and their witnesses for formal written response. This could be particularly
useful in a page "926" documents-only arbitration or where deliberations show
some key gap in the testimony.

12.8.18. Witnesses Present during Other Testimony

A tribunal will need to determine whether one witness is entitled to be present

during the testimony of another. This is otherwise described as ‘sequestration’.
(132) This usually only applies to witnesses of fact, although where experts give

both factual and opinion evidence, similar principles might apply. The parties
could agree to allow or bar access of witnesses to the hearing. In other cases the
tribunal will have to make a determination. As always there are arguments for
and against. The benefit of having other witnesses present is that they hear the
conflicting testimony, they better understand the issues in contention and can
more directly address the matters of most significance to the tribunal. A person
can indicate exactly what they agree or disagree with rather than having this
filtered by lawyers. Hearing the witness might also jog the memory to allow
more accurate and pertinent evidence.

The detriment is that witnesses who are unwilling to be fully honest or at least
who wish to be strategic, are given too much advance warning of the challenges
facing their testimony and can be better able to inappropriately tailor their
comments accordingly. There is a particular disadvantage to the witness going
first as the other witness can be too rehearsed and prepared. Furthermore,
psychological studies show that if people express an opinion openly, a person
with a dissenting view hearing a range of views to the contrary is more reluctant
to present their true thoughts. Where it is the same side's witness, a
disadvantage would be the ability to tailor testimony to maximise consistency
where that would otherwise not have occurred. If one person is in a position of
influence over the other, the one in the weaker position may feel intimidated
and pressured to concur with the more senior person. Conversely, the presence
of such a senior person might induce someone to be more honest when that
otherwise would not be the case. (133) The conflicting hypotheses are also not
mutually exclusive. Hence, every legal system inevitably undertakes a trade-off,
although unfortunately most tend to come to a particular view from a single
issue concern.

Historically, the norm in arbitration was to exclude witnesses until the time that
they gave their own evidence. (134) The IBA Rules of Evidence 2010 do not take a
position either way on this issue. Tribunals will often allow witnesses who have
testified to be present during subsequent parts of the hearing. Some take a
contrary view, arguing that witnesses should have open access at all times. The
argument in page "927" part is that counsel can present prospective witnesses
with transcripts in any event, hence the benefits of exclusion are outweighed by
the detriments.

A special situation is where the witness is a party in person or is a key corporate

officer of a corporate party. As a general rule, it is felt that parties must be given
an opportunity to be present throughout the proceedings, otherwise they might
legitimately argue that they have not had a full opportunity to continually
monitor, and hence present their case. It is only when they can consider all
aspects of the proceedings that they know how best to address issues of concern
to the tribunal and issues raised by opposing witnesses and direct counsel as to
case strategy modifications. Fouchard, Gaillard and Goldman present a contrary
view as to the entitlement of such witnesses to attend hearings. (135) The authors
raise a concern that other witnesses might need to be treated similarly. In some
cases, even though a witness is technically a corporate officer of a party, the
person need not be present for the above-mentioned strategic reasons as they
are not the person with such authority. If tribunals accept the argument that
only party witnesses are entitled to be present throughout, they might at least
schedule them to testify first to minimise their tactical advantage. (136)

12.8.19. Contact with Witnesses during Hearings

It is generally accepted that once a witness has begun his or her testimony, they
should not be approached by counsel, parties or other witnesses during any
breaks. They have already been properly prepared within the boundaries of
what is acceptable and any discussion of the case during their testimony would
be more naturally presumed to be improper coaching. The caution is more
pertinent where cross-examination is concerned. A tribunal might allow social
or work related contact on the express understanding that the case is not to be
discussed.

12.9. Subpoenas, Compulsion and Dealing with Difficult Witnesses or Those

Who Refuse to Testify on Matters
12.9.1. Compulsion

An arbitral tribunal has no direct power over third parties. Arbitral jurisdiction
and power emanates from consent as supported by arbitral statutes. Consent of
the parties to arbitration can never empower a tribunal to exercise coercive
powers over third parties who have not consented.

page "928"

The national legislature that promulgates an arbitral statute may seek to grant
such coercive powers, either directly to the tribunal or to a supervisory court
within the jurisdiction of the seat. Even if a legislature purports to grant such a
power, this does not necessarily mean that it will be effective, particularly in
relation to foreign persons, which will naturally be the case with a neutral Seat.
Some lex arbitri indicate that local courts can be approached to assist in
obtaining the appearance of witnesses who are not willing to come voluntarily.
(137) In some cases the tribunal might choose to hold hearings where it or a party

might apply to a court for a subpoena over a witness. (138) The IBA Rules of
Evidence 2010 allow a party to ask the tribunal to take whatever steps are
legally available to ensure the testimony of the particular person. (139) A party
might also look to a national court that may have jurisdiction.

It would be wrong to presume that compulsion is always concerned with

reluctant witnesses. In some circumstances a witness may be personally willing
to appear but may be concerned to do so lest an employer or other related party
takes an adverse view of it doing so. For such prospective witnesses, a valid
albeit unenforceable compulsion order by a tribunal or supervisory court can
aid them in doing what they would wish to do in any event. (140) In some cases a
tribunal may assist witnesses by merely issuing a letter inviting attendance. This
might also help in getting permission from an employer or in getting a requisite
visa.

12.9.2. Refusals to Answer and Evasive Witnesses

Tribunals will sometimes be faced with witnesses who are uncooperative or

evasive when giving evidence. The proper approach to take in response depends
in large part on the duties of witnesses and powers of compulsion of arbitrators.
Because arbitration is consent based, parties who are witnesses have at least
impliedly consented to arbitral power and discretion, absent evidence to the
contrary via party agreement or exclusionary rules selected. The corollary of the
consent base is that third-party witnesses are generally not obliged to give
evidence, answer questions or indeed continue in attendance at the tribunal's
pleasure. It also means that a bilingual witness cannot be compelled to give
testimony in their non-preferred language. Any powers to the contrary in lex
arbitri would be subject to territorial limitations in an enforcement sense at
least. The key response to evasive witnesses is in relation to the potential to
draw adverse inferences discussed in the following section.

page "929"
A witness should normally be allowed to refuse to answer a question for the
reasons articulated in Article 9.2 of the IBA Rules of Evidence, which were
discussed in section 11.7 above. It may be difficult for a tribunal to decide
whether the objection is validly taken without knowing the nature of the
withheld testimony although in most cases a general discretion should allow a
tribunal to make an informed ruling.

12.9.3. Adverse Inferences

In all cases a tribunal will be able to consider the impact of a witness' failure to
attend or answer, in terms of an analysis of the preponderance of evidence. For
a party with the burden of proof, a failure to respond by a key witness will mean
they might fail for that reason alone. Where the situation is not so clear cut, the
next question is as to adverse inferences against either party and when these
may legitimately be drawn. This was discussed more generally in section
10.4.8.2. Where witnesses are concerned, the issue is in part circular. An adverse
inference is generally to the effect that the evidence would be against interest,
but it needs to be based on a logical presumption to be valid. If the refusal is
because of a legitimate right not to attend or answer, the adverse inference is
not appropriate.

If subpoenas have been utilised and a witness does not attend, a tribunal may
need to consider the validity of the subpoena and whether service has been
appropriate before considering whether adverse inferences may be drawn.

12.9.4. Perjury

Another question is whether statutes which provide criminal sanctions for

perjury apply before arbitral tribunals. That would depend upon the language
and intent of the relevant statute; whether they intend to apply to arbitration
and also to foreign persons or to nationals giving evidence in foreign places.
Even then, there is a question as to which countries' perjury statutes are
intended to apply. A tribunal may also need to consider whether it is entitled to
notify the relevant authorities or whether confidentiality norms preclude this.
Jurisdictions vary as to whether false testimony will lead to criminal sanctions,
which may depend on whether it is sworn or unsworn testimony.

It is normally for the tribunal to determine whether testimony is truthful or not.

A complex situation arises where testimony relied upon is subsequently found
to be false which is then a basis for attacking the award. While a challenge is
conceivable where it can be shown that the false testimony was induced by
some procedural error of the tribunal, (141) this is highly unlikely to be the case.
Deliberate false page "930" testimony by a party or at a party's behest might
form grounds for overturning an award on public policy grounds.

12.10. Expert Witnesses and Expert Assistance

12.10.1. Introduction
A tribunal will often need the input of experts in order to resolve complex
factual disputes. For example, some expertise will typically be required if the
issue is whether a construction had a faulty design, an engineering calculation
was inaccurate or whether the construction itself was negligent. Experts are also
sometimes used for mathematical calculations, such as claims for measured
work in construction disputes, company valuations, present value or projected
profit. Lawyers may give expert evidence as to the content of national systems of
law that are applicable. Forensic scientists may give evidence as to authenticity
of documents where this is challenged. An expert may also assist the tribunal in
considering document production requests where the tribunal does not wish to
be the one assessing grounds for refusal that might compromise it, such as
confidentiality claims, although this is a unique category as there is a debate as
to the status of such persons. (142) In some cases an expert witness or assistant
may be unnecessary as the arbitration clause will lead to an appropriate expert
as arbitrator. An example would be an appointment by the president of an
engineering association in a construction dispute. The very use of expert
witnesses can be questioned in international arbitration, whose flexibility
allows the parties to select experts as arbitrators. (143) Even then the parties may
wish to present conflicting views.

While non-expert witnesses are only heard on issues of fact, and not in order to
express opinions, the same is not true with expert witnesses. Experts often do
not simply provide evidence in the strict sense but instead provide opinions and
reasoning underlying the opinions in relation to evidence otherwise before the
tribunal. (144) In other cases their testimony is as to the status of specialist facts
and theories.

page "931"

In each case there would be questions as to who is an expert; methods of

appointment; suitable qualifications; who can challenge any selection, the
ethical responsibilities of experts including whether and how they can be paid;
the form of presentation of expert witness testimony, including whether they
are entitled to listen to each other's testimony; and the means by which
tribunals can reconcile conflicting expert opinions, including whether they give
evidence individually or whether they are involved in a collective exercise such
as witness conferencing.

12.11. Advantages and Disadvantages of Tribunal Versus Party Experts

In civilian State courts, experts are typically appointed by the adjudicator and
work for that person's benefit. In common law jurisdictions, they are typically
selected by the parties, appear as formal witnesses and their evidence is subject
to cross-examination. (145) Some modern common law rules seek a hybrid
position and also make it clear that an expert is there to help the court and not
to be an advocate for a party's strategic position. (146) Historical differences in
view about party versus tribunal-appointed experts stem in part at least from
cultural differences about adversarial and inquisitorial processes. From an
adversarial perspective, if counsel's job is to present the best possible argument
including the best possible evidence, control over one's experts is important.
Conversely, from an inquisitorial perspective, if expert assistance is only there
to help the adjudicator, then avoiding a battle between party-appointed experts
is the logical corollary. Because there are advantages and disadvantages of each
model, not only are there debates as to which should be preferred but there are
also design issues in optimising the use of whichever is to be utilised. Where
party-appointed experts are concerned, domestic models in common law
countries vary from the English tradition which has sought to articulate
obligations to primarily assist the court, to the US approach which relies on a
preliminary determination as to the standards of proposed scientific expert
testimony. (147) The US system can be argued to be complex and expensive, while
the English system is seen by some as having a misguided belief in the ability to
constrain ‘hired guns’. (148)

The trend in international arbitration is to rely on party-appointed experts

rather than tribunal ones, (149) although there have been important suggestions
and page "932" possibly more recent trends to the contrary. The uncertainty is
because of significant practical problems either way, in addition to differing
historical views between legal families. It is also rarely clear what the marginal
cost implications are of appointing experts as this will depend on the assistance
they give to the tribunal, any reduction in tribunal analysis time where tribunals
are remunerated on an hourly or daily basis and any promotion of efficiency at
the fact finding stage. As always, there is a need to consider any option in
contradistinction to alternatives. For example, Mark Kantor aptly observes that
if parties are denied their own appointed experts, sophisticated counsel can find
witnesses of fact with sufficient expertise to testify on the key issues. (150) It is
also the case that where legal experts are concerned, the relevant person could
simply be used as a co-counsel as opposed to an expert witness. (151)

The key problem with party-appointed experts is thought to be lack of

impartiality. It has been observed that expert testimony can be abused when it
becomes ‘no more than paid advocacy of a party's cause’ . (152) They may
typically be too long and complex, (153) although that may depend on instructions
given by the tribunal. Use of party-appointed experts may also produce obtuse
reports that contain extreme and divergent conclusions and which fail to
prioritise matters of most concern to the tribunal. In addition, by following
independent methodologies and modes of expression, they often make it
difficult to be read alongside other written reports of opposing experts. (154)
There is also uncertainty as to whether party-appointed experts may withhold
adverse information or may refrain from identifying problems with any
assumptions that their brief was required to utilise, although this should not be
the preferred view. There is also the added cost of a larger number of experts.
Even if the parties were able to agree on a single expert, this may not necessarily
reduce costs if they utilise ‘shadow experts’ to help in submissions. (155) A
particular difficulty with party-appointed experts is that if they are each truly
expert and honest but conflict in their views, how is a non-expert adjudicator to
make a choice? The problems are added to when experts cannot provide
compelling proof of their conclusions, but instead, merely state page "933"
conclusions based on their professional experiences. Examples could include the
belief that certain causes were most likely for any damage in a construction
dispute or that certain values are most appropriate in an expropriation claim. A
further problem with conflicting party-appointed experts is where they are
likely to divide on theoretical lines. For example, accounting experts may have
fundamentally different views as to the best way to value a company or an
ongoing business. A tribunal has a judgment problem in deciding which
theoretical view to accept if both theories are well respected in the relevant
professional circles. The challenge will not be diminished simply because one
member of the tribunal is from that expert field, as that person, like other
experts, might well have a preordained view of the theoretical issue.
Furthermore, in a multi-person tribunal, the other members must each form a
view and cannot too readily defer to their expert colleague. In some
circumstances a tribunal may resolve the matter by determining whether the
party with the burden of proof has met the required standard.

There is even a law and economics argument against the use of party-appointed
experts. This is on the basis that the financial incentives to favour the appointing
party, and the difficulty an adjudicator has in assessing where this occurs, could
lead to market failure in the supply of experts by providing greater incentives
for partiality. (156) It has also been observed that the skill-set needed to best
educate a tribunal about matters beyond their expertise is also a skill-set that
would allow partisan experts to engage in advocacy in the guise of objectivity.
(157) However, there are a number of techniques, including proper briefing,

express ethical standards and witness conferencing, which ought to help

minimise any such tendencies.

Tribunal-appointed experts are presumed to be more independent and will be

subject to cost control by the tribunal or by an institution where involved. As
noted, single experts might still add to costs if parties appoint shadow experts.
(158) Concerns with tribunal-appointed experts include lack of control by the

parties, problems with the flow of information from the parties to the tribunal-
appointed expert, difficulties in identifying the key assumptions upon which
certain opinions must rely and a concern that too much of the effective decision-
making is in fact undertaken by the expert. (159) The latter would occur where the
expert is invited to analyse evidence and express an opinion as to the probative
value of conflicting documents that will not be carefully analysed by the
tribunal. Problems again arise with tribunal-appointed experts where experts
are likely to divide on theoretical lines. Here a tribunal has a prejudgment
problem in deciding which theoretical page "934" camp to select an expert from.
Where the key issue is which theoretical camp prevails, selection of the tribunal-
appointed expert will go a long way to deciding the final dispute. Where access
to information is concerned, party-appointed experts will obviously be granted
access to the information they request otherwise they will simply refuse to
present an opinion. Where tribunal-appointed experts are concerned, there may
be problems of selectivity and an inability to ensure compliance without the
assistance of the tribunal. Issues of length and complexity can similarly be a
problem with the reports of tribunal-appointed experts, which can again be
impacted upon by useful directions by the tribunal. This is discussed further in
section 12.14.3 below.
While there is obviously a distinction between tribunal-appointed and party-
appointed experts, where the latter are concerned, if the tribunal exercises
sufficient control and guidance, this can be made to work harmoniously for the
clear benefit of the tribunal. It may even be that the parties could be asked to
agree to changing their status to tribunal-appointed to overcome an impasse.
(160)

Because there are no easy solutions to these problems, leading arbitrators have
experimented with hybrid solutions or other processes to maximise the fairness
and utility of expert input. In some cases, this also mirrors domestic
developments. Because it is relatively easy to find experts that might disagree
on contentious matters, modern case management in common law systems has
led to some encouragement of court-appointed or jointly appointed neutral
experts as has been the tradition in civilian jurisdictions. Even where expert
testimony is still subject to party selection, as noted above, ethical codes
sometimes indicate that an expert must nevertheless see their role as being to
assist the adjudicator. (161) These issues and options are addressed separately in
the sections that follow. It is also important to understand that the two forms of
experts are not mutually exclusive in any particular arbitration. A particular
case might use a tribunal-appointed expert to help the tribunal deal with the
conflicting opinions of party-appointed experts. Here there are again
organisational questions and challenges which flow.

12.11.1. Power of Parties to Appoint

Some arbitral statutes expressly allow parties to appoint their own experts. (162)
Nevertheless, even where the laws only expressly deal with tribunal experts, as
page "935" is the case with Article 26 of the UNCITRAL Model Law, the better
view is that the right to submit evidence, for example pursuant to s 23(1) of the
Model Law allows for evidence by means of party-appointed experts. (163) Some
rules expressly allow for party-appointed experts without tribunal permission.
Article 25(3) of the ICC Rules 2012 requires tribunal permission although this
would invariably be granted. It would be difficult for an arbitrator to deny such
a right given due process concerns. The CIArb Protocol Article 3 also indicates
that permission must be sought before adducing expert evidence. Article 5.1 of
the IBA Rules of Evidence 2010 indicates that a party may rely on a party-
appointed expert. Notification shall be within the time ordered by the tribunal.
Many other rules are silent but the norm is to allow party appointment as of
right. Even without express reference, parties should be able to designate their
witnesses, including experts. Each party's right to appoint their own expert
could be considered a fundamental right in the context of being heard. Given
that a right to representation is sacrosanct, there is nothing in theory to
distinguish an expert presenting an opinion by way of submissions and a lawyer
doing so instead.

While the tribunal has overall control over the number of witnesses and
duration of evidentiary hearings, the mandatory right to present one's case
would mean that restrictions on experts must only be imposed where this right
is not affected. Poudret and Besson suggest a limitation by proposing that there
is no right to an expert opinion flowing from the right to be heard unless it ‘is
necessary and capable of establishing facts which are relevant to the outcome of
the dispute’. The qualification seems broad enough to apply in virtually all
circumstances. (164)

A problem may still arise if there is a difference in view between the parties as
to the value of expert evidence for dealing with a key issue where there is no
express right to appoint. In such cases the tribunal might require the parties to
make submissions in that regard so that appropriate procedural orders could be
made, (165) but the overriding discretion should still remain.

12.11.2. Joint Appointment and Instructions

One possible solution to many of the problems of party-appointed experts is to

jointly agree on a single expert. There are a number of procedural hurdles that
would need to be carefully considered with such a proposal. There may well be
resistance by a claimant who has had to engage an expert to formulate and
crystallise its claim. Not only would there be a need for a clearly expressed joint
appointment but also a joint brief to the expert. The tribunal would inevitably
page "936" control that document, seeking agreement of the parties. Problems
would arise if a joint expert was first appointed but the parties could not then
agree on the instructions. There would also need to be provisions covering
situations where only one party was forthcoming with instructions or materials,
late submission of information by one of the parties, methodology of instructing
and providing access to materials, and conferring and payment of fees. If a single
expert was appointed, it would make sense to allow for cross-examination by
both parties. (166)

12.11.3. Power of Tribunal to Appoint

Arbitral statutes typically make greater reference to tribunal-appointed experts

than party ones, (167) although some key laws are silent on the issue.
Nevertheless, general procedural powers are seen as being broad enough to
cover the entitlement to tribunal appointment. (168) Arbitral rules will also tend
to expressly allow for tribunal appointment of experts. (169) Subject to any
express duty to call for an expert, if a tribunal thinks it has sufficient information
and refuses to order an appointment, this does not violate due process rights.
(170)

12.11.4. The Right to Appoint

Redfern and Hunter suggest that tribunals will rarely appoint experts unless
expressly empowered to do so, (171) although Born suggests that such a request is
not a prerequisite to appointment. (172) The IBA Rules of Evidence require a
tribunal to consult the parties before appointing an expert. (173) Born also
suggests that a page "937" tribunal is not required to appoint an expert where
the parties so request. (174) This will depend on the particular jurisdiction. The
Swiss Federal Supreme Court has considered that in some cases, it is possible
for an individual party to call for a tribunal-appointed expert under mandatory
due process norms. This would be so where there is a specific request in a
proper and timely manner, where the party is prepared to advance the costs if
required by the tribunal and where the expert evidence would be required to
adequately resolve the issue. (175) Redfern and Hunter pose the question as to
whether a power to appoint can be implied under general principles of the lex
arbitri and argue strongly that an appointment power can be implied. (176) This
should be the preferred position under general procedural powers in cases
where the parties express no preference. Nevertheless, there are good practical
reasons to seek party approval wherever possible. Appointment by a tribunal
raises an important policy question as to the proactive nature of tribunal
behaviour, given that absent such an appointment a determination will be made
on the basis of the evidence presented by each of the parties, including their
own appointed experts. A related observation is that wherever a tribunal
appoints its own experts as well as allowing party-appointed experts, costs may
increase significantly. This might only be appropriate in matters of sufficient
significance. (177) Because of the cost involved both directly and indirectly
through the impact upon party witnesses, a tribunal should seek the parties'
views when it is considering appointing an expert, provide reasons and give fair
and reasonable consideration to the parties' responses. (178)

The situation will be more complex where the parties express a common
contrary view or disagree strongly as will often occur. It would seem natural
that tribunals should not appoint experts if the parties do not wish this to occur.
Derains and Schwarz suggest that the general powers in the ICC Rules should
not be interpreted to give a tribunal a right to appoint an expert contrary to the
wishes of the parties as a tribunal ought not to be able to impose such an
expense without their consent. (179) The reasoning may be problematic at least in
the sense that it could be used against many discretionary determinations of
tribunals on a range of issues. This issue may also be affected by questions of
timing. If the tribunal was already page "938" expressly empowered to do so but
the parties subsequently agreed to the contrary, in extreme cases the tribunal
might consider that its entitlement to do justice is being unfairly reduced by the
parties. For example, it has been suggested that international judges in inter-
State disputes have an inherent right to call for experts because of a duty to
resolve the dispute on all relevant data and be in possession of and understand
all relevant evidence. (180) Because commercial arbitration is private and
consent-based and is not seeking to generally establish principles of
international law with broader implications, that logic is not readily
transferable, particularly as it traverses the debate about an arbitrator's duty to
be proactive or whether the arbitrator should merely reconcile the material as
presented by the parties.

12.11.5. The Process of Appointment of a Tribunal Expert

Because differing experts can vary in their views, a tribunal concerned to allow
all parties a full opportunity to present their case has to be very careful on how
it goes about making an appointment of a tribunal expert. The parties might be
invited to comment on selection and certainly must be given an opportunity to
comment on any report. Poudret and Besson suggest that the right to be heard
would imply entitlement to be consulted as to the choice of expert and the terms
of reference. (181) If there are residual concerns, a tribunal might need to allow
the parties to appoint their own expert and then appoint a third to assist the
tribunal, although the added expense might not be appropriate in all cases. An
example might be an accounting expert asked to synthesise the costings in a
construction dispute. There might be little need to allow each party to present
their own calculations as opposed to allowing them to merely cross examine or
make submissions on the report of the tribunal-appointed accountant. Experts
as to foreign law may not be necessary given that the parties are likely to have at
least one counsel expert in the applicable law.

Where a tribunal wishes to appoint experts, it may seek to appoint its own
experts from its own knowledge, or conversely, ask the parties to submit lists of
appropriate names in order to identify a commonly agreed person or persons. If
that approach does not display common names, the lists can be exchanged,
parties given an opportunity to object to a defined number or at least provide
comments, with the tribunal then making a selection from the remainder.
Alternatively, the parties might individually be asked to rank from a list
prepared by the arbitrator. Another possibility is to allow a certain number to
be vetoed as of right, with others able to be challenged for just cause. Another
possibility is to ask the parties to submit their own ordered lists. (182)

page "939"

A difficulty with any list is that the longer the list, the harder it is to ensure that
all candidates are fully suitable and available and the more time consuming it is
to identify this information. In some cases the tribunal might seek
recommendations from a particular professional organisation or utilise lists of
experts such as that promulgated by the International Centre for Technical
Expertise of the International Chamber of Commerce. A number of other
organisations include the Academy of Experts, EuroExpert, Expert Witness
Institute and the Society for Expert Witnesses. The hope is that an institutional
list will provide some quality control, both as to initial expertise and feedback
from performances in other cases. There may be problems in that regard with
issues of confidentiality, defamation and a conservative approach once a person
is already on the list. If the tribunal is making an independent appointment
without utilising recommendations of the parties, it might also wish to provide
the parties with curriculum vitae and afford them an opportunity to make
comments or state any objections.

12.11.6. Criteria for Appointment

In selecting an expert, the tribunal, in consultation with the parties should,

consider independence; expertise; communication and language skills; ability to
undertake any necessary testing and investigation in a timely and efficient
manner; availability for the hearing; fees and expenses. (183) Some take the view
that, wherever possible, a tribunal-appointed expert should not be a citizen of
the country of either party unless that flows as a matter of course from the area
of expertise needed. Ideally expert witnesses will be fluent in the language of the
arbitration as they are attempting to communicate what will often be difficult
concepts to laypersons. This can be particularly problematic where professional
jargon is involved as would be the case with legal experts. (184) While the expert
will generally be an individual, this may not always be so. CIETAC Rules 2012
Article 42.1 indicates that an expert or appraiser may be either an organisation
or citizen.

The tribunal should consider what would be an appropriate number of experts

to appoint. This would depend on the range of matters on which expert opinion
is desirable, and the breadth of expertise of the potential candidates for
appointment.

Where the tribunal appoints an expert, one question is whether a challenge

could be made to the tribunal itself based on bias. In theory at least, if the
evidence of bias ought to have been apparent to the tribunal at the time of the
appointment, such behaviour would appear to offend against the right to equal
treatment. (185)

page "940"

12.11.7. Timing of Appointment

Timing of appointment of a tribunal expert may differ from that of the parties.
The parties independently determine their case strategies, determine the
witnesses that are desirable and will comply with general duties of disclosure in
that regard. An important aspect is to give the other party appropriate warning
of the case that will be put. The tribunal appoints an expert for a different
reason, namely to assist it in coming to a conclusion. A tribunal might find that
the desirability of such assistance arises at different times and for differing
reasons. In some cases a tribunal might know at an early stage that a matter will
be sufficiently complex that it would wish to have its own expert assistance
throughout. In other cases it may only be after the tribunal views the party-
appointed expert reports that it feels the need for help in reconciling and
understanding that material.

A tribunal ought to ensure that it is sufficiently familiar with the material

presented before determining that its own expert would be beneficial. Once that
view is taken, the parties should be informed at the earliest opportunity as this
may impact upon their ultimate choice of witnesses to present and the manner
in which reports will be written. For example, a party-appointed expert might
include more complex material if he or she knows that a tribunal expert will be
involved. The complex material might be the best way to convince the tribunal
expert, but could confuse a non-expert tribunal that does not have such
assistance.

12.11.8. Independence and Challenges to Experts

A tribunal-appointed expert should be independent and impartial and should

have no conflict of interest with any party to the proceedings. (186) While this is
an important general principle that should apply as a matter of course, there is
no uniform express basis for seeking to exclude an expert for lack of impartiality
as is the case with arbitrators, although this is likely to be the emergent norm. If
there is no entitlement to challenge as to a concern about the impartiality of an
expert, it could simply be left for the advocate to challenge the validity of the
expert's evidence on that basis. The IBA Rules of Evidence 2010 allow for a
challenge as to the independence of the tribunal-appointed expert. (187) Even
absent express provisions, such challenges ought to be appropriate (188) and can
be dealt with under broad procedural page "941" powers. (189) The tribunal's
general power of appointment should be seen as having an inherent duty to
consider challenges to the proposed exercise of that power. The expert should
also consider matters needing disclosure and form a view as to independence.

Any challenge to a tribunal expert witness should occur as soon as possible in

order to avoid wasting time and money. More complex is the question of
whether an untimely challenge could constitute a waiver of the challenge right.
That would depend on which principles of waiver would be applicable. A
conclusion as to waiver would be unlikely, given that a compelling argument
that a particular expert does not in fact have the relevant expertise, or has a
fundamental conflict undermining veracity, is hardly likely to be ignored by a
tribunal, at least on questions of weight. That said, Article 6.2 of the IBA Rules of
Evidence 2010 states with respect to tribunal-appointed experts that late
objections may only be made if they relate to reasons of which the party only
became aware after an appointment was made. But even if the parties are
expressly bound by the IBA Rules, barring of a late formal objection cannot be a
bar to submissions as to the lack of weight to be given to the particular expert's
evidence.

It is less clear whether party-appointed experts need to meet the same

standards of independence as tribunal-appointed experts. Some civilian experts
view the whole notion of independence of party-appointed experts as a fiction.
(190) Others argue that party-appointed experts should also be independent. (191)

This is supported by Article 5.2(c) of the IBA Rules of Evidence 2010, which is
significantly different to the 1999 Rules. It is easy to state an obligation of
independence for a party-appointed expert, but it is more difficult to identify
exactly what that entails given that this particular type of witness is expected to
take money from a party and consult with the party and counsel. There is a
difference between impartiality and objectivity, that should apply in any event,
and independence, which should relate to the limits of permissible directions or
suggestions from a party or counsel. All would agree that any opinion they
present should certainly be honest, objective and independent, even though the
relationship itself cannot be described as wholly independent. (192) The opinion
of an expert should not be distorted for the benefit of the party appointing. The
Chartered Institute of Arbitrators Protocol states that ‘(a)n expert's opinion
shall be impartial, objective, unbiased and uninfluenced by the pressures of the
dispute resolution process or by any party’. (193) Conversely, it has page "942"
been suggested that, given the reality of party-appointed legal experts, ‘the
required standard for neutrality of legal experts should be less stringent than
that of arbitrators’. (194) How a less stringent standard might be justifiable is not
easy to discern. The CIArb Protocol makes clear that receiving a fee does not in
and of itself impact upon independence. Nevertheless, most would see a
contingency fee based on success in the proceedings as being an unacceptable
interference with independence. Mark Kantor makes the valid observation that
the desire to please the party hiring the expert and the desire for ongoing
business already provides incentives of this nature that are not easily
distinguished from a contingent or success fee arrangement. Furthermore,
experts are rarely engaged before there has been questioning to determining
whether their opinion is likely to be advantageous. (195) On this view, a tribunal
might not be able to override the wishes of a party, particularly if contingency
fees are the norm for remuneration in particular professions.

If a challenge is made and rejected, the aggrieved party might say that reliance
on a partisan expert breaches due process and transnational public policy.
Conversely, improper exclusion of a party-appointed expert could be argued to
be an interference with its right to be heard and adequately present its case.
Such challenges are unlikely to succeed except on clear-cut factual situations.
Even less likely would be a successful challenge by a party after a tribunal
accepted the opposing party's complaint about lack of independence of the
tribunal-appointed expert. If the tribunal rejected that expert, it would no doubt
appoint another, so due process would hardly be interfered with.

The IBA Rules of Evidence 2010 require a statement of independence before

accepting an appointment for tribunal-appointed experts, (196) and a statement of
independence in the report of a party-appointed expert. (197) The 2010 IBA Rules
now call for disclosure of relationships with the tribunal and legal counsel as
well as in relation to the parties themselves. (198) In some cases an expert may
believe that they are independent on accepting appointment but then will see
potential conflicts arising after viewing the material. The obligation of
independence is an ongoing one. Expert groups may also impose particular
obligations. For example, the ICC International Centre for Expertise calls for a
statement of independence and disclosure of relevant factors. (199)

page "943"

As to the kinds of factors that may be relevant, an expert should not be a

competitor of either party, which in some cases may severely restrict the choice.
(200) Another potential problem is if a leading expert habitually appears in front

of a leading arbitrator or at times sits as a co-arbitrator. There may also be

problems where the expert is knowingly contending for different positions in
one arbitration as opposed to another, when acting for the same party. While a
tribunal might wish to probe such inconsistency, confidentiality duties in
relation to the previous arbitration might pose some challenges. (201)

The Commentary to the IBA Rules of Evidence 2010 raises the special situation
of an expert that may have been appointed by a European national court
immediately after an injury occurs, to determine either causes of damage,
possible remedies or to preserve evidence. This can occur long before an
arbitration commences. Such a person does not presumptively lack
independence but fairness to common law adversaries unfamiliar with this
process may need special sensitivity in such circumstances. (202)

12.12. Challenges as to Expertise

The tribunal should ensure that the hearing is not taken up with challenges to
the expertise of opposing witnesses. Hence, curriculum vitae should be
exchanged well beforehand alongside the expert reports or even earlier so that
if necessary, preliminary rulings can be made as to expertise. In most cases, the
tribunal will simply allow the parties to present the experts of their choice and
allow submissions as to relative expertise to go to questions of weight.

12.13. Powers, Rights and Duties Of Experts

12.13.1. The Differences between Party and Tribunal-Appointed Experts

As noted, there are two broad types of experts, the first being the experts
presented by the parties as witnesses and the second, any experts appointed by
the tribunal. They are discussed together in this section with similarities and
differences noted. There is greater uncertainty with the role of party-appointed
experts. While in each case an expert must present an honest and objective
view, a party-appointed expert is nonetheless part of a team whose objective is
to help the party page "944" win. As noted above, the party-appointed expert is
paid by the party and confers with the party and counsel in preparing a report.
Hence, it is more problematic to determine the nature and extent of any duty of
independence. The emerging albeit still contentious view is that it is still
important to establish that even a party-appointed expert's role is to assist the
adjudicator in a reasoned and independent manner and not advocate the
position of the appointing party. Nevertheless, little is said in arbitral statutes or
rules about any ethical duties of party-appointed arbitrators. Because of the
consensual basis of arbitration, any such ethical standards tend to be limited to
the parties and the tribunal. Even legal counsel are not generally subject to
express ethical norms pertaining to arbitration processes. This poses a
particularly challenging issue for party-appointed experts, particularly in view
of the differences from a comparative law perspective, where lawyers from the
common law tradition see such experts as a natural part of the adversarial
process while many civilians see ‘guns for hire’ with little probative value and
insufficient scope for adequate tribunal control. (203) Mark Kantor also makes the
observation that the ethical obligations on an expert will impact upon the ethical
behaviour of counsel. (204) A party-appointed expert may also need to consider
the contractual obligations with the appointing party. It may even be that there
are tortious obligations to the other party, particularly in relation to
misrepresentation. (205)

All would agree that tribunal-appointed experts are simply there to help the
tribunal come to the correct view, regardless of which side that favours. Redfern
and Hunter describe the role of experts being ‘to assist, educate, and advise the
arbitral tribunal, in a fair and impartial manner in specialist fields (e.g.,
technical, forensic accountancy, legal, etc.) in which the arbitrators (or some of
them) do not themselves have relevant expertise in specific issues in dispute
between the parties'. (206) The key is to maintain objectivity. The comment does
not distinguish between the two types.

The CIArb Protocol indicates that:

An expert's duty in giving evidence is to assist the arbitral tribunal to decide the
issues in respect of which expert evidence is adduced; and page "945" An expert
opinion shall be impartial, objective, unbiased and uninfluenced by the
pressures of the dispute resolution process or by any party. (207)

The IBA Rules of Evidence 2010 do not go as far as the CIArb Protocol in
articulating an express duty to assist the tribunal for party-appointed experts.
Instead, the IBA Rules tend to articulate the specific behaviour rather than the
overriding duty. For example, the Rules call for experts to express their ‘genuine
belief in their opinions as opposed to ‘the truth’. (208) A genuine belief must
articulate a view that reconciles conflicting perspectives. While limited to the
approach of courts, the Reporters for the Principles of Transnational Civil
Procedure involved in developing the ALI/UNIDROIT principles appended their
own proposed rules to those principles. Rule 26.1 requires court-appointed
experts to be ‘neutral’. Rule 26.3 indicates that party-appointed experts are
subject to the same standards for objectivity and neutrality as is the case with
court-appointed experts. (209) A party-appointed expert is also obliged to
perform the task in good faith and in accordance with the standards of the
expert's profession. Where party-appointed experts are concerned, another
issue is the extent to which legal counsel may be involved in helping shape their
reports and testimony. Issues discussed in section 12.5.2 and 12.6 above would
again be relevant. A practical issue is that questions of privilege and limited
document discovery will generally prevent forced disclosure of communications
between counsel and party-appointed experts, (210) which adds to the difficulty
in having any ethical standard impact significantly on the commercial incentives
for partiality and advocacy. Mark Kantor suggests three core duties to promote
the ethical responsibilities of party-appointed experts. He suggests:

(1) a duty of ‘disclosure’: to disclose material relationships with respect to the

parties, their affiliates, counsel or the dispute, including compensation
arrangements;
(2) a duty to provide ‘full information’ even if adverse: to include in any written
and oral evidence all material information, whether supportive or adverse
to the professional analyses and conclusions found in that expert's
evidence; and
(3) a duty to ‘assess reasonableness’: a duty to use diligence to assess, to the
extent the expert has the professional background to do so, the
reasonableness of assumptions provided by counsel or a party on which
that expert relies in the expert evidence. (211)
page "946"

There are practical and strategic considerations as well. Party-appointed

experts need to know from the outset that they are only useful to their
appointing party if they help a particular position be adopted by a tribunal and
that will result from their logic and qualifications. Otherwise it is easy to
wrongly see themselves as having some advocacy role. There may also be some
incentive to downgrade contrary indicators that they might in other
circumstances concede to have more relevance, but if they are caught out doing
so, it can undermine their credibility. It is also important to remember that the
first role a party-appointed expert may play is to give general advice to the party
about the strengths and weaknesses of its case and how the case might best be
presented. (212) That should still be provided in an impartial and objective
manner.

Counsel also need to understand that there is a world of difference between

advocates arguing contrary positions in different cases before the same
arbitrator and party-appointed experts acting in the same manner. An advocate
is simply concentrating on those arguments that best serve their client's
interests. If a leading arbitrator sees the same expert arguing opposing positions
in different cases, that person's veracity is naturally undermined.

12.13.2. The Tribunal's Duty Not to Delegate Decision-Making

It is also important to consider the powers of the expert in the context of the
rights and duties of the tribunal, particularly in the context of tribunal-
appointed experts. A tribunal-appointed expert does not have determinative
power over any factual matter. Experts provide reasoned opinions for the
benefit of the tribunal. A tribunal is not bound by the opinions of experts even if
appointed by the tribunal. An arbitrator's duty to complete the mandate
requires the arbitrator to make a determination even if a tribunal-appointed
expert is involved. At most, the express or implied power to appoint an expert
allows the tribunal to seek such assistance, but does not allow the tribunal to
delegate the adjudicatory function. More generally, legal systems support the
view that a delegated duty cannot generally be further delegated without
permission. (213) The obligation not to delegate decision-making should be
considered in a purposive rather than formalistic manner. For example, page
"947" if an expert is entitled to interview parties in relation to factual matters
for ultimate determination, it is important that the expert does not make
findings of fact in lieu of the tribunal.

The prohibition against delegation of decision-making power is more

problematic if an expert is used because of confidentiality claims so as to shield
material from the parties and the tribunal. For example, in Bechtel Inc, an Iran-
US Claims Tribunal directed that an independent accounting firm inspect non-
public stockholding records to ensure that sufficient stock was beneficially
owned by natural persons who were citizens of the US to ground jurisdiction.
(214)

The final decision on any and all issues must truly be that of the tribunal itself.
The tribunal must consider the evidence of the experts, determine its relevance
and weight and apply independent judgment in resolving conflicting testimony.
If a party-appointed expert does not provide adequate reasons behind an
opinion, a tribunal could be justified in rejecting the evidentiary value of their
report or testimony. Obviously a tribunal must have valid reasons for coming to
a different conclusion. Where one or more of the arbitrators has relevant
expertise in the area where an expert witness is being utilised, it must ensure
that it does not make independent assessments without giving the parties an
appropriate opportunity to know the tribunal's thinking and to respond. (215) If it
is a tribunal-appointed expert, the tribunal can naturally call for reasoning to be
elaborated.

A tribunal needs to ensure that it properly establishes the respective roles of the
tribunal on the one hand and a tribunal-appointed expert on the other. In some
legal systems it is inappropriate to ask the expert the very question that the
adjudicator is to decide as that would appear to usurp the latter's function. (216)
One example noted above is where the tribunal asks the expert to sift through
documents that the tribunal does not wish to analyse. A tribunal at least needs
to understand and accept the methodology and logic and accept the
thoroughness and likely accuracy of the assessment. Allowing parties to
challenge the assessment is a further way to ensure that the tribunal exercises
sufficient independent judgment. It is less contentious if the expert is merely
assisting the tribunal to sort the evidence, understand the technical terms and
collate testimony to the various issues being considered. (217)

page "948"

12.13.3. Tribunal Experts and Tribunal Deliberations

Tribunal-appointed experts should not be allowed to sit in on the tribunal's final

deliberations or assist in any way in the drafting of the final award, (218) although
in Luzon Hydro Corp recourse against the award was denied when the expert
simply reviewed a draft of the award to check for technical errors. (219) Such
assistance should not be problematic where the tribunal has already deliberated
and decided and simply seeks advice as to terminological errors or ambiguities,
but it would always be preferable to ask the parties permission.

While the expert should not be involved in deliberations, a situation might arise
where a tribunal is uncertain of a matter during its deliberations and would
wish for input from the expert. If the tribunal would naturally have asked the
expert the question if it occurred to it during proceedings, there should be a
prima facie entitlement to do so during deliberations. At this stage, however,
there is an additional due process concern. Where this is considered necessary,
it would be preferable to make that request in writing, notify the parties and
allow them a short but reasonable opportunity to comment on the expert's
responses.

12.13.4. The Expert's Right to Evidence and Assistance

Arbitral rules do not generally allow tribunal-appointed experts to question the

parties or their witnesses. Express assistance tends to be limited to requiring
the parties to deliver documents and relevant materials as required. An example
is Article 6.3 of the IBA Rules of Evidence 2010. A party who has been asked to
provide information may object on the grounds stipulated in Article 9.2. While
there is no express reference to questions, if an expert believed certain
questions were necessary for the opinion, it could either explain why this is so,
leaving the tribunal to elicit the information, or invite the tribunal to call for the
information prior to the report being concluded. While experts do not have a
direct right to question factual witnesses, in some cases it may be helpful if they
are allowed to do so. (220) If that was to occur, an appropriate protocol should be
devised by the tribunal to ensure due process.

A tribunal-appointed expert must in fact follow procedures analogous to due

process norms in all appropriate activities. (221) While an expert is entitled to call
for page "949" assistance, all correspondence should be sent to all parties.
Meetings should not be held without representatives of all present, absent
agreement to the contrary. (222) Communication between the tribunal and the
experts should also be communicated to the parties' representatives so that they
have the chance to comment on questions and directions that are contained in
such communications. (223) If a party fails to provide information, this should be
dealt with scientifically by the expert as part of the overall evidentiary matrix.
Any adverse inferences that may be merited should be left to the tribunal.

12.13.5. Party Access to Expert Experimentation

One question is whether the parties have an absolute entitlement to be present

during the tribunal expert's experimentation and analysis or only in relation to
consideration of their testimony and report. The latter approach was taken by
the Paris Court of Appeal in Carter v. Alsthom. (224) Commentators have differed
in their response to this case. (225) The parties' entitlement to be involved in the
report preparation stage is also supported in the European context by Article
6(1) ECHR, although this does not go so far as to require access to every
document viewed or meeting undertaken by the expert. (226) If an expert opinion
is to be based on some experimentation, the parties' representatives could be
invited to be present or some other appropriate record made so that they can
test the process and not be bound by the conclusions reached. The parties
should at least have access to all information considered by the expert.

12.13.6. Experts as to Applicable Law

In common law litigation, questions of foreign law are questions of fact to be

proven via witnesses and/or via exhibits of primary and secondary sources.
Civilian legal systems generally rely on the judges to know foreign law. page
"950" In international commercial arbitration, particularly where there are
three arbitrators, at least one of the tribunal members is likely to be an expert
on the particular law. It is not accurate to describe it as foreign law.

There is no express stipulation as to the way international commercial

arbitration tribunals are to determine the content of applicable law. This is
discussed in sections 10.6 and 13.16. At this point it merely needs to be noted
that experts will often give evidence as to applicable law. One difficulty with
legal experts, whether party or tribunal appointed, is that they may find it
harder to follow their mandate to assist the tribunal with reasoned opinions
rather than make an ultimate determination on the legal question in issue. A key
difference is between providing an opinion on what the law actually is in aid of
the tribunal's subsequent application of such law to the facts as ultimately
found, or conversely tendering an opinion as to the proper conclusion based on
the factual material presented. The latter is inappropriate but in many cases it
will be difficult to easily ensure a distinction. The situation is different where the
key issue in dispute is the content and meaning of that law. A tribunal can deal
with these problems by indicating a number of factual scenarios to the expert
and asking what the legal conclusion would be in each case.

Typically each party will have counsel who are experienced in the relevant law.
Written submissions will cover legal issues. The same will occur with oral
submissions. Because there are no strict rules as to the tendering of
documentary evidence, legal counsel can provide all relevant primary and
secondary sources as to the nature of the relevant law, leaving it to an expert if
necessary to deal with matters of analysis and opinion. In some jurisdictions, a
tribunal might require counsel to submit an affidavit in support of their
contentions. (227) The problem in relying on legal counsel alone is that they are
by definition partisan. While they may not mislead the tribunal, they will
naturally and quite properly argue for the view of the law that suits the client
where there is any grey area involved. While that may be the reality with some
party-appointed experts as well, that is not as ethically legitimate as partisan
counsel submissions.

If all members of the tribunal are qualified in the applicable law, it could be
reasonable to reject an application for any tribunal-appointed legal expert
evidence. If only one member of the tribunal is expert in the particular law, this
is more problematic as it will often be a party-appointed arbitrator. This will
often be because one party succeeded in the negotiations in having their
national law applicable and chose a local person as arbitrator accordingly.

Tribunals should be wary if they are expert in similar but not identical legal
regimes to that applicable in the case as they could too easily overlook nuances
and variations. Examples would be common law arbitrators familiar with UK
law page "951" dealing with Commonwealth countries where the laws are
presumed to be highly derivative. In selecting a tribunal-appointed legal expert,
there might also be prejudices by some arbitrators as to whether they prefer the
opinions of practitioners or law professors. Those from a civilian background
can tend to respect the views of professors. (228) Common law
practitioner/arbitrators can at times have a more negative opinion of academia!

12.14. Controlling Expert Assistance and Testimony

12.14.1. Introduction

Tribunals generally have a broad discretion as to the way they will hear expert
evidence. If the parties are allowed to present their own expert evidence, this
will invariably conflict. There is then the difficult question as to how a tribunal
should resolve which of the conflicting views to prefer. The key challenge is that
a tribunal will typically need to find a way to resolve conflicting expert
testimony on matters that are often outside the expertise of the tribunal
members themselves. Because of this, the traditional approach of hearing
experts individually and then trying to synthesise conflicting testimony, is
thought by many arbitrators to be sub optimal. Avoiding that problem by relying
instead on tribunal-appointed experts raises other contentious issues. A number
of alternative approaches are gaining favour. These include joint appointments,
pre-hearing meetings between experts, expert conferencing at the hearing, an
amalgam of party and tribunal-appointed experts, expert teaming, expert
facilitators and codes of conduct for experts.

If a tribunal intends to engage in any of such ongoing processes, this should be

made clear at the outset as there is a very different mindset needed between
separately presenting an opinion and working collegially, albeit with opposing
party-appointed experts.

12.14.2. Number of Experts

There can be no blanket rule as to the number of party-appointed and/or

tribunal experts. Nevertheless, the ICC recommends that each party is allowed
only one expert for any particular area of expertise. (229) A party advocating a
higher number should need to be able to clearly explain why this is required.

page "952"

12.14.3. Questions for the Expert, the Expert Brief and the Content of Expert
Reports

As noted above, a key issue in expert evidence is to allow the tribunal to make
an independent decision, albeit aided by experts and in doing so, find a method
of assistance that truly allows the tribunal to exercise independent judgment. In
this sense the expert's report and the brief upon which it is based are both
crucial. A tribunal should do everything reasonable within its power to ensure
that each report is prepared in a way that aids the tribunal in its mandate. The
need is the same regardless of whether it is a party-appointed expert or a
tribunal-appointed expert, although the procedure and entitlements may vary in
each case.

The overriding suggestion is that the clearer the brief and directions to any
expert, the more likely that the report will be useful. In turn, the more that all
expert reports are optimised in this way, the less concern there may be as to
whether experts are party or tribunal-appointed. Similarly, the more that the
reports clearly articulate all required material, the less there may be a need for
innovative conferencing options or conversely, the more efficient such options
are likely to be. Hence, the key suggestion is that a tribunal should give
particular attention to the brief it provides to a tribunal-appointed expert and
the directions it may give as to the required content of the reports of all experts,
whether party-appointed or tribunal-appointed.

12.14.4. Tribunal-Appointed Experts

The very appointment of a tribunal expert is based on a tribunal view that some
expert assistance is required. At an appropriate stage, the tribunal can identify
the particular questions on which the expert opinion is sought. The more that
experts get such guidance from the tribunal, the more focused their work is
likely to be. Conversely, too much direction at an early stage may prejudge key
issues, particularly as the direction will typically be given before the entire
evidentiary record is known and the tribunal itself may not be fully across all
technical matters. (230)

The tribunal might identify the questions independently or in some cases seek
to do so in consultation with the parties. (231) Even if the tribunal decides
independently to ask the questions, it would be appropriate to give the parties
an opportunity to comment on those proposed, or suggest additional questions
to be presented to the experts. There is also a difference between posing neutral
page "953" questions or providing questions that show the thinking of the
tribunal and which require confirmation or challenge by the expert. Where
there is more than one expert, the questions might be put to them individually
for their individual response. The parties might be granted an opportunity to
make written submissions to the selected experts prior to the latter rendering
their report. Once the expert reports are provided, the parties might then be
given a further opportunity to make comments on them.

The terms of reference of an expert should consider the questions being asked,
the information to be considered, the tests and investigations if any to be
conducted and the logistical aspects of the process. (232) Logistics should include
the permissible contact or methods of contact between the expert and the
parties; obligations on the parties to assist and provide specific information;
styles, format and length of the opinion; timetable; duties of independence and
confidentiality; fees and expenses including timing and methods of payment. A
tribunal needs to consider whether contact with parties and requests for
information should occur directly or through the tribunal. It has been suggested
that it would be desirable for the tribunal, the parties and the experts to all sign
terms of reference, (233) but this may not always be practical.

12.14.5. The Content of Expert Reports

There are a number of suggestions as to the material to be included in reports.

The IBA Rules of Evidence 2010 Article 5.2 indicates in relation to party-
appointed experts that the report shall contain:

• The full name and address.

• A statement regarding his or her present and past relationship (if any) with
any of the parties, their legal advisers and the tribunal.
• A description of his or her background, qualifications, training and
experience.
• A description of the instructions pursuant to which the opinion and
conclusions are provided.
• A statement of independence from the parties, their legal advisers and the
tribunal.
• A statement of the facts on which the opinions and conclusions are based.
• The opinions and conclusions including a description of the methods,
evidence and information used in arriving at the conclusions.
• If translated, a statement as to the language in which it was originally
prepared and the language in which the expert anticipates giving testimony at
the hearing. page "954"
• An affirmation of a genuine belief in the opinions expressed.
• Signature, date and place.
• Documents relied on that have not already been submitted shall be provided.
If the report is signed by more than one person there is to be attribution of
the entirety or specific parts to each author. (234)

When acknowledging the nature of an independent opinion the Civil Justice

Council Expert Witness Protocol makes a sensible suggestion that ‘a useful test
of “independence” is that the expert would express the same opinion if given the
same instructions by an opposing party’. (235)

Section 12.5.3 above looked at the role of counsel in assisting in the preparation
of witness statements and witness preparation. The same ethical issues arise in
relation to party-appointed expert reports. The main difference is that an
experienced expert will typically draft the report on their own. Nevertheless,
circumstances may arise where counsel would have hoped that the expert
report came to different conclusions or stated conclusions more forcefully. The
Civil Justice Council Expert Witness Protocol suggests that ‘(e)xperts should not
be asked to, and should not, amend, expand or alter any parts of reports in a
manner which distorts their true opinion, but may be invited to amend or
expand reports to ensure accuracy, internal consistency, completeness and
relevance to the issues and clarity’. (236)

There are difficult policy questions as to the disclosure of instructions to

experts. On the one hand, to properly understand the nature of an opinion and
to confirm its independence, one might well wish to see the instructions.
However, for a party-appointed expert, such communications could be claimed
to fall within the policy ambit of legal professional privilege. While that would
depend on the law of privilege that applies, the presumption would be against
privilege applying as the instructions are for a document to be presented to the
tribunal.

Article 5.3 allows for revised or additional expert reports that respond to
matters contained in other reports and witness statements or submissions not
previously presented. Tribunals may need to keep careful control to ensure that
tactical manoeuvring does not misuse this provision to try and get the last word
with a plethora of submissions on or near a cut-off date. (237) Article 6.4 of the
IBA Rules of Evidence deals with the contents of reports of tribunal-appointed
experts. The above headings are again relevant, save that a tribunal appointed
expert does not need to describe the instructions pursuant to which the opinion
was provided (which would have been circulated by the tribunal in any event)
or a page "955" statement of independence (which would have been dealt with
at the appointment stage). (238)

The CIArb Protocol also outlines the matters that ought to be dealt with in an
expert's written opinion. These include:

• Details of any past or present relationship with any of the parties.

• Background, qualifications, training and experience of the expert.
• A statement as to the instructions the expert has received and the basis of
remuneration.
• A statement as to the facts and matters including assumptions which have
been considered in reaching the opinion.
• A statement of the facts and matters including assumptions upon which the
opinion is based.
• A statement of the opinions and conclusions reached and a description of the
methodology and evidence used in reaching the opinions and conclusions.
• A statement as to matters on which the expert has been unable to form an
opinion.
• A statement of the matters, if any, outside of the area of expertise. (239)

The Protocol requires that the opinion only address the issue or issues approved
by the tribunal. It must also contain an expert declaration. (240) The Protocol calls
for reports to be as brief as possible, reference all appropriate documents and
sources, but not unduly extract material from other documents. Annexed
material should be reasonably necessary in support of the opinion reached.
Experts could be asked to declare their availability to perform the specified
tasks as well as their independence.

Because a key part of expert testimony is as to opinion, an important way to

analyse its probative value is to understand the expertise of the person and the
methodology and reasoning behind their conclusions. If the report does not
indicate the qualifications and experience to a significant degree, this should be
tested by the tribunal. However, protracted arguments about qualifications and
expertise would potentially add to the costs and delay.

The most significant aspect of the report is the reasoning, including any
assumptions, the evidence on which it is based, degree of certainty and an
explanation of why contrary opinions are not preferred. Here it may well be
appropriate for a tribunal to give even more detailed guidance than outlined
above. A number of useful suggestions have been made in domestic court
practice directions, protocols or codes of professional bodies. (241) Court rules
will tend to require the expert page "956" to acknowledge reading Court
promulgated guidelines, which in some jurisdictions, will elaborate that the
expert witness has an overriding duty to assist the court and is not an advocate
for a party. The expert might be asked to acknowledge that they understand
their role being to assist the court and that this inevitably has an educative
function and demands enough reasoning and analysis to allow the adjudicator to
make an independent determination. The expert should make clear to what
extent the opinion is based on personal knowledge or on assumptions. Where it
is assumptions, the report should indicate whether the assumptions are
provided through the instructions of counsel or are assumptions of a
professional nature.

If the opinion is based on disputed facts, the report must be particularly clear as
to what assumptions of fact the opinion is based upon. When considering the
facts, material and assumptions upon which the opinion is based, facts and
literature should be identified and the expert should attest to having made all of
the inquiries they believe are desirable and appropriate and have not withheld
any matters of significance. When providing their opinion they should indicate
whether it is provisional or qualified, and whether further information would
allow a more concrete conclusion either way. If the opinion is incomplete or
provisional for any reason, the expert should clearly explain why this is so and
what added material would be appropriate. This will help the tribunal make
assessments as to the weight of evidence. If an expert has relied on the research
of others, this should be explained and an indication given as to why the
methodology of that research is sufficient. Where tests have been undertaken,
the report should indicate the methodology used, who undertook the test, under
what supervision, with what qualifications and ideally a justification for the
methodology employed.

The expert's report should also indicate the factors that affect the certainty of
the proffered opinion. For example, an expert should indicate if the opinion is
based on a theoretical view about which there is a significant divide in the
professional community. Similarly, if the expert was denied access to key
information or did not have time to do the most appropriate tests, this should
also be indicated. This will not only help the tribunal evaluate the opinion and
allow the other parties to consider how to challenge it, but will also assist the
tribunal in the inferences it draws from non-disclosure, in considering further
orders as to testing and even in possibly getting an agreement by the parties to
resolve the matter by the result of the more elaborate test.

An expert's obligations are ongoing. Hence, if an expert subsequently changes

his or her opinion, this should be immediately notified to all relevant parties.
This might occur on simple reflection or after reviewing other reports. In other
cases, further testing or newly published research may lead them to modify
their view.

12.14.6. A Proposed Code of Conduct

The following is a model code of conduct on matters that might be relevant for
proposed experts as proposed by the author. These might be included in
directions or converted into questions for an expert to consider in the report,
with a page "957" requirement of an acknowledgment that the Code has been
followed. Conversely the items could simply be used as an aid to drafting
specific questions, directions or terms of reference. The format should be less
important than the clear articulation of the requirements. It is not intended as a
boilerplate for all cases but merely a guide to aspects that may be appropriate
on a case-by-case basis and is based on the foregoing reflections and some of the
leading models used to date in arbitration and domestic litigation. Many might
disagree with the utility of directions alone. Nevertheless, the key point as noted
at the outset is that the better the instructions, the more likely the utility of the
report. This must be of significant value in its own right and also as a result of
the ongoing debate about party appointed versus tribunal appointed experts
versus conferencing and other methods. The debate deals with what would
appear impressionistically at least, to be some dissatisfaction among the arbitral
community with the way expert testimony is generally dealt with.

Proposed Code of Conduct

The Code aims to cover both party-appointed and tribunal-appointed experts

(although if actually utilised, it would be preferable to separate out for each type.
Thus the following items are applicable to both party-appointed and tribunal-
appointed experts, save where specifically referring to one type only).

1. The expert's function and duties.

I acknowledge that my role is to assist the tribunal to decide issues where

expert evidence is appropriate and to do so in an independent, impartial,
objective, expeditious and efficient manner. I understand that it is not my task
to make a decision in place of the tribunal but, rather, it is to provide the
tribunal with a sufficiently reasoned opinion to allow it to make an
independent determination of the matters in issue before it. I understand that
my role is therefore that of providing reasoned advice so as to not only
present my opinion but to also do so in an educative way that helps the
tribunal understand the reasons behind it and helps the tribunal to form an
independent view on whether and to what extent my opinion should be
accepted.
2. Statement of independence

I acknowledge that I am required to be independent from the parties, counsel

and the tribunal. In accepting the appointment I declare that I have the
necessary independence and there are no conflicts of interest of which I am
aware. I also acknowledge a distinct obligation to disclose any fact of which I
am aware that could allow a party to form a contrary view as to my
independence.

I acknowledge that my duty of independence and the related duty of

disclosure is ongoing until such time as my evidence is completed.
 Consequently, if any circumstance arises or comes to my attention that may
allow any reasonable observer to question my independence, I will disclose it
to the tribunal.

(For party-appointed experts) I also understand that while I may confer with
the party appointing me and its counsel by way of briefing and while my page
"958" fees may be paid for by a party, my role nevertheless remains as
outlined above and that it is not my role to advocate the position of the party
who has appointed me. I acknowledge that the acceptance of a fee for my
report from a party does not itself compromise my independence but it would
if the fee was dependent on the result in the dispute and I acknowledge that
my fee schedule is not of that nature. (For party-appointed experts) I
acknowledge that my opinion was not influenced by any party, party counsel
or any other person. While I am aware that counsel is allowed to invite me to
clarify a draft, ensure completeness, relevance, internal consistency and
accuracy, nevertheless, I have not been asked to and will not construct any
part of my report in a way which distorts my true opinion. The approach I will
take to my task means that the opinion that I will express in this matter would
be the same if my instructions were received from the opposing party or the
tribunal.
4. Expertise

In accepting the appointment I attest to my qualifications and experience

being suitable for the provision of an expert report. In areas where I do not
claim sufficient expertise, these will be articulated and explained in the report.
In appropriate cases, the report will include an indication as to why I believe I
am particularly suitable to provide an expert report of this nature.
5. Availability

I acknowledge that in agreeing to provide this report, I am willing and able to

provide my report and testimony within the timeframe indicated at the time
of my appointment.

I understand that, in giving evidence, I may be asked to do so separately or

may be involved in conferencing with other experts, either before the hearing,
during the hearing or both. In such circumstances I acknowledge that I may
be called on to seek to generate a joint report with other experts, indicating
areas of agreement and disagreement and providing full and adequate
reasons for the latter. I acknowledge my willingness to do so in good faith and
expeditiously.
6. Brief and contact with the parties

I will indicate in my report the instructions on which my report is based (for

party-appointed experts – save for matters properly covered by legal privilege
as applicable in the arbitration).

(For a tribunal-appointed expert) I am entitled to seek documents from the

parties and may do so (directly or via request to the tribunal). I may pose
 questions for the parties (directly or only via request to the tribunal, which
can determine which questions will be put). I am aware that I should not
make contact with either party without the consent and the presence of the
other.
7. Methodology

I acknowledge that an important part of my report is, where appropriate, to

indicate to the tribunal the methodology I have adopted, the reasons for doing
so, my reasons as to why alternative methodologies were not thought
appropriate, the advantages and disadvantages of differing methodologies,
page "959" any presumptions of fact or theory my opinion is based on and an
indication of any methodology I would have wished to have employed where I
was not able to do so. In the latter event, I will explain why the methodology
was not adopted, why I think it would have been preferable and, if possible,
the likely implications if it had been employed.
8. Testing and experimentation

Where testing and/or experimentation is involved in the preparation of my

report, I will give particular attention to explaining why those tests and
experiments were conducted, why they were thought optimal, what if any
problems there are with that methodology and the potential benefits and
problems with alternative methodologies. I will follow any reasonable
procedures directed by the tribunal, such as in relation to access of the parties
to the tests themselves and their results.
9. Findings and reasons

I acknowledge that my key function is to provide a reasoned explanation for

the opinion I have formed. In providing my opinion, I will include any
assumptions on which it is based, explain the basis of those assumptions,
generally indicate the degree of certainty of my opinion and explain why
contrary opinions are not preferred. (For party-appointed experts) If I have
been directed to provide an opinion based on a particular assumption that I
am not to evaluate independently, I will make this clear in my report.

(Conversely, the Code could indicate that a party-appointed expert is

nonetheless under a duty to evaluate the reasonableness of any assumptions
that are required to be utilised.)

In giving an indication as to the level of certainty of my proffered opinion, I

will indicate the factors that affect certainty.

I will indicate the evidence upon which my opinion is based, including the
extent to which it is based on personal knowledge and experience. I
understand that my role is to consider and allude to all material facts,
including those which might detract from my ultimate opinion. If my opinion
is based on the research or experiments of others, this will be explained with
appropriate discussion and citation of the work relied on. If it is based on
 speculation, I will explain the reasons and justification.

I will indicate whether the conclusions are provisional or qualified and the
reasons for this and, in either event, whether further information would allow
a more concrete conclusion, and if so, what information would be
appropriate.

I acknowledge that I am to deal with all of the questions posed by the tribunal
and not deal with matters outside of those questions. If I feel that other
matters are also important or have any concerns about the nature of the
questions, I should seek prompt instructions from the tribunal. If I am unable
to proffer an opinion on any matter, I will make this clear and indicate the
reasons for this.

I acknowledge that my report should reference all appropriate documents

and sources that should reasonably be considered in support of the opinion
reached.

page "960"
10. Confidentiality

I acknowledge that the arbitration proceedings are confidential and I may not
divulge any aspect of the process without the agreement of the parties and the
tribunal.
Signed: (242)

Another way to achieve the same outcome is through terms of reference or

specific questions in a brief. The following is a list of topics that could be
covered in draft terms of reference. The detail could obviously draw on the
above proposed code:

Draft terms of reference for expert reports

1. Role of expert.
2. Duties of independence.
3. Qualification and expertise.
4. Questions for consideration.
5. Required tests and investigations, including protocols.
6. Contact with the parties including requests for information and
questions.
7. Obligations on parties to assist the expert.
8. Time limits for report and ramifications if late.
9. Duty to appear at a hearing and confer with other experts if required.
10. Style, format and length of report.
11. Language of report and oral evidence and translations.
12. Confidentiality.
 13. Fees and expenses.
14. Accompanying documentation.
15. Affirmation of genuine belief.
16. Signature (if joint report, providing indication of individual responsibility
for parts).

12.14.7. Joint Reports of Party-Appointed Experts

As noted above, one problem with the use of party-appointed experts is that
their reports and testimony will invariably conflict. A tribunal then has a
difficulty in knowing how to reconcile the conflicting opinions. One possibility is
to ask them to prepare a joint report for the tribunal, identifying common views,
areas of page "961" disagreement, the basis for that disagreement, and methods
by which the tribunal could legitimately resolve the conflict. Cross-examination
in relation to separate party-appointed expert reports is often time consuming
and repetitive. (243) The above code could be modified for such purposes,
highlighting the need for clear reasons for differences and the way to resolve
them.

Even if a joint report is not possible, it is important to have distinct reports

follow an identical order and explain the reasons for points of difference.
Another suggestion is to have experts exchange draft reports at an early stage.
(244) Another approach is to invite each party to provide specific questions to be

addressed by the opposing experts, (245) subject to tribunal scrutiny and

acceptance.

12.14.8. Pre-hearing Meetings between Party-Appointed Experts

Another methodology in support of refining the opinions to aid the tribunal is to

try and have the experts meet before the hearing so as to identify the issues they
agree upon, the areas of difference, and allow them to concentrate on the latter
before the tribunal. Pre-hearing meetings between opposing experts can ‘(1)
clarify technical and factual issues, (2) outline areas of agreement and
disagreement, (3) focus on relevant points, (4) narrow down the differences
between expert reports, (5) encourage scientific debate and, as a consequence,
(6) render the taking of expert evidence more time and cost efficient’. (246)
Meetings at various stages may also inspire experts to act more independently
and objectively. The ICC Task Force recommended the use of pre-hearing expert
conferencing. (247) Experts may be more willing to present a nuanced view to
their peers than they would in the context of confrontational cross-examination.
Drafting joint reports is more likely to lead to attention being primarily directed
to the most significant issue, with peripheral matters appropriately relegated.
Witness conferences may increase certainty by binding expert positions taken
during conferences, although this could also induce experts to be reticent as a
result.

Some important practical issues arise. The first is timing. Should such meetings
occur before the experts have drafted their first reports or after; should
meetings be allowed with counsel or the parties present; if the parties and/or
counsel are entitled to be present, should there be a protocol as to any
comments they can page "962" make; should the tribunal and/or the parties set
an agenda or list of questions for the expert? It is also important to determine
from the outset of the procedural discussions with the parties whether the
content of any discussion between the experts is privileged or may be referred
to at an ensuing hearing. Most importantly, there is also a need to determine
whether any agreement reached between the experts is binding on the parties,
absent their own agreement to that effect. (248) Absent express agreement by the
parties, a tribunal will need to make decisions on these issues on a case-by-case
basis. It is better to clarify permitted behaviour at the outset than try and
resolve some of these questions after the event.

Article 5.4 of the IBA Rules of Evidence 2010 allows for the tribunal to order
party-appointed experts to meet and confer on such issues. It provides that at
such a meeting the party-appointed experts shall attempt to reach agreement on
those issues as to which they have differences of opinion. The new Rules now
contemplate that such a meeting could be called for either before the first draft
reports or after.

The CIArb Expert Protocol also deals with pre-hearing meetings between
experts and attempts a more elaborate Code. The CIArb Expert Protocol calls on
the experts to meet before they prepare their first report. Experts are to hold a
conference to identify issues on which opinions will be sought, identify tests or
other methods of analysis to be conducted, if possible by agreed methodologies
and the manner of conduct of any such tests. A tribunal may direct the experts to
exchange draft summary opinions for such meetings which will be privileged
from production to the adjudicator and, hence, are without prejudice to the
parties' positions in the dispute. After such a discussion, the experts would then
prepare and serve a statement setting out issues and opinions on which they
agree, tests and methods of analysis agreed upon, and any reasons for
disagreement as to tests, analyses or methodologies. If there is no agreement on
the tests or the methodology, individual tests should still be conducted in the
presence of the other experts. After any necessary testing, written opinions are
completed and exchanged. The Protocol also allows each expert to provide a
further written opinion dealing only with matters raised in the written opinion
of the other experts. The tribunal may also direct further conferral between the
experts and further written reports, either jointly or separately. The tribunal
may also hold preliminary meetings with the experts. Each expert who has
presented a written opinion must be available for oral evidence unless the
parties agree to the contrary and the tribunal accepts the agreement. If the
expert does not appear and does not have an adequate reason, the tribunal shall
disregard the opinion unless the parties agree otherwise and the tribunal
supports the agreement. Agreement that an expert need not give evidence is not
taken to be acceptance of the content of the written opinion. Tony Canham
argues to the contrary that it would normally be better to exchange reports
before the first meeting, as conclusions on complex technical matters are usually
only the result of careful study of the facts page "963" and reasoning in the
opposing expert's report. (249) Obviously it is harder to have an expert change
their mind after they have already prepared a report. (250)

The items of agreement or otherwise are sometimes described as a Scott

Schedule, being a statement between experts as to the matters where they agree
and where they differ and the reasons for differences in view. (251) Wherever
such schedules are used, it is important that the tribunal take a controlling
interest to ensure that matters of disagreement are not simply presented in
adversarial form, but instead, provide the best outline of competing arguments
in order for the tribunal to render its own determination.

It is important to carefully consider that while it might always seem desirable to

attempt to refine the issues in dispute before a hearing, there can be problems
when party-appointed experts are directed to attempt such a process if one or
more take it upon themselves to act as advocates for their appointing party or
are intransigent and refuse to acknowledge facts that ought readily be conceded,
(252) either on their own volition or based on instructions of counsel or would

only contemplate agreement subject to ratification by counsel. (253) This can add
to the costs without commensurate benefit, and can lead to tensions between
expert witnesses. (254) Furthermore, if the experts have already lodged differing
opinions, requiring them to attempt to reach agreement almost forces them into
an adversarial and advocacy role where they remain convinced that they were
right. One can direct that party-appointed experts confer but one cannot easily
force them to leave any mindset they might wrongly have as to an advocacy role
on behalf of their appointing party. A party-appointed expert who truly believes
that they have an ethical duty of complete openness and honesty and an
obligation to assist the tribunal, can write a valuable unilateral report to that
effect. Conversely, a party-appointed expert who believes that they must help
the party win or at least not be of assistance in helping them lose, may be
reticent in discussing technical matters with counterpart experts, be reluctant to
agree on issues that might undermine their party's position, (255) and may
generally see the meeting as an opportunity for advocacy or probing for
weaknesses, rather than a good faith attempt to provide the page "964" tribunal
with clearly delineated points of agreement and clearly explained points of
difference. One response to this would be to demand that disagreement be
based on an expressly reasoned position which could then be analysed in
subsequent proceedings. (256) Pre-hearing meetings might also be taped as an
inducement to proper conduct.

Expert meetings should normally not seek to agree on facts in dispute as this is a
matter for evidence before the Tribunal. (257)

12.14.9. Expert Facilitation

Because of these problems with expert meetings, one possible solution is to

have an independent neutral as co-ordinator. Van Houtte has recommended the
use of an experts facilitator for that task. (258) In complex matters, the facilitator
can co-ordinate a range of expert opinions covering different issues and
disciplines and organise dialogue and timetables. Van Houtte advocates a person
with expertise in the particular area. (259) The advantage is that they speak the
appropriate ‘language’. The disadvantage is that they would need to be careful
to keep their own views to themselves. A substantive expert who has also had
arbitral experience might be ideal, given their understanding of the kind of
reports that will assist the tribunal. Unless there is complete agreement among
the experts, the reasoning is at least as important as the conclusions themselves
as without reasoning, the tribunal has no meaningful way to form an
independent judgment. Van Houtte argues that the parties themselves should
agree on the facilitator. If the parties are not able to come to a speedy
agreement, the tribunal may suggest a list and allow for reasoned objections
leading to an individual's selection. (260) Another possibility is to use a selection
body such as the International Centre for Expertise of the International
Chamber of Commerce. (261)

In a domestic arbitration in Australian, Anaconda Operation Pty Ltd v. Fluor

Australia Pty Ltd which involved numerous expert witnesses, the tribunal
appointed two ‘independent assessors’ whose role was simply to chair meetings
between the experts and produce schedules outlining the key issues and the
views of the party-appointed experts on each of them. The assessors also sought
to identify areas of agreement so that the matters truly in dispute could be
clearly identified. (262)

page "965"

It is important to clarify the exact status of the facilitator, including whether the
expert facilitators could be called to give evidence at an oral hearing and could
be cross examined. That would not be desirable and should be clearly
articulated, especially as the IBA Rules of Evidence 2010 allow the parties to
question a tribunal-appointed expert and without a clear statement to the
contrary, the facilitator could be seen as such.

12.14.10. Experts and Oral Hearings

Similar rules apply to the obligations of experts at a hearing as with other

witnesses. For party-appointed experts, the other party can generally decide
whether to call the expert for cross-examination. If they do and the expert does
not appear, the report will typically only be accepted in exceptional
circumstances. As noted, if a party does not call for an opposing witness, that
does not constitute acceptance of the content of their report where the parties
concurred (263) but care needs to be taken with unilateral decisions to that effect
and the impact on the weight to be given to unchallenged opinions. Parties must
also be entitled to comment on the views of tribunal-appointed experts and seek
to test that evidence. Article 26(2) of the Model Law indicates that if the tribunal
appoints an expert, the parties must be given an opportunity to challenge the
evidence unless they have agreed to the contrary. A number of other statutes
and rules also expressly allow for such entitlements. (264)

Expert witnesses will usually be heard after fact witnesses and be heard back to
back if not being conferenced, so that there is a broader factual record on which
they can comment. Conversely, in some cases a hearing might be bifurcated,
with experts heard at an earlier stage if that will help promote efficiency in the
ensuing proceedings. An example might be hearing experts on applicable law in
support of a partial award on that issue. It might even be the case that a party
seeks to utilise expert evidence in aid of an application for interim measures.

The traditional means of examining and cross-examining experts in the common

law can be highly problematic. Direct examination can too often flow from
rehearsed preparation. Cross-examination can be too controlled in order to
elicit unfavourable responses and bar the expert from expanding on their
opinion in a more nuanced way. (265) One issue is whether experts will be
allowed to provide added arguments in oral testimony over and above their
written opinions. Here a tribunal will need to balance fairness to the other party
and appropriate pursuit of the truth. Where the IBA Rules of Evidence apply,
Article 6.5 indicates that page "966" questions are to be limited to the issues
covered in the expert report. As noted below, this may be more problematic
with expert conferencing.

12.14.11. Expert Conferencing Before the Tribunal

Another methodology to resolve conflicting expert opinions is to have a

roundtable discussion with the entire group of expert witnesses and the
tribunal. The experts would explain and debate the differences in such a way
that the tribunal can make a meaningful determination. (266) Broad tribunal
discretions ought in principle to allow for witness conferencing. Nevertheless,
Redfern and Hunter suggest that a tribunal will rarely employ expert
conferencing techniques without agreement of the parties. If the parties both
want witness conferencing, the tribunal should accept it. (267) Conferencing of
experts or fact witnesses is contemplated by Article 8.3(f) of the IBA Rules of
Evidence 2010.

12.14.12. Advantages and Disadvantages

Many experienced practitioners extol the virtues of witness conferencing. It is

still desirable to consider the advantages and disadvantages of the process for
two reasons. First, different cases suggest different procedures and an
understanding of the advantages and disadvantages will help determine
whether it is an appropriate technique for a particular dispute. Second, where it
is utilised, an understanding of the potential problems should forewarn the
tribunal and the parties how best to engage to optimise the outcome. The
following discussion is mostly concerned with expert witness conferencing but
the policy issues also apply to many aspects of general witness conferencing,
discussed in section 12.8.9 above. Wolfgang Peter, the most influential
proponent of witness conferencing in international arbitration, in fact
contemplated simultaneous conferencing of all witnesses, expert and factual, in
appropriate circumstances. (268) There will obviously be differences in the
cost/benefit and fairness issues depending on whether conferencing involves an
entire witness team against the other, or whether there is a separation between
experts and factual witnesses or a further separation on an issue-by-issue basis.
General policy comments below need to be looked at in that context.
Key advantages of witness conferencing are that a skilled arbitrator can ask
particular experts why they disagree with the views of others, and ask for their
views as to the appropriate process or methodology to resolve the conflict; in
some page "967" cases the mere exchange of views will allow the tribunal to
form an opinion; in other cases the experts may agree on the kind of tests that
might be conducted that would best resolve the issue.; people are less likely to
lie in front of their peers; (269) discussions in conference will often help parties
recollect events (where that is relevant); (270) there should be a diminution of the
utility of unethical coaching in preparation for cross-examination; (271) and it
allows for a more relaxed environment. (272) Witness conferencing can also
identify over-simplifications in expert testimony. (273) Where opinions are based
on theoretical or factual assumptions, conferencing is more likely to bring these
to the attention of the tribunal and allow attention to be shifted to the validity of
the assumption itself. Another advantage of witness conferencing is that experts
are usually more experienced with round-table discussions of technical issues
than with cross-examination by expert legal counsel. Expert conferencing can
also help the experts come to understand the legal issues in contention. When
operating separately, they might see the key issues quite differently to the way a
legal adjudicator will. Conferencing also helps each expert understand the
factual, assumption and methodological bases of conflicting expert opinion
which should help in finding a way forward in resolving the differences.
Conferencing also forces counsel to consider opposing witness testimony and
reports in their own preparation, integrate the evidence in evaluating the
strengths of their case and develop an optimal case strategy. Time efficiencies
may also arise through the use of information technology with witness
conferencing, whereby a number of witnesses can view documents, models,
videos and the like simultaneously. (274)

Conferencing in the presence of parties may also assist settlement, given that all
can see the conflicting body of evidence and opinion that the tribunal will
ultimately need to resolve, (275) although there is also a danger that the inevitable
involvement of the tribunal in controlling the discussion could lead parties to
draw conclusions about the tribunal's supposed predilections. If the assessment
by the tribunal is accurate, this may not be problematic but there is a danger
that a probing question by a tribunal member might be wrongly looked at as a
negative view. Furthermore, concerns that the parties might misconstrue
questions might be a disincentive to tribunals appropriately managing the
discussion. Even if such a process is used, cross-examination is still typically
allowed to ensure that due page "968" process challenges are less likely and to
allow counsel to raise matters that might not have been dealt with under the
tribunal's direction.

Any option must also be evaluated against alternatives. Under the traditional
sequential approach, allowing individual experts to explain why transcripts of
earlier experts are wrong makes it extremely difficult for assessment by
adjudicators. (276) Redfern and Hunter suggest that the transcript of such a direct
dialogue between opposing experts is typically more helpful than cross-
examination between legal counsel and opposing experts. (277) Witness
conferencing may ultimately prove to be a useful means to try and reconcile the
differences in view between common law and civil law jurisdictions as to the
probative value of party-appointed experts. A dialogue through witness
conferencing of experts is not dissimilar to calling for expert meetings and joint
reports which also aim to elucidate the points of similarity and difference for the
benefit of the tribunal. (278) Finally, it is suggested that witness conferencing is
likely to significantly reduce the time taken to deal with witness evidence at the
hearing. (279) Born questions whether witness conferencing actually saves time,
given that it is most appropriately used in addition to traditional cross-
examination. (280) This might depend on the controls, if any, that a tribunal
imposes on counsel in cross-examination when attempting to cover matters
seemingly addressed at length at the conferencing stage.

There are some potential disadvantages as well. Psychological studies show that
if people express an opinion openly, a person with a dissenting view hearing a
range of views to the contrary is more reluctant to present their true thoughts.
An example would be a famous expert giving evidence first followed by a more
junior colleague who would not wish to be seen to contradict such an eminent
person. A further disadvantage of conferencing is that a number of experts on
the one team can hear a question and answer and might tailor their own
answers to maximise consistency in favour of their appointing party. The
problem of tailoring evidence based on what is heard from others is diminished
when written reports have already been written and exchanged. (281)
Conferencing will not work well if the expert is acting as an advocate on
instructions from counsel. If that were the case, that may well be more readily
evident to the tribunal during the conferencing process, rather than if the
witness were cross-examined in the traditional manner. A related issue is that
bringing experts together in an oral discussion will favour those who have page
"969" stronger personalities, who find it easier to present arguments orally and
who think more quickly ‘on their feet’.

12.14.13. Organising Witness Conferencing

For conferencing to be fair and effective, the parties must get sufficient advance
notice that the technique will be utilised and a clear indication as to whether
cross-examination will be allowed. The tribunal must ensure that it knows the
file sufficiently well, so that it can properly manage the discussion both in terms
of timing and direction. A tribunal also needs to be sensitive to the concerns of
counsel unfamiliar with such processes who will naturally be worried about
losing control and losing their ability to challenge adverse contentions, and the
witnesses themselves who may have differing experiences with the process.
That is no different to other more traditional aspects of handling witnesses such
as cross-examination, where counsel may have different experiences and
abilities as may the witnesses themselves. A tribunal also needs to be prepared
to ask questions on matters where they have little understanding and not be
concerned with losing face. (282)

Costs might be wasted if experts are included in discussion of matters not

important to their central testimony. In such circumstances it might be better to
separate the witnesses into discrete issues, (283) although with experts, it may not
be absolutely clear that they cannot meaningfully comment on matters raised
outside of their statements. This raises another important issue with witness
conferencing. As a general rule, a witness ought not to present evidence at a
hearing that clearly goes beyond the written statement without permission to
do so and without the tribunal considering the due process implications for the
opposing party. Yet a round-table discussion under witness conferencing will
typically have one expert commenting on the views of the others in a way which
will inevitably lead to this occurring. The difference is that when the challenge
to one expert comes from a question or comment from counsel or the tribunal,
such interventions are not new evidence. At the very least, forewarning as to
conferencing techniques will mean that the opposing party is aware that its
expert's views may be open to challenge from peers as well as counsel and the
tribunal.

A tribunal chairing a round-table discussion amongst people representing

opposing positions needs to be in control, dissuade inappropriate behaviour and
still allow all parties to feel that they had an adequate opportunity of presenting
their case through their appointed experts. The latter consideration will also
arise in the way a tribunal deals with interventions sought to be made by
counsel. The tribunal needs to ensure that each party is given an appropriate
opportunity to express their views to ensure that there are no potential due
process challenges. page "970" A key skill in conferencing is knowing when to
move back and forward between the opposing teams for questions and
responses. (284) The tribunal will need to explain to each the requirement of
truthful evidence, administer oaths where necessary and identify a protocol for
speaking, in particular to allow the tribunal to remain in control, assist any
transcription that is being conducted and ensure that witnesses do not speak on
top of each other. In controlling such a round-table discussion, the tribunal
needs to ensure that appropriate questions are asked. This will include ensuring
that the order of questioning is appropriate in the context of the claims,
defences and counterclaims as made and the burden of proof that applies, and in
terms of the content of the written statements and the level of expertise. There
may also be challenges in chess-clock arbitration if there is an imbalance in the
level of discussion between each party's witnesses.

There will be added difficulties if experts speak different languages and

simultaneous or sequential translation is required. The more complex the
technical issue, the more difficult it is to have expert translators that can cope
with both legal and scientific jargon and nuances. While this may pose problems,
it is not a particular problem of conferencing per se and would apply if the
witnesses gave separate evidence in any event.

A tribunal will also need to carefully manage its own questions and at the same
time allow counsel for each party to feel sufficiently involved. Raeschke-Kessler
suggests that questioning by the tribunal should not deal with credibility. (285)
The latter should occur during cross-examination. (286) There is also a question
as to whether the witnesses can themselves decide who is best able to answer a
particular question or whether counsel and/or the tribunal ought to be able to
give directions as to who must respond. The tribunal will also need to consider
to what extent one expert can question the other during a conferencing session.
A tribunal will need to be mindful that some experts might take on the role of
cross-examining other experts. While this would not be desirable, in practice, it
may be difficult to distinguish between legitimate questions put by one expert to
another to elicit the reasoning behind a professional opinion and instead, an
expert undertaking an advocacy role, seeking to undermine a perceived
opponent.

Even the shape of the room can be important and should seek to maximise
dialogue between the conferring witnesses, still allowing the tribunal to be the
central focus. Martin Hunter has suggested a variation in the seating
arrangements with the experts on a platform and a semi-circular ‘audience’ of
the tribunal in the centre and advocates and other persons on each side, in the
hope that this would reduce the tendency for inappropriate cross-examination
techniques. (287) page "971" In complex matters, there will need to be an
adequate mechanism for them all to view and discuss key documents, plans,
pictures, videos and models.

12.14.14. Cross-Examination after Conferencing

Common law counsel's concern with witness conferencing may be exacerbated

if their entitlement to cross-examination is held over until after a round-table
discussion with the tribunal, as by that stage key concessions may have already
been made. This also relates to the broader question of efficiency and the extent
to which a tribunal will limit cross-examination of matters it feels were
adequately covered in the conference. On the one hand, undue duplication
should be avoided. However, the very essence of cross-examination is to allow
previous testimony to be challenged. Thus a tribunal would find it difficult to
impose a priori controls as opposed to interceding where cross-examination
seems unproductive.

Where a subsequent cross-examination occurs, there is a question as to whether

the witnesses should then be separated. If not, a further question is whether
other members of the team can answer instead of the witness in whose
direction the cross-examination is targeted. Cross-examination flows from
evidence in chief and witness statements. At times there may be a need to be
clear as to which expert said what during conferencing that may then be
explored on cross-examination.

12.14.15. Expert Teaming of Tribunal-Appointed Experts

Klaus Sachs has proposed a protocol in relation to tribunal-appointed experts as

an alternative to party-appointed expert processes. The suggestion is for an
expert team to combine the advantages of party-appointed and tribunal-
appointed expert models. The tribunal would consult with the parties at an early
stage and invite them to provide the tribunal and the opposing party with a
short list of potential experts. The tribunal might invite the parties to comment
on the experts proposed by the other, particularly as to conflicts of interest. The
tribunal would then choose two experts, one from each list and appoint them
jointly as an ‘expert team’. The tribunal would then meet with the expert team in
conjunction with the parties in order to establish terms of reference. The expert
team would prepare a preliminary joint report based on the terms of reference.
The report would be circulated to the tribunal and the parties with each having
the opportunity to comment on the preliminary report. The expert team would
then review the comments and take them into consideration in preparing a final
joint report to be submitted to the parties and the tribunal. On the request of
either party or the tribunal, the members of the expert team would be present at
the evidentiary hearing and might be page "972" questioned on relevant issues
by the tribunal, the parties or by any party-appointed expert. (288)

The Sachs proposal argues that a team of experts selected by the tribunal
overcomes most of the concerns with tribunal-appointed experts. The selection
is based on lists provided by the parties and not purely on the tribunal's own
selection. The parties are given an opportunity to make comment although it
was noted in section 12.11.5 above that a range of list options are available for
any tribunal appointment, and consultation is not dependent on a team model.
One value of an expert team is that it has internal checks and balances not
available with a single expert. The two experts are more likely to act
independently as assistants to the tribunal as they are not independently
selected and paid for by the parties, although there may be a problem in parties
approaching potential experts prior to drafting the initial list. It would be hard
to proscribe that as counsel could legitimately argue that they need to interview
potential experts to see if they are available and suitable.

The reports are prepared in conjunction with the parties and the tribunal and
hence ought to be more cost effective. Because the parties, tribunal and the
experts meet together, it is less likely that the report would miss the key points.
(289) The parties can still comment on the report and cross examine the expert

team. A major advantage is the fact that there would not be conflicting reports
for the tribunal to handle, although conflicts could of course be included within
the joint report itself. Even here, the differences will be set out more logically
and can be more efficiently understood as compared to synthesising and cross-
referencing multiple expert reports.

One key benefit is likely to be party acceptance where they have influenced at
least one selection, and the hope that two experts will promote broader
coverage of the matters of concern and a more concerted effort to unite in aid of
the tribunal's deliberations. Problems would arise if the two experts cannot
agree, as in many cases they might be unlikely to be able to concur on a
recommended methodology by which the tribunal can resolve the issue. There
are unavoidable problems if the opinions depend on which of competing
respected theories each expert adheres to, although if this is known at the
outset, ideally one expert could be selected from each of the two key
methodological schools so that the tribunal is clearly and impartially seeking the
best possible briefing as to both. Other problems may still arise. Having two
experts rather than one also does not itself overcome the risk that the tribunal
delegates too much of the actual decision-making to the team. There is also the
same issue as to counsel control of strategy and party control over information
flows. In addition to the information in the custody and control of the parties
there are also the assumptions upon which various expert opinions are
commonly based. (290)

page "973"

The utility of the Sachs protocol may vary depending on whether party-
appointed experts are also utilised. The right to be heard must imply a right to
present conflicting evidence to that of the expert team as well as to challenge the
team through cross-examination. At the very least, there will be logistical issues
about the extent to which the expert team can explore the assumptions and
instructions given to the party-appointed experts. Mark Kantor makes the
important observation that a greater amount of information typically flows
between counsel and party-appointed experts than would be provided to a
tribunal expert. In such circumstances, there may be issues as to when and how
broad information may be called for and the applicability of adverse inferences.
(291)

Sachs suggests that it would be advisable that the terms of reference provide,
inter alia that:

(i) both experts retained must be impartial and independent;

(ii) the task of the expert team is to assist the tribunal in deciding the issues
in respect of which expert evidence is adduced;
(iii) the expert team shall only address issues identified in the terms of
reference;
(iv) the expert team is expected to submit a joint report providing only the
joint and mutual findings;
(v) each member of the expert team shall refrain from communicating
separately with the parties, the tribunal or any third party;
(vi) the expert team shall prepare its report ‘from scratch’ and shall rely only
on its own expertise;
(vii) the expert team shall seek any input and assistance required from the
parties;
(viii) in the preparation of the report, the expert team shall carefully examine
all briefs and documents submitted by the parties and shall address the
parties' views and concerns; and
(ix) the expert team shall be prepared to testify during an oral hearing and to
respond to questions asked by the tribunal and the parties, their counsel
and consultants. Areas of disagreement on which the experts cannot
reach a joint conclusion shall be identified and, if necessary, the parties
will be permitted to comment or submit additional (expert) evidence on
these. (292)

The following are suggested modifications to the proposed terms of reference.

Following on from the observations about the expert brief in section 12.14.3
above, the expert team might be advised that their assistance task involves them
carefully explaining the methodology used, the information relied upon, the
reasons underlying the conclusions, including, in particular, any assumptions
made and any impact upon the methodology and findings of schools of thought
about page "974" which experts may legitimately differ. This is important to
ensure that the expert team truly assists the tribunal in coming to its own
conclusions, rather than effectively taking over one key part of the decision-
making process. This is particularly important as a tribunal might feel even
more reluctant to depart from the view of a concurring expert team than a single
tribunal-appointed expert. A tribunal needs to be particularly sensitive to
situations where it is asked to evaluate the assumptions on which the expert
opinion is to be based. The issue may be circular where the parties disagree as
to those assumptions and where the appropriate assumption is itself a matter
for expert opinion.

As to para (iii) above, the expert team might be expressly entitled to invite the
tribunal to consider expanding the terms of reference where this is
subsequently seen to be necessary for a meaningful report. Reference to a joint
report in para (iv) above should not preclude the expert team providing a joint
report indicating that they cannot agree, as long as they give reasons for that. In
some cases the joint report might simply indicate what test they agree would be
appropriate to take the matter further and to finally resolve a key factual
question. For example, a case might concern the cause of a failure of a bridge.
The experts might propose a particular scientific test of a section of a steel beam
to resolve whether this was the likely cause.

As to (vii) the terms may indicate that if the parties do not believe certain
assistance would be appropriate, a procedural ruling would be obtained from
the tribunal. As to (viii), if the expert team believed that the briefs and
documents submitted were unnecessarily voluminous and in part irrelevant, a
ruling might also be sought from the tribunal. Mark Kantor also alludes to an
alternative where each of the two experts is allowed to work with a party with a
view to preparing competing reports if a joint report is not possible. (293)

12.15. Witness Costs and Expenses

The costs of each type of expert are treated differently. Tribunal experts are part
of tribunal costs. (294) Because each party is responsible for presenting its own
witnesses, this implies that each party will cover costs and expenses of its own
witnesses and then seek recovery under a costs order through the tribunal's
discretion. (295) Because witnesses of fact voluntarily agree to assist a party, a
question is then what entitlement they have to fees as opposed to
indemnification page "975" for expenses. (296) There is also a characterisation
question as to whether it is really a fee for the giving of evidence or instead an
indemnification for lost income from the time away from other activities. Where
there is an entitlement to fees, not only is there a problem as to the amount but
also as to whether the evidence might be tainted simply because of the
commercial relationship.

There is no consensus as to whether witnesses of fact are automatically entitled

to fees which can be sought from the losing party. Oetiker, for example, suggests
not, (297) but Gélinas argues in favour of such an entitlement. (298) The situation is
different for experts as their very profession is to provide expert opinions for a
fee. A party-appointed expert will receive a fee as agreed with the appointing
party. Where the costs award is concerned, a tribunal is not bound by the figure
agreed where it is considered to be excessive. Even where indemnification for
expenses is concerned, there are variations in hotels and airline status. There
may also be questions as to cancellation fees for both expert and general
witnesses.

As for tribunal-appointed experts, the tribunal, perhaps with the assistance of

an institution, will appoint the expert. Having agreed on the fee at the outset, the
tribunal is unlikely to review it in the context of a costs award, although it is
conceivable in an extreme case that it may believe that a tribunal-appointed
expert on an hourly rate has done excessive work and that the losing party
should not be responsible for all of this. Even in such a circumstance, the better
view would be a direct challenge to the expert's fee by way of breach of an
express or implied term of the contract. There would be issues as to whether a
tribunal-appointed expert is in a contractual relationship with the tribunal,
institution and/or the parties. In either event there is a question as to the
applicable law of the contractual agreement. (299)

page "976"

1 Laurent Lévy & Lucy Reed, ‘Managing Fact Evidence in International

Arbitration’, in International Arbitration 2006: Back to Basics?, ICCA Congress
Series No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer Law
International, 2007), 637–638.
2 UNCITRAL Arbitration Rules 2010 Art. 17.3; ICC Rules 2012 Art. 25; Swedish

Arbitration Act Art. 24(1); German Code of Civil Procedure Art. 1047; Belgian
Judicial Code Art. 1694(2) and (3); Dutch Code of Civil Procedure Art. 1039(2).
3 See, e.g., IBA Rules of Evidence 2010 Art. 4.7.
4 IBA Rules of Evidence 2010 Art. 4.8.
5 Netherlands Code of Civil Procedure Art. 1039(3).
6 Mauro Rubino-Sammartano, International Arbitration Law and Practice, 2nd

edn (The Hague: Kluwer Law International, 2001), 666.

7 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Thomson/Sweet & Maxwell, 2007), 472.

8 A decision by Professor Lalive as sole arbitrator in ICC Case No. 1512 to refuse

to hear oral evidence because he felt able to decide on the documents was
upheld by the Queen's Bench and appears to have been conceded by the
appellant in the Court of Appeal in Dalmia Dairy Industries Ltd (India) v. National
Bank of Pakistan [1978] 2 Lloyd's Rep 223.
9 Judgment of 7 January 2004, Swiss Federal Tribunal, ASA Bulletin 22, no. 3

(2004): 598. See also Cesare Jermini, ‘Note – Tribunal fédéral, Ire Cour civile, 7
Janvier 2004 (4P.196/ 2003), Witnesses and the Right to Be Heard in
International Arbitration: Some Remarks on Recent Decisions of the Swiss
Federal Court’, ASA Bulletin 22, no. 3 (2004): 605–609.
10 See, e.g., Dombo Beheer BV v. Netherlands, Application No. 14448/88, Case No.

37/1992/382/ 460, Judgment of 27 October 1993 (Merits and Just Satisfaction),

ECHR Series A, Volume 274; (1994) 18 EHRR 213, para. 35.
11 See section 9.2 as to documents-only arbitration.
12 Karl-Heinz Böckstiegel, ‘Presenting, Taking and Evaluating Evidence in
International Arbitration’, in AAA/ICDR Handbook on International Arbitration
and ADR, ed. Thomas E. Carbonneau & Jeanette A. Jaeggi (Huntington: Juris Net,
2006), 138–144.
13 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1843.

14 Section 7 of the US Federal Arbitration Act is an exception.
15 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1901.

16 David D. Caron, Lee M. Caplan & Matti Pellonpää, The UNCITRAL Arbitration

Rules: A Commentary (Oxford: Oxford University Press), 575. The current

provisions are Arts 27 and 28.
17 See Werner Melis, ‘Presentation of Documentary Evidence and Witnesses’, in

Planning Efficient Arbitration Proceedings/The Law Applicable in International

Arbitration, ICCA Congress Series No. 7, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 1996), 512–513.
18 George M. von Mehren & Claudia T. Salomon, ‘Submitting Evidence in an

International Arbitration: The Common Lawyer's Guide’, Journal of International

Arbitration 20, no. 3 (2003): 289–290.
19 Parties should not be presumed to have impliedly consented to a shorter time

period when they know leading arbitrators are working at a lower than normal
fee. Absent express agreement at the time of appointment to a shorter time
frame (which would be entirely acceptable), other reasonable expectations of
the parties seem more likely. Parties should reasonably expect that high quality
arbitrators will always give the appropriate attention to the particular case
based on the demands of legal and factual analysis that the case presents. They
would not presume that effort is dependent on the return for effort that the case
affords the arbitrator. Second, any logical argument to the effect that parties
could expect a shortened time frame when arbitrators receive below average
fees, would have to have a countervailing presumption that they will spend
longer than necessary in more lucrative cases. That is unlikely. Trying to identify
what the parties would reasonably expect should also involve looking
objectively at what an arbitrator could be seen to reasonably agree to when
taking a less lucrative case. The more likely presumption is that leading
arbitrators used to very different fee schedules understand that an acceptable
return for effort is based on an average return. This would entail a mix of ad hoc
arbitrations agreeing to their preferred daily rate, to institutional arbitrations
with a generous scale for large amounts involved that at times will apply to very
simple cases intellectually, through to very significant institutional fee
constraints such as through ICSID schedules. A less lucrative case might simply
have been accepted because there was a gap in the arbitrator's schedule that
would not otherwise be filled.
20 Werner Melis, ‘Presentation of Documentary Evidence and Witnesses’, in
Preventing Delay and Disruption of Arbitration/Effective Proceedings in
Construction Cases, ICCA Congress Series No. 5, ed. Albert Jan van den Berg
(Deventer: Kluwer Law and Taxation Publishers, 1991), 513.
21 Laurent Lévy & Lucy Reed, ‘Managing Fact Evidence in International
Arbitration’, in International Arbitration 2006: Back to Basics?, ICCA Congress
Series No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer Law
International, 2007), 642–643.
22 This is discussed further in sections 12.14.11-13 below.
23 See, e.g., HKIAC Arbitration Rules Art. 23.5. This was formerly the case under

the UNCITRAL Arbitration Rules 1976 Art. 25.2, but is now subject to the broad
discretion in Art. 28.2.
24 Michael Hwang & Andrew Chin, ‘The Role of Witness Statements in
International Commercial Arbitration’, in International Arbitration 2006: Back to
Basics?, ICCA Congress Series No. 13, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 2007), 656.
25 Military and paramilitary activities in and against Nicaragua (Nicaragua v.

USA), Merits, Judgment, ICJ Reports, 27 June 1986, para. 68.

26 Bernard Hanotiau, ‘The Conduct of the Hearings’, in The Leading Arbitrators'

Guide to International Arbitration, 2nd edn, ed. L. Newman & R. Hill (Huntington:
Juris Publishing, 2008), 372; Laurent Lévy & Lucy Reed, ‘Managing Fact
Evidence in International Arbitration’, in International Arbitration 2006: Back to
Basics?, ICCA Congress Series No. 13, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 2007), 636. Lévy and Reed suggest that it is relatively
standard that subject to tribunal discretion, no witness will be allowed to testify
if they have not submitted a written statement.
27 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1843.

28 See Yves Derains, ‘La pratique de l'administration de la preuve dans
l'arbitrage commercial international’ Revue de l'Arbitrage (2004): 796–797. See
also Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th
edn (Oxford: Oxford University Press, 2009), 403–404. Note that, although some
jurisdictions, such as Germany and Austria, do not permit parties to appear in
the capacity of a witness, they are, of course, still permitted to present their case.
See also Charles Brower, ‘Evidence before International Tribunals: The Need for
Some Standard Rules’, International Lawyer 28, no. 1 (1994): 50, n. 14. Brower
notes that the early common law also disqualified interested parties as
witnesses. Civilian systems have also changed significantly. See case of Dombo
Beheer BV v. The Netherlands [1993] ECHR 49.
29 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Thomson/Sweet & Maxwell, 2007), 557. See also
Klaus Peter Berger, International Economic Arbitration (The Hague: Kluwer
Academic Publishers, 1993), 441.
30 IBA Rules of Evidence 2010 Art. 4.2.
31 See, e.g., the comments in W Jack Buckamier v. Islamic Republic of Iran 28

Iran-US Claims Tribunal Reports 53, 74–76 (1992).

32 Michael Straus, ‘The Practice of the Iran-US Claims Tribunal in Receiving

Evidence from Parties and from Experts’, Journal of International Arbitration 3,

no. 3 (1986): 58–63; Charles Brower, ‘Evidence before International Tribunals:
The Need for Some Standard Rules’, International Lawyer 28, no. 1 (1994): 49.
33 David D. Caron, Lee M. Caplan & Matti Pellonpää, The UNCITRAL Arbitration
Rules: A Commentary (Oxford: Oxford University Press), 612–613.
34 Charles Brower, ‘Evidence before International Tribunals: The Need for Some

Standard Rules’, International Lawyer 28, no. 1 (1994): 50.

35 Ibid., 49. In ICC Case No. 7319, the sole arbitrator decided that a party,

including its legal representatives (in this case its directors), could not be heard
as a witness in the arbitration. The sole arbitrator referred in this context to Art.
14 of the then applicable 1988 ICC Rules of Arbitration, which distinguished
between hearing the ‘parties’ and hearing ‘any other person’. Statements made
by the directors would be treated merely as declarations of the party that they
represented. In contrast, an officer of the party – an employee, not its legal
representative – could be heard as a witness.
36 Charles Brower, ‘Evidence before International Tribunals: The Need for Some

Standard Rules’, International Lawyer 28, no. 1 (1994): 49, n. 11.

37 Edoardo F. Ricci, ‘Evidence in International Arbitration: A Synthetic Glimpse’,

in Liber Amicorum Bernado Cremades, ed. M.Á. Fernández-Ballesteros & David

Arias (Madrid: La Ley, 2010), 1027. Attitudes to written witness statements are
also impacted upon by a person's cultural attitude to the entitlement of counsel
to confer with prospective witnesses which is not the norm in civilian systems.
38 See IBA Rules of Evidence 2010 Art. 4.4 and 4.5. This is also dealt with under

ICDR Arbitration Rules 2009 Art. 20.5 which provides a discretion in that
regard.
39 IBA Rules of Evidence 2010 Art. 4.6.
40 Laurent Lévy, ‘Testimonies in the Contemporary Practice: Witness
Statements and Cross-Examination, in Arbitral Procedure at the Dawn of the New
Millennium: Reports of the International Colloquium of CEPANI, October 15, 2004,
ed. Stephen Bond et al. (Brussels: Bruylant, 2005), 118.
41 Charles Brower, ‘Evidence Before International Tribunals: The Need for Some

Standard Rules’, International Lawyer 28, no. 1 (1994): 51. Brower considers
that these advantages and disadvantages led to affidavits being seen as a
suitable and even preferred substitute to oral testimony in some cases before
the Iran-US Claims Tribunal.
42 Laurent Lévy, ‘Testimonies in the Contemporary Practice: Witness
Statements and Cross-Examination, in Arbitral Procedure at the Dawn of the New
Millennium: Reports of the International Colloquium of CEPANI, October 15, 2004,
ed. Stephen Bond et al. (Brussels: Bruylant, 2005), 120.
43 Ibid., 119. See also IBA Rules of Evidence 2010 Arts 4.7, 8.3.
44 This is supported by rules such as: ICC Rules 2012 Art. 25(2) and (3); LCIA

Arbitration Rules 1998 Art. 21.2.

45 IBA Rules of Evidence 2010 Art. 4.8.
46 The most difficult strategic challenge for counsel is where the witness asserts

facts in a very general manner, without sufficient particulars. Absent refutation,

a tribunal might accept the broad testimony. Conversely, calling for the witness
and exploring the particulars might lead to them bolstering an inadequate
written submission. Because counsel cannot know in advance the evidentiary
weight that the tribunal will give to an assertive statement, a tactical choice
needs to be made.
47 See the discussion in footnote 97 below as to the common law rule in Browne

v. Dunn.
48 Cited in Laurent Lévy, ‘Testimonies in the Contemporary Practice: Witness
Statements and Cross Examination’, in Arbitral Procedure at the Dawn of the New
Millennium, Report of the International Colloquium of CEPANI, 15 October 2004,
ed. Stephen Bond et al. (Brussels: Bruylant, 2005), 115.
49 For forceful criticisms of the abuse of written witness statements see V.V.

Veeder, ‘Introduction’ , in Arbitration and Oral Evidence: Dossier of the ICC

Institute of World Business Law, ed. Laurent Lévy & V.V. Veeder (Paris: ICC
Publishing, 2004), 7.
50 For a discussion of control over counsel and counsels' ethical standards see

section 9.7.
51 Michael Hwang & Andrew Chin,‘ The Role of Witness Statements in
International Commercial Arbitration’, in International Arbitration 2006: Back to
Basics?, ICCA Congress Series No. 13, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 2007), 658.
52 Ibid.
53 Gerald Aksen states that he advises counsel to do this in most of his
arbitrations. See Donald Donovan, ‘Act III: Advocacy with Witness Testimony’,
Arbitration International 21, no. 4 (2006): 588.
54 I am indebted to Justice Clyde Croft for this observation.
55 Michael Bühler & Carroll Dorgan, ‘Witness Testimony Pursuant to the 1999

IBA Rules of Evidence in International Commercial Arbitration: Novel or Tested

Standards?’, Journal of International Arbitration 17, no. 1 (2000): 14.
56 Laurent Lévy, ‘Testimonies in the Contemporary Practice: Witness
Statements and Cross Examination’, in Arbitral Procedure at the Dawn of the New
Millennium, Report of the International Colloquium of CEPANI, 15 October 2004,
ed. Stephen Bond et al. (Brussels: Bruylant, 2005), 107, 115.
57 Laurent Lévy, ‘Testimonies in the Contemporary Practice: Witness
Statements and Cross-Examination’, in Arbitral Procedure at the Dawn of the
New Millennium, Reports of the International Colloquium of CEPANI, October 15,
2004, ed. Stephen Bond et al. (Brussels: Bruylant, 2005), 114.
58 Karl-Heinz Böckstiegel, ‘Case Management by Arbitrators: Experiences and

Suggestions’, in Global Reflections on International Law, Commerce and Dispute

Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al.
(Paris: ICC Publishing, 2005), 124.
59 Michael Hwang & Andrew Chin, ‘The Role of Witness Statements in
International Commercial Arbitration’, in International Arbitration 2006: Back to
Basics?, ICCA Congress Series No. 13, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 2007), 653.
60 IBA Rules of Evidence 2010 Art. 4.5.
61 See section 12.3.4.
62 Laurent Lévy, ‘Testimonies in the Contemporary Practice: Witness
Statements and Cross-Examination’, in Arbitral Procedure at the Dawn of the
New Millennium, Reports of the International Colloquium of CEPANI, October 15,
2004, ed. Stephen Bond et al. (Brussels: Bruylant, 2005), 116. See also Laurent
Lévy & Lucy Reed, ‘Managing Fact Evidence in International Arbitration’, in
International Arbitration 2006: Back to Basics?, ICCA Congress Series No. 13, ed.
Albert Jan van den Berg (The Hague: Kluwer Law International, 2007), 636.
63 Michael Hwang & Andrew Chin, ‘The Role of Witness Statements in
International Commercial Arbitration’, in International Arbitration 2006: Back to
Basics?, ICCA Congress Series No. 13, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 2007), 658.
64 Ibid., 657–658.
65 1999 IBA Working Party & 2010 IBA Rules of Evidence Subcommittee,

‘Commentary on the revised text of the 2010 IBA Rules on the Taking of
Evidence in International Arbitration’, (2010), 17.
66 Hans van Houtte, ‘Counsel-Witness Relations and Professional Misconduct in

Civil Law Systems’, Arbitration International 19, no. 4 (2003): 461.

67 The UNCITRAL Notes on Organizing Arbitral Proceedings do not expressly

take a position on interviewing witnesses but instead suggest that the tribunal
should clarify the kinds of contacts the parties may be permitted to have. See
UNCITRAL, ‘UNCITRAL Notes on Organizing Arbitral Proceedings’, para. 67.
68 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 698, 701; Jean-François Poudret & Sébastien Besson, Comparative Law of
International Arbitration, 2nd edn (London: Thomson/Sweet & Maxwell, 2007),
559. See also Swiss Rules 2012 Art. 25.2. Some national Bars have moved to
confirm this position, including the Paris Bar. See Ordre des Avocats de Paris,
‘L'avocat et le témoin dans l'arbitrage international’, Le Bulletin du Barreau de
Paris 9 (2008): 45–46. This tended to be the historical position in civilian
jurisdictions although that is changing over time. See, e.g., Pascal Hollander &
Maartin Dreye, ‘Brussels Bar Lifts the Traditional Prohibition on Preparatory
Contacts between Attorneys and Witnesses’, ASA Bulletin 29, no. 2 (2011): 496–
500. See also Christian Dieryck, ‘Procédure et moyens de preuve dans l'arbitrage
commercial international’, Revue de l'Arbitrage (1988): 270–271.
69 IBA Rules of Evidence 2010 Art. 4.3. That view is also replicated in the Swiss

Rules 2012 Art. 25.2.

70 Otto L.O. de Witt Wijnen, ‘Collection of Evidence in International Arbitration’,

in Liber Amicorum Bernardo Cremades, ed. M.Á. Fernández & David Arias
(Madrid: La Ley, 2010), 358.
71 Hans van Houtte, ‘Counsel-Witness Relations and Professional Misconduct in

Civil Law Systems’, Arbitration International 19, no. 4 (2003): 462.

72 Ibid.
73 C. Mark Baker, ‘Advocacy in International Arbitration’, in The Leading
Arbitrators' Guide to International Arbitration, 2nd edn, ed. L. Newman & R. Hill
(Huntington: Juris Publishing, 2008), 395.
74 Lawrence W. Newman, ‘Cross Examination in International Arbitration –

Opportunities and Challenges’, in The Leading Arbitrators' Guide to International

Arbitration, 2nd edn, ed. L. Newman & R. Hill (Huntington: Juris Publishing,
2008), 412.
75 Michael Hwang & Andrew Chin, ‘The Role of Witness Statements in
International Commercial Arbitration’, in International Arbitration 2006: Back to
Basics?, ICCA Congress Series No. 13, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 2007), 658.
76 Lawrence W. Newman, ‘Cross Examination in International Arbitration –

Opportunities and Challenges’, in The Leading Arbitrators' Guide to International

Arbitration, 2nd edn, ed. L. Newman & R. Hill (Huntington: Juris Publishing,
2008), 412.
77 Paulo Michele Patocchi & Harold Frey Brentano, ‘The Provisional Timetable
in International Arbitration’, in Global Reflections on International Law,
Commerce and Dispute Resolution: Liber Amicorum in Honour of Robert Briner,
ed. Gerald Aksen et al. (Paris: ICC Publishing, 2005), 63.
78 For contrary suggestions see Hans van Houtte, ‘Counsel-Witness Relations

and Professional Misconduct in Civil Law Systems’, Arbitration International 19,

no. 4 (2003): 457–463.
79 See generally Catherine Rogers, ‘The Ethics of Advocacy’, in The Art of

Advocacy in International Arbitration, 2nd edn, ed. Doak Bishop & Edward G.
Kehoe (Huntington: JurisNet LLC, 2010).
80 William W. Park, ‘Arbitration's Protean Nature: The Value of Rules and the

Risks of Discretion’, Mealey's International Arbitration Report 19, no. 5 (2004):

20, n. 74.
81 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1827. Born notes, however, that the practice in this regard
is not uniform. This was also considered in section 11.5 above as to duties of
disclosure generally and section 11.7.2 where consideration was given as to
whether a request for production of documents to be relied on in cross-
examination, is sufficiently specific.
82 But see Procedural Order in ICC Case No. 7170 in Collection of Procedural

Decisions in ICC Arbitration 1993–1996 (1997), ed. Dominique Hascher (Paris:

ICC Publishing, 1997), 55.
83 International Centre for Dispute Resolution, ‘ICDR Guidelines for Arbitrators

Concerning Exchanges of Information’, para. 6(b).

84 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1904–1905.

85 Siegfried H. Elsing & John M. Townsend, ‘Bridging the Common Law-Civil Law

Divide in Arbitration’, Arbitration International 18, no. 1 (2002): 61.

86 Charles N. Brower, ‘Evidence before International Tribunals: The Need for

Some Standard Rules’, International Lawyer 28, no. 1 (1994): 51–52.

87 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1844.

88 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1845.

89 Ibid.
90 Karl-Heinz Böckstiegel, ‘Case Management by Arbitrators: Experiences and

Suggestions’, in Global Reflections on International Law, Commerce and Dispute

Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al.
(Paris: ICC Publishing, 2005), 123.
91 The Working Party which prepared the 1999 IBA Rules considered that

witness statements may be evidence in chief so as to allow examination by the

other party to start immediately. IBA Working Party, ‘Commentary on the New
IBA Rules of Evidence in International Commercial Arbitration’, Business Law
International 2 (2000), 27.
92 See Siegfried H. Elsing & John M. Townsend, ‘Bridging the Common Law-Civil

Law Divide in Arbitration’, Arbitration International 18, no. 1 (2002): 62–64.

93 R. Doak Bishop, ‘Advocacy in International Commercial Arbitration: United

States’, in The Art of Advocacy in International Arbitration, ed. R. Doak Bishop

(Huntington: JurisNet, 2004), 344.
94 Michael Hwang & Andrew Chin, ‘The Role of Witness Statements in
International Commercial Arbitration’, in International Arbitration 2006: Back to
Basics?, ICCA Congress Series No. 13, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 2007), 656.
95 E. David & B. Tavender, ‘To Restore Speed and Cost Effectiveness, Arbitrators

Must Focus on Time Allocation’, Alternatives to the High Cost of Litigation 28, no.
8 (2010): 163.
96 A challenge on that basis was rejected in Generica Ltd v. Pharmaceutical

Basics, Inc., 125 F. 3d 1123 (7th Cir 1997).

97 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 420.

98 See V.V. Veeder, ‘The Lawyer's Duty to Arbitrate in Good Faith’, Arbitration

International 18, no. 4 (2002): 445.

99 English common lawyers are also used to applying the rule in Browne v. Dunn

(1984) 6 R 67 that provides that unless sufficient prior notice is given of a

contention being relied upon, the contention must be put to an opposing witness
who is asserting a contrary position in cross-examination so they have a
meaningful opportunity to respond. The logic is to allow the witness to respond
to the case against his or her testimony. The rule also leads to the corollary that
if the contention is put and the witness does not refute the contention, it is
accepted as made out unless otherwise contradicted. The rule itself does not
apply in arbitration but the obligation in arbitration to present arguments and
documents before a hearing satisfies the same policy aim. If matters had not
been properly notified in arbitration there is a problem either way, undue
surprise if asked and lack of opportunity if not.
100 Section 6.2.4 dealt with the general right of a tribunal to override a
procedural agreement of the parties.
101 See Hilmar Raeschke-Kessler, ‘Witness Conferencing’, in The Leading
Arbitrators' Guide to International Arbitration, 2nd edn, ed. L. Newman & R. Hill
(Huntington: Juris Publishing, 2008), 418. See also the discussion of tandem
witness examination and a draft protocol discussed in Mauro Rubino-
Sammartano, International Arbitration Law and Practice, 2nd edn (The Hague:
Kluwer Law International), 690–691.
102 Wolfgang Peter, ‘Witness “Conferencing”’, Arbitration International 18, no. 1

(2002): 56; Clifford J. Hendel, ‘Witness Conferencing Involving Witnesses of

Fact: A Good Idea Whose Time Has Not Yet Come?’, IBA Arbitration News 16, no.
1 (2011): 88.
103 Doug Jones, ‘Party Appointed Expert Witnesses in International Arbitration:

A Protocol at Last’, Arbitration International 24, no. 1 (2008): 149.

104 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th

edn (Oxford: Oxford University Press, 2009), 423.

105 See, e.g., English Arbitration Act 1996 s. 38(5). The provision does not

indicate the form of any oath or affirmation. The power is also expressly
referred in the Belgian Judicial Code Art. 1696(1) and the Netherlands Code of
Civil Procedure Art. 1041(1). See also Israeli Arbitration Law Art. 14.
106 Swiss law does not allow a private person to administer an oath. Redfern

and Hunter describe such a provision as mandatory. Nigel Blackaby et al.,

Redfern and Hunter on International Arbitration, 5th edn (Oxford: Oxford
University Press, 2009), 317. Likewise, the German Code of Civil Procedure does
not allow arbitral tribunals to administer oaths, although its express prohibition
has been removed from the Code. See Klaus Sachs & Torsten Lörcher, ‘Conduct
of Arbitral Proceedings’, in Arbitration in Germany: The Model Law in Practice,
ed. Karl-Heinz Böckstiegel, Stefan Michael Kröll & Patricia Nacimiento (Alphen
aan den Rijn: Kluwer Law International, 2007), 325. Swedish and Finnish law
also specifically prohibit arbitrators from administering oaths. See Swedish
Arbitration Act s. 25(3); Finnish Arbitration Act s. 27(2). See also French New
Code of Civil Procedure Art. 1467.
107 Roberto Ceccon, ‘UNCITRAL Notes on Organizing Arbitral Proceedings and

the Conduct of Evidence: A New Approach to International Arbitration’, Journal

of International Arbitration 14, no. 2 (1997): 78.
108 German Code of Civil Procedure Art. 1050; Swiss Law on Private
International Law Art. 184(2); Austrian Code of Civil Procedure Arts 306–309;
UNCITRAL Model Law Art. 27. Jurisdictions vary as to whether criminal
sanctions apply to false testimony not made under oath.
109 English Arbitration Act 1996 s. 38(5).
110 ICSID Arbitration Rules Art. 35(3).
111 IBA Rules of Evidence 2010 Art. 8.4.
112 Bernard G. Poznanski, ‘The Nature and Extent of an Arbitrator's Powers in

International Commercial Arbitration’, Journal of International Arbitration 4, no.

3 (1987): 105.
113 Axel H. Baum, ‘International Arbitration: The Path toward Uniform
Procedures’, in Global Reflections on International Law, Commerce and Dispute
Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al.
(Paris: ICC Publishing, 2005), 54. Common law jurisdictions were more inclined
to use verbatim transcripts of oral proceedings.
114 Charles N. Brower, ‘Evidence before International Tribunals: The Need for

Some Standard Rules’, International Lawyer 28, no. 1 (1994): 51.

115 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th

edn (Oxford: Oxford University Press, 2009), 421.

116 Judgment of 16 November 1993, Société Ganz Mozdony et al. v. SNCFT, Cour

d'Appel de Paris (Paris Court of Appeal) in Revue de l'Arbitrage 1995, no. 3: 477.
Even if successful, the party providing such a transcript would be unlikely to be
compensated for the expense in the costs order.
117 I am indebted to Pierre Karrer for this observation.
118 See section 6.16 in relation to on-line arbitration generally.
119 IBA Rules of Evidence 2010 Art. 8.1.
120 1999 IBA Working Party & 2010 IBA Rules of Evidence Subcommittee,

‘Commentary on the revised text of the 2010 IBA Rules on the Taking of
Evidence in International Arbitration’ (2010), 18.
121 I am indebted to Tony Canham for this observation.
122 Charles N. Brower, ‘Evidence Before International Tribunals: The Need for

Some Standard Rules’, International Lawyer 28, no. 1 (1994): 53.

123 Michael W. Bühler & Thomas H. Webster, Handbook of ICC Arbitration:

Commentary, Precedents, Materials (London: Sweet & Maxwell, 2005), 276.

124 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1845, n. 545.

125 Michael Hwang & Andrew Chin, ‘The Role of Witness Statements in
International Commercial Arbitration’, in International Arbitration 2006: Back to
Basics?, ICCA Congress Series No. 13, ed. Albert Jan van den Berg (The Hague:
Kluwer Law International, 2007), 656.
126 Ibid., 652.
127 Ibid., 654.
128 Laurent Lévy, ‘Witness Statements’, in De Lege Ferenda - Réflexions sur le

droit désirable en l'honneur du Professeur Alain Hirsch 95, ed. A. Hériter Lachat &
L. Hirsch (Geneva: Slatkine, 2004), 95, 101.
129 While it is more natural for a civilian lawyer to have the adjudicator direct

questions, common law judges will also intervene on matters of importance.

130 Cited in Laurent Lévy, ‘Testimonies in the Contemporary Practice: Witness

Statements and Cross Examination’, in Arbitral Procedure at the Dawn of the New
Millennium, Report of the International Colloquium of CEPANI, 15 October 2004,
ed. Stephen Bond et al. (Brussels: Bruylant, 2005), 115.
131 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th

edn (Oxford: Oxford University Press, 2009), 420.

132 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1846.

133 These arguments also relate to the advantages and disadvantages of witness

conferencing, discussed below.

134 See, e.g., Swiss Rules 2012 Art. 25.6. Some rules provide an express
discretion to the tribunal.
135 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 700.
136 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1847.

137 UNCITRAL Model Law Art. 27; US Federal Arbitration Act § 7; English

Arbitration Act 1996 s. 43; Swiss Private International Law Art. 184(2); Belgian
Judicial Code Art. 1496(4).
138 Martin H. Hunter, ‘The Procedural Powers of Arbitrators under the English

Act’, Arbitration International 13, no. 4 (1997): 346.

139 IBA Rules of Evidence 2010 Art. 4.9.
140 Pierre A. Karrer, ‘Freedom of an Arbitral Tribunal to Conduct Proceedings’,

ICC International Court of Arbitration Bulletin 10, no. 1 (1999): 24. See also
Laurent Lévy, ‘Testimonies in the Contemporary Practice: Witness Statements
and Cross-Examination’, in Arbitral Procedure at the Dawn of the New
Millennium, Reports of the International Colloquium of CEPANI, October 15,
2004, ed. Stephen Bond et al. (Brussels: Bruylant, 2005), 122.
141 Marianne Roth, ‘False Testimony at International Arbitration Hearings
Conducted in England and Switzerland: A Comparative View’, Journal of
International Arbitration 11, no. 1 (1994): 16–19; Mauro Rubino-Sammartano,
International Arbitration Law and Practice, 2nd edn (The Hague: Kluwer Law
International), 689.
142 See section 11.8.1 These may be in a special category as they are not experts

as to an issue in dispute but experts assisting in the arbitral procedure. Some

express rules as to expert witnesses may not therefore apply, although the
position still has some uncertainty.
143 Yves Derains & Eric A. Schwartz, A Guide to the New ICC Rules of Arbitration

(The Hague: Kluwer Law International), 258.

144 The use of experts in arbitration proceedings is also quite different to the
alternative dispute resolution technique of expert determination. Dana H.
Freyer, ‘Assessing Expert Evidence’, in The Leading Arbitrators' Guide to
International Arbitration, 2nd edn, ed. L. Newman & R. Hill (Huntington: Juris
Pub, 2008), 437. This is also described as ‘expertise arbitrale’,
‘schiedsgutachten’ and valuation; Jean-François Poudret & Sébastien Besson,
Comparative law of International Arbitration, 2nd edn (London: Thomson/Sweet
& Maxwell, 2007), 562.
145 Yves Derains, ‘La pratique de l'administration de la preuve dans l'arbitrage

commercial international’, Revue de l'Arbitrage (2004): 800.

146 See, e.g., UK Civil Procedure Rules, Practice Direction 35 – Experts and

Assessors, but note that the US allows for more in the way of promotion of party
interests.
147 See the discussion in Mark Kantor, ‘A Code of Conduct for Party-Appointed

Experts in International Arbitration - Can One Be Found?’, Arbitration

International 26, no. 3 (2010): 324–325.
148 Ibid., 326–327.
149 Klaus Sachs, ‘Experts: Neutrals or Advocates’, paper given at the ICCA Annual

Conference 2010 (Rio de Janeiro, 23–26 May 2010), para. 8. Klaus Sachs & Nils
Schmidt-Ahrendts, ‘Protocol on Expert Teaming: A New Approach to Expert
Evidence’, in Arbitration Advocacy in Changing Times, ICCA Congress Series No.
15, ed. Albert Jan van den Berg (The Hague: Kluwer Law International, 2011),
135.
150 Mark Kantor, ‘A Code of Conduct for Party-Appointed Experts in
International Arbitration – Can One Be Found?’, Arbitration International 26, no.
3 (2010): 336.
151 Ibid., 364.
152 Arthur Marriott, ‘Arbitrators and Settlement’, in New Horizons in
International Commercial Arbitration and Beyond, ICCA Congress Series No. 12,
ed. Albert Jan van den Berg (The Hague: Kluwer Law International, 2005), 534.
153 Klaus Sachs & Nils Schmidt-Ahrendts, ‘Protocol on Expert Teaming: A New

Approach to Expert Evidence’, in Arbitration Advocacy in Changing Times, ICCA

Congress Series No. 15, ed. Albert Jan van den Berg (The Hague: Kluwer Law
International, 2011), 139.
154 Klaus Sachs, ‘Experts: Neutrals or Advocates’, paper given at the ICCA Annual

Conference 2010 (Rio de Janeiro, 23–26 May 2010), paras 21–23.

155 Doug Jones, ‘Party Appointed Expert Witnesses in International Arbitration:

A Protocol at Last’, Arbitration International 24, no. 1 (2008): 154.

156 See D. Juan & R. Gutiérrez, ‘Expert Testimony, Economic Evidence and

Asymmetry of Information in Antitrust Cases’, CEDEC Competition Law &

Economics Working Paper No. 07-04 (2007).
157 Mark Kantor, ‘A Code of Conduct for Party-Appointed Experts in
International Arbitration – Can One Be Found?’, Arbitration International 26, no.
3 (2010): 335.
158 Professor Doug Jones, ‘International Dispute Resolution in the Global
Financial Crisis’, The Arbitrator and Mediator (October 2009): 45, 46.
159 Klaus Sachs, ‘Experts: Neutrals or Advocates’, paper given at the ICCA Annual

Conference 2010 (Rio de Janeiro, 23–26 May 2010), paras 25–29.

160 Noel G. Bunni, ‘Some Thoughts from Experiences in Construction
Arbitration’, in International Arbitration 2006: Back to Basics?, ICCA Congress
Series No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer Law
International, 2007), 794.
161 The American common law approach still remains somewhat distinct. See

Julian D.M. Lew QC, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 576. For a comprehensive analysis of the ethical basis of party-appointed
expert conduct see Mark Kantor, ‘A Code of Conduct for Party-Appointed
Experts in International Arbitration: Can One Be Found?’, Arbitration
International 26, no. 3 (2010): 323.
162 See, e.g., Austrian Code of Civil Procedure Art. 601(2) and (4).
163 Klaus Sachs & Nils Schmidt-Ahrendts, ‘Protocol on Expert Teaming: A New

Approach to Expert Evidence’, in Arbitration Advocacy in Changing Times, ICCA

Congress Series No. 15, ed. Albert Jan van den Berg (The Hague: Kluwer Law
International, 2011), 137.
164 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Thomson/Sweet & Maxwell, 2007), 561.

165 Doug Jones, ‘Party Appointed Expert Witnesses in International Arbitration:

A Protocol at Last’, Arbitration International 24, no. 1 (2008): 150.

166 UK Civil Justice Council, ‘Protocol for the Instruction of Experts to give

Evidence in Civil Claims’, (2009), para. 17.15.

167 UNCITRAL Model Law Art. 26; English Arbitration Act 1996 s. 37; Swedish

Arbitration Act s. 25; Netherlands Code of Civil Procedure Art. 1042.

168 Klaus Sachs & Nils Schmidt-Ahrendts, ‘Protocol on Expert Teaming: A New

Approach to Expert Evidence’, in Arbitration Advocacy in Changing Times, ICCA

Congress Series No. 15, ed. Albert Jan van den Berg (The Hague: Kluwer Law
International, 2011), 137, referring to the laws of France, Switzerland and the
US where there are no specific provisions for expert evidence.
169 Tribunal-appointed experts are referred to in: UNCITRAL Arbitration Rules

2010 Art. 29 and LCIA Arbitration Rules 1998 Art. 21. For other rules dealing
with tribunal appointed experts, see: ICC Rules of Arbitration 2012 Art 25; ICDR
Rules Art. 22; Swiss Rules 2012 Art. 27; HKIAC Rules Art. 25; SIAC Rules Art. 23;
SCC Rules Art. 29.
170 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 704. However, the authors note that Swiss law would allow a limited
ground of challenge.
171 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th

edn (Oxford: Oxford University Press, 2009), 407.

172 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1861.

173 IBA Rules of Evidence 2010 Art. 6.1.
174 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1861.

175 Decision 102LA493, unpublished Note 8; Decision 4T.23/1991 of 25 May

1992, Note 5B cited in Georg von Segesser & Petra Rihar, ‘The Right to a
Tribunal Appointed Expert’,
<http:////kluwerarbitrationblog.com/blog/2011/08/09...>, 9 August 2011.
176 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th

edn (Oxford: Oxford University Press, 2009), 407.

177 Jean-François Poudret & Sébastien Besson, Comparative Law of International
Arbitration, 2nd edn (London: Thomson/Sweet & Maxwell, 2007), 560.
178 Claus von Wobeser, ‘The Arbitral Tribunal-Appointed Expert’, in
International Arbitration 2006: Back to Basics?, ICCA Congress Series No. 13, ed.
Albert Jan van den Berg (The Hague: Kluwer Law International, 2007), 807.
179 Yves Derains & Eric A. Schwartz, A Guide to the New ICC Rules of Arbitration

(The Hague: Kluwer Law International, 2009), 278–279.

180 Gillian M. White, The Use of Experts by International Tribunals (Syracuse:

Syracuse University Press, 1965), 73.

181 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Thomson/Sweet & Maxwell, 2007), 562.

182 Choosing a candidate ranked more highly by one of the parties would not

offend equal treatment obligations, at least where the other party considered
the candidate suitable as well, albeit less so.
183 Claus von Wobeser, ‘The Arbitral Tribunal-Appointed Expert’, in
International Arbitration 2006: Back to Basics?, ICCA Congress Series No. 13, ed.
Albert Jan van den Berg (The Hague: Kluwer Law International, 2007), 807–808.
184 Kap-you (Kevin) Kim & John P. Bang, ‘Commentary on Using Legal Experts in

International Arbitration’, in International Arbitration 2006: Back to Basics?,

ICCA Congress Series No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer
Law International, 2007), 783.
185 A challenge on this basis was unsuccessful before the Swiss Federal Supreme

Court in part because of lack of evidence of bias and because of the delay in
bringing the challenge. See Christoph Brunner, ‘Note: Federal Supreme Court, 28
April 2000: Procedural Public Policy as a Ground for Setting Aside International
Arbitral Awards’, ASA Bulletin 18, no. 3 (2000): 566.
186 While not directly relevant to international arbitration, the Rules of Conduct

for WTO Dispute Settlement Art. II.1 are a useful guide to the way experts ought
to behave.
187 IBA Rules of Evidence 2010 Art. 6.2. Express provision is made for
challenging tribunal appointed experts in German Code of Civil Procedure Art.
1049(3).
188 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 704.
189 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Thomson/Sweet & Maxwell, 2007), 563.

190 Mark Kantor citing the views of Professor Thomas Walde. Mark Kantor, ‘A

Code of Conduct for Party-Appointed Experts in International Arbitration – Can

One Be Found?’, Arbitration International 26, no. 3 (2010): 334.
191 Julian D.M. Lew QC, Loukas A. Mistelis & Stefan M. Kröll, Comparative

International Commercial Arbitration (The Hague: Kluwer Law International,

2003), 575.
192 Mark Kantor, ‘A Code of Conduct for Party-Appointed Experts in
International Arbitration – Can One Be Found?’, Arbitration International 26, no.
3 (2010): 329.
193 Chartered Institute of Arbitrators, ‘Protocol for the Use of Party-Appointed

Expert Witnesses in International Arbitration’, Art. 4(1).

194 Kap-you (Kevin) Kim & John P. Bang, ‘Commentary on Using Legal Experts in
International Arbitration’, in International Arbitration 2006: Back to Basics?,
ICCA Congress Series No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer
Law International, 2007), 784.
195 Mark Kantor, ‘A Code of Conduct for Party-Appointed Experts in
International Arbitration – Can One Be Found?’, Arbitration International 26, no.
3 (2010): 335.
196 Chartered Institute of Arbitrators, ‘Protocol for the Use of Party-Appointed

Expert Witnesses in International Arbitration’, Art 8. This Article also requires a

declaration of independence. If the tribunal subsequently determines that the
declaration is incorrect, it may disregard the written opinion and testimony
unless in exceptional circumstances it determines otherwise.
197 Article 5.2(c).
198 Article 5.2(a) IBA Rules of Evidence 2010.
199 See ICC, ‘Rules for Expertise’
<http://www.iccwbo.org/court/expertise/id4379/index...>, Art. 3.3. See also
Jean-François Bourque, ‘L'expérience du centre international d'expertise de la
CCI et le développement de l'expertise internationale’, Revue de l'Arbitrage
(1995): 252.
200 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 704.
201 Dana Freyer, ‘Assessing Expert Evidence’, in The Leading Arbitrators' Guide

to International Arbitration, 2nd edn, ed. L. Newman &and R. Hill (Huntington:

Juris Publishing, 2008), 442.
202 1999 IBA Working Party & 2010 IBA Rules of Evidence Subcommittee,

‘Commentary on the revised text of the 2010 IBA Rules on the Taking of
Evidence in International Arbitration’ (2010), 20–21.
203 The difference between the policy models is between establishing a duty of

an expert and relying instead on a court power.

204 Mark Kantor, ‘A Code of Conduct for Party-Appointed Experts in
International Arbitration – Can One Be Found?’, Arbitration International 26, no.
3 (2010): 339.
205 Tony Canham, ‘The Expert Witness in Arbitration and ADR: A Practical

Approach’, The Academy of Experts 27, at 28 available at

<http://www.tonycanham.com>. See also the recent UK Supreme Court decision
holding that there should not be immunity for expert witnesses from being sued
for negligence in Jones v. Kaney [2011] UKSC 13.
206 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th

edn (Oxford: Oxford University Press, 2009), 423. The situation is different with
an expert who is merely part of a legal team helping to prepare the case, with no
intention of giving evidence. Care needs to be taken, however, if such a person is
ultimately selected to give evidence if earlier behaviour has compromised their
independence.
207 Chartered Institute of Arbitrators, ‘Protocol for the Use of Party-Appointed

Expert Witnesses in International Arbitration’, Art. 4.1.

208 IBA Rules of Evidence 2010 Arts 5.2(g), 6.4(e) and 8.4.
209 Cited in Mark Kantor, ‘A Code of Conduct for Party-Appointed Experts in

International Arbitration – Can One Be Found?’, Arbitration International 26, no.

3 (2010): 326, n. 10, who notes that the rules were not formally adopted by ALI
or UNIDROIT but are recommendations of the Reporters.
210 Ibid., 334.
211 Ibid., 375. Kantor also thoroughly analyses the ethical standards of various

professional bodies that may also impact upon the behaviour of experts in
individual cases.
212 Noel G. Bunni, ‘Some Thoughts from Experiences in Construction
Arbitration’, in International Arbitration 2006: Back to Basics?, ICCA Congress
Series No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer Law
International, 2007), 790.
213 See Case No. 24, Award No 314-21-1 of 14 August 1987, Starrett Housing

Corporation v. The Government of the Islamic Republic of Iran, 16 Iran-US Claims

Tribunal Reports 112, 196. In that case, the Tribunal spoke of the use of expert
assistance to ‘guide (the tribunal) in the search for the truth’. It also stated that
‘(i)t is fundamental that an arbitral tribunal cannot delegate to (the expert) the
duty of deciding the case’. See also Corfu Channel (United Kingdom v. Albania)
[1949] ICJ 4, 196. A challenge on the basis that a tribunal expert was too
engaged in the assessment of evidence and did not provide the report to the
parties was rejected in the Singapore High Court in Luzon Hydro Corp v.
Transfield Philippines [2004] 4 SLR 705.
214 Case 181, Award No. 294-181-1, Bechtel Inc et al. v. Islamic Republic of Iran,

14 Iran-US Cl Trib Rep 149 (1987). Such persons are a distinct category and may
not readily be subject to the typical rules applicable to tribunal appointed
experts as they are dealing with a procedural matter and not an ultimate issue in
dispute.
215 Alan Redfern, ‘Experts and Arbitrators – An International Perspective’,
International Arbitration Law Review 4, no. 4 (2001): 106.
216 See, e.g., Longley (James) & Co. Ltd v. South West Thames Regional Health

Authority [1983] 25 BLR 56.

217 Luzon Hydro Corp. v. Transfield Philippines Inc. [2004] SGHC 204 at para. 16.
218 Klaus Sachs, ‘Experts: Neutrals or Advocates’, paper given at the ICCA Annual

Conference 2010 (Rio de Janeiro, 23–26 May 2010), para. 30.

219 Luzon Hydro Corp. v. Transfield Philippines Inc [2004] SGHC 204 at para. 16.
220 Hilmar Raeschke-Kessler, ‘Witness Conferencing’, in The Leading Arbitrators'

Guide to International Arbitration, 2nd edn, ed. L. Newman & R. Hill (Huntington:
Juris Publishing, 2008), 426. One ICC panel appears to have sought to do this,
although one party was very reluctant. See Final Award in ICC Case No. 6497, in
Yearbook of Commercial Arbitration Volume XXIVa, ed. Albert Jan van den Berg
(The Hague: Kluwer Law International, 1999), 77–78.
221 In Tang Ping-Choi and Anor v. Secretary for Transport [2004] 2 HKLRD 284,

an expert's testimony was given no weight, because the expert had obtained the
evidence by secretly recording a conversation. See Doug Jones, ‘Party Appointed
Expert Witnesses in International Arbitration: A Protocol at Last’, Arbitration
International 24, no. 1 (2008): 152.
222 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 705.
223 Entitlement to receive information, attend inspections and examine
documents used by the expert are covered in the IBA Rules of Evidence 2010
Art. 6.3, 6.4 and 6.5.
224 Judgment of 6 February 1997, Carter v. Alsthom, Cour d'appel de Paris, Revue
de l'Arbitrage (1997), no. 4: 556–557.
225 Compare Pierre Mayer, ‘Note – Cour d'appel de Paris’ (1re Ch. C) 6 février

1997), Revue de l'Arbitrage (1997), no. 4: 562, Jean-François Poudret &

Sébastien Besson, Comparative Law of International Arbitration, 2nd edn
(London: Thomson/Sweet & Maxwell, 2007), 479 and Emmanuel Gaillard &
John Savage (eds), Fouchard, Gaillard, Goldman on International Commercial
Arbitration (The Hague: Kluwer Law International, 1999), 704.
226 Mantovanelli v. France, Application No. 21497/93, paras 33–35. See also

Case C-276/01, Joachim Steffensen [2003] ECR I-3735, paras 77–78.

227 Kap-you (Kevin) Kim & John P. Bang, ‘Commentary on Using Legal Experts in

International Arbitration’, in International Arbitration 2006: Back to Basics?,

ICCA Congress Series No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer
Law International, 2007), 780.
228 Ibid., 782–783.
229 ICC, ‘Techniques for Controlling Time and Costs in Arbitration’,
<http://www.iccwbo.org/uploadedFiles/TimeCost_E.pdf...>, 2007, para. 68.
230 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 704.
231 The Iran-US Claims Tribunal at times invited input from the parties in

relation to the expert's terms of reference, called for a preliminary report by the
expert, allowed the parties to comment and directed the expert to take these
comments into account in the final report. See Nigel Blackaby et al., Redfern and
Hunter on International Arbitration, 5th edn (Oxford: Oxford University Press,
2009), 408 referring to Starrett Housing Corp. v. The Government of the Islamic
Republic of Iran (1987) 16 Iran-US Claims Tribunal Reports 117–119.
232 Claus von Wobeser, ‘The Arbitral Tribunal-Appointed Expert’, in
International Arbitration 2006: Back to Basics?, ICCA Congress Series No. 13, ed.
Albert Jan van den Berg (The Hague: Kluwer Law International, 2007), 808.
233 Ibid., 810.
234 This will allow parties to decide which of a group of co-signatories may be

needed to give evidence.

235 UK Civil Justice Council, ‘Protocol for the Instruction of Experts to give

Evidence in Civil Claims’, (2009), para. 4.3.

236 Ibid., para. 15.2.
237 Detlev Kühner, ‘The Revised IBA Rules on the Taking of Evidence in
International Arbitration’, Journal of International Arbitration 27, no. 6 (2010):
673.
238 IBA Rules of Evidence 2010 Art. 6.2.
239 Chartered Institute of Arbitrators, ‘Protocol for the Use of Party-Appointed

Expert Witnesses in International Arbitration’, Art. 4.4.

240 Ibid., Art. 8.
241 See, e.g., Guidelines for Expert Witnesses in Proceedings in the Federal Court

of Australia, Practice Direction May 2008; Civil Justice Counsel, Expert Witness
Protocol; UK Civil Procedure Rules – CPR 35.12; The Academy of Experts Code of
Practice for Experts, 22 June 2005; ICC Rules for Expertise (2003).
242 To similar effect Bunni recommended a draft expert's declaration covering

duty to the tribunal; indication of facts within own knowledge and belief in
truth; identification of adverse issues and qualifications; identification of
sources; attestation of an independent view; confirmation that advice will be
provided if views change; acknowledgement that the report will form the
evidence to be given; acknowledgement that cross-examination may ensue; and
acknowledgement that there may be public adverse criticism if reasonable care
has not been taken. Noel G. Bunni, ‘Some Thoughts from Experiences in
Construction Arbitration’, in International Arbitration 2006: Back to Basics?,
ICCA Congress Series No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer
Law International, 2007), 792.
243 Hans van Houtte, ‘The Document Production Master and the Experts'
Facilitator: Two Possible Aides for an Efficient Arbitration’, in Liber Amicorum
Bernardo Cremades, ed. M. Á Fernández-Ballesteros & David Arias (Madrid: La
Ley, 2010), 1154.
244 Doug Jones, ‘Party Appointed Expert Witnesses in International Arbitration:

A Protocol at Last’ Arbitration International 24, no. 1 (2008): 143.

245 Kap-you (Kevin) Kim & John P. Bang, ‘Commentary on Using Legal Experts in

International Arbitration’, in International Arbitration 2006: Back to Basics?,

ICCA Congress Series No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer
Law International, 2007), 784–785.
246 Klaus Sachs, ‘Experts: Neutrals or Advocates’, paper given at the ICCA Annual

Conference 2010 (Rio de Janeiro, 23–26 May 2010), para. 47.

247 ICC, ‘Techniques for Controlling Time and Costs in Arbitration’,
<http://www.iccwbo.org/uploadedFiles/TimeCost_E.pdf>, 2007, para. 82.
248 As per UK Civil Procedure Rules 35.12(5).
249 Tony Canham, ‘The Expert Witness in Arbitration and ADR: A Practical

Approach’, The Academy of Experts 27 available at

<http://www.tonycanham.com>.
250 Ibid.
251 Noel G. Bunni, ‘Some Thoughts from Experiences in Construction
Arbitration’, in International Arbitration 2006: Back to Basics?, ICCA Congress
Series No. 13, ed. Albert Jan van den Berg (The Hague: Kluwer Law
International, 2007), 791.
252 It would be difficult for a tribunal to make distinct cost orders about

allegedly inappropriate behaviour in pre-hearing meetings where there is no

accurate record of what occurred.
253 Doug Jones, ‘Party Appointed Expert Witnesses in International Arbitration:

A Protocol at Last’, Arbitration International 24, no. 1 (2008): 144.

254 If such a process is being managed by a tribunal appointed expert, there

might be less scope for such parochial behaviour. This is discussed below in the
following section dealing with expert facilitation.
255 Hans van Houtte, ‘The Document Production Master and the Experts'
Facilitator: Two Possible Aides for an Efficient Arbitration’, in Liber Amicorum
Bernado Cremades, ed. M.Á. Fernández-Ballesteros & David Arias (Madrid: La
Ley, 2010), 1155.
256 Doug Jones, ‘Party Appointed Expert Witnesses in International Arbitration:

A Protocol at Last’, Arbitration International 24, no. 1 (2008): 144.

257 Jane Jenkins & Simon Stebbings International Construction Arbitration Law

(The Hague: Kluwer Law International, 2006), 213.

258 Hans van Houtte, ‘The Document Production Master and the Experts'
Facilitator: Two Possible Aides for an Efficient Arbitration’, in Liber Amicorum
Bernado Cremades, ed. M.Á. Fernández-Ballesteros & David Arias (Madrid: La
Ley, 2010), 1155–1159.
259 Ibid., 1157.
260 Ibid., 1156.
261 A fee is charged by the Centre for suggestions made.
262 Andrew Stephenson, ‘Creating Efficient Dispute Resolution Processes:
Lessons Learnt from International Arbitration’, Building and Construction Law
Journal 20, no. 3 (2004): 58.
263 See, e.g., IBA Rules of Evidence 2010 Arts 5.5, 5.6 and 8.1.
264 See ICC Rules 2012 Art. 25(4); UNCITRAL Rules 2010 Art. 29.5; LCIA Rules

Art. 21.2; HKIAC Rules Art. 25.4; SIAC Rules Art. 23.3; ICDR Rules Art. 22.4;
Swiss Rules 2012 Art. 27.3 and 27.4; See also German Code of Civil Procedure
Art. 1049(2).
265 J. Martin & H. Hunter, ‘Expert Conferencing and New Methods’, in
International Arbitration 2006: Back to Basics?, ICCA Congress Series No. 13, ed.
Albert Jan van den Berg (The Hague: Kluwer Law International, 2007), 823.
266 Use of conferencing in certain common law courts has been described as ‘hot

tubbing’.
267 Wolfgang Peter, ‘Witness “Conferencing”’, Arbitration International 18, no. 1

(2002): 56.
268 Ibid., and comments of Clifford J. Hendel, ‘Witness Conferencing Involving

Witnesses of Fact: A Good Idea Whose Time Has Not Yet Come?’, IBA Arbitration
News 16, no. 1 (2011): 88.
269 Hilmar Raeschke-Kessler, ‘Witness Conferencing’, in The Leading Arbitrators'

Guide to International Arbitration, 2nd edn, ed. L. Newman & R. Hill (Huntington:
Juris Publishing, 2008), 418.
270 Ibid., 419.
271 Ibid.
272 Ibid., 421.
273 Gary B. Born, International Commercial Arbitration (The Hague, Kluwer Law

International, 2009), 1850.

274 See Wolfgang Peter, ‘Witness “Conferencing”’, Arbitration International 18,

no. 1 (2002): 55.

275 Doug Jones, ‘Party Appointed Expert Witnesses in International Arbitration:

A Protocol at Last’, Arbitration International 24, no. 1 (2008): 148.

276 Wolfgang Peter, ‘Witness “Conferencing”’, Arbitration International 18, no. 1

(2002): 48.
277 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th

edn (Oxford: Oxford University Press, 2009), 424.

278 See David Rivkin, ‘Towards a New Paradigm in International Arbitration;

The Town Elder Model Revisited’, Arbitration International 24, no. 3 (2008):
384.
279 Hilmar Raeschke-Kessler, ‘Witness Conferencing’, in The Leading Arbitrators'

Guide to International Arbitration, 2nd edn, ed. L. Newman & R. Hill (Huntington:
Juris Publishing, 2008), 420.
280 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1850.

281 Wolfgang Peter, ‘Witness “Conferencing”’, Arbitration International 18, no. 1
(2002): 52.
282 Ibid., 51.
283 Hilmar Raeschke-Kessler, ‘Witness Conferencing’, in The Leading Arbitrators'

Guide to International Arbitration, 2nd edn, ed. L. Newman & R. Hill (Huntington:
Juris Publishing, 2008), 424.
284 Wolfgang Peter, ‘Witness “Conferencing”’, Arbitration International 18, no. 1

(2002): 51–52.
285 Hilmar Raeschke-Kessler, ‘Witness Conferencing’, in The Leading Arbitrators'

Guide to International Arbitration, 2nd edn, ed. L. Newman & R. Hill (Huntington:
Juris Publishing, 2008), 424.
286 Ibid., 427.
287 J. Martin & H. Hunter, ‘Expert Conferencing and New Methods’, in
International Arbitration 2006: Back to Basics?, ICCA Congress Series No. 13, ed.
Albert Jan van den Berg (The Hague: Kluwer Law International, 2007), 824.
288 Klaus Sachs & Nils Schmidt-Ahrendts, ‘Protocol on Expert Teaming: A New

Approach to Expert Evidence’, in Arbitration Advocacy in Changing Times, ICCA

Congress Series No. 15, ed. Albert Jan van den Berg (The Hague: Kluwer Law
International, 2011), 145.
289 Ibid., 147.
290 Mark Kantor, ‘A Code of Conduct for Party-Appointed Experts in
International Arbitration – Can One Be Found?’, Arbitration International 26, no.
3 (2010): 338.
291 Ibid., 339.
292 Klaus Sachs & Nils Schmidt-Ahrendts, ‘Protocol on Expert Teaming: A New

Approach to Expert Evidence’, in Arbitration Advocacy in Changing Times, ICCA

Congress Series No. 15, ed. Albert Jan van den Berg (The Hague: Kluwer Law
International, 2011), 145–146.
293 Mark Kantor, ‘A Code of Conduct for Party-Appointed Experts in
International Arbitration – Can One Be Found?’, Arbitration International 26, no.
3 (2010): 339.
294 This may make a difference when there is differential treatment between

these two broad cost categories.

295 Julian D.M. Lew QC, Loukas A. Mistelis & Stefan M. Kröll, Comparative

International Commercial Arbitration (The Hague: Kluwer Law International,

2003), 573; Bernard Hanotiau, ‘The Conduct of the Hearings’, in The Leading
Arbitrators' Guide to International Arbitration, ed. L. Newman & R. Hill, 2nd edn
(Huntington: Juris Publishing, 2008), 372.
296 Peter F. Schlosser, ‘Generalisable Approaches to Agreement with Experts

and Witnesses Acting in Arbitration and International Litigation’, in Global

Reflections on International Law, Commerce and Dispute Resolution: Liber
Amicorum in Honour of Robert Briner, ed. Aksen et al. (Paris: ICC Publishing,
2005), 775 and 778.
297 Christian Oetiker, ‘Witnesses Before the International Arbitral Tribunal’, ASA

Bulletin 25, no. 2 (2007): 266.

298 Paul A Gélinas, ‘Evidence Through Witnesses’, in Arbitration and Oral
Evidence, ed. Lévy & Veeder (Paris: ICC, 2005), 34.
299 See section 2.3.
Part II: The Process of an Arbitration,
Chapter 13: Procedure and Evidence
in Choice of Law and Interpretation

Jeff Waincymer,
13.1. Introduction

This book deals with procedure and evidence in international arbitration. As

such it is not aiming to contain a comprehensive exposition of any substantive
law issues applicable to the field. These typically cover the range of contract and
non-contractual rights and remedies flowing from the commercial dealings
between the parties. Nevertheless, some coverage of the means by which
substantive laws are identified and interpreted is required. This is because there
are a significant number of crucial procedural and evidentiary issues relating to
choice and application of substantive law. The ILA Report Ascertaining the
Content of Applicable Law in International Commercial Arbitration 2008 even
considered that ascertainment of applicable law is ‘by and large procedural…’. (1)

In addition, in many important questions of jurisdiction and coverage of an

arbitration agreement, many practitioners and scholars will use a conflict of laws
methodology as the key means of resolving contentious questions. This may
include questions such as the law applicable to interpretation of the arbitration
agreement, and whether the agreement can include non-signatories, either as
related entities, controlling shareholders, third-party beneficiaries, principals or
assignees. (2)

page "977"

Another reason to consider some applicable law issues in this book is that a
number of topics are more difficult to classify as either procedural or
substantive. In some cases, classification itself is seen as problematic.
Classification is a key difficulty in private international law. (3) Heiskanen makes
the point that issues of classification permeate legal analysis, in particular as to
the distinction between law and fact and questions as to which legal categories
cover particular facts and which facts fit within particular legal categories. (4)
Characterisation questions arise in relation to such issues as limitation periods,
set-off rights and capacity to enter into a contract. Other uncertain examples
include interest and currency. Different solutions might also apply in arbitration
as opposed to domestic private international law regimes. For example, Born
argues that burden of proof should be based on specialised rules and not
classified as either substantive or procedural. (5)

For the above reasons, this chapter looks at these questions in the context of law
governing substantive rights and the conflicts methodologies and discretions
that may apply, while Chapter 14 looks at this in the context of remedies. A
number of other choice of law issues arise in arbitration which have been dealt
with in previous chapters. These include the law governing the procedure, (6) the
law governing the particular reference to arbitration, the law governing the
arbitration agreement, (7) the law governing each parties' capacity (8) and the laws
dealing with court involvement. In many instances, choice of applicable law may
effectively decide the outcome, for example, where facts are not in dispute but
differing legal systems have differing views as to the rights flowing from such
facts. Of course it should also be acknowledged that many cases are essentially
factual. In some cases, there will thus be little need to identify an applicable law,
as analysis of the terms of a detailed contract and/or the relevant facts will
determine the proper outcome. (9)

In considering how arbitrators can and should ascertain the contents of

applicable law when required, the ILA Report 2008 considered there were four
subsidiary questions as follows:

(1) How should arbitrators acquire information about the content of applicable
law?
(2) How should they interact with the parties about the contents of such law?
(3) How should they make use of the information they receive about the law's
contents?
(4) page "978"How should arbitrators address situations calling for special
treatment regarding the contents of the applicable law? (10)

13.2. Procedural and Evidentiary Issues in Selection and Application of The

Laws Dealing With The Merits

This section seeks to simply introduce some of the procedural and evidentiary
questions that arise in relation to choice of law as a precursor to an analysis of
the nature and sources in the balance of the chapter. There are a host of
procedural and evidentiary questions relating to choice of law. First, if asked to
decide on the presence or validity of the arbitration agreement, a tribunal may
need to consider the law applicable to the agreement itself. Because of the
doctrine of autonomy (or separability), it is entirely possible that the arbitration
agreement has a different applicable law to the balance of a contract even if
found as a mere clause in a general contract between the parties. This is also the
case if the contract has a general choice of law clause. Section 3.2.2 looked at
various possible laws that could apply to interpretation of the arbitration
agreement itself. In these circumstances arbitrators need to consider how to
classify an arbitration agreement and whether to adopt a conflicts approach or
consider interpretation from the basis of general principles. This chapter does
not revisit the question of the law applicable to the arbitration agreement itself,
but concentrates instead on applicable law as to the merits.

As to the merits, the first question is whether the parties have any agreement as
to applicable law. While party autonomy is normally paramount, there may be
interpretation and evidentiary questions where there is a dispute as to the
meaning of any ambiguous agreement on applicable law. This in turn may be
circular, given a need to interpret the contract terms, which itself can require an
applicable law as to interpretation. An example might be a contractual
stipulation that the arbitrators are to use ‘equitable’ principles. Does this simply
mean equity as the term is used in the common law, good faith as per civilian
systems, or did the parties empower (and compel) the tribunal to decide as
amiable compositeur? Another example of uncertainty is whether a specific
choice of law clause includes choice of that nationality's conflicts rules. (11)
Ideally, parties will draft their agreements as to page "979" applicable law clearly
and lawyers will ensure that these matters are unambiguously dealt with. There
is also a need to consider if and when mandatory substantive laws or principles
of public policy should apply to override any agreement of the parties. (12)

In many cases, the parties have not selected an applicable law in their contract.
Even then, party autonomy allows them to make a choice at any stage. One
procedural issue is to remember to offer them an opportunity to reach such an
agreement. If that is not possible and applicable law is hence in dispute, there are
further important procedural questions that arise. Where a tribunal has
discretion as to applicable law, there is a need to consider the procedure and
principles by which such discretion will be exercised as well as its essential
nature. There are also timing and procedural questions that arise. A tribunal may
need to consider whether there should be a preliminary hearing and a partial
award on this issue. This would allow the parties to limit their research and
preparation in order to save overall costs. In some cases, however, a tribunal
may need to hear all of the evidence before deciding on a dispute as to applicable
law, or so as to sensibly apply a discretion in that regard.

Once an applicable law is known, there is then a question of how it is to be

presented and understood by arbitrators who often come from different legal
families. The very fact that international arbitrators may not be familiar with the
applicable law or may not be long-standing experts as to that law, makes issues
of procedure and evidence more central to adequate determinations in this
arena. There are questions as to proof of foreign law discussed in sections 10.6
and 13.16 and the contentious question of the tribunal's own duty to know the
relevant legal principles discussed in section 13.19. If some transnational rules of
lex mercatoria are to be applied, a tribunal will need to understand how to
determine the ambit of such concepts. Where a tribunal has been empowered as
an amiable compositeur, thought will need to be given as to the impact this has
on otherwise applicable legal rules. These issues are discussed in section 13.14.

Once the law is known and all of the relevant documentary material is available,
a tribunal will often need to consider questions of interpretation for treaties,
statutes, contracts and other relevant commercial and investment instruments.
That in turn can raise evidentiary questions in terms of applicable aids to
interpretation. As noted above, interpretation may even be needed as to the
choice of law clause. Interpretation of treaty law can be particularly problematic.
Where investment arbitration is concerned, tribunal determinations are at the
cutting edge of highly contentious and political debates about the ambit of norms
such as compensation for expropriation and non-discrimination and social
welfare exemptions in international treaties. These can also have important
evidentiary and procedural issues such as proof of government intent where that
may be relevant. Some unique aspects of confidentiality also arise where
governments are involved given that investment arbitration is the prime area
where amicus curiae are at times used.

page "980"
An important interpretation question is to determine by what methodology a
tribunal will come to understand the law and interpret it. For example, where a
national law is involved, is a tribunal bound to simply consider how a
conservative domestic judge would apply the law, or may an adjudicator take a
more activist role in developing that law through creative interpretation to make
it consistent with perceived international norms as an appellate judge might do,
at least in common law countries? Similarly, if an arbitrator discerns a majority
view as to the local law, can he or she choose a minority view on the basis that it
is more sensible and more likely to become the dominant view over time? (13)
Arbitral laws and rules have no express guidance on such essential questions.
This is discussed further in section 13.18.

When applicable substantive laws as to the merits are known, there will be
further evidentiary and procedural questions arising. Applicable substantive
laws will commonly define or circumscribe the potential principles pertaining to
the issues in dispute and will often allocate or qualify burden and standard of
proof. (14) Applicable legal standards may also have complex internal evidentiary
questions, such as where they impose minimum quality standards on the
provision of goods or services or as to inspection of goods. There will also be
timing questions as to inspection and notices. Factual questions may also arise
where there is an allegation that rights have been waived.

Finally, there are issues of annulment and enforceability that can arise. These are
not discussed separately in this book, save where they may affect the approach
that a tribunal should take. The point to make is simply that while errors in
application of substantive law would not normally ground valid challenges,
because of the important procedural and evidentiary questions underlying
choice of law, the potential for challenge remains as to any procedural flaws and
a tribunal must be alert to all due process aspects of selection, presentation and
analysis. As just one example, a highly contentious aspect touched on above and
dealt with below is whether a tribunal has a duty to know the law and a right to
apply a law different to that argued by the parties without providing them with
adequate notice. Expansive approaches to articulating public policy norms may
also impact in this area, if they are argued to cover the procedural approach to
identifying applicable law. Public policy may also have a substantive law element
where certain kinds of contracts would not be recognised because of their
subject matter. A related issue is whether a tribunal should consider a question
of public policy even if not argued by the parties. (15) For all the foregoing
reasons, it is important that a tribunal give adequate thought to the proper
procedural and evidentiary approaches it takes to identifying, interpreting and
applying substantive law.

page "981"

13.3. Sources of Law and Choice of Law

As noted, this chapter is primarily concerned with the identification of the

applicable law that will pertain to the substantive issues as between the parties.
It is important to remember that in any international transaction, there will be
terms agreed to between the parties and in addition, some applicable laws that
include rules on the validity and interpretation of those terms. These typically
introduce other applicable principles and outline the range and scope of
permissible remedies. Sources of such laws or qualifications to such laws can
include the arbitration agreement itself or other agreements between the
parties, the lex arbitri and arbitral rules. In some cases, transnational or
international norms and/or customs and usages may also apply, as may
mandatory rules from the Seat or from a foreign jurisdiction. Applicable laws will
also be impacted upon by the claims as made. For example, a claim for
misrepresentation must identify the applicable law of misrepresentation
whether the claimant has a valid claim on this basis or not. To understand the
gamut of relevant law thus requires a tribunal to consider an amalgam of all of
these sources. These are addressed in turn in the following sections.

A tribunal will need to consider when and why each applies and how to properly
integrate them. Lew, Mistelis and Kröll suggest a three-step method:

• Is there a choice of law by the parties? In the affirmative, does it cover the
entire dispute before the tribunal? Is there any contradiction of mandatory
rules or public policy? In any event, is it possible to solve the dispute by
relying exclusively on the contract?
• In the absence of choice of law by the parties, is it possible to determine the
applicable law or non-legal standard without reference to conflict of laws? Is
that an appropriate law or non-legal standard for this particular dispute?
• If it is necessary to employ conflict of laws rules, which conflict of laws rules
should be applied? Is there a conflict of laws system most closely connected
with the dispute? Is it possible to apply cumulatively all relevant conflict of
laws systems? Is it possible to apply generally accepted conflict of laws rules?
(16)

The ILA Report 2008 noted that there is no uniform practice amongst arbitral
tribunals as to determination of applicable law. The ILA Report argued against
the relevance of domestic court approaches to ascertaining applicable law. This
was so for a number of reasons. First, national legal systems vary significantly.
Second, in applying a national rule there would be a conflicts issue as to which
law should apply. There would be difficult characterisation question in
employing a conflict methodology. While many courts will apply their own
domestic law if there is insufficient proof of foreign law, there is no similar
fallback available in page "982" international arbitration. The ILA Report 2008
concluded that the question as to how arbitrators should ascertain the contents
of applicable law ‘is not subject to any meaningful prescriptive laws or rules …’
but that certain recommendations were desirable to at least provide ‘outer
parameters for the exercise of an arbitrator's discretion’. The committee
reiterated the duty to comply with the mandate and to be bound by the scope of
the arbitration agreement. Allied to that is the duty of due process. Parties
should have a reasonable opportunity to address important legal points and the
tribunal should not take the parties by surprise.
The ILA made a number of specific recommendations, but noted that these were
merely guides and commented that failure to follow the recommendations
should not be treated as a departure from acceptable practice or a breach of an
arbitrator's duty. The recommendations were as follows:

General considerations
1. At any time in the proceedings that a question requiring the application of a
rule of law (including a question of jurisdiction, procedure, merits or
conflicts of laws) arises, arbitrators should identify the potentially
applicable law and rules and ascertain their contents insofar as it is
necessary to do so to decide the dispute.
2. In ascertaining the contents of applicable law and rules, arbitrators should
respect the process and public policy, proceed in a manner that is fair to the
parties, deliver an award within the submission to arbitration and avoid
bias or appearance of bias.

Acquiring information.

3. When it appears to the arbitrators that the contents of applicable law might
significantly affect the outcome of the case, arbitrators should promptly
raise that topic with the parties and establish appropriate procedures as to
how the contents of the law will be ascertained (in submissions with
materials attached, through experts, witnesses or otherwise).
4. Arbitrators attempting to ascertain the contents of applicable law should
bear in mind that the rules governing the ascertainment of the contents of
law by national courts are not necessarily suitable for arbitration, given the
fundamental differences between international arbitration and litigation
before national courts. In particular, arbitrators should not rely on
unexpressed presumptions as to the contents of the applicable law,
including any presumption that it is the same as the law best known to the
tribunal or to any of its members, or even that it is the same as the law of the
seat of the arbitration.

Interaction with parties

5. Arbitrators should primarily receive information about the contents of the

applicable law from the parties.
6. In general, and subject to Recommendation 13, arbitrators should not
introduce legal issues – propositions of law that may bear on the outcome of
the dispute – that the parties have not raised. page "983"
7. Arbitrators are not confined to the parties' submissions about the contents
of applicable law. Subject to Recommendation 8, arbitrators may question
the parties about legal issues the parties have raised and about their
submissions and evidence on the contents of the applicable law, may review
sources not invoked by the parties relating to those legal issues and may, in
a transparent manner rely, on their own knowledge as to the applicable law
as it relates to those legal issues.
8. Before reaching their conclusions and rendering a decision or an award,
arbitrators should give parties a reasonable opportunity to be heard on
legal issues that may be relevant to the disposition of the case. They should
not give decisions that might reasonably be expected to surprise the parties,
or any of them, or that are based on legal issues not raised by or with the
parties.

Making use of information about law's content

9. In ascertaining the contents of a potentially applicable law or rule,

arbitrators may consider and give appropriate weight to any reliable
source, including statutes, case law, submissions of the parties' advocates,
opinions and cross-examination of experts, scholarly writings and the like.
10. If arbitrators intend to rely on sources not invoked by the parties, they
should bring those sources to the attention of the parties and invite their
comments, at least if those sources go meaningfully beyond the sources the
parties have already invoked and might significantly affect the outcome of
the case. Arbitrators may rely on such additional sources without further
notice to the parties if those sources merely corroborate or reinforce other
sources already addressed by the parties.
11. If in the course of deliberations arbitrators consider that further
information about the contents of applicable law is necessary to the
disposition of the case, they should consider reopening the proceedings to
enable the parties to make further submissions on the open legal issues,
but only to the extent necessary to address the open legal issues and taking
into account considerations of relevance, time and cost.
12. In applying the rules of the applicable law, arbitrators should give due
regard to available information about the application of the rules in the
jurisdiction from which the rules emanate.

Special circumstances

13. In disputes implicating rules of public policy or other rules from which the
parties may not derogate, arbitrators may be justified in taking measures
appropriate to determine the applicability and contents of such rules,
including by making independent research, raising with the parties new
issues (whether legal or factual), and giving appropriate instructions or
ordering appropriate measures insofar as they consider this necessary to
abide by those rules or to protect against challenges to the award.
14. page "984"In applying the foregoing Recommendations, arbitrators may
take account of the nature of the proceedings, in particular regarding
default and expedited interim relief proceedings, and may take a more
active role than might otherwise be the case in questioning legal
submissions.
15. If after diligent effort consistent with these Recommendations the contents
of the applicable law cannot be ascertained, arbitrators may apply
whatever law or rules they consider appropriate on a reasoned basis, after
 giving the parties notice and a reasonable opportunity to be heard.

13.4. Party Selection

13.4.1. The Value of Autonomy as to Selection of Applicable Law

Virtually all international statutes and rules support party autonomy in selecting
applicable law subject to certain limits and provide broad discretions to the
tribunal in the event that there is no agreement between the parties. (17) Party
autonomy has also been accepted in disputes involving State parties. (18) The
Rome Convention and Rome Regulation (19) also support party choice, although
these do not apply to arbitration per se. It makes good sense to support party
autonomy as to applicable law. The whole essence of contract is that the parties
may agree on the terms and conditions of their commercial relationship. Hence,
they ought to be able to agree on ancillary rules and interpretational principles
pertaining to those matters. The parties could have elaborated on their rights
and obligations encompassing all the norms from a desired system of applicable
law. That would be time consuming and costly. It makes sense instead to allow
selection of an applicable law to be respected as a shorthand means of
incorporating such norms.

There are also efficiency gains from such a choice. Absent agreement, the
tribunal will generally have a discretion, which will follow reasoned argument by
the parties. If the argument occurs at a separate hearing, this can add to the cost
and delay the proceedings. While the a priori preference would still be for a
preliminary determination to save subsequent cost, this may not be seen as
possible or desirable where a full evidentiary record might first be needed, for
example, to understand the nature of a complex transaction and application of a
closest connection test. Yet if the tribunal believes that the matter should not be
bifurcated and questions of applicable law are left to the final hearing, this can
also add significantly to the expense, but in differing ways, given that there
would be a need to research and page "985" argue in relation to alternative legal
systems. Hence, a failure to select law has inevitable unproductive transaction
costs. However, there can be problems in trying to identify an applicable law at a
point in time when the nature of the dispute is unknown. Any selection may not
prove to be the most ideal for the dispute that eventually emerges. Conversely,
an attempt to reach an agreement once a dispute is known may be more difficult,
as each party will be able to determine how any selection advances or interferes
with its interests.

While there are thus policy arguments either way, advance selection is to be
preferred if the selection is as to an advanced, balanced and fair model, although
there can be legitimate differences in view as to such criteria. Special care should
be taken with selections that have serious implications even where facts are
clear, such as very short or long limitation periods.

13.4.2. The Ambit and Limits of Autonomy

Party autonomy relates both to a direct choice of applicable substantive law or a
choice as to the methodology to be employed by a tribunal where that is
appropriate. While lex arbitri or rules will support either form of autonomy, at
times these may also limit choice. They may be drafted in a way that directs
selection of a national system of law, which may adversely impact on the parties'
rights to select an autonomous religious system or lex mercatoria as applying.
Some national laws simply speak of selection of ‘the law’ chosen by the parties.
(20) This is also argued to arise because some religious-based systems may not

support choice of law agreements. There are also countries that apply the Calvo
doctrine and its limits on party autonomy. (21) The US FAA does not expressly
address party autonomy, but State laws and common law support this position.
(22)

While a choice is permissible at any time, parties might not be entitled to select a
law after tribunal appointment that could not have been reasonably anticipated
by the tribunal and which would occasion excessive additional cost and delay
without providing for additional tribunal remuneration. Party autonomy also
needs to be read in the context of custom and usage where a tribunal is
mandated to apply these norms.

Where autonomy is concerned, the parties are free to select a law that otherwise
has no natural connection to the transaction. Choice of an unconnected law
typically arises in industries where there has been a habitual reliance on a well-
developed system, such as English law in relation to transport and insurance.
That is similar to a customary practice or usage. While some national systems
will reject page "986" a choice of law if it is not sufficiently connected to the
jurisdiction, such an argument should not be applicable in arbitration. It would
be contrary to express permission for parties to select applicable law under most
lex arbitri. Potential problems should only arise if there is a tenable argument
that the choice is somehow unfair. The power is subject to mandatory due
process norms so it may still be possible to argue that the selection did not treat
parties equally or fairly but that would rarely be so. An example would be
selection of a law that aims to remove protections for one of the parties that
would be found in a more naturally applicable law. The latter example highlights
the fact that while party autonomy is generally paramount, nevertheless, there
may be limits on party autonomy. A choice must be bona fide and not contrary to
public policy. (23) Because different jurisdictions have different views as to the
nature of public policy, those who take the most expansive view would impose
greater limits on party autonomy.

There might be a related question as to whether the choice improperly seeks to

obviate reliance on a mandatory substantive law. Where mandatory laws are
concerned, this can impact on party autonomy in two ways. First, even if the
parties select a particular law, another jurisdiction's mandatory laws may apply.
This is discussed in section 13.8. Second, in extreme cases, a particular system of
conflicts rules may suggest that a choice of a foreign law not otherwise
connected to the transaction will not be accepted, particularly if it appears to
have been motivated by a wish to avoid otherwise applicable rules. A typical
example is selection of a law that would make an arrangement valid where the
connected jurisdictions consider the transaction to be illegal. In some cases, such
a selection may have occurred because of ignorance of one party as to the biased
nature of the selected law. A party might argue that its acceptance of the choice
was made in ignorance of the other party's wish to improperly avoid or limit
liability as a result of the choice and should, therefore, not be binding. That
would involve factual and interpretational decisions by a tribunal.

On rare occasions, tribunals have chosen not to give full effect to a choice of law
clause where it has been held that the application of a law would be contrary to
the legitimate expectations of the parties. (24) In some cases a tribunal might
reasonably ignore a choice of law made by the parties if that would lead to the
entire contract being invalid. (25) The logic would be that the parties could not
have been taken to intend a choice of law to undermine the very express contract
they entered into. An express choice of law might also be read down if it was
seen to undermine an express agreement of the parties as to elements of the
contract terms. This would be so where the applicable law renders a particular
provision invalid. Often parties page "987" would be unaware of this impact and
the voidable term may have been a crucial element in setting a mutually
agreeable price. (26) A tribunal might legitimately conclude that the express
contract term was intended to prevail over the more general choice as to
applicable law.

While there are thus some justifiable limits to party autonomy, there are cases
where tribunals have improperly rejected party choice. In some situations,
tribunals have unfortunately asserted that selected laws are unsatisfactory and
not appropriate for international commerce. Such an approach cannot be
justified. If that law has a solution to the particular issue, then the parties have
chosen it to apply. If the law has a gap, the tribunal has a designated
methodology or broad discretion on gap filling.

13.4.3. Interpreting the Choice Made and the Ambit of the Choice

If the parties have selected an applicable law, there is a need to determine how
the selection is to be interpreted if it is in any way unclear. In turn there would
be a need to consider which interpretation principles should apply. There is a
circular problem if one looked for applicable law in that context. An alternative
approach is to use broad principles of interpretation as discussed in section
3.2.4. In an extreme case, the parties' designation may be so unclear as to be held
to be inoperable or incapable of interpretation.

While approaches and conclusions may vary, a number of general observations

may be made. Even where the parties do not indicate that their choice is an
exclusive one, this would normally be implied from any selection. When selecting
a national law, parties will typically select the entire system of law. This would
cover all claims and remedies. Some laws would not typically be seen to fall
within a simple choice made. These would include laws as to real property,
capacity or powers of attorney. (27) Under some conflicts rules, choice of law also
does not cover validity of the contract. (28) When parties have selected a national
system of law, an page "988" important question is whether this also includes
that system's conflict of laws rules. Here one problem is that where such rules
apply, they may direct the adjudicator to a foreign law and potentially a situation
of renvoi. (29) This is a question of interpreting intent. If the parties know what
they want, they can express it clearly. In the absence of clear guidance,
arbitrators will generally presume a choice of applicable law to be limited to
substantive law and not choice of law rules to avoid renvoi. (30) Party designation
to this effect would be unnecessary if that presumption is expressly designated
in the lex arbitri or selected rules. Most laws and rules, including Article 28(1) of
the Model Law, expressly indicate that a selection of law does not include the
conflict of laws rules. (31)

As to scope of any party agreement as to applicable law, there may be a question

as to whether a general choice of law clause is intended to apply to the
arbitration agreement or only to the balance of the contract. If parties expressly
indicate that a choice of law applies to the arbitration clause itself, this will be
accepted. In other circumstances, where there is merely a general choice of law,
because of the doctrine of separability, many take the view that the general
selection does not automatically apply to the arbitration clause itself. While
many will apply such a law either through implied intent or for consistency
under discretionary powers, there are a range of potential applicable laws in this
circumstance. This was discussed in section 3.2.2. Even the place where a choice
of law clause is situated in a contract can impact on its meaning. If it is contained
within the arbitration clause, unless clearly expressed, some might think it
should be limited to interpretation of the clause itself. Conversely, if in the
balance of the contract, as noted, many argue that it does not naturally apply to
the autonomous arbitration agreement. Once again, these are simply ambiguities
that can be overcome with clear drafting.

The general presumption is that a choice of law deals with substantive and not
procedural matters. Nevertheless, problems of classification will at times give
rise to uncertainty. This is particularly so with questions such as burden of proof,
statutes of limitation and rights to interest where different legal families have
different views on characterisation. Again, clarity in drafting is the only safe
solution.

In some cases, parties might only have selected contract laws or otherwise
limited the ambit of their selection or this may be arguable. Choice of law clauses
that merely refer to how an agreement is ‘construed’ or ‘interpreted’ make it
harder to have their reach extended to non-contractual claims. It is even
arguable that this does not include distinct substantive rights provided for in the
designated contract page "989" law system external to the contract terms, and
only aids in understanding the rights as expressed in the agreement. In any
event, it is particularly desirable to have the ambit of the choice of law clause
match the ambit of the arbitration agreement. (32) Hence, a broad selection should
be preferred. A contract might also include a saving clause that provides for
alternative applicable laws and gives a preference to that which promotes
viability of the contract.

Some choice of law clauses may seek to apply different laws to different
elements. (33) This is perfectly acceptable, although care should be taken where
there is a potential overlap between the elements and uncertainty at the
margins. Particular problems arise where parties seek to apply a number of
national systems’ laws to the same issues. Examples include cases where
reference is made to principles ‘common’ to more than one legal system or
provisions in one that are ‘not in conflict’ with the other. (34) A related example
would be where the parties expressly select some version of tronc commun in
their agreements, calling for the application of common principles as pertaining
between the relevant States or as between civilised nations. (35) Such choices
proved to be particularly problematic in terms of uncertainty, delay and cost in
the Channel Tunnel arbitration. (36)

Party agreement may also deal with the applicability or non-applicability of

future changes to the relevant laws. It is particularly prevalent in investment
agreements to incorporate stabilisation clauses that seek to limit the impact on
the parties to the laws in existence at the time the agreement was entered into.
While there are different academic views about the value and validity of
stabilisation clauses, these are generally upheld. These are still politically
contentious as critics assert that they can unduly limit the power of governments
to change their laws for social benefit. Stabilisation clauses are discussed further
in section 13.12. Instead of a blanket stabilisation provision, parties may at times
indicate that new national laws shall only apply to the extent consistent with the
contract. That would seem similar in nature and there may be interpretation
questions as to when it would apply. In other cases, parties may simply include a
price variation provision aimed at rebalancing the benefits where laws change.

Other situations which can cause confusion include cases where the parties
might select the contract law of a particular nation, where that nation is a
signatory to the CISC Under that law, it applies automatically unless excluded by
the parties. page "990" Hence, the above designation should be readily
interpreted as agreement to apply the CISC Nevertheless, many counsel and
adjudicators are ignorant of the CISG and see the above designation as a
selection of domestic contract law. Party choice may also be ambiguous in
federal systems where a choice is made as to either State or Commonwealth law
without reference to the other. Constitutional principles may still make the other
potentially applicable and uncertainty may also apply where the European Union
is concerned or other groups of countries are involved as to the applicability of
trans-national as well as national laws. (37)

Because the choice can be express or implied, there are important factual
questions as to the standard of proof that a tribunal will apply. There needs to be
some evidence of an intent to apply a particular law. The mere selection of the
Seat of arbitration is no longer considered to be evidence of this. There is still a
significant difference between an arbitrator who would wish to see some direct
evidence of the parties' intent and other arbitrators who might presume that the
parties impliedly intended anything that the arbitrator would think to be fair and
efficient in the circumstances of the transaction. That is particularly so where, as
in many cases, one would best presume that the parties simply never turned
their mind to this question. (38) The Rome Convention calls for an implied choice
to be ‘demonstrated with reasonable certainty’ (39) while the Guiliano and
Lagarde Report suggests the need for a ‘clear intention of making a choice’. (40)
These do not apply directly to arbitration and it may not be appropriate to add
extra expressions about determinations that are otherwise made on balance or
to the tribunal's satisfaction.

Parties may show an implied choice where they argue the case on the basis of the
same law. (41) In some cases, parties may have reached a limited choice by simply
excluding certain laws. In some cases parties might allege that they have made a
negative choice to exclude certain national laws, leaving the tribunal with a
residual discretion. The better view is that a mere failure to select a law should
not be seen as evidence of an unwillingness to apply either party's national laws
although such a conclusion may be valid if evidence of impasse during the
negotiations leads to this conclusion. (42)

page "991"

13.5. Arbitrator Selection of Applicable Law

Statistics from the ICC show that about 20% of arbitrations do not have an
express choice of law in the relevant contract. (43) In such circumstances a choice
must be made by other means. This will typically be via tribunal discretion.

13.5.1. The Sources of and Constraints on Tribunal Discretion

Modern lex arbitri will generally indicate that a tribunal is bound by an

agreement of the parties and absent such an agreement, a tribunal has a broad
discretion. (44) At times this discretion can be constrained. As noted above, the
parties may have indicated a negative choice, for example by refusing to accept
each other's laws. In other cases the discretion is constrained in the laws or
rules. Some arbitral laws do not expressly empower a tribunal to select a law (45)
while others only allow a tribunal to do so when expressly empowered by the
parties. (46) It would appear that those countries' domestic laws would apply
absent any such agreement of the parties. (47) Some arbitration laws specify the
particular conflicts rule to apply. (48) Some legal systems require that the
arbitrator apply the conflicts rules of the Seat. (49) The Washington Convention
provides that in the absence of party choice, the tribunal must apply the law of
the contracting State party together with such rules of international law as may
be applicable. (50)

Even if the parties have not directly designated a law, they may have directed a
methodology for determination. If they have designated arbitral rules in their
contract, the designation has the same status as other contractual provisions and
such choice is accepted by most lex arbitri. Those rules should then prevail in
terms page "992" of designation of applicable law. (51) Where the lex arbitri seeks
to constrain the choice and the rules selected are more open ended, a question
can still arise as to which prevails. The rules selected should prevail for the same
reason that a direct choice of law is acceptable in such circumstances, (52)
although some jurisdictions might treat the statutory provision as mandatory.
Where the lex arbitri designates a conflicts rule, the better view is that this is not
of a mandatory character. (53)

Historically, some laws and arbitral rules placed an undesirable restriction on

tribunals simply through reference to a choice of ‘law’ or ‘the’ law within the
discretionary power. There may be different permutations depending on what
phrase is used in the arbitration statute and rules, and in addition, whether
different phraseology is used for party choice and for arbitrator choice. (54) On a
plain meaning approach to interpretation, the reference to ‘law’ in the singular
led many commentators to believe that a tribunal could only pick one particular
national system of law. An important corollary of this was that arbitrators felt
unable to select transnational principles, otherwise described as lex mercatoria.
The more restrictive reference to singular law may also make it more difficult for
depecage or for an autonomous non-national religious system to be applicable. In
most cases the discretion has now been broadened, with institutional rules
providing discretion to select applicable ‘rules of law’. (55) The Model Law is seen
as more restrictive as it refers to ‘the’ law determined by the conflict of law rules
selected. Nevertheless, some of the more restrictive lex arbitri have been
interpreted broadly, (56) including the Model Law itself. (57)

An arbitrator's choice should not be subject to annulment or enforcement

challenges except where the choice is contrary to a direction of the parties, in
which case Article V(2)(a)(iii) and (b) of the New York Convention could apply.
(58) If a tribunal ignores a choice of law by the parties or wrongly decides as

amiable compositeur, this should give rise to grounds for setting aside or page
"993" challenging enforcement. There may be rights to appeal against an
arbitrator's choice of law under section 69 of the English Arbitration Act 1996
for misapplication of conflicts of law principles.

13.5.2. The Applicable Methodologies for Tribunal Discretion

If a tribunal has a broad discretion, this will be unconstrained but there ought to
be a number of criteria that are relevant. Where an arbitrator has a broad choice,
Poudret and Besson suggest that the choice should be foreseeable and not be
‘arbitrary or even unusual …’. (59) Where discretion is provided to the tribunal, it
typically falls into one of two groups. The first gives a tribunal discretion over
choice of an appropriate conflicts of law rule. A tribunal adopting this approach
will in theory determine the applicable conflicts rule, apply it as accurately as it
can and identify the system of law so determined. The alternative form of
discretion is to bypass a conflicts methodology and simply indicate that the
tribunal has discretion to directly pick the appropriate substantive law. (60) While
some have suggested that the direct method does not require reasons for the
choice, (61) the better view is that there should be no limitation on the obligation
to give reasons for decisions made. (62) Nevertheless, when making a selection, it
may at times be difficult to clearly articulate the reasoning as to why a particular
law is more suitable. For example, a tribunal would feel uncomfortable in looking
at the outcome of that law, as it may appear to have pre-judged the actual
determination, particularly when choice of law is made as a preliminary
determination.
While the methodologies of direct choice or conflicts are expressly different,
some commentators suggest that arbitrators inevitably follow similar thought
processes in each case. The logic behind the assertion is that whether one
determines what makes a particular conflicts rule most appropriate or a
substantive rule most appropriate, an arbitrator will consider all of the
circumstances of the case and the ramifications of different options. Other
scholars suggest a more conceptual difference. The growth in utilisation of the
direct method has been said to support a more interest- or contextual-based
approach to identification of law than a classic page "994" rules-based approach
under conflicts theories. (63) From a policy perspective, the different approaches
are seen respectively as promoting certainty and neutrality (the rules-based
approach) and fairness and equity (the interest-based approach). (64) The direct
method can also be different to the interest approach as it is not simply looking
at the relevant interests of the jurisdictions concerned, but is instead concerned
to analyse a range of other circumstances, including the reasonable expectations
of the parties. (65) Practical aspects of these alternative approaches are addressed
in the following sections.

13.5.3. Reasons for Discretionary Choice: Conflicts Methodology

Where the tribunal's discretion is via the indirect or conflicts method, a tribunal
may simply have a general preference as to which conflicts rule is the most
appropriate for international transactions. If that is a widespread norm, it
appeals to both fairness and consistency. Conversely, choosing amongst conflicts
rules could look to either their inherent nature or their source. Where source is
concerned, one could look at the conflict of laws rules of the Seat, of potential
enforcement jurisdictions or of the place that would otherwise have had
jurisdiction but for the arbitration. Each is problematic. The Seat may simply
have been selected for neutrality reasons. (66) There will often be uncertainty and
a multiplicity of possibilities where enforcement is concerned. It is similarly
difficult to identify the State with jurisdiction but for the arbitration, as there
may be more than one with problematic forum non conveniens issues in some
systems at least, and in any event, the norms of a system that the parties rejected
should not readily be determinative within their arbitration proceedings.

As to their inherent nature, the two leading options are the closest connection
test, historically the dominant approach in common law countries and the
domicile of the person exercising characteristic performance, typically utilised in
civilian systems. The Rome Convention and Regulation, not directly applicable to
arbitration, attempt to combine the two. The Regulation now confirms a
rebuttable presumption that the domicile of the party effecting characteristic
performance will prevail except where another place is manifestly more closely
connected. (67) Some tribunals consider that a closest connection test is a
transnational principle of page "995" private international law. (68) A closest
connection test appeals to the notion that the law will be tailor-made for the
particular contractual circumstance, although the key problem is that in modern
commerce, there are so many relevant factors, hence application of the test can
be uncertain and unpredictable. There is also the uncertainty as to whether this
approach should lead to depecage where different laws apply to different parts of
the contract. The domicile test still needs a view to be taken as to what
constitutes characteristic performance although that will typically point to the
person providing goods or services. That is, hence, more certain, although it may
not lead to the fairest law as it inherently favours the laws of one type of
contracting party, namely the supplier.

As to other conflicts rules, some such as place of conclusion of the contract are
undesirable as this can be uncertain or arbitrary with electronic commerce and
with ongoing negotiations with a range of potential counter-offers. It can also be
circular, as one may need to know the applicable law to know where and when
an acceptance takes place. (69)

If a conflicts method is utilised, and the parties come from legal systems that
each apply the same method domestically, the tribunal may well consider this to
be applicable based on reasonable expectations. If a contract is involved and the
parties both come from countries that have adopted the CISG, that ought to apply
as there is the identical law applicable to each which has not been excluded by
agreement. In each case this must be closer to the parties' reasonable
expectations. (70) Even then there is a question as to the way particular domestic
case law may have inappropriately interpreted its provisions in a parochial
manner. (71)

A tribunal may feel comfortable employing a conflicts methodology where there

is a false conflict or where cumulative application of relevant conflicts rules leads
to the same outcome. A cumulative method looking for a false conflict is a
sensible initial method, although it provides no guidance when differences are
found. A cumulative method is unsatisfactory if it is merely used strategically to
try and show that the same law would apply under different conflicts methods
where that is not truly the case. An example would be concentrating on those
factors under a closest connection test that lead to the domicile of the person
exercising characteristic performance. Using this method also raises problems
page "996" where there is a difference found, as the tribunal must then choose
between laws at a time when the impact of the differences is known.

13.5.4. Reasons for Discretionary Choice: Direct Choice Methodology

As noted above, one problem is that a choice of applicable law at the time a
dispute is known means that an arbitrator will often understand how the choice
made will impact on respective rights and obligations and potential remedies. An
arbitrator's choice of law could, in fact, decide the case before it is effectively
begun. (72) Areas where arbitrator choice will be likely to have the most
significant impact are in relation to penalty clauses, punitive damages,
limitations and exemptions and good faith obligations. This is discussed further
in the following section in relation to timing and applicable law.

Where the discretion is exercised, if a tribunal knows that a choice will have such
an impact, it is preferable that there is a justifiably neutral policy basis for
making the choice. Where a direct choice is to be made, as noted above, some
tribunals will consider the same criteria as are applicable in choice of law
methodology. Others take a more general view. Arbitrators are said to adopt ‘a
pragmatic and result-oriented standpoint’ and ‘a functional perspective’. (73) The
choice should ‘not defeat reasonable expectations of the parties as expressed in
their agreement’. (74) A tribunal will consider the contract, the circumstances of
the case and the submissions of the parties. In addition to the submissions, other
post-contract behaviour may also be relevant. Other factors are ‘compromise,
neutrality and sophistication of the legal system’. (75) It is suggested that tribunals
should prefer international conventions, instruments or standards where
permissible, to ‘denationalise the dispute’. (76) At other times, a tribunal might
seek to apply what it sees as lex mercatoria or tronc commun. (77) In addition to
lex mercatoria, a tribunal may also consider the application of religious systems
of law where that might seem within the reasonable expectations of the parties.

page "997"

When considering attributes of a system under a direct choice, arbitrators need

to be aware that many national systems of law have not always attempted to
balance the rights between parties evenly, at times being concerned for their
national interest in transactions of a certain nature. For example, where
transport law is concerned, countries are typically divided between carrier
countries and shipper countries, with commensurate differences in liability
norms. In such circumstances, choices will not always be truly neutral. This also
applies to international treaties where national power imbalances skew the
rights and obligations in the treaty for similar reasons, although treaties will
apply as of right.

In other cases, a tribunal might select a law to promote validity, under the logic
that a valid agreement is closer to the original intent of the parties. In Valenciana
the French Cour de Cassation supported a choice of lex mercatoria by a tribunal
where it was not chosen by the parties to uphold validity which would not have
been possible under French law where price of goods had not been specified. (78)
A contentious aspect of such determinations is that this will often involve
bypassing what might appear to be a mandatory substantive law that might
require certain forms for valid transactions or might demand registration. (79)

Some advocate selection of the law most relevant to the commercial

circumstances. For example, if a seller claims force majeure, it may be logical to
apply the notion of that law at the place where the seller asserts the barrier to
performance occurred. Warranty laws may be those of the place where
merchandise is to be used. (80) Some suggest that the search should be for the law
that would have been likely to have been selected if the parties turned their mind
to it. Some arbitrators would come to such a conclusion by a contentious
methodology by finding implied intent without strong evidence in that regard. It
has also been suggested that looking for implied intent is similar conceptually to
looking at the connecting factors in a closest connection test. (81) Arguments
based on legitimate expectations are more realistic if it is about ab initio validity,
but less clear if they only relate to timing or nature of remedies, the latter which
would have been factored into price by the parties. In many other cases it would
be difficult to identify the parties' common expectations. The situation may vary
depending on whether it is an FOB or CIF contract. Whether one party might
have naturally thought about the legal system of the other may also depend on
the type of goods, for example where page "998" foodstuffs might require an
understanding of the regulatory regime in the buyer's country.

In some cases, a tribunal may find that its discretion has been narrowed by a
positive or negative determination of the parties. As noted in the previous
section, an examination of the negotiating history may show that the parties
were unwilling to agree to each other's laws but could not resolve which law
should actually apply. In such circumstances, a tribunal applying a direct
methodology might consider which of the features of each country's legal system
was particularly problematic as a stumbling block to an agreement. That might
point to a more neutral law, although that may be impossible to achieve once the
facts of the case are known.

If the parties come from the same legal family, then selecting laws of that nature
would be closer to their legitimate expectations. Where the parties come from
different legal systems, the tribunal would need to look for other factors such as
trade usage or evidence from past dealings to try and find an implied common
intent. Where this is not possible, a tribunal might select the law it feels more
closely aligns to the optimal requirements of an international commercial
regime. All other things being equal, legal systems that encourage contracts to be
kept on foot and which call for all parties to seek to minimise and redress harm
are preferable in international transactions where the costs of unravelling deals
are far more significant. Nevertheless, tribunals should be careful in properly
integrating general considerations as to laws best suited to international
transactions and the circumstances of the instant case. This is particularly
because the parties may have expressly or impliedly bargained about the very
matter that may be treated differently by a different applicable law. For example,
a contractor who has agreed to heavy penalties for late completion may have
been rewarded by a very high contract price from an employer that needs timely
delivery.

Other scenarios will be problematic because there will be no common view as to

what the most efficient solution is. An example is selection between a law with
strict notice or limitation periods and a law with more generous terms. A
challenging scenario is where a tribunal knows that a buyer and seller come from
countries with very different rules as to the need for prompt notice to claim for
late delivery. The claims will be valid under one national law but not the other. It
is too easy to feel sympathy for an otherwise deserving claimant and opt for the
laws with less restrictive claim procedures, when timeliness, efficiency, cash flow
and business planning may raise at least tenable arguments in favour of the
alternative approach. These reasons were no doubt the cause of those legal
systems opting for strict and shorter periods. This is not to say that one view is
to be preferred, but simply that a consideration of the most appropriate law can
at times require a more nuanced analysis. In other situations, this approach
would be even more contentious. In some cases a choice may open up the
potential for claims that otherwise could not have been made. For example,
application of the UNIDROIT Principles as lex mercatoria would allow for claims
for lack of good faith in pre-contract negotiations, which may not readily apply if
a common law system was selected. The discretion may be particularly
problematic when an arbitrator needs to page "999" consider the application of a
legal system allowing for punitive damages. (82) Another scenario is where
various legal families have different views as to whether they recognise trusts.
This may have impact where a tribunal is exercising discretion as to applicable
law when breach of trust is claimed. However, the tribunal might limit itself to
identifying the applicable law of the contract itself.

Where a State is involved, important issues will be limitations of liability,

sovereign immunity and agency principles. There is even the potential for the
relevant State to alter a law retrospectively.

13.5.5. Timing of Choice of Law

Because applicable law will commonly define the issues, at times allocate burden
and define standard of proof and suggest key aspects of legal research and
evidentiary questions, it is desirable that the parties know the applicable law at
the earliest opportunity. Where there is a dispute on this issue, a tribunal will
commonly consider dealing with this by way of a preliminary award or order.

While the desirability for an early determination always remains, it may not be
appropriate in some circumstances at least. It simply may not be possible where
key factual determinations need to be made to resolve applicable law that will
only best be analysed at a full hearing. An example may be application of a
closest connection test under conflicts principles, where there is a dispute as to
which of a range of contested discussions have led to what if any contracts and
what contractual obligations. A tribunal might feel a need to hear all of the
relevant evidence before making any determination as to applicable law, after
which it would then be fully armed with the evidence needed for a decision on
the merits. Even in such circumstances, a tribunal may wish to give some
guidance as to the way it expects applicable law or alternative applicable laws to
be addressed to promote efficiency without in any way prejudging the applicable
law question. A tribunal might ask the parties to indicate which legal issues are
common to potentially applicable systems, which differ and why and make
particular submissions as to the latter.

The later the choice the more problems that are likely to be faced but there is
even a significant timing problem in all cases where the decision is left to an
arbitrator. There is a fundamental conceptual distinction between selection of
law by an international arbitrator as compared to a national judge that goes
against the predictability and certainty that international commerce aspires to.
On the one hand, provision of a broad discretion as to selection of law by a
tribunal is obviously afforded to try and promote justice on a case-by-case basis.
Yet that very discretion means that in a practical sense at least, contracts subject
to arbitration agreements that do not themselves select an applicable law cannot
be said to have a knowable and particular applicable law from the time that they
were negotiated up page "1000" until the time of a dispute. The situation is
different to cases before domestic judges as the latter must at least follow
domestic conflicts norms. Even if the results of a future determination by a
domestic judge may be uncertain, the methodology is known at the time of
contracting and can be predicted. The differing situation in arbitration has a
number of problems, even though the residual discretion should on balance be
preferable. When seeking to perform ambiguous parts of the contract, parties are
less able to clearly identify what their respective rights and obligations are.
Where a dispute arises, parties are less able to predict the likely outcome, hence
making settlement more difficult. As noted, in many cases, an arbitrator
exercising discretion will know that the choice made will impact significantly on
the respective rights and obligations as to liability or ambit of remedies or both.
Perhaps for these reasons, some lex arbitri provide a residual methodology such
as giving direction as to the particular conflicts rules to be employed. (83)

13.6. Transnational Rules and Lex Mercatoria

Where party autonomy is unconstrained, the parties are free to select

transnational principles or lex mercatoria to apply to their contract. (84) There are
two distinct scenarios. The first scenario is where parties might select such
principles through a general description in that regard or by selecting a discrete
document that aims to encapsulate such norms. Examples of the latter would
include the Vienna Convention on the Sale of Goods, the UNIDROIT principles, (85)
the Principles of European Contract Law or published scholarly lists of principles
of lex page "1001"mercatoria. (86) The other scenario is where a tribunal will
consider the application of such principles under its own broad discretion to
select applicable law. Here the first question is whether the lex arbitri or rules
allow the tribunal to do so and if so, in what circumstances such a choice should
be made. Transnational legal principles are also sometimes described as
‘Principles Common to Several Legal Systems', ‘International Commercial Law’,
or ‘Generally-Recognised Legal Principles’. (87) While much has been written on
this topic, the complexity is disproportionate to the relevance in practice.
Analysis of ICC cases show that lex mercatoria or similar principles have been
selected in well under 1% of ICC cases. (88) However, total numbers are not
insignificant. Reporting in 2000, Molineaux found that the UNIDROIT Principles
had been referred to in over 30 ICC cases. (89)

If a decision is made to apply such principles, the final question is to identify

exactly what these are. Section 13.6.3 below looks further at the content of the
lex mercatoria.

13.6.1. Arbitral Laws and Rules and Lex Mercatoria

As noted, the two scenarios are where the parties select the applicable law and
alternatively where the tribunal does so under a broad discretion. Party
autonomy is generally accepted, although limits noted in section 13.4.2 above
may again apply where the relevant provisions refer to ‘the’ law.

There may also be interpretational issues where party choice is ambiguous.

Where the parties use an expression such as ‘equitable’ or ‘transnational’ legal
principles in an agreement on applicable law there may first be a question of the
tribunal interpreting the expression and seeking to best give effect to the intent
behind the parties' choice. In some cases the parties might direct the tribunal to
decide in equity or apply equitable principles. Here there is some uncertainty as
to whether they are opting for an element of lex mercatoria or are instead
seeking to empower a tribunal to decide as amiable compositeur. Arbitrators may
then vary as to the extraneous evidence that they would employ in undertaking
such a determination. The interpretational questions might be circular. For
example, if the page "1002" choice was expressed in the arbitration agreement
itself, there would be a question as to the methodology of interpreting that clause
and the law applicable to the clause itself. (90)

Similarly, if a tribunal operates under lex arbitri or rules that require selection of
the applicable ‘law’, some argue that this requires choice of a national system
and hence is a barrier to the application of lex mercatoria absent agreement by
the parties. Conversely, Article 28 of the UNCITRAL Model Law, being narrowly
drafted, is still seen as being drafted with sufficient breadth to allow the possible
application of anational rules. (91) French courts have also supported a broad use
of lex mercatoria under ICC discretions. (92) Reference to ‘rules of law’ in the lex
arbitri have consistently been seen by commentators and tribunals as being
broad enough to encompass transnational principles. (93) The revised UNCITRAL
Rules 2010 do not appear to wish to permit a tribunal to apply lex mercatoria
absent agreement of the parties. (94)

A tribunal considering whether to apply lex mercatoria may also turn its mind to
issues in enforceability. (95) The International Law Association Committee on
International Commercial Arbitration resolved that arbitrations based on
transnational rules would be enforceable if the parties agree that the arbitrator
may apply such rules or if the parties were silent as to applicable law. (96) Given
that there is no appellate review of arbitrations and annulment and enforcement
would not naturally apply to simple misapplication of lex mercatoria, a tribunal
has a very significant and potentially unconstrained discretion in identifying the
principles. (97) The more the norms are found within traditionally accepted
sources such as the UNIDROIT Principles or the CISG, the less there could be any
concerns with the tribunal's determination. Even if the tribunal adopted a more
expansive view, this would not be likely to ground any challenges. In Norsolor,
the Vienna Court of page "1003" Appeal did not accept the application of lex
mercatoria, although this decision was reversed by the Supreme Court of Austria.
(98)

Some tribunals have thus selected lex mercatoria or a similar model rather than
select a national system of law. (99) Those in favour of this approach suggest that
it is more consistent with the international and delocalised spirit of international
business to utilise emerging accepted norms of transnational law rather than
parochial legal systems, particularly where the latter would not support the
original intent of the bargain. (100) They might further argue that such
transnational principles, which attempt to identify the common kernel of
different legal families or adopt reasonable compromises in the face of
differences, provide a fairer balance and true equality between the parties. In ICC
Case 8385, Yves Derains argued that it is more important to meet the rightful
expectations of the parties than to adopt a conflicts methodology and that
international principles and lex mercatoria are to be preferred in many instances.
(101)

Poudret and Besson are nevertheless much more sanguine about whether lex
mercatoria should ever apply absent party agreement. The authors suggest that
this should only be so if it corresponded to the expectations of the parties. (102)
Those opposing application without party concurrence question the assertions
about the content of lex mercatoria and raise concerns about uncertainty and
additional disputes where contentious rules of law are sought to apply in
commercial contexts. There is also a danger of subjectivity and lack of rigour if
there is too ready resort to the application of lex mercatoria. Others consider that
the express use of such principles may not add much of significance. Keith Highet
has suggested that lex mercatoria is merely ‘a quasi recognition of the rules of
common sense, equity, and reasonableness that probably would have been used
even in the absence of any page "1004" reference to lex mercatoria’. (103) Others
even criticise the notion of lex mercatoria in the sense of its alleged cultural bias
towards western world norms, hence potentially undermining one of its
supposed virtues, being that of neutrality. (104)

One possibility is to look for trends as to usage, although it is difficult to

inductively argue for use of lex mercatoria by reason of certain cases applying it,
as the arbitrators concerned will often be key scholarly advocates of the concept.
(105) A tribunal might support resort to transnational rules by concluding that the

negotiations showed an unwillingness to agree to any domestic law. (106) In other

cases, a tribunal might select the law to promote validity, under the logic that a
valid agreement is closer to the original intent of the parties. As noted above, in
Valenciana the French Cour de Cassation supported a choice of lex mercatoria by
a tribunal where not chosen by the parties, to uphold validity which would not
have been possible under French law where price of goods had not been
specified. (107) Resort to lex mercatoria in this way may again be problematic if
the aim is simply to avoid a restrictive domestic rule that might bar a claim in a
particular circumstance, as it will not always be reasonable in the context of the
bargain, although the use in Valenciana seems understandable. Examples of
contrary scenarios, as noted, would be where the parties' bargain may have
properly incorporated strict restrictions as found in national laws or trade
usages. Strict notice requirements in sale of goods laws could thus be treated
similarly to express construction notice obligations such as in Clause 20.1 of the
FIDIC-Red Book. (108) page "1005" The logic might be that the efficiency of
construction requires such prompt notice. Where sale of goods is concerned,
there would also be many factual scenarios where it is only reasonable to expect
such prompt notice, for example where a seller's ability to cure will dissipate
quickly over time.

Some tribunals will simply use transnational law to fill gaps in domestic law or
consider both concurrently to see if they each provide identical solutions. (109)
Where tribunals apply transnational principles to fill gaps in domestic law, this
may be problematic depending on whether the domestic law itself has gap filling
measures. If that is so and the domestic law otherwise applies, there seems no
automatic reason to ignore its particular gap filling measures. That would be
particularly so if the parties had selected that law. (110) Nevertheless, some
domestic rules expressly provide that lex mercatoria can be used for gap filling
purposes. (111)

13.6.2. Lex Mercatoria and Due Process

If a tribunal wishes to resort to lex mercatoria, there are likely to be important

aspects of due process involved, given the better view that there should be no
surprise and parties should know and be able to address the particular legal
principles sought to be applied. Because the content and ambit of lex mercatoria
principles can be open to debate, even where the parties have expressly
empowered the tribunal to employ such concepts, there should be a preliminary
stage or other means by which the norms are identified. In some cases, it would
be appropriate to have preliminary submissions and/or a hearing on this matter.
(112) There are a number of reasons why this would be optimal practice. The

parties might disagree with the tribunal's view as to the applicable content. In
such circumstances, the parties could conceivably vary their choice of law
agreement and bind the tribunal to their views as long as this does not cause
undue expense or inconvenience to the tribunal. Where the parties themselves
cannot agree on content but disagree with the tribunal, the latter is entitled to
determine what it sees as appropriate but can only benefit from submissions of
the parties.

13.6.3. Content of Lex Mercatoria

If lex mercatoria is to be applied, the key question is how the content is to be

determined. Jessup described transnational law as ‘(a)ll law which regulates
page "1006" actions or events that transcend national frontiers. Both public and
private international law are included, as are other rules which do not wholly fit
into such standard categories'. (113) Rivkin describes lex mercatoria as ‘(a)n
amalgam of most globally-accepted principles which govern international
commercial relations: public international law, certain uniform laws, general
principles of law, rules of international organisations, customs and usages of
international trade, standard form contracts, and arbitral case law’. (114)

When looking at these sources of lex mercatoria, problems readily arise. Looking
at standard form contracts and general conditions is required in any event if the
parties have already agreed to them. If not, the parties might be said to have
evidenced an intent to have different terms. The same can be so for INCOTERMS.
References to custom and usage ought only be to those that the parties are aware
of or ought to be aware of, otherwise it is not their custom. (115) Where general
principles are concerned, the more these can be discerned from a broad range of
national systems, the more that lex mercatoria will not evidence a differing
approach to those systems, hence resort to such norms will not add much.

One problem with any articulation of lex mercatoria where it might matter is that
the content usually combines principles that are truly accepted by all civilised
countries and other principles that are more aspirational in the context of the
proponents' thoughts about the needs of international commerce. This is the
case with the UNIDROIT principles, which acknowledge that their aim was not
only to codify common principles found through comparative analysis, but also
provide recommendations appropriate for international rules. The Introduction
indicates that the Principles include what are seen as the best solutions for
certain circumstances. Uncertainty as to the scope of lex mercatoria naturally
leads to criticisms. Lord Mustill provided a measured but strong critique of the
lex mercatoria concept. (116) He suggested that because of the diversities of
countries, a truly transnational law is unlikely to be found. Narrowing the ambit
to the relevant countries in the individual dispute can be problematic. Diversities
in standard form contracts and uncertainty as to tribunal behaviour add to the
practical difficulties. He also raised the philosophical objection as to whether it
can truly be seen as a legal system.

If the concept is to be employed for whatever reason, some methodology must be

found. This is certainly so where the parties have expressly selected such
principles. Redfern and Hunter describe the two main alternative means of
identifying the norms as the list method and the functional method. A list
approach would look at such respected sources as the UNIDROIT Principles and
the Lando Principles. A more inclusive and hence contentious list is that of
Professor Berger page "1007" and his Trans-Lex transnational law database. (117)
In support of broader inclusion, Professor Berger has made the important
observation that lex mercatoria cannot be static and closed, hence referring to
the ‘creeping codification’ of the norms. (118) One problem with a list approach is
that there are immediate differences when various lists are compared. In
evaluating the lists, Poudret and Besson suggest some are so general that they
would apply in any event; others do not provide an answer; while some would
lead to controversial solutions. (119) Certainly, if a principle is found in all
respected lists, it is more likely to be accepted as truly part of lex mercatoria.

This chapter does not aim to resolve the debate or advocate core concepts.
Instead it simply seeks to raise awareness as to the kinds of principles that have
been, or could be argued to apply while noting that many may be controversial.
Some of these principles include: (120)

• contracts should be enforced except where there are changed circumstances;

• pacta sunt servanda and rebus sic stantibus are both applicable;
• unfair and unconscionable contracts and contracts clauses may not be
enforced;
• liability may be applied as to behaviour in contractual negotiations, for
example under culpa in contrahendo;
• parties may be required to negotiate in good faith to overcome unforeseen
problems;
• transactions should be performed in good faith;
• contracts with illegal objects or subject to bribes, corruption or dishonesty
may be void or unenforceable;
• a party is not entitled to rely on its own behaviour which causes
 nonperformance by the other;
• a State entity should not rely on its own procedural formalities to avoid
liability;
• damages may be limited to foreseeable results;
• loss of profits is allowable;
• a party must take reasonable steps to mitigate its loss;
• contracts are to be interpreted on a presumption that they are effective. page
"1008"
• In a more restrained listing, Lord Mustill also alluded to the following norms:
• A State entity cannot be permitted to evade the enforcement of its obligations
by denying its own capacity to make a binding agreement to arbitrate, or by
asserting that the agreement is unenforceable for want of procedural
formalities to which the entity is subject.
• One party is entitled to treat itself as discharged from its obligations if the
other has committed a breach, but only if the breach is substantial.
• No party can be allowed by its own act to bring about a non-performance of a
condition precedent to its own obligation.
• A party must act promptly to enforce its rights, on pain of losing them by
waiver.
• Contracts should be construed according to the principle ut res magis valeat
quam pereat, (that the thing may have effect rather than be destroyed).
• Failure by one party to respond to a letter written to it by the other is
regarded as evidence of assent to its terms. (121)

If reference is also made to equitable principles of law in articulating the content

of lex mercatoria, a number can be identified. The ICJ in North Sea Continental
Shelf said ‘it is not a question of applying equity simply as a matter of abstract
justice, but of applying a rule of law which itself requires the application of
equitable principles’. (122) Examples would include estoppel and abuse of rights.
The ICJ considers that acquiescence and estoppel follow from fundamental
principles of good faith and equity. (123)

As can be seen from the extracts above, some key norms are found in each list
and would be expected to be included in most domestic legal systems. Other
principles are more contentious. Some of the more conservative lists may work
well in simple one-off private party arrangements even if they add little to
domestic norms, but may prove to be less helpful in investment and long-term
arrangements. For example, where investment disputes are concerned and the
challenge is to a governmental welfare reform, resort to supposed lex mercatoria
notions of pacta sunt servanda, rebus sic stantibus and good faith, together with
the duty to negotiate to overcome unforeseen difficulties, will not readily point
to an answer as they will pull in different directions. Similar problems apply in
trying to deal with complex distribution and agency scenarios or
build/operate/transfer contracts over long periods of time.

page "1009"
Gaillard argues against the list approach, preferring to see lex mercatoria as a
method of analysis. Fouchard also strongly supports the method approach. (124)
For Gaillard, this would constitute ‘a comparative law analysis which will enable
the arbitrators to apply the rule which is the most widely accepted, as opposed to
a rule which may be peculiar to a legal system or less widely recognised’. (125) He
suggests a three-step process, first giving the utmost attention to the parties'
stipulations. The second is to look at the widest range of comparative sources to
see whether the contentions of the parties are supported by widely accepted
rules. The third step, which is in essence a qualification of the second, is to note
that unanimous acceptance in all legal systems is not required. It is not clear that
the list approach and this method are inherently distinct. Gaillard's steps 2 and 3
are the means by which any arbitrator or scholar would naturally seek to devise
a list. Step 1 is the important qualification in all cases, that party autonomy will
always determine when or why lex mercatoria may be applied in arbitration.

It is also important to compare and contrast lex mercatoria with other notions.
Lex mercatoria needs to be distinguished from amiable composition, although in
practice there may be much overlap between the two. The first technical
difference is that the latter requires express party agreement. In that sense
application of lex mercatoria should not be used as a backdoor means to decide a
case as amiable compositeur or ex aequo et bono absent consent. (126) There is also
a conceptual difference in that lex mercatoria is applied as a recognised system of
law, while amiable composition allows some departure from the law. In terms of
sources, lex mercatoria requires identification of widespread acceptance, while
notions of equity under amiable composition could be that of the arbitrator alone.

13.7. Customs and Usages of Trade

Many arbitration laws and rule systems expressly call for arbitrators to take into
account the customs and usages of a particular trade when rendering awards.
(127) The relevance of customs and trade usages is enhanced because these often

develop ahead of the law. (128) Resort to trade usage is expressly referred to in
Article 28(4) of page "1010" the Model Law and in most rule systems. (129) The
English Arbitration Act 1996 did not expressly direct attention to trade usages,
as this was seen as falling within the designation of modern commercial laws in
any event. (130)

It is easy to call for a reference to trade usage, but harder to determine how it is
to apply over and above cases where it is an express or implied term of the
contract. If the parties have contractually designated resort to trade usages, this
is itself a term of the contract and should be interpreted in the normal way.
Where customs and usages apply for any reason, they need to be integrated with
the terms of the contract and the applicable law. If there is no conflict between
applicable norms, all will easily apply. Some scholars have debated their
respective hierarchy, but circumstances would need to be looked at on a case-by-
case basis as in many instances, usages could be seen as implied terms or as
matters supplanting express contract provisions. The preferable view is to see it
as a question of intent as the parties could easily have expressed the norms
directly or rejected their application. If they have selected rules which conflict
with their express contract norms, it is hard to be sure what was intended.

Where trade usages are concerned, key examples include the Uniform Customs
and Practice for Documentary Credits (131) and INCOTERMS. (132) While these may
be important usages, more typically, this will be so by reason of being expressly
incorporated into the contract. The same would be so with many standard form
contracts, such as FIDIC construction agreements. In the absence of an express
guidance in arbitral statutes or rules, trade usages and custom might also apply
in any event if a tribunal would conclude that these form implied terms of a
contract.

The approach to proof and interpretation may vary depending on whether

customs and usages are express or implied terms of the contract on the one
hand, or whether they are separate factors directed or empowered to be
considered under lex arbitri or arbitral rules on the other. If within the contract,
they would be analysed under the applicable contract interpretation rules
including principles by which terms may be implied. In the alternative scenario
where reliance is on the lex arbitri or rules, a tribunal would be less constrained,
although many may still wish to use the same methodology. The evidentiary
point is that while various rules allow or indeed direct reference to custom and
trade usage, whether there is such applicable usage and its content would need
to be proven by the party with the requisite burden on that issue. The approach
to proof may also vary in a practical sense depending on whether the particular
custom or usage is asserted to be a habitual practice within a trade or profession
that needs to be demonstrated as page "1011" opposed to a recognised and
clearly articulated legal provision such as INCOTERMS. Aspects of proof may also
vary depending on the expertise of the tribunal. Where at least one arbitrator is
an expert in such trade or custom, they may rely on their own knowledge for
such purposes. (133)

The relevance of trade usage may also depend on whether there are standard
form contracts in the particular trade. If so, they may aim to comprehensively
express the relative rights and obligations of typical parties. An example of such
a trade usage is the variation provisions in construction contracts, which might
otherwise be seen as situations of breach instead of expected possible
modifications. (134) Such contracts would also typically call for unanticipated costs
to be borne by the party commercially expected to be responsible. An example
would be allocating costs to the land owner where there are unanticipated
problems in preparing a building site. Some legal systems may also raise factual
questions as to the degree to which a party is aware of elements of standard
forms as this may determine applicability. For example, Article 2:104(1) PECL
requires the party invoking terms that were not individually negotiated to take
‘reasonable steps to bring them to the other party's attention before or when the
contract was concluded’. (135)

In the previous section it was noted that lex mercatoria needs to be distinguished
from other concepts such as amiable composition. In the context of this section, it
does not seem reasonable to equate trade usages with lex mercatoria, although in
some cases that may be the outcome. Lex mercatoria is broader, being general
principles accepted by the vast majority of nations, while trade usages are more
specific to a particular commercial group. While it is true that custom and trade
usage can relate to similar concepts as lex mercatoria, it is generally easier to
identify common usage norms in areas such as international banking and
finance, shipping and commodity trades. (136) Born suggests that requiring
attention to trade usages ‘merely provides for consideration of the practices,
expectations and economic contexts of particular types of business transactions.
This does not supersede applicable legal rules, but instead informs how they are
to be applied to a particular dispute’. (137) However, Taniguchi argues that an
arbitrator required or allowed to take into account trade usage may take into
account the rules of lex mercatoria. (138) page "1012" In ICC Case No. 5721 of
1990, a tribunal considered that the direction to consider trade usages allowed it
to refer to lex mercatoria.

13.8. Mandatory Substantive Law

It is a highly contentious question as to what a tribunal should do when one

party alleges that a particular substantive law must apply, regardless of the
collective wishes of the parties. (139) Even more contentious is whether a tribunal
should address such laws sua sponte. In each case, the argument in favour of
such an approach is that a tribunal cannot ignore a mandatory substantive law.
While there a number of definitions of mandatory rules, each may be contentious
as they may incorporate elements that are subject to debate. A mandatory law is
a law that the promulgating government wishes to apply regardless of the will of
the parties. According to Mayer:

a mandatory rule (loi de police in French) is an imperative provision of law which

must be applied in an international relationship irrespective of the law that
governs that relationship. (140)

The Rome Convention defines mandatory laws as those that ‘cannot be

derogated from by way of contract’. (141) Mandatory laws may apply to matters
such as general capacity, company law, anti-corruption laws, competition laws,
securities law and taxation.

The reason for the existence of mandatory rules is that sovereign States wish
certain rights and obligations to apply regardless of the wishes of private parties.
As such, from the State's perspective at least, whether the rule applies cannot be
affected by parties' choice of law or dispute resolution procedure or forum. The
rule is considered ‘mandatory’ from a conflict of laws perspective because the
State supplying the rule considers it this way. Another point to note at the outset
is that mandatory rules can have differing functions if applied within arbitration.
They can act to impose positive obligations that ground discrete claims within an
arbitration. An example would be a claim for breach of anti-trust or competition
laws. They can also, however, provide defences to a claim. A party might seek to
void the contract or certain obligations under it on the basis of an alleged
mandatory rule to that effect. (142) In the extreme, the mandatory rule may
purport to deny arbitral jurisdiction.
page "1013"

Commentary on this topic has proven to be somewhat difficult to reconcile, in

part because of the fact that many commentators mix two distinct, albeit related
questions. The two questions are what subject matter is appropriate for
arbitration, (generally described as arbitrability and discussed in section 3.2.8,
and what laws cannot be derogated from by way of contract or other agreement.
The latter question is the nub of a true mandatory law issue. The former deals
with matters appropriate for arbitration and some similar conceptual questions,
where the parties seek to agree on a dispute in that context. In the context of the
mandatory rules debate, attention was initially given to whether arbitrators
could even validly consider mandatory rule issues within the framework of
notions of ‘arbitrability’, in the narrowest sense of that term, given that
arbitrators are not expressly empowered by States to safeguard governmental
public policy concerns. As arbitrability has expanded, as evidenced through cases
such as Mitsubishi Motors Co v. Soler Chrysler Plymouth Inc. in the US, (143) and the
Mors v. Labinal case in France, (144) attention has been directed to whether and
when arbitrators should apply mandatory rules.

For a range of reasons, determining the appropriate role of mandatory

substantive rules in international commercial arbitration is complex. First, like
many difficulties that arise in international arbitration theory and practice, it
raises conflicts between the promotion of party autonomy and the protection of
State interests. Given that arbitration is based on consent, an analysis of the
relevance of mandatory rules must consider the hypothesis that they are
incompatible concepts. Conversely, can arbitration ignore the mandates of States
with a sufficient nexus to the transaction under consideration? Second, because
there are a range of States that may have a legal and/or commercial interest in
any international dispute, there are questions (both policy and legal) as to which
State laws should have greatest influence. Third, there are questions as to the
nature of particular mandatory rules that may claim to apply – should certain
kinds of governmental policy norms be more readily applied by international
arbitrators than others? According to what criteria? And finally, what role and
weight should be ascribed to the wishes of the parties to the dispute? For
example, should the parties' wishes add to or detract from the application of
mandatory rules by arbitrators, and by what evidence or inference should their
attitudes be determined?

There have been important contributions to the analysis by a range of

commentators, but they have had different approaches and reached different
conclusions. (145) Some studies have been inductive, looking at a range of arbitral
page "1014" awards and trying to determine the commonly applied pre-
conditions to the application of mandatory rules. (146) Some seek to argue the
importance of a particular factor, such as notions of transnational public policy.
(147) Supporters of this approach argue that it will promote certainty and provide

justifiable limits to the circumstances in which an arbitrator can ignore parties'

consent. Challengers to that view argue that there are other circumstances where
fairness and efficiency ought to allow for the application of mandatory laws and
further, that the concept of transnational public policy is itself uncertain to a
degree. Others propose a conflict of laws methodology. (148) Advocates of this
position find justification for transplanting certain domestic conflicts principles
pertaining to the application of foreign mandatory laws, or support a rigorous
conflicts analysis flowing from choice of law powers within the lex arbitri as a
key means of resolving most legal challenges in arbitration. Others have sought
to argue for a broad discretion, at times with a proposal for a structured
approach. (149) Some relevant factors include closeness of the law to the contract
in issue, the implications of application and non-application and the general
policy aims and worthiness of the mandatory rule involved. Critics of this
approach argue that there is significant uncertainty in broad discretions and no
likelihood that there will be consistent approaches by tribunals. Others have
sought to rank some of the discretionary factors in order of relative importance
as an alternative guide to the exercise of a broad discretion. (150)

page "1015"

In many cases dealing with mandatory laws, the relevant rules argued neither
emanate from either the lex arbitri or the lex contractus. (151) Hence, there is a
debate as to whether there are ever any substantive laws that can truly be seen
as mandatory in the sense that they are required to be applied contrary to the
wishes of the parties. Some authors argue that this should never or almost never
be the case. (152) The majority of scholars, however, see the possibility that
mandatory substantive laws may at times be applicable. One reason why all of
the theories are open to debate (of course including any postulated in this book),
is that there are a significant number of key variables that could apply in any
given situation. The first variable is to consider the source of the alleged
mandatory law. This could emanate from the Seat, the place or places of
performance, one or more of the places of possible enforcement, the home
country of the respondent, the home country of the claimant, or some
international rule or elements of transnational public policy. A second set of
variables relates to the type of law being considered. The broadest distinction is
between procedural and substantive rules. Where procedure is concerned, this
could also be further divided into matters that go to the heart of arbitral
jurisdiction and those which merely go to the conduct of proceedings. Mandatory
procedural laws were considered in section 3.7.

Where substantive issues are concerned, one could identify particular topic
areas, such as competition law or environmental law, or instead consider the
broad features of different groupings, such as laws which criminalise certain
forms of behaviour, civil laws which aim to protect third parties, laws which
simply aim to protect a weaker party in a bargain and laws which aim to support
certain social and political goals. Finally, given the importance of consent as the
foundation of arbitration, there may be a need to integrate the first two variables
into a range of permutations of consent. This could range from situations where
the parties have clearly chosen the mandatory law, (in which case it is irrelevant
whether it is mandatory or not); situations where it is not clear on the evidence
whether they have selected this law or not (e.g., via selection of a country's
contract law without indicating whether other protective laws of that
jurisdiction are to apply); situations where the tribunal has a discretion to select
the substantive laws applicable; situations where the parties have chosen a law
of a jurisdiction other than that of the mandatory law to apply but have said
nothing about the latter; and finally, page "1016" situations where the parties
have expressly excluded the application of the mandatory law which a tribunal
might otherwise consider applicable.

This book argues that whether and when mandatory substantive rules should be
applied are ultimately matters of discretion with a diverse range of factors being
relevant. However, it argues against leaving the conflict of laws methodology at
that point and simply relying on the goodwill, intelligence and experience of
arbitrators to determine what factors should be most relevant in a given case. At
the very least, one ought to be able to articulate the objectives underlying any
discretion in order to justify and aid its employment. In addition to the need to
identify a justifiable policy basis, the more that legitimate guidance can be given
in a general sense, the more likely there is to be consistency, certainty, and
respect for arbitral determinations on this complex issue. This is particularly
important in light of States' sensitivities surrounding the application of their
mandatory rules, and the desire to maintain the broad notions of arbitrability
that have emerged over the last fifty years.

13.8.1. Mandatory Law as a Mere Factual Issue Relating to Non-performance

Mandatory rules questions can arise in a number of ways, only some of which are
contentious and require careful study. The primary concern is with the question
of when and why an arbitrator should apply mandatory rules. This includes
mandatory rules that limit the arbitrability of the subject matter or the capacity
of a party to enter into the contract or an arbitration agreement, or which ground
certain remedies, such as for breach of competition laws. The concern is not with
rules, such as trade embargoes, that can prevent performance of a contract and
which may thereby constitute a force majeure or frustration event. The question
in such cases is what the relevant rules of force majeure or frustration are, and
whether they have been satisfied. The question is not whether the arbitral
tribunal should apply the mandatory rule in question. (153)

13.8.2. Mandatory Rules and the Jurisdiction of an Arbitral Tribunal

Before dealing further with the policy bases for an arbitrator's power to apply
mandatory rules, it is first desirable to distinguish this from broader questions of
arbitral jurisdiction, although they may overlap. Jurisdiction concerns the power
of an arbitrator to determine an issue in dispute. For example, a clear
jurisdictional issue arises when there is a prohibition by the parties against the
application of particular mandatory rules such as general consumer protection
page "1017" provisions aiming to protect third parties. An important question
that arises in this context is whether an arbitrator has the power, or jurisdiction,
to apply the rule. What possible jurisdictional basis can an arbitrator have for
acting contrary to the parties' express consent? What if the prohibition comes
not from the parties, but instead from a mandatory law of the Seat? Given that
the tribunal derives its prima facie jurisdiction via the lex arbitri of the Seat, how
could the prohibition be ignored?
Like many arbitral questions, the answers are coloured by our broad views about
arbitration per se. This question was often situated within broader debates
about delocalised or a-national arbitration. For example, the extreme
contractualist position would only apply mandatory rules where it is consistent
with consent, for example where they form part of the lex contractus. Conversely,
the jurisdiction-alist perspective would be much more sympathetic to the need
to respect at least some governmental regulations purporting to have mandatory
effect. If as many believe, party autonomy does not operate in a vacuum but gains
some support from the recognition of such autonomy by domestic laws, it is
conceivable that at least some of these domestic laws would not recognise a
contrary agreement by private parties. (154) In the first scenario, where the
mandatory rule purports to deny jurisdiction, the tribunal seeks to determine
whether it applies so as to negate the parties' consent to arbitrate. If it is a law
other than that of the Seat, under principles of competence, the tribunal can
consider the question, although the State whose mandatory law is involved,
might not respect a positive determination. In the second scenario, where the
parties are permitted to arbitrate but have expressly sought to bar the
application of a mandatory law to the determination of their mutual rights and
obligations, there are two possible outcomes for such analysis. On the one hand,
one could argue that notwithstanding the possible invalidity of the parties'
agreement, their intentions are clear and an arbitrator should not accept
appointment if it involves running the case contrary to the parties' wishes. (155)
The other view would be that the parties have generally consented in good faith
to arbitral jurisdiction. Implied in that is consent to accept valid directions
whether they are happy with them at the time they are made or not. The
arbitrator accepts the consent to arbitrate but severs the improper constraint on
it. An arbitrator might be more comforted in taking the latter approach where
alternative dispute resolution fora would inevitably apply the mandatory rules in
issue. If everyone would apply such laws and the parties expressly preferred
arbitration, then accepting jurisdiction and then applying the laws would be
closer to the ‘legitimate’ expectations of such parties, and could thus be
articulated under an implied initial consent analysis.

page "1018"

The latter approach may gain support from the fact that conflicts systems
commonly accept that mandatory rules may override an express choice of law or
the law that would otherwise be determined under the basic conflicts principles.
(156) This is also accepted within some substantive systems that may be chosen by

the parties. An example is Article 1.4 of the UNIDROIT Principles which states
that ‘(n)othing in these Principles shall restrict the application of mandatory
rules, whether of national, international or supranational origin, which are
applicable in accordance with the relevant rules of private international law’.
Another way to resolve the issue is to deal with it in an indirect way under the
general approach to determining applicable law. In this sense, determining the
proper role for mandatory rules would simply be resolved by identifying a
conflict of laws theory for all international commercial arbitration issues. For
example, Born sees it as a two-step process; first, were the mandatory rules
intended to apply to such circumstances and; second, do the relevant conflicts
rules allow this. (157) If it is a conflicts issue, the next question is whether
domestic conflicts concepts and solutions should apply to international
commercial arbitration. (158) Mayer points to the respect for mandatory rules of
the lex fori as a natural aspect of domestic conflicts that does not readily apply to
international commercial arbitration. This is because an arbitrator has no
natural forum, only that selected and this is often selected for distinct reasons. A
contrary conclusion would be based on a jurisdictional view that sees the Seat as
paramount in terms of its rules circumscribing the arbitration. Even here, the
broad conflicts discretions in most lex arbitri are arguably different to domestic
conflicts rules and in any event, a direct choice method does not say how
mandatory laws could ever apply contrary to the parties' intent. A further
approach is to see mandatory rule questions as a subset of a broad group of
public policy challenges that are growing in arbitration, because of broader
categories of arbitrability, the growth of investment arbitration and the
emergence of public policy challenges to enforcement. This approach can still be
explained under consent theory or conflicts methodology, but might also look to
solutions consistent with other public policy challenges and inroads into
arbitration.

Valuable though such comparisons might be, some unique aspects of the
mandatory rules challenge should be noted. The significant difference where
mandatory rules are concerned is that the party-appointed arbitrator is at times
asked to apply laws contrary to the parties' apparent choice. It is not always
simply the independent State court system interfering with the choices
purportedly made within the arbitration process, which would be naturally
permitted under most page "1019" constitutional systems. It is also conceptually
different to mere non-enforcement where the New York Convention is
concerned, as the domestic court's application of public policy exemptions is part
of a treaty mandate. (159) Thus, the significant difference when we consider the
rights and obligations of arbitrators to consider and apply some form of
mandatory rule is that the person or persons whose jurisdiction only arises from
clear consent of the parties, is nevertheless allowed to consider ignoring at least
some aspects of that consent for some other policy reason, when they are not
generally empowered as a State officer to pursue these other policy directions.
Defining the proper basis for that determination is the challenge. That involves
due consideration of the restrictive argument that arbitrators should never go
against consent except where the latter is an abuse of a truly transnational norm
that any civilised system should apply.

In addition to the conceptual questions there are significant practical ones.

Mayer makes the practical point that mandatory rule defences can be raised as
spurious delaying tactics. If arbitrators cannot respond to these and distinguish
between meritorious and unmeritorious claims, arbitration will be adversely
affected. (160)

13.8.3. Broad Policy Arguments Supporting the Application of Mandatory

Rules

This section briefly analyses some of the existing arguments that have been
postulated in support of the application of mandatory rules. Some seek to justify
their application on the basis of a duty to maintain respect for arbitration. The
argument would be that if arbitrators are unwilling to show sufficient respect for
mandatory rules, this would undermine State support for such processes. (161) A
related argument is that if mandatory rules were unlikely to ever be applied, it
would encourage greater use of arbitration merely as a means to avoid legitimate
State public policy interests. This could undermine State respect for arbitration.
Such a negative reaction by States towards arbitration is unlikely to be prevalent.
It might be more likely with States who are new to supporting arbitration, but it
would be hard to envisage key arbitration fora pulling back from arbitral support
even in the face of consistently negative determinations. At most, such States
might be encouraged to express their mandatory rules more strongly and clearly
identify where arbitration is not possible or where it must support certain
mandatory rules. There may also be situations where different States have very
different views as to whether a particular mandatory law should apply. A typical
example would be in an page "1020" investment dispute where the host State
has a mandatory rule that purports to restrict the foreign investor's rights. The
investor's home State might well support the investor's view on the issue. Thus,
while the duty to maintain respect for arbitration may be a relevant
consideration, it cannot be determinative and would rarely even give clear
guidance on this issue.

An important obligation commonly referred to as having some bearing on the

debate is the arbitrator's duty to render an enforceable award. Yet this does not
give any a priori indication of whether arbitrators should be in favour of or
against the application of mandatory rules. Wrongly applying purported
mandatory rules would be as much a problem as wrongly not applying them.
There can also be conflicts between mandatory rules. Concerns with
enforceability raise a distinction between mandatory rules of the enforcement
country on the one hand and of the Seat or other jurisdictions on the other if
these do not all lead to the same outcome. There are thus conceptual and
practical problems in relying on this approach to fully guide discretions. Another
approach is to consider the reasons why international public policy may or may
not support recognition or enforcement under the New York Convention. Thus, it
is not enforceability per se that matters, but instead the presence of legitimate
grounds for preventing enforcement that should matter to an arbitrator.
Sheppard (162) suggests in that regard ‘that it is only in those situations where the
dispositive aspect of the award requires the doing of some act which is
unequivocally prohibitive in the forum State that recognition/enforcement
should be refused’.

The key issue is to consider how mandatory law questions can be validly
integrated into a theory of consent. The more one sees arbitration as merely
flowing from the consent of the parties, the less one would see scope for
application of mandatory rules, at least where that would be contrary to the
wishes of both parties. While it is normal to talk about party autonomy versus
the application of mandatory rules, there are in reality a range of scenarios
where different notions of consent could be examined separately. The two
extremes cause no problems. If both parties expressly want a mandatory rule to
apply, then an arbitrator should do so and in that sense its mandatory intent is
irrelevant. The only potential problem would be if an enforcement court would
consider the question non-arbitrable. That would rarely be so given the
broadening of notions of arbitrability and given the express agreement to have
the law applied by the losing party. At the other extreme, if both parties
expressly forbid an arbitrator from applying a mandatory rule that ought to be
applied, the arbitrator could simply not accept the commission. The key question
is whether the arbitrator could instead accept limited jurisdiction or ignore the
parties' prohibition. An arbitrator who somehow felt a desire to apply
mandatory rules contrary to parties' express consent would need to identify a
legal justification. Such an arbitrator could not justify such an approach under a
theory of implied consent unless such consent was identifiable at the outset,
after which the parties decide page "1021" to expressly exclude the rules
concerned. An arbitrator might say that the latter decision is an ineffective abuse
of rights. The more universally accepted the mandatory rule, the more that this
may be so. But what if the parties said that the continuance of the arbitration is
conditional on the arbitrator not purporting to apply mandatory rules? Would an
arbitrator in such circumstances have to withdraw or could they choose to
accept the limited brief? The better view must be that no-one can be required to
arbitrate outside of the parameters of their agreement. (163) An arbitrator must
also consent to the brief. If the parties offer conditional jurisdiction and
competence, an arbitrator can accept or reject but should not be able to ignore
the condition. If the exclusion is serious enough, for example, leading to unjust
results for third parties, the arbitrator should refuse the commission on the
terms provided.

In some cases, party intentions are unclear. Mandatory rule contests often deal
with non-contractual rights and obligations or at least limitations on contractual
autonomy. Where parties have utilised a broadly drafted arbitration agreement
as is now recommended as best practice, and refer to ‘all disputes arising out of
or in connection with the contract’, arbitrators can conclude that tortious and
statutory claims were also within parties' consent. Even this view is overly
simplistic. It is one thing to interpret a broadly worded arbitration agreement to
allow for a range of private legal claims not restricted to contractual claims
arising after the contract and found within its four walls, and conversely,
determining that parties wished their relationship to be subject to statutory
protections of third-party rights that may at times limit the rights of both of the
parties to the agreement. At the extreme, some mandatory rules would deny
arbitral jurisdiction entirely, hence they oppose arbitral consent head-on.
Overall, considerations of implied consent are never likely to be truly
determinative although they should act as a useful filter. If an arbitrator can
rationalise the exercise of a discretion as being consistent with original implied
consent to good faith efficient solutions, then the application of a mandatory law
is less contentious. Nevertheless this does not fully resolve the conceptual
challenge as to when and why mandatory laws might be applicable contrary to
consent. Furthermore, many would see the implied consent argument as a
strained one.

Grigera Naón suggests that mandatory rules' applicability may be justified by

general principles of comparative private international law, perhaps even
forming international public policy. He suggests:

that, under certain circumstances, when vital matters affecting the public or
societal interests of States are required, certain international disputes may be
exclusively governed by substantive mandatory rules of national origin
unilaterally advancing their necessary application primarily on the basis of their
underlying policies and objectives or purposes, irrespective of the contrary page
"1022" will of the parties and of any other conflicting mandatory rules from
other fora also vying for application. (164)

He sees that as part of the inherent jurisdictional function of arbitrators

irrespective of the will of the parties. (165) Such an approach, having a strong link
to jurisdictional theories of arbitration, will not prove to have universal
approval. It also again draws attention to the nature of the rule and the
determination of when the matter is truly ‘vital’. Bernardini supports a tribunal's
power in this regard on the basis that states which grant arbitral authority are
concerned to ensure that international arbitration contributes to the
development of economic relations, in part through optimal dispute resolution.
(166) Nevertheless, he advocates a continuing search for cooperation with the

parties where there are differences in view rather than mere superimposing of a
contrary view by an arbitrator. This approach is supported by the importance of
an arbitrator seeking to assist the parties to continue their business relations.
Mayer argues that arbitrators must take into account ‘the general interests of the
community … ‘ (167) He also suggests that arbitrators need to maintain the
integrity of arbitration and ensure that it is not used as a conduit for illegal acts.
(168) Hanotiaou has suggested that as arbitrators are not guardians of State public

policy, they are not as concerned as judges with the observance of mandatory
rules. (169) Gaillard argues that an arbitrator need not apply mandatory rules per
medium of conflicts rules of the Seat which impose limits on validity of selected
laws. (170)

13.8.4. Broad Policy Arguments against the Application of Mandatory Rules

At the other extreme, some suggest that there is no real justification for the
application of mandatory rules that truly go against the consent of the parties.
Any actions of an arbitrator contrary to the mutually agreed position of the
parties would, the argument goes, be contrary to the nature and function of
arbitration. page "1023" As noted above, that view depends heavily on
understandings as to the notions of arbitration. Furthermore, it is at least
arguable that parties entering into arbitrations do so in good faith and do not
expect any determinations by them to be accepted by a tribunal if the aim is to
overturn or avoid laws which ought to apply. While such an argument is not
necessarily persuasive, it is at least tenable and helps point to the complexities in
the notion of consent per se as noted above.

13.8.5. Should the Application of Mandatory Rules Be Dependent on a

Request by a Party
The next question is whether arbitrators should apply mandatory rules on their
own volition or only where parties make such a request. The latter view is
supported on the basis that an arbitrator's jurisdiction only extends to issues
raised by the parties. The contrary argument is that broad substantive issues
would inevitably have been raised by the parties and it is only a question of
determining the applicable law pertaining to those issues. An arbitrator could
not apply a mandatory rule to raise a cause of action not raised by either party.
But if the parties have alleged certain legal rights, it ought to be the duty of the
arbitrator to apply the proper law to the claim as made, even if that law was not
the one argued by the parties. This notion is discussed more broadly below. At
this stage, note should be taken of the decision of the ECJ in Eco Suisse where
enforcement was successfully blocked when a tribunal did not apply European
competition law, even when not argued by the parties. (171)

13.8.6. Is There a Duty to Apply Mandatory Rules as Opposed to a Discretion?

A key question is whether arbitrators have a duty to apply mandatory rules or

whether they merely have discretion to do so. A whole range of sub-questions
can be identified. If it is a duty, is it a duty in all situations or only in some, for
example where parties have not selected the substantive law or where it is a
mandatory rule of the Seat? If a duty only arises in some instances, is an
arbitrator barred from applying mandatory rules in other cases or is there then
still discretion? Wherever there may be discretion, is it available in all cases? If
not, what circumstances lead to the discretion and what factors should the
arbitrator consider in exercising that discretion?

There may even be a need to consider what is meant by a reference to discretion

in such circumstances as it may not be the right word. Discretion means you may
or may not choose to act after proper consideration of relevant factors. It is to be
distinguished from cases where you must act in a particular way if certain factors
are present, and the issue is simply to determine whether those factors page
"1024" exist. The latter connotes a conditional duty. Mayer has argued that there
will be a positive duty to apply mandatory rules, at least when the following
three factors are all present:

(i) the rule is part of the lex contractus;

(ii) the parties have not expressly excluded the rule; and
(iii) one of the parties invokes the rule before the tribunal. (172)

Such an approach can be tied to express or implied consent, as the parties have
either expressly chosen the lex contractus or have chosen a procedural model
that gave the tribunal a discretion to do so. Mayer's observation suggests that
there should not be a blanket rule as to whether there is a duty to apply
mandatory rules as opposed to discretion, but instead, suggests the need to
consider the facts of the particular case. Yves Derains argued that a tribunal
should even apply a mandatory law contrary to the express wishes of parties
where refusal to do so would be contrary to ‘truly international public policy’.
(173) Hochstrasser suggests that mandatory rules should be applied where the
parties have chosen the applicable law for the sole purpose of circumventing
mandatory rules that would have otherwise applied; where the performance of a
contract is affected by that rule; and there is a close connection between
performance and the rules; and where enforcement would be doubtful or
unlikely if the award does not take into consideration mandatory rules of the
enforcement country. (174) Only the latter two theories would justify acting
contrary to the parties' express intent.

13.8.7. Which Mandatory Laws Should Be Applied?

If mandatory rules are to be applicable absent clear party consent in some

circumstances at least, there are then questions as to which rules from which
jurisdictions are relevant to the arbitrator, either within the consent paradigm
because of the likely wishes of the parties, or by reason of some broader policy
trade-offs within the Seat-based paradigm. The possible sources include the lex
arbitri, the proper law of the contract, the law of the enforcement country, a third
country's law that has some connection to the transaction, or some
supranational law that purports to apply in any event. It may also be important
to consider not only the range of States that may be involved, but also the
attitude of other States as to when and why such foreign mandatory rules ought
to be respected.

Mandatory laws of the Seat are argued to be particularly relevant. Appendix II

Article 6 of the ICC Rules 2012 stipulates that when the ICC International Court
of page "1025" Arbitration scrutinises arbitrations, ‘it considers, to the extent
practicable, the requirements of mandatory law at the place of arbitration’. The
New York Convention allows a discretion against enforcement where an award
has been set aside or suspended by a competent authority of the country in
which it was made. (175) The relevance of the New York Convention is complicated
by the fact that this is merely a discretionary provision. Yet a neutral Seat may
have no wish to apply mandatory substantive norms to transactions that have no
connection with the Seat other than through the arbitration. Some mandatory
laws emanate from international agreements. An example might be a United
Nations embargo. Another example is Article VIII 2(b)(1) of the IMF Convention
which renders unenforceable exchange contracts over a member's currency
contrary to the exchange control regulations of that member. (176) While
arbitrators are not officers of States subject to the IMF Agreement, the case law
of the Iran-US Claims Tribunal has repeatedly said this provision should be
applied by arbitrators. (177)

13.8.8. Categorising Mandatory Rules

In addition to considering the source of the relevant laws, there is then the
question of which type of law should be applied. The two questions are not
unrelated as the importance of a particular country's mandatory rules may vary
depending upon the issue involved. Where substantive rules are concerned, one
approach is to categorise them based on their policy goals. Even here there are
categorisation questions. Mandatory rules could be looked at on a topic-by-topic
basis such as competition or exchange control law, or collected into broader
categories. The latter might differentiate between economic rules, governmental
controls, political rules, health and environmental protections and protections in
the face of unequal bargaining power. (178) Not all would be equally applicable to
international commercial disputes. Most laws aiming to protect weaker parties
deal with consumer contracts, labour and family disputes. As such, they would
normally not form the subject matter of international commercial arbitration.
Some jurisdictions do have laws in this category dealing with commercial
matters, including protection of agents and exclusive distributors. Other
examples include maritime transport and investor protection.

page "1026"

Another differentiating feature is whether the rules in issue merely seek to

protect the rights of the parties themselves or more generally provide
protections for third parties and other societal interests. The latter should be a
strong factor in favour of concern for the application of mandatory rules as the
third parties have not consented to the variation and we would not wish to see
arbitration used to avoid third-party entitlements. The concern would be even
stronger if the parties expressly selected applicable laws to avoid such
mandatory rules and thereby adversely affect the rights of innocent third parties.
Rules might also be categorised on the basis of the extent to which they give
effect to internationally accepted norms of public policy and social regulation.
This should overlap strongly with the previous category looking to third-party
implications. In such circumstances it is again easier to suggest that arbitral
parties have no power to avoid such norms. Here there is a difference between
truly common norms of governmental concern on the one hand and subject areas
that are merely commonly part of governmental policy and particular options for
regulation. Competition policy is an area where most developed countries have a
policy response given that it is an accepted governmental requirement under
classic economic theory, but the nature of the responses differ greatly. An
arbitrator might find that a particular State's policy response does not have
international support even though the underlying problem is of concern to all.
(179)

13.8.9. Interpreting Mandatory Rules

There is also the question of the rule's purported ambit. General questions such
as the intended extra-territorial application of the law and the closeness of the
connection between a particular jurisdiction and the commercial transaction are
relevant. In relation to the extra-territorial application of the law, however,
Grigera Naón suggests that ‘international commercial arbitrators should be
careful not to advance limitations on the powers of national States to unilaterally
extend the extra-territorial application of their mandatory legal rules regarding
vital State interests that are firmly grounded in public international law’. (180)
Some mandatory laws will clearly aim at being applicable e.g. a ban on dealing
with the enemy in time of war. At other times it may be unclear as to the
intended application e.g. antitrust and effects overseas in respect of contractual
negotiations held in a range of countries. In some such cases, an arbitrator may
simply say that the suggested rule was not intended to apply to the particular
facts. In other cases, where such intention is clear, domestic courts have at times
held that the extraterritorial aspiration is excessive.

page "1027"

Questions also arise in relation to whether or not mandatory rules should be

interpreted strictly. According to Voser they should. (181) Blessing also says that
they should be interpreted narrowly in relation to scope as to persons and
situations. However, what if the promulgating country had a wide interpretation
in mind? What if a wide interpretation intent was shared by other countries?
Why should a potentially applicable law be presumptively read down when
considering its application? Grigera Naón also takes a different view. (182)

13.8.10. Structuring a Discretion

One argument against the application of mandatory rules is that this could lead
to great uncertainty if it would be dependent on the exercise of an unduly broad
discretion by arbitrators based on a vast range of admittedly relevant factors and
the need to consider their nuances on a case-by-case basis. More certain
outcomes would flow from a blanket denial of the application of mandatory
rules. Fouchard, Gaillard and Goldman argue that non-application reduces the
chance that arbitrators might exceed their jurisdiction and leave their awards
open to challenge. (183) The converse is also possible, given that arbitrators who
should apply mandatory rules but do not, may equally make their award
vulnerable.

Whatever one's view as to the desirability or otherwise of a broad discretion,

most would agree that the more it can be objectively structured, the more
equitable and efficient it will prove to be. An unguided discretion is likely to lead
to inconsistency, and give greater scope for decisions based on an arbitrator's
personal view as to the contractualist/jurisdictional debate or their personal
attitudes to the worthiness or otherwise of the mandatory rule in issue.
However, too rigid an attempt to objectify the challenge has its own problems.
This is particularly so where any discretionary power draws attention to a range
of relevant factors. In such circumstances it is commonly the case that the factors
may conflict. Too rigid a formula would be unlikely to accurately and efficiently
weigh the varying conflicting factors whose importance may vary on a case-by-
case basis.

In light of these concerns, attempts have been made to identify methodologies to

guide such a discretion. There are three key methods that have been suggested:
the Special Connection Method, the Legitimate Expectations Method and the
Transnational Public Policy method. Brief attention is given to each.

page "1028"

13.8.10.1. Special Connection Method

The Special Connection Method is based on conflict of laws methodologies

employed by courts to assess whether foreign mandatory rules should be
applied. Those methodologies have been adapted to suit international
arbitration. Those referred to most often in this regard are Article 7(1) of the
Convention on the Law Applicable to Contractual Obligations (‘Rome
Convention’) and Article 19 of the Swiss Private International Law Act (‘PILA’).
Both, essentially, determine the applicability of mandatory rules as a function of
their purpose. Article 7 Rome Convention states:

When applying under this Convention the law of a country, effect may be given to
the mandatory rules of the law of another country with which the situation has a
close connection, if and in so far as, under the law of the latter country, those
rules must be applied whatever the law applicable to the contract. In considering
whether to give effect to these mandatory rules, regard shall be had to their
nature and purpose and to the consequences of their application or non-
application.

Nothing in this Convention shall restrict the application of the rules of the law of
the forum in a situation where they are mandatory irrespective of the law
otherwise applicable to the contract. (184)

Article 19 PILA states:

1. If, pursuant to Swiss legal concepts, the legitimate and manifestly

preponderant interests of a party so require, a mandatory provision of a law
other than that designated by this Code may be taken into account if the
circumstances of the case are closely connected with the law.
2. In deciding whether such provision must be taken into account, its purpose is
to be considered as well as whether its application will result in an adequate
decision under Swiss concepts of law.

Examples of how these methodologies have been adapted for international

arbitration can be found in the recommendations of an ICC Working Group and
in the writings of Blessing (185) and page "1029" Berger. (186) The Working Group
of the ICC's Commission on Law and Commercial Practices in 1980 came up with
two alternative Draft Recommendations on the Law Applicable to International
Contracts:

Alternative 1:

Even when the arbitrator does not apply the law of a certain country as the law
governing the contract he may nevertheless give effect to mandatory rules of law
of that country if the contract or the parties have a close contact to that country
and if and in so far as under its law those rules must be applied whatever may be
the law applicable to the contract. On considering whether to give effect to these
mandatory rules, regard shall be had to their nature and purpose and to the
consequences of their application or non-application.

Alternative 2:
Even when the arbitrator does not apply the law of a certain country as the law
applicable to the contract he may nevertheless give effect to the mandatory rules
of the law of that country if the contract or parties have a close contact to the
country in question especially when the arbitral award is likely to be enforced
there, and if and in so far as under the law of that country those rules must be
applied whatever be the law applicable to the contract.

13.8.10.2. Transnational Public Policy Method

An alternate methodology advocated by Fouchard, Gaillard and Goldman is the

Transnational Public Policy Method. This involves applying mandatory rules only
when they:

– form part of the lex contractus;

– form part of the lex arbitri; or
– reflect ‘truly international’ or ‘transnational’ public policy. (187)
page "1030"

Contrary to the Special Connection Method, arbitrators are given very limited
discretion as to which mandatory rules should be applied. This, according to
those authors, is its primary benefit. The lack of discretion makes more
consistent and predictable outcomes more likely:

[the Transnational Public Policy Method] avoids both the subjectivism of leaving
the arbitrators free to apply only the requirements of their own sense of justice,
and the permissiveness of having no public policy reaction at all. (188)

There also appears to be significant support for this approach in practice. After
examining a number of cases, Grigera Naón identified a trend to exclude the
application of a lois de police if it was not part of the expressly chosen proper law
unless its application is demanded by transnational public policy. (189) There are,
however, a number of difficulties with this method. Some of the difficulties with
this method, and with the general role of transnational public policy are the
evidentiary difficulties in determining the degree of acceptance required for a
principle to be seen as truly international. There is also the risk that one
arbitrator might consider something to be part of transnational public policy
while another might not. Commentators generally distinguish between
‘international’ public policy as applied in domestic conflicts systems on the one
hand, and ‘transnational’ or ‘truly international’ public policy on the other.
Transnational public policy is seen as ‘comprising fundamental rules of natural
law; principles of universal justice; jus cogens in public international law; and the
general principles of morality accepted by what are referred to as ‘civilised
nations”. (190) According to Professor Sanders, the concept is ‘confined to
violation of really fundamental conceptions of legal order in the country
concerned’. (191) According to Lalive, it includes policies which reflect ‘the
fundamental values, the basic ethical standards and the enduring moral
consensus of the international business community’. (192) Furthermore, the
important work of the ILA in seeking to limit the grounds of enforcement on the
basis of public policy, has argued strongly in favour of this norm being applicable
to transnational public policy alone. (193) Areas accepted to be within the scope
are prohibitions on corruption, drug trafficking, smuggling and export of cultural
goods subject to export bans. There is not, however, agreement as to where the
line should be drawn. For example, do mandatory rules protecting the
environment page "1031" form part of transnational public policy? What about
the principle of good faith? The Paris Court of Appeal has said that there is a
‘general principle of international public policy whereby contracts are to be
performed in good faith’. (194) Might this also form part of transnational public
policy? If so, there could, as noted above, be significant consequences for the
application of mandatory rules.

Grigera Naón criticises the general approach of limiting the application of

mandatory rules to situations where the rule reflects truly international public
policy on the basis that this goes beyond public international law notions
respecting sovereign legislative rights to promote public policy goals that would
not meet this test. He notes the problems of subjective judgment when seeking to
determine positively which rules of transnational public policy should apply.
Thus it is hard to use the concept to determine the category of lois de police that
deserve extraterritorial application. He considers that only in exceptional cases
will transnational public policy also dictate the application of certain lois de
police. He takes the view that the greater relevance of transnational public policy
is when it contradicts the particular lois de police and should hence operate in a
negative sense. (195) While that would be equally open to interpretive concerns,
he suggests that this would rarely be the case in commercial law. In his view the
proper application of lois de police, thus, depends on a functional balancing and
weighing of their policies, goals and purposes. (196) In other words, he prefers an
approach which then absorbs at least some of the features of the Special
Connection Method.

It is salutary to consider the Final ILA Report on Public Policy as a Bar to

Enforcement of International Arbitral Awards. (197) Valuable though that work is,
it poses problems if it was to be automatically transplanted into a basis for
application or non-application of mandatory laws. For example, being concerned
with enforcement, it is understandable that the Committee recommended
concern for the public policy of the enforcement country alone. Second, when
considering what forms part of international public policy, the Committee
combined some restrictive and broad elements. As to the latter, it included
antitrust laws, where these would differ significantly between countries. Finally,
the Committee understandably drew attention to a range of variables, such as
the connections between the rule and the case and the international consensus
or otherwise in favour of the public policy norms. The report does seek to
distinguish between mere mandatory rules and truly international public policy
and recommends a rule of reason test to the effect that ignoring mandatory
norms would be problematic where recognition and enforcement ‘would
manifestly disrupt the essential political, social or economic interests protected
by the rules’. (198)

page "1032"
13.8.11. Legitimate Expectations

Yves Derains analysed cases dealing with mandatory rules and sought to
inductively identify core principles. In particular, he argued that the key to
determining the application of public policy or mandatory rules to arbitration is
that of the parties' legitimate expectations. (199) He argued that the applicable law
is always determined by parties' legitimate expectations and not merely by their
will, (200) and that, therefore, the same approach should apply to mandatory laws.
Derains distinguishes four situations. First, whether the mandatory law is part of
the lex contractus and second, whether it is outside of that law. He further divides
those situations into cases where the parties have chosen the lex contractus and
those where the arbitrator makes the choice. He says that if the parties have
chosen the law, the issue is to be resolved by the wording of the choice, for
example, express contractual terms that are subject to a stipulated law will be
affected by the provisions of the latter. If instead, what the parties said was
simply to the effect that for matters not covered in this contract, [……] law will
apply …’, public policy under that default law should not render an express
provision void. Where the parties have not chosen a lex contractus, he argues
that, at a minimum, an arbitrator must respect the mandatory laws of the place
of performance. This is within the legitimate expectations of the parties because
if you do business in a particular country you expect to be subject to its
mandatory laws. (201) This broader criterion would include within it a range of
other relevant factors such as the relevance of mandatory laws of the lex
contractus, given that the latter is one part of the likely expectations of the
parties when selecting a choice of law.

If there is a clear choice to avoid a mandatory rule, Derains suggests that an

arbitrator should have recourse to transnational public policy theory. He seems
to suggest that if the mandatory law is part of transnational public policy, the
arbitrator can ignore the parties' will because he or she cannot be a party to
fraud against such laws. While the latter is a valid concern, it does not mean that
the arbitrator should simply apply the mandatory law. An arbitrator must find
jurisdiction. If the parties have expressly barred an arbitrator from applying a
law and the terms of reference are similarly limited, an arbitrator should refuse
to proceed if a proper application of principle suggests that continuing would be
such a fraud. This would be particularly so if continuing would lead to a decision
which simply page "1033" prefers one fraudster over another, rather than one
which protects legitimate third-party interests.

As with the other suggested methodologies, there are uncertainties and

interpretative questions. There are also evidentiary issues as to objective versus
subjective evidence of expectations and express or implied determinations.
There may also be a difference between expectations as to substantive law and
as to applicable conflicts principles or as to procedural norms. Nevertheless, the
theory builds on deference to the need for consent, but using the concept of
legitimacy to justify rejecting some unmeritorious choices.

13.8.12. Conclusions and Recommendations

The above analysis suggests that any attempt to present a rigid formula as to the
applicability of mandatory laws is fraught with danger. When looking at the
source of the relevant laws we have seen that these could emanate from the Seat,
the country of performance, the country of enforcement, the country of the
respondent or claimant or emanate from transnational public policy. As to the
types of rules, these could include procedural rules and a range of substantive
rules such as those barring the commercial actions of the parties, those seeking
to protect the weaker one of the parties, those seeking to protect the interests of
third parties not privy to the arbitral agreement or those seeking to promote
certain social and political goals. Because arbitration is ultimately a matter of
consent, the various sources and typologies need to also be looked at in terms of
the various levels of consent or otherwise to the application of the relevant rules.
These would range from laws clearly chosen by the parties; situations where the
choice made is debatable; to situations where the tribunal exercises a discretion
as to applicable law; to situations where the parties chose another law but were
silent as to the application of mandatory laws; to the extreme case where the
parties expressly chose to exclude the application of the mandatory law in issue.
That gives rise to well over a hundred permutations as to the interrelationship
between the above three classes of elements. Within each element there will
obviously be many further divisions based on nuances dependent on the facts of
particular cases.

The complexity alone provides much support for those who advocate a
simplified approach, whether through limitation of the application of mandatory
rules to situations where they could be described as part of transnational public
policy or otherwise. The practical cautions of Donovan and Greenawalt are also
an important gateway where they suggest that many supposed mandatory law
situations could be dealt with in other ways and that some conceptual arguments
in favour of mandatory laws are founded on a misreading of jurisprudence such
as that within the Mitsubishi case. (202) Part of the confusion is that the many
permutations page "1034" sometimes lead to the wrong question being asked.
For example, if an arbitrator has a broad discretion as to the applicable law and
believes that there is a sufficiently close connection with a deserving foreign
protective law, the application of that law might be justified on broad
discretionary grounds without having to show that it is mandatory. All that
needs to be shown is that it aims to cover this factual situation, and not that it
aims to apply regardless of the parties' wishes to the contrary.

Because all arbitral behaviour must ultimately be considered in the context of

consent, perhaps the best approach is to consider permutations where the
parties have purported to exclude the relevant law or chosen another law
without reference to the mandatory law. Where substantive laws are concerned,
there is much to be said for the view that laws which simply seek to protect one
of the parties as against the other ought only to be analysed under a consent and
choice of law approach. If the parties sought to exclude such a protective
measure, and that has no negative implications for third parties not privy to the
arbitration agreement, then absent fraud, an arbitrator should respect that
choice. An arbitrator should not seek to redress the commercial imbalance
between parties through anything other than laws which would normally be
applicable to the dispute. At the very least, the more one adopts the consent
paradigm of arbitration, the more one would be comfortable in accepting the
parties' wishes to avoid an internal protective law. If in hindsight, the weaker
party believed that there was the exercise of a dominant position, duress or an
abuse of rights, it could raise that claim in proceedings in any event.

The most significant category is where the alleged mandatory rule seeks to
protect innocent third parties, who at the very least, have not consented to any
commercial relationship purporting to avoid the laws' operation. The presence
or absence of negative externalities ought to be one of the pre-eminent factors
considered by a tribunal. Where laws deal with the protection of innocent third
parties much may depend on the facts of the instant case and here there are
again a range of permutations. The argument in favour of application increases if
the law emanates from the country of performance as parties do business in that
place in the expectation that it may have laws constraining their rights and
imposing certain obligations and liabilities on them. People who conduct actual
business within a geographical jurisdiction are presumed to have intended to be
subject to its laws. Arbitral proceedings should not be a mechanism by which
such laws can be readily avoided. If the parties have chosen that law then it is not
a mandatory law issue in any event. If the tribunal has a discretion, the rights of
the third parties may legitimately be a factor in exercising that discretion. If the
parties have chosen another law, this will commonly be limited to the law
applicable to the contract per se. A broadly drafted arbitration agreement will
allow for other claims in due course. In most circumstances, the potential
applicability of that law will simply flow from the facts as pleaded and the
closeness of that law to the facts. Where the parties have sought to exclude the
application of such a protective law to gain a benefit for themselves as against
innocent third parties, arbitrators should not readily allow this to occur.
Depending on one's approach to the notion of consent, an arbitrator should
either refuse to continue with the case, acknowledging that he page "1035" or
she does not have a mandate to act contrary to the parties' intent no matter the
circumstances, or might instead argue that consent in arbitration must always be
consent in good faith consistent with fair and reasonable arbitration as a dispute
resolution methodology. Such an arbitrator might conclude that the parties
clearly consented to arbitrate and the part of their consent that sought to
adversely affect the rights of innocent third parties can be severed from their
general consent. Neither view is inherently correct as a matter of logic, but
simply identifies different theories of arbitration that have been unavoidable
throughout its history.

One ICC case has held that ‘it is the strict duty of the tribunal to examine, even on
its own initiative, the validity in light of public policy … of the agreements that
are the basis for the claimant's rights. Another indirect mechanism would be
where a norm of international or transnational public policy undermines the
validity of the contract in dispute. An example would be an arbitration over a
contract obtained through bribery. (203) The tribunal must assure itself, as a first
matter and before examining the matters at issue, that those agreements are not
tainted by an absolute nullity because they are illegal and contrary to public
policy’. (204) This has been discussed throughout in the context of determining an
arbitrator's rights to override the intent of the parties, as some such situations
may be argued to be an abuse of the arbitral process for improper purposes. This
poses particular difficulties for an arbitrator whose mandate emanates from the
consent of the very parties whose goals may be questionable in such
circumstances. An example of an arbitrator deciding that a contract was null and
void for illegality was the decision of Judge Gunnar Lagergren in ICC Case 1110.
(205) Judge Lagergren considered that commissions were essentially bribes and

involved ‘gross violations of good morals and international public policy …’ (206) A
tribunal in WDF v. Kenya (207) applied principles of international public policy to
conclude that regardless of whether bribery was accepted under applicable
domestic law, it was a breach of international public policy.

Where the place of enforcement is distinct from the place of performance, there
are strong competing arguments. We speak of the duty to render an enforceable
award and at the very least a wish to make arbitration meaningful and effective,
hence making enforceability a key factor in its utility. However, there is simply
too much difference in the way different countries approach enforcement.
Furthermore, the fact that there may be many possible places of enforcement,
with clashes between them as to the approaches to mandatory laws and the
hoped for page "1036" situation where an award is complied with in any event,
all go against this as a key criterion. Nevertheless, it is impossible to ignore
recent European cases dealing with protective economic regulations such as
competition and agency law. The developments from Hilmarton to Accentuate
are significant. In Omnium de Traitement et de Valorisation S.A. v. Hilmarton Ltd,
(208) a French company engaged an English company to provide advice and

coordination in relation to a contract to be performed in Algiers that was

governed by Swiss law. Algerian law prohibited the use of intermediaries. The
arbitrator considered that the English company's behaviour was contrary to that
law and declared it void on that basis and on the basis of offending bonos mores
as articulated in Article 20(1) of the Swiss Code of Obligations. The decision was
overturned by the Court of Justice of the Canton of Geneva, that held that the use
of intermediaries was not contrary to Swiss law. The decision was upheld by the
Swiss Federal Tribunal. (209) In Eco Swiss v. Benetton the ECJ concluded that an
enforcement court may consider mandatory competition laws even where not
argued by the parties in an enforcement action. (210) Cases such as Eco Swiss have
thus put a significant onus on tribunals to consider competition law issues and
other matters that may constitute public policy norms to prevent grounds for
annulment or for blocking enforcement. Because the case determined that a
tribunal should consider such matters even when not raised by the parties, it
also has due process issues in terms of calling for a tribunal to give sufficient
notification to the parties that the issue is to be considered and allowing them an
opportunity to present their views on the matter. There are still key unresolved
questions. For example, what would the ECJ say if those parties directed the
tribunal not to deal with the competition law issue?

Where truly transnational public policy is concerned, while this again can be
justified theoretically on the basis that arbitration ought not to offend against
such principles, if the matters are truly transnational, then there ought to be
domestic laws in all relevant jurisdictions to similar effect. Hence, consideration
of this as a distinct category is to some degree problematic. For these reasons,
the transnational public policy test is not to be preferred as the sole factor,
although it is an important measure by which to consider both positive and
negative effects. It is likely to be most useful in cases where one of the countries
involved has laws that offend against transnational public policy and are
otherwise asserted to apply as mandatory norms.

Another issue is where there may be conflicts between different potentially

applicable mandatory laws. One possible approach would be to consider
independently the application of each individual mandatory law and then resolve
their conflict by determination of which has stronger claims to applicability. A
related approach is to resolve it by looking to whose mandatory law should
prevail based page "1037" on which country has greater claim in that regard. A
converse approach is to say that each must apply by reason of their mandatory
nature and let their results ‘net’ out. One problem would be if they could not both
apply simultaneously. For reasons noted above, many supposed situations of
conflict between mandatory rules could on further analysis be seen as not
leading to that result.

In view of the foregoing, it is suggested that the first question is what is the
evidence for or against consent to the application of the supposed mandatory
law. The more that the arbitrator is unconstrained, the more that a normal
conflicts methodology, utilising the factors suggested by Blessing and Berger are
consistent with fairness and efficiency. Conversely, the more that an arbitrator is
constrained, the more he or she must be concerned with the scenario that the
arbitral process could be used to aid in the avoidance of governmental laws in a
way that challenges the integrity of the arbitral process. The following is a list of
relevant sub-factors and/or questions that an arbitrator might rely upon to help
guide the ultimate exercise of any discretion based on these key questions. These
build on the suggestions of Blessing and Berger and are consistent with the
developments in national legal systems. The factors can be criticised on similar
bases to the above critiques of alternative theories. Without seeking to resolve
this debate at this stage, a book dealing with procedure and evidence should at
least draw attention to the relevant factors that might well be considered.

1. Is it a purportedly mandatory law? If not, it is simply part of the normal

conflicts analysis.
2. Is it intended to apply as of right? How clearly can it be seen that the
enacting State wishes the law to apply to this factual scenario? An accurate
interpretation of intent is to be preferred to an a priori narrow one.
3. Are its preconditions in the law satisfied? If not, it is inapplicable in any
event.
4. Are the parties or a party trying to include or exclude the mandatory law?
The application worthiness may vary depending on whether it is being used
in a positive or negative sense.
5. Are third-party interests involved? This is the strongest instance where a
Tribunal should not automatically look to follow the parties' preferences.
6. What is the nature of those third-party interests; how closely are they
connected to the transaction and how close is their link to the legislature
seeking to protect them?
7. What is the nature of the provisions and to what extent can they be
described as truly transnational public policy elements? The more
universally accepted the subject matter of the mandatory law, the more an
arbitrator should be prepared to apply it, although the more this is so, the
more they would be relevant under any applicable law.
8. Conversely, does the mandatory law conflict with any transnational public
policy?
9. What degree of closeness is there between the relevant rule and the contract
or investment? page "1038"
10. What is the effect of application or non-application? If the result of the case
would be the same in any event, there is no need to even consider the
mandatory rule issue.
11. Was the mandatory law enacted for a bona fide reason and is a party seeking
to invoke it for bona fide reasons?
12. Finally, arbitrators should consider the implications for enforcement in
terms of the exercise of a discretion in relation to mandatory laws, although
the place of performance will be more telling under the above criteria.

These questions are not provided as a rigid formula by which an automatic

outcome can be achieved. Instead they hopefully draw attention to the
relevant factors that at the outset should lead an arbitrator to consider
when, if ever, he or she is entitled to apply a law contrary to the parties'
intent or refuse the commission to arbitrate by reason of being asked to do
so.

13.9. Other Applicable Principles and Laws

13.9.1. Non-Contractual Claims

A broadly drafted arbitration clause will allow for non-contractual as well as

contractual claims. These may include pre-contractual claims such as culpa in
contrahendo, tortious claims, equitable and restitutional remedies and statutory
claims such as under competition law. Once again, in most cases party autonomy
will prevail but parties rarely designate a law applicable to non-contractual
claims. The plain meaning of most arbitral statutes is broad enough to allow for
party choice of such laws. A contrary view is presented by Fouchard, Gaillard and
Goldman (211) indicating that traditionally, private international law did not allow
for prior selection of law applicable to non-contractual claims. If there is a choice
of law clause, the important interpretation question is whether it is broad
enough to encompass these types of claims. Another factor is that many non-
contract provisions could be argued to be mandatory rules. An example would be
competition law norms as discussed in the previous sections.

If the parties have not made a choice as to law applicable to non-contractual

claims the situations may vary depending on whether they have chosen a
contract law. If so, some assert that the chosen system's conflict of law rules for
non-contractual claims should apply. This may appear elegant from the
perspective of harmony, but is more problematic if the choice of law clause
expressly excludes reference to conflict of law rules of the selected jurisdiction.
(212) One might interpret page "1039" the clause to merely mean exclusion of

conflict of law rules for contract, no doubt to avoid renvoi. Another approach is to
look at connecting factors on a case-by-case basis. Unfortunately, there is more
variance between legal families as to conflict of law rules for torts. Some look for
the place where the tort occurred. Where that is complex, some look for the
‘centre of gravity’ of the tort. (213)

13.9.2. Capacity

Questions of capacity may have their own special laws, such as those
determining corporate obligations and capacity to enter arbitration agreements.
Typically these will be taken from the home country of the legal person or
individual concerned.

13.9.3. Fairness and Equity

Some rules expressly include principles of fairness and reasonableness. (214) On

other occasions, parties might use equity clauses to direct the tribunal to adopt
such an approach. This raises similar principles as empowerment as amiable
compositeurs, although the meaning of the latter would be drawn from historical
attitudes to the concepts, while an equity clause may depend on the particular
drafting in the instant case. Tribunals will at times consider that various
equitable principles apply. The ICJ in North Sea Continental Shelf ‘said ‘it is not a
question of applying equity simply as a matter of abstract justice, but of applying
a rule of law which itself requires the application of equitable principles'. (215) As
noted, examples would include estoppel and abuse of rights. The ICJ considers
that acquiescence and estoppel follow from fundamental principles of good faith
and equity. (216)

Some legal systems allow for arguments based on changed circumstances. (217)
This is similar to hardship and renegotiation clauses and raises similar risk
allocation questions as arise with stabilisation clauses. This is discussed further
in section 14.4.6.2 dealing with remedies.

13.10. Changes in The Applicable Law

In some cases, there will be changes in the applicable law during the currency of
the arbitral proceedings. Small changes in the law would naturally be deemed to
page "1040" apply if the promulgating authority intended the laws to be
retroactive. Similarly, in common law jurisdictions, if there is a trend in judicial
attitudes that emerges prior to the determination on the merits, these ought to
be relevant and applicable.

Section 3.9.1 examined situations where parties select arbitral rules which are
then revised by the promulgating authority prior to the dispute. It was noted that
it is a question of intent as to whether the original or new rules will apply. The
parties can make their preferences clear if they wish. The same is so with
changes in substantive law. Parties could expressly indicate whether changes in
the law are or are not to apply. If parties are concerned about potential changes
in the law, they might utilise stabilisation clauses for such purposes. These are
discussed further in section 13.12. This would generally be supported in
commercial matters as it has little difference to express contract terms. The only
exception might be with regard to mandatory norms. The situation is more
problematic with investment arbitration as this would involve a purported
constraint on the impact of a State party's future legislative actions. Another
problematic scenario would be changes by a State in an investment dispute that
are self-serving or changes lobbied for by a private party with significant
political power in a weaker State, with a view to gaining a benefit over another
private party in arbitration. (218) Such scenarios would not only involve questions
of applicable law but would potentially raise other causes of action based on the
behaviour itself.

Where there is insufficient guidance from the parties in private commercial

disputes, the natural presumption would be that the modifications to the
selected law are to be applied where they are intended to apply retroactively to a
prior contract or prospectively to lead to new breaches under it. This is because
parties would naturally expect updates and improvements of law to be
applicable. The situation might be different where there is a total change in the
law contrary to the legitimate expectations of the parties at the outset. In that
circumstance it might be held that the prior law was intended to govern. (219)

13.11. Investment Treaty Law

Where investment disputes are concerned, legal provisions will typically be quite
different in nature to commercial disputes, bringing a host of related questions of
identification of laws, problems of proof and interpretation. This is of course a
whole topic in itself and this section merely aims to touch on some key features,
particularly those raising procedural and evidentiary challenges.

Sources of investment law can be broader, raising questions of proof, and

integration of conflicting norms. Sources are referred to in Article 42(1) of the
Washington Convention. The Convention draws attention to party autonomy, the
Convention, any relevant investment treaty and choice of law rules. page "1041"
The Washington Convention also indicates that if the parties have not selected
the applicable law, the tribunal will apply the law of the contracting State plus
such rules of international law as may be applicable. In Klockner v. Cameroon, an
annulment committee annulled an award because it failed to apply a host
country's law and instead presumed identity with French law. (220) In investment
treaty disputes it will thus often be the case that the parties adopt public
international law even though one party is a private entity that on one view is
not subject to it. The reference to public international law is accepted as
referring to the elements articulated in Article 38(1) of the Statute of the
International Court of Justice. (221) This deals with international conventions,
international custom and general principles of law recognised by civilised
nations. Subsidiary means of determining that law include judicial decisions and
the teachings of the most highly qualified publicists. As has been observed, the
subsidiary means do not create rules, but only assist in determining what the
rules are. (222)

Key norms under investment treaty protection include most favoured nation
obligations, national treatment, fair and equitable treatment, and expropriation
norms. This then raises important interpretation issues as to the meaning of such
concepts. The most challenging issue is to consider the ambit of exceptions.
Investment treaty disputes often involve consideration of the exempting
provisions dealing with emergency and police powers over security, health,
environment and public order. Treaty norms also need to be integrated into
broader notions of public international law, including sovereign and State rights
per se. Other principles would include estoppel and abuse of rights. (223) There
will be evidentiary issues in terms of identifying when there is sufficient
acceptance of a principle to describe it as a general principle of law. Views differ
as to the nature and extent of international law. Historically, some merely saw it
as the law applicable to the rights and obligations between nation States. More
recently it is seen as covering the whole spectrum of persons, including
questions of human rights. The point to the observation is that it is now more
commonplace to see private traders agree to the application of international law
in their transactions with State entities. Such legal principles could be said to
apply in any event under the more expansive view, or as noted above, could be
expressly referred to in applicable treaties. NGOs assert such application when
seeking amicus curiae status. (224)

Redfern and Hunter observe that public international law may apply but may be
inadequate to deal with many contractual style issues such as mistake,
misrepresentation, force majeure, time of performance and effect of bankruptcy
or page "1042" liquidation. (225) There is also the issue that one party, the State
has power to impact upon the transactional rights and obligations with new laws
or regulations. Rules of international law typically impose limits on what a State
can do from time to time to render its own conduct valid. Treaty norms are also
often poorly drafted. Too often a treaty is a record of an unresolved dispute
between negotiators that seeks to create the impression of a successful outcome.
There will also be some uncertainty in relation to the EU, with different
perspectives as to Member States' own rights to enter into BITs and the more
centralised treaty making norms since the Treaty of Lisbon in December 2009.
(226)

13.12. Stabilisation Clauses

Stabilisation or freezing clauses vary from those that involve promises not to
change the law by a State party, to others which simply say that the parties agree
that their respective rights and obligations are limited to those pertaining under
the law applicable at the time of the contract. Because such clauses are drafted
between the parties, it is natural that they can range in nature and validity,
whether intentionally or otherwise. From a commercial perspective, they are
neither more nor less than aspects of certainty and risk allocation, but they have
fundamental philosophical and political elements where they are used to legally
or practically act as disincentives to governmental, social and environmental
regulation. There will also be questions of enforceability on public policy
grounds, particularly where enforcement will need to occur in the host State.
Stabilisation issues are at their starkest when an investment or activity is
occurring in a country with a reputation for fundamental changes in law from
time to time, in part when new governments are in place and who are unwilling
to accept the entitlements granted by previous administrations.

The Texaco arbitration saw investors succeed in a claim for breach of a

stabilisation clause. The case has certainly proved to be contentious. Texaco (227)
is to be contrasted with the decision in Aminoil. (228) As was shown in the Aminoil
case, a tribunal might also hold that a stabilisation clause is itself subject to the
rebus sic stantibus principle, which could justify the change in government
legislation. (229) While stabilisation clauses have thus become highly politicised,
the same outcome page "1043" could be agreed by other clauses that provide
compensatory mechanisms when changes occur. Such clauses may readily be
found in long-term supply contracts. A stabilisation clause can never presume to
be directly effective, as the relevant State could outlaw the clause itself, although
that itself might be subject to international law limitations. The key issue
however, is in defining breach and applicable remedies.

A stabilisation clause can actually be detrimental to the party who proposed it in

circumstances where the change of law is beneficial from its perspective. The
two scenarios may not necessarily be treated identically. Contractual clauses
purporting to prevent a government's beneficial societal changes conflict with
notions of sovereignty and positive externalities for the wider community.
Conversely, where a government improves the law for all, including for the
foreign investor, it may be uncontroversial that it is happy to benefit all,
including the latter. Hence, it might even be arguable that there is an implied
term that it only operates as against unfavourable changes from the investor's
perspective. The drafting could clarify this issue either way.

13.13. Umbrella Clauses

There is often a problem in deciding how contract terms impact upon treaty and
other international law rights where foreign investment is involved. This can
even impact on jurisdiction. Umbrella clauses are at times utilised with a view to
making a breach of a contract provision constitute a breach of the treaty itself.
While these are substantive in that sense, they are essentially procedural,
allowing the contract claim to be brought within the treaty dispute resolution
mechanism.

Tribunals have varied in their approach to umbrella clauses. Cases adopting a

restrictive approach include Salini, (230) Joy Mining (231) and El Paso. (232) Examples
of expansive interpretation include Noble Ventures (233) and Siemens. (234) Some
purport to decide on the ambit of such clauses based on interpretation of the
treaty provisions. Others may adopt a bias for or against investors.
page "1044"

13.14. Amiable Composition or Ex Aequo Et Bono Determinations

13.14.1. Entitlement to Determine as Amiable Compositeur or Ex Aequo Et

Bono

Most arbitral statutes and rules allow the parties to provide a tribunal with the
power to decide as amiable compositeur or ex aequo et bono. (235) While the ambit
of the concepts is open to debate as is discussed below, they essentially allow a
decision other than via strict application of law.

The tribunal itself cannot adopt such an approach unilaterally without

authorisation. The ICSID Convention expressly indicates that a tribunal acting ex
aequo et bono without authority would be subject to annulment. (236) Such
powers depend on entitlement under the relevant arbitral statute or rules and a
selection to that effect by the parties, although some rules and statutes expressly
empower a tribunal to that end regardless of the wishes of the parties. (237)
Section 24(4) of the Sri Lankan Arbitration Act refers to ‘considerations of
general justice and fairness’. Article 56 of the Indonesian Arbitration Law
directly empowers the tribunal to decide ‘based upon justice and fairness’ as an
alternative to the relevant provisions of law. In each case there would be a
question as to whether the parties can proscribe the use of such principles by
agreement, or whether these are mandatory norms of the Seat. (238)

Conversely, some national arbitration laws have deleted reference to amiable

composition. (239) If the parties nevertheless selected rules allowing for such a
power or otherwise empower the tribunal expressly, that latter position should
prevail. A contrary view would be dependent on concluding that the variation of
the lex arbitri was intended as a mandatory procedural norm. This is debatable,
but unlikely, and less so as it can only be implied through the variation itself.
More typically, lex arbitri and institutional rules simply provide that the parties
can allow page "1045" the tribunal to act as amiable compositeur or ex aequo et
bono. In all such cases it is simply dependent on party consent. (240) While the
power is there in most circumstances, parties rarely empower tribunals to
decide as either amiable compositeur or ex aequo et bono. The ICC has noted
approximately three arbitrations per year between 2003 and 2008 where such a
power was awarded. (241)

Whether selected by the parties or empowered via applicable rules, it is then

necessary to consider the nature of the two concepts. While the principle of party
autonomy is clear, the ambit of the concepts is more open to debate. Other
problems arise where parties use other expressions that may or may not intend
to invigorate these concepts. A further issue arises where parties might also
expressly refer to amiable composition and some other system of law. Sigvard
Jarvin poses the following questions where the principles apply:

(a) If the arbitrator may deviate from rules of law in general, is he not bound by
mandatory rules of the place of arbitration and/or enforcement (public
 policy rules)?
(b) Can the arbitrator depart from the terms of the contract and trade usages?
(c) Does the power of amiable compositeur authorise the arbitrator to deviate
from procedural rules?
(d) Must the arbitrator give reasons in the award? (242)

Other questions include whether there is any difference between amiable

composition and deciding ex aequo et bono; whether other phrases used by
parties, such as deciding in equity are different in nature; should a tribunal first
consider how the otherwise applicable law would apply; and what is the
fundamental difference between the concepts on the one hand, and application
of good faith, lex merca-toria, or hardship provisions on the other? These
questions are addressed in the following sections.

13.14.2. Ambiguities in Party Selection

Unless otherwise indicated, amiable composition only applies to the merits and
not procedural matters. (243) In some cases parties simply refer to procedural
matters. (244) page "1046" In other cases the description covers both substantive
and procedural matters. (245) It is thus preferable to clearly indicate in the
drafting whether the power extends to procedural as well as substantive
matters. Where party agreement is required, it makes sense to ensure that there
is a sufficient evidentiary record. Particular care should be taken to confirm in
writing any discussion to this effect at a preliminary conference. (246)

Given that the parties must agree on this power, it will always be a question of
interpretation, particularly if the parties use other expressions that are argued to
be synonymous. The key example is the reference to ‘equity’. (247) Another
confusing expression is an empowerment to decide under ‘principles of law and
equity’. This is problematic in part because different legal systems use this
concept in differing ways. Most legal systems see principles of equity as part of
the law and not as matters outside of it. (248) Conversely, ICC Case 4467
considered that an express entitlement to decide in equity empowered a tribunal
to act as amiable compositeurs. (249)

Where the parties select both the national law and empower tribunals as amiable
compositeur, it is a question of interpreting their intent. In ICC Case 2216, the
tribunal considered that this meant that non-mandatory rules of substantive law
could be overridden where equity made this appropriate. (250) The amiable
compositeur power would make little sense if it had to always defer to the
national law. The converse is also true. To rule as a number of tribunals have,
that they can decide in equity except where mandatory provisions apply, is open
to debate as if they were truly mandatory laws they would apply in any event
even if a tribunal was only asked to decide in equity. Absent extraneous guides to
the parties’ likely intent, the better view is that the dual empowerment should be
looked at on a case-by-case basis to decide when it would be reasonable to
depart from the selected national law. That should only be so where application
of the law would be contrary to the legitimate expectations of the parties at the
time they entered into the contract. In many cases where that would be so, a
tribunal might come to the same conclusion by applying principles of good faith,
estoppel or implying particular terms into the contract. The point is that the
proper approach to such circumstances is to consider questions of interpretation
and evidence as to the page "1047" whole history of the dealings and not simply
follow a broad theoretical perspective as to the nature of amiable compositeurs.

13.14.3. Due Process and Procedural Issues

A tribunal must still fully apply all principles of due process. This should also
entail giving the parties sufficient advice as to the way the amiable composition
power is to be approached. The parties are entitled to know the factors that the
tribunal wishes to consider under this empowerment. Indicating the factors seen
to be relevant allows the parties to make submissions on the key matters that
will impact on the outcome. This will also impact on the range of relevant
evidence, both documentary and oral, particularly when reasonable expectations
and the initial allocation of risk and reward is being analysed. Jarvin suggests the
possibility of applying a lower degree of conviction when accepting facts as
proven under these concepts. (251) There might be circumstances where this
would seem reasonable but this would be rare. One possibility would be where it
would be extremely expensive and onerous to provide proof of a key proposition
that ought to have been admitted by the other side but was not.

Where reasons are concerned, there seems no justification to give any less
reasons when acting as amiable compositeur than when acting strictly according
to law. The losing party still would like to know why it lost. The more equitable
predilections of an amiable compositeur give good guidance for parties in
ongoing commercial relationships as to how to avoid similar disputes in the
future. Institutions that scrutinise awards do not obviate the need for amiable
compositeurs to provide reasons. Understanding the reasoning may also help
ensure voluntary compliance. Annulment or enforcement challenges should not
be required to guess the tribunal's reasoning or assert a lack of reasons as a basis
for challenge itself.

When the tribunal acts as amiable compositeur, another question is whether

there would be any appeal rights where these otherwise pertain and if so, how
these are to be circumscribed. French law indicates that there is to be no appeal
in such circumstances unless the parties expressly reserved their right to do so
in the arbitration agreement. (252) Parties may also agree to limit or waive their
rights to appeal.

Finally, a tribunal may have proactivity concerns if acting as amiable compositeur

when one party does not appear or appears unrepresented. In each case they
may be unaware of what can be argued for in the context in this mandate. (253)

page "1048"

13.14.4. The Nature of the Concepts

ICC Case No. 10728 of 2001 (254) and Case No. 7986 of 1999 (255) both concluded
that there was no distinction between the two concepts, but this is not a uniform
view. To the extent that there is a distinction between the two concepts, this
adds a further difficulty if the parties merely empower the tribunal to decide in
equity. Some jurisdictions treat ex aequo et bono and amiable composition as
identical. French law sees the concepts being similar in meaning, while Swiss law
sees ex aequo et bono as more permissive. (256) Switzerland sees amiable
composition still requiring the application of legal rules but with some tempering,
while ex aequo et bono allows non-legal principles to apply. Poudret and Besson
argue that the terms do in fact have different meanings, citing the discrete
references to each term in various arbitration laws and institutional rules, and
consider that arbitration ex aequo et bono ‘corresponds to arbitration in equity’
(257) whereas amiable composition is limited only by norms of public policy which

must always apply (although Poudret and Besson also clarify that departures
from the law should be made ‘only to the extent that this is necessary’). (258) It is
still a contentious question as to the degree to which an arbitrator may depart
from rules of law when so empowered. Poudret and Besson make the important
observation that given the various solutions and terminological imprecision, the
intent of the parties plays the essential role. (259)

13.14.5. Should the Concepts Be Applied Discretely or Should Their

Application Be Compared to That of Otherwise Applicable Law?

As noted, one uncertainty arises where parties select amiable composition or ex

aequo et bono but call for it to be utilised in conjunction with a national law. In
this case the normal approach suggested is to apply the national law except
where it would produce an unfair result. (260) The above discussion argued for a
more nuanced analysis.

When there is no selection of a national law, tribunals fall into two groups. The
first simply consider that there is no need to identify an otherwise applicable
national law but instead equitable and/or transnational principles may be page
"1049" considered. (261) Others take the view that it is preferable to consider
what law would otherwise apply and then decide when and why equitable
principles might suggest a different conclusion. (262) The clearer the otherwise
applicable law would be in a particular case, the more a tribunal might consider
its potential application in deciding whether that conforms with notions of
equity or not. The duty to provide a reasoned award might naturally lead to such
an approach, with the obligation to explain what principles of equity were
resorted to and why they were necessary or appropriate. This approach was
preferred by an ICC France working group on amiable composition. (263) Where
there are ongoing commercial relationships between the parties, such a
reasoned approach will also guide them in their future decision-making,
including as to applicable law.

Poudret and Besson also take the view that the tribunal should first consult the
law as the method is to utilise equity to depart from the law solely to the extent
necessary. Furthermore, rules of public policy and mandatory rules of law need
to be considered. (264) The other important feature is that tribunals deciding in
this way will commonly reach a different conclusion to that which would be
reached under the otherwise applicable law. In doing so, they need some reason
why the solution offered by that law was unappealing. The converse scenario is
where a tribunal empowered to act as amiable compositeur simply decides
strictly according to law. If the tribunal considered equitable considerations and
concluded that there were no reasons to depart from the law, this is entirely
appropriate. If the tribunal simply failed to engage in this, then the mandate has
not technically been followed. (265)

13.14.6. The Relevance of Mandatory Laws

Another reason why a tribunal should first consider the applicable law is to see
whether there are any mandatory provisions or matters of public policy that
should not be departed from. Some legal systems allow mandatory rules to be
ignored when other than international public policy. That view may be
debateable depending on the permutations of law and rules to that effect.
Poudret and Besson suggest that an arbitrator in equity is not bound to respect
mandatory rules of law. (266) They assert that public policy must be respected.
Jarvin argues that mandatory rules of page "1050" law having a public policy
character still apply to an arbitrator acting as amiable compositeur. He suggests
that this flows from the duty to render an enforceable award. It also ought to
flow from the fact that a mandatory norm cannot be varied by any agreement of
the parties, including an agreement for determination as amiable compositeur.
Thus, allowing a tribunal to determine as amiable compositeur or ex aequo et
bono does not constitute a waiver of fundamental rights. Poudret and Besson
observe that such fundamental rights are even more essential where the tribunal
has such broader powers. (267)

13.14.7. What May a Tribunal Actually do When Applying the Norms?

The greatest difficulty is to consider what a tribunal can or should do when so

empowered. The two questions are to what extent can an arbitrator disregard
the provisions of law that otherwise apply and second what modifications are
permissible to the terms of the contract. (268) Where investment cases are
concerned, there is a need to consider the extent to which to depart from general
principles of international law as well as treaty and contractual norms.

One of the more contentious aspects of these questions is the extent to which the
tribunal can effectively alter the bargain between the parties. This has occurred
from time to time although it has also been suggested that this can breach the
tribunal's mandate. (269) The power to adapt a contract might apply in any event
under some applicable law. (270) This is discussed in the following section. Where
this is not the case and the question is as to the ambit of the concepts, Berger
argues that a tribunal with such power can decide contra legem. (271) He notes
that most, however, feel bound by the rules of applicable law as ratio scripta
referring to ICC decision No 3742. (272) As to the natural argument against
changing the contract, it is suggested that the power is simply to decide disputes
under that contract and within the four walls of the arbitration agreement not in
accordance with strict legal principles. (273) While it is difficult to state what a
tribunal can in fact do, most would agree that the power does not equate to an
adaptation of contract page "1051" power. (274) Another expression is that the
tribunal must not ‘rewrite the structure of the agreement’. (275) Jarvin also
suggests that the terms of the contract may not be modified by an arbitrator
acting as amiable compositeur. (276) However, he notes an ICC case where a
tribunal so empowered reduced the percentage of an agent's commission
although this was expressly defined in the contract. (277) Textually, the position is
unclear as even rules that expressly refer to contractual terms do not express a
position one way or another.

The Paris Court of Appeal has on a number of occasions concluded that the
power to act as amiable compositeur does not allow arbitrators to ‘modify the
economics of the agreement’. (278) Tribunals have nonetheless concluded that
they may bring the contract into its intended balance where the provisions are
clearly against the true intent of the parties or against public policy, (279) may
reduce or remove excessive and unfair provisions given the nature of the
circumstances, (280) and may ‘moderate the rights deriving from the contract…’
(281)

It is easy to suggest that the power to act as amiable compositeur should not be
seen as going so far as to allow an arbitrator to modify the contract but it is
harder to define some meaningful power that does not effectively do that and
which could not be dealt with by other interpretational methodologies. Here it is
much easier to assert what a tribunal should not do rather than state clearly
what it can. While the cases tend to pay lip service to the obligation not to modify
the economics of the contract, it is hard to see what else is being done if a
tribunal goes beyond principles of good faith, implied terms and common law
derivatives such as unconscionability if that is the applicable law. To the extent
that the power goes beyond a purposive and good faith application of the
otherwise applicable law, then a tribunal exercising such a power inevitably
creates new relations between the parties. (282) If a tribunal aims to rebalance the
original value for changed circumstances, then such an approach is similar to
that under the rebus sic stantibus principle. It is difficult to fully justify this
approach based on a finding of party page "1052" intent ab initio simply as a
result of the amiable composition clause. This is because the clause is found in the
same contract as the supposedly imbalanced provision and must, therefore, be
read in context as to intent.

As with all potentially conflicting questions of party intent, tenable arguments

can be made either way. On the one hand, being the supervisory provision for
dispute resolution, it could be argued to take precedence over specific
contractual norms. Conversely, presumptions might better be that specific
provisions take precedence over general ones and that parties were more likely
to have intended the particular clause which would be well understood rather
than the complex legal notion of which many commercial parties would be
unaware. Poudret and Besson make the observation that from an intent basis, it
is somewhat difficult to find the power to modify a contract from the very
contract entered into by the parties. (283) For transactional lawyers, this suggests
that care should be taken in drafting such clauses to indicate what is truly
intended. For tribunals, the better view is that all circumstances should be
considered in order to determine what truly was ‘the economics of the
contract…’ Where a party took on harsh obligations in return for a favourable
price term, that is the essential economics. An example would be a penalty clause
in a building contract awarded to a builder who tendered at a higher price, but
who was prepared to guarantee timely completion. There is a fundamental
difference between rewriting provisions seen to be unfair in circumstances that
would have been envisaged from the outset and instead effectively rewriting a
provision to take into account new circumstances beyond reasonable
contemplation when originally allocating risk and reward. Another scenario is
where the tribunal simply believes that the original contract was unfairly
imbalanced because of the disparity in power between the commercial entities.
While one might feel sympathy with the weaker party, it is harder to justify
rebalancing the contract on the basis of intent through an amiable compositeur
clause when the stronger party could not possibly have intended that outcome.

Even if a power to modify the terms was arguable, there is then a question as to
whether it is modifying the terms in relation to their historical dispute or also
modifying the terms for the future. Fouchard, Gaillard and Goldmann argue that
the latter should not be so. (284) Craig Park Paulsson also suggest that the contract
should not be altered for the future. (285) Nevertheless, if a similar dispute arose
in the future and the same approach applied, this would be the effective outcome.
It would be hard to see what an alternative scenario would entail. Another
question is whether amiable composition requires careful attention to efficiency
as well as fairness, as is the emerging trend in lex arbitri. If that is the case, it is
harder to page "1053" justify ignoring time limits for notices for bringing of
actions under an amiable compositeur power.

13.14.8. Subjective versus Objective Analysis

An important question is the degree of subjectivity in deciding as amiable

compositeur. Because different adjudicators can take different views about the
nature of equity and the way it ought to apply in an instant case, there can be
significant variations on such issues as the extent to which the adjudicator
departs from the law, modifications to the contract, application of limitation
periods and exempting clauses, revising onerous provisions such as interest and
penalty clauses and who should bear the loss when both parties suffer through
the actions of an external event. Furthermore, while it is easy to say that
principles of equity are valid reasons to depart from a strict application of law, it
is harder to decide what that means for an arbitrator simply because legal
systems now incorporate equity norms within them, civilian traditions through
good faith and similar norms and the common law through specific categories
where strict application of express contract terms will not apply. Examples
include mistake, duress and unconscionability. Hence, the challenge is to
determine just what are the equitable principles over and above those within
domestic legal systems that suggest different conclusions to a tribunal. Once
again there is no clear jurisprudence on this question, but tribunals should
search for party intent and reasonable expectations and not simply apply their
own notions of commercial fairness.
13.14.9. The Concepts and Related Notions

13.14.9.1. Gap Filling

It is also important to compare the concepts with the gap filling function of
adjudicators. Because of the principle of non liquet, a tribunal cannot refrain
from deciding a matter simply because there is no obvious rule covering it, either
in the applicable law or the contract. Because of this, all adjudicators must fill
gaps when they are found and this is not dependent on an ex aequo et bono
power. (286) However, the methodology of gap filling may vary when a tribunal is
so empowered. This may be problematic as there would typically be a residual
gap-filling method that would otherwise apply and there needs to be some
justification as to why this would not be employed.

page "1054"

13.14.9.2. The Concepts and Lex Mercatoria

Tribunals acting as amiables compositeurs will at times seek to apply lex

mercatoria. (287) While the notions are different, lex mercatoria arguably contains
notions such as good faith that would be available to an amiable compositeur.

13.14.9.3. The Concepts and Mediation

The notion of amiable compositeur is distinct from that of a mediator. Both the
process and the potential solutions in mediation go far beyond the ambit of
amiable composition. Where the latter is concerned, the obligation is to still
provide a binding decision after due process norms have been satisfied.

While it should be seen as distinct from a mediation function, each were

apparently derived from similar principles with the aim of restoring harmony
through a revised legal relationship. (288) Nevertheless, whatever one's view
about mediation, if the parties truly want a mediated solution, then amiable
composition is not the optimal method.

13.15. Adaptation and Rectification

Chapter 14, dealing with remedies, considers the possibility of adaptation and
rectification as commercial remedies, including hardship clauses used to this
end. For the purposes of this chapter, it is worth merely mentioning how one
might identify an applicable law in that regard to the extent that this is not
sufficiently clear from the terms of the contract itself. In some cases the parties'
agreement may itself empower a tribunal to adapt a contract. (289) The Dutch Act
(290) allows for gap filling or modification while the Swedish Act (291) allows

supplementing of the contracts, but in each case only where authorised by the
parties. Article 6.2.2 of the UNIDROIT Principles deals with hardship. A similar
principle is that of imprévision. Adaptation can also mix questions of procedural
and substantive law. (292) Berger notes potential problems if there is a mismatch
between the procedural provisions allowing for adaptation and the contrary
principles to that effect in the lex causae. (293) Adaptation or hardship cases may
involve determining how the page "1055" contract should be modified, or
consideration of the damages that should be awarded where a party fails to
honour its renegotiation obligations. (294) Damages may be problematic in such
circumstances as there may only be a duty to attempt to negotiate and not any
obligation to achieve a particular outcome.

As noted in the previous section, the better view is that an arbitrator empowered
to act as amiable compositeur is not thereby empowered to modify the contract,
although it was also noted that is difficult to state the power positively in a way
which does do so in substance. Another mechanism would be to apply lex
mercatoria and the doctrine of rebus sic stantibus in particular. It is accepted that
this does not generally apply to transnational contracts. (295) A related situation
where revision of contracts may apply is where the dispute is subject to
principles of public international law, which more readily include the doctrine of
rebus sic stantibus. This may naturally arise in investment treaty arbitration,
although the ambit of the doctrine is unclear as it is in natural tension with the
central doctrine of pacta sunt servanda and may be circumscribed by express
provisions in the treaty itself.

Such a power might apply simply through the express or implied choice of the
parties. For example, if the parties select a substantive system of law that does
allow for rectification in appropriate circumstances, such a right flows as a
matter of course. That may also arise where the tribunal exercises discretion as
to applicable law, although it should be circumspect about granting itself a
rectification power through such a choice alone, absent any evidence of a
reasonable expectation of this outcome by the parties. That should not be a
problem if the choice made was one that would have been reasonably
predictable by the parties. Nevertheless, adaptation absent express consent does
not sit easily with notions of consent generally, given that the very agreement
that one party seeks to vary is the agreement that gives rise to arbitral
jurisdiction in the first place.

One significant difference between arbitrators and courts is that the latter will
often interfere with contracts based on a statutory mandate to protect weaker
parties and to provide minimum standards in commercial activities. Where an
arbitrator is concerned, such a justification would be problematic. Instead, resort
must be had to some notion of original intent and a wish to retain the original
commercial balance of rights and responsibilities. In that context, a tribunal
should be very careful in applying simplistic equitable analysis and instead
should ensure that it reviews all of the evidence about the risk/reward matrix
that each party brought to the negotiating table. (296)

page "1056"

ICC Case No. 7544 considered that gap filling was an important element in long-
term commercial relationships. Berger distinguishes between initial and
supervening gaps in a contract. (297) At the margin there may be little difference
between adaptation or gap filling on the one hand and applying lex mercatoria or
deciding as amiable compositeur on the other hand. (298) This is not to say that the
concepts are identical, but they cover similar factual scenarios. It could also be
said that gap filling and adaptation are fundamentally different concepts. A
related issue will be the application and interpretation of frustration or force
majeure clauses, hardship clauses or other contract revision terms. It is also
difficult to identify the real difference between adaptation on the one hand and
application of norms of good faith and fair dealing on the other as part of
applicable law. Adaptation also comes close to a mediation function in many
instances, albeit with binding power. (299)

13.16. Proof of Applicable Law

In addition to the conflict of laws questions as to the applicable substantive law,

there is a related question as to how such law needs to be proven to a tribunal.
No clear consensus has emerged either in arbitral rules or arbitral practice. Lex
arbitri and arbitration rules generally empower arbitrators as to a methodology
of identifying applicable law but do not give a specific direction on the means of
proof other than providing for broad discretions as to matters of evidence. In
domestic disputes before national courts under national law, the judge is by
definition an expert on the relevant law. Hence, it would never be seen as a
question of proof. Where international disputes are concerned, however, the
presence of different legal systems means that situations will commonly arise
where an adjudicator trained in one legal tradition is asked to apply substantive
law from another jurisdiction.

Courts in domestic legal systems can be grouped into three categories in terms of
their methods of ascertaining applicable foreign law. (300) The first category
involves courts having considerable power to ascertain the contents of foreign
law on their own motion; (301) the second comprises those essentially required to
rely on the parties for proof of foreign law as a factual matter; (302) and
intermediate systems where reliance is primarily placed on proof by the parties
but where the courts also have responsibility. (303) The differences between legal
families relate to page "1057" their different classification of foreign law as
either fact or law. Common law sees foreign law as a question of fact while civil
law systems treat it as law, although each statement is simplistic and depends on
other duties and powers. (304) The corollary of the latter is the doctrine of iura
novit curia, being the presumption that the court must know the law. This not
only impacts on methods of proof but also burden of proof. This is discussed
further in section 13.19.

Differences in approach are in part based on differences between adversarial and

inquisitorial systems. An adversarial system relies on the parties to present all
appropriate evidence and arguments, including in relation to proof of foreign
law. Civilian systems place the obligation on the judge to come to the correct
answer and are based on a premise that the parties do not need the requisite
expertise in law, but simply need to have a justifiable claim on the facts and then
the judge will identify the relevant law. The distinctions are even more complex
as common law systems also utilise a presumption that if foreign law is not
proven, a court may apply its own law on the presumption that the law is the
same, absent proof to the contrary. (305) Over time, common law courts have
accepted that this latter presumption should best be seen as a pragmatic one and
that it is not well supported by a truly logical presumption where the foreign law
is from a civilian country. (306)

In addition to determining who must establish the law, there are also questions
as to how. Courts may seek to determine the content of foreign law by relying on
parties' presentations, making specific requests of the parties, relying on expert
evidence, including an expert appointed by the court or through their own
research (307) and knowledge. There may also be diplomatic approaches to proof
of foreign law. (308) Principle 22.1 of the ALI/UNIDROIT Principles of
Transnational Civil Procedure states:

The court is responsible for considering all relevant facts and evidence and for
determining the correct legal basis for its decisions, including matters
determined on the basis of foreign law.

page "1058"

Where arbitration is concerned there is both a practical and theoretical element

in deciding on which approach to adopt. The practical element arises where the
tribunal is not a legal expert in the system of substantive law that is to apply. The
theoretical element is whether any applicable law in international arbitration
can ever be described as foreign and what implications for methods of proof flow
from this question. Stated differently, when is it ever appropriate for a tribunal
to identify the law itself and not have it proven by one or both of the parties?
There is no such thing as ‘foreign’ law in an international arbitration, but simply
the need to determine the applicable law or laws based on party autonomy or
the tribunal's discretionary powers. Fouchard, Gaillard and Goldman argue
strongly that the content of foreign laws should be treated as an issue of fact in
international arbitration. (309) It is debatable whether a tribunal should apply the
presumption that foreign law is the same as that of the arbitral Seat. This is not
to be preferred. (310)

Further direction is given in some lex arbitri. Article 34 of the English Arbitration
Act 1996 states:

(1) It shall be for the tribunal to decide all procedural and evidential matters,
subject to the right of the parties to agree any matter.
(2) Procedural and evidential matters include–

[…]

(g) whether and to what extent the tribunal should itself take the initiative
in ascertaining the facts and the law.

Article 2.1(c) of the LCIA Rules indicates that subject to contrary agreement
in writing by the parties, the tribunal ‘shall have the power, on the
 application of any party or of its own motion … to conduct such inquiries …
in identifying the issues and ascertaining the relevant facts and the law(s) or
rules of law applicable to the arbitration, the merits of the parties' dispute
and the arbitration agreement’. This is subject to the parties being given a
reasonable opportunity to state their views. Section 27(2) of the Danish
Arbitration Act allows a tribunal to request the courts to request the Court
of Justice of the European Community to give a ruling on a question of
European Union law where it is necessary to enable the tribunal to make an
award. Article 1044 of the Dutch Code of Civil Procedure, applicable to
arbitral tribunals, allows for a request via the President of the District Court
at the Hague for information under the European Convention on Information
on Foreign Law.

Once the identity of the applicable law is known, there are then questions as to
proof of its content and meaning. This is separate to the question of
interpretation although there may also be the need to identify the interpretative
principles themselves. Documents seeking to establish the content of the
applicable law should page "1059" typically be presented, notwithstanding the
suggestion from some civilian lawyers that the tribunal's duty to know the law
would not require this. (311) A range of secondary sources may be admissible to
assist the tribunal in understanding the meaning of the primary texts. Various
legal systems will also differ as to the importance of legislative history and past
cases as aids to understanding or as binding precedents in the case of the
common-law. Even if the domestic legal system does not utilise the doctrine of
precedent, the tribunal ought to pay sufficient respect to municipal case law. For
example, in the Brazilian Loans case, the PCIJ stated that a tribunal of
international law should:

pay the utmost regard to the decisions of the municipal courts of a country, for it
is with the aid of their jurisprudence that it will be enabled to decide what are
the rules which in actual fact, are applied in the country the law of which is
recognised as applicable in a given case. (312)

A related question is the extent to which experts may give opinions on questions
of applicable law. The general rule is that an expert should indicate to the
tribunal the nature and content of the law but should not seek to opine on the
way the law applies to the facts before the particular tribunal. That is the
function of the tribunal itself.

13.17. Scholarly Commentary

The common law tradition involves lawyers making oral representations to a

court in relation to the most significant aspects of previously submitted briefs
and pleadings. Because of the doctrine of precedent, on matters of law, common
law counsel will spend a significant amount of time dealing with cases and
arguing by analogy from other cases based on similarities and differences. The
common law tradition has not had a heavy reliance on legal scholarship in
presentation to courts. Civilian lawyers, dealing in a system without precedent
and often based on codifications, will more typically deal with the code
provisions and scholarly commentaries that seek to expand upon and attempt a
scientific theory underlying more general codified norms.

Once again international arbitration is a hybrid, adopting both forms of

argument. Where scholarly commentaries are concerned, much might depend on
the preferences of particular arbitrators. Here there are practical as well as
theoretical factors at play. Historically, leading arbitrators in the civilian
tradition were eminent law professors, often as well as being eminent
practitioners, who would naturally find at least some comfort in resort to
commentaries. Conversely, in the common law tradition, retired judges, senior
counsel and practitioners have page "1060" tended to dominate and at times
take a more pejorative attitude to scholarly commentary. An optimal
adjudicatory methodology would hopefully adopt the most compelling logic
wherever it may be found.

13.18. Use of Past Cases

While it is common to speak of the doctrine of precedent or stare decisis

applying only to the common law, a number of civilian legal systems see great
respect being given to previous cases and in some circumstances demanding that
particular reasoning be upheld if adopted by a consistent position in a stipulated
number of previous decisions. (313) Hayek argued that following precedent
supports experience gained over generations of experimentation. (314)
Consistency and predictability are part of what has been described as the inner
morality of law. (315) Prior arbitral decisions may be ‘taken into account as
constituting a corpus of principles representing the litigants’ shared
expectations’. (316) Another reason why tribunals must at times consider past
cases is simply because counsel have referred to them in submissions. Giving a
party an adequate opportunity to present a case includes considering the case as
presented. (317) However, slavishly following past cases without attention to
reasoning can itself be problematic.

Prior cases can be respected either for their articulated reasoning, in which case
they are merely a source of an idea, or conversely because of the status and
conscientiousness of the adjudicator and the respect in which the decision has
been held and the consistency with which it has been followed. They are not
mutually exclusive in that respect will flow from acceptance of the reasoning.
Some tribunals will attempt independent reasoning and then conclude that their
decisions are consistent with past cases, giving them further comfort in their
logic. (318) Other cases will openly rely on logic and conclusions in past cases.

While international bodies typically eschew formal precedent (319) nevertheless,

there is a natural tendency to follow previous reasoning and conclusions unless
there is a good reason to vary this. WTO jurisprudence speaks of the ‘acquis’
page "1061" through which Members generate legitimate expectations. (320)
Empirical and analytical research has shown a significant growth in citation of
previous cases, particularly in investment awards. (321) There are three reasons
why this would be more prevalent with investment arbitrations. First the awards
are more widely reported owing to the involvement of governments, and the
application of national freedom of information legislation and ICSID norms.
Second the standard principles are very general and it is hard to determine their
ambit simply from their wording. This applies to key norms such as most
favoured nation treatment, national treatment, fair and equitable treatment and
the like. Hence it makes sense to look for articulations of these concepts,
particularly from eminent arbitrators who in many cases considered these
principles through the ICJ or International Claims tribunals. Third, because
nation States are involved, there is a natural tendency in adjudication to look for
consistency in treatment, as this will at least appear to be less activist from the
perspective of the adjudicator. (322)

It is difficult to accurately evaluate the use of past cases in arbitrations. In

addition to the skewed sample of awards, there is no consistency in approach
(323) and in addition no real way to know whether past cases influenced the

outcome or were selectively used to justify decisions that were otherwise

reached. Furthermore, because many cases refer to scholarly opinion and not
directly to cases, one cannot know the influence of cases without looking at the
particular scholarly material to see how much it is based on an analysis of cases
itself. (324) References may vary depending on the question at issue. For example,
if a tribunal is seeking to employ lex mercatoria, it is natural to look at past cases
to identify what are in fact commonly accepted norms. (325)

One of the difficulties in relying on past cases is that there is not only an overall
problem in terms of publication of awards but in addition, those published page
"1062" are predominantly ICC cases. While this is valuable in the context of being
awards from an institution with the highest reputation, much of the analysis of
collected works have come from an earlier time. In the early years of arbitration,
confidentiality meant that most awards were not published in whole or in part.
The situation is quite different today. Certain institutions regularly publish
awards. That is the case with ICSID. The same was true with the Iran-US Claims
Tribunal both in primary form and also through the volumes of secondary
commentary, largely from those directly involved in the process. The ICC
publishes more and more awards and includes compendia on particular topics in
the ICC Bulletin. Commercial publishers such as Kluwer International provide
internet-based as well as hard copy resources. (326) Where the Model Law is
concerned, the CLOUT database provides a reference point for decisions on
Model Law provisions, hence ensuring transparency and consistency. (327)

13.19. Iura Novit Curia (328)

The comparativist perspective as to proof of applicable law as discussed in the

previous section raised another fundamental difference between legal families,
being the way that foreign law is dealt with by adjudicators. It was noted that in
the civilian tradition, the notion that judges are expected to know the law goes so
far as to imply that they are expected to know foreign law as well as their own
domestic law. This does not mean that they actually must know it, but that they
have an obligation to make appropriate inquiries and the parties are not obliged
to prove its contents. That difference between systems also impacts upon issues
of burden of proof. If it is a judge's duty to know the law, then a party simply
needs to prove the facts and the appropriate legal conclusions will flow as a
natural corollary. Conversely, if foreign law is a matter of evidence, a party who
fails to provide adequate evidence in support of a claim that they are making can
be said to have failed in their burden of proof, although as noted, this is
complicated by the fact that many common law systems also utilise a
presumption that foreign law is identical to domestic law unless there is proof to
the contrary.

Domestic legal systems in the civilian tradition commonly support various forms
of the notion of iura novit curia. Civilian legal systems primarily support the
principle on the basis that an appropriately trained judiciary ought to have the
key responsibility to know and apply the law correctly. The differences between
legal families are affected by variations in the fundamental approach that each
system takes to adjudicatory dispute resolution. It is far more natural to find
civilian jurisdictions supporting the principle of iura, given an inquisitorial
model that makes the judge the primary actor in seeking a just solution.
Conversely, common page "1063" law jurisdictions, which historically saw
judges as passive recipients of adversarial arguments by opposing litigants, were
far less supportive of any notion that a judge could decide on matters not
presented by the parties. (329) This difference in view between civil and common
law systems has been argued to be one of the ‘most significant distinctions’
between the systems. (330)

The key question for arbitration is which approach to adopt as per these
fundamental differences, and whether a tribunal must know and apply the
appropriate law, regardless of the contentions of the parties under the principle
iura novit curia. (331) Like most contentious issues in international arbitration,
part of the conflict also emanates from differences in view as to the essential
basis of arbitral power. To those who opt for a more consent-based paradigm,
any behaviour by an arbitrator outside of the agreement of the parties looks
problematic. Conversely, from a more jurisdictional perspective, arbitrators
appear to take on more of a quasi-judicial role and can be argued to have a duty
to know the law as is the case with at least some domestic judges. There is simply
no consensus view. A number of arbitral cases and scholarly comments have
considered the nature of the principle and whether it applies as a matter of
course to international commercial arbitration. Some assert that there is a duty
to apply the principle. Others consider that there is merely a power to do so.
Some consider that the principle should not apply to international arbitration
and is contrary to arbitrations fundamental nature.

The debate seems to mix a host of significant issues that may explain why it is so
contentious. If there is a duty to know the law, where does it come from? Which
law is a tribunal obliged to know? Is the duty to know the law limited to the
claims as made or does it go so far as to encompass a duty to apply legal
principles that would ground relief beyond that as claimed? If it is limited to the
claims as made, does the duty imply a research function on the part of a tribunal
to evaluate laws regardless of the arguments put by the parties? To what extent
are international arbitral tribunals obliged or able to apply domestic judicial
principles in relation to these questions? To the extent that they are obliged or
entitled to do so, how do they deal with differences in view between different
legal systems? Do the answers differ depending upon the nature of the law
involved? How does the duty to know the law integrate with other principles
such as a duty to apply mandatory law or a duty to consider whether there is
valid jurisdiction, each of which imply the possibility that principles beyond
those argued by the parties may be determinative? If there is a duty, what due
process implications does it have?

These questions show that there are a myriad of issues that need to be
considered. There is also a procedural question that is fundamental to
understanding page "1064" how far reaching the principle might be. If there is a
duty or power to apply the law, the procedural issue is whether the tribunal
should draw the parties' attention to the law it believes to be relevant at an early
enough stage to allow them to make submissions and engage in an adversarial
contest if they do not agree. Conversely, is an arbitrator entitled to or indeed
bound to apply a law after the conclusion of the hearing if it is considered the
appropriate one, notwithstanding the failure of the parties to address it? Stated
in this way, it becomes obvious that the question of iura is impacted upon by
other duties, such as the duty to give all parties an adequate opportunity to
present their case and the duty to treat parties equally. To critics of the concept,
it can never be appropriate to decide matters without appropriate input from the
parties, and that is the view contended for here. To supporters, other duties are
subordinate to, or are qualified by the iura principle, as the correct outcome
must always be the paramount concern.

While some conceptions of due process would at first glance suggest such a duty
to advise exists, it depends in part upon the view one takes about the essence of
the doctrine of iura. At the extreme, if knowledge of the law is seen as the
peculiar expertise of the adjudicator, then due process rights of the parties could
be limited to evidentiary matters. Conversely, a belief that parties and their
counsel are equally able to identify valuable and at times conflicting arguments
as to law, would support a broader due process obligation as to law as well as
fact and potentially undermine the iura principle itself. In other cases, legal
systems might adopt a hybrid, imposing the ultimate obligation on the judge but
only after giving the parties an opportunity to make presentations on the issue,
hence qualifying the notion of iura. The domestic litigation system most
supportive of the iura principle is Switzerland, which adopts it and has not in
every instance required that the parties have an opportunity to be heard. (332)
Germany supports the principle (333) but demands a right to be heard where the
law sought to be applied would be a surprise to the parties. (334) In France, where
parties do not invoke or seek page "1065" to prove foreign law, a court will seek
to apply the law ex-officio if the claims involve matters which cannot be freely
disposed of or where required by international convention. (335)

In common law jurisdictions, the issue is not typically dealt with in relation to
the iura principle itself. Instead it arises in two contexts, the first being how the
content of foreign law is determined and the second being the ability of a court to
raise issues sua sponte, typically on appeal and in the context of establishing
precedent. As to the first issue, there is no uniform position among common law
countries. Under English law, as noted above, foreign law is a question of fact to
be proven by the parties. (336) However, unlike other factual matters, there is also
a presumption that foreign law is identical to local law unless proven otherwise.
Thus the party with the burden of proof on a particular claim does not fail as a
matter of course if they are unable to prove the content of foreign law to
sufficient satisfaction, (337) although as noted in sections 10.6 and 13.6, this view
is now being questioned somewhat. French law also uses such a presumption of
commonality. (338) The US Federal Rules of Civil Procedure provide for more
active involvement by the adjudicator and allow a court in determining foreign
law to consider any relevant material, source or testimony ‘whether or not
submitted by a party or admissible under the Federal rules of evidence. The
court's determination shall be treated as a ruling on a question of law’. (339) This
is closer to the current English arbitral law as noted in the previous section. As to
sua sponte determinations, notwithstanding the general view in common law
systems that iura is not applicable per se, there are numerous circumstances
where a common law page "1066" appellate court raises issues sua sponte.
Scholarly commentary on the desirability or otherwise of this is mixed. (340)

These differences between systems also impacts upon issues of burden of proof
and are hence central to the concerns of this book. Isele notes that ‘at its core the
(iura) principle gives an answer to the question of how to allocate the burden of
presentation and knowledge of the applicable substantive law between the
national judge and the parties'. (341) At its simplest, the principle responds to the
question of whether the parties should have to research and make submissions
about law or whether they can instead rely on the judge to properly understand
it, or engage in research or call for assistance from experts? Whatever the correct
answer, as long as litigants know the rules, they know the minimum
requirements on them if they are to succeed.

This very brief analysis of domestic litigation systems shows that there is
support for the iura principle in civilian systems but with divergent views as to
the obligation to give parties an opportunity to be heard and hence as to the
ambit of the iura concept itself. The common law takes a far more restricted
view. Some systems deal with it broadly, while others concentrate on the
subsidiary question of proof of foreign law.

13.19.1. Arbitral Laws, Iura and Due Process

The principle iura novit curia in fact has different nuances of meaning in various
domestic systems. Douglas Brooker has identified six quite different definitions.
(342) It literally stands for the proposition that the judge knows the law. It relates

to another principle, da mihi facto dabo tibi ius – which means give me the facts
and I will give you the law. Stated in that way, with no further elaboration, it is
difficult to see exactly what ramifications it has as to the rights and obligations of
both page "1067" litigants and adjudicators. Li makes the sensible observation
that one reason why it is difficult to understand is that it is an adage rather than
a direction. (343)

It is appropriate to look to express rules about applicable law if the iura principle
relates to laws that should be known and applied. Any duty to know the law
would encompass questions as to knowledge of procedural and substantive laws.
Most would agree that identifying applicable law is a procedural matter,
although some suggest that there are grey areas at the margin. (344) If iura can
only be seen as a rule of procedure, it would need to be justifiable within the
relevant lex arbitri. A duty to apply appropriate procedural laws should not be
contentious. All relevant procedural norms, whether emanating from the lex
arbitri, from party autonomy generally or from selection of arbitral rules, must
be known and applied. It is contentious as to whether other procedural norms
must ever be applied or can be said to be mandatory. (345) Even if it was valid for
such norms to be applied, these would be publicised to the parties at an
appropriate stage or they simply could not be followed as procedures. An
example might be a decision to hear witnesses on oath if it is felt that public
policy notions in the most likely enforcement country would demand it. Hence,
the key issues concern substantive law, because it is this category that raises the
possibility of applying a law not known or addressed by the parties.

Most lex arbitri say nothing specific about the iura concept. Instead, they deal
generally with selection of applicable substantive law by first acknowledging the
paramount role of party autonomy. Thus provisions such as Article 28 of the
UNCITRAL Model Law indicate that any selection of applicable law by the parties
is binding on the tribunal. (346) Institutional rules express the clear view that the
parties are free to agree upon the laws to be applied and only absent such an
agreement, does the tribunal apply the rules of law which it determines to be
appropriate, either directly or via a conflicts method. Kurkela has suggested that
the principle is part of the procedural framework in arbitrations having their
Seat in Finland or Sweden. (347) The Swedish legislation has not directly
addressed it, although the matter was apparently discussed during its
preparation. Jarvin has page "1068" as a result suggested that the parties clarify
the issue at the outset. (348) French NCCP requires a court to invite comment from
the parties if it intends to rely on a rule of law or construction of facts not
discussed by the parties. This flows from the French principe du contradictoire.
Schneider questions the applicability of iura to arbitration and in any event
argues the need for due process and advice to the parties. (349)

Another complication in the matrix is whether the parties have expressed a view
as to whether the tribunal may apply a law other than that presented in
submissions. Two conclusions ought to be non-controversial as a result of these
provisions. First, if the parties select a particular law or proscribe the application
of a particular law that is otherwise of interest to the tribunal, then the tribunal
ought to be bound by such a choice unless the law in issue is a mandatory norm
that must apply regardless of the wishes of the parties. A choice should be
binding even if it changes a previously applicable law. Mandatory laws are
themselves contentious as discussed in section 13.8. (350) If there is no mandatory
norm involved and the parties are simply making a selection of law, the tribunal
ought to accept it. The situation is also simple where the parties disagree as to
which law should apply. Here a tribunal is simply entitled to make the
determination that it thinks is most appropriate.
The second proposition is that absent any express direction by the parties, a
tribunal with a discretion to apply either an appropriate direct choice or make a
selection via a conflicts rule ought to do so with due care and to the best of its
ability, based on fair and reasonable principles. In that context, a tribunal ought
not to apply a method or rule that it believes to be wrong simply because that
was the rule suggested by the parties in their submissions. It would be a failure
for the arbitrator to follow the mandate if it ignored what it believes to be the
proper approach to such a function simply because of inadequate submissions.
Thus, page "1069" we might accept the principle that an arbitrator should not
apply a law known to be inapplicable and that iura might be analysed in the
context of such a principle.

Once again the contentious question is whether parties deserve notice of the
tribunal's concern and deserve a right to make representations. Lex arbitri make
due process obligations mandatory norms of procedure. Parties are required to
be given at least an adequate or reasonable opportunity to present their
respective cases. If a tribunal is considering which law to apply under a broad
discretion, there are some significant reasons in support of the view that the
parties should know the tribunal's thinking at the earliest opportunity. These
may be compelling even under the iura principle. First, because choice of law is
discretionary absent party choice, there may be arguments for and against
different choices. For example, if a tribunal is applying a ‘closest connection’ test
in complex transactions, there may be connecting factors to a number of
jurisdictions. The parties ought to be entitled to indicate why they believe that
certain factors are closer or more telling than others. Notwithstanding the fact
that conflicts rules are legal principles, this particular conflicts rule builds on
factual analysis, a matter about which parties should always be entitled to make
representations as the facts are solely within their knowledge and relevant
material is solely within their access. A second reason is that parties cannot
know how to frame their claims and defences without knowing the applicable
law. (351) A third reason is that the parties might not be in a position to properly
research and collect factual evidence until they know which legal principles
pertain. Thus there seems no reason in terms of construction of key lex arbitri to
fail to notify parties in all circumstances when a tribunal is considering applying
a different law to that submitted by the parties under a broad discretionary
power. ‘Surprise by a judgment is not allowed. The contents of applicable law
must be known to the parties beforehand and submitted to the parties for
discussion’. (352)

The same result should apply in jurisdictions whose lex arbitri expressly
empower a tribunal to identify applicable law. The main example is the English
Arbitration Act 1996. Powers to investigate the law are expressed in s 34(2)(g).
While the Act allows an arbitral tribunal to take the initiative in determining
questions of law as well as procedural or factual matters, section 34(1) still
subjects the tribunal's discretion ‘to the right of the parties to agree any matter’.
Thus the parties must clearly be entitled to prevent an arbitrator under this Act
from applying laws sua sponte. Furthermore, general principle (b) (section 1) of
the Act indicates that the provisions of that Part shall be construed according to a
number of principles including that ‘the parties should be free to agree how their
disputes are resolved, subject only to such safeguards as are necessary in the
public interest…’. It would be difficult to argue that applying laws sua sponte in
an arbitral setting is inherently in the public interest. That is certainly not part of
the page "1070" common law tradition. Section 34 is not designated as a
mandatory provision, hence the discretion can be removed in its entirety by the
parties should they so desire. Furthermore, section 33(1)(a), which requires
‘giving each party a reasonable opportunity of putting his case and dealing with
that of his opponent’, is a mandatory provision, although one which could be
reconciled with iura if limited to only putting submissions on matters where
submissions are required. Section 33(2) directs the tribunal to comply with the
general duty in conducting proceedings and in its decisions on matters of
procedure and evidence. Section 34 is headed ‘Procedural and Evidential
Matters’. Hence, while it might be suggested that there is nothing in the English
legislation that actively prevents a tribunal applying the law sua sponte, (353) a
proper application of common law style due process norms would imply that
there would almost never be cases where it could justify doing so. (354)

Some similar issues may have to be dealt with as a result of directions in the lex
arbitri or arbitral rules. For example, if a tribunal is bound to consider trade
usages and no evidence is adduced from either party, a tribunal may naturally
wish to hear argument on this issue. Nevertheless, there is again no reason to fail
to give appropriate warning to the parties and give them a full opportunity to
make their own submissions.

Similarly it is possible for a tribunal to ensure the power to decide on applicable

law through terms of reference agreed to with the parties. For example, Carter
has suggested the following terms of reference:

The arbitral tribunal is to resolve all issues of fact and law that shall arise from
the claims and counterclaims and pleadings as duly submitted by the parties,
including, but not limited to, the following issues, as well as any additional issues
of fact or law which the arbitral tribunal, in its own discretion, may deem
necessary to decide upon for the purpose of rendering any arbitral award in the
present arbitration. (355)

This would lead to a similar position as under the English Arbitration Act 1996,
as there would still be an overriding obligation to advise the parties.

Other questions relevant to the iura debate are simply not addressed in typical
lex arbitri or arbitral rules. These include whether a tribunal can find in favour of
an express claim under a different area of law. The rules also do not indicate
what page "1071" inherent knowledge a tribunal should have about questions of
burden and standard of proof. Nor do they indicate whether a tribunal is entitled
to adopt differing interpretations to that of the parties or take a more proactive
approach to research where it is concerned with the adequacy of the material as
presented. These questions will be returned to when the particular categories
are analysed separately at the end of this section. At this stage, however, it is
important to note that due process norms remain central. To the extent that
there are any rights to vary the analysis from that presented by the parties, the
proper conclusion is that no lex arbitri or arbitral set of rules expressly obviates
the need for the tribunal to advise the parties, at least in appropriate
circumstances. The contrary argument can only be made on the basis that a right
to present does not apply where a tribunal has a duty to know the law.

13.19.2. Iura and Party Autonomy

It is important to also consider the issue from the perspective of consent. Party
autonomy is paramount in international arbitration. For an arbitrator to have
either a power or duty to apply the iura principle, it must be sourced directly or
indirectly from party autonomy, unless one argues that the principle is inherent
in any adjudicatory process. That cannot be so given the differences in view
amongst legal families. One might argue in favour of a consent-based justification
for the iura principle on the grounds that parties expect a correct application of
the law. Surveys have shown that a correct and just outcome is a highly valued
aspect of arbitration, (356) although later surveys have been more equivocal.
Nevertheless, if the arbitrator makes the choice, the duty of care is to ensure that
such law is appropriately applied. A converse argument is that except where
both parties come from jurisdictions applying iura domestically, they cannot
collectively expect that it would be applicable.

Where party consent is concerned, an arbitration agreement itself will rarely

address this issue. Instead, party autonomy will lead to a lex arbitri and perhaps
a set of arbitration rules. As noted above, most lex arbitri give a tribunal a broad
discretion to identify the applicable substantive law but only subject to party
agreement to the contrary. Where parties select arbitral rules or select a lex
arbitri via selection of a Seat, they are also engaging in an indirect choice of
applicable law by determining the methodology of the tribunal's discretion. (357)
Party choice is paramount. Parties may choose any law they wish, including one
not connected in any way with the essential commercial transaction. The only
exception is where such a choice might be aimed at avoiding an applicable
mandatory norm. Even then, the norm might apply alongside the parties'
selection. Parties are also entitled page "1072" to use stabilisation clauses,
locking the applicable law into a historical position and not one in force from
time to time. Because of these fundamental rights, it would simply be a denial of
the parties' autonomous powers to make a selection if a tribunal was to consider
an alternative law without warning the parties of its intention. One can only
direct a tribunal not to do so if one knows the issue is being considered. A
negative choice is still a choice that is an entitlement under party autonomy.
Even in an example where the parties are wrong in their view as to the
applicable law and the tribunal is aware of the correct situation, there is
everything to commend the view that the parties still deserve an opportunity to
either address the tribunal in relation to the nature and application of that law or
to agree after the event that their relationship is not to be governed by that law
regardless of the originally correct position.

13.19.3. Court Review of Arbitral Awards Applying Iura

The next question is how national systems treat the iura principle when dealing
with arbitrations within their jurisdiction. This will show what attitude is taken
to the potential tensions between the most extreme versions of iura and due
process norms addressed above. Here again there are two questions. First, do
review courts assert that there is a duty to know the law? Second, must the
parties be notified before it is applied? One problem in analysing any arbitral
cases dealing with the issue is that they will be either annulment or enforcement
cases. This is because the party who has been adversely affected by any arbitral
decision to apply a law not argued will naturally suggest that the tribunal has
gone beyond its mandate and has failed to honour its due process obligations.
They will thus incorporate an analysis of the iura principle within distinct issues
as to the supervisory control of courts. If a supervisory court takes a position
that respect for arbitration and finality of awards should restrict interference to
only egregious circumstances, this cannot be an ideal way to identify the primary
duties and responsibilities of arbitrators as to consideration of legal issues not
raised by the parties.

With this caution in mind, it can be noted that some earlier cases in civilian legal
systems have been more disposed to the view that an arbitrator must apply the
relevant law ex officio and that such a duty is not circumscribed by an obligation
to allow the parties to make submissions. (358) The Swiss Federal Tribunal has
held that the principle applies in arbitration. (359) Swiss law does not habitually
require full notification as of right, by way of analogy with the position of judges.
(360) page "1073" Nevertheless, Swiss courts have at times overturned decisions

for failure of notice. (361) The Swiss Federal Supreme Court reiterated that Article
182.3 of the PILA does not grant a right to be heard as to legal qualification of the
facts except where a law to be applied by the tribunal could not be reasonably
foreseen by the parties. (362) The Swiss Federal Supreme Court in a decision of 21
September 2007 considered the iura principle in the context of an offshore
transaction which was used to shift profits for tax purposes. In the Bahamian
Offshore Structure case, (363) the Court reiterated that an arbitrator is free to
adopt a different legal characterisation as long as the mandate is not exceeded. A
party's right to be consulted on questions of law only arises where it could not
foresee the legal rule being applied and that it comes as a surprise. Whether or
not it should be seen as causing surprise is a matter of judgment which a review
court will approach with restraint. Nevertheless, Swiss law accepts that the
parties may restrict the entitlement (although they must do so clearly). (364) In the
Bahamian Offshore Structure case, it was conceded that the situation would be
different where the arbitration agreement binds the tribunal to rest the decision
solely on legal principles relied upon by the parties. It is difficult to reconcile this
view with the broader iura concept. If there is a duty on an adjudicator, how can
the parties remove it? If they can do so under party autonomy, when is there
even a power let alone a duty, absent their express or implied wishes to that
effect? Poudret and Besson argue, therefore, that the principle cannot be a
mandatory law of procedure, because the parties and/or the arbitral tribunal
could freely adopt or reject its application. (365) Other countries page "1074" have
also been supportive of the principle even absent notification. A decision to apply
the maxim in a Swedish arbitration was upheld by the Svea Court of Appeal. (366)
It was also applied in the Finnish Werfen case. (367) Other jurisdictions have been
more concerned with adequate notice to the parties. A Quebec Superior Court
annulled an award in the Dreyfus case. (368) The dispute was between a French
and German company under an ICC arbitration with the Seat in Montreal. The
agreement was subject to New York law. One of the issues was whether an
agreement between the parties had been terminated. The tribunal held that it
had not, but considered it had the right to make ‘determinations based on ad hoc
application of broad principles of justice and fairness’ in order ‘to find a just
solution’. It applied a doctrine of ‘frustration of purpose’ through which it
identified a buy-out remedy that severed the parties' relationship. The tribunal
had mentioned the doctrine of frustration to the parties, but no hearing had been
held on it. The Superior Court considered that the tribunal had exceeded the
mandate and failed to allow the parties to address the issue.

A similar approach requiring notice in most cases has been taken in France,
although not uniformly so. (369) Article 12 of French NCCP allows for application
of the principle but the right to be heard is paramount under Article 16. Article
16(3) of the NCCP require that parties be given an appropriate opportunity to
consider and respond to new issues of law raised on an ex officio basis by the
tribunal. The principle must also be applied within the context of the relief as
sought. (370) page "1075" Cases have also been overturned in Belgium, (371)
Singapore (372) and England. (373) A Brussels Court of Appeal has taken the view
that a tribunal may not raise new grounds that were not discussed by the parties
in an adversarial proceeding. (374) The parties must be given ‘a fair opportunity to
address its arguments on all of the essential building blocks in the tribunal's
conclusions’. (375) English courts have concerned themselves with the procedural
question and have resolved that tribunals should not apply legal provisions that
are not addressed by the parties. (376) English courts have, however, taken the
position that not all inferences from evidence need to be drawn to the attention
of the parties. (377)

This brief survey of courts supervising arbitral behaviour shows a number of

things. First, there is no consensus between legal families as to whether the iura
principle should apply to arbitration. Second, the vast majority of systems
expressly or impliedly support due process notification rights even where iura
applies. This is expressed in some civil codes. It is necessarily implied in systems
that direct that the parties should not be taken by surprise. Finally, most if not all
systems would accept that the parties can remove an arbitral power to decide
issues sua sponte. No arbitral review system appears to present a consistent or
reasoned justification for ignoring party autonomy and limiting mandatory
procedural due process norms to situations other than those where iura applies.

page "1076"

13.19.4. International Law, Investment Arbitration and the Iura Principle

Some guidance might be gathered from commentary and cases on this issue in
the field of public international law. Iura novit curia has been applied by the
International Court of Justice (ICJ), (378) the World Trade Organization (WTO)
Appellate Body, (379) the Inter-American Court of Human Rights and the European
Court of Human Rights. (380) Judge Mohamed Shahabuddeen, formerly of the ICJ,
has stated that ‘it is clear that, where an issue has been raised, the Court may
competently consider all pertinent arguments and authorities, even if not
presented by the parties'. (381) He said ‘whether or not there has been argument,
the principle iura novit curia requires [the ICJ] to satisfy itself of the legal
proposition before announcing it or acting on it’. (382) In the Nicaragua case, the
ICJ considered that it was for it to interpret and apply relevant rules of
international law, as ‘the law lies within the judicial knowledge of the court’. (383)
Such powers could also be impacted upon by the view a court takes as to its
inherent jurisdiction and the powers flowing from the essential nature of the
adjudicatory function. A number of broad comments have been made by and in
relation to the ICJ in that regard. For example, in the Nuclear Test case it
observed that it:

possesses an inherent jurisdiction enabling it to take such action as may be

required … to provide for the orderly settlement of all matters in dispute, to
ensure the observance of the ‘inherent limitations on the exercise of the judicial
function’ of the Court, and to ‘maintain its judicial character’. Such inherent
jurisdiction, on the basis of which the Court is fully empowered to make
whatever findings may be necessary for the purposes just indicated, derives from
the mere existence of the Court as a judicial organ established page "1077" by
the consent of States, and is conferred upon it in order that its basic judicial
functions may be safeguarded. (384)

In a further case, and in a separate opinion, Judge Higgins considered that:

the court's inherent jurisdiction derives from its judicial character and the need
for powers to regulate matters connected with the administration of justice, not
every aspect of which may have been foreseen….

She further noted that the court has ‘inherent power to protect the integrity of
the judicial process’. (385) Such broad statements say little about the elements of
the powers that flow as a result and whether iura principles would come under
this heading. Pauwelyn suggests that inherent jurisdiction of any international
tribunal includes the power to interpret the parties’ submissions to isolate the
real issues and the object of the claim, determine its competence to hear a
matter, determine whether to refrain from exercising jurisdiction once
established and deciding all issues pertaining to substantive jurisdiction
including burden of proof, due process and other general international rules on
the judicial settlement of disputes. (386) Other examples would include estoppel
and abuse of rights. The ICJ considers that acquiescence and estoppel follow
from fundamental principles of good faith and equity. (387) Brown cautions that an
international court ‘cannot claim to possess an inherent power if that power is
not necessary for the performance of its particular functions’. (388)

Investment arbitration cases have also considered the question. In Klockner v.

Cameroon, an ICSID annulment committee considered that the issue was
‘whether, by formulating its own theory and argument, the Tribunal goes beyond
the ‘legal framework’ established by the Claimant and the Respondent’. (389) It
noted for example that a tribunal could not decide on the basis of tort if the plea
was based on contract. Conversely, if the tribunal remains within the ‘legal
framework’, it ‘must be free to rely on arguments which strike them as the best
ones, even if those arguments were not developed by the parties (although they
could have been)’. (390) Other cases have held that the parties should be notified
of new legal sources and have the opportunity to comment on them. Subject to
that due page "1078" process right, the tribunal can ‘give the legal qualifications
and determine the legal consequences that it deems appropriate, even if they
were not pleaded by the parties’. (391) In the annulment proceedings in Mitchell v.
Congo, the ad hoc Annulment Committee concluded that considering other treaty
provisions exprorio motu or utilising jura novit curia is permitted but not
obligatory. (392) In Dallal v. Iran and BankMellat, Judge Holtzmann in his
dissenting opinion considered that the Chamber ought to have refrained from
relying on arguments flowing from the IMF Agreement. He referred approvingly
to a speech of Judge Sir Robert Jennings, then of the ICJ, who thought it
inappropriate to surprise the parties in this way. (393)

Taken together, these comments add some support for the iura principle in
international investment arbitration. Application by international courts and
tribunals can generally be explained on the basis that these are international
adjudicators applying international law who cannot operate in isolation from
general international law principles regardless of the matters pleaded. Such
cases are particularly concerned with correct application of law, as they have
significant implications for third-party citizens and States. Another reason to
engage in a more fulsome inquiry is where the decision is likely to be treated as a
precedent elsewhere. (394) There may thus be a greater reason to support
application of the principle in investment cases. Even so, the cases have not
articulated a coherent rationale against an obligation to notify issues of concern
and afford parties the opportunity to make submissions on them.

13.19.5. Recommendations

Given the lack of express determinations in typical lex arbitri, the proper
approach requires a consideration of various powers and duties as they would
apply in a range of circumstances, rather than an all or nothing debate about the
iura concept itself. Because of potentially conflicting duties, it is understandable
that the area is contentious, although it ought not to be if appropriate
consideration is given to due process as the magna carta of international
arbitration. (395) From this perspective, the proper approach ought to be to
consider what kind of question is being determined. Is it a question of
jurisdiction; potential application of mandatory laws; page "1079" interpretation
of a statute or treaty; interpretation of a contract term; consideration of a
contract term other than one argued by the parties; re-characterisation of facts
within a chosen legal system; application of a defence not raised; or, application
of a differing area of law to the facts as found?

Regardless of one's view on the controlling impact of consent, if the parties

consent to the behaviour proposed by the tribunal, that will usually be a
complete solution. Consent of the parties ought to prevail in any area where iura
may be conceivable. If the parties are happy to agree on an incorrect law
applying, it is the same as if they modified their contract ex post facto as they are
entitled to do. Conversely, the parties could agree that they are themselves not
required to proffer arguments as to any areas of law, but instead, rely on the
legal expertise and research of the adjudicator. Even then they cannot compel an
already appointed arbitrator if this agreement post-dates the tribunal
constitution in circumstances where this would require extra work by the
arbitrator over and above what would reasonably be contemplated under the
terms of the initial appointment. An arbitrator should know at the outset what
the mandate is.

Thus if consent is paramount, absent bad faith or public policy concerns, the iura
debate can best be seen as one of the many contentious questions in
international arbitration that flow in practice from a lack of clear guidance by the
parties. If the parties both determine that they do not wish the arbitrator to
behave as postulated, again an arbitrator cannot ignore party consent unless
some justification can be found emanating directly or indirectly from the lex
arbitri. If the parties disagree amongst themselves, attention will need to turn to
the tribunal's powers and discretions as per the lex arbitri. One difficulty with
such a methodology is that the powers and duties of an arbitrator are not
generally expressed in clear and unassailable language. Instead, they are either
set out as broad propositions, the application of which can be in dispute, or may
need to be implied from other principles and arguably may conflict. Relevant
duties pertaining to this area discussed above and in Chapter 2 include the duty
of due process, including giving parties an equal opportunity to present their
case, a duty to complete the mandate and not exceed it, a duty of competence and
due care, a duty to render an enforceable award and a duty to remain impartial
throughout. Other possible duties include the duty to respect party autonomy,
the duty to educate and the possible power to override the procedural wishes of
the parties. The issue is also impacted upon by the view one takes as to the duty
or entitlement of a tribunal to be proactive as discussed in Chapter 6. The more
one supports a proactive approach, the more one accepts that a tribunal should
ensure that appropriate legal principles are brought to bear on the dispute,
regardless of the adequacy or otherwise of the parties submissions.

This book argues that iura should not even be described as a broad power,
notwithstanding suggestions to that effect. Isele advocates a mandatory
discretion in favour of tribunals. (396) The difficulty with a discretion of this
nature is that a tribunal will effectively know that it has the power to make a
party win or lose page "1080" depending on whether the discretion is exercised
or not. If there is to be such a discretion, there then needs to be some legitimate
criteria as to why it might apply. Suggestions such as the seriousness of the
matter, the amount involved, or disparity in expertise between the parties would
all be difficult to apply in practice at the margin and are hard to justify as reasons
in and of themselves. Furthermore, if there are concerns about disparity and
ability between the parties, a tribunal can deal with this through proactive
involvement, including in extreme circumstances, inviting a party to consider
certain potential arguments. That itself is contentious, but less so than applying
that very law without notice or submissions.
The vast majority of jurisdictions and commentators are of the view that an
arbitrator who believes there is a power to consider other principles of law than
those cited in submissions should at least warn the parties that he or she is
inclined to do so and allow them to make submissions accordingly. This view is
to be commended. Any contrary view offends against the duty to give parties an
opportunity to best present their case, given that even legal systems that do not
demand legal submissions by parties still allow them to make them. It would also
offend against party autonomy and the entitlement to direct the tribunal as to
the attitude it should take to this question. Strong criticism from scholars such as
Fouchard, Gaillard and Goldman are a proper response to the various concerns
outlined in this article. For example, in considering a practice of not addressing
applicable law under the iura principle, the authors suggest that ‘(t)his practice -
which is unsatisfactory even in court proceedings - should be rejected in
international arbitration’. (397) They make the valid point that a logical corollary
of the principle would be to allow one party to make submissions about law that
are not communicated to the other. (398) There certainly should not be any duty to
apply the principle as such.

The issue was considered in depth by the International Commercial Arbitration

Committee of the International Law Association in a report published in 2008. Its
support of a notice requirement is again commended. Its comments identify the
key concern supporting any iura notion, namely, that arbitrators should not
apply the wrong law when such incorrect or incomplete notions underpin party
submissions. The Committee concluded:

Arbitrators who attempt to develop legal issues in a strict application of the iura
novit curia approach risk taking the arbitration from the parties and appearing
partial. Yet arbitrators who completely fail to seek clarification of legal issues in a
strict laissez faire approach risk not having sufficient guidance when they go to
render an award, or risk making an award that is incorrect on the legal issues in
question. Accordingly, arbitrators should inquire about the applicable law within
the general parameters of the arbitration defined by the parties and, considering
costs, time and relevance of issues, may conduct their own research, provided
the parties are given an opportunity page "1081" to be heard on material that
goes meaningfully beyond the parties' submissions. (399)

In its Recommendations on Ascertaining the Contents of the Applicable Law in

International Commercial Arbitration, the Committee included the following:

Recommendation 7: Arbitrators are not confined to the parties' submissions

about the contents of applicable law. Subject to Recommendation 8, arbitrators
may question the parties about legal issues the parties have raised and about the
submissions and evidence on the contents of the applicable law, may review
sources not invoked by the parties relating to those legal issues and may, in a
transparent manner rely, on their own knowledge as to the applicable law as it
relates to those legal issues. (400)

ILA Resolutions 6 and 13 suggest that new issues of law that arise when
ascertaining applicable law should not be introduced by the tribunal save for the
special case of transnational public policy. To similar effect, Kaufmann-Kohler
recommends that a tribunal should clarify at the outset that the parties shall
establish the content of law, that the tribunal may make its own investigations
but not be under an obligation to do so and if it does so, it shall consult with the
parties about the result. (401)

The following sections complete the analysis by looking at the particular

permutations of issues where the iura principle has been contemplated and seek
to draw conclusions in each case as to the proper approach that a tribunal should
take.

13.19.5.1. New Claims as to Relief

There is a difference between questioning what a tribunal may decide and how it
is to decide it. As to the first question, virtually all agree that a tribunal ought not
to determine an issue not posed for its consideration. However, once an issue has
been posed, a tribunal ought to resolve it in the most appropriate manner. Most
would also agree that a tribunal should not alter the claims as made. This goes so
far as to suggest that if an unsustainable claim is made in contract which could
have succeeded in tort, an arbitrator ought not to suggest the alternative. (402) It
may even be that the tort claim would be outside the terms of the arbitration
agreement where page "1082" that is restrictively drafted. An arbitrator should
also not add to the terms of the relief as sought. For example, if there are no
claims as to interest or costs, this ought not to be raised by the tribunal. (403) If
application of the principle changes the relief, this should also be improper. The
Werfen case referred to above can be criticised on this basis as it actually found a
different form of compensation than that pleaded, albeit at a lesser amount.

At times a tribunal may not need to strictly limit itself to the relief sought if it can
still stay within the essence of that which was claimed. An example would be a
claim for a declaration that money is owed and an award that actually calls for
payment. The parties would have had a full opportunity to present arguments on
the key issues. There should be no due process concern in these circumstances.
Similarly, at times the facts might be recharacterised without due process
concerns. An example would be an allegation of failure to mitigate where the
tribunal thinks the facts show lack of causation as a result. Again due process
requires notice and an opportunity to address the tribunal on this differing
perspective. The better conclusion overall is that, where new claims are
concerned, either a tribunal has no power or, in limited circumstances where it
may clarify, it must do so within full notions of due process rights of the parties.

13.19.5.2. Defences

Identical issues arise in relation to defences as arise in relation to new claims.

Because it is accepted that tribunals are not to determine issues not posed for
consideration, that would be so with a potential defence not raised. The
dominant view, at least where both parties are represented by counsel, is that an
arbitrator should not raise the issue of a limitation defence. (404) The situation
may be more complex with certain kinds of potential defences. An example
would be a defence that can be recast as a factor to be considered in the primary
claim. For example, as noted above, a failure to invoke the notion of mitigation of
damages is more complex, as this could be built primarily into an analysis of the
quantum of damages that can be said to be caused by the breach, as opposed to
the innocent party' s responses to the breach. Once again, there would be no
justification for addressing any issue under this heading without giving
appropriate notice to the parties and giving them an opportunity to make full
presentations.

13.19.5.3. Characterisation and Re-characterisation

Civilian systems that support iura suggest that one area of its application is
where the adjudicator might re-characterise the transaction to afford the same
relief but page "1083" under a different legal principle. There is a need to
distinguish between what was or was not claimed and what is applicable law to
the claims as presented. It is in this context that differences between civilian and
common law legal families are starkest. The ILA Report concludes that both
common law and civil law systems accept that facts are to be pleaded and proven
by the parties. An adjudicator is not to decide on issues that are not raised by the
parties. It noted, however, that there were differences between legal systems as
to the way an adjudicator should treat causes of action and the status of law in
civil litigation. (405) In some civilian jurisdictions, the court may re-characterise
the facts as part of the iura process and substitute a different cause of action
and/or legal reasoning to that pleaded. Considering that the tribunal's power
extends to any law within the relief as claimed becomes particularly problematic
when pleadings include a broad claim for such other and further relief as the
tribunal determines. (406)

Because of differences in view between legal families, the ILA recommendation

that the parties' views are to be sought, with a warning that a tribunal might
make its own inquiries in such circumstances, is the preferable one, although it
still should be the case that once a tribunal knows which law it may apply, the
parties should be given an opportunity to respond in relation to that as well.
Generally speaking, re-characterisation involves some questions of fact, for
example if parties are seen to be in a partnership or fiduciary relationship. These
are legal conclusions depending on key factual determinations. Hence, an
adjudicator ought never to be given the sole right to consider such matters
absent an opportunity for the parties to present evidence and arguments on the
issue.

13.19.5.4. Ambiguous Relief Claimed

Iura is not to be confused with situations where the prayer for relief or terms of
reference may be ambiguous and there is a need to interpret what was in fact
claimed. A tribunal must clarify ambiguities and make determinations where
necessary to adhere to the mandate and to comply with the duty of due care.
Seeking such clarification will naturally support due process rights as well.
Nevertheless, if the duty or power arises from a proactive entitlement to clarify
ambiguous claims, this may give an unfair advantage to less competent counsel,
particularly where a passive adjudicator could validly assert that the burden has
not been made out, in part because of the uncertainty.

page "1084"

13.19.5.5. Selection of Applicable Law

As noted above, the ILA Report concerned itself with selection of applicable law.
In that context it noted that some legal systems allow parties to present facts,
articulate relief and leave it for an adjudicator to determine the relevant law and
the way it ought to apply in the circumstances. As argued above, this scenario
should not be dealt with under a broad notion of iura but should be dealt with by
considering the way a tribunal's powers to determine substantive law are to sit
alongside a duty to respect party autonomy and due process obligations. At the
very least, this will lead to proper notification to the parties to give them an
opportunity both to make representations and also select applicable law if they
have not done so previously.

The most common situation is where no determination as to applicable law has

as yet been made. If there is no choice of law clause and no application for a
preliminary determination, the parties will address the tribunal on a range of
potentially applicable laws. If the tribunal has a broad discretion in that regard,
counsel submissions are merely aimed at swaying the tribunal as to the way that
discretion ought to be exercised. A tribunal that believes a different law applies
ought not to be constrained from doing so. There is still no reason not to warn
the parties of this at the earliest opportunity. The decision the tribunal makes
must be a reasoned one which earns the respect of the losing party. That is so
whether the direct method or a conflicts of law rule is utilised. (407) Here in
particular, a failure of timely warning would mean that the parties have not been
given an adequate opportunity to present their cases in the context of the
applicable law. This is particularly so where selection may involve the
consideration of factual matters such as the connecting factors to particular
jurisdictions. In some cases, once the parties are aware of the applicable law, this
will require extra research or perhaps nuanced arguments on relevant facts. For
example, some legal systems may raise defences of unfair contract provisions.
Other relevant factors that might require consideration and factual argument by
the parties include ‘practical considerations such as the underlying economic
objectives of the contract in dispute, the legitimate expectations and interests of
the parties – especially in upholding the validity of the contract and the possible
consequences of their choice of law decision for the outcome of the case …’ (408) If
the parties accept the relevance of the law as considered by the tribunal, and
address it accordingly, there is no problem. If the parties simply align their
submissions to the law suggested by the tribunal, that could be seen as an
implied agreement on their part. (409) It is important for the tribunal to be clear as
to its intentions in such circumstances. A mere reference to a possible applicable
law might find that the tribunal subsequently considers it to be inappropriate.

page "1085"
If the tribunal makes the parties aware of its preliminary thinking, the parties
might also decide to make a selection of applicable law at that stage or at least
constrain the tribunal from certain options that it might otherwise be disposed
towards. If there is no mandatory norm involved and the parties are simply
making a selection of law, the tribunal ought to be bound to accept it. The same is
so where parties agree that certain laws are not to be applied, leaving the
tribunal with a restricted discretion. (410) The situation is also simple where the
parties disagree as to which law should apply. Here a tribunal is simply entitled
to make the determination that it thinks is most appropriate.

13.19.5.6. Wrong Law Cited for Relief

It sounds intuitively correct to say that a tribunal should be concerned if it is

asked to apply the wrong law. Yet in the context of party autonomy, it is not easy
to understand what that means and how readily it may apply. For example, if the
parties direct the tribunal to apply a law it thinks is not appropriate, that could
be seen as a subsequent agreement by the parties which circumscribes the
tribunal's mandate. The tribunal should not ignore such a direction by the
parties if it is truly based on informed agreement between them. In the extreme,
if the tribunal considers that this is an undue limit on the mandate originally
accepted, it might constitute grounds for valid resignation. That would rarely be
so as the parties are entitled to select a law that otherwise has no connection to
the transaction. If they can do so in their original contract, there is no reason why
they should not be able to do so during a hearing after being notified by the
tribunal as to the law it would otherwise wish to apply. In such circumstances,
the tribunal is not being asked to apply the wrong law. The parties have simply
made a late choice as to the correctly applicable law.

The situation is different if the parties are not agreeing on a particular law but
are independently forming a legal conclusion as to which law ought to apply
where the tribunal simply does not accept the reasoning of either. Situations
where the parties both cite the wrong law would typically arise where there is a
statutory provision designating the law that they are unaware of. Perhaps the
most common example is the application of the Vienna Convention on the
International Sale of Goods (CISG). (411) Countries that have adopted this
Convention have laws that stipulate that the provisions automatically apply
where the preconditions are met unless the parties opt out. Thus such signatory
countries have a domestic contract law, and another one presumptively
applicable to international contracts for the sale of goods. Many lawyers and
traders are unaware of the Convention and its automatic application. An
experienced arbitrator might well know that the relevant countries are
signatories to the CISG and, in the absence of an express page "1086" choice of
law clause, its provisions apply as a matter of course. There seems good reason
to support the view that an arbitrator must apply the correct law absent any
directions from the parties to the contrary. A failure by the parties to opt out
could even be described as objective evidence of intent, notwithstanding that in
the instant case this may have been based on ignorance. Thus a tribunal,
knowing it is bound by a choice of the parties, could conclude that an express
contract between persons from countries that both impose the CISG as of right
except where excluded, have chosen to accept that position where no exclusion
has occurred. Even if that is not the case, an arbitrator with a broad discretion as
to applicable law ought to select the law that both countries believe to be
applicable, absent any contrary express wish of the parties. Even utilising a
conflicts methodology will typically show a false conflict where all countries
apply the CISG or in other cases will lead to a selection of a law that again makes
the CISG apply as of right. Thus the duty to respect the lex arbitri, implied
consent and duties of due care would all suggest that a tribunal should act
accordingly and apply the CISG in such circumstances even if written
submissions wrongly presume other laws apply. The duty to render an
enforceable award would also be relevant as a failure to apply the correct law
might be seen to be a failure to adopt a proper procedure under the lex arbitri,
bringing Article V(1)(b) of the New York Convention into play or might be said to
be contrary to public policy per Article V(2)(b).

Nevertheless, there is simply no possible justification for failing to give the

parties adequate warning. Allied to the duty of due care, a tribunal should be
looking at the materials as soon as they are to hand to ensure that the procedural
steps are shaped appropriately. While some cases may be more complicated, in a
simple sale of goods dispute, a tribunal will know the terms of the relevant
contract from the very first submission, the residence of the parties and whether
the contract is indeed a sale of goods transaction. In those circumstances, any
experienced arbitrator would consider the potential applicability of the CISG if
the parties come from signatory countries. It may even apply in broader
circumstances. Were it to apply and were the early submissions dealing with
different bodies of law, this should be raised at the outset. Informed parties could
then opt out if they wished, which would be a determination binding on the
tribunal.

If an arbitrator only thinks of the potentially applicable law after the hearings are
closed, there seems no reason not to reopen them to give parties at least a short
opportunity to make submissions. Because of duties of due care and diligence, a
failure to consider the CISG at an earlier stage is not only a failure of counsel but
is also a potential failure of the arbitrator. More complex would be a situation
where the parties would want to have expert witnesses giving testimony about
this new area of law. Even then this can be done urgently in writing and could be
based on specific questions asked by the tribunal and only where necessary to
the outcome. If the parties are in agreement in reopening factual issues, there is
no problem. The only concern is if the parties are in disagreement. How a
tribunal would behave in a particular case would depend on the circumstances.
The less important the differing law to the ultimate outcome, the less elaborate
the post hearing procedure needs to be. But the overriding principles should be
that the correct law should be page "1087" applied in such circumstances absent
an agreement to the contrary and the hearing should be reopened if the tribunal
does not believe that due process rights have been afforded in relation to that
law. On this view it is difficult to allow arbitrators to simply rely on their own
personal knowledge of some domestic or international legal system, typically
contract law, and apply it without the benefit of evidentiary submissions as to
primary and secondary sources and perhaps expert witness engagement with
the key issues.

13.19.5.7. Considering Other Parts of Law Cited and/or Other Contract

Clauses

If the substantive law has only been partially addressed and is inaccurate
without reference to other provisions, then for similar reasons, a tribunal should
apply the law correctly. Once again, however, there is no reason not to warn the
parties and give them an opportunity to make submissions. The same principles
apply where other parts of the relevant contract undermine submissions as
made. Wetter refers to an example of a claim for lost profits under a contract
which expressly prohibits such damages and instead limits a claim to expenses
incurred. The better approach in such circumstances is for the tribunal to invite
the claimant to consider why the relevant provision does not bar the claim. A
simple reading of the clause would at least show that they ought to amend the
claim to cover the expenses. (412)

13.19.5.8. Mandatory Laws

A subset of the question as to whether there is a duty to apply the law is whether
that varies depending on the type of law involved. The most obvious example
would be a law that purports to apply regardless of the will of the parties or a
tribunal. This relates to mandatory laws. While a previous section has
considered that party consent may be determinative, that will not be so where
mandatory norms are concerned, although this is a controversial area. While
mandatory procedural norms of the Seat must always apply, more contentious is
whether a procedural norm of other jurisdictions can ever be mandatory. This is
quite a distinct question to that of iura although they overlap in practice. Most
would take the view that a purported mandatory procedural norm from another
jurisdiction, such as a potential enforcement country, is not binding on an
arbitrator, although in cases where it is not difficult to comply with such a norm,
a tribunal might do so as part of a duty to render an enforceable award. (413) Even
in this contentious category, there is little problem in the context of this section
as the parties will be notified of the procedure and given a full opportunity to
present in relation to its relevance and under its norms were they to apply.

page "1088"

An even more contentious question is whether arbitrators must apply

mandatory substantive laws in all circumstances. This was discussed in section
13.8. Examples include situations where a tribunal would be aware that the
European Court of Justice (ECJ) believes that failure to apply European
competition law may render an award unenforceable. (414) Once again there
seems no reason to fail to give parties an opportunity to address an issue. If a
tribunal thinks about the issue in time, which ought to occur under duties of
diligence and due care, then the parties will naturally be asked to address the
issue. If the tribunal only thinks about it after the proceedings are closed, the
flexibility of international arbitration procedure would make it very easy for the
parties to be invited to make urgent written submissions on a matter of concern
to the tribunal.

13.19.5.9. Invalidity of Laws

If the parties have purported to make a selection of law that would be invalid or
may be thought so by a tribunal, broad duties under the lex arbitri in relation to
enforceability and public policy and duties of due care and good faith might
suggest that a tribunal could adopt a differing law. Once again, there is no reason
not to advise the parties as soon as possible. In particular, a hypothesis that the
parties have selected an inappropriate law for a bad faith purpose, perhaps to
avoid some provisions for the benefit of revenue avoidance, raises important
factual questions that the parties should be entitled to address. An example
might include a circuitous set of contracts thought to be designed to route a
commercial transaction through a tax haven or so as to come within a favourable
investment treaty. Upon being challenged, the parties may have valid
justification for doing so in what otherwise might appear to be questionable
circumstances. Even if they do not have any valid justification, an adverse finding
by the tribunal should only arise after they have been given an opportunity to
make representations.

A duty to know the law may also arise in a situation where a tribunal has
suspicions about a legal principle asserted by both parties. An extreme example
would be where there is a suspicion of illegality and both parties assert that the
transaction is fully above board. (415) Arbitrators may also feel bound to apply
relevant law when they consider that transactions are contrary to international
public policy, for example where bribery is involved. Here again the very notion
of page "1089" bribery is based on a factual allegation about which the parties
may have differing submissions or corroborative or countervailing evidence.

13.19.5.10. Reading Authorities Fully

It was suggested above in relation to submissions that only deal with parts of
relevant statutes or contracts that a tribunal is obliged to make a correct
assessment and should be entitled to address those other elements that would
alter the result. The same logic ought to apply with selective reference to
authorities. A tribunal ought to be able to refer to other parts of a judgment
submitted as relevant by both parties, but where they have not addressed those
comments in their written submissions. The tribunal is obliged to consider the
entire judgment in context to understand its meaning.

Nevertheless, if submissions have been allowed from both sides in relation to

certain parts of the judgment and others seem to alter the meaning of the
authorities as contended, there seems little reason not to give each party an
opportunity to address the tribunal's concerns. Once again, if a tribunal is
reading the material as it comes to hand or at least before the final hearing,
parties can be warned as soon as they make incomplete submissions. In
circumstances where a tribunal only notices something after the proceedings are
closed, a short opportunity to make follow-up written submissions adds very
little in the way of unjustifiable costs. It may even be that if the fault was on the
part of the successful party, some adjustment as to cost entitlement could flow as
a result.

13.19.5.11. Referring to Other Authorities

A related but more complex situation to that of the previous sub-section is where
a tribunal researches some of the cases cited in such an authority to better
understand the principles articulated. A similar situation is where a tribunal
reads some secondary sources for the same purpose. In these circumstances, if
the tribunal can legitimately say that the parties have been given an adequate
opportunity to make submissions on the principle to be applied, then additional
work by the tribunal does not interfere with their due process rights. (416)

A distinction might need to be drawn between situations where the research

merely confirms the submissions made and simply bolsters the quality of the
reasoning in the award and conversely, where the research undermines the
submissions as made. In the latter event, prudence would again suggest that the
parties be invited to address the contradictory authorities. Just as the tribunal
might have found fault with the research of the parties, it may be that after
adequate notice, they could show that their position is the preferred view for one
reason or another.

page "1090"

13.19.5.12. Different Interpretation

An arbitrator should not be constrained in interpreting applicable law simply

because of limited arguments by counsel. Adjudicators must interpret statutes
and treaties correctly even if the parties wrongly assert the meaning. For
example, if both counsel only address a tribunal in relation to plain meaning of a
bilateral investment treaty, the tribunal cannot be constrained thereby from
exercising a purposive approach to interpretation as permitted under the Vienna
Convention on the Law of Treaties. Once again, however, there is no reason for a
tribunal to not warn counsel that they may need to address the tribunal on this
issue.

13.19.5.13. Special Areas Where Notice is Not Needed

While this section has argued strongly that the parties should be notified of the
tribunal's inclination in virtually all circumstances, there are some areas where a
tribunal is expected to know the law and all would understand that law so that
even if a question is not expressly raised with the parties, there should not be
any entitlement to a challenge. An example would be in relation to a question as
to the burden of proof. Many cases are heard without anyone referring expressly
to who has the burden of proof as this is presumed by all. A tribunal has the duty
to apply the burden and standard of proof correctly and any failure by the
parties to address the issue is better understood as an implied agreement that
the issue speaks for itself. If the tribunal applies it incorrectly then any review
rights should be considered in that context. All legal systems place the burden of
proof as to facts on the party seeking to assert a proposition. While judicial
behaviour may vary as between passive or inquisitorial modes, this does not
alter the basic burden as to factual matters.

13.19.5.14. Jurisdiction

One special case where a duty to know the law may arise is as to a duty to ensure
valid jurisdiction. (417) Courts and tribunals will commonly raise questions of
jurisdiction even if not argued by the parties. This could be justified on the basis
of the duty to complete the mandate and not exceed it. For example, is there a
duty to know rules of capacity or rules as to when third parties can be brought
in? Wetter argues that a tribunal ought to take the same approach as the New
York Convention in allowing ex officio consideration of the matters in Article
V(2), that is arbitrability of the dispute and the role of public policy. He suggests
that a tribunal should act on its own motion and assume total inquisitorial
powers to decide jurisdictional aspects of the case regardless of the defences
raised. The tribunal must also verify the correct identity of the parties, be
satisfied that they are properly page "1091" represented and that they have
appropriate standing. (418) In most cases, however, that is only a limited
responsibility calling for some prima facie satisfactory documentation. (419) That
can be particularly important in situations where there may be a group of
companies. Nevertheless, this should come with full disclosure to the parties. (420)

13.19.5.15. Amiable Compositeur

The principles ought not to be different when an arbitrator is empowered as

amiable compositeur, where there is thought to still be a need to consider
applicable law and then decide to what extent it should be tempered under this
approach. (421)

13.19.5.16. One Party Not in Attendance

A tribunal may wish to exercise such powers to properly test the allegations
made when one party refuses to attend. At least in ex parte proceedings, Wetter
argues that a tribunal's duty could go so far as to request a claimant to produce
additional documentation and legal argument and not simply make a
dispassionate ruling as to whether a prima facie case was made out or not. (422) In
BP Exploration Co, Judge Lagergren considered that he was entitled to conduct
legal research and evaluate claims in the absence of one of the parties. (423) In
such circumstances the tribunal is naturally giving warning to the party in
attendance and under broad duties to communicate, would be notifying the
other party of each key step in any event. Default proceedings are discussed
further in section 6.15.

13.19.5.17. Iura and Proactivity Generally

As the previous sub-section dealing with default proceedings implies, the

question whether a tribunal deciding when to itself seek to identify applicable
law rather than rely on the parties is a subset of proactivity generally. In relation
to identifying applicable law, the ILA Report 2008 noted a number of practices
where tribunals may be more proactive. These included where the solution from
the rules invoked by the parties appears manifestly wrong, where there is an
imbalance between the page "1092" parties regarding knowledge of the law and
where at least one of the tribunal members is an expert on the applicable law.
The Report neither criticised such practices nor wished to incorporate them into
recommendations. (424)

13.20. Analysis and Interpretation

In addition to considering what laws apply, an important question is how their

content will be proven and analysed. As to proof, one issue is whether legal
experts should be appointed by the parties and/or the tribunal. This is discussed
further in sections 10.6 and 13.16 and Chapter 12. Use of legal experts is not a
uniform position, as some legal systems noted above see an inherent duty in
adjudicators to know the law in a more direct manner.

Once the applicable law is known, there will obviously be many factual and
interpretational questions thrown up by its norms. It is also important to
understand that the methodology of legal analysis will vary between different
applicable systems. Common law systems tend to see counsel rely on analysis of
cases in making inductive submissions and arguments by analogy in relation to
the meaning of primary rules or principles otherwise applying. The civilian
tradition concentrates more on the words of an applicable Code, aided by the
insights in leading scholarly commentaries. Differential views about the value of
scholarly commentaries are thus also important.

An important question rarely discussed is whether an arbitrator ought to take a

more conservative approach to interpreting substantive law than would be the
case with a judge in the particular jurisdiction whose law applies. In one sense,
an arbitrator's role is inherently different to that of a judge. This is particularly
so when compared to common law judges. In appropriate circumstances, the
latter are entitled to develop principles of law. An arbitrator dealing with a
particular common law country’s substantive law is not entitled to try and
interpret it in ways which are felt more sound from a policy perspective, but
which are beyond a position already taken by domestic judges. However, where
there are differences in view among domestic judges, an arbitrator ought to have
leeway to select amongst the options. Even here the arbitrator is not seeking to
identify the preferred systemic outcome as may be the case with an appellate
judge in a common law jurisdiction. The arbitrator is instead trying to make the
most accurate determination of what the emergent legal position actually is
under that substantive law. This is particularly so given the fact that foreign law
under common law approaches is in many cases determined by arbitrators as a
question of fact to be proven if necessary by experts and primary and secondary
sources. (425) Civilian judges are more likely to take an expansive approach to
interpretation of foreign law where they see it as page "1093" their obligation to
know the law and where they have such expansive powers in interpretation of
their own domestic laws. Conversely, if a common law judge is merely looking
for proof of the law by way of fact, it is trying to understand what the law
actually is at a particular point of time.

More contentious is the question whether a tribunal can develop its own
substantive legal principles. Veeder uses the development of the law as to the
awarding of interest in arbitrations as an example of such an approach. (426)
William Park suggests that arbitrators ‘fix their eyes on existing legal norms in
determining what the parties have a right to expect’, rather than engaging in
social engineering. (427)

An arbitrator may also wish to consider interpreting the applicable law in the
context of international practice and developing norms. While the parties may
have selected a particular national system, many systems draw inspiration from
other systems where international matters are concerned with a view to
harmonisation, consistency and predictability. (428)

Where the content of applicable law is a factual question, an issue may also arise
as to whether the intent was to apply the law as stated at the time of the contract,
or as applicable at the time of the dispute where there are differences between
the two. Party autonomy would allow them to clarify that in any event. An
example of a preference for the first situation is the use of stabilisation clauses
indicating that the applicable law is that which pertained at the time of the
contract. The validity of stabilisation clauses is discussed further in section
13.12. In the absence of any evidence of a direction by the parties, the better
view is that selection of the law is to be that in force at the time of the dispute.
This is because parties would expect that modifications to the law will generally
improve and clarify issues and therefore be for their overall best interests,
ignoring the peculiar features of the dispute not known at the time of
contracting. That would be the purported intent of the law in any event and it
would be for the tribunal to determine whether new laws are only intended to
have prospective effect.

page "1094" page "1095"

1 ILA Report, International Law Association, International Commercial

Arbitration Committee, Final Report: Ascertaining the Contents of the Applicable
Law in International Commercial Arbitration, Reporters: Geraldo Vidigal Meto
and Jonathan Cardenas (2008): 16. The report notes only that it may have a
substantive element, in terms of the fallback applied when the content of
applicable rules cannot be sufficiently ascertained. Even this might be described
as a procedural method of identifying a fallback.
2 Some will of course resolve similar questions by confining themselves to

evidence of consent and the implied intentions of the parties from the outset.
3 Classification is sometimes described as ‘characterisation’, or ‘qualification’.

Veijo Heiskanen, ‘And/or: the problem of qualification in international

arbitration’, Arbitration International 26, no. 4 (2010): 441.
4 Ibid., 444.
5 Gary B. Born, International Commercial Arbitration (The Hague, Kluwer Law
International, 2009), 2148–2149.
6 See sections 3.6–3.12.
7 see section 3.2.2.
8 see section 7.2.5.
9 For a case where the tribunal consider it was unnecessary to identify the

applicable law see ICC Case No. 1990 of 1972, YBCA III (1978) 217.
10 ILA Report 2008:7–8. Sections 10.6 and 13.16 note differences between

jurisdictions that leave foreign law to be proven by the parties, at times with
default to national laws (e.g., England and France) and conversely, systems
which mandate that the adjudicator is to establish the contents of foreign law
(e.g., Swiss and US law). Gabrielle Kaufmann-Kohler, ‘Globalization of Arbitral
Procedure’, Vanderbilt Journal of Transnational Law 36, no. 4 (2003): 1332.
11 One area wrongly seen as ambiguous is choice of a national law where the

country has adopted the CISG. In such circumstances, they have agreed to that
law as it applies automatically unless excluded, although in some cases,
ambiguous wording might allow for an argument that exclusion was intended.
12 See section 13.8.
13 ILA Report 2008:7.
14 Some would argue instead that burden and standard are procedural issues or

should be determined from specialised rules as argued by Gary B. Born,

International Commercial Arbitration (The Hague, Kluwer Law International,
2009), 2148–2149.
15 For a discussion of this question in the context of French annulment decision

see Thomas Webster, ‘Terms of Reference and French Annulment Proceedings’,

Journal of International Arbitration 20, no. 6 (2003): 561.
16 For a discussion of the various conflicts rules see Julian D.M. Lew QC, Loukas

A. Mistelis & Stefan M. Kröll, Comparative International Commercial Arbitration

(The Hague, Kluwer Law International, 2003), 428.
17 Article 1496 French CCP; Arbitration Act 1996 (UK) s 46(1); Swiss PILA 1987

Art. 187(1); German ZPO 1998 Art. 1051(10). Art. 22.3 LCIA Rules; Art. 28.1
ICDR Rules; §23(1) DIS Rules; Art. 33(1) Swiss IAR.
18 Institute of International Law Resolution on Arbitration Between States, State

Enterprises or State Entities, And Foreign Enterprises 1989, 12 September 1989,

XVI YBCA 236 (1991) 237. Party autonomy is also supported by Art. 42 of the
Washington Convention.
19 80/934/EEC: Convention on the law applicable to contractual obligations

opened for signature in Rome on 19 June 1980; Regulation (EC) No. 593/2008 of
the European Parliament and of the Council of 17 June 2008 on the law
applicable to contractual obligations (Rome I).
20 For example s. 46(1) of the English Arbitration Act 1996. It goes on to provide

as an alternative that the case is to be decided in accordance with ‘such other

considerations’ which the parties agree upon. The latter has been interpreted to
allow the application of lex mercatoria where so agreed.
21 Gary B. Born, International Commercial Arbitration (The Hague, Kluwer Law

International, 2009), 2156.

22 Ibid., 2158.
23 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn
(Oxford: Oxford University Press, 2009), 197.
24 Gary B. Born, International Commercial Arbitration (The Hague, Kluwer Law

International, 2009), 2206 citing ICC Case No. 9009 where the tribunal ignored
the applicable law denying interest on the basis that this was contrary to most
countries’ practices and the reasonable expectations of the parties.
25 Emmanuel Gaillard & John Savage (ed.), Fouchard, Gaillard, Goldman on
International Commercial Arbitration (The Hague: Kluwer Law International,
1999), 797.
26 Gaillard cites the example of ICC Case No. 10625, Award of 2001, unpublished,

Emmanuel Gaillard, ‘The Role of the Arbitrator in Determining the Applicable

Law’, in The Leading Arbitrators' Guide to International Arbitration, 2nd edn, ed.
L. Newman & R. Hill (Huntington: Juris Publishing, 2008), 174–175.
27 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 578.

28 Locus Regit Actum, or ‘the place governs the act’, a conflict of laws rule

providing that certain sets of facts should be ‘governed by the laws of the place
where the facts occurred’ (per R.E.B. Totterman, ‘Functional Bases of the Rule
Locus Regit Actum in English Conflict Rules’ The International and Comparative
Law Quarterly 2 no. 1 (January 1953):27. In the case of the contract, this choice of
law rule would provide that the contract is governed by the law of the place in
which it was signed (per W Anderson, Ballentine’s Law Dictionary, 3rd edn
(Rochester: LCP, 1969), p752). See also Emmanuel Gaillard & John Savage (ed.),
Fouchard, Gaillard, Goldman on International Commercial Arbitration (The Hague:
Kluwer Law International, 1999), page 367 and footnote 566, citing ICC Award
No. 5832 (Zurich, 1988), 115 J.D.I. 1198(1988).
29 Pierre Mayer, ‘The Limits of Severability of the Arbitration Clause’, in
Improving The Efficiency of Arbitration Agreements and Awards: 40 Years of
Application of the New York Convention 261, ICCA Congress Series No. 9, ed.
Albert Jan van den Berg (Paris:ICC Publishing, 1999), 239.
30 Julian Lew, Loukas Mistelis & Stefan Kröll, Comparative International
Commercial Arbitration (The Hague: Kluwer Law International, 2003), 416 citing
ICC Case No. 1455 (1967) III YBCA 212 (1978); and partial award in ICC Case No.
5073 (1986) US Exporter v. Argentine Distributor, XIII YBCA 53 (1988).
31 See, e.g., Art. 28(1) UNCITRAL Model Law.
32 Gary B. Born, International Commercial Arbitration (The Hague, Kluwer Law

International, 2009), 2217.

33 This is known as depeçage.
34 See, e.g., Channel Tunnel Group Ltd v. Balfour Beatty Construction Ltd [1993]

AC 334 (House of Lords).

35 See, e.g., the Aminoil Case, Award of 24 May 1982, Kuwait v. The American

Independent Oil Company (AMINOIL), YBCA 1984 71. In the Aminoil case the
choice of law was left to the tribunal, stipulating that the tribunal should
consider ‘the quality of the parties, the transnational character of their relations
and the principles of law and practice prevailing in the modern world’.
36 Channel Tunnel Group Ltd v. Balfour Beatty Construction Ltd [1992] 1 QB 656

(CA); Channel Tunnel Group Ltd v. Balfour Beatty Construction Ltd [1992] AC 334
(HL); see also Reymond, ‘The Channel Tunnel Case and the Law of International
Arbitration’ (1993) 109 LQR 337.
37 Julian Lew, Loukas Mistelis & Stefan Kröll, Comparative International
Commercial Arbitration (The Hague: Kluwer Law International, 2003) at 445,
give the example of a vague reference to British law without indicating whether
English, Scottish or Northern Irish law is truly intended to apply.
38 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 231.

39 Article 3(1).
40 [1980] Official Journal of the European Union No III 282, 17.
41 Julian Lew, Loukas Mistelis & Stefan Kröll, Comparative International
Commercial Arbitration (The Hague: Kluwer Law International, 2003), 416 citing
ICC Case No. 1434, Multinational Group O v. State B 103 Clunet 978 (1976).
42 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 577. Such an argument
was rejected on the facts in ICC Case No. 5835 (1996). For an ICC case presuming
such a negative determination see ICC Case No. 4650 XII YCA (1987), 111.
43 Lew, Mistelis and Kröll note 82.1% in 1998, 82% in 1999, 77% in 2000 and

78% in 2001. One question is why there is a slight decrease over time. One would
normally expect that there would be increased selection as parties begin to
understand the value of such an approach. One reason could be that many
transactions will have a known law applicable even absent express party
agreement. For example, arguably the take-up of the CISG means that the parties
know that the same law applies without any express reference. Julian Lew,
Loukas Mistelis & Stefan Kröll, Comparative International Commercial Arbitration
(The Hague: Kluwer Law International, 2003), note 2 at 411.
44 See, e.g., Art. 28 UNCITRAL Model Law.
45 For example, Indonesian Arbitration Law Art. 56(2).
46 For example, s. 24(3) Sri Lankan Arbitration Act.
47 Simon Greenberg, Christopher Kee & J. Romesh Weeramantry, International

Commercial Arbitration: An Asia-Pacific Perspective (New York:Cambridge

University Press, 2011), 104.
48 See, e.g., Art. 36(2) Japanese Arbitration Law; Art. 29(2) South Korean
Arbitration Act and see also Art. 1051(2) German Arbitration Law.
49 An example is Art. 18 of the Law of the People's Republic of China on the Laws

Applicable to Foreign-Related Civil Relations; in contrast, s. 30(4) of the

Malaysian Arbitration Act requires the application of ‘the conflict of laws rules’,
but does not specify which particular conflict of laws rules.
50 Washington Convention Art. 42(1).
51 A closest connection test is directed by DIS Rules Arts. 23(2) and 33.1 Swiss

Rules 2012.
52 Gary B. Born, International Commercial Arbitration (The Hague, Kluwer Law

International, 2009), 2119.

53 Pierre Lalive, ‘Le droit applicable au fond par rarbitre international’, in Droit

international et droit communautaire: actes du colloque, ed. De Lourdes Belchior,

Paris, 5 et 6 avril 1990, Paris, Fondation Calouste Gulbenkian Centre culturel
portugais, 1991, at 44–45 and cited in Jean-François Poudret & Sébastien Besson,
Comparative Law of International Arbitration, 2nd edn (London: Sweet &
Maxwell, 2007), 585, fn 77.
54 This is the case with the ACICA Rules which referred to ‘the’ law designated by

the parties and the ‘rules’ of law in respect of the tribunal's discretion. See Art.
34(1) ACICA Rules.
55 See, e.g., ICC Rules 2012 Art. 21.
56 See, e.g., Halpern v. Halpern [2007] EWCA Civ 291 in relation to the English
Arbitration Act 1996 and the application of the reference to ‘such other
considerations’ to allow a religious law to apply.
57 H.M. Holtzmann & J. E. Neuhaus, A Guide to the UNCITRAL Model Law on

International Commercial Arbitration: Legislative History and Commentary (The

Hague: Kluwer Law and Taxation Publishers, 1989), 767–768.
58 Tibor Varady, ‘Can Proceeding “Not in Accordance with the Agreement of the

Parties” Be Condoned? Remarks on Article V(1)(b) of the New York Convention’,

in International Arbitration and International Commercial Law: Synergy,
Convergence and Evolution: Liber Amicorum Eric Bergsten, ed. Kröll Mistelis et al.
(The Hague: Kluwer Law International, 2011), at 467 et seq.
59 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 585.

60 For direct choice see Art. 1496 French NCCP and Art. 1054(2) 1986
Netherlands Arbitration Statute; Art. 22(1) Stockholm Arbitration Rules; AAA
Rules Art. 28(1); ICC Rules 2012 Art. 21(1); UNCITRAL Rules 2010 Art. 35.1, the
latter as changed from the earlier version of the Rules.
61 Emmanuel Gaillard & John Savage (ed.), Fouchard, Gaillard, Goldman on
International Commercial Arbitration (The Hague: Kluwer Law International,
1999), 865.
62 Klaus Peter Berger, ‘Evidentiary Privileges: Best Practice Standards
versus/and Arbitral Discretion’, Arbitration International 22, no. 4 (2006): 501–
502; Jean-François Poudret & Sébastien Besson, Comparative Law of
International Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 587.
63 Veijo Heiskanen, ‘And/or: the problem of qualification in international
arbitration’, Arbitration International 26, no. 4 (2010): 451.
64 For the United States interest-based approach see Symeon C Symeonides, ‘The

American Revolution and the European Evolution in Choice of Laws: Reciprocal

Lessons’, Tulane Law Rev 80, no. 2(2008): 1.
65 Veijo Heiskanen, ‘And/or: the problem of qualification in international
arbitration’, Arbitration International 26, no. 4 (2010): 451.
66 Some still opt for this approach. See Gary B. Born, International Commercial

Arbitration (The Hague, Kluwer Law International, 2009), 2208.

67 Regulation (EC) No. 593/2008 of the European Parliament and of the Council

of 17 June 2008 on the law applicable to contractual obligations (Rome I) Art. 4;

see also 80/934/EEC: Convention on the law applicable to contractual
obligations opened for signature in Rome on 19 June 1980 (The Rome
Convention) Art. 4(2).
68 ICC Case No. 7071, Interim Award of 2 March 1994, unpublished, cited in

Emmanuel Gaillard, ‘The Role of the Arbitrator in Determining the Applicable

Law’, in The Leading Arbitrators' Guide to International Arbitration, 2nd edn, ed.
L. Newman & R. Hill (Huntington: Juris Publishing, 2008), 174.
69 Compare the postal rule under common law and the requirement for receipt

by offeror under civilian systems.

70 It is slightly challenging doctrinally as the parties have not expressly selected

laws that apply automatically, although one might assert that a failure to exclude
is an implied choice, hence there would be no residual discretion for the tribunal.
71 See, e.g., Franco Ferrari, ‘Homeward Trend and Lex Forism Despite Uniform
Sales Law’, Vindobona journal of International Commercial Law and Arbitration
13, no. 1 (2009):15.
72 Peter Binder, International Commercial Arbitration and Conciliation in
UNCITRAL Model Law Jurisdictions, 3rd edn (London Sweet & Maxwell,
2010):336; see also Marc Blessing, ‘Choice of Substantive Law in International
Arbitration’, Journal of International Arbitration 14 no.2 (1997):39, 49.
73 Horacio Grigera Naón, Choice-of-Law Problems in International Commercial

Arbitration, 289 Collected Courses, Hague Academy of Intl. L. 9 (2001).

74 Julian Lew, Loukas Mistelis & Stefan Kröll, Comparative International
Commercial Arbitration (The Hague: Kluwer Law International, 2003), 424.
75 Ibid., 426.
76 Ibid., 440.
77 Rubino-Sammartano, ‘Le Tronc Commun des lois nationales en présence

(réflexion sur le droit applicable par l'arbitre international), 1987 Clunet 133.
Cases where tribunals have applied the Tronc Commun method include ICC
Cases No. 2272 and 2886, Collection of ICC Arbitral Awards 1974–1985, 11 and
332 and ICC Case No. 5103, Collection of ICC Arbitral Awards 1986–1990, 361.
78 Civ 1 ère 22 October 1991, Valenciana, Rev Arb 1992, 457. Cases where choice

of law discretions were used to make the contracts valid include ICC Case No.
4996, Collection of ICC Arbitral Awards 1986–1990, 293 and ICC Case No. 4145,
Collection of ICC Arbitral Awards 1986–1990, 53.
79 Other cases include protection of the claims rather than the contract in terms

of statutes of limitation. ICC Case No. 7154, Collection of ICC Arbitral Awards
1991–1995, 555. Such an approach is consistent with Rome Convention norms,
including Articles 8 and 9.
80 Andreas F Lowenfeld, ‘The Two-Way Mirror: International Arbitration as a

Comparative Procedure’, Michigan Yearbook of International Legal Studies 7

(1985): 163, 179.
81 Simon Greenberg, Christopher Kee & J. Romesh Weeramantry, International

Commercial Arbitration: An Asia-Pacific Perspective (New York:Cambridge

University Press, 2011), 111.
82 ICC Case No. 8450, unpublished, cited in Emmanuel Gaillard, ‘The Role of the

Arbitrator in Determining the Applicable Law’, in The Leading Arbitrators' Guide

to International Arbitration, 2nd edn, ed. L. Newman & R. Hill (Huntington: Juris
Publishing, 2008), 175.
83 This is discussed in section 13.5.1.
84 Berthold Goldman (Les Conflits de lois dans l'Arbitrage International de droit

priv?, 109 Recueil des Cours 347, 542 (Hague Academy of International Law,
1963) was one of the first key exponents of lex mercatoria. Subsequently
significant work has been done by Joachim Bonell (see Joachim Bonell, The
UNIDROIT Principles in Practice: Caselaw and bibliography on the UNIDROIT
principles of international commercial contracts (Ardsley, NY: Transnational
Publishers, 2006) and Klaus Peter Berger (K.P. Berger (ed.), The Practice of
Transnational Law, Wolters Kluwer Law & Business, 2001). See also Clive M.
Schmitthoff, ‘International Business Law: A New Law Merchant’, in Clive M.
Schmitthoff s Select Essays on International Trade Law, ed. Chia-Jui Cheng
(Dordrecht:Martinus Nijhoff Publishers, 1998), 3; Thomas E. Carbonneau, Lex
Mercatoria and Arbitration: A Discussion of the New Law Merchant (The Hague,
Kluwer Law International, 1998). For a criticism of lex mercatoria see FA. Mann,
‘The Proper Law and the Conflict of Laws’, The International and Comparative
Law Quarterly 36, no. 3 (1987): 437. Another more circumspect view is that of
L.J. Mustill, ‘The New Lex Mercatoria: The First Twenty Five Years’, Arbitration
International 4, no. 2 (1988): 86.
85 Joachim Bonell, The UNIDROIT Principles in Practice: Caselaw and bibliography

on the UNIDROIT principles of international commercial contracts (Ardsley, NY:

Transnational Publishers, 2006; Pierre Mayer, ‘The Role of the UNIDROIT
Principles in ICC Arbitration Practice’, in ICC Bulletin –Special Supplement:
UNIDROIT Principles of International Commercial Contracts –Reflections on their
Use in International Arbitration (2002) at 108. See generally Andreas F
Lowenfeld, Lex Mercatoria: An Arbitrator's View in Thomas E. Carbonneau, Lex
Mercatoria and Arbitration a Discussion of the New Law Merchant (The Hague,
Kluwer Law International, 1998) 37; Ole Lando, ‘The Lex Mercatoria in
International Commercial Arbitration’ ICLQ 34, no. 4 (1985) 747.
86 See, e.g., the Central project of Professor Klaus-Peter Berger. See Klaus Peter

Berger, ‘The TransLex Principles: An Online Research Tool for the Vis Moot and
International Arbitration’, in International Arbitration and International
Commercial Law: Synergy, Convergence and Evolution, ed. S. Kröll et al. (The
Hague: Kluwer Law International, 2011).
87 Emmanuel Gaillard & John Savage (ed.), Fouchard, Gaillard, Goldman on
International Commercial Arbitration (The Hague: Kluwer Law International,
1999), 805.
88 Simon Greenberg, Christopher Kee & J. Romesh Weeramantry, International

Commercial Arbitration: An Asia-Pacific Perspective (New York:Cambridge

University Press, 2011), 133.
89 Charles Molineaux, ‘Applicable Law in Arbitration – The Coming Convergence

of Civil and Anglo-Saxon Law via UNIDROIT and Lex Mercatoria’, Journal of World
Investment 1, no. 1 (2000): 130.
90 These issues were discussed in sections 3.2.2 to 3.2.4.
91 H.M. Holtzmann & J.E. Neuhaus, A Guide to the UNCITRAL Model Law on

International Commercial Arbitration: Legislative History and Commentary (The

Hague:Kluwer Law and Taxation Publishers, 1989), 767–768.
92 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 601–602.

93 Emmanuel Gaillard & John Savage (ed.), Fouchard, Gaillard, Goldman on
International Commercial Arbitration (The Hague: Kluwer Law International,
1999), 802–803.
94 UNCITRAL Rules 2010 Art. 35.
95 David W. Rivkin, ‘Enforceability of Arbitral Awards based on Lex Mercatoria’

Arbitration International 9, no. 1 (1993): 67. See also Filip de Ly, International
Business Law and Lex Mercatoria (Amsterdam:Emerald Group Publishing ltd,
1992); Yves Derains, ‘Transnational Law in ICC Arbitration’, in Klaus Peter
Berger (ed.), The Practice of Transnational Law (The Hague: Kluwer Law
International, 2001) at 43.
96 International Law Association Committee on International Commercial
Arbitration, ‘The Applicability of Transnational Rules in International
Commercial Arbitration’, ILA Report of the 64th Conference, Queensland
Australia (1990).
97 Hilmar Raeschke-Kessler ‘The UNIDROIT Principles in Contemporary Contract
Practice’, in ICC Bulletin – Special Supplement, UNIDROIT Principles of
International Commercial Contracts – Reflections on Their Use in International
Arbitration 2002, 111.
98 Norsolor SA v. Pabalk Ticaret Sarketi, Award Rendered 26 October 1979,

Award No. 26, Addendum 1 to Appendix V: Court of Appeal of Vienna, 29 June

1982: Supreme Court of Austria, 18 November 1982, 1983 Recht der
Internationalen Wirtschaft 29, 868, Klunet 1983, 645, Rev Arb 1983, 516, 519, IX
Yearbook Commercial Arbitration 159. See also David W Rivkin, ‘Enforceability
of Arbitral Awards based on Lex Mercatoria’ Arbitration International 9, no. 1
(1993): 67.
99 Cases applying transnational law include ICC Case No. 3540 (Award of 3

October 1980) Year Book of Commercial Arbitration Volume VII (The Hague:
Kluwer Law International (1982), 124; ICC Case No. 3131 (The ‘Norsolor’ Case)
Year Book of Commercial Arbitration Volume IX (The Hague: Kluwer Law
International (1984), 109; ICC Case No. 5953, Partial Award of 1 September
1988, 1990 Review Arb 701; ICC Case No. 9246, Award of 8 March 1996, Year
Book of Commercial Arbitration Volume XXII (The Hague: Kluwer Law
International, 1997), 28; ICC Case No. 8486, Year Book of Commercial
Arbitration Volume XXIVa (The Hague: Kluwer Law International, (1999), 162. In
Andersen Consulting v. Arthur Andersen (ICC Case 9797 (2001) 12:2 ICC ICArb.
Bull. 88) a tribunal applied the UNIDROIT Principles, absent a choice of law by
the parties.
100 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 606.

101 Julian Lew, Loukas Mistelis & Stefan Kröll, Comparative International
Commercial Arbitration (The Hague: Kluwer Law International, 2003), 426.
102 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 606–607.

103 Keith Highet, ‘The Engima of the Lex Mercatoria’, in Lex Mercatoria and

Arbitration, Tulane Law Review 63, no. 3 (1989): 613, ed. Thomas E. Carbonneau
(The Hague:Kluwer Law International, 1998), 107–108. For the view that most
elements are already captured by developed national legal systems see also Gary
B. Born, International Commercial Arbitration (The Hague, Kluwer Law
International, 2009), 2235.
104 A Rogers, ‘Contemporary Problems in International Commercial Arbitration’,

International Business Lawyer 17, no. 4 (1989): 154, 158; M Sornarajah, ‘The
Uncitral Model Law: A Third World Viewpoint’, Journal of International
Arbitration 6 no. 4 (1989): 7, 16.
105 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 596.

106 The Norsolor Case came to this conclusion but seemed to reverse the
standard of proof in finding that there was not a sufficiently clear intent to
localise applicable law. ICC Case No. 3131 (The ‘Norsolor’ Case) Year Book of
Commercial Arbitration Volume IX (The Hague: Kluwer Law International
(1984), 108, 109.
107 Cour de Cassation Civ lere 22 October 1991, Valenciana, Rev Arb 1992, 457.
108 First edition 1999. This provides as follows:

‘If the Contractor considers himself to be entitled to any extension of the Time
for Completion and/or any additional payment, under any Clause of these
Conditions or otherwise in connection with the Contract, the Contractor shall
give notice to the Engineer, describing the event or circumstance giving rise to
the claim. The notice shall be given as soon as practicable, and not later than 28
days after the Contractor became aware, or should have become aware, of the
event or circumstance.

If the Contractor fails to give notice of a claim within such period of 28 days, the
Time for Completion shall not be extended, the Contractor shall not be entitled to
additional payment, and the Employer shall be discharged from all liability in
connection with the claim. […]’.

109 Filip de Ly, International Business law and lex Mercatoria

(Amsterdam:Emerald Group Publishing Ltd, 1992).
110 N. Kornet, Contract Interpretation and Gap Filling: Comparative and
Theoretical Perspectives (Ius Commune Europaeum) (Belgium:Intersentia, 2006).
111 See, e.g., US Uniform Commercial Code s. 1-103.
112 An example of a tribunal issuing a procedural order as to applicable law in

the context of CISG and UNIDROIT Principles arose in ICC Case No. 8817 of 1997,
XXV ICCA Yearbook 355 (2000), 355-358.
113 P C Jessup, Transnational Law (New Haven: Yale University Press, 1956), 2.
114 David W. Rivkin, ‘Enforceability of Arbitral Awards based on Lex Mercatoria’

Arbitration International 9, no. 1 (1993): 67.

115 See, e.g., Art. 9(2) CISG.
116 L.J. Mustill, ‘The New Lex Mercatoria: The First Twenty Five Years’,
Arbitration International 4, no. 2 (1988): 86.
117 The Trans-Lex database can be accessed at <www.trans-lex.org/>.
118 Klaus Peter Berger, The Creeping Codification of the Lex Mercatoria (The

Hague: Kluwer Law International, 1999), 210-2111.

119 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 598–599.

120 The list is derived from Julian Lew, Loukas Mistelis & Stefan Kröll,
Comparative International Commercial Arbitration (The Hague: Kluwer Law
International, 2003), 458 et seq. and Klaus Peter Berger, The Creeping
Codification of the Lex Mercatoria (The Hague: Kluwer Law International, 1999),
210–211.
121 In ICC Case No. 8365 a tribunal listed eight principles of the lex mercatoria,

namely pacta sunt servanda, good faith, a duty to negotiate in good faith,
termination of contract in case of substantial breach, prohibiting of a party acting
to prevent its own obligation performance, venire contra factum proprium,
interpreting contracts under the principle of ut res magis valeat quampereat and
implied consent by conduct. ICC Award No. 8365, Clunet 124 (1997), 1078.
122 Article 1496 French CCP; Arbitration Act 1996 (UK) s 46(1); Swiss PILA 1987

Art. 187(1); German ZPO 1998 Art. 1051(10). Art. 22.3 LCIA Rules; Art. 28.1
ICDR Rules; § 23(1) DIS Rules; Art. 33.1 Swiss Rules 2012.
123 Gulf of Maine (Canada v. United States) (Judgment) [1984] ICJ Rep 246, 305.
124 Emmanuel Gaillard & John Savage (ed.), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 813.
125 Emmanuel Gaillard, ‘Transnational Law: A Legal System or a Method of

Decision-Making?’, in The Practice of Transnational law, ed. Klaus Peter Berger

(The Hague: Kluwer Law International, 2002), 62–63; Emmanuel Gaillard, ‘Thirty
Years of Lex Mercatoria: Towards the Selective Application of Transnational
Rules’, ICSID Review: Foreign Investment Law Journal 10 (1995): 208.
126 Allan Philip, ‘The Duties of an Arbitrator’, in The Leading Arbitrators' Guide to

International Arbitration, 2nd edn, ed. L. Newman & R. Hill (Huntington: Juris
Publishing, 2008), 80.
127 Article 28(4) UNCITRAL Model Law; Art. 21(2) ICC Rules 2012; Art. 28.2 AAA

Rules.
128 Laurence Craig, William W. Park & Jan Paulsson, Annotated Guide to the 1998

ICC Arbitration Rules with Commentary, 3rd edn (Oxford:Oxford University Press,
2000), 330.
129 UNCITRAL Rules 2010 Art. 35.3; Art. 21.2 ICC Rules 2012; Art. 1496 French

NCCP; German ZPO § 1051(4).

130 Gary B. Born, International Commercial Arbitration (The Hague, Kluwer Law

International, 2009), 2146. The view was taken that if that was not the case the
drafters would not want to express a provision that might make usages take
precedence over the law.
131 ICC Uniform Customs and Practice for Documentary Credits (UCP 600), ICC

Publication No. 600, 2006 Edition (Paris: ICC Publishing, 2006).

132 Incoterms 2010, ICC Publication No. 715, 2010 Edition (Paris: ICC Publishing,

2010).
133 Nigel Blackaby, Constantine Partasides, Alan Redfern & Martin Hunter,
Redfern and Hunter on International Arbitration, 5th edn (Oxford: Oxford
University Press, 2009), 224.
134 Similarly, they could be seen as rebus sic stantibus circumstances as
applicable in that industry.
135 The Principles Of European Contract Law 1998, Parts I and II, prepared by

the Commission on European Contract Law.

136 Uniform Customs and. Practice for Documentary Credits (UCP 400), 1983

revision, ICC Publication No. 400 (Paris: ICC Publishing, 1983).

137 Gary B. Born, International Commercial Arbitration (The Hague, Kluwer Law

International, 2009), 2147.

138 Yasuhei Taniguchi, ‘The Obligation to Mitigate Damages’, in Evaluation of

Damages in International Arbitration, Dossier of the ICC Institute of World

Business law, ed. Yves Derains & Richard H Kreindler (Paris: ICC Publishing,
2006), 668, 690.
139 George A. Bermann, ‘Introduction: Mandatory Rules of Law in International

Arbitration’, American Review of International Arbitration 18, no. 1 (2007): 1.

140 Pierre Mayer, ‘Mandatory Rules of Law in International Arbitration’,
Arbitration International 2, no. 4 (1986): 275. English writing sometimes speaks
of ‘international public policy’ instead of mandatory substantive laws. French
awards and scholarly writings speak of ‘lois de police’ or ‘lois d'application
immediate’.
141 Article 3(3) 80/934/EEC: Convention on the law applicable to contractual

obligations opened for signature in Rome on 19 June 1980 (The Rome

Convention), although the Convention does not apply to arbitration.
142 Gary B. Born, International Commercial Arbitration (The Hague, Kluwer Law
International, 2009), 82, fn 3.
143 Mitsubishi v. Soler 473 US 614 (1985).
144 Cour d'Appel de Paris, 19 May 1993 [1993] Rev Arb 645; decision of Cour de

Cassation, 5 January 1999.

145 M. Blessing, ‘Mandatory Rules of Law Versus Party Autonomy in International

Arbitration’, Journal of International Arbitration 14, no. 4 (1997): 23; M. Blessing,

‘Arbitrability of Intellectual Property Disputes’, Arbitration International 12, no.
2 (1996): 191; M. Blessing, ‘Impact of the Extraterritorial Application of
Mandatory Rules of Law on International Contracts’, in Swiss Commercial Law
Series, Vol 9, ed. Nedim Peter Vogt (Basel: Helbing & Lichtenhahn, 1999), 69;
Pierre Mayer, ‘Mandatory Rules of Law in International Arbitration’, Arbitration
International 2, no. 4 (1986): 274.; P. Mayer, ‘Reflections on the International
Arbitrator's Duty to Apply the Law - The 2000 Freshfields Lecture’, Arbitration
International 17, no. 3 (2001): 235; D. Hochstrasser, ‘Choice of Law and ‘Foreign’
Mandatory Rules in International Arbitration’, Journal of International
Arbitration 11, no. 1 (1994): 57; S. Lazar-eff, ‘Mandatory Extraterritorial
Application of National Law’, Arbitration International 11, no. 2 (1995): 137; N.
Voser, ‘Mandatory Rules of Law as a Limitation on the Law Applicable in
International Commercial Arbitration’, American Review of International
Arbitration 7, no. 3/4 (1996): 319; F-B. Weigand, ‘Evading EC Competition Law
by Resorting to Arbitration?’, Arbitration International 9, no. 3 (1993): 249; D.F.
Donovan & A.K.A. Greenawalt, ‘Mitsubishi After Twenty Years: Mandatory Rules
before Courts and International Arbitrators’, in Pervasive Problems in
International Arbitration, ed. L. A. Mistelis & J.D.M. Lew (The Hague: Kluwer Law
International, 2006), 11.
146 For example, Yves Derains, ‘Public Policy and the Law Applicable to the

Dispute in International Arbitration’, in Comparative Arbitration Practice and

Public Policy in Arbitration, ed. P. Sanders (The Hague: Kluwer Law and Taxation
Publishers, 1987), 228.
147 For example, Emmanuel Gaillard & John Savage (ed.), Fouchard, Gaillard,

Goldman on International Commercial Arbitration (The Hague: Kluwer Law

International, 1999), 852–854.
148 For example, Gary B. Born, International Commercial Arbitration (The Hague:

Kluwer Law International, 2009), 2133-2135.

149 M. Blessing, ‘Mandatory Rules of Law Versus Party Autonomy in International

Arbitration’, Journal of International Arbitration 14, no. 4 (1997): 23; Klaus Peter
Berger, ‘Acts of State and Arbitration: Exchange Control Regulations’, in Acts of
State and Arbitration, ed. K.H. Böckstiegel (Köln: C. Heymanns, 1997), 99 et seq.
150 In an earlier joint paper the present author attempted this approach as a first

round analysis of the issue. A. Barraclough & J. Waincymer, ‘Mandatory Rules of

Law in International Commercial Arbitration’, Melbourne Journal of International
Law 6, no. 2 (2005): 205. This section and its revised conclusions draw on and
extract from a later paper, J. Waincymer, ‘International commercial arbitration
and the application of mandatory rules of law’ Asian International Arbitration
Journal 5, no. 1 (2009): 1–45.
151 For example, Mitsubishi v. Soler 473 US 614 (1985), arbitration in Japan

under Swiss law with US Antitrust laws purported to apply; Soleimany v.

Soleimany [1998] 3 WLR 811, arbitration subject to Jewish law where Iranian
law was argued to invalidate the contract; Westacre Investment Inc v. Jujoimport
–SPDR Holding Co Ltd and Others [1998] 2 Lloyd's Rep 111 (QBD Comm Ct),
arbitration under Swiss law with Kuwaiti law purported to apply; and Omnium
de Traitement et de Calorisation SA v. Hilmarton Ltd [1999] 2 All ER (Comm) 146,
arbitration under Swiss law with Algerian law purporting to invalidate the
agreement.
152 D.F. Donovan & A.K.A. Greenawalt, ‘Mitsubishi After Twenty Years:
Mandatory Rules before Courts and International Arbitrators’, in Pervasive
Problems in International Arbitration, ed. L.A. Mistelis & J.D.M. Lew (The Hague:
Kluwer Law International, 2006), 58.
153 Pierre Mayer, ‘Mandatory Rules of Law in International Arbitration’,
Arbitration International 2, no. 4 (1986): 281, fn 14, refers to three cases of this
nature.
154 Grigera Naón suggests that a mandatory provision of the forum might also

limit a tribunal's power to apply foreign mandatory rules. H. Grigera Naón,

Choice-of-law Problems in International Commercial Arbitration 289 Collected
Courses, Hague Academy of Intl. L. 9 (2001).
155 The argument would be that the arbitrator would at the very least be acting

contrary to express or implied terms in his or her contract with the parties.
156 For example, 80/934/EEC: Convention on the law applicable to contractual

obligations opened for signature in Rome on 19 June 1980 (The Rome

Convention) Arts 3 and 7; The US Restatement (Second) Conflict of Laws § 187
(1971); Albert V. Dicey & John H.C. Morris, The Conflict of Laws, 12th edn
(London: Sweet & Maxwell, 1993), 1239 ff.
157 Gary B. Born, International Commercial Arbitration (The Hague, Kluwer Law

International, 2009), 88, fn 3.

158 Pierre Mayer, ‘Mandatory Rules of Law in International Arbitration’,
Arbitration International 2, no. 4 (1986): 282–283.
159 It could even be said to be part of implied party consent when the parties

choose to arbitrate in a way that makes the NYC relevant. Parties can even avoid
such questions simply by honouring the award in full, so their overriding
autonomy is non-contentious in that context.
160 Pierre Mayer, ‘Mandatory Rules of Law in International Arbitration’,
Arbitration International 2, no. 4 (1986): 285.
161 Ibid., 285.
162 A. Sheppard, ‘Interim ILA Report on Public Policy as a Bar to Enforcement of

International Arbitral Awards’, Arbitration International 19, no. 2 (2003): 217.

163 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 86-87, fn. 3.

164 H. Grigera Naón, Choice-of-law Problems in International Commercial
Arbitration (289 Collected Courses, Hague Academy of Intl. L. 9 (2001), 207.
165 Ibid., 208.
166 Piero Bernardini, ‘The Role of the International Arbitrator’, Arbitration
International 20, no. 2 (2004): 115.
167 Pierre Mayer, ‘Reflections on the International Arbitrator's Duty to Apply the

Law - The 2000 Freshfields Lecture’, 17 Arb Int 235 (2001) 246.
168 Pierre Mayer, ‘Mandatory Rules of Law in International Arbitration’,
Arbitration International 2, no. 4 (1986): 274, 285–286.
169 Bernard Hanotiau, ‘The Law Applicable to Arbitrability’, in Improving the
Efficiency of Arbitration and Awards: 40 Years of Application of the New York
Convention, ICCA Congress Series No. 9 (Paris 1998), ed. Albert Jan van den Berg
(Paris: ICC Publishing, 1999), 146, 158.
170 Emmanuel Gaillard, ‘The Role of the Arbitrator in Determining the Applicable

Law’, in The Leading Arbitrators' Guide to International Arbitration, 2nd edn, ed.
L. Newman & R. Hill (Huntington: Juris Publishing, 2008), 182.
171 Eco Swiss China Time Ltd v. Benetton International NV, 1 June 1999, ECJ, Case

C-126/97, [1999] ECR I-3055 ECJ.

172 Pierre Mayer, ‘Mandatory Rules of Law in International Arbitration’,
Arbitration International 2, no. 4 (1986): 280.
173 Yves Derains, ‘Public Policy and the Law Applicable to the Dispute in
International Arbitration’, in Comparative Arbitration Practice and Public Policy
in Arbitration, ed. P. Sanders (The Hague: Kluwer Law and Taxation Publishers,
1987), 251–252.
174 D. Hochstrasser, ‘Choice of Law and ‘Foreign’ Mandatory Rules in
International Arbitration’, Journal of International Arbitration 11, no. 1 (1994):
57.
175 Article v (1)(e).
176 See also Art. 23(1) of the UN Convention on the Carriage of Goods by Sea

1978 (Hamburg rules) which declares that provisions of agreements null and
void where those provisions contradict the mandatory rules of that Convention.
177 Klaus Peter Berger, ‘Acts of State and Arbitration: Exchange Control
Regulations’, in Acts of State and Arbitration, ed. K.H. Böckstiegel (Köln: C.
Heymanns, 1997), 99, 120.
178 Blessing refers to rules aimed solely at protecting State monetary interests,

e.g., monetary controls; policing rules; rules safeguarding a State's vital interests
and peoples welfare; market protection, e.g., competition policy. M. Blessing,
‘Mandatory Rules of Law versus Party Autonomy in International Arbitration’,
Journal of International Arbitration 14, no. 4 (1997): 23.
179 But note again the decision in Eco Swiss China Time Ltd v. Benetton
International NV, 1 June 1999, ECJ, Case C-126/97, [1999] ECR I-3055 ECJ.
180 H. Grigera Naón, Choice-of-law Problems in International Commercial
Arbitration 289 Collected Courses, Hague Academy of Intl. L. 9 (2001), 353.
181 N. Voser, ‘Mandatory Rules of Law as a Limitation on the Law Applicable in

International Commercial Arbitration’, American Review of International

Arbitration 7, no. 3/4 (1996): 349.
182 H. Grigera Naón, Choice-of-law Problems in International Commercial
Arbitration 289 Collected Courses, Hague Academy of Intl. L. 9 (2001), 338.
183 Emmanuel Gaillard & John Savage (ed.), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 857.
184 80/934/EEC: Convention on the law applicable to contractual obligations

opened for signature in Rome on 19 June 1980 (The Rome Convention). See also
Art. 9, 80/934/EEC: Convention on the law applicable to contractual obligations
opened for signature in Rome on 19 June 1980; Regulation (EC) No. 593/2008 of
the European Parliament and of the Council of 17 June 2008 on the law
applicable to contractual obligations (Rome I).
185 According to Blessing, for a mandatory rule to apply, there are six relevant

criteria:
1. the rule must be a norm of mandatory character;
2. the rule must impose itself regardless of the applicable law;
3. the rule's preconditions must be satisfied and should be construed narrowly
e.g. is it intended to apply to these facts;
4. a close connection between the subject matter of the dispute and the State
whose rule is sought to be applied;
5. the rule must appear to be ‘application worthy’ in the context of shared
values of an essential character, being based on transnational public policy
and recognised universally. Here the implications for all parties under
application would be considered;
6. the result must be seen as an ‘appropriate result’.

Blessing recognises that the last two provide for a type of rule of reason test (as
opposed to a per se rule). He suggests that practice shows that the test is not met
for mandatory rules only aiming to protect a State's fiscal or political interests.

186 Berger has put forward a similar set of criteria. He argues that mandatory
rules should be applied if the following five conditions are met:
1. the provision in question must be qualified as a mandatory norm;
2. the pre-conditions of this norm have to be met;
3. the provision wants to be applied irrespective of which law is applicable to
the contract;
4. there must be a ‘close connection’ between the contract and the jurisdiction
that has promulgated the provision; and
5. the contents of the provision, i.e. the socio-political values and policy
considerations inherent in it must be in accordance with the general opinion
of the international community of states (‘shared values’—requirement).

187 Emmanuel Gaillard & John Savage (ed.), Fouchard, Gaillard, Goldman on
International Commercial Arbitration (The Hague: Kluwer Law International,
1999), 853.
188 Ibid.
189 Ibid.
190 ‘International Law Association Report on Public Policy as a Bar to
Enforcement of International Arbitral Awards’ 70 bit'l L. Ass'n Rep. Conf. 352
2002, 364.
191 P. Sanders ‘Commentary’, in 60 years of ICC Arbitration – A Look at the Future,

ed. International Chamber of Commerce. Court of Arbitration (Paris: ICC

Publishing, 1984).
192 P. Lalive, ‘Transnational (or Truly International) Public Policy’, in
Comparative Arbitration Practice and Public Policy in Arbitration, ed. P. Sanders
(The Hague: Kluwer Law and Taxation Publishers, 1987), 257, 285.
193 P. Mayer & A. Sheppard, ‘Final ILA Report on Public Policy as a Bar to

Enforcement of International Arbitral Awards”, Arbitration International 19, no.

2 (2003): 249.
194 Republique de Cote d'lvoire v. Norbert Beyrard, 12 January 1993, 1994 Rev Arb
4, 685.
195 H. Grigera Naón, ‘Choice-of-law Problems in International Commercial
Arbitration’ 289 Collected Courses, Hague Academy of Intl. L. 9 (2001), 323.
196 Ibid., 320.
197 P. Mayer & A. Sheppard, ‘Final ILA Report on Public Policy as a Bar to

Enforcement of International Arbitral Awards”, Arbitration International 19, no.

2 (2003): 249.
198 Ibid., 261.
199 Yves Derains, ‘Public Policy and the Law Applicable to the Dispute in
International Arbitration’, in Comparative Arbitration Practice and Public Policy
in Arbitration, ed. P. Sanders (The Hague: Kluwer Law and Taxation Publishers,
1987), 228.
200 Ibid., at 233. The distinction may not always be made by tribunals. In ICC

Case No. 8385, Collection of ICC Arbitral Awards 1996–2000, 474 a tribunal
chose not to apply the Racketeer Influenced and Corrupt Organisation Act to a
contract where the parties had selected New York law, in part, on the basis that
this conformed with parties' expectations.
201 Yves Derains, ‘Public Policy and the Law Applicable to the Dispute in
International Arbitration’, in Comparative Arbitration Practice and Public Policy
in Arbitration, ed. P. Sanders (The Hague: Kluwer Law and Taxation Publishers,
1987), 228.
202 D.F. Donovan & A.K.A. Greenawalt, ‘Mitsubishi After Twenty Years:
Mandatory Rules before Courts and International Arbitrators', in Pervasive
Problems in International Arbitration, ed. L.A. Mistelis & J.D.M. Lew (The Hague:
Kluwer Law International, 2006), 11.
203 World Duty Free Company Ltd v. The Republic of Kenya, Award, ICSID Case No

Arb/00/7 IIC 277 (2006).

204 Racine, L’Arbitrage Commercial International et L'Ordre Public (1969) 398–

399 cited in Piero Bernardini, ‘The Role of the International Arbitrator’,

Arbitration International 20, no. 2 (2004): 122, fn 21. A similar approach was
taken in relation to public policy rules by the Paris Court of Appeal in Sari Enodis
v. SNC Prodim (16 March 1995, 1996 RevArb) 1, 146.
205 The Lagergren Award of 1963 in ICC Case 1110 (1994) 10 Arb International

3, 282 et seq.
206 Ibid., 293.
207 World Duty Free Company Limited v. The Republic of Kenya, ICSID Case No

ARB/00/7, Award of 4 October 2006, 48.

208 Award in ICC Case No. 5622 (1988) YCA 1994, at 105 et seq.
209 Court of Justice Decision Hilmarton v. OTV, 17 November 1989, XIX YBCA 214

(1994); Swiss Federal Tribunal, 17 April 1990, OTV v. Hilmarton, 1993 Rev Arb
315.
210 Eco Swiss China Time Ltd v. Benetton International NV, 1 June 1999, ECJ, Case

C-126/97, [1999] ECR I-3055 ECJ.

211 Emmanuel Gaillard & John Savage (eds.), Fouchard Gaillard Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 859.
212 Simon Greenberg, Christopher Kee & J. Romesh Weeramantry, International

Commercial Arbitration: An Asia-Pacific Perspective (New York:Cambridge

University Press, 2011), 117.
213 For example, see Babcock v.Jackson 12 NY2d 473 (1963).
214 Article 47.1 CIETAC Rules 2012.
215 Article 1496 French CCP; Arbitration Act 1996 (UK) s 46(1); Swiss PILA 1987

Art. 187(1); German ZPO 1998 Art. 1051(10). Art. 22.3 LCIA Rules; Art. 28.1
ICDR Rules; § 23(1) DIS Rules; Art. 33.1 Swiss Rules 2012.
216 Gulf of Maine (Canada v.United States) (Judgment) [1984] ICJ Rep 246, 305.
217 For example, the French Theorie de Imprevision. For an overview of this

doctrine, see Sidonia Culda ‘The Theory of Imprevision’, Fiat Iusticia 2 (2010): 40
et seq.
218 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 576–577.

219 Ibid., 576 citing ICC Awards 3099 and 3100, VII YCA (1982), 87.
220 Klockner v. Cameroon ARB/81/2, Decision on Annulment, 3 May 1985, 2ICSID

Reports 90 at para. 91.

221 Statute of the International Court of Justice, 26 June 1945, 59 Stat 1031, 1060.
222 Bin Cheng, General Principles of Law as Applied by International Courts and

Tribunals (New York: Cambridge University Press, 2006 reprint), 23.

223 Gulf of Maine (Canada v. United States) (Judgment) [1984] ICJ Rep 246, 305.
224 As to amicus curiae see section 7.16.
225 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 209.

226 Treaty of Lisbon amending the Treaty on European Union and the Treaty

establishing the European Community, signed at Lisbon, 13 December 2007. For

some background on the possible effects of the Lisbon Treaty on EU Member
State BITs, see Stephen Woolcock, ‘The Potential Impact of the Lisbon Treaty on
European Union External Trade Policy’, European Policy Analysis 8 (2008): at 4.
227 (1981) VI YBk Comm Arbitration 89.
228 American Independent Oil Company Inc (Aminoil) v. Government of the State of

Kuwait [1982] 21 ILM 976.

229 Aminoil 21 ILM 976 (1982) 1065, para. 26.
230 Salini Costruttori S.p.A. and Italstrade S.p.A. v. Hashemite Kingdom of Jordan,

ICSID Case ARB/02/13, Award 31 January 2006.

231 Joy Mining Machinery Ltd v. Arab Republic of Egypt, ICSID Case ARB/03/11

Award on Jurisdiction, 6 August 2004.

232 El Paso Energy International Co v. Argentine Republic, ICSID Case ARB/03/15,

Decision on Jurisdiction, 27 April 2006.

233 Noble Ventures, Inc v. Romania, ICSID Case ARB/01/11, Award 12 October

2005.
234 Siemens A.G. v. Argentine Republic, ICSID Case ARB/02/08, Award 6 February

2007.
235 Article 35(2) UNCITRAL Rules 2010; DIS Rules 1998 Art. 23(2); HKIAC Rules

2008 Art. 31.2; Art. 28(3) ICDR Rules 2009; Art. 30(1) JAMS Rules 2005; Art.
22(4) LCIA Rules 1998; Art. 22(3) SCC Rules 2010; Art. 27.2 SIAC Rules 2010;
Art. 33.2 Swiss Rules 2012; Art. 24(3) VIAC Rules 2006; Art. 34(2) ACICA Rules
2005. Art. 38 of the Statute of the International Court of Justice also allows the
Court to be empowered to decide ex aequo et bono. It does not appear that the
court has to date decided on this basis. Leon Trakman, ‘Ex Aequo et Bono:
Demystifying an Ancient Concept’, Chicago Journal of International Law 8, no. 2
(2008): 625. The German translation of amiable compositeurs is nach billigem
Ermessen.
236 Article 42(3).
237 Redfern and Hunter, referring to Ecuador’s law of arbitration. Nigel Blackaby

et al., Redfern and Hunter on International Arbitration, 5th edn (Oxford: Oxford
University Press, 2009), 228.
238 It is suggested that the parties cannot bind the tribunal in this way in China.

Jingzhou Tao, Arbitration Law and Practice in China (The Hague: Kluwer Law
International 2008): 105–106.
239 See, e.g., Malaysia.
240 See ICC Working Group ‘Amiable Composition: Report of the ICC France

Working Group’, Revue de Droit des Affaires Internationales 6 (2005): 753 et seq.
241 Simon Greenberg, Christopher Kee & J. Romesh Weeramantry, International

Commercial Arbitration: An Asia-Pacific Perspective (New York: Cambridge

University Press, 2011), 143.
242 Sigvard Jarvin, ‘The Sources and Limits of the Arbitrator's Powers’,
Arbitration International 2, no. 2 (1986): 158.
243 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 625.

244 For example, in ICC Case No. 7986 the tribunal was to act as amiable

compositeur ‘without being bound to comply with the deadlines and ordinary
rules of procedure’. Final Award Case 7986, 1 ICC International Court of
Arbitration Bulletin 18, 72.
245 For example, in ICC Case No. 7913 the tribunal was ‘…not be required to

apply the provisions of any law whatsoever, whether relating to procedure or to

the merits of the case…’ Final Award Case 7913, 1 ICC International Court of
Arbitration Bulletin 18, 67.
246 See, e.g., Hewitt v. McKensey [2003] NSWSC 1186 (16 December 2003).
247 In some cases, the parties will empower an arbitrator to decide ‘according to

the principles of equity'. For example, see ICC Case 9704, 1 ICC International
Court of Arbitration Bulletin 18, 87.
248 See, e.g., the Statute of the ICJ. For a discussion of equity in international law

see Thomas Franck, Fairness in International Law (Oxford: Clarendon Press,

1998), 47 et seq.
249 ICC Case No. 4467 S. Jarvin & Y. Derains, Collection of ICC Arbitral Awards

1974–1985 (Paris: ICC Publishing, 1990), 501.

250 ICC Case No. 2216 S. Jarvin & Y. Derains, Collection of ICC Arbitral Awards

1974–1985 (Paris: ICC Publishing, 1990), 224.

251 Sigvard Jarvin, ‘The Sources and Limits of the Arbitrator's Powers’, 2 (1986)

Arbitration International 2, pl57.

252 French NCCP Art. 1482.
253 See, e.g., ICC Case No. 12099 cited in Laurence Kiffer, ‘Amiable Composition in

ICC Arbitration’, ICC International Court of Arbitration Bulletin 18, no. 1 (2007)
62.
254 ICC ICArb. Bull. 18:1 (2007), 98.
255 ICC ICArb. Bull. 18.1 (2007), 72.
256 Nigel Blackaby, Constantine Partasides, Alan Redfern & Martin Hunter,
Redfern and Hunter on International Arbitration, 5th edn (Oxford: Oxford
University Press, 2009), 227–228.
257 Jean-François Poudret & Sébastien Besson, Comparative Law of International
Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 619.
258 Ibid., 623.
259 Ibid., 621.
260 Ibid., 472.
261 See, e.g., ICC Case No. 9669 1 ICC International Court of Arbitration Bulletin

18, 54 and ICC Case No. 12070 1 ICC International Court of Arbitration Bulletin
18, 111.
262 See, e.g., ICC Case No. 10049 1 ICC International Court of Arbitration Bulletin

18, 89.
263 E Bertrand, ‘Amiable Composition: Report of the ICC France Working Group’,

Revue de Droit des Affaires Internationales 6 (2005): 753 et seq. at 756.

264 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 623.

265 Gary B. Born, International Commercial Arbitration (The Hague, Kluwer Law

International, 2009), 2241.

266 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 626, and see the
decisions cited in footnote 275.
267 Ibid., 477.
268 Yves Derains & Eric A. Schwartz, A Guide to the ICC Rules of Arbitration, 2nd

edn (The Hague: Kluwer Law International, 2005), 245.

269 L.J. Mustill, ‘The New Lex Mercatoria: The First Twenty Five Years’,
Arbitration International A, no. 2 (1988): 103.
270 Norbert Horn & Stefan Kröll, Arbitrating Foreign Investment Disputes (The

Hague: Kluwer Law International, 2004), 436–437.

271 Klaus Peter Berger, ‘Integration of Mediation Elements into Arbitration:

‘Hybrid’ Procedures and ‘Intuitive’ Mediation by International Arbitrators',

Arbitration International 19, no. 3 (2003): 397.
272 ICC Decision No. 3742 (1984) Clunet 910.
273 Bernard G. Poznanski, ‘The Nature and Extent of an Arbitrator's Powers in

International Commercial Arbitration’, Journal of International Arbitration 4, no.

3 (1987): 79–80.
274 Simon Greenberg, Christopher Kee & J. Romesh Weeramantry, International

Commercial Arbitration: An Asia-Pacific Perspective (New York: Cambridge

University Press, 2011), 142 and citing Coderre v. Coderre (2008 QCCA 888),
Court of Appeal of Quebec, 13 May 2008.
275 Gary B. Born, International Commercial Arbitration (The Hague, Kluwer Law

International, 2009), 2241.

276 ICC Case No. 3938, 111 Clunet 926 (1984).
277 ICC Case No. 4206 published in Assarsratt, No. 3, 1984.
278 6 May 1988, Rev Arb 1989.63; 19 April 1991, Rev Arb 1991. p673 (annot. E

Loquin) cited in Laurence Kiffer, ‘Amiable Composition in ICC Arbitration’, ICC

International Court of Arbitration Bulletin 18, no. 1 (2007) footnote 29.
279 ICC Case No. 3267 Year Book of Commercial Arbitration Volume VII (1982)

96.
280 ICC Case No. 7913 cited in Laurence Kiffer, ‘Amiable Composition in ICC

Arbitration’, ICC International Court of Arbitration Bulletin 18, no. 1 (2007) 58.
281 Paris 6 March 1988 Rev Arb 1989. p83 (annot. E Loquin) cited in Laurence
Kiffer, ‘Amiable Composition in ICC Arbitration’, ICC International Court of
Arbitration Bulletin 18, no. 1 (2007): footnote 33.
282 Hersch Lauterpacht, The Development of International Law by the
International Court (London: Stevens and Sons, 1958), 215.
283 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 624.

284 Emmanuel Gaillard & John Savage (ed.), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), p839, para. 1507.
285 Laurence Craig, William W. Park & Jan Paulsson, Annotated Guide to the 1998

ICC Arbitration Rules with commentary, 3rd edn (Oxford: Oxford University Press,
2000), 113.
286 Leon Trakman, ‘Ex Aequo et Bono: Demystifying an Ancient Concept’, Chicago

Journal of International Law 8, no. 2 (2008): 625.

287 See, e.g., Award of 3 November 1977, 1980 Rev Arb 560.
288 René David, Arbitration in International Trade (The Hague: Kluwer Law and

Taxation Publishing, 1985), 334–335.

289 See, e.g., ICC Case No. 7544 (1999) Clunet 1062 with note by Hascher where

the tribunals' authorisation was ‘including a change of the contract itself.

290 Section 1020(4)(c) 1986 Dutch Arbitration Act.
291 Section 1(2) 1999 Swedish Arbitration Act.
292 Klaus Peter Berger, ‘Power of Arbitrators to Fill Gaps and Revise Contracts to

Make Sense’, Arbitration International 17, no. 1 (2001) p7.

293 Klaus Peter Berger, ‘Power of Arbitrators to Fill Gaps and Revise Contracts to

Make Sense’, Arbitration International 17, no. 1 (2001) p12.

294 See, e.g., ICC Cases Nos. 6515 and 6516 XXIV YBCA (1999), p80.
295 Bernard G. Poznanski, ‘The Nature and Extent of an Arbitrator's Powers in

International Commercial Arbitration’, Journal of International Arbitration 4, no.

3 (1987): 79–80.
296 In ICC Case No. 5754 (unpublished) cited in Laurence Craig, William W. Park

& Jan Paulsson, Annotated Guide to the 1998 ICC Arbitration Rules with
commentary, 3rd edn (Oxford: Oxford University Press, 2000), 112 a tribunal
held that a standard ICC arbitration clause could cover adaptation of contracts
where these were long term in nature including provisions that would naturally
require adjustment over the period of the contract.
297 Klaus Peter Berger, ‘Power of Arbitrators to Fill Gaps and Revise Contracts to

Make Sense’, Arbitration International 17, no. 1 (2001): 1.

298 Ibid., 7.
299 Ibid., 14.
300 International Law Association Final Report: ‘Ascertaining the Contents of the

Applicable Law in International Commercial Arbitration’, in 73 Int'l L. Ass'n Rep.

Conf. (2008), 2008, at 861.
301 This is the approach in most civilian jurisdictions.
302 This is the norm in English based common law systems.
303 The US position.
304 Sections 1.8 and 10.2 deal with differences between common law countries

and some fundamental due process questions.

305 Mustill and Boyd suggest: ‘First the arbitrator should recall that it is for the

parties to allege that the foreign law differs from English law. If they are content
to have their disputes decided according to English law, it is no part of his
function to multiply trouble and expense by suggesting that the two laws differ’.
Michael J. Mustill & Stewart C. Boyd, The Law and Practice of Commercial
Arbitration in England, 2nd edn (London: Butterworths, 1989), 72.
306 The practical presumption that foreign law is the same was applied in

Huffman (Europe) Ltd v. Al Ameen Dev & Trade Co [2000] EWHC 210 para. 42.
307 Rule 44.1 of the US Federal Rules of Civil Procedure provides that ‘[…] In

determining foreign law, the court may consider any relevant material or source,
including testimony, whether or not submitted by a party or admissible under
the Federal Rules of Evidence. […]’. Accordingly, the US position is not limited by
evidentiary admissibility rules and the court is entitled to engage in its own
research.
308 See, e.g., 1968 Council of Europe Convention on the Information on Foreign

Law. Even in jurisdictions that consider it a question of fact, it is left to the

determination of judges alone in jury trials.
309 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 692.
310 Gary B. Born, International Commercial Arbitration (The Hague, Kluwer Law

International, 2009), 2210, but noting some contrary cases such as ICC Case No.
3540 in S. Jarvin & Y. Derains (eds), Collection of ICC Arbitral Awards 1974–1985
(1990): 105.
311 Emmanuel Gaillard & John Savage (ed.), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 690.
312 Payment in Gold of Brazilian Federal Loans Contracted in France (France v.

Brazil), 1929 PCIJ. (ser. A) No. 21 (July 12), 124, para. 81.
313 Kaufmann-Kohler cites the examples of Spain and Mexico. Gabrielle
Kaufmann-Kohler, ‘Arbitral Precedent: Dream, Necessity or Excuse?’ Arbitration
International, 23 no. 3 (2007): 359.
314 FA Hayek, Law, Legislation and Liberty: Rules and Order, 119 (1973). W Mark

C Weidemaier, ‘Toward a Theory of Precedent in Arbitration’, Wm & Mary L Rev.

51 (2010): 1895.
315 Lon Fuller, The Morality of Law, revised edn (London: Yale University Press,

1969).
316 William W. Park, ‘Arbitrators and Accuracy’, Journal of International Dispute

Settlement 1, no. 1 (2010): 25, 49.

317 See, e.g., Mitchell v. Democratic Republic of Congo, ICSID Case No. Arb/99/7,

Decision on the Stay of Enforcement of the Award, 30 November 2004, para. 23.
318 For a detailed discussion of the use of past cases in investment arbitration

see AES Corp v. Argentine Republic, ICSID Case No. Arb/02/17, Decision on
Jurisdiction, 26 April 2005 paras 17–33.
319 See, e.g., Art. 59 of the ICJ Statute.
320 United States Import Prohibitionist Certain Shrimp and Shrimp Products,

WTO Appellate Body Report, 12 October 1998, WT/DS58/AB/R, para. 108.

321 Jeffery P. Commission, ‘Precedent in International Treaty Arbitration: The

Empirical Backing’, Transnational Dispute Management 4, no. 5 (2007): 6;

Gabrielle Kaufmann-Kohler, ‘Arbitral Precedent: Dream, Necessity or Excuse?’,
Arbitration International 23, no. 3 (2007): 357.
322 Article 59 of the ICJ Statute indicates that court decisions have ‘no binding
force except between the parties and in respect of that particular case’. Art.
1136(1) of NAFTA is the similar effect. Art. 53(1) of the ICSID Convention simply
provides that the award is binding on the parties, which some take to also mean
that it has no other effect. The more limited reference to the binding nature of
the award as between the parties is found in numerous arbitral rules.322 It
seems highly unlikely that the intent behind such phraseology is to make a
comment about the precedent effect of decisions. Instead, such comments are
about the parties acknowledging through the express wording of the rules that
they agree to be bound by the award.
323 Gabrielle Kaufmann-Kohler, ‘Arbitral Precedent: Dream, Necessity or
Excuse?’, Arbitration International, 23 no. 3 (2007): 362.
324 Ibid.
325 A tribunal will look to such cases for evidence of acceptance of the norms and

not for creation of the norms per se. Gabrielle Kaufmann-Kohler, ‘Arbitral
Precedent: Dream, Necessity or Excuse?’ Arbitration International, 23 no. 3
(2007): 364.
326 See <www.kluwerarbitration.com>.
327 See <www.uncitral.org/uncitral/en/case_law.html>.
328 The following sections draw on and extract from J. Waincymer, ‘International

Arbitration and the Duty to Know the Law’, Journal of International Arbitration,
28 no. 3 (2011): 201.
329 Michael J. Mustill & Stewart C. Boyd, The Law and Practice of Commercial

Arbitration in England, 2nd edn (London: Butterworths, 1989), 299.

330 F.A. Mann, ‘Fusion of the Legal Profession’, Law Quarterly Review 93, no. 3

(1977): 367, 375. Mann considered that the rule did not apply in the common
law, which instead effectively applied a principle of judicial unpreparedness. This
is because under the adversarial system, the court relies on counsel for the
points of law on which the case will be decided.
331 A number of authors describe it as jura novit curia.
332 Article 16, Swiss Private International Law states: ‘The contents of the

foreign law shall be established ex officio. The assistance of the parties may be
requested. In the case of pecuniary claims, the burden of proof on the content of
the foreign law may be imposed on the parties.’
333 Article 293 of the German ZPO states: ‘The law which is in force in another

state, customary law and by-laws require proof only to such extent as they are
unknown to the court. In the establishment of these legal norms, the court is not
limited to the evidence brought forward by the parties; it is empowered to make
use of other sources of knowledge and to order whatever is necessary for the
purposes of such utilisation’.
334 K.H. Böckstiegel, S. Kröll & P. Nacimiento, Arbitration in Germany: The Model

Law in Practice (The Hague: Kluwer Law International, 2008), 282–283. See also
Judgment of 6 December 2001, XXIX YB Comm Arb 742 (Oberlandesgericht
Stuttgart) (2004). Wetter makes a distinction between the iura principle and the
notion of materielle Prozessleitung known in Germanic legal systems as
‘directions pertaining to the substantive submissions and argument of a party …’
See J. Gillis Wetter, ‘Procedures for Avoiding Unexpected Legal Issues’, in
Planning Efficient Arbitration Proceedings: The Law Applicable in International
Arbitration, ICCA Congress Series No. 7, (Vienna 1994), ed. Albert Jan van den
Berg (Paris: ICC Publishing, 1999), 95. The latter relates to the proactivity of an
adjudicator in explaining concerns with existing submissions and inviting
attention to areas of particular concern. A decision of the European Commission
of Human Rights had considered that the German court practice applying iura
without necessarily inviting submissions from the parties did not constitute an
infringement of the ‘fair hearing’ requirement under Art. 6 of the European
Convention on Human Rights. European Commission of Human Rights,
Application 3147/67, Yearbook of Human Rights or Collection of Decisions 27,
119. The European Court of Justice in jeroen Van Schijndel and Johannes Van Veen
v. Stichting Pensioensonds voor Fysiotherapeuten [1996] 1 CMLR 801 noted
academic criticism and differing national perspectives and stated ‘…in many
systems, if a court does raise a new point, it will, or must, invite the parties to
address argument to the point…’ (para. 34).
335 Dr Rainer Hausmann, ‘Pleading and Proof of Foreign Law – A Comparative

Analysis’, The European Legal Forum, no. 1 (2008):I-1-I 13, available at

<www.simons-law.com/library/pdf/e/878.pdf>.
336 See Julian D.M. Lew, ‘Proof of Applicable Law in International Commercial

Arbitration’, in Festschrift für Otto Sandrock zum Mai70. Geburtstag, ed. Klaus
Peter Berger & Otto Sandrock (Frankfurt am n: Verlag Recht und Wirstchaft,
2000), 581, 595.
337 John Morris & Lawrence Collins (eds), Dicey and Morris on The Conflict of

Laws, 13th edn (London: Sweet & Maxwell, 2000), 221; Richard Fentimin,
Foreign Law in English Courts – Pleading, Proof and Choice of Law (Oxford: Oxford
University Press, 1998), 60; Trevor Harrley, ‘Pleading and Proof of Foreign Law:
The Major European Systems Compared’, International and Comparative Law
Quarterly 45, no. 2 (1996): 283.
338 Gabrielle Kaufmann-Kohler, ‘Globalization of Arbitral Procedure’, Vanderbilt

Journal of Transnational Law 36, no. 4 (2003): 1313, 1331.

339 Rule 44.1 US Federal Rules of Civil Procedure. See further Loise E. Teitz,

‘From the Courthouse in Tobago to the Internet: The Increasing Need to Prove
Foreign Law in US Courts’, Journal of Maritime Law and Commerce 34 (2003): 97.
340 See, e.g., Barry A. Miller, ‘Sua Sponte Appellate Ruling: When Courts Deprive

Litigants of an Opportunity to be Heard?’, San Diego Law Review 39, no. 4 (2002):
1253.
341 Teresa Isele, ‘The Principle Iura Novit Curia in International Commercial

Arbitration’, Int ALR 13, no. 1 (2010): 14. Li describes the question as being
‘(o)nce the choice of the applicable law is made, how does one know the parts of
the law which bear immediate relevance to the dispute at hand? In other words,
how should the burden of ascertaining the contents of the applicable law be
allocated between the parties and the arbitrators?’ Tao Li (2009), Penetrating the
Myths – Jura Novit Curia in International Commercial Arbitration (Master's
Thesis, Stockholm University), 3. Kalnina asks the question ‘once the arbitrator
has determined the applicable substantive law … how should the arbitrator
proceed in ascertaining its content?’ leva Kalnina, ‘Iura Novit Curia: Scylla and
Charybdis of International Arbitration?’, Baltic Yearbook of International Law 8,
no. 1 (2008): 89, 90. See also Christian P. Alberti, ‘Iura Novit Curia in
International Commercial Arbitration: How Much Justice Do You Want?’, in Liber
Amicorum Eric Bergsten: International Arbitration and International Commercial
Law: Synergy, Convergence and Evolution, ed. Kröll et al. (Alphen aan den Rijn,
Kluwer Law International, 2011), 3.
342 Douglas Brooker, ‘Va Savoir! – The Adage “Jura Novit Curia” in Contemporary

France’ (30 October 2005), bepress Legal Series. Working Paper 845, 7–8.
343 Tao Li, Penetrating the Myths – jura Novit Curia in International Commercial

Arbitration (Master's Thesis, Stockholm University, 2009), 10.

344 An ILA Report in 2008 considered that ascertainment of applicable law is ‘by

and large procedural…’. The report notes that it may have a substantive element,
in terms of the fallback applied when the content of applicable rules cannot be
sufficiently ascertained, but this could still be argued to be procedural. See The
Final Report of the International Commercial Arbitration Committee of the
International Law Association in the Rio de Janeiro Conference (2008),
Ascertaining the Contents of the Applicable Law in International Commercial
Arbitration, 16.
345 This is discussed in section 3.7.
346 Section 56(7) of the Hong Kong Arbitration Ordinance provides that the

‘tribunal may decide whether and to what extent it should itself take the
initiative in ascertaining the facts and the law relevant to those proceedings’.
347 MattiS. Kurkela, “Jura Novit Curia’ and the Burden of Education in
International Arbitration – A Nordic Perspective’, ASA Bulletin 21, no. 3 (2003):
499.
348 S Jarvin, ‘La Nouvelle Loi Suédoise sur 1'Arbitrage’, Rev Arb 1, 27 (2000): 58–

59, cited in Jean-François Poudret & Sébastien Besson, Comparative Law of

International Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 483.
349 Michael E Schneider, ‘Combining Arbitration with Conciliation’, in Planning

Efficient Arbitration Proceedings/The Law Applicable in International Arbitration,

ICCA Congress Series No. 7, ed. Albert Jan van den Berg (The Hague: Kluwer Law
International, 1996), 61.
350 If the parties accept the relevance of the law that is being considered by the

tribunal, and address it accordingly, there is no problem. Two other possibilities

are where the parties expressly direct the tribunal not to apply the law it thinks
is most appropriate or to similar effect, direct the tribunal to apply a law it thinks
is not appropriate. In such circumstances, that is at the very least a subsequent
agreement by the parties which circumscribes the tribunal's mandate. The
tribunal should not ignore such a direction by the parties. In the extreme, if the
tribunal considers that this is an undue limit on the mandate originally accepted,
it might constitute grounds for valid resignation. Such an improper limit on the
jurisdictional mandate would arise inevitably where the parties seek to
proscribe the application of mandatory procedural norms. Arbitrators must
always apply mandatory procedural principles regardless of whether these are
referred to by the parties. A tribunal might also continue with the original
mandate where mandatory norms are sought to be avoided, on the basis that
such a mandate expressly or impliedly allows for applications of such laws and
any subsequent direction to the contrary by the parties is invalid. The position is
far more contentious with substantive laws given that some question whether
these should ever be mandatory. See further section 13.8.
351 For example, application of a particular legal system as substantive law may

lead to the opportunity to claim that certain contract terms are void for
unfairness.
352 Yasuhei Taniguchi, ‘The Obligation to Mitigate Damages’, in Evaluation of
Damages in International Arbitration, Dossier of the ICC Institute of World
Business Law, ed. Yves Derains & Richard H. Kreindler (Paris: ICC Publishing,
2006), 90.
353 Teresa Isele, ‘The Principle Iura Novit Curia in International Commercial

Arbitration’, Int ALR 13, no. 1 (2010): 14.

354 To similar effect is LCIA Art. 22.1(c). It indicates that subject to contrary

agreement in writing by the parties the tribunal ‘shall have the power, on the
application of any party or of its own motion … to conduct such inquiries … in
identifying the issues and ascertaining the relevant facts and the law(s) or rules
of law applicable to the arbitration, the merits of the parties' dispute and the
arbitration agreement’. This is subject to the parties being given a reasonable
opportunity to state their views.
355 James H. Carter, ‘The Rights and Duties of the Arbitrator: Six Aspects of the

Rule of Reasonableness’, in The Status of the Arbitrator, ICC International Court of

Arbitration Bulletin: 1995 Special Supplement, ed. Jean-François Bourque (Paris:
ICC Publishing, 1995), 32 citing a sample used in an ICC Arbitration Program,
New York, 27 May 1992.
356 R.W. Naimark & S.E. Keer, ‘What Do Parties Really Want in International

Commercial Arbitration?’, Dispute Resolution Journal 57, no. 4 (2002): 78, 80.
357 Klaus Peter Berger, International Economic Arbitration (The Hague: Kluwer

Law and Taxation Publishers, 1993), 506.

358 See Jean-François Poudret & Sébastien Besson, Comparative Law of
International Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 475–476,
dealing with such an approach by Belgian courts, a more equivocal position in
Germany and some resiling from such a presumption in Swiss courts.
359 See the decision of the Swiss Federal Tribunal in case 4P 146/2004 (28

September 2004), 24 ASA Bulletin 2/2006, 318; 25 ASA Bulletin 3/2007, 582.
360 See the decision of the Swiss Federal Tribunal in case 4P 100/2003 (30

September 2003), 22 ASA Bulletin 3/2004, 574; 4P 4/2007 (26 September

2007), 26 ASA Bulletin 1/2008, 152.
361 Swiss Federal Court, CAS Decision, 30 September 2003:22 22 ASA Bulletin

3/2004, 574; 9 February 2009, 4A 400/2008, 27 ASA Bulletin 3/2009, 495. The
Swiss Tribunal Fédéral considered that the parties were entitled to comment in
Urqurjo Grotia v. Da Silva Muñiz (No 4A) 400/2008, Rre Cour de droit civil, 9
February 2009.
362 BGE 130 III 35; BGE 4P 260/2000 (2 March 2001). The Court considered that

there had not been surprise in the case before it as one party had referred to the
relevant provisions indirectly through referencing the consequences of their
application, ratherthan explicitly mentioning the provisions themselves. In an
earlier decision, the Swiss Federal Tribunal made a broader assertion as to the
obligation in stating that ‘the arbitrator, like ajudge of a State court, is not obliged
to submit to the discussion by the parties the legal principles on which he
willbase his decision. However, according to doctrinal opinion, the arbitrator
who specialised and who has access to sources and knowledge which are not
always at the disposal of the parties, has an obligation to bring in advance to the
attention of the parties the fundamental technical elements on which his decision
will be based’. (Judgment of 8 February 1978, Chrome Resources SA v. Léopold
Lazarus Ltd, XIYB Comm Arb 538 (Swiss Federal Tribunal) (1986).) In the CAS
decision (Case Reference 4A 400/2008). the Swiss Federal Tribunal quashed the
Court of Arbitration's award when the tribunal invoked a Swiss statute which
annuls exclusivity clauses in agency agreements that relate to employment
contracts. The provision had not been invoked by the parties. The case is unusual
as the tribunal concluded that the statute could not apply in any event as it only
applied to residents of Switzerland. Switzerland was selected as the Seat but all
parties were foreign. The application of the law was erroneous and hence would
have to have been a surprise to the parties.
363 Mrs Y v. Z and Ad Hoc Arbitral Tribunal Swiss International Arbitration Law

Report 2007 Vol 1 Nos 1 & 2, 265.

364 BGE 4P 260/2000 (2 March 2001).
365 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 482.

366 Antonias Dimolitsa, ‘The Equivocal Power of the Arbitrators to Introduce Ex

Officio New Issues of Law’, ASA Bulletin 27, no. 3 (2009): 429, fn 9 referring to an
award under the Arbitration Institute of the Stockholm Chamber of Commerce,
<http://ita.law.uvic.ca/document/Bogdanov-Moldova-2...>, upheld Svea Court of
Appeal 28 November 2008 – not published.
367 Decision of the Supreme Court of Finalnd, Judgment Rendered 2 July 2008 in

Case No S2006/716, Werfen Austria GmbH v. Polar Electro Europe BV Zug Branch.
368 Louis Dreyfus SAS v. Holding Tustulum BV 2008 QCCS 5903, Superior Court of

Canada, 8 December 2008.

369 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 475.

370 Nouveau Code de Procédure Civile (France) Art. 4. Poudret and Besson suggest

that ‘the majority of French decisions seem to admit that the arbitrator can rely
on reasons which are based on the applicable law and are implicitly included in
the debate without inviting the parties to express themselves thereon’. They cite
a range of cases, including one where the arbitrators relied on the usages of
international trade. Rev Arb 1982, 183 (Second Case) with a note by Couchez.
They also cite a case where the principle of good faith was applied. French cases
which supported arbitral behaviour contrary to the principle of contradiction
were criticised by Kessedjian, although Poudret and Besson take a different view
(Catherine Kessedjian, ‘Principe de la Contradiction et Arbitrage’, Rev Arb (1995)
381). When dealing with those French cases that overturn arbitral behaviour,
Poudret and Besson suggest that they involve instances of decisions ultra petita.
They conclude from the case law that an ‘arbitral tribunal does not have the duty
to provoke explanations by the parties on the legal provisions which result from
the applicable law and on the legal characterisation of the acts or agreements
which are the subject matter of the dispute’. In these circumstances there can be
no element of surprise. Ibid.
371 Brussels Court of Appeal, 24 January 1995, XXII YB Comm Arb (1997), 655.
372 Singapore Court of Appeal, 9 May 2007, Soh Beng Tee & Co Pte Ltd v.

Fairmount Development Pte Ltd [2007] 3 SLR 86.

373 Modern Engineering v. Miskin [1981] 1 Lloyds Rep 135 (CA).
374 Judgment of 24 January 1997, Inter-Arab Inv Guarantee Corp v. Banque Arabe

et Internationale d'Investitssements, XXII YB Comm Arb 643 (Brussels Cour

d'Appel) (1997).
375 OAO Northern Shipping Co v. Remolcadores De Marin FL (the Remmar Case)

[2007] EWHC 1821 (Comm) quoting ABD AG v. Hochtief Airport GmbH [2006] 2
Lloyd's Rep 1, para. 70.
376 Modern Engineering v. Miskin [1981] 1 Lloyd's Rep 131; Pacolv. Rossakhar

[2000] 1 Lloyd's Rep 109; Sanghiv. The International Investor(CKFC) [2000] 1

Lloyd's Rep 480. Another case where an arbitrator's decision was overturned
because the tribunal decided on an issue not addressed was ‘Re Vimeira’. [1984]
2 Lloyd's Rep 66.
377 Bulfracht (Cyprus) Ltd v. Boneset Shipping Co Ltd. ‘MV Panthilox’ [2002] 2

Lloyd's Rep 681 (QB). English cases that have overturned or remitted
arbitrations for a lack of an opportunity to be heard include the Remmar case,
Vee Networks Ltd v. Econet Wireless International Ltd [2004] EWHC 2909
(Comm) and Pacol Ltd v. Joint Stock Co Rossakhar [2001] 1 Lloyd's Rep 109. Lord
Hodson in the House of Lords considered that there is an ‘inestimable benefit of
critical argument…’ In re K (Infants) [1965] AC 201, 234 (HL) per Lord Hodson.
See also Goldsmith v. Sperrings Ltd [1977] 1 WLR 478, 486. Kaplan LJ refused
enforcement on due process grounds where an arbitrator's independent fact
finding powers were utilised without giving the parties appropriate opportunity
to comment. Paklito Investment Ltd v. Klockner East Asia Ltd 15 January 1993,
cited in ILA Report, above 20, 5, footnote 10. In Huffman (see, e.g., Huffman
(Europe) Ltd v. Al Ameen Lev & Trade Co [2000] EWHC 210 para. 42), a tribunal
appointed its own expert on foreign law and met with him without discussing it
with the parties. The Court considered this to be in error but not an irregularity
for the purposes of s. 68 of the English Arbitration Act 1996.
378 The Lotus case PCD Series A No. 10 (1927), 26; Fisheries Jurisdiction case (UK

v. Iceland) ICJ Reports 1974, 9; Case Concerning Military and Paramilitary

Activities in and against Nicaragua (Nicaragua v. USA) Merits, ICJ Reports, 1986,
24.
379 The WTO Appellate Body in Conditions for the Granting of Tariff Preferences

to Developing Countries (European Communities) WT/DS246/AB/R referred to

ICJ cases articulating the Iura principle but it seems not to be readily applicable
given that the reference was to legal interpretations of provisions clearly in
dispute. See also WTO Appellate Body Final Report on EC Measures Concerning
Meat and Meat Products (Dormones) AB-1997-4 para. 156.
380 ILA Report, supra n. 20, 10. Moiwana Village v. Suriname 15 June 2005, Inter-

Am Ct DR (Ser C) No. 124, 102. The Inter-American Court of Duman Rights has
invoked the principle in relation to violations where the parties have had the
opportunity to express their positions in relation to the relevant facts.
381 Mohamed Shahabuddeen, Precedent in the World Court (New York:
Cambridge Dniversity Press, 1996), 140.
382 Ibid., 137. See also Fisheries jurisdiction (United Kingdom v. Iceland) 1974 ICJ

3, 9.
383 Case Concerning Military and Paramilitary Activities in and against Nicaragua

(Nicaragua v. USA) Merits, ICJ Reports, 1986, 25.

384 Nuclear Test (Australia v. France) (Judgment) [1974] ICJ Rep 253, 259-260

citing in part Northern Cameroon (Cameroon v. United Kingdom) (Preliminary

Objections) [1963] ICJ Rep 15, 29.
385 Separate Opinion of Judge Higgins in Legality of Use of Force (Serbia and

Montenegro v. Belgium) (Preliminary Objections) [1999] ICJ Rep 279, 338–339.

386 Joost Pauwelyn, Conflict of Norms in Public International Law: How WTO Law

Relates to Other Rules of International Law (New York: Cambridge University

Press, 2003), 447–448; Joost Pauwelyn, ‘The Role of Public International Law in
the WTO: How Far Can We Go?’, American Journal of International Law 95, no. 3
(2001): 535, 555.
387 Gulf of Maine (Canada v. United States) (Judgment) [1984] ICJ Rep 246, 305.
388 Chester Brown, A Common Law of International Adjudication (New York:

Oxford University Press, 2007), 79.

389 Klockner v. Cameroon Decision on Annulment, 3 May 1985, 2 ICSID Reports

95 at para. 91.
390 Ibid.
391 Yurii Bogdanov, Agurdino-Invest Ltd and Agurdino-Chimia JSC v. Republic of

Moldova, SCC, Arbitral Award 22 September 2005 para. 2.2.1.

392 Antonias Dimolitsa, ‘The Equivocal Power of the Arbitrators to Introduce Ex

Officio New Issues of Law’, ASA Bulletin 27, no. 3 (2009): 430, fn 9 referring to
Mitchell v. Congo ICSID Case Arb/99/7 1 November 2006, Decision on the
Application for Annulment of the Award, para. 57.
<http://ita.law.uvic.ca/document/mitchellannulment....>.
393 Dissenting Opinion of Judge Howard M Holtzmann in Dalla v. Iran and Bank

Mellat, Award No. 53-149-1, 3 Iran US CTR 10 at 28 (1983).

394 Antonias Dimolitsa, ‘The Equivocal Power of the Arbitrators to Introduce Ex

Officio New Issues of Law’, ASA Bulletin 27, no. 3 (2009): 429.
395 Julian Lew, Loukas Mistelis & Stefan Kröll, Comparative International
Commercial Arbitration (The Hague: Kluwer Law International, 2003), 95.
396 Teresa Isele, ‘The Principle Iura Novit Curia in International Commercial

Arbitration’, Int ALR 13, no. 1 (2010): 24–25.

397 Emmanuel Gaillard & John Savage (ed.), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 692.
398 Ibid.
399 The Final Report of the International Commercial Arbitration Committee of

the International Law Association in the Rio de Janeiro Conference (2008),

Ascertaining the Contents of the Applicable Law in International Commercial
Arbitration, 880.
400 See Resolution No. 1/2008 in ILA Report, 2008, at 59. The ILA Committee

identified some examples, being where one party is not represented, where the
tribunal can get no guidance as to the content of applicable law or when interim
relief is sought.
401 Gabrielle Kaufmann-Kohler, “Iura Novit Arbiter’ – est-ce bien raisonable?

Réflexions sur le statut de droit de fond devant l'arbitre international’, in De Lege

Ferenda Etudes Pour le Professeur Alain Hirsch, ed. A. Heritier Lachat & L. Hirsch
(Geneva: Slatkine, 2004), 71, 78.
402 J. Gillis Wetter, ‘Procedures for Avoiding Unexpected Legal Issues’, in
Planning Efficient Arbitration Proceedings: The Law Applicable in International
Arbitration, ICCA Congress Series No. 7, (Vienna 1994), ed. Albert Jan van den
Berg (Paris: ICC Publishing, 1999), 91 describes this as based on ‘professional
prudence’.
403 Ibid., 90.
404 Ibid.
405 The Final Report of the International Commercial Arbitration Committee of

the International Law Association in the Rio de Janeiro Conference (2008),

Ascertaining the Contents of the Applicable Law in International Commercial
Arbitration, 2.
406 This occurred in a Swiss Federal Tribunal decision in 2002 which was further

criticised by Poudret and Besson. Jean-François Poudret & Sébastien Besson,

Comparative Law of International Arbitration, 2nd edn (London: Sweet &
Maxwell, 2007), 745–746 referring to a decision of the Swiss Federal Tribunal
reported in 3 ASA Bull 20 (2002) 493.
407 Klaus Peter Berger, International Economic Arbitration (The Hague: Kluwer

Law and Taxation Publishers, 1993), 502.

408 Ibid., 500.
409 Ibid., 494.
410 Problems only arise with mandatory laws, discussed in section 13.8.
411 See Lisa Spagnolo, ‘Green Eggs and Ham: The CISG, Path Dependence, and the

Behavioural Economics of Lawyers’ Choices of Law in International Sales

Contracts', Journal of Private International Law 6, no. 2 (2010): 417–464.
412 J. Gillis Wetter, ‘Procedures for Avoiding Unexpected Legal Issues’, in
Planning Efficient Arbitration Proceedings: The Law Applicable in International
Arbitration, ICCA Congress Series No. 7, (Vienna 1994), ed. Albert Jan van den
Berg (Paris: ICC Publishing, 1999), 91.
413 This was considered in section 3.7.2.
414 Eco Swiss China Time Ltd v. Benetton International NV, 1 June 1999, ECJ, Case

C-126/97, [1999] ECR I-3055 ECJ. In Mostaza Claro, the ECJ considered that a
national court was required to consider whether an award was in violation of EU
consumer law even when not pleaded by the parties. Eliza Maria Mostaza Claro v.
Centro Movil Milenium SL, 26 October 2006, ECJ, Case C168/05, [2006] All ER (D)
322 (Oct). The ECJ considered that where a national court is considering
annulment of an arbitration award, it must determine whether the agreement is
void and annul the award if the agreement contains an unfair term contrary to
the Council Directive on Unfair Terms in Consumer Contracts (see paragraph 39
of the judgment).
415 ILA Report, 2008, 7.
416 See, e.g., unreported award dated February 8, 1994, ICC Case No. 7071 cited

in ILA Report, 2008, 17.

417 Gary B. Born, International Commercial Arbitration (The Hague, Kluwer Law

International, 2009), 1626.

418 J. Gillis Wetter, ‘Procedures for Avoiding Unexpected Legal Issues’, in
Planning Efficient Arbitration Proceedings: The Law Applicable in International
Arbitration, ICCA Congress Series No. 7, (Vienna 1994), ed. Albert Jan van den
Berg (Paris: ICC Publishing, 1999), 93.
419 Ibid.
420 Ibid., 92–93.
421 Antonias Dimolitsa, ‘The Equivocal Power of the Arbitrators to Introduce Ex

Officio New Issues of Law’, ASA Bulletin 27, no. 3 (2009): 432.
422 J. Gillis Wetter, ‘Procedures for Avoiding Unexpected Legal Issues’, in
Planning Efficient Arbitration Proceedings: The Law Applicable in International
Arbitration, ICCA Congress Series No. 7, (Vienna 1994), ed. Albert Jan van den
Berg (Paris: ICC Publishing, 1999), 97.
423 BP Exploration Co (Libya) Ltd v. The Government of the Libyan Arab Republic,

10 October 1073, Award on the Merits, 53 ILR 297 (1979).

424 ILA Report, 2008:18.
425 J. Gillis Wetter, ‘Procedures for Avoiding Unexpected Legal Issues’, in
Planning Efficient Arbitration Proceedings: The Law Applicable in International
Arbitration, ICCA Congress Series No. 7, (Vienna 1994), ed. Albert Jan van den
Berg (Paris: ICC Publishing, 1999), 97.
426 V.V. Veeder, ‘Compound Interest and Specific Performance: “Arbitral
Imperium” and Sections 49 and 48 of the English Arbitration Act 1996’, in
Interest, Auxiliary and Alternative Remedies in International Arbitration, Dossier of
the ICC Institute of World Business law, ed. Laurent Lévy & Filip De Ly (Paris: ICC
Publishing, 2008), 344 ff. Veeder also cited Lord Wilberforce's extra-judicial
speech in support of the Arbitration Bill when it received its Second Reading in
the House of Lords referring to his wish to see arbitration ‘regarded as a free
standing system, free to settle its own procedure and free to develop its own
substantive law…’. Hansard, Cole 778, 18 January 1996.
427 William W. Park, ‘Arbitrators and Accuracy’, Journal of International Dispute

Settlement 1, no. 1 (2010): 25, 43.

428 Domestic use of notions of comity to similar effect may be less relevant

where an arbitrator is concerned, not being concerned with intergovernmental

relationships through being a servant of one of them.

Part III: The Award, Chapter 14:

Remedies and Interest
Jeff Waincymer,

14.1. Introduction and Classification

This book concentrates on procedural and evidentiary matters in arbitration.

Hence, it does not seek to outline key principles of substantive law that are
essential to the arbitral process. Nevertheless, as noted in Chapter 13, the
application of substantive law commonly depends heavily on procedural issues
as to the way substantive law is determined and on evidentiary matters as to
how the law is proven and how salient facts are found. This is also the case
where remedies are concerned, as these are generally a subset of substantive
applicable law. Another reason to consider these issues is that not all systems
see all aspects of remedies as substantive. Lex arbitri or rules may deal with
some matters at least, such as interest. Classification issues in conflicts of law
can also treat some issues as either procedural or substantive. An example is
limitation of actions. (1) The law relating to damages is classified by some
systems as partly procedural and partly substantive. (2) Equitable doctrines may
present particular difficulties with regards to classification. Tracing, for
example, involves the identification of assets which could be regarded as an
evidentiary and, therefore, procedural matter. However, Dicey and Morris
suggest that tracing is a substantive issue as it is an essential step in the bringing
of substantive actions in the law of property. (3) Party autonomy can also come
into play, with contracts and treaties at times prescribing or proscribing certain
forms of relief. In some cases there may be a need to interpret ambiguous page
"1097" provisions in that regard, which in turn can raise evidentiary as well as
legal challenges.

Another reason to deal with practical aspects of remedies at least in an

introductory fashion is that it is a topic not usually dealt with in much detail in
treatises. Given that arbitration is essentially about seeking practical and timely
solutions to complex international commercial problems, it is perhaps
surprising that so little has been written about the nature of relief at the same
time as there is a degree of criticism of tribunals in relation to uncertainty,
inconsistency, lack of reasoning and the like where remedies are concerned. (4)

Some contentious issues covered in this chapter include what choice of law
should be made where a discretion exists, whether and to what extent non-
pecuniary remedies are permissible or appropriate, (5) methods of calculation of
pecuniary remedies; appropriate standards of proof, the enforceability of
liquidated damages and penalty provisions, issues of currency and timing and
the related principles of awards of interest.

14.2. Procedural and Evidentiary Aspects of Remedy Determinations

A gateway issue in many cases is that where tribunals have discretion as to

applicable law, this choice can have a major impact on the nature and extent of
remedies and hence a major impact on the prior rights and obligations of
contracting parties. While this can also occur with international litigation, the
difference with judges is that they are bound by their national systems' conflicts
rules. Even if it is uncertain what law will ultimately be applied, the rules are in
existence and known at the time of the contract. Hence, parties can at that stage
predict what law is likely to apply. Where arbitration is concerned, because an
arbitrator's discretion as to applicable law can come much later and is not
constrained by any national law, key decisions on substantive matters, including
the applicable law of remedies, can have major effective retroactive impact. This
can either open up or bar certain remedies or have significant impact upon
measurement of damages. For this reason alone, parties should always make a
selection consistent with their reasonable expectations as and when they set
price and other terms. An example of such an impact would be choosing
between a contract system that opts for damages as a primary remedy as
opposed to one which has specific performance as the standard norm. These
choices have both philosophical and commercial elements. For example,
Friedmann argues that specific performance is concerned with protection of the
page "1098" subjective value of the promised performance in the eyes of the
aggrieved party, while damages is concerned with objective value. (6) Where
there is a disparity between objective and subjective value, one form of remedy
will be preferable to the other. Such factors can add to the difficulty and
certainty as to tribunal determinations of applicable law.

In making discretionary choices between different systems of applicable law

and within systems as to which remedies to award, arbitrators are also making
decisions about the role of contract law and the interests worthy of protection
by legal systems. It is important to understand at the outset that within most
legal systems, there is an ongoing debate as to the ultimate aim of remedial
measures. Legal theorists have considered that damages in contract may be
concerned with restitutionary interests, which can be divided into reliance
interest and expectation interest. (7) Even when a particular substantive law has
been identified, it will often be the case that there are major debates within that
system as to the exact nature and form of remedies that should apply. An
important question is how an arbitrator determines the ambit of an applicable
national law that is in the middle of such flux and uncertainty. What is an
arbitrator to do when applying a law of contract damages where he or she
knows there is a major debate in the literature and case law about whether the
aim is to meet the performance interest, expectations interest, or reliance
interest? What attitude is taken to requests for disgorgement of profits when the
party in breach makes a windfall gain from that very activity, without causing
commensurate loss to the aggrieved party where the applicable law has not
taken a firm view on the question? Will arbitrators tend to be more conservative
than domestic judges in dealing with cutting edge questions of this nature?
Tribunals may not even be limited to a national law. Some tribunals seek to
bring in lex mercatoria, for example UNIDROIT Principles, through the notion of
usages of international trade, referred to in provisions such as Article 28(4) of
the UNCITRAL Model Law. (8)

There are also evidentiary issues as to how an applicable law is to be identified.

Chapter 13 looked at the distinction between common law systems which treat
it as a question of fact, albeit with some residual presumptions, and civilian
systems that may apply the iura novit curia principle. These issues are equally
applicable to the determination of the relevant laws on remedies. Similarly,
Chapter 2 looked at the duty to render an enforceable award and the possibility
that an enforcement country might see certain awards as against public policy.
In some cases, particular remedies, such as punitive damages, may be seen as
being against public policy in some systems.

page "1099"

Once the relevant laws have been determined, there are important procedural
and evidentiary questions as to the factual elements of remedy determinations,
particularly in relation to damages, where calculations are complex. A further
discussion of specific evidentiary challenges in damages assessments is
contained in section 14.5.4. Calculation of aspects of damages assessments are
discussed further in sections 14.7 and 14.8.

14.3. Party Autonomy, Lex Arbitri and Institutional Rules

As always, the powers of relief held by a tribunal can arise from the lex arbitri,
arbitral rules or the agreement of the parties. Party agreement can operate both
positively and negatively in terms of either granting or barring particular forms
of relief. Remedy conditions may be drafted by the parties in their initial
agreement or agreed to subsequently. Any alleged agreement of the parties as to
remedies could itself be subject to dispute and the ideal is that there is a clear
stipulation of the parties' wishes. (9) For example, there may be uncertainty
where the drafting uses a term such as ‘consequential damages’, typically in an
exclusion clause, which may have a different meaning in different legal systems.
(10) From a procedural perspective, any limitation on remedies should usually be

acceptable unless it is so imbalanced as to offend equal treatment of the parties.

Challenges are more likely under substantive norms. Excessive damages could
be seen as unreasonable or unenforceable penalties. Where contractual
provisions seek to limit damages, they might be subject to substantive law
principles that at times invalidate limitation clauses. (11)

While consent would generally be paramount, this is only vis-à-vis the rights of
the parties themselves. For example, the parties cannot simply agree to grant a
tribunal a power of disposition over third-party property. Relief may only be
granted for or against a party to the arbitration agreement, although that notion
might be extended to non-signatories under theories such as group of
companies, alter ego or assignment where such claims have been brought within
the page "1100" arbitration. More debateable would be remedies as against a
party that also adversely affect third parties. An example would be an order for
specific performance against a disputed grantor of a patent licence, when the
person instead intends to licence a third party allegedly in breach of the
agreement subject to arbitration. In many such cases, a tribunal might have a
power to order the remedy, but might choose not to do so because of third-party
impact and/or inability to have the third party assist in effecting the remedy. A
further aspect of party choice is that a tribunal is only entitled to grant the relief
as sought. A tribunal should only make an award in relation to the amount
sought even if the tribunal believes that a higher amount is appropriate. This is
also the case if interest is sought from a date later than could otherwise have
been argued. (12) Consequently, if a claim is only for declaratory relief, damages
cannot be awarded. However, if declaratory relief is that payment of purchase
price should be made good, the declaration itself leads to a monetary obligation.

Tribunal powers do not have to be exhaustively expressed in the lex arbitri or

rules. Consent to final and binding arbitral awards also implies consent to
dispositive powers. The general view is that every remedy available in litigation
is also available in arbitration, although that may be impacted upon by the
applicable law. (13) Some lex arbitri express this principle. For example, section
48(2) of the Arbitration Act 1996 (UK) indicates that absent a contrary
agreement of the parties, a tribunal may order payment of money, grant
declaratory relief and grant the same relief as an English court with regard to
injunctive relief, specific performance and rectification. (14) There are differences
in view as to whether a tribunal may even go beyond substantive relief that
could be afforded by a court, perhaps based on implied consent of the parties.
(15) Obviously that is contentious and would need to be based on some a priori

logic, given that at the time the relief is granted, the losing party would not be
supportive. While the entitlement to award monetary damages is not
contentious, there is more debate about some forms of non-monetary relief.
Here there are questions as to the relevant power, the role of discretions and the
criteria to apply. Care should be taken to ensure that there are no possible
challenges for exceeding authority, although such challenges are unlikely to be
successful. (16) To the extent that remedies are substantive, this will limit the
possibility for enforcement challenges to decisions made. page "1101" Even
then, while the award itself might not be challenged, an arbitrator should be
careful in asserting a power that does not exist.

One contentious issue is where a tribunal wishes there to be some form of

ongoing scrutiny of the suitability of the relief granted. For example, a tribunal
might award a certain form of relief on the understanding that concurrent
declaratory orders will be obeyed. To the extent that they are not, this may
adversely affect the suitability of the damages assessment that was made
contemporaneously. The starting presumption is that awards are meant to be
self-contained and tribunals are functus officio when the final award is rendered
save for correction, interpretation and gap filling powers. In some cases,
however, the fact that a tribunal seeks to maintain a supervisory function should
not automatically be a ground for challenge, although the tribunal may well be
functus officio, hence rendering the supervisory power inoperative. Some
national courts have considered that retaining jurisdiction does not contravene
the functus officio principle. (17) In some circumstances, the parties might be able
to commence a new arbitration on the basis that the failure to honour the
declaration was a breach of the arbitration agreement, hence giving a newly
constituted tribunal an opportunity to effectively alter the net damages within
the award.

Where there is a long term remedy proposed, the tribunal should certainly
consider how to frame this to maximise enforceability by an appropriate court if
the identity of that court is known. That should certainly be known in some
circumstances, such as an obligation to transfer a licence or patent after a
defined time period. A tribunal might always wish to carefully consider whether
to grant a form of relief where there may be no suitable oversight of long-term
remedies. (18)

14.4. Non-Pecuniary Remedial Measures

14.4.1. Restitution

While this book aims to deal with both commercial and investment matters
indiscriminately, it is important to at times differentiate between the two.
Remedies are perhaps the area where this is particularly important given the
significant difference between investment treaty norms of non-discrimination
and compensatory expropriation as opposed to typical contract breaches. Many
comments about non-pecuniary remedies have been made in the context of
investment and international cases and some caution should apply when
considering their potential application to private commercial disputes. Anne van
Aaken, for example, writes page "1102" that ‘it is clear’ that restitution is an
available remedy in international law. (19) Having said that, core principles
should prima facie apply in each arena. Generally speaking, arbitration is
concerned to grapple with questions of restitution when considered in its
broadest sense.

As a general rule, laws that apply to the availability of remedies in the law of
restitution are characterised as substantive rather than procedural. (20) In the
Chorzów Factory case the PCIJ said ‘reparation must, as far as possible, wipe out
all the consequences of the illegal act and re-establish the situation which
would, in all probability have existed, if that act had not been committed.
Restitution in kind, or, if this is not possible, payment of a sum corresponding to
the value which a restitution in kind would bear; the award, if need be, of
damages for loss sustained which would not be covered by restitution in kind or
payment in place of it – such are the principles which should serve to determine
the amount of compensation for an act contrary to international law’. (21) The
notion of restitution is thus also important in terms of calculating damages even
where physical restitution is not possible. (22) Article 36 of the ILC Articles on the
Responsibility of States for Internationally Wrongful Acts similarly indicates
that:

A State responsible for an internationally wrongful act is under an obligation to

make restitution, that is, to re-establish the situation which existed before the
wrongful act was committed, provided and to the extent that restitution:

(a) is not materially impossible;

(b) does not involve a burden out of all proportion to the benefit deriving from
restitution instead of compensation. (23)
It should be noted, however, that these norms were not drawn up with a view to
application in investment disputes involving private parties. (24) Nevertheless,
such principles have been applied in investment disputes, although not without
controversy. In an ad hoc award in Texaco Overseas Petroleum Co, the tribunal
page "1103" considered that ‘restitution in integrum is… the normal sanction for
nonperformance of contractual obligations …’ It considered that it was only
inapplicable ‘to the extent that restoration of the status quo ante is impossible’.
(25) A further example where restitution was ordered is one of the cases

concerning the rights of Britons in Spanish Morocco. (26) In that case, the tribunal
ordered the Spanish Government to provide replacement premises for the
British Consul in Tetuan because the Government was responsible for the
destruction of the Consul's previous premises.

Some international treaties expressly refer to restitution, including NAFTA (27)

and the Energy Charter Treaty. (28) Nevertheless, some express provisions allow
for a State to pay damages in lieu of other remedies as granted. (29) The 1955
Washington Convention does not expressly address the issue, although some
have questioned the implications of Article 54(1) of the Convention by which
parties promise to ‘recognise an award rendered pursuant to this Convention as
binding and enforce the pecuniary obligations imposed by that award…’.
Christoph Schreuer has analysed the history of the drafting and concluded that
this was a compromise position to accommodate the views of some delegates
that non-pecuniary penalties might not be enforceable on public policy grounds
but did not wish to proscribe such remedies and still wished to allow such
awards to have res judicata effect. (30)

Even if it is permissible to award restitution in kind, there is still a question as to

whether it is desirable. Restitution is rarely ordered in international arbitration,
as it is often not possible to undo the effects of breaches made and tribunals
‘quite rightly tend to avoid making awards that are difficult to enforce’. (31) The
award of remedies other than damages by international arbitral tribunals is,
therefore, extremely unusual. (32) The difficulties associated with restitution are
considerable, and include the length of time that has elapsed since the original
unlawful act may make restitution impossible; restitution may not constitute an
adequate remedy for all damage involved and it cannot repair moral or personal
injury. (33) Redfern and Hunter suggest that monetary compensation in lieu of
restitution is a better page "1104" alternative. (34) Moreover, they regard the case
of Texaco as an ‘apparent exception’ which is ‘difficult to accept… as a precedent
for the effective granting of restitution in international commercial arbitration’.
(35) They note that the case has been severely criticised (36) and that a different

conclusion was reached on the same facts in the case of British Petroleum
Company (Libya) Ltd v. The Government of the Libyan Arab Republic. (37)

The debate is impacted upon by the type of case. Most controversial are
situations of expropriation where sovereign rights have to be balanced with
notions of just compensation. In the BP case just mentioned, Judge Lagergen, as
sole arbitrator, considered that even where the nationalisation was unlawful,
restitution was inapplicable because the ‘nationalisation is de jure an exercise of
territorial sovereignty, coupled with the low likelihood that a nationalising State
will reprivatise a nationalised entity …’. (38) It has been noted that there are
practical problems in enforcing non-pecuniary remedies against States as this
would require a State's cooperation within its own territory while damages
awards may be enforceable anywhere. (39) Such practical problems were the
source of the arbitrator's decision in the Walter Fletcher Smith (40) case, where
the claimant US requested the restoration of property that had been illegally
seized by Cuba, or in the alternative pecuniary compensation. The arbitrator
held that it was in the best interests of the parties and the public that the
remedy be damages. Sovereign rights may be considered to be more important,
the more the investment relates to vital industries such as food, resources,
health, communications or transport. (41) Moreover, it has been questioned
whether the award made in Texaco was even intended page "1105" to be
enforceable, in that the claimants in that case were only seeking an authoritative
legal opinion on the merits of the case, not an enforceable award. (42)

For these reasons, in most cases an investor will only seek monetary damages
and not wish to attempt to continue the investment within an unfavourably
disposed host State. That will not always be so, particularly when the
investment is intact but some important entitlement has been removed, as was
the case in Goetz v. Burundi. In that case the arbitrators gave the responding
State the option of agreeing on monetary compensation or re-awarding a free
trade zone certificate wrongly revoked. (43) The tribunal made the comment in a
partial award on liability. The tribunal's suggestions were not remedies for
illegal acts, but instead, suggested compensation to make the expropriation
lawful. (44)

14.4.2. Specific Performance

Specific performance is typically akin to the restitutionary norms discussed in

the previous section where the aim is to effectively restore the position prior to
breach by calling for the action that should have occurred but for the breach.
Whether a tribunal may order specific performance is generally seen as a
question of substantive law, although there is some uncertainty simply because
some instruments suggest that this may at times be impacted upon by the law of
the forum. For example, international instruments will often reserve some
powers to the law of the forum to determine whether adjudicators can grant
specific performance. Examples include Article 12.1(c) of the Rome I regulation
and Article 28 of the CISC (45) Nevertheless, Rome I does not apply directly to
arbitration and the page "1106" better view is that one should look to the lex
causae to determine this matter. (46) The CISG may itself be the applicable law
but again Article 28 does not speak of arbitration. Any limitation on the right to
award specific performance in the lex arbitri would take precedence over the
substantive law unless it is subject to a contrary stipulation of the parties. An
example is section 48(5)(b) of the Arbitration Act 1996 (UK) which allows a
tribunal to order specific performance generally, but not over a contract relating
to land. (47)

In commercial disputes, specific performance might typically be granted in sale

of goods contracts, loan agreements and transfers of intellectual property and
even licences on the breakdown of joint-venture, distribution or similar
arrangements. As long as the subject of a specific performance obligation is
within the arbitration agreement, it need not necessarily be the primary
contract that the dispute revolves around, although this would be the norm.
Specific performance orders do not in and of themselves cover the damage if any
from the delay in performance. A separate remedy would normally cover any
damage arising from the delay. (48) Principles of remoteness of damage are not
generally relevant to specific performance, nor are considerations of mitigation
of loss. (49)

According to Redfern and Hunter ‘the question of whether an arbitral tribunal is

empowered to order specific performance is …rarely an issue in international
arbitration’. (50) The reason for this is that specific performance is a principal
remedy for breach of contract in civil law jurisdictions, (51) and arbitral tribunals
in common law are either empowered to award this remedy or have seen
tribunals and supervisory courts supporting this broad approach. (52) Specific
performance is also page "1107" allowed for under the CISG (53) with the
qualification noted above and under the UNIDROIT Principles. (54) The relevant
substantive law will have its own internal principles as to when specific
performance would be a principal remedy. Under common law systems, it was
traditionally thought appropriate to award the remedy of specific performance
only when damages would not suffice. For example, specific performance might
be ordered when a buyer has sought to acquire the final rare postage stamp to
complete a particular set. It may be that some common law jurisdictions are
moving to a more open discretionary approach, allowing for specific
performance where it is an appropriate remedy in the circumstances of the case.
(55) Civilian systems generally start with a rebuttable presumption that it is the

preferred remedy. Notwithstanding the obvious differences between the legal

traditions, Chappuis suggests that the differences in practice are smaller than
one would suppose. (56) Some restrictions apply in all legal systems. Under both
common law and civil law systems, specific performance will not be granted to
force purely personal actions such as the creation of an industrial design or
other work of intellectual property. (57) Specific performance cannot be directed
so as to call for assistance by third parties, although specific performance can at
times have benefits for third parties who are not otherwise entitled to arbitral
determinations or remedies. (58)

One discretionary issue is whether contractual promises are best susceptible of

specific performance, particularly in the context of the ability to supervise and
enforce them within an arbitral context. Some regard the remedy of specific
performance as placing too high a responsibility on enforcing courts, arguing
that the New York Convention aims to provide for the minimal involvement of
courts at the page "1108" place of performance and cautioning that tribunals
should be loath to provide relief that requires ongoing supervision as to
compliance. (59) However, Schneider suggests that possible or real difficulties at
the level of enforcement should not be taken as the sole criterion for deciding
whether to grant the remedy of specific performance. (60) He points out that
‘experience and some studies’ have confirmed that the parties comply
voluntarily with most decisions made by arbitrators and that difficulties in
enforcement should not, therefore, deprive a party of obtaining an order for
specific performance, assuming the party is willing to assume the risk of future
difficulties in enforcement. (61) Some enforcement courts might even employ an
astreinte power, which would greatly increase the likelihood of compliance. (62)

There are discretionary factors to consider. Important situations include

ongoing joint-venture agreements where there has been a breakdown in the
interpersonal relationships between the parties. An adjudicator may not wish to
force unwilling parties to return to a collaborative endeavour. Schreuer suggests
that where there remains an ongoing relationship, there may be more
justification in applying non-pecuniary remedies. (63) It has been conversely
argued that ‘the more refined and complicated the human relationship, the less
reasonable it is to force the parties to be bound thereto against their will’. (64)
This is a particularly important issue given the high percentage of modern
contracts that are long term in nature, calling for close personal relationships
among contracting parties. Difficulties may also be exacerbated where
enforcement would be needed in a range of jurisdictions, for example where a
regional exclusive distributorship agreement was wrongly repudiated. (65) As
noted above, a further problem is where a grant of specific performance would
adversely affect the rights of third parties page "1109" such as a purchaser in a
substitute transaction by the defaulting party. Not only may a tribunal be
unwilling to adversely affect that party, but in some cases that party needs to
support the remedy award for it to be effective. Other tribunals might take the
view that the party is entitled to the full range of remedies and it is not the
tribunal's concern whether the award could be enforced as against the third
party.

There are examples of arbitrations involving States where specific performance

has been considered inappropriate. (66) Some tribunals take the view that an
order for specific performance would offend against notions of State sovereignty
and that ultimately, States merely need to be prepared to provide appropriate
compensation. (67) Other tribunals might not impose a blanket rule but might
consider that specific performance would be rejected if it imposed too heavy a
burden on the defaulting party, contrary to Article 35 of the ILC Articles on State
Responsibility. (68) Much may depend on the circumstances as it should be far
easier to gain an award of specific performance against a State where this does
not entail a change to internal regulatory measures. An example might be return
of personal items after a multinational and its personnel have been ejected from
the host State.

Another issue is whether specific performance may be granted in relation to the

arbitration agreement itself. This is not generally a relevant remedy where an
arbitrator is concerned, but may be a vexed question for a national court in the
context of anti-suit injunctions. It is generally not an appropriate direct remedy
for an arbitrator regardless of which party has breached the agreement. If a
claimant fails to honour an arbitration agreement, it simply has not proceeded
with the case appropriately and the case will in due course be dismissed. Where
a respondent fails to take part, relevant lex arbitri and rules direct that the
tribunal proceed to hear the dispute. The tribunal only needs to give the
respondent regular notice but should not seek to compel its involvement. In a
rare case, it might be that a court has ordered damages in lieu of specific
performance for breach of an arbitration agreement and the tribunal may in due
course have to ensure that there was no double counting when its own damages
awards are made. Another relevant scenario where a tribunal itself might
consider breaches of an arbitration agreement is where one party has failed to
provide the required page "1110" advance on costs. Arbitrators disagree as to
whether this can become a claim under the arbitration agreement, but if so,
damages would be the more typical remedy. This is because the norm is to first
have the other party pay the defaulting party's share and then seek
compensation one way or another. Finally, any anti-suit injunction by the
tribunal in the context of litigation commenced in breach of an exclusive
arbitration agreement, operates indirectly as an incentive to specific
performance although it is not an enforceable remedy as such and the party
might abandon the case as of right.

Conversely, where the issue is whether damages would be an adequate remedy,

a relevant factor would be whether these are too uncertain. Calculating damages
for breach of longer term contracts and distribution and licence agreements is
more complicated than for one-off sale of goods transactions. However, damages
might still be more appropriate than specific performance. Global sums may be
appropriate where the heads of claim necessarily overlap and there would be no
objective way to distinguish between categories. An example was Pabalk v.
Norsalor where there was a claim for lost customers and goodwill, clearly
overlapping categories. (69)

A final consideration concerns the procedural approach of arbitral tribunals

when defining the performance obligations of a particular party. The level of
specificity of an order may depend on the corresponding level of detail in the
contract in question. Where an obligations is expansively defined in a contract, it
may be possible for a tribunal to order its specific performance without issue,
however this will depend on the circumstances of the case. In some cases, the
performance obligation may be defined in general terms. For example, in the
Avena case, the International Court of Justice ordered the US to review and
reconsider the conviction and sentencing of certain Mexican nationals ‘by means
of its own choosing’. (70) Alternatively, an arbitral tribunal may direct a party to
offer to the other a number of different options for performance from which to
choose. For example, in a case heard before the ICC, the respondent was ordered
to choose from amongst three specified countries to which the claimant could
return the machines which the respondent claimed. (71)

14.4.3. Satisfaction

The concept of satisfaction is largely concerned with inter-State disputes and

hence does not apply readily in investment cases between a host State and a
foreign page "1111" investor, (72) and certainly not in private commercial
disputes. Article 37 of the ILC Articles indicates:
1. The State responsible for an internationally wrongful act is under an
obligation to give satisfaction for the injury caused by that act insofar as it
cannot be made good by restitution or compensation.
2. Satisfaction may consist in an acknowledgment of the breach, an expression
of regret, a formal apology or another appropriate modality.
3. Satisfaction shall not be out of proportion to the injury and may not take a
form humiliating to the responsible State.

14.4.4. Declarations

Declaratory relief has become a common remedy in international arbitration. (73)

The PCIJ considered that a declaratory judgment ensures ‘recognition of a
situation of law, once and for all and with binding force as between the parties;
so that the legal position thus established cannot again be called in question
insofar as the legal effects ensuing therefrom are concerned’. (74) Article 48(3) of
the English Arbitration Act 1996 states that a tribunal ‘may make a declaration
as to any matter to be determined in the proceedings’. Declaratory relief may be
awarded ‘as a result of the law governing the arbitration proceedings or… an
agreement of the parties’. (75) While declaratory relief is not capable of
enforcement, ‘it is capable of recognition’. (76) There are a number of typical
scenarios. Parties seeking contractual damages may request a declaration of
breach of contract as well. Parties may also limit the scope of the arbitration
agreement to the question of determining their legal position, in which case the
arbitral tribunal may only give declaratory relief. (77) A well-known example of
the provision of declaratory relief occurred in the Aramco arbitration, (78) in
which the parties had expressly agreed that the arbitrators page "1112" should
not award damages to either side, but may make a declaratory award only.
Acknowledging this restriction, the tribunal stated:

There is no objection whatsoever to Parties limiting the scope of the arbitration

agreement to the question of what exactly is their legal position. When the
competence of the arbitrators is limited to such a statement of the law and does
not allow them to impose the execution of an obligation on either of the Parties,
the Arbitration Tribunal can only give declaratory relief. (79)

In some cases such as distribution, franchise and joint-venture agreements,

there might be a disagreement about proper prices or profit share
arrangements. An arbitral decision may specify the prices for a particular period
of time. A declaratory award ‘may be useful for further negotiations or
adaptation of contracts in long-term or relational contracts’. (80) Formal
declarations of an arbitral tribunal may also constitute steps in the tribunal's
process leading to a final award. (81) At the domestic level, such limited
restrictions on the issuing of declaratory awards as do exist stem from the idea
that the resources of the judiciary should be used efficiently and not wasted on
baseless actions. (82) As a consequence, a number of civil law systems require a
claimant to show a legitimate legal interest in bringing an action before
declaratory relief is sought. (83) However, as Schneider points out, such
considerations may not necessarily apply to arbitral tribunals whose services
have been sought and paid for by the parties to the arbitration. (84) He concludes
that declaratory actions ought to be admissible in international arbitration
without the claimant having to show a legal interest such as that required by
some civil codes. (85)

14.4.5. Injunctive Relief

Few arbitral statutes expressly empower a tribunal to award injunctive relief. As

noted an exception is section 48(2) of the Arbitration Act 1996 (UK) which
provides a tribunal with a power to grant similar injunctive relief as an English
court. Nevertheless, such powers are presumed by many to be inherent and
have page "1113" been upheld by a number of courts and tribunals. (86) At times
the power is subsumed into a broad power to grant any relief deemed just and
equitable. (87) Where a power exists, a tribunal is permitted to award injunctive
relief as against a party where it also indirectly impacts adversely on third
parties although this circumstance may be a key factor in leading to a
discretionary decision not to do so. (88)

Injunctive relief is more likely when the breach and damage is ongoing and an
historical assessment will not suffice. Thus, in the Trail Smelter case, where
there was international pollution, the tribunal not only awarded damages but
ordered the producer to refrain from future damage. (89) Where investment
arbitration and public international law is concerned, the tribunal in the
Rainbow Warrior case considered that such an order requires, ‘… two essential
conditions intimately linked, namely that the wrongful act has a continuing
character and that the violated rule is in force at the time at which the order is
issued’. (90) An ICSID tribunal in Enron v. Argentina accepted that it had the
power to provide injunctive relief against stamp taxes alleged to be tantamount
to expropriation. (91) In that case, the tribunal explicitly stated that ‘in addition to
declaratory powers, it has the power to order measures involving performance
or injunction of certain acts’. (92)

Where ongoing relief is concerned, such as that by way of injunction, the

question is again what ongoing supervisory power a tribunal can or should seek
to retain. The issue is whether a tribunal is functus officio in that regard. If
necessary, new proceedings may need to be instigated where that is thought to
be a concern.

page "1114"

14.4.6. Gap Filling and Contract Variation

Redfern and Hunter consider that there are four reasons a tribunal might be
asked to adapt a contract made by the parties: to fill a perceived gap in the
contract; to change the contract to meet changed circumstances; due to a
‘hardship’ situation; or, to redress the equilibrium of the contract. Redfern and
Hunter also discuss the remedy of rectification. (93) The remedy is virtually
unknown in civil law countries. It is allowed under the Arbitration Act 1996
(UK), s 48(5)(c) and under the LCIA Rules, Article 22.1(g). The following
sections expand on these scenarios.

14.4.6.1. Adaptation

The power of the tribunal to adapt a contract may derive from the law
applicable to the substance of the dispute. (94) In some civil codes, the doctrine of
rebus sic stantibus is implied. This doctrine also exists in public international
law. (95) Under the doctrine, the contract is binding ‘so long as things stand as
they are’. In other words, contractual terms may need to be amended to adapt to
a significant change in circumstances. The ICC had formulated special rules on
the adaptation of contracts. (96) Though these allowed for the adaptation of
contractual terms where expressly agreed to by the parties, the ICC suggested
the applicable law may in some cases deny an arbitrator this power. (97) The ICC
comments came in for criticism and are no longer published. (98)

In discussing the difference between filling a gap in a contract and amending it

to meet new circumstances, Redfern and Hunter remark that it is ‘a smaller step
for an arbitral tribunal to imply a power to fill a gap in the agreement than to
imply a page "1115" power to change it’. (99) Most tribunals are reluctant to
change the terms of a contract unless they have an express power to do so under
the arbitration agreement. Red-fern and Hunter explain that ‘the ability to adapt
contracts effectively expands the competence of the arbitral tribunal into the
area of varying or substituting the legal instrument that is the source of the
tribunal's own jurisdiction’. This can also raise questions as to interpretation of
the arbitration agreement. Have the parties employed the arbitrator to change
the contract or merely apply it? Redfern and Hunter conclude that regardless of
the legal basis for adapting a contract, ‘arbitral tribunals have proved very
reluctant to substitute their own views of a fair allocation of contractual risk for
that of the parties at the time the contract was originally concluded’. (100)

14.4.6.2. Hardship

The UNIDROIT Principles incorporate the concept of hardship, under which a

contract may be amended to restore equilibrium in certain circumstances where
‘the occurrence of events fundamentally alters the equilibrium of the contract
either because the cost of a party's performance has increased or because the
value of the performance a party receives has diminished’. (101) Concepts of
hardship are also an expression of rebus sic stantibus. Common law systems do
not recognise hardship per se but would limit consideration of such facts to
questions of frustration and similar excuses for impossibility of performance.
Nevertheless, hardship per se does not meet the normal standards under
common law frustration tests. (102)

While the UNIDROIT Principles are commonly cited as encapsulating lex

mercatoria, its hardship provisions do not reflect a widespread domestic norm
and hence are rarely applied as such, absent agreement of the parties. (103) As
such, page "1116" hardship provisions will typically apply when agreed to by
the parties. The ICC has published a model clause on hardship which parties
could incorporate into their agreements. (104) Some contracts expressly provide
for adjustment or renegotiation in circumstances of hardship, such as Article
34.12 of the Model Exploration and Production Sharing Agreement of Qatar of
1994. In Article 34.12 it was stated that where any future law, decree or
regulation affected the contractor's financial position, both parties would enter
into negotiations in good faith to maintain the economic equilibrium of the
agreement. However, ICC Case No. 7365 considered that hardship should be
regarded as a general principle of law. (105) A contrary view was taken in ICC
Case No. 8873. (106) Article 6.2.2 of the UNIDROIT Principles stipulates the
preconditions for a hardship claim:

(i) The event must occur or become known after the conclusion of the
contract;
(ii) The event could not reasonably have been taken into account at the time
the contract was entered into;
(iii) The event is beyond the control of the party claiming hardship; and
(iv) The risk of the event was not assumed by that party.

Each of these elements will involve questions of fact and hypotheses about likely
attitudes of the parties by an adjudicator.

Article 6.2.3(1) of the UNIDROIT Principles entitles the disadvantaged party to

request re-negotiation in case of hardship. This would not apply if the contract
contains its own adaptation rules. (107) If re-negotiation is not effective, the
parties can resort to an adjudicator who may terminate the contract or adapt it.
(108) Any adaptation ought to be based on an analysis of the initial bargain, of the

variations from the hardship events and of the most efficient way in which to re-
balance the cost/benefit from both parties' perspectives.

14.4.7. Sequestration of Assets

It is suggested that the tribunal has no power to order sequestration of the

defaulting party's assets, (109) although it can certainly order specific
performance over page "1117" promises as to assets. It would not be able to
sequester assets merely as a way to satisfy a damages award.

14.4.8. Committals for Contempt

An arbitral tribunal has no power to order committal for contempt as against

any tribunal orders as against any party. (110) Such court committal powers allow
for gaol terms until the party relents as per the judge's order.

14.5. Damages and Monetary Payments

14.5.1. Introduction

There are a number of reasons why damages are the most significant remedy in
international commercial dispute settlement. First, they are the primary remedy
for breach of contract under the common law where that substantive law
applies. Second, in investment arbitration, monetary compensation would be the
norm where expropriation occurs as by definition the relevant State has sought
to take over the investment. Third, where there is a breakdown in long-term
commercial arrangements, even legal systems that try to give primacy to
specific performance would be reluctant to make orders forcing unwilling
parties to work harmoniously together in the future.

14.5.2. Lex Arbitri and Rules Provisions

Unless the parties agree to the contrary or the claim is for other than monetary
relief, a tribunal naturally has the power to award damages and related
monetary orders. A tribunal can also indicate the time for payment, the mode
and the currency. (111) A tribunal would normally look to the substantive law to
determine how damages are calculated. Civilian systems commonly treat the
calculation of damages as determined by the law of the contract, while English
courts have at times seen it as a hybrid. (112) Investment arbitrations have seen a
greater tendency to purport to resort to general principles of law (113) or to
principles of ‘equity … in conformity with international law standards’. (114)

page "1118"

14.5.3. Party Autonomy and Agreements to Limit Damages Levels

Parties might seek to make certain agreements in relation to damages. There are
a range of means by which parties could modify the otherwise applicable law in
determining damages. At the extreme they could totally proscribe damages.
Parties might agree to exclude consequential damages. Construction contracts
will typically put a cap on damages as do international transport conventions.
This might apply to amount or to the cause, for example excluding loss from
simple negligence. The parties could impose a global cap on any individual claim
or on the totality of claims over a defined period. They could try and obviate a
calculation phase by using a liquidated damages provision as discussed in
section 14.18. This provides for a fixed amount, which might either be higher or
lower than the actual loss. If higher, there is a question as to whether it might be
an improper penalty or might be reduced to a reasonable level. Because the
entitlement to damages is generally seen as a substantive matter, such
agreements will be effective where permitted under the particular substantive
law.

An extreme limitation on damages would be a limitation of liability clause.

Historically, legal systems have varied in their approach to such clauses.
Limitation of liability clauses typically arise in international carriage rules and
treaties. To some, such clauses offend against the fundamental notion of a
bargain where each should have rights as well as responsibilities. To others,
these clauses are simply a matter of risk allocation and would have been
accounted for in the price, at least where parties have equal bargaining power.
Price is likely to vary as and when risk varies as well. Some legal systems apply
particular interpretation norms such as the contra proferentem principle. Under
this approach, adjudicators might read down the subject matter that the clause
relates to or the type of claim or both. In the extreme, they could be held to be
invalid. There is also a question of interpretation and coverage. A typical
question is whether the clauses merely apply to intentional wrongdoing, gross
recklessness or are broad enough to cover mere negligence, if this has not been
stated with clarity.

14.5.4. Procedural and Evidentiary Issues and Damages Assessment

Section 14.2 above looked generally at some procedural and evidentiary issues
with all forms of remedies. The point was made that it would be misleading to
simply see this as an area where a tribunal simply has to find and apply a
suitable substantive law. There are instead a host of procedural and evidentiary
issues that have a major impact on the result and hence on the fairness and
efficiency of the process. This section expands on these issues in the context of
damages assessment, where the challenges are most significant. Assessment of
damages will include a range of issues of law, fact and procedure. These issues
have both theoretical and practical elements. It is too often the case that both
parties put the bulk of their efforts towards respectively proving or disproving
liability. This means that factual and legal preparation for arguments on
remedies are often given less page "1119" attention than is their due. Where this
occurs, damages evidence is not dealt with early enough in the arbitral
proceedings and may be left until near the end of the case.

The first issue is as to the quality and quantity of evidence expected. To some
arbitrators, rigorous best evidence proof is needed for all matters. To others, if a
party is truly deserving and pedantic damages measurement could be too time
consuming and costly, reasonable estimates are more in tune with the
commercial expectations of the parties. The point to be made is not that one
view is necessarily to be preferred, but simply, that the need for even a
subconscious choice demonstrates the importance of discretionary procedural
and evidentiary questions to the essentially substantive question of remedies.
Other discretionary decisions can also have a significant impact on this aspect of
sufficiency of evidence. One example is where the claimant provides less than
ideal evidence of damages for delay in a construction dispute and the
respondent seeks broad-ranging document production on that matter. Does the
tribunal limit the respondent's production rights only to those matters on which
it has the burden of proof? Alternatively, does it leave it to an analysis of
whether the claimant has met appropriate standards of proof? Does the tribunal
warn the claimant that more evidence might be needed? At the time of the
document production request, it is difficult for the tribunal to reject it simply on
the basis that the tribunal would prefer to find against claimant if it fails to
establish a prima facie case. At that stage a tribunal would not wish to bind itself
to such a commitment, which might not be valid once all evidence is presented
at the hearing. Document production has other relevant challenges. For
example, there may be added confidentiality issues in relation to documents
necessary to prove or disprove damages, for example, third-party contracts
where an account of profits is sought. Key documents of this nature may also be
in the hands of third parties where there is a claim for lost profits or recovery of
damages paid to sub-purchasers. (115) There is also the question as to the extent
that a tribunal chooses to analyse the validity of such damages.

More controversial is the tendency of some tribunals to simply award a global

lump sum where it is felt that more detailed or itemised calculation would be
difficult or would lead to problematic overlap. (116) Article 7.4.3 paragraph 3 of
page "1120" the UNIDROIT Principles indicates: ‘Where the amount of damages
cannot be established with a sufficient degree of certainty the assessment is at
the discretion of the court’. Section 10.4.3 looked at issues of standard of proof,
including where damages assessments are concerned. It noted that some
commentators and cases argue that there are justifications for lesser standards
for damages assessment, while others argue that the standards should be the
same but arbitrators commonly fail to meet them when granting awards. One
argument in favour of laxer standards on proof of damages is that at that stage
the adjudicator has determined that the claimant is in fact correct in its liability
argument (or respondent as the case may be). If the tribunal is equally confident
that there is some damage worthy of compensation, but notes that there is too
much difficulty in clarifying the amount, some reasonable assessment is much
closer to a just outcome than a rejection in toto on the basis of the failure to
meet the burden and standard of proof and may even be more cost effective. A
contrary argument is that the same standards of proof should apply to all
elements and respondent has just as much entitlement to protection in that
regard where damages are concerned as is the case with primary liability.

Standards of proof are particularly problematic to consider in many damages

assessments, dealing as they do with hypotheticals about future scenarios. Loss
of profits is particularly problematic given that there are so many variable
factors that could determine whether profit would have ensued and if so to what
degree. Appropriate standards and the tribunal's involvement in calling for
and/or testing evidence also vary where damages are concerned as the focus of
analysis shifts from one party to the other. A liability claim is generally about
claimant making assertions about actions or omissions of the respondent.
Where damages claims are concerned, it is the claimant's (117) own business
affairs, prospects and options that are central. The more a party is arguing about
evidence in its own possession, the more that adverse inferences or strict
application of standards of proof could be applied against it where the evidence
is not forthcoming. The situation may be very different if all of the relevant
records are under the control of the party in breach, for example where there
has been an improper expropriation and the investor has been forced to leave
the host State and leave all records behind. Even in such circumstances there is
an obligation to do whatever is reasonable and necessary to reconstruct the
evidentiary record.

The second key evidentiary issue relates to expert testimony as to damages.

Does the tribunal appoint an expert to evaluate the damages claim and merely
check the logic in that person's report? What degree of challenge is allowed to
that expert if sought by one or both parties? How is a tribunal to resolve
conflicting assessments by party-appointed experts? As with any aspect of
expert evidence, it is important that the tribunal itself makes the ultimate
determination. The more complex and trade-specific the calculations and
analysis, the more important it is for the tribunal to understand the merits of
different methodologies and ensure that appropriate data and assumptions
were used as the basis of the calculations. In many cases, there will be no right
or wrong view as to various assumptions page "1121" such as discount rates,
but the tribunal must still determine what is most appropriate in the
circumstances after being educated by the reports and evidence of the experts.
Such a discussion would seem particularly appropriate for a witness
conferencing approach. The tribunal can then determine which assumptions it
thinks are more likely and invite completion of the calculations accordingly.

There are a number of other strategies that arbitrators may employ to expedite
or rationalise the giving of expert evidence. At the initial stages of damages
discussions, the tribunal, the parties, their counsel and potentially the experts
can examine ways to narrow the damages issues. (118) Such discussions may
enable the parties to reach a consensus with regards to the theories and models
that will be used to determine the valuation of damages. Kantor makes a
number of suggestions for managing the expert report process for valuation
purposes. (119) Chapter 12 outlined various alternatives more generally for
expert witnesses including exchange of reports, pre-hearing meetings and/or
witness conferencing. Kantor suggests that after preliminary exchanges the
arbitral tribunal may then request supplemental valuation reports seeking the
use of more comparable methodologies and assumptions. However, arbitrators
should be very sensitive to the costs involved. (120) As soon as an arbitrator
becomes aware of a situation in which experts are using divergent approaches
to valuation, arbitrators should consider whether to request the experts to
provide valuations on a similar basis. (121)

The CPR Protocol on Determination of Damages in Arbitration provides useful

guidance to arbitrators, counsel and their clients concerning the efficient and
fair development and presentation of damages evidence in arbitration
proceedings. (122) In relation to the issue of experts using different valuation
methods, the CPR Protocol suggests that arbitrators require experts to make
presentations early in the proceedings in a way that permits arbitrators to
understand the methodologies by which the experts reached their conclusions.
Chapter .12 also outlined codes of conduct and possible directions for expert
witnesses. There may also be codes of conduct from within the valuation
industry which may also affect an expert's duties. For example, the AICPA
Professional Ethics Division's Code of Professional Conduct binds AICPA
members and rejects the position that when providing expert witness services,
its members can advocate on behalf of the client. (123) However, as Kantor
acknowledges, regardless of any professional codes of conduct, expert witnesses
in many arbitration proceedings will still page "1122" conduct themselves as
partisan advocates on behalf of the party retaining them. (124) Hence the
directions recommended in Chapter 12 remain desirable on valuation issues as
well.

Another important evidentiary issue is the way experts are selected. This itself
can wholly shape the process. For example, if a valuation method such as
Discounted Cash Flow (DCF) (125) requires a number of elements of subjective
judgment which cannot easily be articulated by experts, then it is difficult for the
tribunal to evaluate conflicting expert testimony. Questions of burden of proof
will be significantly impacted upon if instead of relying on party-appointed
experts as to valuation, the tribunal appoints its own expert. Two completely
contradictory experts on DCF might lead a tribunal to conclude that the claimant
has not met its burden of proof. Conversely, if there is only one tribunal-
appointed expert who identifies a figure, the tribunal is more likely to accept it.
There may also be other issues of contradictory evidence for a tribunal to
evaluate. For example, assertions of a party's expert as to lost profits may
contradict published accounts, notifications to stock exchanges or prospectuses
to investors. A claimant may have undervalued an investment where this is
hoped to reduce a licence fee or property tax, only to find that this undermines
its case at the time of an expropriation. A tribunal may try and determine the
correct position or rely on the inconsistency to attack credit overall.

A third issue relates to efficiency of proceedings and damages assessments.

Choices as to bifurcation may have a significant impact. Heilbron suggests that
bifurcation should not be agreed to automatically. Relevant factors would be
whether the damages claim is likely to be straightforward, whether there will be
overlapping witnesses and other evidence, the nature of expert evidence and the
likely delay. A further caution is to be careful as to whether bifurcation is a
tactical manoeuvre by either party, either a further effort at delay by respondent
or deferral by a claimant with inadequate or unprepared damages claims. (126)
For example, a tribunal might feel compromised if an otherwise deserving
claimant makes a last minute application for bifurcation when it is obvious that
the damages claim is ill-prepared because of counsel's inexperience and could
otherwise be lost on burden of proof grounds. A tribunal would also wish to
consider how to assess damages and other remedies in the most efficient
manner both as to time and cost. For example, if a case is likely to be bifurcated,
this may impact on the extent of pleadings and submissions in the early stages
as it may be wasteful to call for damages arguments if there is to be a separate
hearing in any event. (127) If the proceedings are bifurcated into liability and
quantum phases, the tribunal might consider similar division with regards to the
document production process. One page "1123" concern is that while claimants
and respondents generally bear equal weight for the production of documents
necessary to establish liability, if that phase is completed before a quantification
of damages, it may lead to noticeably different attitudes towards document
production for the quantum phase, (128) where it is generally claimant's
documents that are relevant. There is also an issue in determining the
appropriate market if a market price is used in calculating damages. Once that is
done there will also be evidentiary issues in identifying a market price where
there are a range of prices in relation to a particular commodity.

The next set of procedural and evidentiary issues flow from substantive
elements in the calculation of damages. For example, there will be evidentiary
issues when applying foreseeability tests in terms of consequential loss and
reliance on subjective or objective evidence. How claims are characterised may
also have a significant impact. For example, different outcomes may apply
depending on whether damages are valued at the date of judgment, inclusive of
compensation for delay or instead at the date of breach with a separate award of
interest for late payment. Are damages measured on the date of the award or at
the time of the breach? From a policy perspective, the standards in each case
should lead to the same or similar result, although this may be varied by
national laws or express agreements of the parties as to applicable legal
provisions. Other questions may include whether the particular legal system
allows for liquidated damages as a matter of law; if it does so, does it have an
exception for provisions that are more truly characterised as penalty provisions;
if so, on the facts how is a determination made as to whether it is an a priori
assessment of likely damages or an excessive penalty; if there is little evidence
from either party, what attitude should a tribunal take?

Another typical scenario noted above is where the purchaser of defective goods
claims compensation for damages owed to a sub-purchaser. There are then
evidentiary questions of proof of the validity of those losses and whether they
arise by way of negotiated agreement or by way of some adjudicatory forum. In
the latter event, will the tribunal re-examine those findings as part of the
assessment? An added complication is if the claim from a sub-purchaser was
through arbitration, subject to confidentiality rights. There is also an overlap
with issues of mitigation, given that the claimant might have obviated such
losses if it had responded to the breach in differing ways. The ability to mitigate
via substitute purchases or sales may involve an evidentiary analysis of likely
circumstances in a particular market from time to time. Chapter 10 examined
general questions of burden and standard of proof. While the burden is on the
party seeking to establish a damages entitlement, the burden is also on the
respondent to show that there was a failure to mitigate.

Fluctuations in value are particularly difficult to deal with fairly. For example,
there may be problems where the only damage was short-term decline in the
market value of an asset still held by the aggrieved party. There may also be
page "1124" questions of whether changes in monetary value should be applied
in assessing damages. (129) Complex valuation issues will also arise where the
question is damage to goodwill or business or personal reputation, breach of a
joint-venture or merger agreement or misrepresentation in pre-contractual
negotiations. Here there will first be a question as to whether the relevant
applicable law allows for claims of that nature and second by what methodology
is the claim to be proven. There would also be an intermediate need to prove
that there is such reputation or goodwill and that the breach has caused damage
to it. In the business context, it is hard to demonstrate such damage without at
the same time being able to show loss of revenue or customers. The scenario is
more likely to be relevant where there is a growing business and the damage to
reputation prevented the business reaching the desired level. Diminution in
value of reputation could be measured by lost income, variation in market value
or the costs required to repair the harm. There is then a need to discount
damages on account of future contingencies, and a question as to the degree to
which hindsight should be used in making such calculations. (130)

The overriding point is that so much of the remedy assessment stage is bound
up in key procedural and evidentiary determinations and discretionary
decisions by tribunal. Without sufficient attention to these factors,
determinations are likely to be suboptimal. This applies from the outset through
to final deliberations. The proactive arbitrator should be alert to these issues
from the outset. Where deliberations are concerned, because of the complexities
of damages assessment, a multi-person tribunal should ensure there is some
appropriate time and methodology of deliberating on these issues.

14.6. General Principles of the Law of Damages

This chapter does not aim to outline all of the substantive elements of damage
assessments in various potentially applicable legal systems. Nevertheless, some
broad observations are appropriate, in particular as they would relate to
procedural and evidentiary questions.

There are important differences between legal systems as to the principles by

which damages for economic loss are circumscribed. Legal systems seek to
circumscribe true damages primarily because if there are no reasonable limits
on consequential damages, this would be a great disincentive to engaging in
commerce. English and French law look to questions of foreseeability. To some,
fore-seeability tests have significant economic logic. This is because the
foreseeable consequences are those which the parties would have contemplated
when page "1125" negotiating price and other conditions. The foreseeability test
has been adopted in Article 74 of the CISC Article 7.4.4 of the UNIDROIT
Principles also requires foreseeability of damage. Foreseeability may have both
subjective and objective elements, looking to what reasonable people ought to
have contemplated at the time and also considering what actually was discussed
and contemplated by the parties concerned.

Not all domestic systems follow this approach. German law has been concerned
with questions of causation. Some civilian systems include notions of ‘certainty’
although those that do, do not employ the concept in a literal fashion. (131) The
UNIDROIT Principles in Article 7.4.3 refer to harm established with a reasonable
degree of certainty where certainty relates to the extent of the harm as well as
to its existence. (132)

Terms such as ‘certain’ or ‘uncertain’ do not readily differentiate between

acceptable and non-acceptable damages, given that these terms must be
considered in the context of contested facts and accepted standards of proof.
These are essentially factual questions. (133) The UNIDROIT Principles also deal
with a standard of proof issue by indicating that the court has the discretion to
make the assessment. (134) Default on the part of the wrongdoer may also be
relevant depending on whether the breach is by way of simple or gross neglect,
recklessness or fraud. (135) Foreseeability tests may not apply where there is
some fraud involved. Again categorising behaviour in this way is essentially a
factual question.

Where remoteness and foreseeability are concerned, distinctions are at times

made between direct and indirect losses, the latter being subject to more strict
limitations. Legal systems tend to distinguish between direct consequential
damages which are naturally recoverable, and non-recoverable categories that
may be described as ‘too remote’, ‘unforeseeable’ or ‘speculative’ and an in-
between category of indirect damages that are more controversial. (136) It is the
margin between this category and the rejected category that is problematic.

Contracts and treaties will often seek to limit recoverability of indirect or

consequential losses. A buyer of faulty goods may suffer a loss of profits claim
from an ultimate purchaser. Defective machinery may lead to lost output and
page "1126" further customer complaints and loss of market share. It is virtually
impossible for statutory provisions or case law to set up bright-line tests
between acceptable and non-acceptable degrees of connection. The tests include
questions of degree about which adjudicators may differ in approach. The tests
also rely on factual assessment of hypotheticals such as foreseeability, which in
turn allows much scope for differing approaches by adjudicators. It is suggested
that tribunals are likely to take a more expansive approach unless constrained
by the applicable law. (137)

The conceptual issues are intertwined with practical issues in the context of
burden and standard of proof. At times, an adjudicator may argue that a
particular form of loss is too remote or speculative and is rejected as a matter of
principle. The CPR Protocol, for example, states that ‘where assessing damages
would require speculation, they should not be awarded’. (138) In other cases, an
adjudicator may simply say that the claimant has failed to establish to the
requisite degree that a particular loss was more likely than not to arise or that a
particular future profit has in fact been obviated by the breach.

There may also be complex questions of causation, remoteness and

foreseeability where there is concurrent liability in contract and tort.
Contributory negligence in a tort claim can also have analogies in contract
disputes where it is arguable that part or all of the damage was in fact caused by
the behaviour of the claimant. The idea of contributory actions is included in
Article 80 of the CISC UNIDROIT Principles Article 7.4.7 also calls for non-
attribution where harm is partly caused by the act or omission of the aggrieved
party. There is also the issue of mitigation, discussed in section 14.17. Legal
systems will generally require that the breach be a proximate cause of the loss,
hence turning attention in some cases to whether there was an intervening
event which breaks the chain of causation. (139) Legal systems grapple with
notions of proximity and how to determine what is truly a break in the chain of
causation in these circumstances when life has a myriad of related causal events.
(140) Situations may vary from cases where an external factor truly caused part

or all of the damages which should not be attributable, and instead, external
factors that must have been in the parties' reasonable contemplation and which
made the breach naturally lead to the damage that ensued.

page "1127"

Where groups of companies are concerned, there may also be a need to ensure
that the loss is actually that of the claimant and not associated entities. (141) A
tribunal may also need to consider whether there is joint and several liability in
the case of multiple respondents, (142) or whether damages should be allocated
pro rata.

14.7. Measurement of Damages

14.7.1. The Compensatory Principle as the Basis for Assessment

The basic principle in common law and civil law systems is that an award of
damages should ‘as nearly as possible get at that sum of money which will put
the party who has been injured, or who has suffered, in the same position as he
would have been in if he had not sustained the wrong for which he is now
getting his compensation’. (143)

14.7.2. Elements of Damages Awards

Damages will generally protect a claimant's positive expectation (loss of bargain

or of the expected performance) (144) and reliance, or out-of-pocket expense (145)
interests. There are thus two forms of pecuniary compensation, one relating to
damnum emergens, meaning the damage suffered by the aggrieved party and the
second being lucrum cessans, meaning loss of profit. Depending on the
circumstances, damages might be awarded for loss of value in defective goods or
services, loss of value in property, business and investments, loss of profits, and
delay.

While the two broad forms of damages are accepted, the problems of proof vary
significantly. There is a difference between assessing damage based on proof of
the concrete losses of the aggrieved party and instead by an abstract measure
usually based on identifying market value. From an evidentiary point of view, a
tribunal is assessing historical events up until the hearing but also projecting
into page "1128" the future. (146) Where loss of profits or diminution of the value
of a business are concerned, a tribunal will need evidence as to past activities,
future plans and past and future market conditions. Where the future is
concerned, there will need to be an assessment of various contingencies and
appropriate discount factors for likely eventualities.

Some breaches of contractual commitments will give rise to great uncertainty as

to the appropriate level of damages. The CPR Protocol provides as an example
contracts for the acquisition of a business which produce claims by the buyer
based on breaches of warranties in which damages are sought for the difference
between what was expected in terms of future earnings and what was obtained.
(147) A further example is acquisition agreements with ‘earn-out’ provisions.

Claims may arise in which the seller alleges that the buyer has not operated the
business in the contractually specified manner during the earn-out period,
giving rise to a damages claim for the difference between what was actually
earned and what should have been earned. In such cases, arbitrators are
confronted with the difficult task of determining what might or should have
happened but did not. Accordingly, compensation is by nature imprecise. It
cannot undo the harm done or perform the unperformed contract. It is premised
on the simple rationale that ‘the established wrong and the material imbalance
it generates between the parties cannot be factually undone, so that the award
satisfactorily re-adjusts that tangible imbalance such as it can’. (148) There may be
problems of calculation within each category of damages and problems of
overlap between them. In Pabalk v. Norsolor, an ICC tribunal considered that a
global lump sum should be identified in such circumstances. (149) In some cases, a
contract will provide a very detailed formula for calculation of loss or damages.

14.7.3. Unjust Enrichment: Quantum Meruit and Quantum Valebat

The concept of unjust enrichment exists in one form or another in almost every
legal system. (150) It becomes relevant where the application of other rules would
lead to an unjust result, and is based on notions of justice and equity. page
"1129" The conditions that must be present for the principle of unjust
enrichment to apply are well-summarised by the Iran-US Claims Tribunal, which
stated in its Sea-Land award:

There must have been an enrichment of one party to the detriment of the other,
and both must arise as a consequence of the same act or event. There must be no
justification for enrichment, and no contractual or other remedy available to the
injured party whereby he might seek compensation from the party enriched.
(151)

Where an action in unjust enrichment is for the return of money paid, the action
is one for money had and received. If the performance of the contract is in the
nature of services, the claim is in quantum meruit. If the title to property has
been transferred under an invalid contract, the action for its recovery is one in
quantum valebat. (152)

14.7.4. Net Benefits

If a claimant has been benefited in some way by the breach, that would need to
be taken into account in calculating damages. (153) This might arise either
through obtaining some concrete benefit or through being relieved of the
obligation of future outgoings. In some cases this is discussed under the notion
of mitigation, although that seems undesirable in the context of direct net
damages and the claimant's general burden of proof in that regard. Mitigation is
discussed further in section 14.17.

Comparing the situation arising on breach with what would have happened if
the contract was properly performed will at times involve a cost-benefit
analysis. The claimant will no doubt highlight the extra costs and losses but in
some cases there may even have been savings of expenses that would have
arisen if the contract had been validly performed, for example hypothetical
storage costs in relation to goods that have not been delivered. A tribunal will
have to consider how to approach such issues if they seem obvious but are not
raised by the respondent, once again as they can be viewed as part of the
claimant's burden.

page "1130"
14.8. Valuation

14.8.1. Introduction

An integral part of calculating any damages award will be a valuation of the

harm that has been inflicted upon a claimant. Ultimately, if a tribunal
determines that damages have been incurred, it should award them, even if they
are difficult to establish with precision. (154) The biggest challenges are with
expectation loss. A tribunal must consider what it is trying to assess, namely the
concept of value and the applicable measure, and also the evidence by which it is
to be determined. In complex matters, there will typically be conflicting expert
evidence as to valuation. Since many lawyers and arbitrators are not themselves
experts in accounting or financial analysis, it can be difficult for arbitrators to
understand how they may use and interpret expert evidence in determining
damages. (155) As one leading arbitrator and scholar recently stated, an arbitrator
must arrive at ‘an independent determination of value, regardless of whether
the competing experts have provided widely divergent estimates of value’. (156)
The challenge for the arbitrator then, is to reach a reliable valuation that has ‘at
best the virtues of a good faith attempt at estimation’. (157) The tribunal must
come to an informed decision and not a mere compromise.

14.8.2. The Approaches to Valuation

A starting point for this process of valuation is identified by Dr Shannon Pratt in

his treatise on valuation: (158)

An intuitively appealing method of concluding the value estimate is for the

analyst: (1) to use subjective but informed judgment and decide on a percentage
weight to assign to the indications of each meaningful valuation approach or
method and (2) to base the final value estimate on a weighted average of the
indications of the various methods. (159)

It is suggested that arbitrators take a practical approach to valuation questions

that incorporates all the relevant facts as well as elements of common sense and
page "1131" reason. (160) This process should enable the arbitrator to weigh the
relevant facts and determine their aggregate significance. Parlade briefly
mentions ‘Quodient Awards’ (161) saying they are not permitted unless the
parties agree to be bound. They are awards where ‘in the absence of agreement
on the award, the arbitral tribunal awards an amount which is arrived at by
aggregating the sums decided by each member to be due and dividing the sum
by the number of arbitrators’. Where experts offer differing models for the
calculation of damages, tribunals may request a reconciliation showing the key
differences between the models. Joint reports of party-appointed experts is
discussed in section 12.14.7.

Complex valuations may need to be made in such things as a breach of a joint-

venture or merger agreement or misrepresentation in pre-contractual
negotiations. Calculating damages for breach of longer-term contracts and
distribution and licence agreements are more complicated than one-off sale of
goods transactions. Global sums may be appropriate where the heads of claim
necessarily overlap and there would be no objective way to distinguish between
categories. An example was Pabalk v. Norsalor where there was a claim for lost
customers and goodwill, clearly overlapping categories. (162)

14.8.3. Three Approaches to Valuation

In many investment disputes, a central issue will be whether and how the
conduct alleged to have injured the claimant has affected the ongoing prospects
for the claimant's business. (163) In answering this question, arbitrators must
consider future events such as lost revenue, the additional capital required to
maintain operations and the reliability of such estimates. The International
Valuation Standards Committee (IVSC) has published the Valuation Standards to
be used in financial statements and to promote the worldwide observance of
valuation approaches. (164) Organisations in over fifty countries subscribe to the
IVSC Principles. The three approaches suggested by the IVSC are ‘The Income
Based-Approach’; ‘The Market-Based Approach’; and ‘The Asset-Based
Approach’. (165) It has been page "1132" suggested that these approaches have
been accepted by the valuation community to create a ‘Valuation Mercatoria’.
(166)

14.8.3.1. The Income-Based Approach

5.14.2.1 (IVSC): The income capitalisation approach estimates the value of a

business, business ownership interest or security by calculating the value of
anticipated benefits. The two most common approaches are capitalisation of
income and discounted cash flow or dividends method.

14.8.3.2. The Market-Based Approach

5.14.1.1 (IVSC): The market approach compares the subject to similar

businesses, business ownership interests, and securities that been sold in the
market'.

5.14.1.2 (IVSC): The three most common sources of data used in the market
approach are public stock markets in which ownership interests of similar
businesses are traded, the acquisition market in which entire businesses are
bought and sold, and prior transactions in the ownership of the subject
business'.

5.14.1.3 (IVSC): There must be a reasonable basis for comparison with and
reliance upon the similar businesses in the market approach….

14.8.3.3. The Asset-Based Approach

5.14.3.1 (IVSC): In business valuation the asset-based approach may be similar

to the cost approach used by Valuers of different types of assets.
5.14.3.2 (IVSC): The asset-based approach is founded on the principle of
substitution, i.e., an asset is worth no more that it would cost to replace all of its
constituent parts.

5.14.3.3 (IVSC): In the execution of the asset-based approach, the cost basis
balance sheet is replaced with a balance sheet that reports all assets, tangible
and intangible, and all liabilities at Market Value or some other appropriate
current value.

The Féderation des Experts Comptables Européens, the representative

organisation for the accountancy profession in Europe, adopts the approach that
the valuation of a business for purely financial objectives is based on the present
value of net cash flows from the business to the owner. It is a value based on the
profits earned by the business which the business will continue to accrue into
the future. (167) It might be thought that the Income-Based and Market-Based
approaches will page "1133" generally offer the best measure of the value of a
business, however particular facts and circumstances and problems of proof
may necessitate the use of other valuation approaches.

In reality, all three Approaches take into account future earnings capacity. The
Income-Based Approach does this explicitly by reference to a company's
earnings capacity, while the Asset-Based and Market-Based approaches may
implicitly take account of future earnings through the incorporation of market
values. (168) While at first sight the Asset-Based approach may not appear to do
so, it typically requires consideration of the market value of tangible and
intangible assets. Tangible, fixed assets will usually depreciate over time;
however intangible assets like goodwill and intellectual property can and do
appreciate and depreciate over time. Thus to calculate the value of all of a
business' assets, the value of intangibles may need to be adjusted to reflect
market values, thereby incorporating future earnings capacity into the valuation
process. (169) Hence it is not fully distinct from the other methods. Kantor
explains that as a result of the reliance by valuers on the market's perception of
future earnings potential, the Income-Based and Market-Based Approaches
converge towards a single fundamental measure, namely, earnings. If the
different valuation methods are correctly applied, each could be expected to
generally produce results consistent with the others. (170) It is thus important to
remember that valuation methods are not mutually exclusive. As the US
Supreme Court said in CSX Transport, Inc. v. Georgia State Bd. Of Equalization,
(171) valuation is not ‘a matter of mathematics…Rather, the calculation of true

market value is an applied science, even a craft. Most appraisers estimate

market value by employing not one methodology but a combination. These
various methods generate a range of possible market values which the appraiser
uses to derive what he considers to be an accurate estimate of the market, based
on careful scrutiny of all the data available’. (172)

In the case of Compañía de Aguas del Acnoquija S.A. & Others (CAA) v. Argentine
Republic, an ICSID Tribunal did not accept the valuation methods proposed by
the parties and instead used a method of its own choosing. (173) In that case,
claimants argued that the fair market value should be based on the profits that
CAA would have obtained had the concession in question not been undermined.
(174) The Tribunal declined to adopt this approach, finding that the record of the

concession's operations failed to demonstrate with sufficient certainty that it

would have been profitable. Ultimately the Tribunal adopted an ‘amounts
invested’ approach as the measure of the fair market value which on appeal, the
ad hoc page "1134" Committee held to be ‘well within the margin of
appreciation of the Tribunal’. (175) Such an approach will be similarly appealing
in joint-venture transactions that break down at an early stage, where one party
invested capital and the other invested knowhow. In each case, damages for loss
of the use of the money or interest in lieu thereof could seem appropriate.

Given the range of estimates that are likely to be put forward by party-
appointed experts, it is necessary for arbitrators to have a practical approach in
mind when coming to a final valuation. There will also be a need to consider a
range of possible estimates of future earnings as conflicting views can be based
on key variables about which there can be legitimate debate. For example, the
value of a mining concession will depend heavily on future exchange rate
projections and general economic conditions. In such circumstances, the
adjudicator may wish to consider alternative scenarios and pick the best one or
attempt some form of averaging. (176) It may also be necessary for arbitrators to
direct parties and expert witnesses to maintain an accurate and detailed audit
trail so that the records and sources of information upon which a valuation is
based, and the ties between the valuation and the evidence presented in the
case, can be easily identified. (177)

14.8.4. Actual Transaction Cost

Income-based methods like Discounted Cash Flow valuations are often

contrasted with market-based methods that rely on stock prices or transactions.
However, arbitrators are not required to use the actual or potential sale price
for a company as the best indication of value. Actual transaction cost refers to
the price paid for a part of or whole company in a recent arm's length
transaction. If the tribunal is of the opinion that the price paid for a company or
an asset is ‘unreasonably high’ (178) then it may use the price that it believes a
reasonable person would have paid for the company or asset as a foundation for
the calculation of damages. The logic would be that the defaulting party only
caused damage to the reasonable component. The balance was wasted money in
any event. Similarly, where the price negotiated for a company or asset is
privately negotiated and the market in which the transaction occurred is limited
or uncompetitive, a tribunal may choose not to use actual or potential
transaction prices to measure the value of an investment. (179)

page "1135"

14.8.5. Fair Market Value

Each of the valuation methods considers the market value of a business, asset or
investment. Courts and commercial codes have offered different definitions of
fair market value. It is useful to set out some of these definitions in full. The
ICSID arbitrators in CMS v. Argentina adopted the definition of fair market value
offered by The American Society of Appraisers, a major international
organisation of professional appraisers:

The price, expressed in cash equivalents, at which the property would change
hands between a hypothetical willing and able buyer and a hypothetical willing
and able seller, acting at arms length in an open and unrestricted market, when
neither is under compulsion to buy or sell and when both have reasonable
knowledge of the relevant facts. (180)

The World Bank Guidelines define fair market value as:

an amount that a willing buyer would normally pay to a willing seller after
taking into account the nature of the investment, the circumstances in which it
would operate in the future and its specific characteristics, including the period
in which it has been in existence, the proportion of tangible assets in the total
investment and other relevant factors pertinent to the specific circumstances of
each case. (181)

Ultimately, the best evidence of the fair market value of a business may be the
price agreed to by a willing buyer and a willing seller, each with knowledge of
the relevant facts in a recent arms' length transaction. (182) The ‘fair market
value’ compensation principle contains an implicit assumption about the
treatment of ‘rare events’ whether having positive or negative impact on value.
Kantor explains that future ‘rare events’ are ‘incorporated into a market
valuation only insofar as the market price at the valuation date (rightly or
wrongly) considers the prospects for the future rare event’. (183) Thus, in
assessing market value, a tribunal is not required to shut its eyes to events
subsequent to the date of injury if these shed page "1136" light in more concrete
terms on the value applicable at the date of injury. (184) As noted by the tribunal
in Metalclad, ‘the fair market value of a going concern which has a history of
profitable operation may be based on an estimate of future profits subject to a
discounted cashflow analysis …’. (185) The tribunal thought this method would
only be appropriate to determine market value where there had been a
sufficient period of time and actual experience of profits. However, blanket rules
about the time that a business has been active may be misleading. In some
businesses it would be very easy to project likely future profit. An example
would be a new business with a concrete long-term supply agreement with a
high quality sole customer with fixed prices or price formulae.

In Enron v. Argentina, the ICSID Tribunal considered the existence of actual

willing sellers and buyers to be ‘meaningful’ in relation to the Tribunal's
findings. Transactions between these buyers and sellers were held by the
tribunal to accurately reflect the current market value of the company. (186) The
tribunal stated that whilst the use of the Discounted Cash Flow Method (a form
of Income-Based valuation) has been applied by other tribunals, the ‘real value’
obtained in transactions involving the company ‘better reflects the current value
of such participation’. (187) In that case, Enron Corporation and Ponderosa Assets,
through a series of complex transactions, made a number of investments in the
Argentinean gas transportation company Trasportadora de Gas del Sur (TGS).
The tribunal found that it was able to assess the current value of TGS and the
current value of Enron's investment in TGS by reference to the actual sale of
Enron's 15.2% stake in TGS to a third party.

In the case of Ioannis Kardassopolous & others v. The Republic of Georgia, the
Tribunal found it appropriate to rely on the unusual circumstance where it had
evidence of three comparable transactions to arrive at the fair market value of
the claimant's percentage interest in the business. (188)

In many cases, there is no way to establish market value, because there is no

established market for sales of a particular interest. This occurs regularly with
respect to equity interests. As Dobbs has said:

In many instances no appropriate market can be found for the property or

entitlement in question. This occurs for many reasons, ranging from the fact that
no one is buying or selling property of the kind involved, to the fact that a
monopoly may control prices. There would almost never be a ‘market’ in real
page "1137" property, only sales of more or less comparable properties.
Businesses, opportunities and trade secrets might theoretically be bought and
sold, but unique advantages and risks again make standard market prices
unlikely. So in many cases, market value, one of the most common bases for
damages decisions, is not an existing fact but a legal construct or even a
convention. (189)

If an actual market does not exist, arbitrators and the parties will search for
substitutes. This process may involve consideration of ‘comparable sales,
income produced by the property, replacement costs, cost of obtaining a
functional substitute, risks of loss and chances of gain’ (190) as well as other data
for constructing this substitute market. (191)

Other valuation methods include the net value of individual assets, the cost of
the assets, the amount required to replace the assets injured and the sale value,
either as a whole or under a liquidation process. (192) Each has flaws and
assumptions. Looking at cost and replacement ignores important intangibles.
Methods that look for net balance sheet value are dependent on accounting
standards, depreciation allowances and the like and can see a timing mismatch
between accounting valuations and adjudicatory valuations. (193) Tribunals
should encourage experts to identify and disclose assumptions underpinning
their presentations in order to allow for quicker and easier comparison of
positions taken by opposing experts. (194) This point is discussed further in
section 12.14.3 While there are different methods of valuing assets, in a
properly informed market, all must be relevant to calculations both because of
their inherent relationship and because an analysis of alternatives should help a
tribunal's confidence level in complex matters. As to the first, while one can
speak of the value of an asset being the price available in an appropriate market,
that price will be a reflection on the earning potential of the asset itself.

One issue is whether a claimant is entitled to choose whichever method gives it

the greatest return. One view is that the adjudicator must apply the most fair
and reasonable measure. For example, if a claimant invested USD 100,000,000
in building a hotel which was then expropriated, where evidence shows that the
building was grossly over-priced and would only be worth USD 70,000,000 even
without expropriation, a respondent might assert that it is not reasonable to
award claimant full compensation for its inflated expenditure. Conversely, page
"1138" Paulsson seems to argue to the contrary when he states ‘the wasted cost
is what the claimant has spent in reliance on the agreement, without reference
to how judicious or providential those expenditures turned out to be. No further
explanation is necessary to understand why victims of contractual breaches
tend first and foremost to articulate a plea for damnum emergens’. (195)

14.8.6. Discounted Cash Flow

The Discounted Cash Flow Method (DCFM) referred to in the Enron case above
by the tribunal is explained by Weisburg and Ryan:

The discounted cash flow or ‘DCF’ is a tool that measures the value of a business
by projecting the net cash flow for a fixed period of time into the future and then
discounting it back to present value as of the date of injury. The discount rate
should reflect the time value of money in the host country and the relative risk
associated with the particular investment. (196)

The DCFM recognises that the true value of a going concern is the profits it will
generate over its operative life in present value terms. It requires arbitrators to
look forward to project a company's future performance by extrapolating from a
company's current financial position. (197) Forward looking methods of
calculation, such as the DCFM, require an adequate past record available to build
on for an acceptable future profit forecast. (198)

It is important to understand that methods such as DCF are built on a range of

highly speculative elements. While these may be the best way of assessing value
in a market environment, the application of such a principle must be looked at in
the context of the requisite burden and standard of proof in an adjudicatory
forum. Because the method is built on future predictions as to issues such as
currency, asset and commodity prices, inflation, interest and political and
commercial risk, results can vary greatly. Other criteria under DCF methods
include analysis of general market trends and likely behaviour of existing and
prospective members of the relevant industry. With large conglomerates that
move in and out of different page "1139" activities depending on how best to
utilise their capital, this can be a particularly uncertain exercise. Discount rates
need to factor in inflation, market volatility and lending rates. (199) The discount
rate should reflect the time value of money in the host country and the relative
risks associated with a particular investment. (200) Evaluating host country
factors are particularly problematic in non-market economies, (201) or those with
volatile political landscapes, oscillating between pro- and anti-business forces.

If the individual elements are too uncertain to give a tribunal even confidence on
balance, then the ultimate findings of the method may be similarly problematic.
Applying the DCF method might also call for consideration of a vast amount of
information not otherwise relevant to the determination of liability. The Iran-US
Claims Tribunal questioned the utility of the method in Amoco. (202) If DCF is too
uncertain because of a start-up operation, then a more appropriate assessment
may be one of loss of a chance, discussed in section 14.10.2, although DCF may
be the means to assess the chance. (203)

While the burden of proof is on claimant, in complex calculations such as DCF, a

tribunal may consider various elements and assumptions, accept those where
the burden is satisfied and err in favour of the defaulting party in other
circumstances. (204) Where discounted cash flow analysis will occur, while the
tribunal will typically analyse the logic and methodology of conflicting experts, it
would rarely wish to make its own calculations but would instead ‘determine
and identify the extent to which it agrees or disagrees with the estimates of both
parties and their experts concerning all these elements of valuation…’. (205)

page "1140"

14.9. Timing and Assessment Dates

14.9.1. Timing and Damages Assessments

As the above cases suggest, the date on which to measure the impact of the harm
suffered will have a significant effect on the ultimate calculation of
compensation. It is, therefore, desirable that the arbitrator's decision about the
applicable valuation date is made and communicated as a preliminary step in
proceedings. The possibility of settlement should be enhanced where there is no
dispute remaining about the applicable valuation date. Situations such as the
one that arose in Santa Elena v. Cost Rica (206) will then be avoided. In that ICSID
case, the claimant presented a 1993 property appraisal and a fair market
discounted cash flow analysis based on a December 1997 valuation date, only to
have the panel rule that the operative valuation date was May 5,1978. (207) If a
particular date cannot be determined at the outset, the tribunal might still
usefully limit the number of potential dates. In this way arbitrators can focus the
parties on the impact of different valuation dates. (208) Thus, it will enable the
parties' own experts to give evidence relevant to all prospective valuation dates
rather than focussing exclusively on the one contended for by the party calling
them.

An adjudicator should also be aware that the timing of payment of damages

under an award will rarely reflect the way moneys would have flowed but for
the breach. Hence tribunals will need to take this into account in making the
calculations. The normal situation is that damages come later than would have
been the norm, hence the entitlement to interest or damages in lieu of interest.
In some cases, however, the reverse is the case, with adjudicatory compensation
for loss of profits dealing with the future profit stream that will not arise
because of the breach. In these circumstances, the tribunal will need to apply a
discount rate to take account of the present value of that future stream. Leaving
aside taxation matters, the typical formula for identifying present value of a
future income stream at simple interest is P = A/(1 + nr) where P = present
value, A = amount due, n = number of years till due and r = rate of interest in
decimal form. (209) Present value with annual compound interest is P = A/(1 +
r)n.

14.9.2. The Date of Assessment of Value

The predominant rule across legal systems is that damages for breach of
contract are assessed as at the date of breach, (210) or restated, the calculation of
compensation page "1141" values the impact of the injury at the date of harm,
(211) although this is not a uniform position, some systems requiring notice of

claim. This will be determined by the applicable law.

However, in order to accurately assess damages, and thus satisfy the

compensatory principle, facts that occur subsequently to the breach or injury
should not be ignored. (212) For example, in the Golden Victory, a contract
between a ship-owner and a charterer contained a ‘war clause’ giving each party
the right to cancel the contract if war broke out. The respondent charterers
wrongfully repudiated the contract three years into its seven year term. A war
subsequently broke out fourteen months after the repudiation. The majority in
the English House of Lords held that damages were to be measured by taking
into account that the owners would only have had the benefit of the charter until
the outbreak of the war. (213) Furthermore, steps taken by an injured party to
mitigate loss may be taken into account in calculating compensation. (214) In
Sinclair Refining, the US Supreme Court held that post-injury events may at
times appropriately be considered when calculating damages:

The law will make the best appraisal that it can, summoning to its service
whatever aids it can…. We find no rule of law that sets a clasp upon its pages,
and forbids us to look within…. To correct uncertain prophecies is not to charge
the offender with elements of value nonexisting at the time of his offense. It is to
bring out and expose to light the elements of value that were there from the
beginning. (215)

Further, in CME Stockholm, the Tribunal adopted claimant's argument that the
investment's actual subsequent performance was relevant to test the
reasonableness of the company's earlier pre-injury forecasts: (216)

Settled law makes clear that the Tribunal's valuation of the Claimant's
investment in CNTS based on the company's market value should be
accomplished by reference to conditions at the time of loss, since a willing buyer
would not have known what future events would bring in negotiating a
purchase with a willing seller at that time…. Nevertheless, comparison of CME's
projections against actual results for the period since the forecasts were
prepared reinforces the reasonableness of using those forecasts in valuing
CNTS.

page "1142"
International expropriation law has varied in its approach to these principles, in
particular when dealing with valuation of expropriated property. (217) In Santa
Elena v. Costa Rica, an ICSID Tribunal rejected the claimant's argument that the
value of the expropriated property should be assessed as of the date of
judgment and instead held that compensation would be calculated as of the date
the expropriation was effected. The tribunal in that case held that ‘factors that
arose thereafter… must be disregarded’. (218) However, two recent ICSID tribunal
decisions have derogated from the ‘settled law’. In both ADC v. Hungary and
Siemens v. Argentina, it was concluded that the Chorzów Factory (219) reparations
standard required compensation for the unlawful conduct to be computed as at
the date of the award rather than the expropriation date. The decisions were
made on the basis that in order to put the Claimants in the same as position as if
the expropriation had not occurred, it was necessary to capture the subsequent
increase in value in calculating the award. (220) Finally, in the case of Amco Asia v.
Indonesia (Amco II), the ICSID tribunal reasoned:

It may, on one view, be the case that in a lawful taking, Amco would have been
entitled to the fair market value of the contract at the moment of dispossession.
In making such a valuation, a tribunal in 1990 would necessarily exclude factors
subsequent to 1990. But if Amco is to be placed as if the contract had remained
in effect, then subsequent known factors bearing on that performance are to be
reflected in the valuation technique.

14.10. Lost Profits

Problematic areas in relation to damages include whether lost profits are

generally permissible and particularly so when identifying adequate or full
compensation in expropriation cases. Article 7.4.2 of the UNIDROIT Principles
allows for full compensation which would encompass lost profit. That can be
impacted upon by whether one identifies part of the asset as goodwill, which
itself reflects profit potential to some degree. There must not be double
accounting. A further complication where lost profits is concerned is that it has
even been argued that loss of profit stream can itself be expropriation under
investor protection norms, a view taken by the dissenting arbitrator in En Cana
v. Equador. (221)

page "1143"

In all cases the claimant will need to have sufficient evidence that lost profits
resulted from the breach and that profits would have in fact on balance arisen.
In Vivendi v. Argentina, the Tribunal stated that ‘compensation for lost profits is
generally awarded only where future profitability can be established (the fact of
profitability as opposed to the amount) with some level of certainty. (222)
However, the aggrieved party cannot be expected to have clear evidence as to
the amount of the loss. (223) The fact that the exercise is inherently uncertain is
not a reason for the tribunal to decline to award damages for loss of profits. (224)
This point was made by the sole arbitrator in Sapphire International Petroleums
Ltd v. National Iranian Oil Co (225) who stated:
It is not necessary to prove the exact damage suffered in order to award
damages. On the contrary, when such proof is impossible, particularly as a result
of the behaviour of the author of the damage, it is enough for the judge to be
able to admit with sufficient probability the existence and extent of the damage.

The issue of uncertainty was similarly held not to preclude an ICSID tribunal
from assessing the quantum of damages in South Pacific Properties (Middle East)
Ltd v. Arab republic of Egypt. (226) In that case the tribunal stated that ‘…it is well
settled that the fact that damages cannot be assessed with certainty is not
reason not to award damages when a loss has been incurred’. (227) Tribunals are
generally allowed a considerable measure of discretion in determining issues of
quantum so as to facilitate a prompt and effective resolution to the dispute.
Thus, in Wena Hotels Ltd v. Arab Republic of Egypt (228) the ad hoc Committee
held:

With respect to determination of the quantum of damages awarded, it may be

recalled that the notion of ‘prompt, adequate and effective compensation’
confers to the Tribunal a certain margin of discretion, within which, by its
nature, few reasons more than a reference to the Tribunal's estimation can be
given, together with statements on the relevance and the evaluation of the
supporting evidence.

page "1144"

The determination of the precise amount of damage is a matter for the tribunal's
‘informed estimation’. (229) This is a principle that is widely accepted in
municipal as well as international law. (230)

Nevertheless, the assessment needs to be as reasonable as possible, including

appropriate discount factors. Where lost profits are concerned, in addition to
discounting for the time value of the money there should also be a discount in
terms of the risk premium that the profits may not eventuate. However, if a
party is entitled to lost profits and is able to prove them to a sufficient degree,
there should be no diminution simply because the potential profits were
particularly high in the context of typical returns on investment. Often profits
will be high because the risk is high. A contrary view would be based on an
inappropriate reading of the Aminoil case which granted a reasonable rate of
return as opposed to loss of profits because it found the parties had adopted this
standard. (231) In Aminoil v. Kuwait (232) the tribunal accepted claimant's argument
for compensation based upon the replacement value of the business as a going
concern, including lost profits. The tribunal consequently accepted in principle
that future earnings prospects should form part of the damages award. (233)
However, in calculating the lost profits that were to form part of the award, the
tribunal rejected an Income-Based method in favour of an Assets-Based method.
The tribunal awarded the net amount of an adjusted balance sheet plus a rate of
return until the award date as the measure of compensation. (234)

There will also be evidentiary issues as to how to identify lost profits, given that
it will often be a hypothetical exercise. Calculation problems will arise for
example where a mining company wrongly loses a concession or licence. (235)
With longer term commercial ventures such as construction contracts, mining
exploration or similar joint ventures, there will be complex risk factors to take
into account in assessments. (236) Tribunals may reject loss of profits claims
where there is insufficient evidence in support. (237) If there is no track record of
performance, a tribunal page "1145" might conclude that a loss of profits claim
is too speculative. (238) If profits are less than likely, a claimant may characterise
the claim as loss of an opportunity, particularly where the opportunity would
have a market value. Venture finance transactions and mining concessions are
examples.

Where lost profits are concerned, there is a question as to whether a claimant

should present direct evidence of this or rely on a commonly used formula such
as the Hudson or Eichleay Formulae. (239) Other questions include whether lost
profits should be awarded without any discount for the effort that would have
been required if contracts had been validly performed? Given that virtually all
legal systems look at cash flow and not utilisation of time, this seems an
unavoidable result. Lost profits should be calculated on the basis of net profit
after deduction of expenses. It is also important to ensure that there is no double
counting. This will typically arise in an avoided construction contract where the
builder argues that there have been wasted costs and also seeks lost profits.

14.10.1. Account of Profits

The remedy of an account of profits differs from lost profits in that, if awarded, it
requires the respondent to transfer to the claimant any profits that the
respondent may have obtained from his or her breach. The question that arises
is whether a damages assessment would require a party in breach to disgorge
profits from its breach of contract. An example would be an exclusive
distribution agreement where the distributor promises not to on-sell the goods
in other markets. If the distributor does so profitably, the seller has not suffered
any actual loss unless it can show that its own sales were displaced, but in
addition to that, the buyer has made a profit from its breach.

At common law, the position has long been that an account of profits is not
available for a breach of contract. This is based on the principle that a claimant
must show the damage that he has suffered if he is to claim compensation.
However, recent UK decisions such as Attorney General v. Blake (240) and
Experience Hendrix (241) in which an account of profits was been awarded for a
breach of contract indicate otherwise. Arguably, a principle now exists that
where a party to a contract deliberately breaches a contractual term and makes
a profit from the breach without causing the innocent party any provable loss,
the innocent party may be able to obtain as damages a reasonable payment
calculated as a proportion of the contract breaker's profit. In civilian legal
systems, the remedy of an account of profits is not generally available, as it is
understood to page "1146" be a punitive remedy. The French Civil Code, for
example, provides that the principal objective of contract damages is to
compensate loss, (242) and that accordingly, the intention of the party in breach
and the extent of any profit accruing by way of the breach are irrelevant.
14.10.2. Loss of a Chance or Opportunity

As noted above, it is important to separately consider the case where there is a

significant uncertainty as to whether there would be any profit or not. Legal
systems have to decide whether a party who lost a 20% chance of making profit,
should gain an appropriately discounted amount of compensation or instead
should be told that on balance they were unlikely to be profitable and hence no
compensation of that nature should be provided. While legal systems grapple
with the theoretical concepts, the proper approach may indeed depend on the
particular facts in dispute and whether the rights impaired by the breach would
have an independent market value. For example, a licence to explore for oil or
minerals might only be based on geological evidence that there is a 5% chance
of making a finding and hence of making a profit, but the licence would have a
value on the open market. The returns if successful would presumably vastly
outweigh the costs. Hence overall the entitlement is positive in value even if
actual success is less likely. Loss of an opportunity was allowed for in Sapphire
International Petroleum Ltd where it was held that it was only necessary for the
complainant to show ‘sufficient probability (of) the existence and extent of the
damage’. (243)

It has been suggested that it may be too difficult to assess damages for loss of
chance where a party has breached an obligation to arbitrate. (244) In any event,
the problem should not easily arise. If the claimant refuses to commence an
arbitration, the putative respondent could do so. If a respondent refuses to
attend, the arbitration proceeds in any event and there is no obligation to render
an award of specific performance. The arbitration promise is not to attend but
merely to allow arbitration to proceed, including the standard approach where a
respondent is not in attendance. This is separate to the question of damages for
breach of the agreement by commencement of legal proceedings in court.

14.10.3. Measuring Expectancy: Discounted Cash Flow

The discounted cash flow or ‘DCF’ method mentioned in section 14.8.6 is a tool
that measures the value of a business by projecting the net cash flow for a fixed
page "1147" period of time into the future and then discounting it back to
present value as of the date of the injury. As such it may be used to measure an
expectancy discussed in the previous section. A particularly problematic area in
that context which sometimes arises is loss of a right to tender in a construction
or distribution situation. Here there is the double uncertainty as to whether the
tender would have been successful and if so, what profits would have arisen.
There is also a question of net gains, as the aggrieved party has been alleviated
of the cost of going through the tender and setting up the outgoings of the
activity.

14.11. Calculation of Damages in Expropriation and other Investment

Standards

Expropriation cases are different to commercial disputes as BITs and customary

international law attempt to define valuation standards that arguably modify
standard market value analysis. However, as noted above, general comments as
to remedies from investment and expropriation cases are commonly resorted to
in other disputes. The measure of compensation may vary depending on
whether the expropriation is legal or not. A legal expropriation is within the
contemplation of the parties from the outset and is based on a compensatory
formula that they would have been aware of at the time of the agreement.

While complete restitution is well respected by most if not all legal systems,
application of this principle in expropriation cases has always been
controversial, in part because of what has become known as the Hull Formula,
named after then US Secretary of State Cordell Hull who advocated ‘prompt,
adequate and effective’ compensation when Mexico nationalised foreign owned
oil fields in the 1930s. The application of the Hull Formula has been questioned
by developing States over the validity of its application. (245) Customary
international law was not clear in articulating the nature of expropriation and
its remedies. ‘In the context of expropriation claims… discussions of fair market
value calculations inevitably overlap with the perennial dispute as to whether
customary international law requires compensation for an expropriation that is
‘prompt, adequate and effective,’ ‘fair,’ ‘just,’ ‘full,’ or ‘appropriate’ or otherwise’.
(246)

A number of general principles for a background to this debate. Article 34 of the

International Law Commission's Articles on State Responsibility indicates that
full reparation for the injury caused by the internationally wrongful act shall
take the form of restitution, compensation and satisfaction, either singly or in
combination. Under Article 36, a ‘state responsible for an internationally
wrongful act is page "1148" under an obligation to compensate for the damage
caused thereby, insofar as such damage is not made good by restitution’ and that
such compensation ‘shall cover any financially assessable damage, including loss
of profits insofar as it is established’ UN General Assembly Resolution 3171
declared that a State ‘is entitled to determine the amount of possible
compensation and the mode of payment, and … any disputes which might arise
should be settled in accordance with the national legislation of [the] State’ when
expropriating foreign property. (247) Nevertheless, such a Resolution cannot alter
customary norms of international law, hence the debate as to the extent of value
remains. Some tribunals have moved towards more equivocal terminology such
as just, appropriate or equitable compensation as opposed to full compensation.
No matter the terminology, full compensation ought to be the starting point, in
the sense of an accurate assessment of value, with arguments then permissible
about any circumstances that might indicate why some lesser amount would
seem fair and reasonable in the circumstances.

Investment treaties now typically aim to clarify the standards that are to apply.
(248) In ADC, an ICSID Tribunal stated that ‘there is general authority for the view

that a BIT can be considered as a lex specialis whose provisions will prevail over
rules of customary international law’. (249) Thus in the case of a BIT that specifies
the standard of compensation in the case of both lawful and unlawful
expropriation, the Tribunal will most likely apply the standard stipulated in the
BIT. However, where the treaty, BIT or alternative governing instrument does
not contain any lex specialis rules governing the issue of compensation, the
tribunal will apply the customary international law standard prescribed in the
Chorzow Factory case. In that case, the court famously said:

The essential principle contained in the actual notion of an illegal act – a

principle which seems to be established by international practice and in
particular decisions of arbitral tribunals – is that reparation must, as far as
possible, wipe out all the consequences of the illegal act and re-establish the
situation that would, in all probability, have existed if that act had not been
committed. (250)

Nevertheless, absent specific guidance, it is also possible to take different views

as to the aim of damages in lieu of restitution and what is fair. Professor
Brownlie in page "1149" dissent in CME Stockholm (251) interprets the phrase
‘just compensation’ in the Czech-Netherlands BIT as taking into account the
hardship that a large award would cause to the Czech State. (252) He concludes
that the injured investor's recovery should have been limited to the amount
actually invested in the expropriated venture, profits retained by the venture
and not distributed to the investor, and ‘foreseeable profits’ based on ‘a
reasonable rate of return’. (253) This valuation approach relies on a calculation of
the sunk investment costs along with a commensurate return from the injury
date until the recovery date calculated at a rate the tribunal considers
reasonable in the circumstances. The important difference between the two
approaches is that Prof. Brownlie would place the investor back into a position
as if the investment had never occurred while the Chorzow Factory dictum
focuses on putting the investor back in a position as if the investment had been
made but the injury had never occurred. (254) The recent case of Ioannis
Kardassopolous & others v. The Republic of Georgia (255) provides a useful example
of how an ICSID tribunal defines the appropriate standard of compensation for
unlawful expropriation. In that case, the tribunal had available to it three
contemporaneous arm's length transactions to assist in valuing the expropriated
investment. Accordingly, the tribunal found that it was appropriate in the case
to assess the fair market value of the investment by reference to the three
transactions.

A further consideration is whether a claimant is entitled to ‘full compensation’

or whether ‘public interest’ should be taken into account. In James v. UK, the
European Court of Human Rights ruled that the European Convention on Human
Rights does not ‘guarantee a right to full compensation in all circumstances.
Legitimate objective of ‘public interest’ such as pursued in measures of
economic reform or measures designed to achieve greater social justice, may
call for less reimbursement than full market value’. (256) This approach has
however been rejected by an ICSID Tribunal, which stated that international
investment law should not follow the ECHR perspective because ‘Article I of the
First Protocol to the European Convention on Human Rights permits a margin of
appreciation not found in customary international law or the Treaty’. (257) The
traditional approach has since been followed by other ICSID tribunals, including
the tribunal in Siemens, in which it was stated: page "1150"
The law applicable to the determination of compensation for a breach of such
Treaty obligations is customary international law. The Treaty itself only
provides for compensation for expropriation in accordance with the terms of the
Treaty.

The key difference between compensation under the Draft Articles and the
Factory at Chorzow case formula, and Article 4(2) of the Treaty is that under the
former, compensation must take into account ‘all financially assessable damage’
or ‘wipe out all the consequences of the illegal act’ as opposed to compensation
‘equivalent to the value of the expropriated investment’ under the treaty. Under
customary international law Siemens is entitled not just to the value of its
enterprise as of May 18, 2001, the date of expropriation, but also to any greater
value that enterprise has gained up to the date of this Award, plus any
consequential damages' (258)

Some cases have considered whether these principles apply to non-

expropriation situations. The tribunal in LG&E v. Argentina in its Award on
Damages held against this:

In the Tribunal's view, this type of valuation [Fair Market Value assessed
through DCF] is appropriate in cases of expropriation in which the claimants
have lost the title to their investment or when interference with property rights
has led to a loss equivalent to the total loss of investment. However, this is not
the case. The Tribunal rejected the claim for indirect expropriation put forward
by the Claimants … For the Tribunal, compensation in this case cannot be
determined by the impact on the asset value; it does not reflect the actual
damage incurred by Claimants. The measure of compensation has to be
different. (259)

Other recent awards have instead considered that the absence of explicit
standards of compensation for non-expropriatory breaches of international
investment law does not preclude the use of standards applicable in case of
expropriation. In CMS v. Argentina (260) it was said that:

the Tribunal is persuaded that the cumulative nature of the breaches discussed
here is best dealt with by resorting to the standard of fair market value. While
this standard figures prominently in respect of expropriation, it is not excluded
that it might also be appropriate for breaches different from expropriation if
their effect results in important long-term losses. Moreover, precisely because
this is not a case of expropriation, the Claimant has offered to transfer its page
"1151" shares in TGN to the Argentine Republic, and the Tribunal will address
this question in due course.

This is an approach that has been followed in Azurix, (261)

Enron (262) and Sempra. (263) A recent essay by Pierre Yves Tschanz and Jorge E
Vinuales explores the rules applicable to the determination of damages for
breaches of investment protection standards other than expropriation. (264) They
suggest that as a general matter ‘the rules and methodologies… that may be
used to assess damages in non-expropriatory breaches of international
investment law are flexible, contrary to those applicable in case of
expropriation, which are much more precise’. (265) In the absence of any specific
language contained in an arbitration agreement to the contrary, the ICSID
Tribunal in Vivendi stated as follows:

Based on these principles, and absent limiting terms in the relevant treaty, it is
generally accepted today that, regardless of the type of investment, and
regardless of the nature of the illegitimate measure, the level of damages
awarded in international investment arbitration is supposed to be sufficient to
compensate the affected party fully and to eliminate the consequences of the
state's action.

Of course, the level of damages necessary to compensate for a breach of the fair
and equitable treatment standard could be different from a case where the same
government expropriates the foreign investment. The difference will generally
turn on whether the investment has merely been impaired or destroyed.

This approach has been endorsed by other ICSID Tribunals and by other arbitral
investment tribunals. (266)

Assessment of compensation is easier where an investment is totally destroyed

or completely expropriated than where it is adversely impacted upon by a
measure, such as a new environmental regulation that renders a product less
valuable in the market place. Another complication where damages for
expropriation are concerned is the relevant time both as to the expropriation
page "1152" itself and also in determining value. (267) A further reason why
assessing compensation and damages in investment cases is more complicated
is because there is a very broad definition of investment typically covered,
which includes intellectual property and many legal instruments such as shares,
rights to use property, loans and the like as well as physical assets, often difficult
to assess without a ready market for such assets.

Because of the broad legal principles under investment treaties and customary
norms, these can at times be sought to be argued in relation to domestic
litigation remedies. An example was the Loewen dispute where an applicant
failed under a NAFTA investment arbitration claim when it complained in part
as to USD 400,000,000 in punitive damages and USD 74,000,000 in
compensation for emotional distress in a State court trial, where the initial
dispute was over some USD 4,000,000. (268)

While there are cases indicating that expropriation standards of damages should
not apply to non-expropriation cases, this may arise in any event where the
value of the relevant investment has been completely destroyed. This also arises
because in many cases, claims of discriminatory behaviour can also be argued in
the alternative as indirect expropriation of part of the value.

14.12. Currency of Damages

Generally speaking, issues of currency are substantive and flow from the
applicable law and the terms of any contract between the parties. However, the
discretion as to currency of remedies is at times enshrined in arbitral statutes
and rules. (269) Where broad procedural discretions are concerned, there are a
range of factors the tribunal should consider in determining what currency to
award damages in. A tribunal might look to the reasonable expectations of the
parties; or the currency that would best meet restitutionary concepts in the
substantive law; or give adequate concern to convertibility where there would
be major obstacles in some jurisdictions, although this should be in the context
of anticipated convertibility in the underlying contract if there had been no
dispute. The permitted currency in an award may not be the same as the
currency required for enforcement purposes, although the latter may also be a
factor in a tribunal deciding on the applicable currency.

page "1153"

Redfern and Hunter note that money sums are usually awarded in the currency
of the contract or the currency of the loss. (270) However, in some cases, tribunals
may need to receive written or oral submissions as to the currency or currencies
in which the award should be made, for instance, in large international projects
in which reference is made to several different currencies. If a remedy is to be
provided in a currency requiring conversion, an adjudicator will typically select
either the date of breach, the date of the award or the date of payment. There is
also a need to pick a correct rate of conversion. If there is more than one that is
listed, this may depend on which would be consistent with the commercial
abilities of the relevant parties to attract that rate. The logic flows from broad
restitutionary principles. There might also be concern with the problem of
currency fluctuations and selection of a fair conversion rate. The latter is more
difficult as it may involve a predictive exercise as to the likely value at the time
an award is honoured. (271)

14.13. Some Specific Scenarios in Damages Assessment

14.13.1. Arbitration Costs as Damages

Costs are dealt with separately in Chapter 15. At this stage it is merely
appropriate to note that a party could in theory seek to construct an argument
that breach of contract that includes an arbitration agreement leads to
consequential losses in relation to the costs of the arbitration itself. This would
not be the normal way that a tribunal would approach this issue. A particular
concern would be where the parties have expressly agreed that there is to be no
costs award by the tribunal. It would undermine the intent if the successful
party was allowed to recharacterise costs as damages.

14.13.2. Failure to Pay an Advance on Costs

There are two approaches to seeking recovery by one party when it has had to
advance a share of costs of its opponent to allow the arbitration to proceed. One
approach is based on a ‘contractually agreed procedural duty’ while the
alternative is to consider it as some form of interim measure. (272) Whether it is
possible to claim damages for breach of the supposed contractual duty may
depend upon whether it is institutional or ad hoc arbitration. Where an
institution is concerned, an express page "1154" agreement to the rules that
expressly directs contributions to advances could be argued to be an agreement
between the parties, a breach of which might lead to a damages entitlement.
However, this is not uniformly supported. ICC Case No. 12491 considered that
there was no damage at that stage and allocation of costs would only be
determined at the time of the award. (273) This is discussed further in section
6.11.

14.13.3. Damages for Litigation Expenses Caused by Breach of an

Arbitration Agreement

If one party commences court action in breach of an arbitration clause, the

breach itself could cause loss to the innocent party, at the very least in seeking
to bar the court action on the basis of the exclusive arbitration clause. While the
relevant court may itself award costs, these may be limited and the balance
could be argued to be damages in the arbitration. Common law courts have
accepted this approach. (274) It is suggested that in the civilian law tradition, the
better view is that forum selection clauses are procedural, the breach of which
does not give an entitlement to damages. (275) It is even suggested that the
innocent party may need to defend the court proceedings by way of mitigation
of loss. (276)

14.13.4. Delays in Construction Disputes

A problematic area of damages calculation is when there are claims for delay
and disruption, most typically in a construction dispute. Delays can be
considered in the context of breach or simply in the context of contractual
entitlements to extension of time. (211) Assessing damages for delay where all
relevant cash flows will ultimately ensue, may simply be a present value analysis
of delayed profits. Delay in construction contracts will generally lead to
increased overheads in terms of staff and office costs. Different formulae are
typically applied so as to avoid the need to page "1155" specify the exact
marginal impact on overhead expenses. (278) Because extensions of time or
expenses for delay and disruption usually involve establishing certain
conditional requirements and following stipulated procedures, there will be
interpretation issues as to both entitlement and amount. Another evidentiary
issue is whether final certificates in construction matters constitute binding
evidence as to their content. (279)

Disputes about delaying completion for construction contracts are particularly

problematic as they often involve cross-claims and difficult questions of
causation. For example, the contractor might argue that the delay was caused by
changes called for by the employer or the latter's failure to provide necessary
information and assistance. The employer might blame the contractor for failing
to proceed at an appropriate rate. Each may blame third parties or external
circumstances for their own lack of performance. As with other matters in this
chapter, the actual rights and obligations in a particular transaction will depend
on the applicable law and terms of the contract. Many questions are simply
about liability and would not otherwise impact on calculations. Nevertheless,
some broad proposition in the context of evidence and procedure can be
appropriately made. While it is relatively easy to state a date for completion, the
degree of completion is more open to debate as it will typically involve a
contractual stipulation of a concept such as ‘practical completion’. Even the date
for completion is problematic, as it needs to be read in the context of
entitlements for extensions. Evidentiary problems are exacerbated by the fact
that it rarely makes commercial sense to demand that a contractor stick to a
rigid and detailed programme but in the absence of such a document it is harder
to identify and compare what was intended and what did in fact happen. (280)

Even if there is no detailed contractual programme, there should be appropriate

evidence of what has occurred from time to time as to the basis of delay claims.
In some cases a question may also arise whether a programme is approved and
whether an employer was entitled to refuse to approve a tendered programme.
Where external circumstances are involved, delay analysis will involve
interpreting the contract to determine who ought to have borne the risk for this
external event.

14.13.5. Reasonableness, Good Faith and Abuse of Rights

Principles of reasonableness and good faith could be applied to articulating the

interface between primary damages entitlements and mitigation duties. page
"1156" The Himpurna tribunal considered that on the facts before it, a claim for
lost profits on investments not as yet made was so unrealistic in terms of the
potential detriment to the host State that it was ‘likely to constitute an abuse of
right inconsistent with the duty of good faith that is fundamental to the
Indonesian Law of Obligation’. (281)

It is undesirable to use the concept of abuse of right to limit the damages in

Himpurna. The abuse of rights doctrine relates to concepts of good faith and
contemplates rights being used in ways which would not be legitimately
contemplated. This is quite different to the case of contractual terms that are
seen as being unbalanced from the outset. (282) It is difficult to employ an abuse
of rights logic absent a consideration of more direct concepts such as that of
hardship. If a party repudiates a contract that has become grossly unfavourable
because of changed circumstances, most would agree that the expenditure of the
innocent party should be recoverable but it is more problematic to consider
whether it should be entitled to full loss of profits when the new circumstances
make the risk/reward for both parties grossly different to their original
expectations. (283)

14.14. Non-Pecuniary and Moral Damages

Two issues will typically arise, the first whether such damages are claimable and
second whether the claimant has satisfied its burden of proof. (284) A number of
civilian legal systems allow various categories of non-pecuniary loss, including
pain and suffering, emotional distress and moral harm. (285) The ILC Articles
specifically require reparation for moral as well as material damages. Article
31(2) provides that ‘[i]njury includes any damage, whether material or moral,
caused by the internationally wrongful act of a State’. Computational issues will
obviously be different. The evidentiary basis of claims for pain and suffering or
loss of general reputation cannot be as precise as such items as wasted
expenditure, although measuring loss of reputation in a commercial context
ought to be similar to valuing diminution in goodwill. (286)

page "1157"

Investment treaty provisions rarely provide for claims by the State, including
claims for loss of its own reputation as a result of improper prosecution. (287)
There may be private law principles that could otherwise apply. Whether an
investor is entitled to moral damages may depend on whether the relevant BIT
refers to nondiscriminatory and similar treatment to the investor as well as the
investment. (288) In turn that may depend on whether the investor is an
individual or a corporation. Even non-discrimination in relation to an investor
may be limited to treatment of the investment itself. In Desert Lion Projects LLC
v. Yemen (289) an arbitral tribunal awarded moral damages for stress as well as to
injured reputation when armed gunmen expelled the investor from the relevant
country.

One risk to be avoided in awarding compensation for moral harms is double-

counting, in case such harms have already been compensated as material
damage. (290) The risk is more likely to be present when the fair market value
(FMV) of a business is used as the measure of compensation. This is because
among the various components of FMV is goodwill, which includes the value of
the business's reputation. (291) According to Sabahi and Birch, when an arbitral
tribunal uses some other valuation method which does not take into account the
goodwill, or more specifically damage to reputation, then the recovery of moral
damage to reputation would be justified. (292)

14.15. Set-off

Whether other claims can be set off as against damages entitlements is a vexed
question in international arbitration, primarily because there is no consensus as
to whether set-off should be seen as a matter of procedure or substance. Brower
refers to a case where an ICC tribunal did not set off a monetary award to an
insolvent claimant against a far greater award to a counterclaiming sovereign
respondent. The defaulting claimant assigned its entitlement to an affiliated
company which page "1158" sought to enforce it, while at the same time seeking
to evade its own award debt. (293) A rare example of a rule seeking to expressly
allow for set off rights is contained in Article 21.5 of the Swiss Rules of
International Arbitration 2012. Set off is discussed in detail in section 4.4.3.

14.16. Force Majeure and Frustration

In most legal systems, proof that the cause of the loss was a force majeure or
frustrating event will relieve the respondent of the obligation to pay damages.
Key legal and evidentiary issues would include analysing questions of
impossibility and foreseeability at the time the contract was entered into. If the
contract or applicable law does not expressly deal with this defence, some
would assert that it is a principle of lex mercatoria in any event, although that
would be open to debate.

14.17. Mitigation of Damages

Most substantive laws will require the innocent party to take reasonable steps
to mitigate losses, including loss of profits, arising from breach. This may
overlap with general duties of good faith, duties to renegotiate for changed
circumstances and hardship provisions. It may also overlap with causation and
foreseeability analysis. Comparative law suggests that different legal systems
address the various categories either through notions of mitigation,
contributory negligence or causation. Civilian legal systems may have been less
concerned with principles of mitigation simply because specific performance
was the primary remedy, with damages only being residual. Many civilian legal
systems do not speak of mitigation per se but instead consider notions of fault
on behalf of the aggrieved party. (294) From a conceptual perspective, mitigation
can simply be seen as an aspect of the need for a causal link, as arises for
example under French law. (295) This has important implications if there is an
inadequate evidentiary record and a tribunal wishes to apply a strict approach
to burdens of proof. Such an approach is not to be preferred. The duty to
mitigate page "1159" could also be asserted to arise as a matter of custom and
usage (296) and constitute an element of lex mercatoria. (297)

While mitigation is often spoken of as a duty, it is not a duty in the sense that it
give rights to claim breach to other persons, but instead it is a means of reducing
damages otherwise proven. (298) Concepts of mitigation are also commonly built
into the primary rules of assessment of damages even subconsciously. For
example, if a purchaser of goods has entered into a substitute transaction, the
actual loss will be the difference between the two, the substitute being a
mitigating event. It could also be dealt with under foreseeability notions if
unmitigated loss is ‘outside of the bona fide expectation of the non-performing
party’. (299) It would normally be the case that if the parties have agreed on
liquidated damages, a duty to mitigate would not be a basis of decreasing the
entitlement. (300)

Article 9.504 of the Principles of European Contract Law directs that the non-
performing party is not liable for loss suffered by the aggrieved party if the
latter contributes to the non-performance effects. Article 9.505 also indicates
that the non-performing party is not liable for loss ‘to the extent that the
aggrieved party could have reduced the loss by taking reasonable steps and that
the aggrieved party is entitled to recovery any expenses reasonably incurred in
attempting to reduce the loss’. Article 7.4.8 of the UNIDROIT Principles states:

1. The non-performing party is not liable for harm suffered by the aggrieved
party to the extent that the harm could have been reduced by the latter
 party's taking reasonable steps.
2. The aggrieved party is entitled to recover any expenses reasonably incurred
in attempting to reduce the harm.

Article 77 of the CISG states:

A party who relies on a breach of contract must take such measures as are
reasonable in the circumstances to mitigate the loss, including loss of profit,
resulting from the breach. If he fails to take such measures, the party in breach
may claim a reduction in the damages in the amount by which the loss should
have been mitigated.

While it is easy to state the general principles of mitigation in this way, the more
difficult question, including legal and evidentiary perspectives, is what standard
of behaviour is to be expected. In a commercial context the difficulty is that
many page "1160" steps would be expensive, uncertain and may be problematic
if the claimant does not have a strong liquidity position. How much risk and
expense should a claimant undertake in order to ensure that damages otherwise
payable are not reduced under mitigation principles? Laws, cases and
commentators naturally fall back on broad expressions such as a requirement to
take ‘reasonable’ measures, that the claimant need not take every imaginable
step, but will invariably suggest that quite significant and costly steps will at
times be appropriate.

The relevant substantive law will also have its own particular test as to the kind
of behaviour that would be considered reasonable and the extent to which the
innocent party must take risks in seeking to minimise the loss. Important issues
of timing include whether steps only need to be taken once the breach occurs or
should at times be taken previously, when breach can be reasonably anticipated.
(301) This might be answered by the applicable substantive law although any

tests will ideally depend on the circumstances, as there must be situations

where an impending breach is so clearly the case and reasonable alternative
steps are immediately available that this should be taken into account. Once the
legal standards are determined, what did happen and what ought to have
happened would be questions of fact for the adjudicator's determination.
Evidentiary challenges are also compounded by the fact that the burden of proof
will typically be on the breaching party to show that the claimant failed to
mitigate its loss, but it naturally does not have access to the key relevant data.
For example, a seller of defective goods seeking to argue that alternative
supplies were readily available, would be looking for supportive information
from its competitor manufacturers. An adjudicator will also face document
production issues in such circumstances. One problem is that much of the
market information may be confidential and not readily available from third
parties. The innocent party itself may not wish to divulge confidential internal
information in relation to which such assessments might be made.

There are a number of aspects of mitigation calculation. First, a party will not be
entitled to damages which could reasonably have been avoided. If reasonable
steps have been taken, which consequently reduced the loss, the cost of those
steps should also be borne by the person in breach. Only a net analysis would be
commercially realistic. A party should also not do anything to aggravate the loss.
If the claimant did so, this would break the causal link as to that part of damages
in any event. However, the corollary of the duty to act reasonably to minimise
loss should be that if the steps were seen as reasonable at the time but in fact
increased the loss, that should still be at the expense of the party in breach. That
could also be justified under causation and foreseeability analysis.

It is generally accepted that the burden of proving a failure to mitigate should be

on the breaching party. (302) The key aspect will be comparing what the innocent
party did do with what its opponent alleges it ought to have done. This involves
a page "1161" combination of factual market information and hypothetical
assessment. In considering the kinds of activities that may or may not be
reasonable, there may be a need to balance a number of factors including cost,
time, risk and impact upon goodwill. Where the latter is concerned, it may be
reasonable to seek all damages within a confidential arbitration as opposed to
going into the market-place to highlight a problem with one's business. In some
cases an adjudicator might hold that while steps were taken which did reduce
the loss, more reasonable or timely steps could have been taken which should
again be a means of reducing residual damages. An example might be a delayed
substitute purchase by an innocent buyer in a market with steeply rising prices.
If the purchases had occurred earlier, the cost of the alternative would have
been lower. In making the assessment, it is important to consider what would
have been reasonable to foresee or to consider as alternatives in the
circumstances known to the parties at the time or which could have been known
on reasonable inquiry. It should not be examined through hindsight. Other
problematic scenarios might include an improper breach of a distribution
agreement that offered generous terms to the distributor, with the party in
breach offering a far less favourable but still profitable arrangement in lieu
thereof. A distributor who has worked full time for that supplier would not
necessarily be expected to continue in such unpleasant circumstances. There
may also be moral and ethical issues that arise, such as whether a builder should
lay off construction workers when a supplier of materials will significantly delay
delivery.

A duty to mitigate will often imply a duty to give the party in breach an
opportunity to cure the defect. There may then be factual questions as to how
long one should give the party in breach to do so and what assistance both in
time and money should be provided. At other times the duty to mitigate entails
the innocent party actively working with the party in breach to try and
consummate the transaction. For example, if a seller of goods does not have the
funds to organise transport, the buyer would need to do so even if it was a CIF
contract. In other cases where a seller provides goods not in accordance with
the requirements of the contract, the buyer may still need to take delivery, on-
sell the goods at a profit and call for a new shipment of the intended
specifications. If a third-party sale was at a loss, that might be problematic in
terms of being able to recover the difference as damages, particularly if the
seller could have instead taken back the non-complying goods.
In cases where the seller has breached the contract but promises to make
amends in time if the buyer keeps the contract on foot, situations may arise
where it would be needed to determine whether the buyer should legitimately
have taken the risk, given the circumstances and the seller's overall reputation
and likely ability to deliver on its new promises. A tribunal might consider that
loss might only be mitigated after a considerable time and would allow full
damages in the meantime, including loss of profits. (303)

Most legal systems will not require a claimant to refrain from avoiding a
contract if valid grounds exist simply because this might reduce the total
damages, although in the latter circumstances it becomes arguable that the
breach was not page "1162" sufficiently fundamental to constitute an
entitlement to terminate. This is essentially a question of applicable substantive
law. Ultimately, an adjudicator should make the best possible factual assessment
of the impact of the hypothetical reasonable steps on the actual loss suffered
and should not simply pluck a round figure as a discount.

When considering what degree of risk a party could hypothetically be expected

to take, one uncertainty is whether that party's own historical risk aversion
preferences should be taken into account. Risk aversion should certainly take
into account the liquidity position of the innocent party. An example of a
situation of risk and uncertainty would be delayed payment in a foreign
currency that is falling drastically on exchange rate markets. Whether the
innocent party should undertake a hedging contract to stem the potential losses
is difficult to determine as the exchange market could turn around. Because the
burden is on the breaching party to show breach of a duty to mitigate, in that
circumstance it might need to show that consistent expert advice was that the
currency situation would continue to deteriorate and hence a hedging contract
would stem the losses.

In some cases a failure to take out insurance might be seen as a failure to

mitigate or at least a contributory cause of the loss, although this would not be
so under legal systems that require no mitigating activities until breach occurs.
A contentious example would be a failure to take political risk insurance in an
investment scenario. Given that such insurance is often provided by the
investor's home State on favourable terms, one argument is that a reasonable
person would have naturally undertaken such a step. Another example of risk is
where the issue is whether an investor should continue with an investment after
some bad experiences where the host State has breached its obligations in
drastic ways. Even if the investor is entitled to cease activities on this basis,
there would be a duty to sell its interest at the best possible price, although in
some circumstances, a tribunal could rightly conclude that there would be no
likely buyer. (304)

Where avoiding loss from a sale of goods contract is concerned, the innocent
party, whether buyer or seller, would normally look to make a substitute
contract. (305) As long as the substitute contract is profitable, there would be a
duty to undertake one on less favourable conditions as any profit will mitigate
the damage. In some cases, however, it will not be readily apparent whether the
substitute contract is reasonable or not. For example, if a party habitually buys
high grade materials with challenging specifications from one reliable supplier
in a particular country who fails to deliver, what expense should be undertaken
to find suitable suppliers in other countries and what risks should be taken
where there is insufficient time to properly test for quality? At the extreme, one
might be able to say that the hypothetical substitute is not truly of that nature,
particularly where quality and reliability is crucial. (306) Substitute transactions
are also difficult page "1163" to analyse from an evidentiary point of view as
traders are always trying to make as many sales as possible, hence it is at times
difficult to distinguish between a truly substitute purchase which could have
mitigated the loss and another transaction which could have been undertaken in
any event but for the breach. (307)

Where a respondent establishes that the claimant may have failed to mitigate its
loss, the CPR Protocol recommends that tribunals should consider granting
requests from the respondent for the claimant to supply information concerning
its activities following the breach. (308) In this way the arbitrators can better
determine the financial impact of mitigation measures that the claimant might
have taken as well as any benefits derived by the claimant from the breach.

14.18. Liquidated or Agreed Damages and The Prohibition on Penalties

Both civil and common law families allow for liquidated damages but each
allows challenges to be made where the amounts are excessive. Civilian systems
will typically allow for reductions, while common law systems will seek to
determine whether the clause is void or voidable as an improper penalty
provision. Under French law, a judge may vary the amount even on an ex officio
basis, which cannot be prevented by the parties. (309) Islamic principles would
incline to this approach. Article 7.4.13 of the UNIDROIT Principles adopts the
civilian approach allowing for specified damages but allowing the sum to be
reduced to a reasonable amount if grossly excessive in relation to the harm
resulting from non-performance and in the context of other circumstances. (310)
Liquidated damages were accepted in ICC Case No. 3267. Even if struck down,
this does not preclude a claim based on proper proof of actual damage if brought
within time and in the appropriate manner.

Liquidated damages or penalty provisions are essentially calling for

determinations of a factual nature. Under the civilian approach, where amounts
can be reduced if they are unreasonable, there would usually need to be
comparable evidence to show that the amount is actually excessive. Under the
common law approach, attention is given to such data but also such factors as
whether it was truly an attempt to fix compensation or instead to provide for a
punishment. This is because common law looks at the factual issue of the intent
of the clause which to be valid requires ‘a genuine covenanted pre-estimate of
damage’. (311) Distinguishing between acceptable levels of preordained damages
and penalties to promote enforcement will raise evidentiary issues as to
methods of calculation and appropriate discounting for a range of contingencies.
Other evidentiary elements that page "1164" may come into play are whether
the provision in issue was adequately compensated for through the price or
other mechanisms. Where that is so, a provision that might look problematic in
isolation may be highly commercial when viewed in context. An owner might
accept a building tender at a higher quote where the builder has shown a belief
that tight timeframes can be met in the context of high-end liquidated damages.

Conversely, where liquidated damages provisions provide for a capped lower

level, they become akin to partial exclusion clauses, (312) although some legal
systems allow claims for additional damages over and above the liquidated
amounts. (313) Liquidated damages provisions also mean that as long as the
provision is valid, the claimant will not need to prove actual damage. It would
normally be the case that if the parties have agreed on liquidated damages, a
duty to mitigate would not be a basis of decreasing the entitlement, although
each case should be looked at on the merits as it may be a question of drafting
and interpretation. As noted, liquidated damages are commonly provided for in
construction contracts for late completion where delay is problematic but hard
to quantify. A liquidated damages clause can also induce the builder to stay on
track as the agreed damages may soon remove the builder's profit on the deal.
There may be debates as to whether the clause is truly a liquidated damage
clause or is instead a formula for proper calculation of loss. There may also be
questions of interpretation such as whether it should be construed narrowly.
(314) In some cases it is not the level of damages that may render the clause penal

in nature under common law analysis but rather factors such as timing of
payment as against stages of the work or a lack of indication as to how
cumulative delays are properly addressed. (315) For example, a lack of
appropriate extension of time mechanisms where the fault might be that of the
employer can undermine a liquidated damages provision. (316)

From a procedural perspective, penalty provisions and limitation of liability

clauses may prove problematic for arbitrators where there is no choice of law by
the parties, the matter is left to the arbitrators' discretion and there are tenable
arguments for at least two legal systems, only one of which supports such
provisions. Similar problems arise where one system is more liberal than the
other. In making the choice as to applicable law, the arbitrator will be aware that
he or she is making a choice as to the validity, vulnerability or ambit of the
clause page "1165" itself. There may also be a conflict between trade usage and
the applicable law if trade usage is more readily accepting of incentive and
disincentive provisions that are not tied to actual damages. For example, we
readily accept agency and distribution agreements that may give variable
benefits depending on turnover. On one view, there is no reason to see
construction contracts as a priori problematic simply because they also provide
incentives and disincentives for early or late completion. Informed people will
build all of this into price and their risk assessment. The argument proceeds that
if limitation of liability clauses are enforced under the applicable law, the
converse situation which crystallises liability may not need paternalistic
interference by adjudicators.

14.19. Exclusionary Clauses

As with other issues, it will be a question for the applicable law as to the validity
and treatment of exclusionary clauses. Provisions which limit remedies will
typically raise important evidentiary issues such as evaluation of pre-
conditional scenarios such as where parties limit damages to fraud, deliberate
acts or gross recklessness. Where there is a limitation on consequential loss, this
may require interpretation as to the meaning of the term. (317) There may be
general interpretation questions such as the application of contra proferentem.

14.20. Punitive Damages and Statutory Formulae

The notion of punitive damages involves an award of more than a compensatory

amount to respond to inappropriate behaviour by the losing party. Where such
damages are sought in arbitration, there will always be a need to consider
express statutes and rules and any agreement of the parties. Historically,
punitive damages have been treated differently in different legal families. They
have been accepted at times in common law jurisdictions and are often used for
public policy reasons such as in the common law of tort and in statutory
antitrust regimes. In each case this is with a view to deterring inappropriate
behaviour and in the latter case, also providing incentives for weaker
commercial parties to take action against major entities. (318) Conversely, civilian
legal families tended to consider such damages to be improper, although there
have been recent moves to at least consider such developments in certain
circumstances.

page "1166"

Where there is no express guidance, some writers (319) have argued against such
powers for arbitrators on the basis that punitive damages were only for judges
so authorised by the State, that arbitral consent does not go to punishment, that
the absence of an appeal mechanism makes punitive damages inappropriate,
and finally, that the confidential nature of arbitration means that the broader
educative value will not be felt. Those who argue against punitive damages often
rely on the explicit refusal of certain tribunals to award punitive damages. Gray
notes that the most famous of these is perhaps Umpire Parker's discussion of
the problem in the 1923 US-Germany Commission. (320) A converse view is that if
a particular issue is arbitrable and it naturally has punitive damages in a
business setting, to deny an arbitrator that power would mean that the expected
commercial outcome cannot be achieved through the arbitral process.

At times, rules of lex arbitri will be determinative. For example, Article 28.5 of
the AAA International Arbitration Rules prohibits the tribunal from awarding
‘punitive, exemplary or similar damages…’ unless agreed to by the parties or
unless such relief is required by an applicable statute. Care should be taken with
some civil law jurisdictions which view punitive damages as contrary to public
policy. Some lex arbitri will prohibit the awarding of punitive damages even if
the applicable law provides for this. Some take the view that in appropriate
circumstances, a tribunal's broad powers do extend to the granting of punitive
damages. In Sedco Inc v. NIOC, Judge Brower in a separate opinion, considered
that punitive damages may be acceptable where there is an unlawful taking. (321)
The view was rejected subsequently in Amoco Int'l Fin Corp v. Iran. (322)
The suggested approach by Redfern and Hunter is for the arbitral tribunal to
examine whether punitive damages may be awarded under the law applicable to
the substance of the dispute, and to also consider ‘the threshold question’ of
whether they have the power to award punitive damages by examining the lex
arbitri, the law of the place of arbitration, and the terms of the arbitration page
"1167" agreement. (323) Regarding enforceability concerns, Redfern and Hunter
argue that such concerns ‘should be left for the courts at the place of
enforcement’ but suggest tribunals ‘treat any award in respect of punitive
damages or any other penalties as an entirely separate claim, in order to ensure
that the punitive portion of the award is severable in the event of a successful
challenge in the courts at the place of enforcement’. (324)

Because of the problems of classification, selection of an applicable law does not

necessarily resolve all issues. For example, if the parties select a law that does
not allow for punitive damages, a tribunal may still conclude that this only deals
with the criteria for determining when punitive damages may be awarded and
not whether the arbitrator has a power in the first place. (325) The contrary view
is that arbitrators should not ignore an express choice of law by the parties.
Parties cannot be presumed to agree to one without the other. The situation
may be different with statutory damages where the relevant statute is
considered arbitrable. The best example is US antitrust law, where relevant
statutory provisions allow a claimant to seek treble damages. The US Supreme
Court upheld an arbitrator's ability to determine punitive statutory damages in
a domestic arbitration in Mastrobuono v. Shearson Lehman Hutton Inc. (326) There
may also be questions whether certain jurisdictions will enforce awards of
punitive damages under public policy norms. An award for punitive damages
has been refused enforcement in Japan as being contrary to public policy. (327) It
has also been suggested that in Germany and Switzerland, punitive damages
awards may be considered incompatible with domestic public policy and may be
set aside for that reason. (328) Thus, party agreement may not necessarily
empower the award of punitive damages where that would be against public
policy. (329) In a controversial decision, a Swiss tribunal page "1168" applying
New York law considered that punitive or exemplary damages were contrary to
Swiss public policy and should not be applied by a tribunal sitting in
Switzerland. (330) The decision is rightly criticised by Born who points out that
setting aside is only possible under transnational public policy and further that
if the parties have selected a law allowing for punitive damages, this should be
respected. (331)

A number of situations may vary depending on whether the parties have

expressly allowed for punitive damages or conversely, where one of the parties
makes a controversial claim in law giving rise to similar effect. For example, if
the parties expressly agree to arbitrate a US antitrust dispute they are agreeing
to all of the remedies including triple damages. A converse situation is a contract
with an express choice of law of one of the parties, where the other argues that
its own antitrust law applies extraterritorially notwithstanding the choice of law
of the contract itself. The latter scenario raises issues of mandatory law. If the
antitrust law is in fact mandatory and comes with those remedies, then an
arbitrator may feel bound to apply it, although the ambit of mandatory laws is
also controversial. (332) A consideration of mandatory laws even adds further
complexity, given that it may at times be argued that a mandatory law of the
forum is against the application of punitive damages under consensual
arbitration.

It is important to understand that a tribunal can be challenged either way. For

example, an arbitrator sitting in the US where antitrust or RICO laws apply may
be forced to consider awarding punitive damages or could otherwise be alleged
to have failed to complete the mandate or otherwise be challenged on the
grounds of public policy. (333) Conversely, if the Seat of arbitration is in a place
that does not support punitive damages, the tribunal may find it difficult in
deciding how to proceed, particularly when the parties have not selected the law
imposing punitive damages but where the argument is that it is a mandatory
substantive law in any event. The Chartered Institute's Guidelines suggest that
where a page "1169" tribunal is in doubt it might ‘make the award according to
the applicable law chosen by the parties, make it clear which parts of the award
relate to any punitive damages element and risk partial annulment or limited
enforcement of the award’. (334)

14.21. Judicial Penalties

It is not clear whether arbitrators are entitled to impose judicial penalties

otherwise known as astreintes. (335) This involves an adjudicator ordering a
payment of a specified sum of money in the event of a failure to meet another
adjudicatory direction. The aim is to induce compliance with the latter. Judicial
penalties are different to liquidated damages or penalty provisions in contracts,
although the latter can also aim to provide incentives for timely compliance. A
judicial penalty will typically be a stipulated amount for each day, week or
month that the obligation remains unmet. Penalties may relate to substantive
issues such as specific performance of contractual obligations, or bringing to an
end actions in contravention of contractual obligations. In these circumstances,
there is compensatory logic. A penalty to induce performance could in part
compensate for damages for delay. A penalty seeking to prevent continued
breach could cover the lost profit or unjust enrichment from the breach.
Nevertheless, the essential nature of the judicial penalty is an amount set more
as an incentive or disincentive as the case may be. A judicial penalty could also
relate to procedural matters, such as responding to document production
orders.

Whether arbitrators can or should award punitive damages was considered in

various US cases including Mitsubishi Motors Corp v. Soler Chrysler Plymouth Inc
(336) and Mastrobuono v. Shearson Lehman Hutton Inc. (337) On the one hand, it

might be argued that only judicial officers ought to have the power to punish
through damages awards for wider societal benefits. A converse view is that if a
particular issue is arbitrable and it naturally has punitive damages in a business
setting, to deny an arbitrator that power would mean that the expected
commercial outcome cannot be achieved through the arbitral process. Judicial
penalties are more likely to arise in conjunction with injunctive relief. Thus, in
ICC Case No. 7895 a tribunal granted an injunction against sales of goods subject
to an exclusive distribution agreement coupled with a fine for each product sold
in violation of the injunction. (338) Intellectual property, distribution and
licensing arrangements would be examples where this should at least be
considered. Mourre refers to a number of unpublished awards that ordered
penalties in support of orders page "1170" or injunctions, (339) although he
concluded that these were insufficient to show a common arbitral practice.

Where substantive penalties are concerned, there may be more of a question of

applicability where the primary obligation is as to monetary payments, as these
can be enforced by other means and late payment can be subject to an interest
direction. (340) The application would be less contentious if the monetary
obligation was to pay an amount to a third party or open an escrow account as
per direction, (341) although the latter could be part of a procedural order in any
event. The penalty set could be provisional or final.

The entitlement to apply a procedural penalty would naturally flow from

procedural powers applicable to the arbitration. Where substantive penalties
are concerned, they would naturally seem to relate to the power to do so under
the applicable substantive law, although in some cases this could still be seen as
a procedural issue, relating as it does to the primary order on a substantive
matter. Poudret and Besson acknowledge the distinction by suggesting that
astreintes are permissible to reinforce merits determinations where allowed by
the lex causae, but this should not apply to procedural orders. (342) In some cases,
therefore, a judicial penalty will be akin to a provisional measure. If that is so,
there is a need to ensure that any action taken is consistent with express
provisional measure rules. If the entitlement to judicial penalties depends on the
lex causae, any limits on the tribunal's powers to award specific performance
would naturally undermine the penalty entitlement, as the latter strongly
relates to this primary remedy.

Such powers are allowed for under the UNIDROIT Principles Article 7.2.4 which
refers to court orders. (343)

(1) Where the court orders a party to perform, it may also direct that this party
pay a penalty if it does not comply with the order.
(2) The penalty shall be paid to the aggrieved party unless mandatory
provisions of the law of the forum provide otherwise. Payment of the
penalty to the aggrieved party does not exclude any claim for damages.

Article 1.11 of the UNIDROIT Principles defines ‘court’ to include an arbitral

tribunal. The Commentary to the UNIDROIT Principles makes the following
observation:

While a majority of legal systems seem to deny such a power to arbitrators,

some modern legislation and recent court practice have recognised it. This page
"1171" solution, which is in keeping with the increasingly important role of
arbitration as an alternative means of dispute resolution, especially in
international commerce, is endorsed by the Principles. Since the execution of a
penalty imposed by arbitrators can only be effected by, or with the assistance of,
a court, appropriate supervision is available to prevent any possible abuse of
arbitrators' power. (344)

Most arbitration statutes are silent on this although the entitlement is

proscribed under the Swedish Arbitration Act, (345) and probably Italian law as
well. Statutory provisions or case law seem to support the power in Belgium,
France, The Netherlands and Switzerland. (346) There are differences in view
whether the power to impose penalties is inherent, absent agreement to the
contrary by the parties, (347) or absent a contrary direction in the lex arbitri. The
inherent power argument appears somewhat strained given the lack of
worldwide consensus on this issue. Even more strained is an implied consent
argument. (348) This may depend on whether the matter is seen as a procedural
or substantive penalty. Laurent Lévy (349) and Alexis Mourre (350) both support
the view that judicial penalties should apply in international arbitration to assist
the efficacy of arbitral determinations. There may also be problems in seeking to
rely on courts to impose such penalties as part of post-award relief, (351)
although in many cases that may be different to the kind of conduct during the
arbitral proceedings themselves.

An important question where any procedural issue is concerned is whether an

enforcement or annulment court would take issue with what it felt was an
erroneous utilisation of a judicial penalty. Some enforcement courts might also
see a judicial page "1172" penalty as a determination contrary to public policy.
The Court of Appeals of Paris chose not to overturn an award even where the
tribunal imposed a penalty without being requested to do so by claimant. (352)
This also raises the question as to whether there will be a difference between
requests made by a party and ex officio determinations by a tribunal. An
application ex officio may be more problematic in jurisdictions that otherwise
allow for the power. (353) Because of the uncertainties around the entitlement,
parties negotiating contracts where directions, declarations or injunctions may
be central to an efficient outcome, may well wish to expressly grant the tribunal
such powers.

Because such penalties are independent of damages assessments, and are paid
to the aggrieved party, they may lead to the latter being more than fully
compensated, (354) although this may depend on the applicable law as to the
penalty and whether an adjudicator would wish to take this into account in the
assessment of restitutionary damages. Where that is concerned, Mourre makes
the important observation that these might lose their deterrent effect if the
party in breach knew they were merely an advance on damages. (355) In one
sense, they can never be purely seen as an advance, as if the aggrieved party
loses the case, the party subject to the judicial penalty does not get
compensation for the penalty paid. If the aggrieved party wins and the judicial
penalty is higher than the amount of damages that would otherwise have been
assessed, the difference is not refunded either. While the amount of the penalty
assessed may to some extent be arbitrary, most would assess it at a level
sufficiently high to clearly outweigh any commercial benefits for the defaulting
party if it fails to comply with the primary obligation. In setting the level in this
manner, attention is naturally given to either the ongoing loss suffered by the
aggrieved party from non-compliance, or the unjust enrichment for such
noncompliance by the defaulting party.

If a judicial penalty is a final determination, it should be rendered as an award

so it may be enforced. (356) Where evidentiary questions arise as to the amount of
the penalty, there is a question as to whether the tribunal can resolve disputes
as to the amount. A Paris Court of Appeals considered that this was an
enforcement issue outside of the tribunal's jurisdiction, (357) but this confuses
finalisation of the exact amount as awarded by the tribunal and the enforcement
of its payments. (358) If a judicial penalty is imposed, there may be an evidentiary
question as to when a page "1173" primary obligation has been completed so
that the penalty no longer applies. Where there is any dispute, it may be that the
defaulting party would need to prove that performance had been completed.
(359)

14.22. Application of Equity and The Assessment of Damages

Damages calculations should not vary when a tribunal is empowered to

determine the outcome et aqueo et bono, although there might be grounds to
discount or pro rata a calculation made. (360) This is discussed further in section
13.14

14.23. Security, Timing of Effect and Ongoing Supervision by the Tribunal

A tribunal might wish to order a security arrangement prior to the date for
honouring the final award. Generally speaking, such a determination should be
upheld. One issue discussed above is where a tribunal wishes to have some form
of ongoing scrutiny of the suitability of the relief granted. For example, a
tribunal might award a certain form of relief on the understanding that
concurrent declaratory orders will be obeyed. To the extent that they are not,
this may adversely affect the suitability of the damages amount. The fact that a
tribunal seeks to maintain a supervisory function should not of itself be a
ground for challenge, although the tribunal may well be. functus officio, hence
rendering the supervisory power inoperative. Some national courts have
considered that retaining jurisdiction does not contravene the functus officio
principle. (361) It was suggested that in some circumstances, the parties might be
able to commence a new arbitration on the basis that the failure to honour the
declaration was a breach of the arbitration agreement, hence giving a newly
constituted tribunal an opportunity to effectively alter the net damages within
the award.

Where there is a long-term remedy proposed, the tribunal should certainly

consider how to frame this to maximise enforceability by an appropriate court if
the page "1174" identity of that court can be anticipated. That should certainly
be known in some circumstances, such as an obligation to transfer a licence or
patent to a specified party after a defined time period.

14.22. Interest
There are a number of challenging issues in determining the entitlement and
amount of interest as a remedy. Key questions include whether interest is an
entitlement; if so, whether it is an element of damages or an addition to damages
entitlement; is it a procedural or substantive issue; what rate of interest is to
apply; what are the commencement and completion dates for the obligation;
should different rates apply to pre and post award interest; and, will interest be
calculated on a simple or compound basis?

Arbitral case law does not display any uniform solutions or unifying principles.
(362) Writing in 1996, Gotanda asserted that the various methods used by

tribunals in awarding interest have led to inconsistent and arbitrary awards.

(363) One of the many problems preventing a harmonised solution is that there

have been significant differences between investment and commercial

arbitration, in part because of different sources of law and the differences in the
inherent nature of the disputes. (364) In addition, various national systems and
courts have historically had mixed views about the awarding of interest. This is
understandable given religious opprobrium to money lending. Nevertheless,
over time virtually all systems have come to understand the importance of
borrowing for trade and investment purposes and in turn, the need to count the
value of time lost in truly awarding damages to a successful arbitral party. A
failure to provide for interest in times of high inflation may also be an incentive
towards breach. (365) An award of interest can also prevent unjust enrichment by
the party in breach. (366) page "1175" An entitlement to interest also promotes
prompt settlement. (367) From a practical point of view, the longer it takes to
complete an arbitration, the more significant that claims of interest will become.
Similarly, the longer any applicable limitation of action statute and the longer
the delay in commencing arbitration within that time frame, the more significant
pre-award interest could prove to be.

It is important to distinguish between cases where an entitlement to interest

arises from express contract provisions and separately where it arises as part of
some general applicable remedy. For example, interest may be a contractual
entitlement where there is late certification or non-certification in building
contracts. Here it flows from party autonomy and not a mandatory discretionary
remedy. Party agreement may still be overridden where it offends against
fundamental public policy, is against the parties' true intentions or involves
extreme prejudice or injustice to one party. (368) There may also be equitable or
good faith considerations that may militate against an award of interest such as
fraud, duress or bad faith. (369)

Because there are significant differences between legal families as to the

entitlement itself and the means of calculation, little clear guidance is contained
in many statutes and rules. Statutory provisions and rules will typically leave
the awarding of interest to the discretion of the tribunal. The UNCITRAL Model
Law does not contain a provision concerning interest, although some countries
which have adopted it have added a provision to that effect. (370) The NAFTA
Agreement also allows for ‘any applicable interest’. (371) Many national legal
systems allow for interest to be awarded in addition to damages to account for
the delay in receipt of appropriate compensation for default. Article 49 of the
Arbitration Act 1996 (UK) provides a broad discretion as to interest. There are
no express provisions in Swiss PILA and French NCCP. Many institutional rules
are also silent as to the question of interest. (372) Article 60 of the WIPO Rules
empowers the tribunal to award simple or compound interest. The tribunal is
given a broad discretion as to the rate and period. The AAA Arbitration Rules
provide a more constrained discretion in page "1176" Article 28.4 referring as it
does to an award of interest ‘as it considers appropriate, taking into
consideration the contract and applicable law’. (373)

The question is not necessarily resolved by an examination of the lex arbitri and
rules. It has been suggested that the right to interest ‘is an accepted
international legal principle’, (374) is part of the inherent authority of tribunals,
(375) and the entitlement may have become part of lex mercatoria. (376) The Iran-

US Claims Tribunal considered that it was customary to award interest even

where not expressly referred to in the compromise. (377) Nevertheless, the Claims
Tribunal did not manage to develop a consistent jurisprudence with regard to
interest. (378) It has been suggested that the express direction in various laws to
consider custom and trade usage may also be broad enough to encompass the
awarding of interest. (379) That may be dependent on what trade usage is
relevant to, as there is a major difference between considering usage to evaluate
a remedy within power and instead considering usage as the basis for the
remedy itself. That will be dependent on the wording of the mandate in the
relevant law. If permitted there will also be a factual question as to the norms of
the particular industry.

While some see the authority to determine interest as a procedural matter

governed by the lex arbitri and any arbitral rules, some cases have taken
differing approaches, looking to the applicable substantive law. The majority of
national systems would see interest as a substantive matter, being part of
damages. (380) English law distinguished between the entitlement to interest
which it saw as a page "1177" matter of substantive law and the rate, being seen
as a procedural matter. (381) Red-fern and Hunter assert that the right to interest
flows generally from the parties' contract or the applicable law. (382) They note
that in some regions, the power of tribunals to award interest is governed by the
law of the place of arbitration (383) but in others the law of the contract governs
this power, such as under German conflict of laws rules. (384) At times the
substantive law itself will directly deal with this. An example is Article 78 of the
Vienna Convention on the International Sale of Goods. (385) A pure conflicts
approach may have to consider the applicable conflicts rule as a first step.
Gotanda points to particular choice of law difficulties where differing legal
families treat the payment of interest as either substantive or procedural or
treat different aspects of the entitlement in differing ways. (386) Those applying a
substantive law conflicts approach have looked at the applicable law as chosen
by the parties; (387) or to the State with the closest connection. (388) In some cases
a tribunal might not consider such a choice to be effective, if it believes that the
relevant domestic law is not intended to apply to arbitrations. (389) A conflicts
approach would depend on whether the lex arbitri or applicable rules invited an
indirect choice of law via a conflicts methodology or a direct choice such as per
the ICC Rules. In either event, tribunals are likely to apply the same law on the
basis that an essential contract term is as to interest, although the possibility of
depecage remains.

If a tribunal is exercising a discretion to determine the applicable law, one issue

will be its impact on choice if one of the possibilities otherwise meritorious,
page "1178" leads to an entitlement to interest that could infringe on
international public policy in the country where execution is likely to take place.
(390) Born suggests that arbitrators in practice have generally looked to the

substantive law to determine the amount of interest. (391) This may also mean
that a tribunal might refuse to award interest if the applicable law is Islamic and
forbids such a remedy. (392) Some Islamic legal systems might proscribe the
entitlement to interest as Shari'a expressly prohibits the taking of interest or
(riba). (393) In Islamic countries it is generally the case that for interest to be
permitted, there needs to be some positive rule to that effect either directly or
indirectly via drawing express distinctions between commercial and non-
commercial matters. Some Islamic systems allow for interest in transactions
with foreigners whose laws permit such payments. (394) More often, Islamic
systems allow exceptions in favour of interest or allow for similar service or
administrative fees. (395) It is suggested that where the law of the arbitration
agreement or of the contract under which a dispute arises prohibits an award of
interest, a tribunal will have to follow that law. (396) Born questions the
consistency of such provisions with the New York Convention provisions in
Article III and V(1)(d). (397)

To the extent that the issues are substantive, there should again be less scope
for annulment or challenges to enforcement, although the fact that some
proscriptions are argued to be mandatory or matters of public policy leaves
open the potential for challenges. Consideration would also need to be given as
to whether an offending part of the award is severable or whether the entire
award could be tainted under a challenge to an order of interest. In regards to
the law of the place page "1179" of enforcement as a possible barrier to the
entitlement to interest, Redfern and Hunter argue it would too onerous to
expect an arbitral tribunal to take into account ‘the likely consequences of [an
award of interest] in a potential place of enforcement’; they suggest a tribunal
should only consider the issue if it is brought to its attention. (398) Even where a
tribunal is applying the substantive law of the contract, if the law of the place of
arbitration prohibits an award of interest, and especially if such provisions are
mandatory, Redfern and Hunter suggest the tribunal separate any award of
interest from the rest of the award, in case their award is attacked under the law
of the place where it was made. (399) Lew, Mistelis and Kröll make the same
suggestion for any award expected to be enforced in an Islamic country. (400) A
party can affect this issue with its claims, simply by seeking the time value of
money as part of its damages claim. If it seeks interest as well as such damages
in the alternative, it will still be for a tribunal to determine which approach is
the most appropriate in the circumstances. An important difference is when
interest is claimed as such, proof of loss is not required. If claimed as damage, it
must be subject to normal standards of proof and assessment. Claim by way of
damages may still be preferable where the innocent party has had to borrow in
mitigation at higher than standard interest rates owing to their own liquidity
problems, although the applicable law may not necessarily allow this.

Finally, while the following sections deal separately with questions of what
interest may be applied to, the applicable rate, questions of timing and whether
compound or simple interest should apply, the answers to these questions can
depend heavily on the view a tribunal takes as to the aim of the entitlement. A
full reparation approach to interest led the tribunal in Siemens to identify the
starting date, the rate and to award compound interest. (401)

14.24.1. What Is Interest Applied To?

Interest naturally applies to the damages awarded or an unpaid contractual

sum. Where pre-award damages are concerned, tribunals might be reluctant to
accept arguments that there should be interest on costs awarded as from the
time the costs were incurred. The situation might be different if the costs were
entitled to be claimed as damages. Interest may not necessarily flow on non-
pecuniary elements of damages as the tribunal might instead have identified an
amount that seems reasonable at the time of the award in the light of the likely
payment date.

page "1180"

14.24.2. The Appropriate Rate

There are difficulties in selecting the appropriate rate. Rates vary between
countries and within countries over time. Rates also vary depending on the
financial circumstances of borrowers and lenders. Differences will also arise
depending on whether one looks at depositor rates or borrower rates.
Borrowing rates may be more applicable for many businesses that have lost
liquidity through breach, but in some cases may be unfair if the relevant
claimant has sufficient capital. (402) Borrowing rates also depend on the credit
rating of the relevant party, which would be difficult to assess by a tribunal
unless there were comparable borrowings. In a modern world, people may
borrow in a range of jurisdictions, a range of currencies and can choose to hedge
against currency or interest fluctuations. A tribunal would not wish to make an
interest award that leads to windfall gains or undeserved losses but the
disparity in rates creates inevitable problems.

While the difficulty remains, some broad principles should apply. If the parties
agree on the rate of interest in their contract or during the arbitration, this
would be accepted by most tribunals, unless the rate was seen as an excessive
charge or inappropriate penalty by the relevant legal system. Excessive interest
rates as agreed between the parties may also be against public policy. In some
cases the parties have not agreed on a rate of interest for awards but instead for
payments contemplated under the contract. In some cases a tribunal will utilise
this for award amounts as well. (403) There may be exceptions and conflicting
instruments may need to be reconciled. For example, in the CMI case, a tribunal
considered that the express broad discretion in the Iran-US Claims Settlement
Declaration allowed the tribunal to depart from the express choice of the
parties. (404) In some cases a rate agreed to by the parties may be higher than
that stipulated in the applicable law. Where a government ordains a ceiling, that
is an aspect of domestic public policy which could have some impact in
annulment or enforcement actions. In ICC Case No. 8874, a tribunal acting as
amiable compositeur considered that principles of equity meant that a
contractually agreed monthly interest rate of 1.4% was too high. (405) It is
questionable on what basis a tribunal should ignore such an agreement of the
parties. (406) Where discretion applies, uniformity does not necessarily need to
apply to the exact rate, but ideally, to the method of calculating it. A uniform rate
applied to people in different countries with different credit ratings may not in
fact be truly equal or consistent as a matter of policy. If the aim is to compensate
the relevant party for the loss of the use of the money, there may even be a
difference in page "1181" choice of type of rate depending on whether the party
would have invested the money in a financial instrument, used it as capital in its
business or used it to obviate the need for borrowings otherwise made.

Some lex arbitri specifically direct the determination of a reasonable rate. (407) If
a rate is not directed, there is a conflicts issue as to whether the rate of interest
is determined by the rate applicable in the forum, the rate under the applicable
contractual law, the rate at the place of payment or such other reasonable rate
as determined by the tribunal. (408) There have been cases that have applied the
statutory interest rate in the country whose currency is the currency of
payment. (409) Another case looked to the interest rate customarily awarded by
courts in the place where the award was likely to be enforced. (410) Another
approach is the habitual rate in the currency of the debt. An alternative is the
currency of the award. Another possibility is to look at interest rates in the
domicile of the aggrieved party as this is where that party could most easily
invest the money. There are cases covering each approach. (411) This may be
impacted upon by whether the tribunal sees it as a question of procedure or
substance. One of the problems with a conflicts approach is that given the
disparity of interest rates around the world, in part because of floating
currencies, no single national system will necessarily meet the reasonable
expectations of the parties at the time of their initial contract.

Lew, Mistelis and Kröll suggest that parties should provide for the rate of
interest on monies due in their agreement, but assert that if no such agreement
exists, tribunals should look to the law governing the arbitration or the
arbitration rules. (412) Redfern and Hunter assert that most applicable laws leave
questions such as the rate, start date and currency of interest to be awarded to
the discretion of the tribunal. (413) Ordinarily, tribunals will invite parties to
make submissions and page "1182" present evidence on these questions.
Gotanda advocated a two-step approach for resolving interest claims with a
view to promoting consistency, while still allowing for appropriate flexibility.
The first step would involve a determination of whether the parties expressly
included a provision as to payment of interest or specified the law applicable to
the merits. In either case, the tribunal should resolve interest issues in
accordance with that specification. Where the contract is silent or sufficiently
ambiguous, then the second step would involve resolution on the basis of
certain presumptions namely ‘(1) the debtor is liable for the payment of
interest; (2) interest runs from the date of default; and (3) interest accrues at a
rate corresponding to that of a commonly used savings vehicle in the country of
the currency in which payment is to be made and it is compounded quarterly’.
(414)

Some tribunals look to general principles of law in awarding interest, (415)

although those relying on general principles of international law would find it
difficult to identify an exact rate of interest and any timing rule that has become
accepted in that way. The same would be true in regard to arguments based on
entitlement under lex mercatoria. Some tribunals simply look at what is
considered to be fair and reasonable in the circumstances. (416) In ICC Case No.
8521, the tribunal considered that the trend in practice and scholarly analysis is
to see an arbitrator having broad freedom in fixing rates and that the tribunal is,
therefore, not bound by conflicts of laws rules or by a need to select a particular
national system. (417) In McCollough & Co, the Iran-US Claims Tribunal applied a
reasonableness standard in the light of all surrounding circumstances including:

(i) any pertinent contractual stipulations (which, when they exist, are usually
followed for the determination of the rates);
(ii) the rules and principles of the law applicable to the contract; page "1183"
(iii) the nature of the facts generating the damage;
(iv) the nature or level of the compensation awarded, particularly if it extends
to the lost profit or includes a profit in the costs to be reimbursed;
(v) the knowledge that the defaulting party could have had of the financial
consequences of its default for the other party;
(vi) the rates in effect on the markets concerned; and
(vii) the rates of inflation. (418)

A restitutionary approach would look at the rate the successful party might have
earned on the funds if paid in time. (419) However, a modern globalised world
where funds can be invested anywhere, chasing best interest rates and at times
speculating on currency movements, makes this less of a logical corollary of a
restitutionary model. In Sylvania, the tribunal considered that interest should be
‘based approximately on the amount that the successful claimant would have
been in a position to have earned if it had been paid in time and had the funds
available to invest in a form of commercial investment in common use in its own
country’. (420)

Secomb suggests a three-step approach to promote certainty and consistency,

namely:

(1) Apply the parties' agreement, if it deals with interest;

(2) If not, the parties should be able to prove their actual borrowing costs; and
(3) If they cannot or do not, a commonly used commercial rate for the relevant
currency should be applied. (421)
The corollary of the third principle is that arbitrators should not simply apply
statutory rates from a national law and should not simply take it upon
themselves to determine a reasonable rate from a range of sources.

The UNIDROIT Principles 2004 Article 7.4.9 suggest that the rate should be ‘the
average bank short-term lending rate to prime borrowers prevailing for the
currency of payment at the place for payment, or where no such rate exists at
that place, then the same rate in the State of the currency of payment’. Article
7.4.10 indicates that ‘unless otherwise agreed, interest on damages for non-
performance of non-monetary obligations accrues as from the time of non-
performance’. (422) The Principles of European Contract Law 1998 Article 9:508
adopts a similar approach to the UNIDROIT Principles. An EC Directive (423) seeks
to set interests page "1184" rates at a level significant enough to discourage
delay in payment under commercial transactions. This may impact upon
otherwise applicable law in the European context. (424) This goes beyond
restutitionary notions.

14.24.2.1. Simple or Compound Interest

If the most widely accepted rationale for the awarding of interest is to put the
successful party in the position it would have been in if the contract had been
validly performed, then there is an argument in favour of the award of
compound interest as a matter of course. Compound interest can be justified
from a restitutionary perspective on the basis that it equates to opportunity
costs. (425) Compound interest would also be a natural element of damage if the
funds underlying the damages award had been earned or saved under the
contract if it had been performed. (426) Compound interest is not intended to be
punitive but instead, to provide commercially realistic restitution as if payments
had all been made promptly, enabling further investment or other productive
use as and when received.

Gotanda notes the potential problem of proof for a claimant to show how it
normally deals with cash surpluses and at what rates and the proper way to deal
with differing situations where excess cash is reinvested, paid out as dividends
or otherwise returned to shareholders. (427) Fénéchal argues that under a
commercially realist compensation principle, the opportunity cost exercise
should aim at making the injured party whole, should adjust for the passage of
time, should reflect market reality and go beyond pure inflation based
approaches to identify a fair market rate taking into account risk. He suggests
that cost of capital could be a proxy, noting that this could be a problem for non-
listed companies and for some regions. Compound interest is recommended,
using an annual basis as a conservative estimate. (428) This would apply both to
substantive damages and also any award as to costs.

page "1185"

Nevertheless, many tribunals would restrict themselves to the award of simple

interest. In part this is because there was previously the view that compound
interest was not acceptable under customary international law, a principle if
correct, that would be most influential in investment arbitration. (429) In addition,
the ability to accurately compound interest to equate to what would really have
applied if the contract was not breached is only possible after tax liabilities have
been met. Hence, a basic compounding formula will typically be unrealistic or
difficult to determine. The entitlement to compound interest on top of damages
may also be affected by whether the aggrieved party is itself suffering damage
by way of compound interest. (430) However, compound interest has also become
more acceptable over time as it more readily replicates the real commercial
environment and hence meets the reasonable expectations of commercial
parties. Redfern and Hunter explain this growing acceptance of the
appropriateness of awarding compound interest by suggesting that
international tribunals are increasingly realising that simple interest may not
fully compensate a party for the loss suffered by them. (431) Redfern and Hunter
contend, ‘sources of international law are no clearer’. (432) Nevertheless, the
authors claim that the award of compound interest is becoming ‘more common’
and ‘is no longer an exception to the rule’. (433) They cite the cases of Santa Elena
(434) and Wena Hotels. (435) In Santa Elena the tribunal found ‘no uniform law has

emerged from the practice in international arbitration as regards the

determination of whether compound or simple interest is appropriate in any
given case’. Wena Hotels found compound interest to be generally appropriate
page "1186" in modern arbitration. These cases have since been followed by
many tribunals. (436) Brower and Sharpe do not necessarily criticise the policy
outcome, but provide a forceful critique as to the way these more expansive
principles have been developed by investment arbitration tribunals through
questionable use of comments in earlier cases. (437) Most problematic is reliance
on the Aminoil Awards where a tribunal awarded USD 83,000,000 in net
damages plus USD 96,000,000 by way of 7.5% interest and 10% as per level of
inflation, each compounded annually. Brower and Sharpe argue that it is sui
generis, using compounded levels of inflation and compound interest to
effectively compensate Aminoil for an unlawful expropriation while purporting
to hold that the expropriation was lawful. (438)

Recent arbitration legislation in some common law countries expressly provides

for an award of compound interest. (439) However, there remain common law
countries in which the power to award compound interest is limited (440) or
varies within the country from one State or region to the next. (441) A number of
arbitral rules expressly allow for compound interest. (442) In civil law
jurisdictions, arbitral tribunals often have the power to award a ‘statutory’
interest rate, which is simple interest at a rate statutorily defined, (443) however
not all civil law jurisdictions allow for this to occur. (444)

page "1187"

If compounding is to occur, there is also a need to consider the relevant time

period on which to compound. There is no inherently correct time. Annual
compounding would be typical and is a conservative approach. (445)

14.24.2.2. Timing of Interest

Most legal systems would allow interest to accrue from the date of breach or
from when a debt becomes payable. ICC Case No. 6281 calculated interest from
the date of the breach. (446) Some relevant statutory periods provide an express
starting point. (447) Some only allow interest to accrue from the date of a claim.
(448) Some only allow for interest from the date of the award. (449) Where the

relevant time period is from that of the breach or default, there may be
evidentiary questions as to when that has arisen. Timing may also be affected by
issues of waiver and acquiescence or agreed extensions as to the time of
performance. Another approach is to look at the date proceedings commenced.
While loss of funds occurred before-hand, in some cases a tribunal might feel
that the claimant has taken too long to make a claim and is hence partially
responsible for loss of the use of its money. (450)

Timing can also be dependent on other contractual provisions. For example, in a

construction dispute, if an engineer fails to adequately certify entitlements
which are then confirmed by an arbitral tribunal, the engineer's decision is not
necessarily one of fault, but is simply part of an escalating adjudicatory process.
(451) In some cases there may be a need to make a specific demand to commence

entitlement to interest. This may flow from the substantive law or from general
principles of reasonable entitlement. (452) This can be because the applicable law
might find no entitlement to any relief until certain notice of non-performance
has been provided. Where the EU Directive of 2000 is concerned, this seeks to
specify the relevant dates and does not require notice. There is also a need to
consider the end date. (453)

page "1188"

Domestic litigation systems also commonly differentiate between prejudgment

and post-judgment interest. This is because pre-judgment interest involves an
analysis by the adjudicator of a fair and reasonable rate in all the circumstances.
Pre-award interest should be further divided into true arbitral interest and post-
request and pre-award interest. Post-judgment interest is not scrutinised by the
judicial adjudicator and is typically at a statutory rate set by government that
may for policy reasons be set at a high level to encourage prompt payment. An
arbitral tribunal will often simply concern itself with finding the appropriate
rate for both pre and post-award interest. However, Redfern and Hunter
contend that it is open to arbitrators to set a post-award interest rate they deem
appropriate, which is ‘often the rate that would apply to a judgment in the
country where the award is made’. (454) At the stage of enforcement, when the
award is enforced in a court judgment, a post-award interest rate may be
replaced by the rate applicable to such judgments. (455) If a tribunal denies
entitlement to pre-award interest then interest will flow from the date of the
award. This is more likely to arise where the claimant has itself acted in a way
which might undermine its entitlement to earlier interest. (456)

It is not appropriate to consider that a tribunal is functus officio and hence

unable to grant post-award interest. (457) Any ongoing remedy, including
declaratory and injunction relief can postdate the tribunal's termination of office
as can be the case with payment of damages themselves.
page "1189"

1 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative

International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 649. See also Lawrence Collins (ed.), Dicey, Morris & Collins on The
Conflict of Laws, 14th edn (London: Sweet & Maxwell, 2006), 177–206.
2 Lawrence Collins (ed.), Dicey, Morris & Collins on The Conflict of Laws, 14th edn

(London: Sweet & Maxwell, 2006), 192.

3 Ibid., 180.
4 See, e.g., Felix Praendl, ‘Measure of Damages in International Commercial

Arbitration’, Stanford Journal of International law 23, no. 1 (1987): 263; Andreas
Reiner, ‘Burden and General Standards of Proof, in Alan Redfern et al., ‘The
Standards and Burden of Proof in International Arbitration’, Arbitration
International 10, no. 3 (1994): 317, 328.
5 Where non-pecuniary remedies are concerned, some have already have been

considered in the context of provisional measures discussed in section 8.2

above. This chapter only deals with final remedies of that nature.
6 Daniel Friedmann, ‘Economic Aspects of Damages and Specific Performance

Compared’, in Contract Damages: Domestic and International Perspectives, ed.

Djakhongir Saidov & Ralph Cunnington (Oxford: Hart Publishing, 2008), 65.
7 L. Fuller & W. Perdue, ‘The Reliance Interest in Contract Damages’, Yale Law

Journal 46, no. 1 (1936): 52.

8 These issues were discussed in sections 13.6 and 13.7.
9 In some cases there may be issues about uncertainty in drafting and whether

certain heads of damage have been successfully excluded.

10 Paul-A Gelinas, ‘General Characteristics of Recoverable Damages in
International Arbitration’, in Evaluation of Damages in International Arbitration,
Dossier of the ICC Institute of World Business Law, ed. Yves Derains & Richard H.
Kreindler (Paris: ICC Publishing, 2006), 17. See also ICC Case No. 9950, Final
Award 2001, the UNIDROIT Principles: New Developments and Applications,
2005 Special Supplement – ICC Int'l Ct of Arb Bull 77–9.
11 Felix Praendl, ‘Measure of Damages in International Commercial Arbitration’,

Stanford Journal of International Law 23, no. 1 (1987): 296. For a case where the
parties unsuccessfully sought to bar set-off rights see USSR Project Arbitration
(France v. Yugoslavia) ICC Case No. 3540, Award of 3 October 1980, Journal de
Droit International 4 (1981): 914, excerpted in Pieter Sanders (ed.), Yearbook
Commercial Arbitration, vol. VIII (The Hague: Kluwer Law International, 1982),
124. An example of a contractual limitation is cl. 17.6 of the General Conditions
of Contract for Construction, FIDIC 1998.
12 Judgment of 28 June 1988, Total Chine v. E.M.H. et G.S.C, Cour d'appel de Paris,

Revue de l'Arbitrage (1989): 328.

13 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 649.
14 This is consistent with the traditional English approach and conflicts rules

Kremezi v. Ridgway [1949] 1 All ER 662, 664. More recently, English courts have
looked to the substantive law to determine what losses may be claimed but look
to procedural law to determine assessment. Felix Praendl, ‘Measure of Damages
in International Commercial Arbitration’, Stanford Journal of International law
23, no. 1 (1987): 263, 268.
15 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2479.

16 ibid.
17 Ibid., 2483 citing Dreis & Krump Mfg Co. v. Int'l Assoc of Machinists & Aerospace

Workers, etc., 802 F. 2d 247 (7th Cir 1986); Engis Corp. v. Engis ltd, 800 F. Supp.
627, 632 (N.D. 111. 1992).
18 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2482.

19 Anne van Aaken, ‘Remedies in International Law and Investment Law’, in

International Investment Law and Comparative Public Law, ed. Stephen W. Schill
(Oxford: Oxford University Press, 2010), 731.
20 Where courts are concerned, notwithstanding that matters of procedure are

governed by the lex fori, Cook suggests that in private international law, a court
should apply as much of the foreign law as it can without unduly
inconveniencing itself. Walter Wheeler Cook, Logical and Legal Bases of the
Conflict of Laws (Cambridge: Harvard University Press, 1942), 154–193. See also
Joanna Bird, ‘Choice of Law’, in Restitution and the Conflict of Laws, ed. Francis
Rose (Oxford: Mansfield Press, 1995), 84, 166.
21 Chorzów Factory (Poland v. Germany), PCIJ 1926 Merits, Judgment No. 13,

1928, PCIJ Series A No. 17, 29, available at <www.worldcourts.com>.

22 CNS Gas Transmission Company v. The Argentine Republic (ICSID Case No.

ARB/01/8, Award, 12 May 2005) 406.

23 ILC Articles, Art. 35.
24 Carole Malinvaud, ‘Non-Pecuniary Remedies in Investment Treaty and
Commercial Arbitration’, in 50 Years of the New York Convention, ICCA
International Arbitration Conference, ICCA Congress Series no. 14, Dublin 2009),
ed. Albert Jan van den Berg (Paris: ICC Publishing, 2009), 209, 225.
25 Award of 19 January 1977, Texaco Overseas Petroleum Co v. The Government

of the Libyan Arab Republic and others in Pieter Sanders (ed.), Yearbook
Commercial Arbitration, vol. IV (The Hague: Kluwer Law International 1979),
177, 184.
26 1923 Spain/United Kingdom 2 RIAA 615 at 722.
27 NAFTA Art. 1135.
28 Article 26(8).
29 Such a formulation is also utilised in bilateral investment treaties. See, e.g.,

the Draft US Model BIT, 5 February 2004, Art. 34.

30 Christoph Schreuer, The ICSID Convention: A Commentary (Leiden: Cambridge

University Press, 2001), 1115; Christoph Schreuer, ‘Non-Pecuniary Remedies in

ICSID Arbitration’, Arbitration International 20, no. 4 (2004): 325.
31 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 532.

32 Christine D. Gray, Judicial Remedies in International Law (Oxford: Clarendon

Press, 1987), 12.

33 Ibid., 15.
34 Article 1135 of Chapter 11 of the NAFTA provides that although a tribunal
may award ‘restitution of property’, such awards ‘shall provide that the
disputing party may pay monetary damages and any applicable interest in lieu
of restitution’.
35 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 532.

36 Brigitte Stern, Trois Arbitrages, Un Même Problème, Trois Solutions: Les

Nationalisations Pétrolières Libyennes Devant l'Arbitrage International' (1980)

Revue de l'Arbitrage (1980): 1, 36. See also Amco Asia Corp. and others v. The
Republic of Indonesia (ICSID Case No. ARB/81/1, Award on the Merits, 21
November 1984) 202 in 24 International Legal Materials, 24 (1985) 1022.
37 (1979) 53 ILR 297.
38 Extracted and translated in Carole Malinvaud, ‘Non-Pecuniary Remedies in

Investment Treaty and Commercial Arbitration’, in 50 Years of the New York

Convention, ICCA International Arbitration Conference, ICCA Congress Series no.
14, Dublin 2009), ed. Albert Jan van den Berg (Paris: ICC Publishing, 2009), 209,
222.
39 Carole Malinvaud, ‘Non-Pecuniary Remedies in Investment Treaty and
Commercial Arbitration’, in 50 Years of the New York Convention, ICCA
International Arbitration Conference, ICCA Congress Series no. 14, Dublin 2009),
ed. Albert Jan van den Berg (Paris: ICC Publishing, 2009), 209, 228–229.
40 1929 United States/Cuba 2 RIAA 913.
41 Carole Malinvaud, ‘Non-Pecuniary Remedies in Investment Treaty and
Commercial Arbitration’, in 50 Years of the New York Convention, ICCA
International Arbitration Conference, ICCA Congress Series no. 14, Dublin 2009),
ed. Albert Jan van den Berg (Paris: ICC Publishing, 2009), 209, 229.
42 The parties reached a damages settlement after the award and specific

performance was never accepted. Carole Malinvaud, ‘Non-Pecuniary Remedies

in Investment Treaty and Commercial Arbitration’, in 50 Years of the New York
Convention, ICCA International Arbitration Conference, ICCA Congress Series no.
14, Dublin 2009), ed. Albert Jan van den Berg (Paris: ICC Publishing, 2009), 209,
222.
43 Antoine Goetz et. al. v. The Republic of Burundi (Decision of 2 September 1998)

ICSID Review, 15 (2000) 457, paragraph 133.

44 The suggestion of creation of a new free trade zone was based on an
agreement of the parties and not necessarily an example of awarding non-
pecuniary relief. Christoph Schreuer, ‘Non-pecuniary Remedies in ICSID
Arbitration’, Arbitration International 20, no. 4 (2004): 325, 330.
45 Regulation (EC) No. 593/2008 of the European Parliament and of the Council

of 17 June 2008 on the law applicable to contractual obligations (Rome I). A

provision with similar effect was set out at Art. 10.1(c) of the Convention on the
Law Applicable to Contractual Obligations, opened for signature in Rome on 19
June 1980: ‘The law applicable to a contract by virtue of Articles 3 to 6 and 12 of
this Convention shall govern in particular:… (c) within the limits of the powers
conferred on the court by its procedural law, the consequences of breach,
including the assessment of damages in so far as it is governed by rules of law…’:
United Nations Convention on Contracts for the International Sale of Goods,
opened for signature 11 April 1980, 1489 UNTS 3, Art. 28: ‘If, in accordance with
the provisions of this Convention, one party is entitled to require performance
of any obligation by the other party, a court is not bound to enter a judgement
for specific performance unless the court would do so under its own law in
respect of similar contracts of sale not governed by this Convention’.
46 Alexis Mourre, ‘Judicial Penalties and Specific Performance in International

Arbitration’, in Interest, Auxiliary and Alternative Remedies in International

Arbitration, Dossiers ICC Institute of World Business Law, ed. Filip de Ly &
Laurent Lévy (Paris: ICC Publishing, 2008), 67, referring to similar issues with
regard to the Rome Convention. In particular, arbitration agreements are
expressly excluded from the scope of the Regulation of Art. 1.2(e).
47 This is not a mandatory provision per s. 4 of Schedule 1 and hence may be

overridden by the parties.

48 B. Friedmann, ‘Economic Aspects of Damages and Specific Performance
Compared’, in Contract Damages: Domestic and International Perspectives, ed.
Djakhongir Saidov & Ralph Cunnington (Oxford: Hart Publishing, 2008), 65, 69.
49 Ibid., 65.
50 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 531.

51 See, e.g., Art. 1184(2) French Civil Code; Switzerland Art. 97 Code des
Obligations.
52 The powers were recognised by tribunals in Award on Merits of 10 October

1973, British Petroleum Exploration Company (Libya) Ltd v. The Government of

the Libyan Arab Republic, in Pieter Sanders (ed.), Yearbook Commercial
Arbitration, vol. V (The Hague: Kluwer Law International, 1979), 143, 150;
Award of 12 April 1977, Libyan American Oil Company (LIAMCO) v. The
Government of the Libyan Arab Republic, in Pieter Sanders (ed.), Yearbook
Commercial Arbitration, vol. VI (The Hague: Kluwer Law International, 1981) 89,
105, however restitution was not awarded in these cases due to practical
impossibility. The US Federal Arbitration Act and the Uniform Arbitration Act
(UAA) do not expressly specify the remedies available in international
arbitrations taking place in the US; nonetheless US courts have confirmed the
power of arbitrators to award specific performance, even where this remedy is
not specified in the arbitration agreement. See, e.g., Brandon v. MedPartners Inc,
203 FRD 677, 686 (SD Fla 2001). Also, there are examples where State courts in
common law countries have upheld tribunal orders for specific performance, for
example, In re Staklinski 6 N.Y.2d 159, 160 N.E.2d 78, 188 N.Y.S.2d 541 (1959)
where the granting of specific performance of an employment contract was held
not to be contrary to public policy. Julian D.M. Lew, Loukas A. Mistelis & Stefan
M. Kröll, Comparative International Commercial Arbitration (The Hague: Kluwer
Law International, 2003), 650.
53 Articles 28, 46 and 62.
54 Article 7.2.2 and 7.2.3.
55 Edwin Peel, The law of Contract, 12th ed, (London: Sweet & Maxwell, 2007),

1107.
56 Christine Chappuis, ‘A Comparative Overview on Performance as a Remedy: A

Key to Divergent Approaches’, in Performance as a Remedy: Non-Monetary Relief

in International Arbitration, ed. Michael E. Schneider & Jaochim Knoll (New
York: Juris, 2011), 70 citing Lord Hoffmann in Co-operative Insurance Society ltd
v. Argyll Stores (Holdings) ltd [1997] C.L.C. 1114, 1117.
57 Carole Malinvaud, ‘Non-Pecuniary Remedies in Investment Treaty and
Commercial Arbitration’, in 50 Years of the New York Convention, ICCA
International Arbitration Conference, ICCA Congress Series no. 14, Dublin 2009),
ed. Albert Jan van den Berg (Paris: ICC Publishing, 2009), 209, 213. This position
is an application of the principle nemo praecise cogi potest ad factum.
58 B. Friedmann, ‘Economic Aspects of Damages and Specific Performance
Compared’, in Contract Damages: Domestic and International Perspectives, ed.
Djakhongir Saidov & Ralph Cunnington (Oxford: Hart Publishing, 2008), 65.
59 This viewpoint is noted by Lew, Mistelis Kröll: Julian D.M. Lew, Loukas A.

Mistelis & Stefan M. Kröll, Comparative International Commercial Arbitration

(The Hague: Kluwer Law International, 2003), 650-651, and see Troy E. Elder,
‘The Case against Arbitral Awards of Specific Performance in Transnational
Commercial Disputes’, Arbitration International 13, no. 1 (1997): 1, 7.
60 Michael E. Schneider, ‘Non-monetary Relief in International Arbitration:
Principles and Arbitration Practice’, in Performance as a Remedy: Non-Monetary
Relief in International Arbitration, ed. Michael E. Schneider & Jaochim Knoll
(New York: Juris, 2011), 5.
61 Ibid.
62 This power is discussed in section 14.21.
63 Christoph Schreuer, ‘Non-Pecuniary Remedies in ICSID Arbitration’,
Arbitration International 20, no. 4 (2004): 325, 332.
64 F. Bellivier & R. Sefton-Green, ‘Force Obligatoire et Exécutionen Nature du

Contrat Men Droit Français et Anglais: Bonnes et Mauvaises Surprises du

Comparatitne’, in he Contrat au Début du XXème Siècle Etudes Offertes à J Ghestin,
ed. G. Goubeaux et al. (LGDJ, 2001), 91 at 111 cited and translated in Carole
Malinvaud, ‘Non-Pecuniary Remedies in Investment Treaty and Commercial
Arbitration’, in 50 Years of the New York Convention, ICCA International
Arbitration Conference, ICCA Congress Series no. 14, Dublin 2009), ed. Albert Jan
van den Berg (Paris: ICC Publishing, 2009), 209, 214. See also Troy E. Elder, ‘The
Case against Arbitral Awards of Specific Performance in Transnational
Commercial Disputes’, Arbitration International 13, no. 1 (1997): 1, 8.
65 Troy E. Elder, ‘The Case against Arbitral Awards of Specific Performance in

Transnational Commercial Disputes’, Arbitration International 13, no. 1 (1997):

1, 24.
66 Cases where the remedy of specific performance has been deemed
inappropriate include, e.g., ICSID Final award, 31 March 1986 and Interim
award, 24 October 1984, Liberian Eastern Timber Corporation (‘LETCO’) v. The
Government of the Republic of Liberia, 26 ILM 647 (1987), Yearbook of
Commercial Arbitration, vol. XIII, 35 (1988); Zurich arbitration award in Case
No. 273/95, 31 May 1996, raw material processor (Hungary) and processing
Group (Argentina) v.raw material seller (Russian Federation), Yearbook of
Commercial Arbitration, vol. XXIII, 128 (1998); ICC Award No. 8032 (1995),
Yearbook of Commercial Arbitration, vol. XXI, 113 (1996).
67 Libya American Oil Company (LIANCO) Ltd v. Libyan Arab Republic 20 ILM 1

(1981) 63; BP Exploration Company (Libya) Limited v. Libyan Arab Republic, 52

ILR 297 (1974) 354.
68 Occidental Petroleum Corporation and Occidental Exploration and Production

Company v. The Republic of Ecuador, ICSID Case No. ARB/06/11, Decision on

Provisional Measures, 17 August 2007, §§ 82–84.
69 Pabalk Ticaret Ltd v. Norsalor SA (Turkey v. France) ICC Case No. 3131, Award

of 26 October 1979, 1985 Yearbook of Commercial Arbitration 109; upheld in

Norsalor SA v. Pabalk Ticaret Ltd, Oberster Gerichtschof (Austrian Supreme
Court) Judgment of 18 November 1982, 1984 Yearbook of Commercial
Arbitration 159.
70 Avena and Other Mexican Nationals (Mexico v. United States) Judgment, ICJ

Reports 2004, 23.

71 Francesca Mazza, ‘International Chamber of Commerce (ICC)’, in Performance

as a Remedy: Non-Monetary Relief in International Arbitration, ed. Michael E.

Schneider & Jaochim Knoll (New York: Juris, 2011) 149.
72 Carole Malinvaud, ‘Non-Pecuniary Remedies in Investment Treaty and
Commercial Arbitration’, in 50 Years of the New York Convention, ICCA
International Arbitration Conference, ICCA Congress Series no. 14, Dublin 2009),
ed. Albert Jan van den Berg (Paris: ICC Publishing, 2009), 209, 221.
73 Nigel Blackaby et al., Hunter, Redfern and Hunter on International Arbitration,

5th edn (Oxford: Oxford University Press, 2009), 535. Arbitration legislation
often makes express provision for the granting of declaratory relief. See, e.g., the
English Arbitration Act 1996 s. 48(3).
74 Case Concerning the Factory at Chorzów, PCIJ Series A, No. 13, 20 (PCIJ 1928).
75 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 652; they cite s. 48(3) of the English Arbitration Act as an example of a
law providing for declaratory relief.
76 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 535.

77 Saudi Arabia v. Arabian American Oil Company (Aramco) (1963) 27 ILR 117,

at 145. As no party wished to jeopardise their business relationship, the tribunal

was asked to make a declaration that the agreement had been breached without
claiming damages.
78 Saudi Arabia v. Arabian American Oil Company (Aramco), 27 ILR 117 (1963).
79 Ibid., 145.
80 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 652.
81 Michael E. Schneider, ‘Non-Monetary Relief in International Arbitration:
Principles and Arbitration Practice’, in Performance as a Remedy: Non-Monetary
Relief in International Arbitration, ed. Michael E. Schneider & Jaochim Knoll
(New York: Juris, 2011), 28.
82 Ibid., 30.
83 New Code of Civil Procedure (France) Art. 31; German Code of Civil
Procedure (ZPO) s. 256; Swiss Code of Civil Procedure Art. 59(2)(a) which
prescribes in a general manner that a claimant must have a justified interest in a
legal action.
84 Michael E. Schneider, ‘Non-Monetary Relief in International Arbitration:
Principles and Arbitration Practice’, in Performance as a Remedy: Non-Monetary
Relief in International Arbitration, ed. Michael E. Schneider & Jaochim Knoll
(New York: Juris, 2011), 30.
85 Ibid.
86 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2481 cites Adamas Mgt & Services Inc v. Aurado Energy Inc,
Yearbook of Commercial Arbitration, vol. XXX, 479 (New Brunswick QB 2004)
(2005); IPOC Int'l Growth Fund Ltd v. LV Finance Group Ltd, Civil Appeal No. 30 of
2006 (B .V.I. Court of Appeal, 18 June 2007); LV Finance Group Ltd v. IPOC Int'l
Growth Fund Ltd [2006] Bda LR 69 (Bermuda Commercial Court); Ethiopian Oil
Seeds & Pulses Exp Corp v. Rio del Mar Feeds Inc [1990] 1 Lloyd's Rep 86 (QB);
NSW Racing v. TAB [2002] NSWSC 742 (New South Wales Supreme Court);
Enron Corp and Ponderosa Assets LP v. Argentine Republic, Decision on
Jurisdiction, ICSID Case No. ARB/01/3 (14 January 2004) para. 81, available at
<http://icsid.worldbank.org>.
87 For example, Rule 43 AAA Rules as interpreted in Island Creek Coal Sales Co v.

City of Gainesville, 729 R2d 1046, 1049 (6th Cir 1984).

88 Most international and institutional rules make clear that the arbitration

clause is not to be taken as excluding the jurisdiction of national courts to make

orders for interim measures of protection: see, e.g., ICC Rules 2012 Art. 28;
UNCITRAL Rules 2010 Art. 26.
89 Trail Smelter Case (United States v. Canada), Award, 16 April 1938 and 11

March 1941, 3 RIAA 1905.

90 Rainbow Warrior RIAA vol. XX (1990) 217 at 270.
91 Enron Corp and Ponderosa Assets IP v. Argentine Republic ICSID Case No.

ARB/01/03 Decision on Jurisdiction, 14 January 2004, available at

<www.asil.org/ilib/Enron.pdf>.
92 Enron Corp and Ponderosa Assets LP v. Argentine Republic ICSID Case No.

ARB/01/03 Decision on Jurisdiction, 14 January 2004, para. 76.

93 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 535–536.

94 Laws which expressly or impliedly provide the power to fill a gap in, or

amend, a contract include Bulgaria, Arbitration Act 1988/1993 Art. 1(2);

Netherlands, CCP Art. 1020(3); Sweden, Arbitration Act Art. 1(2); CEPANI Rules
Art. 7(4); and query Art. 7(3) UNCITRAL Model Law, cited in Kröll, Ergänzung
und Anpassung von Verträgen durch Schiedsgerichte. Kröll concludes at 305 that
the English and German arbitration laws also provide the tribunal with this
power.
95 Bin Cheng, General Principles of Law as Applied by International Courts and

Tribunals (reprinted edn, Cambridge University Press, 2006), 118–119. In

Himpurna California Energy Ltd v. PT (Persero) Perusahaan Listruik Negara,
UNCITRAL Award of 4 May 1999, (2000) Yearbook of Commercial Arbitration,
vol. XXX, 13, at 58 et seq, rebus sic stantibus was one basis upon which the
Indonesian State parties sought to change their payment obligations under
various independent power project contracts following the South-East Asian
economic crisis of the late 1990s.
96 ‘Rules for Adaptation of Contracts’ (1978) ICC Publication No. 326.
97 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 539–540 referring to ICC Publication

No. 326.
98 See W. Laurence Craig, William W. Park & Jan Paulsson, International
Chamber of Commerce Arbitration, 3rd edn (Oxford: Oxford University Press,
2000), 709 for an outline of the rules and the critique by Derains.
99 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 537.

100 Ibid., 539. See also Himpurna California Energy ltd v. PT (Persero) Perusahaan
listruik Negara, UNCITRAL Award of 4 May 1999, (2000) Yearbook of
Commercial Arbitration, vol. XXV, 13, at 61 and 62, where an arbitral tribunal
commented ‘… The arbitrators cannot usurp the role of government officials or
business leaders. They have no political authority, and no right to presume to
impose their personal view of what might be an appropriate negotiated solution.
Whatever the purity of their intent, arbitrators who acted in such a fashion
would be derelict in their duties, and would create more mischief than good. The
focus of the Arbitral Tribunal's inquiry has been to ascertain the rights and
obligations of the parties to the particular contractual arrangements from which
its authority is derived’. See generally James Otis Rodner, ‘Hardship under the
UNIDROIT Principles of International Commercial Contracts’, in Global
Reflections on International law, Commerce and Dispute Resolution: liber
Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al. (ICC Publishing,
2005), 677.
101 UNIDROIT Principles Art. 6.2.2.
102 Davis Contractors ltd v. Fareham (JVC [1956] AC 696, 729.
103 James Otis Rodner, ‘Hardship under the UNIDROIT Principles of
International Commercial Contracts’, in Global Reflections on International law,
Commerce and Dispute Resolution: liber Amicorum in Honour of Robert Briner, ed.
Gerald Aksen et al. (ICC Publishing, 2005), 678.
104 ICC Hardship Clause 2003 – ICC Publication 650.
105 M.J. Bonell (ed.), The UNIDROIT Principles in Practice: Case Law and
Bibliography on the Principles of Commercial Contracts (New York:
Transnational, 2002), 491.
106 (1999) 10:2 ICC ICArb Bull 78 cited in James Otis Rodner, ‘Hardship under

the UNIDROIT Principles of International Commercial Contracts’, in Global

Reflections on International Law, Commerce and Dispute Resolution: Liber
Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al. (ICC Publishing,
2005), 677, 685. ICC Case No. 7367, it was considered a general principle of
law.106 A converse view is found in lCC Case No. 8873 (1999) 10:2 ICC ICArb
Bull 78.
107 Article 6.2.3 Comment 1, UNIDROIT Principles.
108 Article 6.2.3(4).
109 Alexis Mourre, ‘Judicial Penalties and Specific Performance in International

Arbitration’, in Interest, Auxiliary and Alternative Remedies in International

Arbitration, Dossiers ICC Institute of World Business Law, ed. Filip de Ly &
Laurent Lévy (Paris: ICC Publishing, 2008), 53.
110 Such powers may apply on enforcement applications.
111 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2480.

112 See Kremezi v. Ridgway [1949] All ER 662, 663; Kohnke v. Karger [1951] 2

K.B. 670, 677; D'almeida Araujo v. Becker [1953] 2 QB 329.

113 For example, CMI Int'l Inc v. Iran (US v. Iran), Iran-US Claims Tribunal Case

No. 245 (99-245-2) (Award of 27 December 1983) 1985 Yearbook of

Commercial Arbitration 316, 319.
114 Award of 23 July 1981, Shipping Arbitration (Norway v. Belgium), in Pieter

Sanders (ed.), Yearbook Commercial Arbitration, vol. VIII (The Hague: Kluwer
Law International, 1983), 89, 91. Iran-US Claims Tribunal cases need to be
looked at in the context of the Claims Settlement Declaration which itself has a
choice of law provision. Award of 27 December 1983, CMI Internationall Inc v.
Ministry of Roads and Transportation, the Islamic Republic of Iran (US v. Iran), in
Pieter Sanders (ed.), Yearbook Commercial Arbitration, vol. X (The Hague:
Kluwer Law International, 1985), 316, 319; North Sea Continental Shelf (West
Germany v. Denmark; West Germany v. Netherlands) (1969) ICJ 3, judgment of 20
February 1969, reprinted in (1969) 8 ILM 340, 379.
115 Hilary Heilbron QC, ‘Assessing Damages in International Arbitration:
Practical Considerations’, in The leading Arbitrators' Guide to International
Arbitration, ed. L.W. Newman & R.D. Hill (Huntington: Juris Publishing, 2008),
454.
116 Pabalk Ticaret Ltd v. Norsalor SA (Turkey v. France) ICC Case No. 3131,

Award of 26 October 1979, 1985 Yearbook of Commercial Arbitration 109;

upheld in Norsalor SA v. Pabalk Ticaret Ltd, Oberster Gerichtschof (Austrian
Supreme Court) Judgment of 18 November 1982, 1984 Yearbook of Commercial
Arbitration 159.
117 Or respondent's on a counter-claim.
118 International Institute for Conflict Prevention & Resolution, CPR Protocol on

Determination of Damages in Arbitration (2010) 3.

119 Mark Kantor, Valuation for Arbitration: Compensation Standards, Valuation

Methods and Expert Evidence (The Hague: Kluwer Law International, 2008), 283.
120 Ibid.
121 Ibid.
122 International Institute for Conflict Prevention & Resolution, CPR Protocol on

Determination of Damages in Arbitration (2010).

123 American Institute of CPAs, Code of Professional Conduct, Rule 101
<www.aicpa.org/Research/Standards/CodeofConduct/Pag...>.
124 Mark Kantor, Valuation for Arbitration: Compensation Standards, Valuation

Methods and Expert Evidence (The Hague: Kluwer Law International, 2008), 289.
125 This method is discussed in section 14.8.5.
126 Hilary Heilbron QC, ‘Assessing Damages in International Arbitration:
Practical Considerations’, in The Leading Arbitrators' Guide to International
Arbitration, ed. L.W. Newman & R.D. Hill (Huntington: Juris Publishing, 2008),
449.
127 Ibid., 450.
128 Mark Kantor, Valuation for Arbitration: Compensation Standards, Valuation

Methods and Expert Evidence (The Hague: Kluwer Law International, 2008), 282.
129 Charles Proctor, ‘Changes in Monetary Values and the Assessment of
Damages’, in Contract Damages: Domestic and International Perspectives, ed.
Djakhongir Saidov & Ralph Cunning-ton (Oxford: Hart Publishing, 2008), 65,
459.
130 Michael Bridge, ‘The Market Rule of Damages Assessment’, in Contract

Damages: Domestic and International Perspectives, ed. Djakhongir Saidov &

Ralph Cunnington (Oxford: Hart Publishing, 2008), 65, 430, 435.
131 Ole Lando & Hugh Beale, Principles of European Contract Law, Parts I and II

(The Hague: Kluwer Law International, 2000), 2443.

132 Official Comment on Art. 7.4.3.
133 Alexander S Komarav, ‘Mitigation of Damages’, in Evaluation of Damages in

International Arbitration, Dossier of the ICC Institute of World Business Law, ed.
Yves Derains & Richard H. Kreindler (Paris: ICC Publishing, 2006), 253.
134 Article 7.4.3(3).
135 Paul-A Gélinas, ‘General Characteristics of Recoverable Damages in
International Arbitration’, in Evaluation of Damages in International Arbitration,
Dossier of the ICC Institute of World Business Law, ed. Yves Derains & Richard H.
Kreindler (Paris: ICC Publishing, 2006), 20.
136 Jan Paulsson, ‘The Expectation Model’, in Evaluation of Damages in
International Arbitration, Dossier of the ICC Institute of World Business Law, ed.
Yves Derains & Richard H. Kreindler (Paris: ICC Publishing, 2006), 59.
137 Hilary Heilbron QC, ‘Assessing Damages in International Arbitration:
Practical Considerations’, in The Leading Arbitrators' Guide to International
Arbitration, ed. L.W. Newman & R.D. Hill (Huntington: Juris Publishing, 2008),
462.
138 International Institute for Conflict Prevention & Resolution, CPR Protocol on

Determination of Damages in Arbitration (2010) 2.

139 Hilary Heilbron QC, ‘Assessing Damages in International Arbitration:
Practical Considerations’, in The Leading Arbitrators' Guide to International
Arbitration, ed. L.W. Newman & R.D. Hill (Huntington: Juris Publishing, 2008),
460.
140 S D Myers v. The Government of Canada, partial award of 13 November 2000,

§ 316; Pope & Talbot v. The Government of Canada, award in respect of damages
of 31 May 2002, § 80.
141 Hilary Heilbron QC, ‘Assessing Damages in International Arbitration:
Practical Considerations’, in The Leading Arbitrators' Guide to International
Arbitration, ed. L.W. Newman & R.D. Hill (Huntington: Juris Publishing, 2008),
470.
142 Ibid., 464.
143 Livingstone v. Rawyards Coal Co. [1880] 5 AC 25, at 39 per Lord Blackburn.

The position taken at English law is that there exists a fundamental principle ‘to
restore the injured party to the same position he would have been in but for the
breach’ but that this principle should not be misapplied to unjustly enrich a
plaintiff. See Golden Strait Corp v. Nippon Yusen Kubishika Kaisha (The Golden
Victory) [2007] UKHL 12, 83.
144 Robinson v. Harman [1848] 154 ER 363, 365: ‘[W]here a party sustains loss

by breach of a contract, he is, so far as money can do it, to be placed in the same
position, with respect to damages, as if the contract had been performed.’
145 Anglia Television v. Reed [1972] 1 QB 60.
146 Hilary Heilbron QC, ‘Assessing Damages in International Arbitration:
Practical Considerations’, in The Leading Arbitrators' Guide to International
Arbitration, ed. L.W. Newman & R.D. Hill (Huntington: Juris Publishing, 2008),
466.
147 International Institute for Conflict Prevention & Resolution, CPR Protocol on

Determination of Damages in Arbitration (2010) 1.

148 Olusoji Elia, Judicial Remedies in the Conflict of Laws (Oxford: Hart
Publishing, 2001), 76.
149 Pabalk Ticaret Ltdv. Norsalor SA (Turkey v. France) ICC Case No. 3131, Award

of 26 October 1979, 1985 Yearbook of Commercial Arbitration 109; upheld in

Norsalor SA v. Pabalk Ticaret Ltd, Oberster Gerichtschof (Austrian Supreme
Court) Judgment of 18 November 1982, 1984 Yearbook of Commercial
Arbitration 159. As to assessment of global sums see Window Arbitration,
Netherlands Arbitration Court Building Industry Award of 16 April 1975,
Summary in 1976 Yearbook of Commercial Arbitration 137.
150 Norber Wuhler, ‘Application of General Principles of Law’, in Planning
Efficient Arbitration Proceedings/The Law Applicable in International Arbitration,
ICCA Congress Series No. 7 (Vienna 1994), ed. Albert Jan van den Berg (The
Hague: Kluwer Law International, 1996), 559.
151 Sea-Land Service v. The Government of the Islamic Republic of Iran (1984) 6

Iran-US Claims Tribunal Reports 149, 168.

152 Phillip Landolt, Modernised EC Competition Law in International Arbitration,

(The Hague: Kluwer Law International, 2006), 351.

153 Alexander S Komarav, ‘Mitigation of Damages’, in Evaluation of Damages in

International Arbitration, Dossier of the ICC Institute of World Business Law, ed.
Yves Derains & Richard H. Kreindler (Paris: ICC Publishing, 2006), 38.
154 International Institute for Conflict Prevention & Resolution, CPR Protocol on

Determination of Damages in Arbitration (2010) 2.

155 Ibid.
156 Mark Kantor, Valuation for Arbitration: Compensation Standards, Valuation

Methods and Expert Evidence (The Hague: Kluwer Law International, 2008), 1.
157 Ibid., 2.
158 Dr Pratt's approach was cited with approval by the ICSID Tribunal in Ioannis

Kardassopolous & others v.The Republic of Georgia (ICSID 3 March 2010) at para.
601.
159 Shannon P. Pratt, Robert F. Reilly & Robert P. Schneihs, Valuing a Business:

The Analysis and Appraisal of Closely Held Companies, 4th edn (New York:
Mcgraw-Hill, 2000), 445.
160 Mark Kantor, Valuation for Arbitration: Compensation Standards, Valuation

Methods and Expert Evidence (The Hague: Kluwer Law International, 2008), 2.
161 Custodio O. Parlade, ‘Arbitral Awards’, in The Asian Leading Arbitrators'

Guide to International Arbitration, ed. Michael Pryles & Michael Moser

(Huntington: JurisNet LLC, 2007), 389.
162 Pabalk Ticaret Ltd v. Norsalor SA (Turkey v. France) ICC Case No. 3131,

Award of 26 October 1979, 1985 Yearbook of Commercial Arbitration 109;

upheld in Norsalor SA v. Pabalk Ticaret Ltd, Oberster Gerichtschof (Austrian
Supreme Court) Judgment of 18 November 1982, 1984 Yearbook of Commercial
Arbitration 159.
163 Mark Kantor, Valuation for Arbitration: Compensation Standards, Valuation

Methods and Expert Evidence (The Hague: Kluwer Law International, 2008), 13.
164 Ibid., 3.
165 IVSC, Guidance Note GN 6 on Business Valuation (International Valuation

Standards, 8th edn 2007).

166 Mark Kantor, Valuation for Arbitration: Compensation Standards, Valuation

Methods and Expert Evidence (The Hague: Kluwer Law International, 2008), 9.
167 Ibid., 10.
168 Ibid., 15.
169 Ibid.
170 Ibid., 27.
171 CSX Transport., Inc. v. Georgia State Bd. Of Equalization 128 S. Ct. 467 (4

December 2007).
172 Ibid.
173 Compañía de Aguas del Acnoquija S.A. & Others v. Argentine Republic, ICSID 10
August 2010 para. 32.
174 Ibid., para. 173.
175 Ibid., para. 255.
176 Mark Kantor, Valuation for Arbitration: Compensation Standards, Valuation

Methods and Expert Evidence (The Hague: Kluwer Law International, 2008), 286.
177 Ibid.
178 Azurix v. Argentina, ICSID Case No. ARB/CAF/01/12, Award dispatched 14

July 2006, at para. 426: An ICSID Tribunal found the price of USD 438.55 million
paid by Azurix to the Province of Buenos Aires for a water concession to be
unreasonable high. In calculating damages, the tribunal founded its award on
the lower fair market value of USD 60 million, plus verified additional
subsequent capital investments.
179 See Southern Pacific Properties (Middle East) Limited v. Arab Republic of

Egypt, ICSID Case No. ARB/84/3 at para. 197.

180 CMS Transmission Company v. The Argentina Republic, ICSID Case No.
ARB/01/08, Award at 402. The definition employed by the tribunal is found at
International Glossary of Business Valuation Terms, available at
<www.bvapprasiers.org/glossary/glossary.pdf>, This definition was also
adopted by the tribunal in National Grid pic v. Argentina UNCITRAL, Award, 3
November 2008, para. 263, note 99.
181 World Bank Guidelines IV 5, International Bank for Reconstruction and

Development ‘Report to the Development Committee and Guidelines on the

Treatment of Foreign Direct Investment (1992) 31 ILM 1366.
182 Mark Kantor, Valuation for Arbitration: Compensation Standards, Valuation

Methods and Expert Evidence (The Hague: Kluwer Law International, 2008), 18.
183 Ibid., 35. For further discussion of rare events (or Black Swans as they are

known) see Mark Kantor, Valuation for Arbitration: Compensation Standards,

Valuation Methods and Expert Evidence (The Hague: Kluwer Law International,
2008), 24–25; Taleb, The Black Swan: The Impact of the Highly Improbably, at
xvii (Random House 2007).
184 Republic of Kazakhstan v. Rumeli Telekom A.S. ICSID Case No. ARB/05/16, 25

March 2010 at para. 151; Mark Kant or, Valuation for Arbitration: Compensation
Standards, Valuation Methods and Expert Evidence (The Hague: Kluwer Law
International, 2008), 68–69; Golden Strait Corp v. Nippon Yusen Kubishika Kaisha
(The Golden Victory) [2007] 2 WLR 691 at para. 36.
185 Metalclad Corporation v. The United Mexican States, ICSID Case No. ARB(AF)

97/1, Award of 30 August 2000, §§ 119–120.

186 Enron Corporation and Ponderosa Assets, LP. v. Republic of Argentina, ICSID

Case No. ARB/01/3, Award may 22 2007.

187 Ibid., at 388.
188 Ioannis Kardassopolous & others v. The Republic of Georgia (ICSID 3 March

2010) at para. 603.

189 Dobbs, Law of Remedies, 2nd edn (St Paul: West Publishing Co., 1993), at 3.5.
190 CME Czech Republic B.V. v. The Czech Republic, Final Award issued in
Stockholm, Sweden, in the UNCITRAL Arbitration Proceedings, March 14 2003
at 161.
191 Mark Kantor, Valuation for Arbitration: Compensation Standards, Valuation

Methods and Expert Evidence (The Hague: Kluwer Law International, 2008), 59.
192 Henry Weisburg & Christopher Ryan, ‘Means to be made Whole: Damages in
the Context of International Investment Arbitration’, in Evaluation of Damages in
International Arbitration, Dossier of the ICC Institute of World Business Law IV, ed.
Yves Derains & Richard H. Kreindler (Paris: ICC Publishing, 2006), 179.
193 Ibid., citing World Bank Report, 31 ILM 1377 (1992); Aminoil, 21 ILM 1038-

1039.
194 International Institute for Conflict Prevention & Resolution, CPR Protocol on

Determination of Damages in Arbitration (2010) 4.

195 Jan Paulsson, ‘The Expectation Model’, in Evaluation of Damages in
International Arbitration, Dossier of the ICC Institute of World Business Law, ed.
Yves Derains & Richard H. Kreindler (Paris: ICC Publishing, 2006), 65.
196 Henry Weisburg & Christopher Ryan, ‘Means to Be made Whole: Damages in

the Context of International Investment Arbitration’, in Evaluation of Damages in

International Arbitration, Dossier of the ICC Institute of World Business Law IV, ed.
Yves Derains & Richard H. Kreindler (Paris: ICC Publishing, 2006), 173.
197 Ibid., 174.
198 BorzuSabahi & Nicholas J. Birch, ‘Comparative Compensation for
Expropriation’, in Stephen W. Schill, International Investment Law and
Comparative Public Law (Oxford: Oxford University Press, 2010) 764; for
examples of arbitral tribunals that have used the DCF method, see CME Czech
Republic BV v. Czech Republic Ad Hoc UNCITRAL Arbitration Rules, Final Award
and Separate Opinion, 14 March 2003, paras 595–620; CMS Gas Transmission Co
v. Argentina ICSID Case No. ARB/01/8, Award, 12 May 2005, para. 403.
199 Henry Weisburg & Christopher Ryan, ‘Means to be made Whole: Damages in

the Context of International Investment Arbitration’, in Evaluation of Damages in

International Arbitration, Dossier of the ICC Institute of World Business law, ed.
Yves Derains & Richard H. Kreindler (Paris: ICC Publishing, 2006), 174; Amoco
International Finance Corporation v. Islamic Republic of Iran, Partial Award of 14
July 1987, 27 ILM 1320, 1988, 1365.
200 Henry Weisburg & Christopher Ryan, ‘Means to be made Whole: Damages in

the Context of International Investment Arbitration’, in Evaluation of Damages in

International Arbitration, Dossier of the ICC Institute of World Business law, ed.
Yves Derains & Richard H. Kreindler (Paris: ICC Publishing, 2006), 176. Paul B.
Friedland & Eleanor Wong, ‘Measuring Damages for the Deprivation of Income-
Producing Assets: ICSID Case Studies’, ICSID Review – Foreign Investment law
Journal 6, no. 2 (1991), 400, 406–407.
201 Henry Weisburg & Christopher Ryan, ‘Means to be made Whole: Damages in

the Context of International Investment Arbitration’, in Evaluation of Damages in

International Arbitration, Dossier of the ICC Institute of World Business law, ed.
Yves Derains & Richard H. Kreindler (Paris: ICC Publishing, 2006), 174.
202 Amoco International Finance Corporation v. Islamic Republic of Iran, Partial

Award of 14 July 1987, 27 ILM 1320, 1988, §238.

203 Hilary Heilbron QC, ‘Assessing Damages in International Arbitration:
Practical Considerations’, in The leading Arbitrators' Guide to International
Arbitration, ed. L.W. Newman & R.D. Hill (Huntington: Juris Publishing, 2008),
469.
204 See, e.g., Himpurna California Energy ltd v. PP Persero Perusahaan listruik

Negana (PL) ad hoc, 4 May 1999, Yearbook of Commercial Arbitration, vol. XXV,
13 (2000), 84.
205 Iran-US Claims Tribunal, Philipps Petroleum Award, Factory at Chorzów
(Germany v. Poland) (Indemnity), 1928 PCD (Ser A) No. 17, § 111, 114.
206 Compania del Desarrollo de Santa Elena, S.A. v. Costa Rica, ICSID Case No.

ARB/96/1, Final Award (17 February 2000), 15 ICSID Rev.-FILJ 169 (2000).
207 Mark Kantor, Valuation for Arbitration: Compensation Standards, Valuation

Methods and Expert Evidence (The Hague: Kluwer Law International, 2008), 281.
208 Ibid.
209 A flat discount rate was applied in ICC Case No. 6829, 1991, in J.J. Arnaldez,

Y. Derains & D. Hascher, Collection of ICC Arbitral Awards 1996 – 2000, vol. 5
(The Hague: Kluwer Law International, 1997), 282.
210 Golden Strait Corp v. Nippon Yusen Kubishika Kaisha (The Golden Victory)

[2007] UKHL 12.

211 Mark Kantor, Valuation for Arbitration: Compensation Standards, Valuation

Methods and Expert Evidence (The Hague: Kluwer Law International, 2008), 61.
212 Golden Strait Corp v. Nippon Yusen Kubishika Kaisha (The Golden Victory)

[2007] UKHL 12.

213 Ibid.
214 Enron Corporation and Ponderosa Assets, L.P. v.Republic of Argentina, ICSID

Case No. ARB/01/3, Award may 22 2007 at para. 200.

215 Sinclair Refining Co. v. Jenkins Petroleum Process Co., 289 US 689, 698 (1933)

(citations omitted).
216 CME Stockholm, Final Award, at 131.
217 See Borzu Sabahi & Nicholas J. Birch, ‘Comparative Compensation for
Expropriation’, in International Investment Law and Comparative Public Law, ed.
Stephen W. Schill (Oxford: Oxford University Press, 2010), 766–771.
218 Compania del Desarollo de Santa Elena, S.A. v. Costa Rica, ICSID Case No.

ARB/96/1, Final Award (17 February 2000) at 76–83.

219 Case concerning The Factory at Chorzów, PCIJ Series A, No. 13 (PCIJ 1928).
220 Mark Kantor, Valuation for Arbitration: Compensation Standards, Valuation

Methods and Expert Evidence (The Hague: Kluwer Law International, 2008), 67.
221 En Cana v. Equador, UNCITRAL/LCIA Case No UN3481 (Award) (3 February

2006) reprinted in (2006) 45 ILM 895 with comment by Dev Krishan.

222 Compañia de Aguas del Aconquija SA and Compagnie Générate des
Eaux/Vivendi Universal (Vivendi') v. Argentine Republic (ICSID Case No.
ARB/97/3), Second Award, 20 August 2007, para. 8.3.3 (emphasis in original).
223 Jan Paulsson, ‘The Expectation Model’, in Evaluation of Damages in
International Arbitration, Dossier of the ICC Institute of World Business Law, ed.
Yves Derains & Richard H. Kreindler (Paris: ICC Publishing, 2006), 60 citing
Kyocera v. Prudential-Baché, 299 F. 3d 769, 790 (9th Cir 2002).
224 Republic of Kazakhstan v. Rumeli Telekom A.S. ICSID Case No. ARB/05/16, 25

March 2010 at para. 144.

225 (Award) (1963) 35 ILR 136, 187–188.
226 ICSID Case No. ARB/84/3, Award 20 May 1992, 3 ICSID Rep 189.
227 South Pacific Properties (Middle East) Ltd v. Arab republic of Egypt, ICSID

Case No. ARB/84/3, Award 20 May 1992, 3 ICSID Rep 189, para. 215.
228 (ICSID Case No. ARB/98/4), Decision on Annulment, 5 February 2002, 6

ICSID Rep 129, para. 91.

229 Per Brennan J, Sellars v. Adelaide Petroleum NL (1992–1994) 179 CLR 332,

368 (HCA).
230 Gotanda ‘Recovering Lost Profits in International Disputes’ (2004–2005) 36
Georgetown JIL 61, 101.
231 Jan Paulsson, ‘The Expectation Model’, in Evaluation of Damages in
International Arbitration, Dossier of the ICC Institute of World Business Law, ed.
Yves Derains & Richard H. Kreindler (Paris: ICC Publishing, 2006), 61 citing
Aminoil v. Kuwait, ad hoc Award of 24 March 1982, 66 International Law
Reports 518 (1984).
232 American Independent Oil Company v. Government of Kuwait, Award of 24

March 1982, 66 I.L.R. 518 (1984).

233 Mark Kantor, Valuation for Arbitration: Compensation Standards, Valuation

Methods and Expert Evidence (The Hague: Kluwer Law International, 2008), 21.
234 Ibid.
235 See, e.g., Mechema Ltd v. MNM (UK v. Belgium) Award of 3 November 1977,

1982 Yearbook of Commercial Arbitration 77.

236 See, e.g., FPP (Middle East) Ltd v. Egypt (Hong Kong v. Egypt) ICC Case No.

3493, Award of 16 February 1983, Summary in 1984 Yearbook of Commercial

Arbitration 112, 123.
237 Paul-A Gélinas, ‘General Characteristics of Recoverable Damages in
International Arbitration’, in Evaluation of Damages in International Arbitration,
Dossier of the ICC Institute of World Business Law, ed. Yves Derains & Richard H.
Kreindler (Paris: ICC Publishing, 2006), 21, fn 18.
238 Metalclad (fn. 43).
239 Jane Jenkins & Simon Stebbings, International Construction Arbitration Law

(Kluwer Law International: 2006) 196.

240 [2001] 1 AC 268; see also Devenish Nutrition Limited v. Sanofi-Aventis SA

(France) & Ors [2008] EWCA 1086.

241 Experience Hendrix LLC v. PPX Enterprise Inc [2003] EWCA 323.
242 French Civil Code Art. 1149.
243 Sapphire International Petroleum ltd v. National Iranian Oil Co, Arbitral

Award, 15 March 1963, reprinted in 35 International Law Reports 136, 187–

188.
244 Emmanuel Gaillard & John Savage (eds), Fouchard Gaillard Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 384.
245 Henry Weisburg & Christopher Ryan, ‘Means to be made Whole: Damages in

the Context of International Investment Arbitration’, in Evaluation of Damages in

International Arbitration, Dossier of the ICC Institute of World Business Law, ed.
Yves Derains & Richard H. Kreindler (Paris: ICC Publishing, 2006), 166, fn 8.
246 Mark Kantor, Valuation for Arbitration: Compensation Standards, Valuation

Methods and Expert Evidence (The Hague: Kluwer Law International, 2008), 49.
247 UN General Assembly Resolution 3171 (XXVII) Permanent Sovereignty Over

Natural Resources, 28 UN GAOR Supp, No. 30, at 52, UN Doc A/390, 1973,
reprinted in 13 ILM 238, 1974.
248 Pierre Bienvenu & Martin J. Valasek, ‘Compensation for Unlawful
Expropriation, and Other Recent Manifestations of the Principle of Full
Reparation in International Investment Law’, in 50 Years of the New York
Convention (ICCA International Arbitration Conference, ICCA Congress Series no.
14, Dublin 2009), ed. Albert Jan van den Berg (Paris: ICC Publishing, 2009), 231.
For example, the 2004 US Model Bilateral Investment Treaty indicates that
expropriation may only occur ‘upon payment of prompt, adequate and effective
compensation’ measured as the fair market value of the property before
expropriation too place. available at
<www.state.gov/document/organisation/39710.pdf>.
249 ADC v. Hungary, ICSID Case No. ARB/03/16, Award (2 October 2006) para.

481.
250 Case Concerning the Factory at Chorzów, PCIJ Series A, No. 13, 20 (PCIJ 1928)

at 47.
251 CME Czech Republic B.V. v. The Czech Republic, Final Award issued in
Stockholm, Sweden, in the UNCITRAL Arbitration Proceedings, March 14 2003.
252 Mark Kantor, Valuation for Arbitration: Compensation Standards, Valuation

Methods and Expert Evidence (The Hague: Kluwer Law International, 2008), 50.
253 Ibid.
254 Ibid., 52.
255 Ioannis Kardassopolous & others v. The Republic of Georgia Case No.
ARB/05/18 (ICSID 3 March 2010).
256 James v. UK, European Court of Human Rights, 1986, 8 EHRR 123, at 48.
257 Siemens A.G. v. The Argentine Republic, ICSID Case No. ARB/02/08, Award (6

February 2007) at para. 346.

258 Ibid., at paras 349 and 352. This approach was also adopted by the tribunal

in Vivendi v. Argentina ICSID Case No. ARB/97/3, Award (20 August 2007) at
8.2.2-8.2.5.
259 LG&E v. Argentina, ICSID Case No. ARB/02/1, Damages Award of 25 July

2007, paras 35-36.

260 CMS Gas Transmission Co. v. Argentina, ICSID Case No. ARB/01/08 (US-

Argentina BIT Award of 12 May 2005 para. 410.

261 Azurix Corp v. Argentina, ICSID Case No. ARB/01/12, Award of 14 July 2006

paras 419–424.
262 Enron and Ponderosa Assets v. Argentina, ICSID Case ARB/01/3, Award of 22

May 2007 paras 359–363.

263 Sempra Energy v. Argentina, ICSID Case No. ARB/02/16 Award of 28
September 2007 paras 402–404.
264 Pierre Yves Tschanz & Jorge E. Viñuales, ‘Compensation for Non-
expropriatory Breaches of International Investment Law’, Journal of
International Arbitration 26, no. 5 (2009): 729–743.
265 ibid., 737.
266 See, e.g., D. Myers, Inc. v. Canada, UNCITRAL (NAFTA), Partial Award (13

November 2000), at paras 311–315; Metalclad Corporation v. United Mexican

States, ICSID Case No. ARB(AF)/97/l, Award (30 August 2000), at para. 122;
MTD Equity Sdn. Bhd. & MTD Chile S.A. v. Chile, ICSID Case No. ARB/01/7, Award
(25 May 2004) at para. 238; Enron Corporation and Ponderosa Assets, L.P. v.
Republic of Argentina, ICSID Case No. ARB/01/3, Award 22 May 2007 at paras
359–361.
267 Timing is particularly problematic with creeping or indirect expropriation.

William W. Park, ‘Framing the Case on Quantum’, World Arbitration and

Mediation Review 2, no. 4 (2008): 59, 62; John Y Gotanda, Damages in Private
International Law, 326 Recueil des Cours 73 (2007); Mark Kantor, Valuation for
Arbitration: Compensation Standards, Valuation Methods and Expert Evidence
(The Hague: Kluwer Law International, 2008), 61.
268 The Loewen Group Inc. v. United States of America, ICSID Case No.
ARB(AF)/98/3, Final Award of 26 June 2003.
269 See, for instance, s. 48 of the Arbitration Act 1996 (UK), which was applied in

Lesotho Highlands Development Authority v. Impregilo SpA and others [2006] 1

AC 221.
270 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th

edn (Oxford: Oxford University Press, 2009), 327.

271 See Smit, Judgments and Arbitral Awards in a Foreign Currency: A Means of

Dealing with Currency Fluctuations in International Adjudication (1996) 7 Am

Rev Int'l Arb 21.
272 José Rosell, ‘Arbitration Costs as Relief and/or Damages’, Journal of
International Arbitration 28, no. 2 (2008): 122.
273 ICC Case No. 12491, Partial Award No. 2, 1 June 2004, 24 ASA Bull 281

(2006).
274 Mantovani v. Carapelli S.p.A. [1980] 1 Lloyd's Rep 375 (CA); CMA, COM, S.A. v.

Hyundai Mito Dockyard Ltd [2008] EWCH (Comm) 2791; Paramedics

Electromedicina Comercial Ltda v. GE Medical Systems Information Technologies,
Inc 369 F. 3d 645 (2d Cir 2004).
275 José Rosell, ‘Arbitration Costs as Relief and/or Damages’, Journal of
International Arbitration 28, no. 2 (2008): 124.
276 Ibid.
211 Varying approaches, particularly in common law jurisdictions are discussed

in Mr Justice Vivian Ramsey, ‘Problems of Delay and Disruption Damages in

International Construction Arbitration’, in Evaluation of Damages in
International Arbitration, Dossier of the ICC Institute of World Business Law, ed.
Yves Derains & Richard H. Kreindler (Paris: ICC Publishing, 2006), 193; and
citing The Great Delay Analysis Debate, King's College, London, 18 October
2005, organised by the Society of Construction Law and the Centre for
Construction Law and Management and Concurrent Delay, John Marrin QC,
Society of Construction Law, London, 5 February 2002, available at
<www.scl.org.uk>.
278 The key formulae are the Hudson, Enden and Eichleay formulae, described in

Mr Justice Vivian Ramsey, ‘Problems of Delay and Disruption Damages in

International Construction Arbitration’, in Evaluation of Damages in
International Arbitration, Dossier of the ICC Institute of World Business Law, ed.
Yves Derains & Richard H. Kreindler (Paris: ICC Publishing, 2006), 209.
279 Jane Jenkins & Simon Stebbings, International Construction Arbitration Law

(Alphen aan den Rijn: Kluwer Law International, 2006), 46–47.

280 Ibid., 217–218.
281 Yearbook of Commercial Arbitration, vol. XXV, 2000 at 89–90.
282 Henry Weisburg & Christopher Ryan, ‘Means to be made Whole: Damages in

the Context of International Investment Arbitration’, in Evaluation of Damages in

International Arbitration, Dossier of the ICC Institute of World Business Law, ed.
Yves Derains & Richard H. Kreindler (Paris: ICC Publishing, 2006), 178.
283 See the discussion in Jan Paulsson, ‘The Expectation Model’, in Evaluation of

Damages in International Arbitration, Dossier of the ICC Institute of World

Business Law, ed. Yves Derains & Richard H. Kreindler (Paris: ICC Publishing,
2006), 72–73.
284 Felix Praendl, ‘Measure of Damages in International Commercial
Arbitration’, Stanford Journal of International Law 23, no. 1 (1987): 263, 292.
285 Jacques Werner, ‘Punitive and Exemplary Damages in International
Arbitration’, in Evaluation of Damages in International Arbitration, Dossier of the
ICC Institute of World Business Law, ed. Yves Derains & Richard H. Kreindler
(Paris: ICC Publishing, 2006), 102.
286 Ingeborg Schwenzer & Pascal Hachem, ‘Moral Damages in International

Investment Arbitration’, in International Arbitration and International

Commercial Law: Synergy, Convergence and Evolution, Liber Amicorum Eric
Bergsten, ed. S. Kröll et al. (Alphen aan den Rijn: KluwerLaw International,
2011), 425.
287 Ibid., 426.
288 For example, Art. 5 Model Norwegian Bilateral Investment Treaty (2007)

cited in Ingeborg Schwenzer & Pascal Hachem, ‘Moral Damages in International

Investment Arbitration’, in International Arbitration and International
Commercial Law: Synergy, Convergence and Evolution, Liber Amicorum Eric
Bergsten, ed. S. Kröll et al. (Alphen aan den Rijn: Kluwer Law International,
2011), 419.
289 Desert Lion Projects LLC v. Yemen, ICSID Case No. ARB/05/17 (Award) (6.

February 2008).
290 See Borzu Sabahi & Nicholas J. Birch, ‘Comparative Compensation for
Expropriation’, in International Investment Law and Comparative Public Law, ed.
Stephen W. Schill (Oxford: Oxford University Press, 2010), 771.
291 International Valuation Standards Committee, International Valuation
Standards, 8th edn (London: IVSC, 2007), 208.
292 See Borzu Sabahi & Nicholas J. Birch, ‘Comparative Compensation for
Expropriation’ in International Investment Law and Comparative Public Law, ed.
Stephen W. Schill (Oxford: Oxford University Press, 2010), 772.
293 Charles Brower, ‘Correction and Completion of Awards; Enforcement of

Partial and Final Awards; Collaboration by Courts for an Award to Be Effective;

‘Impact of International Public Policy' on Arbitration’, in International
Arbitration in a Changing World, ICCA Congress Series No. 6 (Bahrain 1993), ed.
Albert Jan van den Berg (Paris: ICC Publishing, 1994), 213, 214.
294 Alexander S Komarav, ‘Mitigation of Damages’, in Evaluation of Damages in

International Arbitration, Dossier of the ICC Institute of World Business Law, ed.
Yves Derains & Richard H. Kreindler (Paris: ICC Publishing, 2006), 256.
295 Ibid., 39.
296 Ibid., 40.
297 Middle East Cement Shipping and Handling Co v. Republic of Egypt, Award of

12 April 2002, ICSID Case No. ARB/99/6.

298 Alexander S Komarav, ‘Mitigation of Damages’, in Evaluation of Damages in

International Arbitration, Dossier of the ICC Institute of World Business Law, ed.
Yves Derains & Richard H. Kreindler (Paris: ICC Publishing, 2006), 37.
299 Yasuhei Taniguchi, ‘The Obligation to Mitigate Damages’, in Evaluation of

Damages in International Arbitration, Dossier of the ICC Institute of World

Business Law, ed. Yves Derains & Richard H. Kreindler (Paris: ICC Publishing,
2006), 84.
300 UNCITRAL Digest of Case Law on the UN CISG (A/CN.9
SEI.C/Digest/CISG/77, Hof Arnhem, The Netherlands, 22 August 1995, Unilex.
301 Alexander S Komarav, ‘Mitigation of Damages’, in Evaluation of Damages in
International Arbitration, Dossier of the ICC Institute of World Business Law, ed.
Yves Derains & Richard H. Kreindler (Paris: ICC Publishing, 2006), 46.
302 Ibid., 51.
303 See, e.g., ICC Award No. 7006, Yearbook of Commercial Arbitration, vol. XVIII

1993, at 58.
304 ICSID Final Award of 5 June 1990, Case No. ARB/81/8 and Decision on

Supplemental Decision and Rectification of 17 October 1990, Yearbook of

Commercial Arbitration, vol. XVII, 1992, 73.
305 A substitute purchase is described as a cover purchase under American law.
306 Alexander S Komarav, ‘Mitigation of Damages’, in Evaluation of Damages in

International Arbitration, Dossier of the ICC Institute of World Business Law, ed.
Yves Derains & Richard H. Kreindler (Paris: ICC Publishing, 2006), 42.
307 Ibid., 44.
308 International Institute for Conflict Prevention & Resolution, CPR Protocol on

Determination of Damages in Arbitration (2010) 2.

309 Article 1152 French Civil Code.
310 See also Art. 9:509 Principles of European Contract Law.
311 Dunlop Pneumatic Car Co Ltd v. New Garage & Motor Co Ltd [1915] PC 79, 86.
312 See generally ICC Guide on Penalty and Liquidated Damages Clauses,
Publication No. 478 (1999).
313 For an annex on published awards in relation to liquidated damages see

Antonias Dimolitsa, ‘Contractual Remedies: Clauses Penales and Liquidated

Damages Clauses’, in Interest, Auxiliary and Alternative Remedies in International
Arbitration, Dossiers ICC Institute of World Business Law, ed. Filip de Ly &
Laurent Lévy (Paris: ICC Publishing, 2008), 27.
314 Hilary Heilbron QC, ‘Assessing Damages in International Arbitration:
Practical Considerations’, in The Leading Arbitrators' Guide to International
Arbitration, ed. L.W. Newman & R.D. Hill (Huntington: Juris Publishing, 2008),
465.
315 Jane Jenkins & Simon Stebbings, International Construction Arbitration Law

(Kluwer Law International 2006) 39.

316 Ibid., 40.
317 Hilary Heilbron QC, ‘Assessing Damages in International Arbitration:
Practical Considerations’, in The Leading Arbitrators' Guide to International
Arbitration, ed. L.W. Newman & R.D. Hill (Huntington: Juris Publishing, 2008),
465 citing Clause 17.6 General Conditions of Contracts for Construction FIDIC
1998.
318 From a consent-based policy perspective, a government's general desire to

encourage whistleblowers is not a dominant factor where arbitral jurisdiction is

concerned. The situation may be very different where the parties agree on
applicable law as opposed to situations where instead the tribunal applies its
own broad discretion, absent an agreement between the parties. Redfern and
Hunter note that under English law punitive damages may only be awarded in
tort actions and only in three category of cases, which are ‘(1) abuse of power by
servants of the government; (2) conduct which was motivated by the pursuit of
profits; and (3) where punitive (aka exemplary) damages are expressly
authorised by statute’: Nigel Blackaby et al., Redfern and Hunter on International
Arbitration, 5th edn (Oxford: Oxford University Press, 2009), Rookes v. Barnard
[1964] AC 1129. The situation is different in the US where there are statutes
expressly allowing for the payment of multiple damages, such as the Racketeer
Influenced and Corrupt Organizations Act (18 USCS) (RICO) and antitrust laws
that provide for treble damages.
319 See, e.g., Bin Cheng, General Principles of law as Applied by International

Courts and Tribunals (reprinted edn, Cambridge University Press, 2006), 234,
236.
320 Christine D. Gray, Judicial Remedies in International law (Oxford: Clarendon

Press, 1987), 26.

321 Sedco Inc v. NIAC 10 Iran-US Claims Trib Rep 180, 1986.
322 Amoco Int'l Fin Corp v. Iran 15 Iran-US Claims Trib Rep 189, 1987.
323 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th

edn (Oxford: Oxford University Press, 2009), 331.

324 Ibid.
325 See the US Federal Court of Appeals decision in Bonar v. Dean Witter

Reynolds 837 F. 2d 1378, 1387 (11th Cir 1988). As to the desirability of a higher
award of damages for unlawful as opposed to lawful expropriation see the
separate opinion of Judge Brower in Sedco Inc v. NIOC, 10 Iran-US CTR 180, 205
(1986).
326 Mastrobuono v. Shearson Lehman Hutton Inc 514 US 52 (US S. Ct. 1995); the

award of punitive damages in this case was made for wilful fraud in the
inducement of a contract. See also Willoughby Roofing Supply Co v. Kajima
International Inc, 776 F 2d 269 (11th Cir 1985), approving awards of statutory
treble damages for antitrust violations and Mitsubishi v. Soler Chrysler-Plymouth
Inc, 473 US 614 (1985). However, in Garrity v. Lyle Stuart Inc, 40 NY 2d 354; 353
NE 2d 793 (1976) a US court held that ‘the prohibition against an arbitrator
awarding punitive damages is based on strong public policy indeed’.
327 Emmanuel Gaillard, ‘The Role of the Arbitrator in Determining the
Applicable Law’ in The leading Arbitrators' Guide to International Arbitration, ed.
L.W. Newman & R.D. Hill (Huntington: Juris Publishing, 2008), 176.
328 Jacques Werner, ‘Punitive and Exemplary Damages in International
Arbitration’, in Evaluation of Damages in International Arbitration, Dossier of the
ICC Institute of World Business law, ed. Yves Derains & Richard H. Kreindler
(Paris: ICC Publishing, 2006), 107.
329 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 651.
330 Final Award in ICC Case No. 5946, Yearbook of Commercial Arbitration, vol.

XVI, 97 (1991). Also, in a leading judgment in Germany in 1992, which has since
been affirmed, the Federal Supreme Court (Bundesgerichtshof) refused to
enforce that part of a US court decision allowing recovery of punitive damages
on the grounds that it was contrary to German public policy: Bundesgerichtshof
(Neue Juristische Wochenschrift, 1992), 3096 et seq. Redfern and Hunter
suggest German courts would react similarly to an arbitral award providing the
recovery of punitive damages: Nigel Blackaby et al., Redfern and Hunter on
International Arbitration, 5th edn (Oxford: Oxford University Press, 2009), 330–
331. In the District Court of Rotterdam, 17 February 1995, NIPR 1996, 205 et
seq. (207) a Dutch courts held that a judgment to pay punitive damages could
not be recognised and enforced in the Netherlands without further enquiry.
331 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law
International, 2009), 2487 fn 362.
332 See Andrew Barraclough & Jeff Waincymer, ‘Mandatory Laws and
International Commercial Arbitration’, Melbourne Journal of International Law 6,
no. 2 (2005); Jeff Waincymer, ‘International Commercial Arbitration and the
Application of Mandatory Rules of Law’, Asian International Arbitration Journal
5, no. 1 (2009); section 13.8.
333 Chartered Institute of Arbitrators: Guidelines for Arbitrators Dealing with

Jurisdictional Problems in International Cases 123 available at

<www.ciarb.org>.
334 Ibid.
335 There is also the question as to whether national courts might impose such

penalties as part of enforcement processes. This is outside the scope of this

book.
336 473 US 614 (1985).
337 115S.Q. 1212(1995).
338 11 ICC Court Bulletin 2000, No. 1, 57.
339 Alexis Mourre, ‘Judicial Penalties and Specific Performance in International

Arbitration’, in Interest, Auxiliary and Alternative Remedies in International

Arbitration, Dossiers ICC Institute of World Business Law, ed. Filip de Ly &
Laurent Lévy (Paris: ICC Publishing, 2008), 58.
340 Ibid., 53.
341 Ibid.
342 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 540.

343 Article 7.2.4 (Judicial Penalty): No similar provision is contained in the

Principles of European Contract Law.

344 UNIDROIT. Principles of International Commercial Contracts 2004, 215.
345 Section 25 Swedish Arbitration Act 1999.
346 Alexis Mourre, ‘Judicial Penalties and Specific Performance in International

Arbitration’, in Interest, Auxiliary and Alternative Remedies in International

Arbitration, Dossiers ICC Institute of World Business Law, ed. Filip de Ly &
Laurent Lévy (Paris: ICC Publishing, 2008), 58.
347 Carole Malinvaud, ‘Non-Pecuniary Remedies in Investment Treaty and
Commercial Arbitration’, in 50 Years of the New York Convention, ICCA
International Arbitration Conference, ICCA Congress Series no. 14, Dublin 2009),
ed. Albert Jan van den Berg (Paris: ICC Publishing, 2009), 209, 216 referring to
the conflicting views of Charles Jarrasson and Pierre Mayer. Mourre citing
Besson makes the point that if it was an inherent power, it may not be excluded
by the parties, which seems an unappealing outcome. Alexis Mourre, ‘Judicial
Penalties and Specific Performance in International Arbitration’, in Interest,
Auxiliary and Alternative Remedies in International Arbitration, Dossiers ICC
Institute of World Business Law, ed. Filip de Ly & Laurent Lévy (Paris: ICC
Publishing, 2008), 59 citing Sebastien Besson, Arbitrage international et mesures
provisoires: Etude de droit compare, Swiss Studies in International Law, vol. 105
(Zurich: Schulthess Polygraphischer Verlag, 1998), 131.
348 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 540 argue against such
an implied term.
349 L. Lévy, ‘Les Astreintes et 1'Arbitrage International en Suisse’, ASA Bulletin
19, no. 1 (2001), 24.
350 Alexis Mourre, ‘Judicial Penalties and Specific Performance in International
Arbitration’, in Interest, Auxiliary and Alternative Remedies in International
Arbitration, Dossiers ICC Institute of World Business Law, ed. Filip de Ly &
Laurent Lévy (Paris: ICC Publishing, 2008), 53.
351 Ibid, 58.
352 Paris 24 May 1991 (1992) Rev Arb 636, with a note by J Pellerin.
353 L. Lévy, ‘Les Astreintes et l'Arbitrage International en Suisse’, ASA Bulletin

19, no. 1 (2001), 32.

354 Alexis Mourre, ‘Judicial Penalties and Specific Performance in International

Arbitration’, in Interest, Auxiliary and Alternative Remedies in International

Arbitration, Dossiers ICC Institute of World Business Law, ed. Filip de Ly &
Laurent Lévy (Paris: ICC Publishing, 2008), 54.
355 Ibid.
356 There may be procedural questions as to enforcement with escalating
amounts where the primary obligation is still unmet.
357 Paris, 11 October 1991, Rev Arb 1992, 636, Note by J Pellerin.
358 Alexis Mourre, ‘Judicial Penalties and Specific Performance in International

Arbitration’, in Interest, Auxiliary and Alternative Remedies in International

Arbitration, Dossiers ICC Institute of World Business Law, ed. Filip de Ly &
Laurent Lévy (Paris: ICC Publishing, 2008), 62.
359 Ibid., 54.
360 As to the use of equitable standards to determine damages see ICJ
Administrative Tribunal case. Judgment for the Administrative Tribunal of the
ILO upon Complaints against UNESCO 1956 ICJ 75, 100 (Advisory Opinion of 23
October 1956). See Bernard Hanotiau, ‘La Détermination et L'évalutation du
Dommage Réparble: Principes Généraux et Principes en Émergence’, in
Transnational Rules in International Commercial Arbitration, Dossiers of the
Institute of International Business Law and Practice, ICC Publication No. 480/4,
ed. E. Gaillard (Paris: ICC Publishing, 1993), 209.
361 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2483 citing Dreis & Krump Mfg Co v. Int'l Assoc of
Machinists & Aerospace Workers, etc., 802 F.2d 247 (7th Cir 1986); Engis Corp. v.
Engis ltd, 800 F.Supp. 627, 632 (N.D. I11. 1992).
362 Andrea Giardina, ‘Issues of Applicable Law and Uniform Law on Interest:

Basic Distinctions in National and International Practice’, in Interest, Auxiliary

and Alternative Remedies in International Arbitration, Dossiers ICC Institute of
World Business Law, ed. Filip de Ly & Laurent Lévy (Paris: ICC Publishing,
2008), 131, 150–151.
363 John Y Gotanda, ‘Awarding Interest in International Arbitration’, American

Journal of International Law 90, no. 1 (1996): 55. See also John Y Gotanda, ‘A
Study of Interest’, in Interest, Auxiliary and Alternative Remedies in International
Arbitration, Dossiers ICC Institute of World Business Law, ed. Filip de Ly &
Laurent Lévy (Paris: ICC Publishing, 2008), 169.
364 John Y Gotanda, ‘A Study of Interest’, in Interest, Auxiliary and Alternative

Remedies in International Arbitration, Dossiers ICC Institute of World Business

Law, ed. Filip de Ly & Laurent Lévy (Paris: ICC Publishing, 2008), 186.
365 Yves Derains, ‘Conclusion’, in Evaluation of Damages in International
Arbitration, Dossier of the ICC Institute of World Business Law, ed. Yves Derains &
Richard H. Kreindler (Paris: ICC Publishing, 2006), 225.
366 John Y Gotanda, ‘A Study of Interest’, in Interest, Auxiliary and Alternative

Remedies in International Arbitration, Dossiers ICC Institute of World Business

Law, ed. Filip de Ly & Laurent Lévy (Paris: ICC Publishing, 2008), 131, 170.
367 Charles N. Brower & Jeremy K. Sharpe, ‘Awards of Compound Interest in

International Arbitration: The Aminoil Non-Precedent’, in Global Reflections on

International law, Commerce and Dispute Resolution: liber Amicorum in Honour of
Robert Briner, ed. Gerald Aksen et al. (ICC Publishing, 2005), 156.
368 John Y Gotanda, ‘Awarding Interest in International Arbitration’, American

Journal of International law 90, no. 1 (1996): 40, 57.

369 John Y Gotanda, ‘A Study of Interest’, in Interest, Auxiliary and Alternative

Remedies in International Arbitration, Dossiers ICC Institute of World Business

Law, ed. Filip de Ly & Laurent Lévy (Paris: ICC Publishing, 2008), 131, 174.
370 Australian International Arbitration Act ss 25–26; Hong Kong Arbitration

Ordinance Art. 79, 80.

371 North American Free Trade Agreement, 17 December 1992, Art. 1135(1),

32ILM (1993) 289, 646 (entered into force 1 January 1994).

372 A notable exception is the LCIA Rules which allows for compound interest

Art. 26.6 LCIA Rules.

373 AAA Arbitration Rules Art. 28.4.
374 John Yukio Gotanda, Supplemental Damages in Private International haw: The

Awarding of Interest, Attorneys' Fees and Costs, Punitive Damages and Damages in
Foreign Currency Examined in the Comparative and International Context (The
Hague: Kluwer Law International, 1998).
375 Islamic Republic of Iran v. United States of America, Decision No Dec65-A19-

F3 (30 September 1987) (1998) 16 Iran-US CTR 285, 289–290; Plywood Trade
Arbitration (Syria v. Ghana) ICC Case No. 4237, Award of 17 February 1984,
excerpted in 1985 Yearbook of Commercial Arbitration 52, 59–60.
376 J.D.N. Lew, ‘Interest on Money Awards in International Arbitration’, in
Making Commercial Law: Essays in Honour of Roy Goods, ed. Ross Cranston
(Oxford: Oxford University Press, 1997), 558. Charles M. Brower & Jeremy K.
Sharpe, ‘Awards of Compound Interest in International Arbitration: The Aminoil
Non-Precedent’, in Global Reflections on International Law, Commerce and
Dispute Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen
et al. (ICC Publishing, 2005), 155.
377 Islamic Republic of Iran v. United States of America, Decision No DEC65-A19-

FT (30 September 1987), (1998) (16 Iran US CTR 285 at 289-290, cited in
Brower and Sharpe, ibid., 155).
378 Charles M. Brower & Jeremy K. Sharpe, ‘Awards of Compound Interest in

International Arbitration: The Aminoil Non-Precedent’, in Global Reflections on

International Law, Commerce and Dispute Resolution: Liber Amicorum in Honour
of Robert Briner, ed. Gerald Aksen et al. (ICC Publishing, 2005), 156; J.G. Wetter,
‘Interest as an Element of Damages in the Arbitral Process’, International
Financial Law Rev 5, no. 12 (1986): 20, 21.
379 John Y Gotanda, ‘Awarding Interest in International Arbitration’, American

Journal of International Law 90, no. 1 (1996): 40, 57.

380 ICC Case No. 9771 of 2001 cited in J.D.N. Lew, ‘Interest on Money Awards in

International Arbitration’, in Making Commercial Law: Essays in Honour of Roy

Goods, ed. Ross Cranston (Oxford: Oxford University Press, 1997), 543.
381 J.D.N. Lew, ‘Interest on Money Awards in International Arbitration’, in
Making Commercial Law: Essays in Honour of Roy Goods, ed. Ross Cranston
(Oxford: Oxford University Press, 1997), 543.
382 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th

edn (Oxford: Oxford University Press, 2009), 540.

383 See Arbitration Act 1996 (UK) s. 49.
384 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th

edn (Oxford: Oxford University Press, 2009), 541. (They provide Bermuda, Hong
Kong, England, and Scotland as examples.)
385 While the CISG refers to interest in Art. 78, it provides little detail as to

timing and rate as the contracting States would not have been able to agree on a
set of principles.
386 John Y Gotanda, ‘Awarding Interest in International Arbitration’, American

Journal of International Law 90, no. 1 (1996): 40, 52.

387 See, e.g., Guinea v. Maritime International Nominees Establishment – MINE

(Liechtenstein), ICSID Decision on Application for Annulment (14 December

1989), Yearbook of Commercial Arbitration, vol. XVI40, 45, 51 (1991). ICC Case
No. 6527 of 1991, Final Award in Yearbook of Commercial Arbitration, vol. XVIII
(1993), 44, ICC Award in Case No. 9448 of July 1999, Final Award.
388 See, e.g., Dr Horst Reineccius v. Bank for Int'l Settlements, Final Award in

Permanent Court of Arbitration (19 September 2003), Yearbook of Commercial

Arbitration, vol. XXVIII, 100, 151-152 (2003).
389 See, e.g., Grove-Skanska v. Lockheed Aircraft International AG, ICC Award No.

3903, cited in John Y Gotanda, ‘Awarding Interest in International Arbitration’,

American Journal of International Law 90, no. 1 (1996): 40, 51 where the
arbitrators considered that the New York laws as selected only related to court
actions. The tribunal relied on general principles of international law.
390 Nayla Comair-Obeid, ‘Recovery of Damages for Breach of an Obligation of

Payment’, in Evaluation of Damages in International Arbitration, Dossier of the

ICC Institute of World Business Law, ed. Yves Derains & Richard H. Kreindler
(Paris: ICC Publishing, 2006), 146.
391 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2505 and cases cited in fn. 466.

392 Interim Award in ICC Case No. 5277, Yearbook of Commercial Arbitration,

vol. XIII, 80, 89–90 (1988). Abdul Hamid el-Ahdab, ‘Enforcement of Arbitral
Awards in the Arab Countries’, Arbitration International 11, no. 2 (1995): 169.
393 See Samir Saleh, ‘The recognition and enforcement of foreign arbitral awards

in the States of the Arab Middle East’, in Contemporary Problems in International

Arbitration, ed. Julian D.M. Lew (Dordrecht: Martinus Nijhoff Publishers, 1986),
348, 349. However, see the decision of the English courts in Sanghi Polyesters
Ltd (India) v. The International Investor KCFC (Kuwait) [2000] 1 Lloyd's Rep 480,
which suggests that, in some Islamic jurisdictions, interest may be awarded
under another name.
394 Nayla Comair-Obeid, ‘Recovery of Damages for Breach of an Obligation of

Payment’, in Evaluation of Damages in International Arbitration, Dossier of the

ICC Institute of World Business Law, ed. Yves Derains & Richard H. Kreindler
(Paris: ICC Publishing, 2006), 141. Saudi Arabia maintains the strictest regime
opposing the potential for interest.
395 John Y. Gotanda, ‘Awarding Interest in International Arbitration’, American
Journal of International Law 90, no. 1 (1996): 40, 47–48.
396 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th

edn (Oxford: Oxford University Press, 2009), 543–544.

397 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2504, fn 460. Interest was routinely awarded against Iran
and claimed by Iranian claims and counterclaims before the US-Iran Claims
Tribunal. John Y. Gotanda, ‘Awarding Interest in International Arbitration’,
American Journal of International Law 90, no. 1 (1996): 40, 49–50.
398 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th

edn (Oxford: Oxford University Press, 2009), 544.

399 Ibid.
400 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 657.
401 Siemens AG v. The Argentine Republic, ICSID Case No. ARB/02/8, Final Award

of 6 February 2007, 126–127.

402 John Y. Gotanda, ‘Awarding Interest in International Arbitration’, American

Journal of International Law 90, no. 1 (1996): 40, 60.

403 See, e.g., ICC Award 7622, ICC International Court of Arbitration Bulletin

15(1) (2004) 79.

404 4 Iran-US Claims Tribunal Rep 263, 267–268 (1983).
405 Final Award ICC Case No. 8874, ICC International Court of Arbitration Bulletin

Vol. 10 No. 2, 1999, 85.

406 Nayla Comair-Obeid, ‘Recovery of Damages for Breach of an Obligation of

Payment’, in Evaluation of Damages in International Arbitration, Dossier of the

ICC Institute of World Business Law, ed. Yves Derains & Richard H. Kreindler
(Paris: ICC Publishing, 2006), 136.
407 See, e.g., s. 25 International Arbitration Act (Australia).
408 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2506.

409 ICC Case No. 2930, Final Award (1982) reprinted in 9 Yearbook of
Commercial Arbitration 105 (1984).
410 Final Award in Case No. 5460 of 1983, reprinted in 13 Yearbook of
Commercial Arbitration 104, 106 (1988).
411 Compare Case No. 5731, ICC International Court of Arbitration Bulletin Vol. 3

No. 1, 1992; Case No. 7078, ICC Final Award, ICC International Court of
Arbitration Bulletin Vol. 15 No. 1, 70; Case No. 5324 of 1989, ICC Final Award,
ICC International Court of Arbitration BulletinWoX. 3 No. 2, 1992, 53; ICC
International Court of Arbitration Bulletin Vol. 10 No. 2, 1999, 77; ICC Final
Award, Case No. 7063 of 1993, ICC Arbitral Awards, 1996–2000, 75; Yearbook of
Commercial Arbitration Final Awards, Case No. 9771 (2001) Yearbook of
Commercial Arbitration, vol, XXIX, 2004, 46.
412 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 655. They also state in footnote 169, 656, that ‘there is a general
presumption that interest accrues at a rate corresponding to that of a commonly
used saving vehicle in the currency in which payment is to be made and is to be
compounded quarterly’.
413 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th
edn (Oxford: Oxford University Press, 2009), 541. Section 49(3) of the English
Arbitration Act 1996, empowers a tribunal seated in England to award interest
‘from such dates, at such rates and with such rests as it considers meets the
justice of the case’. Almost identical provisions are found in the Irish Arbitration
(International) Commercial Act 1998 s. 10(2). Australian law permits a tribunal
to award interest ‘at such reasonable rate as the tribunal determines for the
whole or any part of the money, for the whole or any part of the period between
the date on which the cause of action arose and the date on which the award is
made’ and thereafter ‘from the day of the making of the award or such later day
as the tribunal specifies, on so much of the money as is from time to time
unpaid’: Australian International Arbitration Act ss 25(1)and 26; equivalent
provisions are found in the Maltese Arbitration Act 1996 ss 63(1) and 64. Other
jurisdictions such as Hong Kong, India, and Singapore have enacted laws giving
arbitrators similar discretion in the award of interest: Hong Kong Arbitration
Ordinance 2011 ss 79 and 80; Indian Arbitration and Conciliation Act 1996 s.
31(7)(a) and (b); Singapore International Arbitration Act (Ch 143A) 2002 ss
12(5)(b) and 20.
414 John Y Gotanda, ‘A Study of Interest’, in Interest, Auxiliary and Alternative

Remedies in International Arbitration, Dossiers ICC Institute of World Business

Law, ed. Filip de Ly & Laurent Lévy (Paris: ICC Publishing, 2008), 131.
415 See, e.g., ICC Case No. 3572, Final Award 1982, reprinted in 14 Yearbook of

Commercial Arbitration 111 (1989).

416 See, e.g., Asian Agricultural Products Ltd v. Sri Lanka, reprinted in 17
Yearbook of Commercial Arbitration 106 (1992).
417 ICC Final Award of April 1997, Case No. 8521, ICC International Court of

Arbitration Bulletin Vol. 15 No. 1, 2004, 90.

418 McCollough & Co v. Ministry of Post, Telegraph and Telephone 11 Iran-US

Claims Tribunal Reports at 28.

419 See, e.g., Iran-US Claims Tribunal Case 64 of 1985 (fn. 103).
420 Sylvania Technical Systems Inc v. Government of the Islamic Republic of Iran,

Award No. 180-64-1 17 June 1985 in (1985) 8 Iran-US CTR 298, 320.
421 Matthew Secomb, ‘A Uniform, Thee-Step Approach to Interest Rates in

International Arbitration’, in International Arbitration and International

Commercial law: Synergy, Convergence and Evolution, liber Amicorum Eric
Bergsten, ed. S. Kröll et al. (Alphen aan den Rijn: Kluwer Law International,
2011), 432.
422 The UNIDROIT Principles were applied in Petrobart ltd v. Kyrgyzstan, SC3

Case No. 126/2003 113 184, 29 March 2005.

423 2000/35/EC of 29 June 2000 OJ L 200 8 August 2000 Art. 3.
424 See Andrea Giardina, ‘Issues of Applicable Law and Uniform Law on Interest:

Basic Distinctions in National and International Practice’, in Interest, Auxiliary

and Alternative Remedies in International Arbitration, Dossiers ICC Institute of
World Business Law, ed. Filip de Ly & Laurent Lévy (Paris: ICC Publishing,
2008), 131, 150–151.
425 John Y Gotanda, ‘A Study of Interest’, in Interest, Auxiliary and Alternative

Remedies in International Arbitration, Dossiers ICC Institute of World Business

Law, ed. Filip de Ly & Laurent Lévy (Paris: ICC Publishing, 2008), 131, 188;
Hungerfordv. Walker (1989) 171 CLR 125 at 143.
426 See, e.g., Mqffezini v. Kingdom of Spain, ICSID Case No. ARB/97/7, 13
November 2000, §§96.
427 John Y Gotanda, ‘A Study of Interest’, in Interest, Auxiliary and Alternative

Remedies in International Arbitration, Dossiers ICC Institute of World Business

Law, ed. Filip de Ly & Laurent Lévy (Paris: ICC Publishing, 2008), 131, 188.
428 Thierry J Fénéchal, ‘Present-Day Valuation in International Arbitration: A

Conceptual Framework for Awarding Interest’, in Interest, Auxiliary and

Alternative Remedies in International Arbitration, Dossiers ICC Institute of World
Business Law, ed. Filip de Ly & Laurent Lévy (Paris: ICC Publishing, 2008), 131,
219.
429 A contrary analysis was presented by F. A. Mann, ‘Compound Interest as an

Item of Damage in International Law’, in F.A. Mann, Further Studies in

International Law (New York: OUP, 1990), 377 at 381; F.A. Mann, ‘Compound
Interest as an Item of Damage in International Law’, Davis LR 21, no. 3 (1998):
577.
430 Holtzmann J in Starrett Housing Corporation v. The Government of the Islamic

Republic of Iran, Award No. 314-24-1, 14 August 1987, reprinted in 16 Iran-US

CTR 112 at 254.
431 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th

edn (Oxford: Oxford University Press, 2009), 534. They note that this view was
reached in 1983 by Judge Howard Holtzmann in a dissenting judgment in
Starrett Housing Corporation v. Iran, Iran–US CTR 122, 269. Further, they note
the view of Gotanda that ‘almost all financing and investment vehicles involve
compound interest … if the claimant could have received compound interest
merely by placing its money in a readily available and commonly used
investment vehicle, it is neither logical nor equitable to award the claimant only
simple interest’: John Y. Gotanda, ‘Awarding Interest in International
Arbitration’, American Journal of International Law 90, no. 1 (1996): 40, 61.
432 Redfern and Hunter note the example of Art. 38 of the International Law

Commission's Draft Articles on Responsibility of States for Internationally

Wrongful Acts, providing that ‘interest shall be payable on any principal sum
when necessary in order to ensure full reparation’: Official Records of the
General Assembly, F50, 6th Session, Supplement No. 10 (AR/56/10), Ch IV, E.l).
433 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th

edn (Oxford: Oxford University Press, 2009), 543.

434 Compañia del Desarollo de Santa Elena SA v. Republic of Costa Rica (2000) 39

ILM 1317, para. 103.

435 Wena Hotels Limited v. Egypt, Award, ICSID Case No. ARB/98/4; IIC 273

(2000); 41 ILM 896 (2002).

436 Redfern and Hunter cite the following examples, available online at
<http://icsid.worldbank.org> and <www.investmentclaims.com>: with respect
to NAFTA claims, Pope Talbot v. Government of Canada, Award on Damages, Ad
hoc UNCITRAL Rules; IIC 195 (2002); and SD Myers Inc v. Government of Canada,
Second Partial Award, Ad hoc UNCITRAL; IIC 250 (2002). Other awards include:
Middle East Cement Shipping and Handling G SA v. Arab Republic of Egypt, Award,
ICSID Case No. ARB/99/6); IIC 169 (2002); CME Czech Republic BVv. The Czech
Republic, Final Award and Separate Opinion, Ad hoc-UNCITRAL Rules; IIC 62
(2003); Azurix Corporation v. The Argentine Republic, Award, ICSID Case No.
ARB/01/ 12; IIC 24 (2006); ADC Affiliate Ltd, ADC & ADMC Management Limited
v. The Republic of Hungary, Award, ICSID Case No. ARB/03/16, IIC 1 (2006);
Siemens AG v. The Argentine Republic, Award, ICSID Case No. ARB/02/8, IIC 227
(2007); and LG&E Energy Corp, LG&E Capital Corp, and LG&E International Inc v.
The Argentine Republic, Award, ICSID Case No. ARB/02/1; IIC 295 (2007).
437 Charles M. Brower & Jeremy K. Sharpe, ‘Awards of Compound Interest in

International Arbitration: The Aminoil Non-Precedent', in Global Reflections on

International Law, Commerce and Dispute Resolution: Liber Amicorum in Honour
of Robert Briner, ed. Gerald Aksen et al. (ICC Publishing, 2005), 155.
438 Ibid.
439 For example, in England, Ireland, Hong Kong, and Bermuda. See the English

Arbitration Act 1996 s. 49, which provides for the award of simple or compound
interest unless the parties agree otherwise.
440 Australia and New Zealand are good examples.
441 This is the case in Canada and the US.
442 For example, AAA International Arbitration Rules Art. 38.4; LCIA Rules Art.

26.6; WIPO Arbitration Rules Art. 60(b).

443 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th

edn (Oxford: Oxford University Press, 2009), 542.

444 Under the Dutch and Japanese civil codes, ‘statutory interest is automatically

capitalised at the end of each year’: Nigel Blackaby et al., Redfern and Hunter on
International Arbitration, 5th edn (Oxford: Oxford University Press, 2009), 542.
445 Thierry J Fénéchal, ‘Present-Day Valuation in International Arbitration: A

Conceptual Framework for Awarding Interest’, in Interest, Auxiliary and

Alternative Remedies in International Arbitration, Dossiers ICC Institute of World
Business Law, ed. Filip de Ly & Laurent Lévy (Paris: ICC Publishing, 2008), 131,
219.
446 ICC Arbitral Awards, 1986–1990, 394.
447 See, e.g., Flourmill Insurance Arbitration, 1981 Yearbook of Commercial
Arbitration 121, 124.
448 See, e.g., I J Reynolds Tobacco Co v. Iranian Tobacco Co, 1985 Yearbook of

Commercial Arbitration 258, 261.

449 See, e.g., lianco v. libya, 62 ILR 140 (1982), 20 ILM 1 (1981).
450 See, e.g., ICC Case No. 6162 of 1990 and Case No. 10329 of 2000. ICC Case No.

6162, Yearbook of Commercial Arbitration 1990, 85; ICC Case No. 10329 of 2000,
Yearbook of Commercial Arbitration Final Award.
451 Different common law cases have taken different views on this issue. See Mr

Justice Vivian Ramsey, ‘Problems of Delay and Disruption Damages in

International Construction Arbitration’, in Evaluation of Damages in
International Arbitration, Dossier of the ICC Institute of World Business law, ed.
Yves Derains & Richard H. Kreindler (Paris: ICC Publishing, 2006), 207–208.
452 Yearbook of Commercial Arbitration, vol, XVII, 1992, 221; ICC Case No. 6754

of 1993, ICC Arbitral Awards 1991–1995, 600; ICC Case No. 7331 of 1994 and
ICC Case No. 7078.
453 2000/35/EC of 29 June 2000 OJ L 200 8 August 2000.
454 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th

edn (Oxford: Oxford University Press, 2009), 544.

455 Ibid.
456 See, e.g., ICC Case No. 8264 of 1997.
457 In England, s. 49(3) of the England Arbitration Act 1996 expressly permits

the arbitral tribunal to exercise its discretion to award interest up to the date of
payment.

Part III: The Award, Chapter 15: Costs

in Arbitration
Jeff Waincymer,

15.1. Introduction and Policy

The awarding of costs in international arbitration is a troubling topic for a

number of reasons. First, there is no consensus among legal families as to the
way costs should be awarded in adjudicatory processes. Some follow a loser-
pays principle; (1) at the other extreme some require each party to bear its own
costs; some adopt a mid-position, differentiating between different types of
costs, and only awarding some, the most contentious being legal fees and
expenses. (2) A key reason for the differences is that there are important trade-
offs between fairness and efficiency under any costs model that make it difficult
to identify an optimal policy prescription. The promotion of efficiency calls for
controls of both time and cost by parties and arbitrators, (3) and suggests some
limitation on the way cases are to be prepared and presented. Yet this needs to
be tempered with considerations of fairness and the pursuit of truth, which
imply that parties should be given a full opportunity to present potentially costly
submissions if they could materially affect the result. That opportunity needs to
be viewed from the perspective of the parties at the page "1191" outset, when
their ultimate aim is to win the case. At that point in time it will often be perfectly
reasonable to adopt a broad ranging set of strategies in order to win, even
though some will ultimately be unsuccessful.

Economic analysis also suggests that whatever costs rules are employed, this will
have a significant impact on the kinds of claims brought and hence the essential
character of the particular adjudicatory system. (4) Some argue for rules which
promote particular investment outcomes. (5) The analysis is further complicated
because some see fairness largely from the winner's perspective, implying the
desirability of a full indemnity of costs incurred by the winner akin to a damages
entitlement. Others see the issue primarily from the loser's perspective in the
context of determining a fair contribution from it as to costs. To some, this
requires no more than a reasonable cost contribution to the winner's expenses.
It may even provide support for the view that each party should bear its own
transaction costs of the adjudicatory process, just as they accept the time loss
and stresses as dead-weight costs without expectations of compensation.
Considering both perspectives simultaneously leads to an immediate conflict
between these considerations.

While these are all challenging conceptual issues that militate against a
consensus position as to arbitral best practice, there is an important need to find
practical and consistent solutions for two key reasons. First, it has long been
argued that the total cost of arbitration has grown to a level which is causing
dissatisfaction amongst its users. (6) The suggestion is that tribunals need to have
a principled basis on which to contain costs. Second, there is little consistency
and hence little predictability as to the attitudes of arbitral tribunals to cost
awards and cost control strategies. (7) Uniform principles in deciding on which
costs to award would thus be desirable.

page "1192"

Michael Bühler suggests that there are two main reasons leading to the
uncertainty and lack of consensus about the type and level of costs that should be
awarded. (8) He suggests that not all arbitrators adopt the same principles in
determining which party should pay which costs and second, that they take
different views on which categories of costs are reimbursable. As noted above, a
third reason is that there is no consensus as to whether arbitral tribunals should
seek to impose reasonable limits based on efficiency concerns and/or fairness to
the loser, or to allow all costs so as to indemnify the winner.

A lack of predictability is likely to add to dissatisfaction when the final outcome

is known and is also likely to operate as a disincentive to settlement. This is
because the less predictable the outcome of any adjudicatory process, the further
apart the parties are likely to be in settlement negotiations and hence the less
successful those negotiations are likely to be. However, tribunal and institutional
fees are generally a small percentage of counsels' fees, hence the key element can
be as certain as the parties and their counsel wish. Nevertheless, tribunal and
institutional fees are not insignificant and the latter are a key driver as to which
institution is selected. A lack of predictability on cost allocation may also impact
upon decisions taken during the proceedings. An example would be a party
deciding whether to advance the costs unpaid by the other party where the latter
chooses to refrain from involvement in proceedings. If the party wishing to
continue does not know whether it will be reimbursed that cost, this will be a
disincentive to them choosing to continue the proceedings. The other problem
with a lack of certainty is that because parties, counsel and arbitrators may all
come from differing legal cultures, they may have vastly different expectations as
to the treatment of costs, which could also lead to unnecessary tension when the
disparity of views comes to the fore. When uncertainty in any legal system is too
acute, it affects the overall respect for and ultimately the legitimacy of the
system.

However, any attempt to be too prescriptive as to the way costs should be

allocated might prevent justice being done on a case-by-case basis through
proper consideration of a range of variable factors. An overly prescriptive regime
would also inevitably involve a priori choices between conflicting meritorious or
at least tenable values, such as the wish to adequately compensate winners but
also provide reasonable limits as to what a loser should be asked to pay and the
separate wish to consider procedural behaviour as well as relative success in
deciding who deserves compensation. There would be no consensus as to the
way these a priori choices should be made.

In addition to these broad policy challenges, there is the added problem of

determining the powers and duties of arbitrators in seeking to control costs and
how their wish to do so may be impacted upon by party autonomy and/or lex
arbitri and/ or potential annulment or enforcement challenges. Any attempt to
control costs has page "1193" to be considered both in terms of express and
implied powers and also subject to mandatory due process norms that would
prevent undue interference. The issues raised in this chapter thus overlap with
the broader issue of tribunal control over procedure considered in Chapter 6 and
the rights and duties of arbitrators considered in Chapter 2.

This chapter discretely considers what powers a tribunal has to award or

otherwise control costs, how a tribunal ought to award costs, how tribunals
otherwise control proceedings to promote cost effectiveness, and where
institutions are involved, how they may also engage in cost control.

15.2. Arbitral Laws and Rules

15.2.1. Is It a Question of Procedure or Substance?

In considering the rights and duties as to costs in arbitration, there is a

preliminary issue as to whether this is a procedural or substantive question, as
this determines which arbitral rules and principles will be relevant. It is
generally accepted that determinations as to costs and fees are matters of
procedure, (9) although a number of US courts have still divided as to whether
such awards are governed by procedural or substantive law. (10) Some see a
distinction between the general power to make costs awards, a commonly
agreed procedural matter and second, the standards to apply. For example, the
key justification for a loser-pays approach is substantive, in seeking to fully
indemnify the winner and put it in a position it would have been in financially
but for the default on the part of the losing party.

As to the proper standards, while Born accepts that the authority to award costs
is procedural, he notes that it has been suggested that the substantive law
governing the parties' underlying contract or dispute ought to provide the
standard for awards of legal costs. In some cases the applicable substantive law
might also have an express provision in relation to costs. (11) Nevertheless, Born
advocates the adoption of international standards. He suggests that the sui
generis standards of institutional page "1194" rules and arbitral practices
typically ‘provide that (a) the prevailing party is presumptively entitled to a costs
award; (b) only reasonable costs will be reimbursed; and (c) expenses that were
inefficient or unnecessary will not be reimbursed, while costs resulting from the
need to respond to unreasonable or uncooperative actions will be recoverable’.
(12) Such standards are as arguable as anything as an optimal trade-off between

fairness and efficiency.

15.2.2. Costs versus Damages

While costs powers will largely depend upon an analysis of the lex arbitri,
arbitral rules and any agreement between the parties, a claimant might also look
to the substantive law of damages for breach as an alternative ground for
seeking indirect compensation for costs. The damages approach is of course not
procedural and would depend on the substantive law as to damages entitlement
and quantum. Key issues would be the applicable test of causation and whether a
party's choices as to cost expenditure break the chain of causation. There would
also be questions as to the duty to mitigate and how this impacts upon choices in
the proceedings.

One suggestion is that this may at least arise in circumstances where procedural
law does not allow for redress. (13) It is not clear why the entitlement should be
limited to these circumstances alone as a damages right is an independent right
if made out, although double compensation should obviously not be permissible.
Furthermore, while parties might sensibly be more inclined to consider claims
for consequential damages where there are procedural limits on costs awards,
there is no guarantee that this would be effective. While unlikely, if there was a
clear proscription against certain costs in the lex arbitri, (14) a tribunal may well
take a substance over form approach and deny an award of costs when
recharacterised as consequential damages. The better view should be that each
claim is treated on its merits.

Examples of circumstances where costs might be properly characterised as

damages include claims for breach of the arbitration agreement itself if a party
subject to it nevertheless attempts to commence court proceedings. Each
adjudicator would again need to ensure that there is no double dipping between
a costs claim in the arbitration and any costs sought from the court in an
application to bar page "1195" the litigation. (15) A party who was forced to pay
the whole advance on costs might also sue for recovery of a half share on the
basis that the other party is in breach of the arbitration agreement. Such a claim
might be easier with institutional arbitration where an agreement to the
institution constitutes agreement to a specific costs schedule, although a well-
drafted ad hoc arbitrator's contract might lead to the same outcome. This is
discussed further in section 6.11.

The consequential damages route only applies to a successful claimant and not a
respondent who merely defends a claim. (16) Where the respondent successfully
defends, its claim for costs must flow from the procedural rules of the arbitration
itself as it is not based on any breach by claimant, (17) unless the claim was seen
as vexatious or frivolous, in which case, damages may have relevance. (18)

15.2.3. Cost Criteria in Arbitral Statutes and Rules

Because of the lack of consensus among domestic jurisdictions, it is

understandable that drafters of international arbitral rules provided little
guidance on the approach to costs issues, although virtually all provide
discretion for these to be awarded. Before considering a range of models, it is
important to consider their status as either mandatory rules or conversely, those
that are subject to contrary party agreement.

15.2.4. Mandatory Rules as to Costs

Generally speaking, there are few, if any, express mandatory rules in relation to
cost determinations by tribunals. This means that the contents of the lex arbitri
and arbitral rules can be varied by party agreement. Section 60 of the English
Arbitration Act 1996 is an exception and provides that ‘(a)n agreement which
has the effect that a party is to pay the whole or part of the cost of the arbitration
in any event is only valid if made after the dispute in question has arisen’. (19) Pre-
dispute cost agreements of this nature would even exclude each party agreeing
in advance of a dispute to bear their own expenses and share tribunal costs. (20)

page "1196"

Due process norms such as Article 18 of the UNCITRAL Model Law are also
mandatory. They impact directly on tribunal procedural discretions aiming to
promote efficiency. It is at least arguable that they also indirectly impact upon
efficiency-based cost determinations if a party could legitimately argue that it
was induced to refrain from availing itself of due process rights based on a
warning by the tribunal as to an adverse cost determination were it to do so.

15.2.5. Costs Provisions, Discretions and Discretionary Criteria

While most institutional rules ultimately leave cost decisions to the discretion of
the arbitrator, they vary between those that are wholly silent on the matter of
costs, those that simply provide a broad discretion, (21) and those that provide
guidance on the criteria for awarding costs. (22) The latter typically rely on
presumptions such as that costs follow the event or that costs should be awarded
in proportion to relative success. Even where rules raise a presumption that
loser pays (at least in part), these will invariably empower an arbitrator to
depart from the presumption in appropriate circumstances. Some also seek to
limit and/or articulate the range of costs to be considered. The rules may also
seek to impose limits as to level of costs, most particularly allowing only for the
award of ‘reasonable’ or ‘necessary’ costs.

The working group dealing with the UNCITRAL Model Law did not consider that
costs were appropriate to be dealt with in the Model Law. (23) While the Model
Law is itself silent as to costs, a number of States that have adopted the Model
Law have added provisions expressly granting tribunal jurisdiction over costs
awards. (24) There is also no reference to costs in Swiss or French arbitral
legislation. The US Federal Arbitration Act does not mention costs but some US
state laws do, as do the ICDR Rules. The proper approach where the lex arbitri is
silent as to costs is considered in section 15.2.7 below. Rebuttable presumptions
that costs follow the event are included in English and German arbitral
legislation. (25) Similar presumptions are contained in various arbitration rules.
(26) Article 42 UNCITRAL Rules 2010 states that the costs of arbitration shall in

principle be borne by the page "1197" unsuccessful party. Nevertheless, it has

been noted that there remains a broad discretion under the UNCITRAL Rules. (27)
The Japanese Arbitration Law provides for the opposite presumption, namely
that each party shall bear the costs it has disbursed with respect to the arbitral
proceedings, absent contrary agreement. (28)

Where some form of loser pays is involved, there is some difference between the
English Arbitration Act 1996 referring to costs following ‘the event’ and Article
28.4 LCIA Rules which speaks about ‘relative success and failure’. The difference
may be between awarding costs to the party determined to be the overall winner
under the Act or alternatively, looking at it from an issue-by-issue basis under
the Rules, although it is also arguable that costs following the event is a vague
expression which can be interpreted to also encompass relative success
considerations. These two approaches are discussed further in sections 15.7 and
15.8 below. Some rules are drafted in a way which makes it arguable that costs
following the event is the defining factor even though this is not directly
articulated. For example, Article 44 of the SCC Rules indicates that the tribunal
may at the request of a party, order one party to pay another party's costs
‘having regard to the outcome of the case’. Similarly, Article 31 of the ICDR Rules
defines legal costs as ‘the reasonable costs for legal representation of a successful
party.’ (emphasis added)

Article 37 of the ICC Rules 2012 does not contain a presumption that costs follow
the event but indicates expressly that one relevant factor may be ‘the extent to
which each party has conducted the arbitration in an expeditious and cost-
effective manner’ . (29) As noted, some rules also seek to limit the level to
reasonable costs. Non-mandatory provisions of the English Arbitration Act 1996
provide in principle for limitations of recoverable costs and fees to reasonable
amounts. Doubts as to reasonableness are to be resolved in favour of the paying
party. (30) Article 41 of the UNCITRAL Rules 2010 allows for a party to refer
tribunal fees to an appointing authority to determine if the fees are inconsistent
with the tribunal's proposal or are otherwise ‘manifestly excessive’, in which
case the authority can make binding adjustments.

Finally, some rules seek to articulate the various categories of costs to be

considered. In some cases this is defined exhaustively. (31) Some rules also seek to
distinguish between the treatment of different types of costs. For example, the
difference in attitude between legal families as to tribunal and party costs,
particularly as to legal representation, was thought to be reflected in the way the
UNCITRAL Rules were formerly divided. (32) Articles 40-42 of the UNCITRAL
Rules 2010 no longer maintains any presumptive distinction.

page "1198"

15.2.6. The Relevance of Domestic or National Laws

It has been suggested above that costs issues are procedural and that the arbitral
statutes and rules applicable should be the gateway to determining costs issues.
For non-mandatory rules, party autonomy is thus controlling. As noted above,
Born has also advocated the application of sui generis international principles.
Based on these approaches, while it would be undesirable for a tribunal to apply
domestic law in considering costs matters, domestic practices may influence
their discretion from time to time.

In some cases this would arguably be inappropriate. For example, in Him-purna,

(33) legal representation costs were not awarded because ‘recovery of significant

legal costs is foreign to the legal system of Indonesia, where the parties chose to
hold the arbitration’. Arguably, the domestic litigation law of the place of
arbitration itself ought not to be determinative. Certainly if the parties to an
arbitration with a seat in Indonesia were American and Indonesian respectively,
the same conclusion could be based on their understanding of the common
norms in each country's court processes and hence their likely expectations.
Even then they may have selected arbitration in large part to opt instead for the
international norm, where costs often follow the event. However, given the size
of the award and the decision to resolve many complex questions in favour of
Indonesia, the outcome may well have been a very sensible and practical one.

One undesirable inroad from domestic practice would be if arbitrators simply

adopt the approach they are familiar with domestically. Derains and Schwarz,
speaking of ICC practice, suggest that national background often influences
arbitrators. (34) While domestic laws do not apply as of right and might at times
be inappropriately influential, at times they can aid in the analysis. For example,
it is inevitable that modern litigation developments based on comprehensive
reform analysis might have influence over time. Thus, civil procedure rules such
as those incorporated in England following the Woolf reforms can influence the
kind of logical arguments that might be put to a tribunal with a broad discretion.
There is nothing problematic in this as long as it is based on adequate reasoning
with proper attention to whether the principles should properly be applied to
international arbitration. In this context, the English Civil Procedure Rules
introduced in 1999 now call for a court to have regard to all of the circumstances
of the case including relative success, party conduct, payments into court and
admissible offers of settlement, conduct before proceedings, including time
limits of raising issues, and accuracy of the claim. The court is also asked to
generally consider whether the costs incurred were proportional and
reasonable. (35)

page "1199"

Price and Stans argue that many of these principles should be applied in
international arbitration. (36) This section is not arguing for or against that
proposition, but simply suggesting that counsel might well ask a tribunal to
consider the reasoning behind a carefully considered domestic reform analysis.

Domestic laws might also indirectly impact on certain types of costs. For
example, the residence of counsel may be relevant, given that legal fees will
typically be calculated based on national practices. (37) This may be relevant as a
test of reasonableness where that is a criterion, although matching international
levels when domestic ones are lower is not necessarily unreasonable.

15.2.7. What are the Powers If there is No Mention of Costs in the Lex Arbitri
or any Selected Arbitral Rules?

Note was taken above of rules such as the UNCITRAL Model Law and US, Swiss
and French statutes that make no mention of powers over costs. Here, there is a
foundational question as to whether a tribunal nevertheless has a residual
discretion to award costs. The question assumes no specific power emanating
from the arbitration agreement or selected rules. Because arbitrators are always
given an overriding express discretion to determine procedure unless proscribed
in the lex arbitri, rules, or party agreement, it is commonly accepted that these
general procedural discretions are broad enough to cover costs awards. (38) Born
also suggests that the power to award costs can be implied from the parties'
arbitration agreement, citing Bühler and Webster, (39) although this would
depend on the method of interpreting that agreement, including an analysis of
legitimate expectations to that end (40) and the background of the parties. Born
further suggests that this ‘is a natural and inherent aspect of page "1200" the
tribunal's authority (absent contrary agreement)’, which if correct, supports the
implied agreement analysis. (41)

The situation may be different if the particular jurisdiction generally does not
support costs awards to the winning party. Jurisdictions that may not have
viewed costs as an inherent adjudicatory power would typically provide express
powers in their arbitration statutes. (42) As noted, the Federal Arbitration Act (US)
has no provision dealing with allocation of costs and attorneys' fees. Because of
the broad support in that jurisdiction for the rule requiring each party to bear
their own costs, this has led some courts to consider that arbitrators might lack
the authority to award such fees. Courts in the US have at times overturned
awards on the basis that awards of costs go beyond arbitral power, although
some courts have taken the opposite approach, concluding that an arbitral
tribunal has an inherent power unless prevented by contract and Born,
discussing these cases, suggests that the American rule should not apply just
because the parties have selected a seat in the US or selected US substantive law.
(43) He suggests that it is meant specifically for domestic litigation, although he

notes some contrary US authorities. His view is certainly to be preferred.

With the possible exception of the US, the prevailing view is, therefore, that
broad procedural powers encompass costs entitlements and that an enforcement
court should not hold in such circumstances that a tribunal has exceeded powers.
Nevertheless, because there are possible conflicting views and because party
autonomy is paramount, the most sensible solution is to be clear in the
arbitration clause or a submission agreement or through selection of arbitral
rules as to what the tribunal is empowered to do where the lex arbitri is silent.
(44)

15.2.8. What If there is a Difference in Terminology between the Lex Arbitri

and Rules Selected by the Parties?

As a general principle, lex arbitri provisions such as Article 19 of the Model Law
indicate that the tribunal is bound by a choice made by the parties. Thus, if they
page "1201" select a set of rules with differing costs provisions than the lex
arbitri itself, that should prevail unless provisions in the lex arbitri are
mandatory. An example of a rare mandatory provision noted above is section 60
of the English Arbitration Act which proscribes pre-dispute agreements to
impose costs on a party.

Where non-mandatory provisions are concerned, an important question is

whether detailed principles in the lex arbitri still apply where the parties have
then selected a set of arbitral rules that include a very broad discretion. An
example would be an arbitration with its seat in Japan where the lex arbitri has a
presumption that each party bears its own costs subject to an agreement to the
contrary. What if the parties selected the ICC Rules to apply which has a general
discretion as to costs but no presumption either way? Have they reversed the
presumption in the lex arbitri, have they left it entirely to the discretion of the
arbitrator or have they failed to do enough to displace the statutory presumption
that each bears their own costs? While the second alternative seems the proper
one absent evidence to the contrary, questions of this nature are simply
questions as to determination of the parties' true intent in the absence of
unambiguous guidance. If parties know what they mean they can say so
expressly. At the very least, counsel and arbitrators need to be aware when
uncertainty may be caused simply through differences between the lex arbitri
and the rules. If a tribunal raises this issue with the parties at an early stage, they
can ideally come to some agreement. If they are unable to do so, they can still be
asked to make submissions about the tribunal's proper powers in the face of
their differences in view. Absent further guidance it is still the case that a broad
discretion in selected rules should not be read down by non-mandatory
provisions in the lex arbitri.

Absent any such clarification, where a very prescriptive code such as that
contained in the English Arbitration Act 1996 also sees selection of a very
general set of rules, this is not seen by some commentators as sufficient evidence
of opting out. (45) Mustill and Boyd also suggest that a contracting out of the entire
1996 Act cost provisions without any replacement may not be permitted, (46)
although that again should be seen as a matter of interpretation, as the parties
could surely expressly agree to not be bound by limits or presumptions in non-
mandatory aspects of any lex arbitri.

15.2.9. Court and Institutional Assistance in Cost Determinations

Some arbitral rules call for court involvement in cost assessment, at least on a
voluntary basis. Under the English Arbitration Act 1996 there is an entitlement
to apply to the court in relation to the fees and expenses of arbitrators and also
where page "1202" the tribunal does not determine recoverable costs. (47) The
Singapore International Arbitration Act provides for taxation of costs directed by
an arbitral award. Section 21 indicates that unless the award otherwise directs,
the costs are taxable by the Registrar of the Singapore International Arbitration
Centre. (48) Section 27 of the Australian International Arbitration Act provides
that the tribunal may tax or settle the costs. If the tribunal does not tax or settle
costs that are to be paid by an award, then a court having jurisdiction under
Article 34 of the Model Law may hear an application for setting aside the award.
The revised UNCITRAL Rules 2010 allow for review by the appointing authority
or in default by the Secretary-General of the PCA, although review is limited to
either analysing the consistency with prior notification of the conditions or
determining whether the fees are ‘manifestly excessive’. (49) The express review
mechanisms in institutional or UNCITRAL Rules should take precedence over
mechanisms in the lex arbitri.

Even if there is an option of referring the matter to a court or court Registrar, if

this is not mandatory this might not necessarily be preferred as the arbitrator
knows the case better than any such third person. This needs to be taken into
account in deciding whether to exercise such rights. (50) Institutions will also
typically control costs. For example, they will fix the arbitrators' fees and
expenses rather than the tribunal itself. (51) Institutional control over costs is
discussed further in section 15.16 below.

15.3. Costs where a Tribunal Denies Jurisdiction

It has been suggested above, that regardless of whether express or implied in the
lex arbitri, a tribunal has power to award costs. One distinct situation that
requires further analysis is where a tribunal denies jurisdiction and the
respondent seeks costs. While the doctrine of competence-competence is readily
accepted in virtually all lex arbitri, allowing a tribunal to determine that it has no
arbitral jurisdiction, more problematic is whether it can award costs concurrent
with making such a negative determination. There are competing conceptual
views: those in favour argue that competence-competence extends to all aspects
of an adjudicatory determination, including any relevant collateral order as to
costs. The contrary view is that the jurisdiction to award costs only flows from a
valid arbitration agreement which is not present when jurisdiction is denied. The
view that an incidence of the competence-competence power is page "1203" the
power to make a costs award in favour of a prevailing party is to be preferred. (52)
A number of cases have seen arbitrators consider that they have inherent power
to award costs in a successful challenge to jurisdiction. (53) Bühler makes the
sensible observation that if the respondent had not taken part in the proceedings
to argue against jurisdiction, the claimant would have had to pay the fees in any
event. (54) Born also observes that the conceptual problem should not arise if
there is a valid agreement and the tribunal is merely determining that the
particular claim is outside of its ambit, although that could depend upon one's
view as to the nature of jurisdiction. (55) Schwarz and Konrad also support the
view taken by Born and note acceptance in German supervisory courts. (56) The
power is now expressly granted in section 609(2) of the Austrian Code of Civil
Procedure, although the wording is limited to situations where there is no
arbitration agreement and not other circumstances where jurisdiction may be
lacking. (57)

An express power to award such costs contained in the lex arbitri, selected rules
(58) or the parties' express agreement would certainly support such an award

although one could even take the extreme view that rules only apply where there
is a valid agreement to arbitrate. (59) One possible approach is to obtain the
parties' express written consent to the assessment of costs in such
circumstances. A respondent would probably not wish to refuse such consent as
it might make the challenge look weaker. This could also be dealt with through
terms of reference where applicable. (60)

15.4. Agreement of the Parties as to Costs

15.4.1. Choosing Arbitration

The first and most important agreement of the parties impacting upon costs is
the very choice of arbitration over other dispute settlement processes. Questions
of page "1204" costs may be an important element in deciding whether to engage
in arbitration as opposed to litigation. For example, an American party used to
each party bearing their own costs in domestic litigation, has to consider the
extra risks and potential rewards in an arbitration where the rules provide that
costs are likely to follow the event.

As to the likely total costs of arbitration relative to litigation, a number of factors

ought to be considered by parties and their legal advisers both in terms of choice
and in terms of mechanisms of control where arbitration is selected. Some costs
are inevitable simply because of the international nature of the adjudicatory
process and would apply equally to litigation. An example is where there is a
need for translation of documents and interpreters. Some costs are peculiar to
arbitration or are higher in that arena. The first of them is the systemic need to
pay for the services of the tribunal, pay for a hearing room and even the
inevitable travel and accommodation expenses in international matters,
generally higher than in international litigation if a neutral seat is selected. There
may also be costs of an arbitral institution. Filing fees with arbitral institutions
would often be higher than court costs although each may be subsidised by state
tax revenue. A second reason why costs may escalate in arbitration is that
concerns for due process and cross-cultural sensitivity have led to international
arbitration adopting key features of both civil and common law traditions,
arguably allowing for undue repetition and overly expensive processes. For
example, in an admittedly extreme scenario, where parties are allowed to have
two rounds of pre-hearing written submissions and a post-hearing final written
submission stage, with full examination and cross-examination of witnesses, this
is far more elaborate than most national adjudicatory models.

However, there are a number of factors that may lead to lower costs in
arbitration than litigation. One is the lack of an appeal mechanism in arbitration,
although annulment and enforcement challenges may lead to similar costs. A
second aspect is that some jurisdictions allow very expensive procedural
mechanisms in litigation, such as discovery of documents as of right, while
international arbitration may adopt a more constrained approach. There are also
some other domestic litigation processes that are not replicated in arbitration,
hence reducing the costs of the latter. Examples would be presenting or opposing
motions for summary judgment, witness deposition in the US and empanelling
juries where these are involved in commercial matters. International litigation
may also involve other arguments such as forum non conveniens where more
than one court purports to have jurisdiction. Arbitrators are also less able to
compel evidence from third parties, which is not only an evidentiary problem but
also an uncertain costs factor depending on whether courts can assist.

There is little in the way of empirical evidence to identify the likely relative costs
in the context of these conflicting analytical considerations. One reason is that it
is difficult to compare litigation and arbitration costs without excluding
variations in the cases themselves. Commercial arbitration and investment
arbitration will often also involve numerous complex matters that do not
typically arise in domestic litigation. Comparative costs would also need to take
into account the page "1205" time value of money and the time taken to render
and enforce a final judgment or award. The overriding point is not to suggest a
systemic advantage of one form over the other but, rather, help identify issues
that may impact upon choice or design and which can be addressed by the
parties as early as possible.

15.4.2. Party Control over Arbitral Costs

As noted above, with the exception of mandatory rules, party autonomy is

paramount on all procedural matters, including cost control and cost
determinations. While parties often do not make specific agreements as to costs
in drafting their arbitration clauses, they should at least give consideration to
doing so. Considering costs at the outset could be part of a general approach to
conducting arbitration in the most efficient manner. The parties themselves can
consider such matters as selecting one arbitrator rather than three and selecting
expedited procedures or at least setting benchmarks as to when these should
apply. Early consideration of costs issues may also impact upon selection of an
arbitrator if a particular costs approach is preferred. An agreed costs protocol
would also act as a submission to its terms.

15.4.3. The Relationship between Party Autonomy, Lex Arbitri and Arbitral
Rules

It must always be remembered that party agreement can either add to or take
away tribunal entitlements as to the award of costs. (61) Generally speaking,
where the parties have come to an agreement on an issue of costs, this will be
binding on the tribunal unless the agreement offends against a mandatory law of
procedure. (62) An example of a mandatory provision, as noted above, is the
English Arbitration Act 1996 which renders inoperative, cost agreements that try
and shift the costs to one party prior to the dispute evolving. An agreement after
tribunal constitution that undermines a prior agreement with the tribunal would
be problematic as there needs to be consensus to vary such an agreement. Even
if there was no agreement, page "1206" problems may also arise where a
subsequent agreement of the parties as to costs undermines the legitimate
expectations of the tribunal as to the return on its effort, although it would need
an extreme set of facts for this to be a barrier where the expectation was not an
agreed term.

In other cases, a lex arbitri may simply proscribe something but not indicate
expressly whether it is mandatory and hence whether the parties can waive the
relevant provision. For example, the arbitration statute of Thailand implies that a
tribunal cannot award attorneys' fees. (63) What is the situation if the parties
expressly empower the tribunal to make such costs awards in an arbitration that
is subject to Thai arbitral law? A tribunal might need to form a view as to
whether the law is truly intended to be mandatory, namely, one which operates
contrary to the will of the parties. While a simple agreement on costs might not
at first sight appear to offend against mandatory procedural norms, governments
will tend to proscribe cost measures if they feel that these can interfere with
access to justice. For example, a cost agreement which tries to put all the costs on
one party becomes a disincentive to that party seeking redress before an
adjudicatory body. Requiring each party to bear its own costs might be seen by
some governments as a means to encourage costs to be kept low and also leave
costs out of settlement negotiations, hence aiding their successful outcome.

A party who has entered into a cost agreement and then seeks to argue that it
offends against mandatory laws might also invite consideration of whatever
rules of waiver or estoppel might be applicable in the instant arbitration,
although some would say it is impossible to waive a mandatory law entitlement.
A tribunal might also disregard the parties' agreement where it violates some
transnational principle of public policy or was otherwise significantly prejudicial
to one party, or where the express agreement can be seen to be contrary to the
parties' true intentions. (64) In the first instance, it is likely that an agreement that
offends transnational public policy also offends against mandatory procedural
due process norms in any event. In turn, this would limit the types of cases
where public policy could be legitimately and distinctly utilised to interfere with
costs agreements. For example, the better view is that the American rule that
each party is to bear their own costs is not a fundamental principle of public
policy per se and hence parties can agree to provide for costs orders in that
jurisdiction, (65) although issues of unconscionability or unequal bargaining
power might be raised in certain circumstances. Absent such factors, from a
consent perspective, the presence of presumptive costs rules in particular lex
arbitri might be the reason why a US practitioner would choose arbitration over
litigation or vice versa.

page "1207"

15.4.4. Drafting Cost Agreements

If a power to award costs and fees only emanates from the parties' agreement,
care needs to be taken with the drafting, as a supervisory court or tribunal that is
not used to such a power might take a restrictive approach to interpretation. (66)
Care is also needed as there are so many potential items to cover, that
ambiguous phrases might not be held to have covered each relevant matter. (67)
For example, if an agreement referred to ‘arbitral costs’, would this be
interpreted to cover legal fees as well as costs of the tribunal? Applicable rules
may also raise interpretative presumptions applicable to party agreements. For
example, section 52 of the English Arbitration Act 1996 indicates that, unless
otherwise agreed by the parties, any obligation under an agreement between the
parties as to how the costs of the arbitration are to be borne or under an award
allocating costs extends only to the costs as are recoverable.

Even if lex arbitri provisions are not mandatory, in some cases it is not clear
whether they are opt-in or opt-out provisions. That may impact upon the
evidentiary standards required to show contrary party intent. Absent mandatory
norms or public policy considerations, if there is any doubt as to the entitlement
of a tribunal to award costs under the lex arbitri, the parties should make clear in
any agreement that they are expressly allowing for or barring fee shifting. (68)

Some arbitration clauses give a claimant an option as to which seat will be used
to underpin the arbitration. Given that costs principles differ greatly, a claimant
armed with such a power is given an important tactical advantage to select a
costs model that would suit its particular interests, although there are other
important factors in selecting a seat.

Gotanda has suggested a model clause that might be inserted into the parties'
agreement to clarify issues as to the awarding of costs and fees. (69) His model
covers express authority, a presumption that the costs of the arbitration shall in
principle be borne by the unsuccessful party, but also provides that the tribunal
may apportion such costs, in whole or in part, where appropriate. His model also
requires that the costs of the arbitration shall be reasonable in amount. Finally,
he advocates a non-exhaustive list of items being:

(a) the fees and expenses of the arbitral tribunal;

(b) any fees and expenses of the appointing authority; page "1208"
(c) the costs of expert advice and of other assistance required by the arbitral
tribunal;
(d) the travel and other expenses of witnesses; and
(e) the legal or other costs of the parties.

Whatever the final preferred position, the parties need to consider the extent of
the authority, any presumption or other general principles determining
entitlement, the range of relevant factors, whether reasonable limits should be
imposed, and which types of costs should be covered. If such matters are not
included in the arbitration agreement they could be included in a costs protocol
at an early stage.

15.5. Arbitral Discretion

15.5.1. Is there a Duty to Decide on Costs?

Because an arbitrator must complete the mandate, an arbitrator with a power to

determine costs ought to do so unless the lex arbitri or rules entitle it to avoid a
determination. (70) In some cases, a failure to award costs may be a ground for
overturning an award. (71) However, the tribunal's discretion should only relate
to claims actually made. A tribunal might award costs sua sponte if there is
egregious behaviour by one of the parties and the new ICC Rules are certainly
broad enough to allow this. Arbitral rules may give directions on this issue. Some
may be slightly ambiguous in that regard. For example, Article 42.2 of the
UNCITRAL Rules 2010 indicates that the tribunal ‘shall… determine any amount
that a party may have to pay as to costs’. Previously, the UNCITRAL Rules had
indicated that the costs of legal representation needed to be claimed in the
arbitration. This stipulation is no longer referenced in the new Rules. Hence, it is
at least arguable that there is now a direct duty to consider costs in all cases,
although this was unlikely to be the intent. In a practical sense, there needs to be
some evidence in support of a positive costs order so it would be hard to imagine
a tribunal awarding costs without an application before it.

Fouchard et al. observe that even where the parties initially agree to advance the
costs, the tribunal is generally required to rule on the ultimate allocation. (72) That
is certainly so if the agreement is limited to the notion of an advance. The
situation is different if the parties have agreed on final allocation.

page "1209"

15.5.2. The Nature of the Discretion

While it is clear that in virtually all cases tribunals have a broad discretion
subject to party agreement and mandatory norms, there are differences in view
as to the way this discretion should be applied. In addition to the differences in
compensatory models, such as between loser-pays and the American-litigation
approach at the two extremes, and differences between indemnity and
reasonable limits approaches, some also question whether the legitimate
expectations of the parties should be considered. This has led some to suggest
that tribunals might consider the laws and practices in the seat or in the parties'
own jurisdictions. Looking to the parties' jurisdictions may be sensible if both
national systems have an identical costs rule. The more familiar the parties are
with a particular approach, the more likely it may fit within their reasonable
expectations, (73) although as noted above, even that may be an inappropriate
presumption if they have selected arbitration in part to avoid this feature of their
domestic litigation systems. Any choice of arbitration is a rejection of some
aspects of litigation systems.

While most legislation and rules provide a tribunal with a broad discretion, it is
not unlimited. Some supervisory courts would consider that the discretion must
be exercised judicially. (74) Limitations on the discretion can also be impacted
upon by directions in the lex arbitri. For example, under the English Arbitration
Act 1996, because of the presumption that costs follow the event, a costs award
could be annulled where it wrongly applied that principle. (75)

15.5.3. Tribunal Cost Control

To the extent that they are permitted to do so, arbitrators may seek to control
costs directly, through limiting costs awards or indirectly, through their
procedural control. For example, as to the latter, limits on the length of hearings,
the number of witnesses and total pages in written submissions must indirectly
impact upon the costs incurred by the individual parties. The issue of cost
control must be considered in the context of party autonomy and the parties'
own entitlement to frame their arbitration to achieve desired aims, including in
this context, cost efficiency. The duty of efficiency as to cost is enshrined in Rule
7 of the IBA Ethics Rules indicating that arbitrators ‘shall do their best to conduct
the arbitration in such a manner that costs do not rise to an unreasonable
proportion of the interests at page "1210" stake’. A number of arbitral statutes
and rules also direct tribunals to take steps to promote efficiency.

Chapter 2 considered the duty of efficiency more generally. Chapter 6 considered

a number of actions on behalf of the parties or tribunal that might promote
efficiency. It was therein suggested that the proper approach is to manage
proceedings to promote efficiency but without prejudicing fairness, including
avoiding disincentives to the parties' legitimate strategies for winning the case
which can arise through certain cost allocation presumptions. (76)

Efficiency options by the tribunal might include:

1) preliminary rulings that will limit the ambit of research and submissions, e.g.
ruling on applicable law at an early stage;
2) constraints on time and length such as chess-clock arbitration;
3) being open with the parties as to the key concerns of the tribunal so that
lengthy submissions are not made on irrelevant matters or on matters that
are not seen as contentious;
4) invitations to the parties to agree on facts;
5) document control strategies such as use of information technology systems;
and
6) actual determinations on costs or warnings at the outset that procedural
behaviour will impact upon final cost determinations.

The last element in the above list deals with the use of costs awards by a tribunal
as an efficiency promoting mechanism. Redfern and Hunter suggest that ‘a costs
order is one of the few means at a tribunal's disposal to discourage, and in
appropriate circumstances to punish, a party's wasteful procedural tactics
during an arbitration’. (77) The ICC Report on Techniques for Controlling Time and
Costs in Arbitration similarly suggests that ‘(t)he allocation of costs can provide a
useful tool to encourage efficient behaviour and discourage unreasonable
behaviour.’ (78) It also makes the important point that for this to work
successfully as a deterrent, a tribunal should specify at the outset that it will take
unreasonable behaviour into account in exercising its costs discretion. (79) ICC
Rules 2012 Article 37(5) also expressly deals with such allocations.

Nevertheless, there are a number of difficulties in inviting tribunals to directly

control efficiency through costs awards. Because of concerns of due process,
arbitrators may feel less able than domestic judges to strictly control party
behaviour and thus may find it harder to regulate inefficient practices. This is
particularly so when inefficiency is the result of a hybrid procedural model
adopting what each page "1211" party believes to be the most useful processes
from its legal culture. Parties who see a tactical advantage in a particular
procedural approach might also threaten challenge on due process grounds if the
procedural request is not granted. Key examples of the problems facing
arbitrators include parties who seek adjournments shortly before key
procedural deadlines, unnecessary repetition in written submissions, excessive
number of witnesses, over-broad document production applications, written
witness statements overlapping with time-consuming evidence in chief, and the
utility or otherwise of post-hearing briefs.

There can also be specific problems with some of the above-mentioned

procedural cost cutting steps. For example, one problem with tribunals directing
the parties as to the matters of greatest interest is that any premature indication
can be seen as prejudging what later evidence may indicate. The same may even
be so in relation to choice of law, for example, if a closest connection test is to be
applied in a complex factual scenario. A further problem arises in multi-person
tribunals. For the tribunal as a whole to indicate the areas of greatest concern
requires some deliberation at an early stage between tribunal members. (80) If a
tribunal exercises a controlling hand too early in the process, it may get it wrong.
If it leaves it too late, it may have little impact on the actual fees incurred.

A particular problem is that arbitrators do not have direct control over the major
source of costs, namely the fees of counsel. An arbitrator can indirectly control
such costs through procedural directions as to written submissions and length of
hearings but has no real ability to directly control the amount of work counsel
does in preparing the case. Indirectly dealing with this after the event through
costs awards is a poor second-best solution as it will be too late to ensure that
total costs are not an unreasonable proportion of the interests at stake. Where
total costs are unduly high, but the winning party is awarded a lesser percentage
of costs accordingly, both parties are likely to come away from the arbitration
being decidedly unsatisfied. While this is a problem, the potential for over-
servicing by counsel is also a problem in domestic litigation systems, again with
little control by the courts save for a power to report counsel to local Bar
Associations in egregious circumstances. Arbitrators even have limitations in
that regard, given their general duty of confidentiality and the lack of an express
power over counsel. Such a power is considered further in section 9.7. Costs
discretions are discussed further in section 15.9. A further problem is that while
all good arbitrators will follow ethical principles in individual cases, they would
have to realise that if they develop the reputation of severely reducing counsels'
fee entitlements on a regular basis, such counsel may not recommend them for
arbitrations in the future. Furthermore, costs awards in arbitration would never
entitle punitive costs absent party agreement. At most the tribunal can seek to
indemnify an innocent party.

A more radical approach in order to contain costs in an appropriate case, would

be for the arbitrator in conjunction with the parties at an early stage to fix the
total sum of money it believes should reasonably be spent on the conduct page
"1212" of the arbitration. The parties would be at liberty to spend more if they
wish, but they would be on notice that recoverable costs are likely to be limited
to the estimates made. It is certainly desirable that the parties have a clear
understanding that costs can too easily escalate to an unacceptable percentage of
the amount in dispute. It is also desirable that they have the opportunity to cap
the costs by agreement. There are tenable arguments either way as to whether
the tribunal should ultimately do so unilaterally. A tribunal that establishes very
early and rigid cost caps can have a very positive impact but might also
encourage guerrilla behaviour by a party who knows it will lose ultimately, in the
hope that it can make the process so costly that a claimant may be discouraged
from proceeding. Similarly, if a tribunal is considering applications from the
parties as to how to cap costs at an early stage, there is an incentive for the party
with a weaker case to advocate strict self-serving caps in the guise of efficiency
norms. (81) There is a further problem in that a tribunal may not be in a position
at an early stage to make an accurate assessment as to a total costs figure.

An alternative approach is to utilise a percentage cap that guarantees that

transaction costs are not too large when compared to the amount in issue.
Bernstein argues in favour of a rule of thumb to cap at 20% if there is no reason
for a higher amount, with 10% preferable. (82) Suggestions of this nature can be
motivated by the fact that parties often do not understand that legal preparation
can expand indefinitely if winning is the only aspiration. If full cost recovery is
not applicable, this will often lead to all parties being unhappy with the outcome
when all bear an excessive level of net losses from the process. There are
problems with an a priori figure, however. Would the 20% be of the claim or the
amount awarded? If the former, claims will be inflated; if the latter, it will be less
effective as the result is not known when the expenditure is incurred, although it
may still be justified as a reasonable limit from the loser's perspective. The need
to consider factors such as complexity, and procedural behaviour would also
mean that the presumption would commonly be rebutted. Nevertheless, the wish
to warn parties that costs can easily match the disputed amount is a laudatory
one.

15.6. Presumptions in Costs Awards

15.6.1. Current Practice

Redfern and Hunter suggest that ‘(i)n general… the procedure in international
commercial arbitration is for the arbitral tribunal to have the power to require
page "1213" the losing party to pay or contribute towards the legal costs of the
winning party.’ (83) Other scholarly opinions also support the view that the loser-
pays principle is becoming the norm in international arbitration, (84) although the
suggestion that it should be seen as best practice has been questioned by Richard
Kreindler. (85) There are also many instances where the power exists, but the
discretion is not exercised in this way. (86)

Some survey research and harmonisation work support this emerging principle.
For example principle 25.1 of the Draft Principles of Transnational Civil
Procedure of the Joint Working Group of the International Institute for the
Unification of Private Law (UNIDROIT) and the American Law Institute suggest
that ‘the winning party ordinarily should be awarded all or a substantial portion
of its reasonable costs’, although this exercise was not dealing with arbitration.
(87) Note was previously taken of Born's suggestion that the sui generis standards

of institutional rules and arbitral practices typically ‘provide that (a) the
prevailing party is presumptively entitled to a costs award; (b) only reasonable
costs will be reimbursed; and (c) expenses that were inefficient or unnecessary
will not be reimbursed, while costs resulting from the need to respond to
unreasonable or uncooperative actions will be recoverable.’ (88)

Conversely, Gotanda suggests that while the conventional wisdom in

international arbitration is that tribunals adopt the costs follow the event
approach, the few studies that have been made and analysis of recent cases
suggest that this may not be so and ‘awards of costs and fees are often arbitrary
and unpredictable.’ (89) In Waguih Elie, George Siag and Clorinda Vecchi v. The
Arab Republic of page "1214"Egypt, an ICSID tribunal stated that it ‘has also taken
due note of the decisions made by previous ICSID tribunals, in light of which it
appears that the practice of such tribunals has not been uniform and that the
present tribunal therefore, has a broad discretion to apportion costs.’ (90) Care
should be, taken in comparing comments in commercial arbitration cases with
those in investment arbitration cases, the latter having less of a tradition in costs
following the event, with some arguing for a contrary position from a policy
perspective. Investment arbitration is also an area where there are more likely to
be frivolous claims that might challenge legitimate sovereign rights or try and
blame the host State for an investment failure truly caused by other factors.
Nevertheless, some more recent investment arbitration cases have argued in
favour of a loser-pays norm. (91) Susan Franck has undertaken an empirical
analysis of the data. (92) A more recent study was undertaken by David Smith. (93)
Susan Franck's earlier studies had seen 32.6% involving cost or fee shifts
between 1990 and 2006, with 12% shift in arbitral costs but not legal fees, 2%
shift in legal fees but not arbitral costs, and 18.4% shift in some of each. Smith's
analysis of cases in 2008-2009 saw the figure of some costs shifting increasing
from 32.6% to 41.9%. He observed that the bulk of the increase was in awards
that shifted legal fees but not arbitral expenses. (94)

15.6.2. Policy Considerations Underlying Cost Discretions

Because tribunals generally have a broad discretion as to the awarding of costs

and the models and methods to use, it is appropriate to consider the alternatives
and their relative merits as a guide to arbitral practice and as an aid to party
advocacy on costs applications. This section looks at broad policy factors as to
competing costs principles. The following sections look at how each principle
may be applied in practice. Subsequent sections look at individual cost elements.

While there is no consensus worldwide as to the optimal costs model in domestic

litigation, as noted in the previous section the most common approach is said to
be ‘loser pays’, otherwise described as ‘costs follow the event’. This is the basic
approach in common law countries other than the US. At the other extreme is
page "1215" the American approach where each party bears its own legal costs
no matter who wins, although this is not a uniform approach in all American
litigation and may be affected by party agreement to the contrary. (95) Civilian
jurisdictions differ primarily as to their treatment of legal costs.

The loser-pays principle is sought to be justified from the perspective of the

winner based on restitution principles. But for the losing party's improper
behaviour, the winner would have been better off financially. On this approach,
the role of the adjudicator is to return the winner to that position through an
amalgam of damages, interest and cost awards. In the extreme, that perspective
would only concern itself with the bona fides of the winner's costs and not
whether the level was reasonable or not. If the winner had to pay, then the
winner should be reimbursed. One argument against this is to suggest that it is
not fully analogous with restitution and damages caused by breach, as in any
damages assessment there is also a need to consider new intervening events that
should not be attributed to the losing party and also consider mitigation
arguments. Where counsel and other arbitration costs are concerned, the winner
has made a decision as to how much to spend. Where a decision is made to
engage in more costs than seems reasonably necessary, not all of the expense can
be said to have been caused by the losing party's original breach.

Another possible reason to apply the loser-pays principle is to punish the loser
for its behaviour. (96) Imposing costs on a losing party may also be an added
disincentive to parties maintaining a losing position in an arbitration or
breaching their agreements in the first place. This should not be dependent on
whether it is the claimant or respondent which wins. Awarding costs in favour of
a successful respondent not only indemnifies it for its losses but also provides a
disincentive to inappropriate claims being brought. One suggestion is that this
may be less relevant in arbitration where the respondent has willingly signed an
arbitration agreement, (97) but there is a difference between agreeing to arbitrate
and agreeing to spurious claims. Furthermore, some respondents may not have
signed any agreement, for example where there is an erroneous claim of agency,
alter ego or group of companies involved. As a general rule it would be
undesirable if there was a systemic preference in terms of costs awarded to a
winning claimant as opposed to a winning respondent. (98)

However, slavishly following a costs follow the event approach limits the
tribunal's ability to discourage inappropriate procedural steps such as excessive
page "1216" discovery, requests for adjournment and late provision of
submissions. At the other extreme, fairness arguments in favour of the American
litigation model might include the suggestion that even losing parties can have
justifiable reasons to present claims that are ultimately unsuccessful. An example
in international arbitration would be submissions as to applicable law when a
tribunal has a broad discretion. Another example might be a jurisdictional claim
against a non-signatory under a group of companies theory. Allied to that is the
fact that many complex legal and factual arguments may have costs grossly
disproportionate to the amount of money in dispute.

The American litigation perspective is also concerned more with systemic

efficiency and promoting access to justice. Access is enhanced through reducing
transaction costs. This costs rule goes hand in hand with the contingency fee
system that has arisen commercially as a result. At the outset, neither party is
committed to paying legal fees, although each knows if they win, a contingency is
to be paid. A priori each party knows that if they win they will not be fully
compensated as to costs but they are likely to be happy with a reduced amount
where they win in return for not having to pay either parties' costs if and when
they lose. While the upside and downside potentials are arithmetically identical,
most people are risk averse, so the American system is seen by many as
commercially desirable. (99) When the US Supreme Court articulated the rule, it
justified it because of the uncertainty of litigation; access to justice and the added
costs of assessing costs claims themselves. (100) The access to justice argument is
less relevant to arbitration where the arbitration itself has to have been consent
based. (101)

Because both loser-pays and the American litigation rule appeal to conflicting
economic efficiency arguments, it is understandable that the true situation may
be more complex. For example, it is arguable that where the parties know that
the winner will be awarded costs, a higher amount is, therefore, in dispute,
which in turn might encourage higher expenditure on legal preparation. (102) The
more a party thinks they will succeed, the more they will discount the marginal
costs of such legal services, thus providing a further page "1217" incentive
towards their expenditure. (103) Costs following the event may also discourage
settlement, as anything which increases the disparity between the parties'
assessment of the outcome will make it harder for them to find a mutually
agreeable figure, although that is a matter of some debate among law and
economics scholars. (104) While it is suggested that loser pays should encourage
litigants to be more circumspect in deciding what cases to bring or defend, the
risk of adverse costs awards may also be a disincentive to the bringing of
meritorious claims, particularly by those with limited resources. Conversely,
broader economic efficiency considerations would suggest that in most cases
where a decision is made on the merits, the loser-pays principle promotes the
proper performance of contractual obligations. If a party can benefit financially
from breach of contract, it is more inclined to wish to do so. Increasing the costs
of such breach will act as a disincentive. There are obviously conflicting tenable
hypotheses. As with most aspects of comparative legal analysis, there are thus
understandable reasons why different national systems adopted quite different
solutions to common problems. The problem in all cases is to find the happy
medium. The following sections look at the way the different approaches should
apply in practice.

15.7. Costs Following the Event

15.7.1. The Nature of the Loser-Pays Principle

The notion of costs following the event traditionally means that if a party wins as
to part, it should obtain its entire costs unless the amount awarded was de
minimis, trivial or nominal, (105) although that is not a uniformly held view. Most
domestic jurisdictions follow the loser-pays principle with the clear exception
being the US. (106) To some, the notion that loser pays or costs follow the event is
sufficiently broad to allow a range of variations, including a consideration of
relative success, discussed in the following section under issue-by-issue costs
awards. Furthermore, even in systems that adopt the costs follow the event
principle, they will invariably allow judges to vary costs orders, for example
where the winning party raised spurious claims. In such circumstances they
might deny the winner part of their page "1218" costs and even order them to
compensate the loser in relation to defending the inappropriate claims. (107)
Some would justify a winner-take-all approach on the basis that the loser's
reluctance to pay forced the arbitration and hence forced the arbitration costs.
That would be so for the costs of the arbitration itself but does not deal with the
proportionality of the legal fees. (108) Some, therefore, distinguish between
procedural and legal costs. Where the claimant succeeds as to part, this still
might justify the full costs of the procedure as these needed to be incurred to
obtain any relief. The same logic would not support compensation for legal fees
in respect of the unsuccessful claims. Another argument in favour of the
distinction is that because the parties have agreed on the tribunal and the
institution if any, they have agreed to the costs that arise as a result and it more
accords with legitimate expectations that the loser should fully compensate the
winner as to these. Where legal costs are concerned, each makes a separate
contract with their own lawyers. Because there is often such disparity between
the financial status of parties and charge out rates of law firms, similar
presumptions might not readily apply where each party has no involvement in
their opponent's selection of counsel. (109)

Whatever the merits of the arguments, the differing treatment of procedural and
legal costs by tribunals is borne out by an earlier ICC study which showed that
successful claimants obtained part or all of procedural fees in 81 % of cases but
were only awarded legal costs in 50% of cases. (110) A study of CIETAC tribunal
decisions concluded that arbitration costs were awarded in approximately 90%
of cases, with attorneys' fees only awarded in 28% of such cases. (111)

15.7.2. Determining Who Won under Loser-Pays Cost Criteria

Even if a tribunal resolves to apply the loser-pays principle, in some cases it may
not be clear as to who the winner is and to what extent. As noted, there will be a
page "1219" different outcome if one took the view that the overall winner
deserves full compensation, or instead, whether an allocation should be made on
an issue-by-issue basis and even then subject to possible reduction if the amount
awarded is significantly less than the amount claimed. An obvious example is a
claimant that succeeds in showing liability but only obtains nominal damages.
Another complication is if there are claims, counterclaims and set-off
applications. Is each individual cause assessed as to costs individually or should
costs be considered in relation to the net winner overall? This is discussed
further in the following section dealing with issue-by-issue cost awards.

15.8. Issue-by-Issue Cost Awards

Some advocate a costs approach which does not simply seek to identify the
overall winner, but attempts instead to apportion costs in proportion to the
degree of success. Schwarz and Konrad recommend allocation of costs based on
relative success unless special circumstances warrant an exception or the parties
otherwise agree. (112) Even here there are still differing approaches that could be
considered. One approach would be to look at the ultimate outcome and the
degree to which the successful party achieved the remedies that it sought. On
this basis a party who claimed USD 1,000,000 via a range of heads of claim and
succeeded as to USD 900,000 on one claim only, was nevertheless 90%
successful. An alternative approach is to look at how many claims were raised,
which were successfully pursued and which were not. Under the above example,
if the claimant succeeded on one argument but failed on five, an issue-by-issue
approach could suggest that they need to compensate the ultimate losing party
who won on five out of six legal arguments and itself only be compensated as to
the argument which won. An issue-by-issue approach or a proportion of success
approach might thus concentrate on the procedural costs and their
reasonableness or the ultimate level of success with very different outcomes. (113)

Where tribunals look at relative success, a number of differing computational

options have also been adopted. Some simply try and assess the percentage page
"1220" of success and provide the winning party with that proportion of the total
costs and fees claimed. (114) Another approach is to net one amount off against the
other. For example, if a claimant succeeded as to 80%, it failed as to 20%, the two
figures netting out at 60% which might be a basis for a partial award of costs and
fees. (115) A quite different result applies if the tribunal considers both success
and failure and makes an allocation in relation to each. Biihler uses an example of
a claimant only succeeding as to 50% of the claim. In this circumstance, it is
arguable that each party was equally successful, claimant in proving half of its
claim and respondent in defending the other half. (116) In Final Award No. 7047, a
claimant was held to have succeeded on 50% of the claims. The tribunal awarded
it 50% of its costs and fees but also obliged it to compensate respondents for
50% of their costs and fees on the basis that each party had the same relative
level of success. (117)

There are no unassailable arguments that could be made for or against any of the
above options. Regardless of the computational method adopted, from a policy
perspective, any form of issue-by-issue approach alerts parties to the cost
downside of unsuccessful claims and defences. As such, a possible efficiency
advantage of an issue-by-issue approach is that it might deter respondents from
taking an undue number of procedural and substantive defences and provides an
equal disincentive against claimants adding a multiplicity of alternative and
perhaps speculative claims to their central issues of concern. (118) That would
presume that the parties know in advance that this is the costing model that will
be utilised.

Nevertheless, there are a range of problems. The first is selecting the particular
computational approach out of the range outlined above, given that they lead to
vastly differing outcomes. Another problem with an issue-by-issue approach is
that it is difficult in many cases to apportion the costs between various claims.
For example, interviewing witnesses will typically cover a range of potential
claims. The same would be so for general research. It would also be
inappropriate to presume that the time taken on an issue is proportional to the
words devoted to it page "1221" in written submissions. Another problem is
whether an issue-by-issue approach would concentrate on success alone or also
on the complexity and reasonableness of the money spent. There may be no
correlation between legal effort and the relative success of particular heads of
claims. It would also be hard to substantiate any time-based break-up of global
figures.

There is also a distinction between success or otherwise as to the basic claim and
the degree of success as to the measure of damages claimed. That in turn can be
impacted upon by the amount of time and effort the parties devoted to arguing
liability and quantum. In addition, a party might only have failed to succeed on
some alternative claims that were argued superficially and which took little time
and effort on the part of the other party and the tribunal to respond to. There
might then be little cause to discount entitlements from an efficiency or
indemnity perspective. (119) Considering both complexity and percentage of
success is also complicated if consideration is given to whether a party ought
reasonably to have taken issue with certain matters or whether it responded to
an obviously weak claim with excessive zeal.

A further problem with a rigid issue-by-issue approach is that it might

discourage counsel from arguing in the alternative where it would otherwise
seem reasonable to do so. For example, in some contract cases it is difficult to
determine whether an adjudicator would be more inclined to accept an
argument based on a favourable interpretation of ambiguous express contract
terms, or on implied contract terms including good faith, or on tortious issues of
misrepresentation or on possible ancillary statutory provisions dealing with
good faith and/or misleading and deceptive conduct. If the claimant is ultimately
successful on only one ground, it is difficult to assert that a priori its alternative
claims were inappropriate and should not have been pursued. The better view is,
therefore, to not simply look at the outcome but look at what could be said to be
reasonable behaviour at the time of bringing the case. Another problem with an
issue-by-issue approach is that it itself can add to the costs of the arbitration
through parties then arguing about costs in great detail, although the tribunal
can impose reasonable limits proportional to the amount in dispute at this stage
as well.

15.9. Other Reasons to Award or not Award Costs

The following sub-sections look at various reasons why tribunals might depart
from a loser-pays, issue-by-issue or relative-success presumption. The following
situations identify a range of factors that could be considered on a case-by-case
basis. This can also be impacted upon by applicable laws and rules.

page "1222"

15.9.1. Problems of Proof

In all cases the party seeking costs must provide adequate proof of the
entitlement. Given the overriding discretion as to the reasonableness of costs
and their relationship to the arbitration, a tribunal may require proof as to the
nature and justification for each type of expenditure. At the extreme, a tribunal
might reject a claim simply through lack of evidence. (120) Some contentious
evidentiary issues may also arise. For example, attorney fee details should not be
subject to privilege claims.

15.9.2. The Difficulty of the Case

Some tribunals have taken the view that where cases are finely balanced,
notwithstanding that one party must prevail, a full costs award would not be
appropriate. The same result might be possible where the tribunal feels that the
unsuccessful party simply failed in terms of sufficiency of evidence in a close
contest and not because it was unlikely to have had a meritorious claim. (121) The
parties' differences might also have arisen from good faith tenable differences in
view. An example might be a BIT dispute where the tribunal believes that the
treaty negotiators were responsible for ambiguous drafting that caused a dispute
adversely affecting investors.

15.9.3. Party Responsibility for the Dispute

Another area where tribunals might choose to apportion or reduce costs is

where each is responsible for the underlying dispute, typically because of
inadequate drafting in their commercial agreement. For example, if parties enter
into a joint-venture agreement with a clear indication of the inter-party fees to
be paid in the first year but with no designation thereafter, resort to arbitration
is effectively resort to an expert ruling on a matter that they should have dealt
with themselves. A tribunal might thus consider that they should each bear some
responsibility for giving rise to the uncertainty. (122)

page "1223"

15.9.4. Considerations of Process

Tribunals might vary success-based cost presumptions to deal with undesirable

procedural tactics that have unreasonably added to the overall transaction costs
of the arbitration. Examples of conduct by the parties that might lead to adverse
costs awards include failure to comply with deadlines requiring follow-up
communications and amendments to timetables, late delivery of materials
requiring postponement of the hearing, failure to produce documents when
required, non-appearance without notice, failure to agree or sign reasonable
terms of reference, and waste-fully revisiting matters already decided by the
tribunal. (123) A tribunal might also order each party to bear its own costs where it
feels that both have created a process where the legal costs far outweigh the
amounts in dispute. In the extreme, a tribunal might even say that a successful
party has presented its claims in such an obtuse and disorganised way, that the
opponent should be given some compensation for excessive costs thereby
occasioned. (124) A tribunal might also refuse certain costs of evidence where it
believes that evidence was irrelevant or otherwise unnecessary. (125)

Where problematic tactics are concerned there is a need to consider whether

costs are to be used merely as an indemnity for losses that result, or as a
sanction. (126) Such a sanction can hopefully operate as a deterrent, although
again only if the parties are aware at the outset that this approach might be
taken. Where indemnity is concerned, there is a need to consider the impact of
the behaviour on costs and whether the behaviour has actually led to wasted
costs as a result. If so, it is less contentious to make an adverse costs order.
However, if the behaviour is unreasonable or disrespectful of the tribunal's
orders but does not occasion significant extra cost, a tribunal might be less
inclined to make an adverse costs award that simply looks as if it is being
punitive, although under most discretions it would still have the power to do so.
An example would be a responding party who refuses to attend the scheduled
hearing. While it might waste some time at the outset while the tribunal and
claimant tries to identify the respondent's whereabouts, ultimately it may lower
the cost of the process.

page "1224"

15.9.5. Inflated Claims

A tribunal might vary a costs presumption as a response to inflated claims that

are otherwise successful as to part. There are two aspects to the notion of
inflated claims. The first is where the claims are inflated from the outset. The
second is where the losing party makes a settlement offer part way through the
proceedings where this is rejected, but the claimant does not succeed in being
awarded a higher amount. This section deals only with the first aspect. The
relevance of settlement negotiations is considered in section 15.9.7 below,
although it can be noted at this stage that from a policy perspective, there may be
inconsistency if a tribunal was willing to reduce entitlements for excessive
claims from the outset but not take into account settlement offers during the
course of proceedings.

Tribunals will at times reduce the amount of costs awarded or even ask a
claimant to bear its own costs where the amount awarded is substantially less
than the amount claimed. (127) This may flow naturally under a number of rules
which impose reasonableness limits, at least subject to contrary agreement of
the parties. (128) Gotanda points out the difficulty in using this approach in
complex cases with a multiplicity of claims and counterclaims and where there
are also claims for non-monetary relief. (129) This situation is not to be confused
with cases where a claimant is wholly unsuccessful. In such a case respondent
can be compensated under the costs follow the event principle, although if the
claimant loses on jurisdictional grounds there may be arguments as to the
tribunal's powers in that regard. This was discussed in section 8.1.

Once again there is a need to consider whether the principle is based on

deterrence, indemnity or punitive considerations. For example, it will not always
be the case that inflated claims will increase the costs of the process, suggesting
that in such cases, full indemnity might still be appropriate if that is the guiding
criterion. Where an institution is involved and claimant's exaggerated claim has
caused an unnecessarily high advance to be paid by respondent, this might be
held against it in the award of costs. (130) One reason why a tribunal might wish to
reduce costs awards when there were inflated claims even absent marginal costs
in the proceedings is because that might be the sole reason why the losing party
chose to page "1225" defend. If the loser denies reasonable claims, there is much
to be said in favour of a full indemnity for a winning party which was forced to an
adjudicatory process. Conversely, where claims are inflated, the loser had no
alternative but to defend and in fact won in relation to all inflated components.
From an indemnity perspective, the unreasonable claim caused the partially
successful defensive action.

15.9.6. Failure to Admit and Needlessly Putting the Winner to Its Proof

Another possible factor of relevance is whether a party has unnecessarily forced

issues of proof which ought to have been admitted. (131) Thus, if the respondent
successfully defends a claim but took issue with all aspects, some of which
should have been admitted, costs might still be awarded for wasting everyone's
time and expense. (132)

The converse situation is where a successful party only provides its opponent
with adequate proof of its entitlement some way through the proceedings. A
tribunal might take the view that a more carefully articulated and proven
demand at an earlier stage might have led to a settlement or at least led to
significant admitted facts. Similar issues arise if a claim is insufficiently clear
when first made and is only properly articulated after directions by the tribunal.
(133)

15.9.7. The Relevance of Settlement Offers

A more contentious question is whether a tribunal should take account of any

settlement offers. Some domestic litigation systems indicate that such offers can
be produced to the court after judgment on an application for costs. If they show
that a successful party who rejected such an offer did not obtain a significantly
higher amount, the loser may not have to pay all of the costs of the litigation
subsequent to the time of the offer. The logic is that the overall winner caused
those wasted costs by not accepting the reasonable offer. In common law
jurisdictions these are described as ‘Calderbank letters’ or ‘sealed offers’. (134)
Arbitral rules page "1226" or party agreement might also expressly cover this.
For example section 6(2)(a) of the New Zealand Arbitration Act 1996 allows an
offer to be taken into account where ‘the award of the arbitral tribunal is no
more favourable to the other party than was the offer’. The Construction
Industry Model Arbitration Rules (CIMAR) rule 13.9 allows the arbitrator to have
regard to offers of settlement and invites the application of the general principle
that if a party recovers no more than the offer, then costs after that event should
be awarded to the offeror. (135)

Absent express guidance, the general question was raised in an LCIA award
involving an English claimant and a US respondent where a Calderbank offer was
made. As to the admissibility of settlement offers in relation to the award of
costs, the tribunal considered the principles as to general admissibility of
evidence found in section 34(f) of the English Arbitration Act 1996 and Article
22.1 of the LCIA Rules. The tribunal considered that it was not bound by strict
rules of evidence, that it could consider the substantive approach in the party's
jurisdiction but ultimately concluded that the settlement offer made by claimant,
which was expressed to be ‘without prejudice, save as to costs’, would not have
been sufficiently understood as to its implications by a US party not used to
Calderbank offers in domestic litigation. On this basis the tribunal did not
consider that the evidence should be admitted. That factual conclusion seems
sensible and certainly implies that parties should make clear the intended
implication of settlement offers. Having said that, however, it is suggested that
the tribunal should not have begun with an analysis of admissibility of evidence
rules, but rather, a policy consideration as to whether a settlement offer should
be relevant as to costs. Any adjudicatory body must first consider what
principles are relevant and only then can it consider what style of evidence may
be admissible in support of those principles. (136) There is also a problem in using
such an approach in arbitration where cost decisions are generally included in
the final award and not after separate representations as to costs. Bühler makes
the sensible suggestion that a tribunal intending to take such offers into
consideration should advise the parties at the outset. (137) Another approach is to
have a distinct award on costs after the tribunal is notified of the rejected
settlement offer.

If Calderbank principles are to apply, it should be noted that domestic courts do

not generally try and make net assessments setting off future costs against
amounts received in deciding whether offers should be accepted, but simply look
arithmetically at whether the party who rejected the settlement offer page
"1227" ultimately achieved more than a trivial extra amount. (138) That is also the
express approach in the New Zealand Act and CIMAR rules noted above. A
converse approach would look at whether the subsequent proceedings were
justified on a net cost/benefit basis. Even if they were still justified, some would
even argue that subsequent costs should be reduced if they were
disproportionate to the marginal recovery. (139)

15.9.8. Guerrilla Tactics and Other Examples of Bad Faith

Because arbitration is based on consent, which is rightly presumed to be in good

faith, it is easy to assert that a party is in breach of its obligations if it engages
unduly in delaying tactics (140) or tactics aiming to increase the costs of the
proceedings to embarrass the opponent or otherwise acts in bad faith. (141) For
example, several investment arbitration cases have awarded costs against
claimants that were held to have made fraudulent misrepresentations in relation
to the relevant investments. (142) Another example may be where a party
knowingly makes inconsistent allegations of fact in different fora. Tribunals
might also refrain from awarding costs to a successful defending party if it
succeeded on the basis of demonstrating that the entire agreement which it
voluntarily entered was corrupt or otherwise against public policy. (143) In some
cases a tribunal might even conclude that the winner has a clear entitlement
based on unfair contract terms and might not wish to compensate them as to
costs. While such propositions are easy to state, there can sometimes be a fine
line between what might be described as bad faith tactics and conversely, what
could be argued to be highly competent and strategic adversarial behaviour. An
example of a contentious scenario might be a respondent that demands that the
claimant better particularise its claim in the absence of which it would assert
that it is unable to properly prepare its defences. It is hard to determine at the
time whether that is merely a dilatory tactic or a proper means page "1228" to
use all available procedural mechanisms to best prepare the defence. A
legitimate aim may also be to identify weaknesses in the claimant's strategy, in
part to gain sympathy with the tribunal and/or promote a more favourable
settlement if that is being pursued.

In domestic litigation, some courts will even award costs against a legal
representative in their personal capacity where it is felt that their own behaviour
has been the cause of problems. The whole issue of control over counsel by an
arbitrator is a difficult one but it seems clear that there is no direct power to
award costs against counsel. If the behaviour was extreme enough, the tribunal
might seek to bar that counsel, although that is also contentious. The whole
question of control over counsel is discussed in section 9.7. A tribunal might
indirectly seek to impose costs on counsel by refraining from awarding
otherwise applicable costs to their client, stating the reason for refusal to do so,
being concerns as to the behaviour of counsel. That might indirectly lead to an
indemnity for the party but could not guarantee it.

15.9.9. Hardship and Other Factors

While variations to costs following the event are usually aimed at diminishing
compensation for inappropriate behaviour, other considerations may apply. A
contentious question is whether a tribunal should depart from a loser-pays
principle on the basis that this would cause undue hardship. (144) Such an
approach might have political overtones in investment arbitration where the
host State is the losing party. An ICSID tribunal resolved to only award tribunal
costs but not legal fees against the Republic of Zimbabwe because of economic
hardships it faced at the relevant time. (145) Hanotiau refers to an instance of an
unsuccessful claimant from a developing country that was not asked to bear all
of the costs of an unsuccessful claim brought in good faith in Paris. (146)

The special procedural regimes of an arbitration might also have relevance as to

questions of fairness. For example, Chamber 2 of the Iran-US Claims Tribunal
consistently refused to award costs, the key reason being that the security
account underpinning the Tribunal guaranteed prompt and full payment of
awards against page "1229" Iran, when if disputes had been proceeded with in
American courts, the litigants would have had to bear their own costs and there
would be further problems of potential enforcement. (147)

15.10. Full Indemnity or Reasonable Limits

The previous sections looked at factors such as issue-by-issue approaches or

procedural behaviour that might militate against full costs indemnity. These
were factors external to the successful claims and the entitlement flowing from
those alone. This section looks at whether indemnity itself is appropriate even
for wholly successful claims or whether tribunals should generally impose
reasonable limits on costs awards. This could be justified for two reasons, first,
that losers should not have to pay more than reasonable costs, and second, that
the imposition of such limits may discourage unreasonable cost expenditure in
the first place.

This approach is mandated by certain arbitral rules. Article 37(1) of the ICC
Rules 2012 limits recovery to reasonable legal expenses. Article 40.2 of the
UNCI-TRAL Rules 2010 defines ‘costs’ in terms of ‘reasonable’ specific elements.
A similar approach is taken in the English Arbitration Act 1996 and the LCIA
Rules. Article 35(1) DIS Rules considers costs incurred ‘which were necessary for
the proper pursuit of claims or defences (emphasis added). Article 32(3) of the
Vienna Rules speaks about the ‘appropriate’ expenses of the parties for their
representation. This might include considerations of proportionality. (148) Section
609(1) of the Austrian Code of Civil Procedure also speaks about ‘reasonable
costs appropriate for bringing or defending against the action’. Some rules tried
to set percentage limits rather than rely on broad concepts. Article 59 CIETAC
Rules previously sought to impose a cap by indicating that compensation for
expenses shall not in any case exceed 10% of the total amount awarded to the
winning party. The current Rules provide the tribunal with a discretion to order
compensation of expenses ‘reasonably incurred’ by the winning party. (149) Even
where some form of reasonableness test is not expressly referred to in the
arbitral rules, a tribunal's broad discretion could be argued to allow it to impose
reasonable limits. Craig, Park and Paulsson suggest that it is wholly within the
arbitrator's discretion to determine what is reasonable, although they were
writing about ICC arbitration where reasonableness is expressly referred to in
the rules. (150) A contrary view is that the consent basis of arbitration implies each
party being free to organise legal representation as it sees fit as long as this is
done in good faith. (151)

page "1230"

Because of the uncertainties where the rules are silent, the tribunal could invite
submissions from parties on this and determine its position at an early
procedural directions meeting. (152)

It is more difficult to assess what makes expenses reasonable, necessary or

adequate and how a determination is to be made. These phrases need to be
interpreted in a common sense way, taking into account what each party would
reasonably expect of its counsel in order to win the case. Adequate procedures
would be those that ought to be taken on a cost-benefit basis in order to achieve
the outcome sought. At the outset there is also the question as to whether a
reasonableness test should be based on objective or subjective standards or a
combination of the two. Most would consider an objective approach to be
appropriate, looking to see what costs a reasonable person would have been
expected to incur to properly pursue the claims or defences as the case may be.
From an implied consent perspective, that might go one step further and
consider what the other party might reasonably have anticipated and hence what
it impliedly accepts as its risks in an adverse costs award. (153) The tribunal may
be aware from an early stage which matters it considers to have been most
contentious and hence worthy of more effort, but may not have given any such
indication to the parties. Where simultaneous submissions are involved, each
party does not know how much work and effort the other party is putting in,
hence an advocate would be very brave if it was too concerned with economy of
effort. Because of these factors, a tribunal should concern itself with the
reasonable behaviour of a party at the time the expense was incurred that is
presumed to rightly wish to take appropriate steps to win the case when it does
not know the responses that the other party will take. When considering
expenditure at the time it was incurred, a tribunal ought to consider the
complexity of the issue, the importance of the issue to the likely outcome and the
risks involved as to success or failure on that aspect in terms of the ultimate
proceedings.

One difficulty in assessing the reasonableness or otherwise of fees is that in

many cases, what a party spends is in part affected by its opponent's behaviour.
A related difficulty with arbitration is that even where a claim is not defended,
the tribunal must still call on the claimant to meet its burden of proof. There is no
equivalent of automatic judgment in default of appearance in international
arbitration. (154) Another factor which might militate against tribunals
considering the reasonableness of costs is their own impact on this through
procedural rulings, including determinations as to the length of the hearing and
entitlement to document production. Another related problem with determining
reasonableness is in relation to the tribunal's own costs and fees. There is
obviously a conflict of interest where a page "1231" tribunal and/or an
institution have already set the amount or range of fees involved. Another
problem is that costs which may reasonably be incurred to improve the
efficiency of the arbitral tribunal's own deliberations, may actually add to the
costs of the parties. An example would be a direction that post-hearing briefs are
properly integrated with earlier submissions, witness statements and
transcripts, rather than allowing counsel to utilise excessive repetition, leaving it
for the tribunal to put things in an elegant order. Conflicts may also arise where
the tribunal makes procedural determinations for its own convenience, for
example calling for hearings to be held in the home State of the tribunal
members.

A commonly cited exposition of the notion of reasonableness is that of Judge

Holtzmann. He considered that the following principles should guide costs
determinations:

1. Were costs claimed in the arbitration?

2. Was it necessary to employ lawyers in the case?
3. Is the amount of costs reasonable?
4. Are the circumstances of the particular case such as to make it reasonable to
apportion such costs?

The first two will invariably be affirmatively answered in most arbitrations. As to

‘reasonableness’, he made the following observations:

A test of reasonableness is not... an invitation to mere subjectivity. Objective tests

of reasonableness of lawyers' fees are well known. Such tests typically assign
weight primarily to the time spent and complexity of the case. In modern
practice, the amount of time required to be spent is often a gauge of the extent of
the complexities involved. Where the tribunal is presented with copies of bills for
services, or other appropriate evidence, indicating time spent, the hourly billing
rate, and the general description of the professional services rendered, its task
need be neither onerous nor mysterious. The range of typical hourly billing rates
is generally well known... Just how much time any lawyer reasonably needs to
accomplish a task can be measured by the number of issues involved in the case
and the amount of evidence requiring analysis and presentation. While legal fees
are not to be calculated on the basis of pounds of paper involved, the tribunal by
the end of a case is able to have a fair idea, on the basis of submissions made by
both sides, of the approximate extent of the effort that was reasonably required.
Nor should the Tribunal neglect to consider the reality that legal bills are usually
first submitted to businessmen. The pragmatic fact that a businessman has
agreed to pay a bill, not knowing whether or not the Tribunal would reimburse
the expenses, is a strong indication that the amount billed was considered
reasonable by a reasonable man spending his own money, or the money of the
corporation he serves. That is a classic test of reasonableness. (155)

page "1232"

As to the latter observation, there are a number of presumptions behind it, not
all of which are necessarily realistic in the modern world. To suggest that it is
presumed reasonable because it was accepted, presumes that there was an
actual opportunity of the client to negotiate a lesser fee. If instead, that is
effectively the norm of all top law firms, a client wishing to use their services has
little practical choice. Second, there is no presumption that the selection of a top
law firm that typically devotes maximum resources to major cases is necessarily
reasonable from a cost effectiveness point of view in every case where it is hired.
Third, even the most sophisticated business person is in the hands of their legal
advisers. If the adviser says that significant effort is required, it would be a brave
business person that would second guess their counsel. In addition, one party
might engage in excessive expenditure to try and pressurise the opposing party's
cash-flow in trying to keep up with their legal manoeuvres. Finally, there is no
automatic identity between the commercial reasons why a party might wish to
pay its own lawyers and the policy reasons why a losing party should
compensate them for doing so. As Bühler and Jarvin have noted ‘since each party
is free to manage its litigation resources as it sees fit, the issue is whether it is
reasonable to place a financial burden of that party's decision onto the other
side’. (156)

Thus, while a tenable observation, it would be undesirable if it was a strong

albeit rebuttable presumption on the part of adjudicators. Were it to be a
presumption, it would also effectively shift the onus of proof on the losing party
to show why costs were unreasonable. Judge Holtzmann's separate opinion
needs to be looked at in context and was in fact his commentary on the
majority's view that the reasonableness of legal fees should be considered in a
cautious manner. (157) In his separate opinion he noted that ‘the almost casual,
and certainly inconsistent, way in which costs have been awarded, and not
awarded, in so many of the Tribunal's past cases tends to obscure the fact that
basic rights are involved’ . (158) He was justifiably arguing in favour of legal cost
reimbursement in an environment where many adjudicators had an adverse
view. He was not suggesting a rule of thumb for all commercial arbitral
proceedings.

There are other questions and cautions as to the methodology and criteria by
which to determine what are reasonable levels. Tribunals should be careful not
to simply compare the costs of each side as they each may have very different
things to prove or disprove. However, capping costs equal to the losing parties
costs can be a disincentive to excessive costs by the winner and can minimise the
chances page "1233" that the loser could argue that fees are excessive. The
economic logic might still be outweighed by restitutionary principles. Tribunals
also need to be mindful that assessments of reasonable costs can themselves
become costly without some kind of presumptive regimen as is found in cost
scales in some domestic courts. In most cases it would be undesirable to have a
practice of tribunals scrutinising the files and considering the amount of time it
took to interview witnesses, draft documents and engage in research, to try and
reduce cost awards. However, in appropriate circumstances, to fail to even
consider such an approach might be unfair to the losing party and would
constitute an inadequate disincentive to excessive costs in the future.

A factor promoting consistency in many domestic litigation systems is that

presumptions and principles are often employed by a particular court officer
such as a Taxing Master or Registrar, rather than being left to individual judges
where inconsistent approaches may typically apply. As noted in section 15.6,
some arbitral rules allow for reference to a court for taxation of arbitral costs.
The advantage would be consistency, but a problem could be lack of experience
with arbitration per se or with the intricacies of the particular case. The option
noted is the use of an arbitration institute registrar or person of similar
experience.

One way to better consider the reasonableness or otherwise of costs is to break

the analysis down into various stages, although this presumes a procedural
paradigm in assessing costs rather than an indemnity one. The Chartered
Institute of Arbitrators in its recent survey of fees and costs asked parties to
break down the percentage between pre-commencement, commencement,
completion of exchange of written pleadings/memorials, document production,
fact witnesses, expert witnesses, hearing preparation, hearing, and post-hearing.

15.11. Factual Assessment and Verification

Arbitral rules rarely indicate the procedure by which cost determinations should
be made. As always, a tribunal should be bound by the agreement of the parties.
The tribunal should also be concerned with mandatory procedural laws,
primarily giving the parties an adequate opportunity to present their case as to
costs. Equal treatment also needs to be considered, although the concept applies
somewhat differently to other submissions depending on whether the cost
procedure is commenced before or after a decision on the merits is known.

A claim for costs, like any other claim, requires proper attention to burden of
proof and adequacy of evidence. The burden ought to be on the person seeking
the costs. While the burden remains, the onus may shift depending on the
material presented or even depending on the presumptions in the rules. For
example, a presumption in the rules that costs follow the event, in a practical
sense at least, calls for the losing party to explain why the circumstances ought to
be different in the instant case. There is also a difference between proving that
the costs were incurred and proof as to whether they are reasonable or not. In a
practical sense, the onus may be on the loser to explain why the fees are
unreasonable.

page "1234"

Questions of proof raise issues as to procedure, timing and the type of material to
be provided. As to the evidentiary material, this will comprise legal bills, bills for
other professional services and witnesses, copies of travel bills and perhaps
tickets, accommodation and sustenance, phone, fax, email, secretarial, translation
and interpretation bills. Otherwise meritorious claims may properly be denied
on the basis of a lack of sufficient evidence. The Chartered Institute of
Arbitrators' Guidelines for Arbitrators on Making Orders relating to the Costs of
the Arbitration makes a number of recommendations. (159) As to procedure, it
suggests that there first be a list of claimed costs together with their justification,
an entitlement for the paying party to raise objections, an entitlement of the
receiving party to reply and then an assessment stage by the tribunal,
culminating in the issue of an award. The Guidelines suggest that where there are
large sums in issue, a short oral hearing may be held to receive representations.
Another question is even whether document production rights are applicable to
identify the work actually done. From a due process point of view, if each party
must be given an opportunity to be heard on all aspects of the case, that should
also relate to cost claims being made against it. Yang points out that the
Guidelines do not give any indication about disclosure or witness statements in
relation to such oral hearings. He also raises the problem of privilege in relation
to such evidence. (160)

Timing is also vital, in particular, the views of the tribunal being made known at
the earliest possible stage. It would be most desirable for the tribunal to give an
indication as to the type of documentation and time recording it requires for its
ultimate costs decision so the parties are forewarned and can take appropriate
action. If the tribunal makes known its requirements at the outset this will also
avoid it perhaps being placed in a difficult situation where inadequate
documentation is provided, but an extension is sought in order to complete the
record.

The question of the degree to which a party needs to substantiate its costs claims
can depend on policy considerations and also considerations as to the legal
entitlement. From the latter perspective, it has been argued that because of a
tribunal's broad discretion ‘the parties do not have to give full evidence on the
parties' costs and therefore have only a limited duty to substantiate’. (161) Such a
view mixes the notion of the relevant criteria and the evidence by which such
criteria are to be determined. This is not to suggest whether this conclusion is
preferable or not from a policy perspective, but rather, to simply say it does not
flow inexorably from the mere presence of discretion. The field of administrative
law is based on bureaucratic discretion, but there are still burdens of proof and
evidentiary burdens applying in a whole host of such areas. Craig Park and page
"1235" Paulsson suggest that cost details should be subject to proof akin to any
other proof of damage. (162) Many arbitrators would appear to simply accept a
general fee note from counsel. Others have rejected claims for lack of supporting
evidence. (163) One reason to not impose a strict evidentiary burden is if it is
considered to merely be a question of estimation by the tribunal and not a
determination of an issue of fact. (164)

One element of the assessment of costs should be causation: were the costs truly
caused by the adjudicatory process? In some instances, costs might have been
incurred in any event, for example accounting calculations required for taxation
purposes. In other circumstances the actions might have been undertaken but
not to the degree chosen for the adjudicatory process. Here some fair and
reasonable apportionment might need to be made. Another question is whether
costs should be assessed on a global basis or whether there should be an attempt
to evaluate the entitlement and individualise it. A global assessment will be
quicker and cheaper but may not be as accurate. It may also hide a lack of
reasoning which could lead to dissatisfaction by the parties and even a challenge
as to that aspect of the award. One argument in favour of a global assessment is
that where there are a range of matters worthy of consideration, there is no clear
objective formula that can lead to a particular figure. This is so where a tribunal
wishes to consider relative success, procedural behaviour, approach to
settlement, excessive claims and the reasonableness of the fees of counsel and
other key expenses. Even then, a tribunal ought to articulate which factors led to
the global assessment. An alternative method would be to make an assessment
as to the appropriate number of hours, an assessment as to the appropriate fee
for different activities and compute a total.

Given that there are a range of approaches, a tribunal ought to advise the parties
of its predisposition when seeking costs submissions from them. From the
parties' perspective, it is important to ensure that there is sufficient
documentary evidence to support ultimate costs claims that may be brought. It is
particularly difficult trying to calculate costs and identify an evidentiary trail
after the event. Appropriate detailed bills should be constructed by counsel on an
ongoing basis. This is particularly so if in-house counsel's costs are to be claimed,
where both time devoted and a methodology of computing sums allocated are
necessary. Raising the suggested evidentiary process at an early stage may also
give a tribunal an opportunity to give guidance where more is thought to be
needed, although tribunals may be reluctant to prejudge a future factual
determination in this way.

One complex scenario is where a tribunal is asked to assess costs of litigation,

typically as damages for breach of an arbitration agreement. The applicable rule
on damages assessments should typically apply, and not the relevant court scale,
although that scale may be a guide to reasonable levels and hence what has truly
been caused as damage.

page "1236"

15.11.1. Calling for Evidence of Costs before the Result

The tribunal can assist its ultimate determination by calling for an appropriate
breakdown of fees and costs. Many leading arbitrators will call for information as
to the legal costs of each party prior to the rendering of the award. (165) At that
point in time, at least in cases that are finely balanced and where no one can yet
predict the outcome, each is likely to make the fullest claim. If there is no
significant difference between the level of fees, a tribunal would find it easier to
reject subsequent submissions that the winner's costs are excessive. If costs are
significant, each party might be invited to make submissions in relation to the
other side's cost assessment. Submissions in favour of or challenging costs
assessments would need to provide explanations and justifications as well as the
broad data. (166)

15.11.2. Costs after an Award Might Be Functus Officio

If the tribunal does not make a determination as to costs in its final award, it may
have become functus officio and be unable to render such a decision at a later
stage. Some lex arbitri expressly allow for the parties to subsequently call for a
costs determination. (167) In other cases, entitlement to an additional award could
be utilised. (168)

15.12. Cost Elements

Previous sections have looked at costs in general, although note was taken of
differing views as to general entitlement to legal costs as opposed to tribunal
costs. This section looks at particular types of costs in more detail. Biihler
categorises arbitral costs into two groups, procedural costs, including fees and
expenses of arbitrators and any institution or tribunal appointed experts; and
parties' costs, being those incurred by the parties in the preparation of the case.
(169) The kinds of costs that may be recoverable can in part depend on the way

costs are referred to in applicable rules. For example, some rules such as Article
37(1) page "1237" ICC Rules 2012 are drafted broadly to encompass ‘the
reasonable legal and other costs incurred by the parties for the arbitration’.
Others are more restrictive in language. Under some rules, designated costs are
exhaustively defined. (170) An example is Article 40.2 UNCITRAL Rules 2010
which refers to costs ‘only’ including enumerated items. The following sections
look at particular categories and their suggested treatment in more detail, with
the caveat that express provisions in statutes, rules and party agreement would
be determinative.

15.12.1. Cost before Commencement of the Arbitration

In any adjudicatory system, there is a danger in making significant expenditure
prior to commencing an action where concern is for cost recovery of these
amounts. An adjudicator may refuse to award such costs on the basis that they
are premature or that they are better characterised as relating to legal demands
of settlement rather than the adjudicatory process itself. (171) They could also
suggest that they were excessive in trying to promote settlement without legal
action. Limitations may also apply through a textual analysis of the rules. For
example, if a tribunal is considering ‘the reasonable legal and other costs
incurred by the parties for the arbitration’ (emphasis added) under Article 37(1)
of the ICC Rules 2012, expenses other than those necessary for the preparation
of the case and those directly linked to the filing of the request may not be seen
as costs of the arbitration. (172)

There may be an added problem in arbitration. Because a tribunal's cost powers

might be limited to matters flowing from the particular submission, uncertainty
may exist in relation to entitlement for costs incurred before the commencement
of the arbitration. (173) That is certainly a conservative view as to the powers of an
arbitrator and does not fit well, for example, with an arbitrator's jurisdiction
over pre-contractual negotiations under a broadly drafted arbitration
agreement. Nevertheless, because the situation is arguable, care should be taken
in such circumstances. The unfortunate corollary is that parties may be induced
to commence proceedings at an earlier stage, which might undermine the
potential for mutually agreeable settlement.

page "1238"

15.12.2. Counsel 's Fees

Section 15.5.3 above, which considered whether tribunals should impose

reasonable limits on costs claims, was as much as anything concerned with
counsel's fees. In part this is because studies have shown that counsel costs are
easily the most significant element of the total cost of arbitration. For example,
the ICC reported that legal costs were approximately 82% of the total arbitration
costs. (174) Thus, such fees will normally be the key part of a costs application,
although it has been noted that in some jurisdictions, a tribunal is not entitled to
award attorneys' fees. (175) It is also the category where legal families generally
differ the most, some tribunals are more reluctant to award counsel's fees than
arbitration fees and expenses. While the norm has tended to be to award
winning parties their legal fees, this is not so for investment arbitration. (176)

If legal fees are to be awarded but with reasonable limits imposed, relevant
factors would appear to be complexity and novelty of the issues involved, time
required, level of skill demanded, the amount in dispute, the expertise and
reputation of the attorney, charge out rates, and fee awards in similar
circumstances.

The comments about the challenges in articulating and applying reasonableness

tests in section 15.5 above are not repeated here but remain relevant. Some
additional observations are appropriate. If tribunals are to try and assess
reasonable counsel costs, it should be an assessment that is realistic and not
paternalistic. Even if a tribunal is disposed or required to impose some
reasonable limits, there is no justification for imposing domestic litigation scales
on the parties. For example, they ought not to follow the very rough rules of
thumb in non-US common law jurisdictions that distinguish between party/party
costs and solicitor/client costs, regularly only awarding the former and often as
little as 40% of the fee actually charged. (177) The suggested policy's aim in
domestic litigation was to encourage settlement, but if so, it is a very rough
measure and would operate unevenly on parties depending on their financial
wealth. Domestic practice also shows no consistency between jurisdictions. (178)

page "1239"

It would also be overly simplistic to concentrate solely on the hourly rate that
counsel charges to determine if the fees are reasonable. For example, the more
expert the counsel, the less work they might need to do at that higher rate to
achieve the same result. However, they may address more issues and hence lead
to higher costs again. A tribunal might take note of domestic practice to
determine whether counsels' behaviour in the instant case seems atypical and
potentially unreasonable without further explanation. One problem is that a
tribunal might face arguments as to confidentiality where it seeks to identify a
law firm's normal charging practices to compare with the bill in the instant case.
A further problem in attempting to identify reasonable counsel fees is that there
is no consensus worldwide as to how legal counsel should charge. The key
difference is between time costing and value costing. The former looks at the
actual time spent, regardless of the anticipated or eventual relevance of the work
done. Value costing can look at the importance of the matter to the parties or the
importance of particular issues and arguments that are prepared. Because there
are tenable arguments in favour of each approach, there is hence no consensus
about the method of determining what reasonable attorney fees should be.

Some may charge a flat fee. A flat fee based on an estimation of the time spent
could be looked at in two ways. It could be awarded in full on the basis that it
was a reasonable estimate and a proper commitment. The alternative approach
is to look at the time actually taken to see whether it was an underestimate or an
overestimate. That would be inelegant as there would be no power to uplift the
award, only reduce it. The situation is even more problematic with value-based
costing. That can be a very reasonable commercial decision on the part of the
client, particularly if it is outcome based. Once again, however, there is no
presumption that this is a fair and reasonable amount to award against a losing
party. Where the value fee is similar to time-based estimates, then there is little
practical difference. In other circumstances, a higher value fee could not be
presumed to be reasonable and necessary for the pursuance of the legal right.

Another difficulty with international adjudication is that there will typically be

very significant differences in levels of costing and approach to legal matters in
different jurisdictions. A multinational law firm headquartered in a developed
country may use large teams of senior and junior lawyers to scour every possibly
relevant document and argument and interview and prepare a host of potential
witnesses. Hourly rates even for paralegals are often extremely high by
community standards. Conversely, lawyers in some developing countries at
times have to hold a number of jobs as they cannot earn a sufficient amount from
legal practice alone. Developing States that are parties in investment arbitrations
may rely heavily on departmental officers in preparing cases, although this has
had mixed success. While such differences are disconcerting, they should not ipso
facto be a basis for grossly interfering with the commercial reality as to the costs
incurred. The underlying commercial transaction giving rise to the dispute was
motivated in large part by differences in comparative advantage and costing.
Consensual arbitration ought not to ignore those commercial realities. This is not
to say that a tribunal ought to accept excessive legal costs, but it cannot a priori
presume that one party's costs are too page "1240" high and/or the other's are
too low simply because of marked national differentials. Nevertheless, tribunals
have at times cut down recoverable fees where the disparity in rates between
counsel for each side were alarmingly high. (179)

There may also be issues as to the number of counsel that would be considered
reasonable. It is not uncommon in international arbitrations for a party to
employ more than one legal firm, perhaps its domestic legal firm working in
conjunction with an international arbitration practice. There might also be an
agency arrangement in a foreign country where that country's substantive law
applies or where there are important evidentiary materials within that
jurisdiction. In these circumstances a tribunal might reduce the costs awarded
where it feels that there was unnecessary duplication of effort. (180) In addition, if
parties change counsel in the middle of a case, the award might be reduced in
relation to overlap between the work of the two firms, particularly if the reason
is because of dissatisfaction with the first law firm.

Other multi-counsel issues in determinations of reasonableness might include

whether single or co-counsel should be involved. In the common law this may
also involve determining whether a Queen's Counsel or a more junior barrister
should be briefed. Once again, tribunals should not readily wish to be
paternalistic in such cases. A more contentious category would be an expense
incurred to obtain an expert opinion from an independent lawyer that is not
intended as a witness.

15.12.3. Contingency Fees

Where a law firm is charging on a contingency basis, an added complication in

international arbitration is whether the winning party should be compensated
by the loser for the contingency fees it has paid to its legal firm. Under a
contingency fee arrangement, the winner pays the law firm more than a typical
fee to account for the fact that reduced fees or even no fees are paid for running a
losing case. A related situation is if a law firm has a variable fee arrangement
with its client depending on a successful outcome, more typically described as a
success fee. For example, a firm might charge lower than its normal rate, but
have a surcharge above the normal rate in the event of success. The complication
is that generally speaking, contingency or success fees are thus higher than
normal to account for the fact that cases are lost as well as won. (181) Should the
loser in international page "1241" arbitration, even subject to a loser-pays
principle, fully compensate the winner for that kind of a systemic trade-off not
agreed to by the loser?

A preliminary issue concerning contingency fees is whether there is even power

to award costs at all, given that such arrangements commonly arise in systems
that require each party to bear their own costs. As noted previously, if the
arbitration has a seat in the US, at least some courts might consider that there is
no such power unless expressly granted. Other legal systems that generally do
not award costs to the winner include Indonesia, Japan, the People's Republic of
China and the Philippines. (182) Another complication is that in some jurisdictions,
contingency fees are against public policy or are in breach of express ethical
rules. There would be a range of permutations depending on whether this is the
rule of the seat or the attorney's home country or of an enforcement jurisdiction.
The better view would be that a tribunal still has a discretion in each scenario
unless the rule in the seat should be read as intending to apply to international
arbitration. Wehrli argues that domestic ethical rules about attorney practice
should apply to international arbitration. (183)

Some arbitral rules might impact upon the question indirectly. For example, it
might be even more difficult to justify a contingency fee under a lex arbitri or
arbitral rules that only allow for the awarding of ‘necessary’ costs. (184) Other
guiding principles look at the level of charges, although generally from the
perspective of the lawyer and client and not the losing party under a costs award.
For example, Article 18 of the IBA Code of Ethics, 1988 edition states:

The contract for a contingent fee, where sanctioned by the law or by professional
rules and practice, should be reasonable under all circumstances of the case,
including the risk and uncertainty of the compensation and subject to
supervision of a court as to its reasonableness. (185)

Where there are no constraints, a range of options are possible. Assume for
example that an English company is in arbitration with an American company
with the seat in England. The presumption is that costs follow the event. The
English company pays an hourly rate to its lawyers while the US company is on a
contingency fee basis promising a percentage of its winnings but with no costs
obligation were it to lose. If the US party loses, it indemnifies the English party
for the hourly rate paid to its law firm, subject to the fees being seen as
reasonable in the circumstances. There are then three ways to deal with the page
"1242" contingency fees in the converse situation where the US company wins.
At one extreme, it could be provided with full indemnity on the basis that
contingency is just as reasonable a methodology of payment as is an hourly rate.
The mid-position is to suggest that the contingency fee needs to be apportioned
between that part that is truly reasonable costs of preparing the case and the
component that is akin to a premium for the promise of non-payment if a result
is not obtained. Here the aim would be to equate to the reasonable fees as paid
by the English party. This might be calculated by way of a reasonable hourly rate
for the work done or a broader assessment of the contingency margin over
normal fees. The most extreme situation is to say that a party entering into a
contingency fee arrangement with a law firm does so on the assumption that no
fees will be paid if it loses and no costs recovery will be obtained from the other
side were it to win. Thus, the party went into the arrangement knowing that
there will be something less than full indemnity where there is a successful
outcome. While the latter is certainly the background to contingency contracts in
American domestic litigation, this presumption should not apply in arbitration,
particularly when the seat expresses a presumption that costs follow the event.

Where there are no overriding constraints, the key question is thus whether a
tribunal should award full indemnity in favour of a winning party as to the
contingency fee element. One justification for doing so is based on a restitution
perspective. The winning party incurred that expense as a result of the losing
party's breach and should be put in the position it would have been in if the
commercial arrangement had been properly undertaken. The converse argument
is that the breach did not cause that loss. Instead it was the independent decision
to engage in that kind of legal representation contract. Another possible
justification for contingency fees is that they operate akin to an ad valorem basis,
not dissimilar from institutional charges and arbitrators' fees under many
institutional rules. If kept to reasonable percentages, they can be consistent with
desirable transaction cost ratios.

Conversely, a number of commentators argue against the view that a loser-pays

principle leads to an obligation to compensate for a full contingency fee. Bühler
has rightly pointed out that if a claimant subject to a contingency fee
arrangement sues a respondent who pays on a time basis and loser pays is
applied formalistically, the entire cost risk of claimant's counsel is shifted to one
party alone. (186) O'Reilly supports the view that recovery should be limited to an
amount that would have been payable on a time charge basis. (187) Wehrli argues
that under an objective test of reasonableness, costs incurred by reason of
personal circumstances would not generally be seen as the basis of reasonable
reimbursement. page "1243" Under this principle, contingency fees exceeding
otherwise reasonable fees should not be reimbursable. (188)

Another way to consider the issue is to think of the commercial aim of the
contingency payment when made. Arbitrators might be legitimately reluctant to
award the higher amount as it is not truly indemnifying the winner for its
legitimate expenses in running a winning case but is instead compensating it for
a promise similar to an insurance policy it took out privately with its counsel in
case it was not as successful as it turned out to be. Paying a premium when a
party wins to avoid any legal costs were they to lose, is not dissimilar to a
premium for an independent insurance policy against losing. If a party took out
such an insurance policy with a separate company, the premium should not be
recoverable as costs were it to win. Such a premium is neither damages flowing
from the breach or reasonable costs involved in pursuing a successful outcome
through an adjudicatory process. (189) Hanotiau describes the fees as a reward for
success which similarly should not be compensated for. (190) Another problem
with providing a full indemnity to the winner under a contingency arrangement
in arbitration, is while it may have felt that it obtained commercial value in
offering a significant percentage of winnings to avoid legal costs in the event of a
loss, the opposing party was never invited to consent to this arrangement. The
more that all the parties conducted themselves in this way, the more a tribunal
might take a contrary view, based on their presumed legitimate expectations.

A practical question is whether there is even a duty to disclose the fact of a

contingency style arrangement or whether the winning party can simply proffer
a bill setting out the amount of the charges. Duties of honesty and the duty to act
in good faith suggest that disclosure should occur so that a tribunal can properly
address the issue. (191) Newmark also notes the important observation that
success fee arrangements can compromise the legal adviser when being asked
about settlement and can even give rights to the lawyer in terms of the client
using its best endeavours to pursue the adjudicatory process, lest the lawyer
unfairly lose the potential fee entitlement, (192) although this observation is more
about whether they can be charged and less about a tribunals response when
this occurs.

page "1244"

15.12.4. Litigation Funding by Third Parties and Reimbursed Costs

It is becoming more common for parties to enter into arrangements with

litigation funding businesses to cover the costs of proceedings in return for a fee,
either based on interest on costs advanced or a percentage of winnings or a
hybrid of the two. Similar cost award considerations apply as with contingency
fees in apportioning between reasonable fees and added costs for ancillary
services.

There is one added concern which is the potential in common law countries at
least for the overall arrangement to be considered illegal or unenforceable under
the principles of champerty and maintenance. The common law historically took
issue with many such arrangements, particularly where unmeritorious claims
were handed over to wealthy and powerful individuals in the hope that
involvement of such persons would influence the court towards a favourable
outcome. (193) Common law courts in more recent times have tempered their
views about these doctrines, particularly in relation to litigation funding
arrangements. They have rightly shifted their policy concerns from wealthy
persons taking over unmeritorious claims, to a concern to allow financial
institutions to help those with worthy claims but inadequate resources to be able
to exercise their legal rights. In arbitration, there is a further issue as to whether
such principles developed in the court system are necessarily relevant to private
consensual arbitration. Here there have been quite different views. (194) One view
would be that concepts of champerty and maintenance should only be of concern
in arbitration in the most extreme circumstances, but that in the normal case,
litigation funding should be accepted as little different to factoring of debts.
Importantly, litigation funders are likely to be a barrier to unmeritorious claims
as they will consider all funding decisions on a careful cost/ benefit basis, thus
indirectly helping parties evaluate claims. (195) The issue remains controversial,
however. (196)
Once again as with contingency fees, in considering both the legality or
otherwise or the appropriateness of litigation funding, care needs to be taken not
to resort too quickly to the consensual basis of arbitration as a reason for page
"1245" distinguishing arbitration treatment of the issue. Most aspects of
arbitration are in fact agreed to between the parties but the way a particular
party chooses to engage counsel and fund its adjudicatory activities will almost
never come with express consent of its opponent. At most, one could argue from
an implied perspective. Here there might be two competing a priori
observations. On the one hand, an opponent that has agreed to arbitrate disputes
should not feel aggrieved if the other party has cash-flow problems and needs to
go to a litigation funder to pursue the very agreed dispute resolution mode.
Conversely, unless it was part of the legitimate expectations at the time of
entering into the arbitration agreement, that opponent cannot be presumed to
have agreed to pay a cost premium on losing, simply because its opponent chose
to pay a premium for funding support or for a contingency arrangement. The
better view is that the profit that the litigation funder takes, either as interest on
a loan or as a percentage of the result, should not be compensable as part of a
costs order. Where costs awards are concerned, a decision to use a litigation
funder will generally be a personal decision about cash-flow and about use of
available resources. The entire fee could not be described as reasonably
necessary to pursue the adjudicatory process. Thus, there ought to be an
apportionment on a similar basis to that of contingency fees, or simply a denial of
the element over and above the cost of legal counsel to account for the cost of
adjudicatory financing. It could alternatively be claimed by way of damages if it
can be shown under the relevant test of damage that resort to financing support
was necessary and hence directly caused by the losing party's breach. There may
also be questions of foreseeability in such circumstances. Costs as damages is
itself contentious.

Where costs are first reimbursed by way of an insurance or indemnity right, one
tribunal considered that counsel's costs as retained by the party were
recoverable but the successful party would then need to reimburse the insurer.
(197) It seems debatable whether the tribunal should have effectively looked for

costs recovery by the insurer instead of considering the out-of-pocket expenses

of the successful party, including the costs of the insurance policy. The situation
may vary depending on whether the insurer has a subrogation right vis-à-vis the
insured.

15.12.5. In-House Counsel

There is no clear view as to the costs entitlement in relation to in-house counsel,

save to say that a tribunal is much more likely to allow costs of external advisers.
(198) A contrary view was taken in a number of ICC cases which accepted the

possibility page "1246" in principle, (199) although other cases have rejected such
claims. (200) There has also been scholarly support for such entitlement. (201)
Depending on the country concerned, there may also be ethical and other
statutory principles that either limit or empower the entitlement to in-house
lawyer costs. (202)
Even where there is entitlement to the cost of in-house counsel, there will be
difficult questions of apportionment and substantiation. (203) Where a party
intends to claim the costs of in-house counsel, careful time records should be
kept. (204) An added difficulty is to determine the rate of such cost entitlement.
Calculations may need to be based on salary plus on-costs, perhaps including
overheads. Any hypothetical figure based on what they might have charged if
external, would not be true costs incurred and ought not to be the basis of any
calculation. In addition, it would be hard to identify a benchmark normal fee to
use in any such hypothetical assessment. A tribunal might well impose higher
substantiation requirements where in-house counsel are concerned. (205) Another
approach is not to consider in-house counsel on an all-or-nothing basis but to at
least consider some allocation where the particular adjudicatory proceeding
calls for more than the average amount of time as contemplated for in-house
counsel. (206) There will also be issues if external counsel are also used. There is
no reason in principle why a party cannot allocate work between in-house and
external counsel and have recovery for both where there is no unreasonable
overlap. (207) In other cases there may be a need to distinguish preparation costs
from mere reporting costs. As to the latter, it has also been observed that an
over-zealous in-house counsel can indirectly increase legal fees simply because
of the need for external counsel to deal with the internal counsel's queries. (208)

page "1247"

Because of the uncertainty as to the entitlement of in-house counsel, a tribunal

might well indicate its views at an early stage or at least flag this as a matter for
submissions. At the very least this can warn the parties as to the type of
evidentiary records that the tribunal would wish to see.

15.12.6. Self -Representation

Domestic litigation systems will rarely allow self-represented litigants to obtain

costs. At most they will allow for necessary out-of-pocket expenses. (209) The
situation may be different where the party is legally trained but even here there
is a difference between legal systems. (210) One possible justification for such
expenses is that arbitration does not generally demand representation by legally
qualified persons. The issues are a sub-element of the entitlement of a party to
its own costs and lost time, discussed in section 15.12.12 below.

15.12.7. Costs of Document Production

While document production is normally seen as part of counsel fees, there are
two reasons to consider it separately. First, it is an area of particular difference in
view among legal families, some arguing for the fullest production obligation in
aid of promotion of truth, while others take a libertarian approach, arguing that a
party should not be compelled to assist the very person that has dragged it into
an adjudicatory process. The second reason is that the work of one counsel is
impacted upon by the production request made by the other. A related
complication is that the tribunal will typically have ruled on the production
request. It is difficult for the tribunal to make a particular order for production
and then later say it was excessive and should require a negative cost order,
although this may be so when subsequent evidence makes this clear. The final
complication is that the effort required to identify and collect the relevant
documents is often undertaken by the parties and their internal employees,
raising the broader question whether any such time and effort of such persons
should be compensable by way of costs orders. This is again discussed in section
15.12.12.

The costs of document production can include costs of searching, photocopying,

collating and distribution. There may also be inspection costs where the other
party is entitled to view a range of documents to determine which are to be
copied.

page "1248"

15.12.8. Witnesses, Experts and Advisers

Witness costs can include travel, accommodation, ancillary expenses and fees.
Other expenses of witnesses would include subsistence costs. (211) The original
committee drafting the UNCITRAL Arbitration Rules saw different views
presented as to whether witness costs should be allowed, some being concerned
with the overall cost of arbitration. (212) Distinctions need to be made between
ordinary witnesses and expert witnesses and, in the latter case, between
tribunal-appointed experts and party-appointed experts as discussed in Chapter
12.

15.12.9. Claims Consultants

Parties will sometimes use claims consultants to help prepare and advise on
claims. An example would be quantity surveyors used in construction disputes.
In principle such costs should be able to be awarded where reasonable and
necessary for the pursuance of the claim. (213) There will still be issues of
causation as at times, it might involve an argument that the survey should have
been undertaken anyway as part of the construction process. As to what is
reasonable, an argument may arise if a claims consultant charges more than
would normally be the case for legal counsel. In some circumstances, claims
consultants may also charge a percentage of the award. Here the same issues
would arise as with contingency fees for legal counsel discussed in section
15.12.3.

15.12.10. Advisers

In some cases, parties may seek the advice of a range of professionals, many of
whom are not ultimately selected to be witnesses. These should be looked at on
the merits in terms of cost recovery. If parties are able to show that they are
necessary for the presentation of their arguments or are at least reasonable in
proportion to the amount in dispute and the complexity of the case, and there is
no significant duplication, cost recovery may be appropriate. For example,
accountants might be used to assist in calculating measures of damages.

A complex area may be where a party seeks the advice of an independent expert
on quantum to decide whether the arbitration is worth proceeding with on a
cost-benefit basis. This might be an engineer in a construction dispute or a
forensic accountant on a commercial valuation issue. Care may be needed to
distinguish between situations where a person is an expert witness and
conversely where they page "1249" are providing expert advice and assistance
about the strategy and conduct of the adjudicatory process. In some cases this
might be seen as doubling up with the role of counsel and may lead overall to
unreasonably high total fees. This would not normally be so for damages
assessments where the expert typically is delegated the task of computation. An
example of a problematic scenario would be dress rehearsal arbitrations and
expert opinions from a range of leading arbitrators and retired judges as to the
likely outcome of the actual tribunal process.

Expert reports might also at times be considered from the perspective of

damages, for example, where they indicate how a claimant should mitigate or
repair the fault occasioned by the opposing party. (214)

15.12.11. Expert Witnesses

Costs will include professional fees, travel, accommodation and ancillary

expenses. There needs to be a distinction between tribunal-appointed and party-
appointed experts. This is because a tribunal-appointed expert is part of the
tribunal's own fees and expenses and is not based on a particular choice of the
parties. The distinction may also be important where the parties have an express
agreement as to the way tribunal expenses will be allocated or where the
tribunal wishes to make a distinction between party costs and tribunal costs.

While Article 37(1) of the ICC Rules 2012 only refers to tribunal-appointed
experts, the accepted view under the same provision of the 1998 Rules was that
it is not intended to exclude the costs of party-appointed experts. (215) Article 6.8
of the 2010 IBA Rules on the Taking of Evidence in International Arbitration,
treats fees of a tribunal-appointed expert as costs of the arbitration and leaves it
to tribunal discretion as to allocation.

15.12.12. Time and Opportunity Costs of Parties

Domestic courts do not generally compensate parties for disruption to their

business from the time allocated to the litigation itself. The same is true of
arbitration. Background preparatory work by employees digging out corporate
documents or conferring with lawyers and experts would not normally be
recoverable. The observations in relation to in-house counsel or self-
representation also apply to any other in-house expertise utilised to advance the
case. Arbitral rules that articulate sub-elements of recoverable costs invariably
do not make page "1250" any reference to the time involvement of the parties
and their employees and corporate executives. (216)
While the norm is not to compensate parties for their own internal costs and loss
of time allocated to the arbitration, (217) some cases have taken a different
approach. (218) A number of commentators have also questioned the justification
for the distinction between external and internal costs. (219) Bühler argues that
where document production is required, the cost of managerial time devoted to
that task should be recoverable if it can be identified and accounted for. (220) On
the one hand, the traditional view could be said to be an outmoded notion as to
the entitlement of lay persons, but on the other, it is not necessarily a marginal
cost, as key personnel engage in all aspects of the business for a fixed annual
salary, hence it would be hard to assess the true loss to the company caused by
re-prioritisation of tasks occasioned by the adjudicatory process. In some cases it
may be easier to justify a party's own costs as a head of damages. Where allowed
as costs, one domestic judge considered that this should be ‘actual and direct’
costs excluding overheads which are incurred in any event. (221)

15.12.13. Tribunal Fees and Expenses

Most systems allow for full recovery of tribunal fees and expenses. Where there
is a panel of arbitrators, the presiding arbitrator or Chair might receive a general
loading or be compensated at a higher level on an hourly rate simply because of
the extra work in providing the first draft of the award and dealing with certain
procedural rulings and administrative matters. Some rules require arbitrator
fees to be individually stated in the award. (222)

The level and methodology for calculating tribunal fees will be impacted upon
where an institution is involved. Institutions vary as to whether they promote
hourly rates or flat daily fees or a global amount depending on the amount in
dispute and the level of complexity. Even in an ad hoc arbitration, a tribunal page
"1251" will select from a range of fee options when negotiating with the parties.
Article 41.2 of the UNCITRAL Rules 2010 calls on the tribunal to take into
account the appointing authority's methodology to the extent the tribunal
considers it appropriate. Institutional ad valorem fee schedules typically have a
broad range allowing the institution or its supervisory court to set fees within
the range at a level commensurate with the process undertaken and the
complexity of the case. (223)

Another approach is to allocate fees on an hourly or daily basis as is the common

approach of many private lawyers. The LCIA operates on an hourly fee. The
current LCIA rate is ‘not exceeding £400 per hour’, though in certain
circumstances fees can exceed this. (224) ICSID operates on a daily fee schedule,
currently USD 3,000 per day. (225) In some cases, these fees may be varied by
agreement of the parties. ICDR adopts a more flexible approach, inviting
potential arbitrators to identify their rates for agreement by the parties when
making selections. These rates are looked at alongside the size and complexity of
the case and the amount of service provided in calculating a final fee, (226)
although the fees are generally passed on as charged.

The final approach, sometimes negotiated in ad hoc arbitrations, is a flat fee. In

such cases all relevant persons must make an assessment of the likely length of
proceedings and whether such a fee is fair and reasonable. There would also be a
question as to whether there is a final agreement on the flat fee or whether there
is also liberty to apply for a higher fee if the matter proves to be more complex
than anticipated, for example by a party adding new claims or defences. This may
depend on the contractual rights between arbitrator and parties, discussed in
section 2.3 above. While respected and ethical arbitrators will approach each
case with a view to adopting the fairest and most efficient procedures regardless
of the fee methodology utilised, it is still appropriate to address the insights from
economics as to the incentives of each. In the case of private lawyers and a range
of other professionals, an hourly rate encourages some to heavily work a file to
maximise fees. A flat fee provides the opposite incentive.

Finally, a decision by a tribunal as to its own fees cannot prevent the parties from
challenging the level under their separate contractual entitlements. (227) A
tribunal would be bound by an institution's determination as to costs, although
there may be a separate contractual claim if the institution is behaving
unreasonably. Article 41 of the UNCITRAL Rules 2010 provides a review
mechanism to the appointing authority both as to the proposed fees and later as
to the computed fees.

page "1252"

15.12.14. Secretary

Tribunal fees would also include the costs of an appointed secretary, although
that would be subject to prior consent of the parties. If instead a tribunal
appoints a secretary out of its own fee entitlement, it would not be reasonable to
make a costs order against the losing party in respect of that component
separately as absent consent, this has just been a private expense of the tribunal.

15.12.15. Appointing Authority

There may also be expenses of an appointing authority where one has been
utilised. The same would be so where the Secretary-General of the Permanent
Court of Arbitration designates the appointing authority under the UNCITRAL
Rules. (228)

15.12.16. Travel and Accommodation

Travel expenses will include those of tribunal, counsel and any witnesses.
Business class travel is generally the norm for matters of significance, (229)
particularly where distances are also significant. Where parties are witnesses or
assist or give instructions, their travel expenses would be covered for that
reason. If not, these may be included in a broad discretion although they might
equally be excluded under rules that only expressly refer to travel for the
tribunal, counsel and witnesses. It is useful to resolve these issues in a costs
protocol when both hope to win, when the parties are only instructing or
assisting.
15.12.17. Hearing Room

The rental cost of the arbitral hearing room would naturally be recoverable.

15.12.18. Transcription, Interpreters and Translation

Transcription and translation services are commonly organised by the parties

themselves. It would be rare for a tribunal to want to organise such functions.
Where the parties make the arrangements, there will often be some agreement at
that stage about fee sharing. Here it is important to be clear as to whether the
agreement is intended to remain as originally articulated regardless of the final
page "1253" outcome, or whether it is simply an agreement as to preliminary
payment to be varied by the tribunal's final costs award. The same issue may
arise with such items as institutional fees and room hire.

Normally these costs will be included in the overall costs of the arbitration, but
other approaches may arise where a translation is only for the convenience of
one of the parties.

15.12.19. Institutional Costs

This section looks at the institutions own fees and not the controls they impose
on arbitrator fees. As is the case with arbitral fees, there are a number of
differing costing approaches. The ICC adopts an ad valorem approach. LCIA
administrative fees are based on an hourly rate. It does not require terms of
reference and has no equivalent to the ICC s International Court of Arbitration
scrutiny of draft awards. ICDR fees are in proportion to the amount in dispute.
ICSID has a fee schedule.

Institutional rules can also impact indirectly on costs. For example, the LCIA
Rules allow for third-party intervention (230) and for tribunals to be constituted
on an expedited basis. (231)

15.12.20. VAT and Taxes

The cost of arbitration could include taxes imposed on the service fees of
professionals, including the tribunal itself. Any unavoidable on-costs should
similarly be compensable. More problematic is a registration fee on the award at
the enforcement stage as this is not a cost of the arbitration itself. It should be
left to the court procedures to determine recoverability and then only if
enforcement is necessary.

15.12.21. Security for Costs

The entitlement to security for costs was discussed in section 8.3 above as part of
the process of establishing the procedure. The only relevant factor for this
chapter is whether the costs of security should be awarded to a winning party
who was required to post security. That issue might have been dealt with by the
tribunal as part of the determination of the security application. If it was not
addressed at that stage or was left to be considered at the end, it is arguably a
cost of proceedings, as without security, the claim could not be pursued. It is also
obviously within the loser's reasonable contemplation as it called for the
security.

page "1254"

15.12.22. Costs of Contemporaneous Court Proceedings

It is not always clear whether a tribunal would seek to make an order over the
costs of ancillary court proceedings, such as applications for stays of
proceedings, anti-suit injunctions, interim measures, assistance in obtaining
evidence or the like. Technically speaking, many of these are not costs of an
arbitration but would instead normally be damages flowing from the arbitration
agreement itself where the court case is pursued in breach of the exclusive
arbitration agreement. (232) Conversely, where it is ancillary behaviour of the
successful party in supporting the arbitration, such as the costs of court-ordered
provisional measures, this may potentially be argued to be costs of the
arbitration. (233) However, Hanotiau suggests that such costs are generally not
accepted and should be claimed in the relevant procedures or as damages in
appropriate circumstances. (234) To the extent permissible, a tribunal should be
concerned to ensure that there is no duplication in the award of costs vis-à-vis
the court involved.

Concurrent court or arbitral proceedings may also be a reason to reduce fees on

the basis that the work was already done for the other matter. (235)

15.12.23. Costs in Relation to Settlement Negotiations

There is no consensus as to whether costs of mediation or other settlement

negotiations ought to be recoverable via the arbitration. (236) Once again, this
ought to be a matter for prior agreement of the parties. If a formal mediation is
held, it ought to indicate the cost basis upon which it is being undertaken. In
informal settlement negotiations, the issue can be complicated by the fact that
some work can indiscriminately benefit the settlement but also benefit the
ultimate proceedings if they ensue. When a tribunal renders an award following
a settlement agreement, it should turn its mind to the issue of costs. Article 40.3
of the UNCITRAL Rules 2010 directs it to do so.

page "1255"

15.12.24. Correction and Interpretation of Awards

It is arguable that in many cases, separate costs awards ought not to be made
against a losing party as a result of interpretation, correction or completion of
awards as this is likely to arise because of a failure of the tribunal itself. (237)
Nevertheless, rules such as Article 40.3 of the UNCITRAL Rules 2010 expressly
allow for limited costs awards in such circumstances but not tribunal fees. If the
application was ill-founded, there is no reason not to allow for costs. In some
cases, cost advances may have already been paid out, making recovery more
difficult.

15.12.25. Annulment and Enforcement

Costs relating to court challenges to the award or its enforcement would be dealt
with under the rules of the particular court. The tribunal itself is functus officio at
that stage. Such costs are costs of enforcement and not of the arbitration itself.
(238)

15.12.26. State Parties and Investment Arbitration

In many investment disputes where States are parties, there would need to be
separate consideration of governmental costs and those of private individuals.
(239)

It has been suggested that where there is arbitration between two sovereign
States, it is established practice to require each party to bear its own costs. (240)
That may often apply by way of express agreement between the sovereign States
as for example with the agreements establishing the Iran-US Claims Tribunals.
(241) Christoph Schreuer has suggested that ‘the practice of ICSID tribunals in

apportioning costs is neither clear nor uniform.’ (242) John Gotanda has suggested
that there be page "1256" default rules for both commercial and investment
arbitration. As to the former he suggests the default rule is for a rebuttable
presumption in favour of costs following the event. In investment disputes,
however, his suggested default rule is that parties should bear their own costs. In
addition to the economic argument against fee shifting generally, (which would
apply equally to both types of arbitration), he draws attention to a number of
factors in support of a differing rule in investment cases. First, because awards
against governments are borne by constituents, they may not readily be able to
bear the burden. From the investor's perspective, the risk of adverse costs
awards can be a disincentive to investment itself. (243) He also suggests that
parties have more opportunity to tailor their costs entitlements on a case-by-
case basis in commercial arbitration than in investment cases, which are often
circumscribed by provisions in a BIT, multilateral treaty, the ICSID Rules or the
ICSID Additional Facilities Rules. That may be so but there are no mandatory
rules as to costs and parties can certainly make choices, including selecting the
UNCITRAL Arbitration Rules to apply. He also alludes to the greater uncertainty
in investment arbitration both as to legal norms such as expropriation and
measurement of quantum of damages in such circumstances. Schill argues for a
pro-claimant rule, suggesting that there is a hierarchical relationship between
States and investors with the former being able to unilaterally bind the latter
through administrative orders or legislation, supported by international law and
principles of sovereignty. (244) However, presumptions as to relative size and
power of investors and States are unlikely to be uniformly applicable.

As noted above, empirical studies have shown an increasing trend towards loser
pays in investment arbitration. Some investment cases have argued strongly for
a harmonised approach between commercial investment arbitration. (245) Others
have followed a textual approach when applying the UNCITRAL Rules in their
earlier version, finding a presumption in Article 40.1 of the 1976 Rules without
the same presumption in Article 40.2, thus shifting arbitral expenses but not
representation costs. (246) Nevertheless, the key cases proposing costs following
the event contain strong dissents to the contrary, and the revised UNCITRAL
Rules do not mention the distinction. (247) Policy arguments against costs
following the event in page "1257" investment arbitration tend to be based on
certain presumptions, such as governments being more able to cover costs than
investors and arguments as to economic ramifications as to the promotion of
effective investment. Because the respective size between investor and host
State can vary significantly, any blanket presumptions are problematic.
Arguments as to the implications for investment per se are also problematic as
entitlements and uncertainties will all be factored into assessment of investment
value. Empirical analysis has also suggested that successful respondents rarely
receive costs awards.

15.12.27. Other

Other expenses might include typing fees, photocopying, facsimile and courier.

Interest is not awarded in relation to the pre-award historical aspect of cost

expenditures as costs only become due when awarded. (248) Post-award interest
might be sought where permitted by law and pre-award interest may apply if
costs are successfully claimed as damages.

15.13. Reasons and Costs Awards

It is a contentious question in arbitration as to the degree of reasoning to be

incorporated in an award, although absent agreement to the contrary, the better
view is that all awards should be adequately reasoned. Nevertheless, there is no
presumption or consensus that the same principles necessarily apply to cost
determinations. The UNCITRAL drafting history of the 1976 Rules suggested that
apart from arbitrators' fees, all other costs may be stated as a lump sum. (249) Best
practice would nevertheless suggest that adequate reasoning be provided
notwithstanding the lack of any stipulation in lex arbitri. The parties are entitled
to know the reasons for the outcome, particularly the losing party. Providing
reasons can also help in the educative function of published arbitral awards.
More elaborate reasoning might also be expected when draft awards are to be
submitted to institutions for their scrutiny.

page "1258"

15.14. Timing of Costs Awards: Interim Stage and Final Cost Awards

15.14.1. Deposits of Costs

There is a difference between the payment of the costs and the actual award in
relation to final allocation. Arbitral rules generally require the costs of the
tribunal and any institution to be provided upfront. (250) This is discussed in
section 6.11

15.14.2. Interim Costs Determinations

While costs awards are normally made as part of the final award, it is entirely
possible that costs decisions could be made at an earlier stage. Some tribunals
will reserve the question of costs on interim decisions. Some rules clearly allow
for interim or partial costs awards. (251) While tribunals might well consider the
imposition of interim costs awards to promote efficiency, it is commonly
expected that even such orders may be revised directly or indirectly when the
final costs award is made, as at that later stage, the tribunal can consider relative
success as well as procedural behaviour and can consider the relationship
between the two. Tribunals should also indicate clearly whether this is likely to
be a final ruling on the particular issue. For example a party who seeks excessive
discovery but is ultimately successful would be treated differently if the tribunal
wishes to indemnify the winner as the dominant factor or instead wishes to
promote fair and reasonable behaviour as to discovery of documents. (252)

If a tribunal wishes to be able to reconsider costs issues at a later stage, it may be

preferable to make an interim decision by way of a mere procedural order so
that the earlier determination might not be considered res judicata, although
that approach may undermine enforceability and would encourage rearguing of
costs perhaps unnecessarily. (253) Conversely, if the determination is to be
enforceable, the tribunal should register the cost decision as part of a partial
award. That might not always be easy as the notion of an award requires a final
disposition of all or part of a particular claim. Unless a broad view was taken to
suggest that the costs claim is itself a matter finally disposed of, an interim costs
award might need to relate to another dispositive determination. A problematic
area would be a costs award in relation to an application for a conservatory
measure. (254)

page "1259"

Where a tribunal intends to utilise interim costs awards to promote timely and
reasonable behaviour, it ought to make this very clear to the parties from the
outset. For example, if a party applied to change a hearing date at the eleventh
hour causing unfair loss to other parties, it could be entirely reasonable for the
tribunal to award wasted costs at that stage. The same can relate to preliminary
determinations as to jurisdiction. Another simple reason to award interim costs
is that parties have incurred the expenses. Cash-flow considerations would
demand prompt reimbursement where there is a basic entitlement for this.
Biihler suggests that if the tribunal wishes to make cost determinations on an
interim basis, this might be included in the terms of reference where applicable.
(255)

Where institutions are involved and fix the administrative costs and tribunal
fees, these will not be known by the tribunal at an interim stage and cannot
readily be allocated. Nevertheless, counsels' fees can still be allocated at that
stage in appropriate circumstances. (256)

15.14.3. Final Costs

Gerald Aksen criticises the tendency to reserve cost decisions until an award on
the merits. While this saves the need for the losing party to present an indication
of their costs, if both parties simultaneously submitted costs this would help in
identifying whether each is reasonable. (257) Court scrutiny by the ICC Court tends
to encourage one draft final award including costs issues. (258) There are
advantages and disadvantages either way. If submissions on costs are called for
prior to announcing the award on the merits, then in cases where each party has
some confidence of success, there are likely to be broad costs claims and less of a
difference between the parties. However, one party may be preparing and
submitting costs information needlessly. Conversely, once the final result is
known, parties are clearly opposed on costs applications, which may draw out
the proceedings and add further costs in arguing costs entitlements. However,
evidentiary material is only needed from the party entitled to costs. There is
even a need to consider distinct costs awards at that stage. For example, a
successful party who claims excessive costs, might find that there is a basic costs
order in their favour as to the merits, a reduction on reasonable grounds and a
costs order against them for the argument as to costs itself.

page "1260"

While the norm is to leave costs determinations until the final stages, as noted
above, arbitrators should ensure that this is not left so late as to allow an
argument that the tribunal has become functus officio prior to the costs
determination. This might occur if a tribunal simply forgot to deal with a
particular cost element in what it described as its final award, although most
rules allow for additional awards within strict time-limits. (259)

15.15. Enforcement of Costs Awards

One advantage of costs awards by tribunals rather than delegation to courts is

that the decisions of the latter may not readily be enforceable in an international
context. (260)

15.16. Review of Costs Awards

Generally speaking, costs decisions by tribunals will not be subject to annulment

or appeal. Where as in the UK, there can be limited appeals on errors of law, in
some circumstances at least it is conceivable that a tribunal might base the cost
decision on such an error. Limitations on the discretion can be impacted upon by
directions in the lex arbitri. For example, under the English Arbitration Act 1996,
because of the presumption that costs follow the event, a costs award has been
annulled where it improperly applied that criterion. (261) Annulment and
enforcement challenges could also proceed in any jurisdiction if the arbitrator
failed to follow due process in coming to the costs determination. For example an
aggrieved party ought to be able to challenge if the procedure of costs
determination did not allow an equal opportunity to be heard or took
inappropriate issues into account. Less clear is the situation where a tribunal
simply makes no decision as to costs. In turn that would depend on whether
costs were claimed by a party. In the latter event, a failure to rule on costs could
be said to be a failure to complete the mandate, allowing for challenges to that
extent. (262)

As noted, one aspect of uncertainty in such procedural challenges is whether

there is a need for similar reasons on a costs determination as there is for the
balance of the award. This was discussed in section 15.13 above. Because costs
orders are part of an award, there seems no reason to have a fundamentally
different page "1261" approach to the level of reasoning required as to other
elements of the award, particularly when in many cases the costs can be a very
significant part of the total compensation awarded.

15.17. Conclusion

It is unlikely that lex arbitri and arbitral rules will in the foreseeable future seek
to be more prescriptive about the nature of recoverable costs and the basis upon
which this is to be determined. The most that can be expected is increased
confirmation that loser pays is the starting presumption, that other factors such
as procedural behaviour can be taken into account and that reasonable limits
may also be appropriate. Because there are a range of potentially relevant factors
and because of the broad discretions left to a tribunal, the most desirable way
forward is for parties to give appropriate attention and directions and where
they cannot agree, for the tribunal to make its approach known at the earliest
possible stage in the arbitral process and where possible, attempt to develop a
costs protocol at an early stage.

page "1262"

1 The loser-pays principle, while associated with Commonwealth countries, can

be traced to Roman law. John Y. Gotanda, 'Awarding Costs and Attorneys' Fees in
International Commercial Arbitrations', Michigan Journal of International Law
21, no. 1 (1999): 5.
2 Historically, the civilian legal tradition was more inclined to award court and

procedural costs rather than reimbursement of legal fees. Even in litigation,

there now appears to be a strong trend towards greater reimbursement of legal
costs to successful parties in such jurisdictions. See Charles Price & Yves Stans,
'Using Costs as a Case Management Tool in International Arbitration', ASA
Bulletin 25, no. 4 (2007): 704. For a current database on various jurisdictional
approaches to costs see the work of the Oxford University Centre for Socio-Legal
Studies available at <www.csls.ox.ac.uk/CivilJusticeSystems.php>.
3 Christopher Newmark, ‘Controlling Time and Costs in Arbitration’, in The

Leading Arbitrators' Guide to International Arbitration, ed. Lawrence W. Newman

& Richard D. Hill (Huntington: Juris Pub, 2008), 81.
4 Richard A. Posner, Law and Legal Theory in England and America (New York:
Clarendon Press, 1996), 70-73. See also Avery Katz, 'Measuring the Demand for
Litigation: Is the English Rule Really Cheaper?', Journal of Law, Economics and
Organization 3, no. 2 (1987): 143. Katz looks at the financial incentives in terms
of the size and strength of claims that each costs rule tends to favour.
5 See, e.g., Stephan W. Schill, ‘Arbitration Risk and Effective Compliance: Cost-

Shifting in Investment Treaty Arbitration’, Journal of World Investment & Trade 7,

no. 5 (2006): 653.
6 There are no significant empirical studies although the Chartered Institute of

Arbitrators has recently conducted a survey amongst users. The results were not
available at the time of publication. Industry groups have also given particular
attention to concerns as to costs. An example was the establishment of the
Corporate Counsel International Arbitration Group. See <www. cciag.com/>.
7 Gotanda has made this observation in a range of detailed studies. John Y.

Gotanda, ‘Awarding Costs and Attorneys’ Fees in International Commercial

Arbitrations', Michigan Journal of International Law 21, no. 1 (1999): 1; John Y.
Gotanda, ‘Attorneys’ Fees Agonistes: The Implications of Inconsistency in the
Awarding of Fees and Costs in International Arbitration', The Villanova University
School of Law Working Paper Series, Paper 144, 2009,
<http://law.bepress.com/villanovalwps/paper/artl44>; John Y. Gotanda,
Supplemental Damages in Private International Law (The Hague: Kluwer Law
International, 1998). George H. Aldrich also noted that the three Chambers of the
Iran-US Claims Tribunal adopted different approaches with respect to the
awarding of costs of legal representation. George H. Aldrich, The Jurisprudence of
the Iran-United States Claims Tribunal (Oxford: Oxford University Press, 1996),
479.
8 Michael B ü hler, ‘Awarding Costs in International Commercial Arbitration: An

Overview’, ASA Bulletin 22, no. 2 (2004): 249.

9 Michael B ü hler, ‘Awarding Costs in International Commercial Arbitration: An

Overview’, ASA Bulletin 22, no. 2 (2004): 250. See also decision of an arbitral
tribunal seated in England operating under the LCIA Rules discussed in Jonathan
Wood, ‘Protection against Adverse Costs Awards in International Arbitration’,
Arbitration 74, no. 2 (2008): 139.
10 See, e.g., Bill's Coal Co. v. Board of Public Utilities, 887 F. 2d 242 (10th Cir

1989); Davis v. Prudential Sec. Inc. 59 F. 3d 1186 (11th Cir 1995). See also John Y.
Gotanda, ‘Awarding Costs and Attorneys’ Fees in International Commercial
Arbitrations', Michigan Journal of International Law 21, no. 1 (1999): 16. A
conflicts methodology that seeks to begin with characterisation as either
procedural or substantive could also lead to renvoi situations where, for example,
the arbitration takes place in a seat which considers costs to be a matter of
substantive law and the tribunal selects a substantive law that treats costs as a
procedural matter.
11 This is sometimes the case in regulatory legislation seeking to encourage

private litigants. See, e.g., the US Racketeer Influenced and Corrupt Organisations
Act 18 USC § 1964(c) (Sup page "11996") as applied in Triumph Painters Ltd v.
Kerr McGee Refining Corp, in Albert Jan van den Berg (ed.), Yearbook Book of
Commercial Arbitration, Volume XVIII (The Hague: Kluwer Law International,
1995), 120.
12 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law
International, 2009), 2495.
13 Michael Bühler, ‘Awarding Costs in International Commercial Arbitration: An

Overview’, ASA Bulletin 22, no. 2 (2004): 251.

14 Note the Arbitration Act BE 2545 Thailand, Chapter VIII. Fees, expenses and

remuneration s. 46 ‘Unless otherwise agreed by the parties, the fees and

expenses incurred in the arbitration proceedings, including the remuneration of
the arbitrator, but excluding attorney's fees and expenses, shall be in accordance
with those stipulated in the award of the arbitrator.’
15 The arbitration agreement might also have a different applicable law to the

balance of the contract, impacting upon applicable damages principles. See Marc
Blessing, ‘The Law Applicable to the Arbitration Clause’, in Improving the
Efficiency of Arbitration Agreements and Awards: 40 Years of Application of the
New York Convention, ICCA Congress Series No. 9, ed. Albert Jan van den Berg
(The Hague: Kluwer Law International, 1999), 168.
16 As opposed to one who wins via a counterclaim or set-off. See Michael Bühler,

‘Awarding Costs in International Commercial Arbitration: An Overview’, ASA

Bulletin 22, no. 2 (2004): 251.
17 In an extreme case, a respondent might try and argue that a spurious claim is

a breach of a good faith agreement to arbitrate or is a general misrepresentation.

18 I am indebted to Mark Kantor for this observation.
19 This provision is mandatory per s. 4(1) and Schedule 1 of the Act and thus

cannot be contracted out.

20 On one view this seems overly paternalistic given the law and economics

arguments for the American rule to similar effect discussed below.

21 ICC Rules 2012 Art. 37(5); SIAC Rules Art. 31.1; ICDR Rules Art. 31; ICSID

Convention Art. 61(2); Vienna Rules Art. 31.

22 UNCITRAL Rules 2010 Art. 42; LCIA Rules Art. 28.4.; HKIAC Rules Art. 36.4;

Swiss Rules 2012 Art. 40.1; ACICA Rules Art. 41.

23 See Howard M. Holtzmann & Joseph E. Neuhaus, A Guide to the UNCITRAL

Model Law on International Commercial Arbitration: Legislative History and

Commentary (Deventer: Kluwer Law and Taxation Publishers, 1989), 1118-1119.
See also Gary B. Born, International Commercial Arbitration (The Hague: Kluwer
Law International, 2009), 2489-2490.
24 See, e.g., Australian International Arbitration Act s. 27; Hong Kong Arbitration

Ordinance 2010 ss 57 and 74.

25 English Arbitration Act 1996 s. 61(2); German Code of Civil Procedure Art.

1057(1).
26 DIS Rules Art. 35.2; NAI Rules Art. 61(2); UNCITRAL Rules 2010 Art. 42; LCIA

Rules Art. 28.4; HKIAC Rules Art. 36.4; Swiss Rules 2012 Art. 40.1; ACICA Rules
Art. 41. See also SCC Rules Art. 43.
27 David Smith, ‘Shifting Sands: Cost-and-Fee Allocation in International
Investment Arbitration’, Virginia Journal of International Law 51, no. 3 (2011):
750.
28 Japanese Arbitration Law 2003 Art. 49(1).
29 Article 37(5).
30 English Arbitration Act 1996 s. 63(5)(b).
31 See UNCITRAL Rules 2010 Art. 40.2; ACICA Rules Art. 39; HKIAC Rules Art. 36.
32 David D. Caron, Matti Pellonpää & Lee M. Caplan, The UNCITRAL Arbitration

Rules: A Commentary, 2nd edn (Oxford: Oxford University Press, 2006), 948.
33 Himpurna California Energy Ltd and PT (Persero) Perusahaan Listruik Negara,
(Final Award), in Albert Jan van den Berg (ed.) Yearbook of Commercial
Arbitration, Volume XXV (The Hague: Kluwer Law International, 2000), 107.
34 Yves Derains & Eric A. Schwartz, A Guide to the ICC Rules of Arbitration, 2nd

edn (The Hague: Kluwer Law International, 2005), 370.

35 English Civil Procedure Rules, Parts 44 and 47.
36 Charles Price & Yves Stans, ‘Using Costs as a Case Management Tool in

International Arbitration’, ASA Bulletin 25, no. 4 (2007): 711.

37 Michael W. Biihler & Sigvard Jarvin, ‘ICC Arbitration Rules’, in Practitioner's

Handbook on International Commercial Arbitration, 1st edn, ed. F.B. Weigand &
E.H. Betts (Oxford: Oxford University Press, 2002), 297.
38 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2490; Michael Bühler, ‘Awarding Costs in International

Commercial Arbitration: An Overview’, ASA Bulletin 22, no. 2 (2004): 252-253
and 256, n. 42; Howard M. Holtzmann & Joseph E. Neuhaus, A Guide to the
UNCITRAL Model Law on International Commercial Arbitration: Legislative
History and Commentary (Deventer: Kluwer Law and Taxation Publishers, 1989),
1119.
39 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2498, citing M. Bühler & T. Webster, Handbook of ICC

Arbitration: Commentary, Precedents, Materials, 2nd edn (London: Sweet &
Maxwell, 2005), 361.
40 Marc Blessing, ‘The Law Applicable to the Arbitration Clause’, in Improving the

Efficiency of Arbitration Agreements and Awards: 40 Years of Application of the

New York Convention, ICCA Congress Series No. 9, ed. Albert Jan van den Berg
(The Hague: Kluwer Law International, 1999), 178.
41 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2491.

42 An example is England where the power to award costs was seen as entirely

based on statute. Michael O'Reilly, ‘Provisions on Costs and Appeals: An

Assessment from an International Perspective’, Arbitration 76, no. 4 (2010): 709-
720.
43 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2492-2493. See also ReliaStar Life Insurance Co. of New
York v. EMC National Life Co.., 546 F. 3d 81, US Court of Appeals for the Second
Circuit, 8 April 2009 where a majority panel of the US Court of Appeals for the
Second Circuit held that the inclusion in an arbitration agreement of a general
provision that each party will bear the expenses of its own arbitrator and
attorney fees does not deprive an arbitration panel of its inherent authority to
award such expenses and fees against a party who was determined to have failed
to arbitrate in good faith.
44 Clear party autonomy will prevail, in part because the American costs rule is

not considered to be part of US public policy in the event of challenge to

enforcement. See Gary B. Born, International Commercial Arbitration (The Hague:
Kluwer Law International, 2009), 2491.
45 Bruce Harris, Rowan Planterose & Jonathan Tecks, The Arbitration Act 1996: A

Commentary, 4th edn (Oxford: Blackwell Publishing, 2000), 276.

46 Lord Mustill & Stewart Boyd, Commercial Arbitration, 2nd edn (London:

LexisNexis Butter-worths, 2001), 39.

47 English Arbitration Act 1996 s. 63.
48 Section 21(3) indicates that a certificate signed by the Registrar on the
amount of costs or fees taxed shall form part of the award of the tribunal. Section
21 (4) indicates that the Chief Justice may, if he thinks fit, by notification
published in the Gazette, appoint any other person to exercise the powers of the
Registrar under that section.
49 UNCITRAL Rules 2010 Art. 41.4.
50 Phillip Yang, ‘The Organisation of International Arbitration Proceedings’, in

The Asian Leading Arbitrators' Guide to International Arbitration, ed. Michael

Pryles & Michael J Moser (Huntington: Juris Pub, 2007), 220.
51 See, e.g., ICC Rules 2012 Art. 37(1) and Appendix III; SCC Rules, Appendix III.
52 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2209-2502.

53 See ICC Case No. 3383 (1979), in Pieter Sanders (éd.), Yearbook of Commercial

Arbitration Volume TO (The Hague: Kluwer Law International, 1982), 123; ICC
Case No. 5896 (1992), ICC Court of Arbitration Bulletin 4 (1993): 37; SCC Case No.
129/2000 (2002), Stockholm Arbitration Report (2003), 127.
54 Michael Biihler, ‘Awarding Costs in International Commercial Arbitration: An

Overview’, ASA Bulletin 22, no. 2 (2004): 258.

55 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2501. See also S. Kröll, ‘Recourse against Negative Decisions
on Jurisdiction’, Arbitration International 20 (2004): 55.
56 Franz T. Schwarz & Christian W. Konrad, The Vienna Rules: A Commentary on

International Arbitration in Austria, (The Hague: Kluwer Law International,

2009), 717.
57 Ibid., 718.
58 See, e.g., NAI Rules Art. 52(3).
59 Michael Bühler, ‘Awarding Costs in International Commercial Arbitration: An

Overview’, ASA Bulletin 22, no. 2 (2004): 257-258.

60 See, e.g., ICC Case No. 10574 (2000), Mealey's International Arbitration Reports

16 (2001): 62.
61 Final Award in ICC Case No. 6320, in Albert Jan van den Berg (ed.), Yearbook of

Commercial Arbitration Volume XX (The Hague: Kluwer Law International, 1995),

108-109.
62 Cases that have upheld cost agreements by the parties include ICC Final

Award No. 3572 (1982), in Albert Jan van den Berg (ed.), Yearbook of Commercial
Arbitration Vol XIV (The Hague: Kluwer Law International, 1989), 111; ICC Final
Award No. 6320 (1992), in Albert Jan van den Berg (ed.), Yearbook of Commercial
Arbitration Volume XX (The Hague: Kluwer Law International, 1995), 62; ICC
Final Awards Nos 7385 & 7402 (1992), in Albert Jan van den Berg (ed.),
Yearbook of Commercial Arbitration Volume XVIII (The Hague: Kluwer Law
International, 1993), 68; Final Award of 17 November 1994 (ad hoc), in Albert
Jan van den Berg (ed.), Yearbook of Commercial Arbitration Volume XXI (The
Hague: Kluwer Law International, 1996), 13. See also Award of 27 May 1991 (ad
hoc), in Albert Jan van den Berg (ed.), Yearbook of Commercial Arbitration
Volume XVII (The Hague: Kluwer Law International, 1992), 11. The latter case
enforced an agreement to bear all costs and fees equally.
63 Section 46 Arbitration Act BE 2545.
64 John Y. Gotanda, ‘Awarding Interest in International Commercial Arbitration’,
American Journal of International Law 90 (1996): 57.
65 John Y. Gotanda, ‘Awarding Costs and Attorneys’ Fees in International
Commercial Arbitrations', Michigan Journal of International Law 21, no. 1 (1999):
30.
66 For example in Transvenezualian Shipping Co. SA v. Czarnikow-Rionda Co. Inc.

(1982) AMC 1458, 1460 (SDNY 1981) the US District Court held that a clause
allowing for the award of the ‘expenses and costs of arbitration’ did not cover
attorneys' fees and legal expenses.
67 See Murray L. Smith, ‘Costs in International Commercial Arbitration’, Dispute

Resolution Journal 56, no. 1 (2001): 33-34.

68 Michael Bühler, ‘Awarding Costs in International Commercial Arbitration: An

Overview’, ASA Bulletin 22, no. 2 (2004): 257.

69 John Y. Gotanda, ‘Awarding Costs and Attorneys’ Fees in International
Commercial Arbitrations', Michigan Journal of International Law 21, no. 1 (1999):
49.
70 Michael Bühler, ‘Awarding Costs in International Arbitration: An Overview’,

ASA Bulletin 22, no. 2 (2004): 256. There is a slight confusion in the terminology
in the UK legislation with s. 61(1) speaking of what the tribunal may do while s.
61(2) indicates what it shall do in certain circumstances. Sheridan argues that
they must so decide. Peter Sheridan, Construction and Engineering Arbitration
(London: Sweet & Maxwell, 1999), 430.
71 See, e.g., Australian International Arbitration Act s. 27. For the position in New

Zealand, see Casata Ltd v. General Distributors Ltd [2006] NZSC 8.

72 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 685.
73 Michael Bühler, ‘Awarding Costs in International Commercial Arbitration: An

Overview’, ASA Bulletin 22, no. 2 (2004): 255.

74 Previously the situation in England was that arbitrators were required to act

judicially in dealing with costs but this is now questioned. See, e.g., SKI pic v.
Metro-Cammell Hong Kong Ltd (1996) CILL 1162; Fence Gate v. NEL Construction
Ltd [2002] 82 Con LR 41 per Thornton J, para. 33; Hew R. Dundas, ‘Recent
Developments Regarding Costs in Litigation: Are They Applicable in
Arbitration?’, Arbitration 69 (2003): 90.
75 See Newfleld Construction Ltd v. John Lawton Tomlinson [2004] EWHC 3051

(TCC) for an example of error found based on a tribunal's cost reasoning.

76 Christopher Newmark, ‘Controlling Time and Costs in Arbitration’, in The

Leading Arbitrators' Guide to International Arbitration, ed. Lawrence W. Newman

& Richard D. Hill (Huntington: JurisPub, 2008), 82.
77 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 548.

78 ICC, ‘Techniques for Controlling Time and Costs in Arbitration’,
<www.iccwbo.org/uploaded-Files/TimeCost_E.pdf>, 2007, para. 85.
79 Ibid.
80 Michael Bühler, ‘Costs of Arbitration: Some Further Considerations’, in Global

Reflections on International Law, Commerce and Dispute Resolution: Liber

Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al. (Paris: ICC
Publishing, 2005), 183.
81 Peter Sheridan, Construction and Engineering Arbitration (London: Sweet &
Maxwell, 1999), 436.
82 Ronald Bernstein, ‘Some Thoughts on When and How an Arbitrator Should

Use the Power to Cap the Recoverable Costs of the Arbitration, or of Any Part of
the Arbitral Proceedings, Given to Him by Section 65 of the Act of 1996’,
International Arbitration Law Review 1, no. 3 (1998): 51.
83 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 546. There are numerous arbitral cases
applying the loser pays principle. See, e.g., ICC Case No. 8486 (1996), in Albert
Jan van den Berg (éd.), Yearbook of Commercial Arbitration Volume XXIV (The
Hague: KluwerLaw International, 1999), 171; ICC Case No. 7645 (1995), in
Albert Jan van den Berg (ed.) Yearbook of Commercial Arbitration Volume XXVI
(The Hague: Kluwer Law International, 2011), 151; ICC Case No. 8528 (1996), in
Albert Jan van den Berg (ed.), Yearbook of Commercial Arbitration Volume XXV
(The Hague: Kluwer Law International, 2000), 352.
84 See, e.g., Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative

International Commercial Arbitration (The Hague: Kluwer Law International,

2003), 655.
85 Richard H. Kreindler, ‘Final Rulings on Costs: Loser Pays All?’, ASA Special

Series 26 (2006): 41.

86 Susan D Franck, ‘Empirically Evaluating Claims about Investment Treaty

Arbitration’, North Carolina Law Review 86 (2007): 1; Susan D. Franck,

‘Rationalizing Cost Awards in Investment Treaty Arbitration’, Washington
University Law Review 88, no. 4 (2011).
87 UNIDROIT, ‘Draft Principles of Transnational Civil Procedures, Study LXXVI,

Doc 11', <www.unidroit.org/english/documents/2004/study76/s-76-ll-e.pdf>,

February 2004. Gotanda also identifies the following countries that employ the
costs follow the event principle: Australia, Austria, Brazil, Canada, Chile,
Colombia, Costa Rica, Denmark, Dominican Republic, France, Germany, Greece,
Hungary, India, Italy, Iran, Luxembourg, Mexico, The Netherlands, New Zealand,
Portugal, Romania, Switzerland, Turkey and Yemen.
88 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2495.

89 J.Y. Gotanda, ‘Bringing Efficiency to the Awarding of Fees and Costs’, in Liber

Amicorum Eric Bergsten. International Arbitration and International Commercial

Law: Synergy, Convergence and Evolution, ed. S Kröll et al. (Alphen aan den Rijn:
Walder Wyss, 2011), 147.
90 Waguih Elie, George Siag and Clorinda Vecchi v. The Arab Republic of Egypt,

ICSID Case No. ARB/05/15 (Italy/Egypt BIT), Award and Dissenting Opinion (1
June 2009), available at
<http://ita.law.uvic.ca/documentsAVaguihElieGeorgeS...> para. 617.
91 See, e.g., EDF (Services) Ltd v. Romania. ICSID Case No. ARB/05/13, Award (8

October 2009), para. 327.

92 See Susan D. Franck, ‘Empirically Evaluating Claims about Investment Treaty

Arbitration’, North Carolina Law Review 86 (2007): 1; Susan D Franck,

‘Rationalizing Cost Awards in Investment Treaty Arbitration’, Washington
University Law Review 88, no. 4 (2011).
93 David Smith, ‘Shifting Sands: Cost and Fee Allocation in International
Investment Arbitration’, Virginia Journal of International Law 51, no. 3 (2011):
749.
94 An earlier study by Noah Rubins in 2003 had also found little in the way of
applications of costs following the event. Noah Rubins, ‘The Allocation of Costs
and Attorneys Fees in Investor-State Arbitrations’, ICSID Review 18 (2003): 109.
95 Gotanda notes numerous US federal statutes allowing for attorneys' fees

awards, fee shifting for malicious claims and party agreement to fee awards. See
John Yukio Gotanda, ‘Awarding Costs and Attorney's Fees in International
Commercial Arbitrations’, Michigan Journal of International Law 21, no. 1 (1999):
13.
96 John Y. Gotanda, ‘Awarding Costs and Attorneys’ Fees in International
Commercial Arbitrations', Michigan Journal of International Law 21, no. 1 (1999):
5.
97 Michael O'Reilly, ‘Rethinking costs in commercial arbitration’, Arbitration 69,

no. 2 (2003): 122.

98 There is a separate question as to whether there is a power to award costs to

the respondent who may succeed in showing that there is no arbitral

jurisdiction. This was discussed in section 8.1.
99 American commentators have argued that the development of the rule may

have a more parochial gestation, being intended to reduce lawyers' costs through
general distrust of the profession (e.g., Ronald Braeutigam et al., ‘An Economic
Analysis of Alternative Fee Shifting Systems’, Law & Contemporary Problems 47
(1984): 173, 174) or that it flowed from general deregulation of lawyers' fees
after the American Revolution where excessive fees worked well if these were
recovered from the winning party out of moneys received (John Leubstors,
‘Toward a History of the American Rule on Attorney Fee Recovery’, 47 Law &
Contemporary Problems 9 (1984).
100 John Y. Gotanda, ‘Awarding Costs and Attorneys’ Fees in International
Commercial Arbitrations', Michigan Journal of International Law 21, no. 1 (1999):
11, citing Arcambel v. Wiseman 3 US 306 (1796) and Fleischman Distilling Corp. v.
Maier Brewing Co., 386 US 714 (1967).
101 Murray L. Smith, ‘Costs in International Commercial Arbitration’, Dispute

Resolution Journal 56, no. 1 (2001): 31-32.

102 Michael Bühler, ‘Awarding Costs in International Commercial Arbitration: An

Overview’, ASA Bulletin 22, no. 2 (2004): 268.

103 Clinton S. Becker III & Avery Katz, ‘The Incentive Effects of Litigation Fee

Shifting when Legal Standards Are Uncertain’, International Review of Law &
Economy 15 (1995): 205-206.
104 John Y. Gotanda, ‘Awarding Costs and Attorneys’ Fees in International
Commercial Arbitrations', Michigan Journal of International Law 21, no. 1 (1999):
37 and articles cited in fn. 172.
105 Peter Sheridan, Construction and Engineering Arbitration (London: Sweet &

Maxwell, 1999), 432.

106 David P. Riesenberg, ‘Fee Shifting in Investor-State Arbitration: Doctrine and

Policy Justifying Application of the English Rule’, Duke LJ 60 (2010-2011): 977 at

1000; Werner Pfennig-storf, ‘The European Experience with Attorneys Fee
Shifting’, Law & Contemporary Problems 47 (1984): 37.
107 See, e.g., In re Elgindata (No 2) [1992] 1 WLR 1207 CA.
108 Even the arbitration costs can be significantly affected. Neil Kaplan uses the

example of a successful claim for USD 1,000,000 with a successful counterclaim

for USD 900,000. Is the net balance of USD 100,000 a true win? See Neil Kaplan,
‘Problems at Both Ends’, in Liber Amicorum Eric Bergsten. International
Arbitration and International Commercial Law: Synergy, Convergence and
Evolution, ed. S Kröll et al. (Alphen aan den Rijn: Walder Wyss, 2011). An
institution would consider that the combined claims are USD 1.9 million and
assess these accordingly. The net figure is simply USD 100,000 that ought to have
been paid from the outset.
109 However, Bühler questions whether procedural and parties' costs should be

treated differently. Michael Bühler, ‘Awarding Costs in International Commercial

Arbitration: An Overview’, ASA Bulletin 22, no. 2 (2004): 261.
110 See Study reported in Eric A Schwarz, ‘The ICC Arbitral Process-Part IV: The

Costs of ICC Arbitration’, ICC International Court of Arbitration Bulletin 4 (1993):

21.
111 See J.Y. Gotanda, ‘Attorneys’ Fees Agonistes: The Implications of
Inconsistency in the Awarding of Fees and Costs in International Arbitration',
The Villanova University School of Law Working Paper Series, Paper 144, 2009),
11.
112 Franz T. Schwarz & Christian W. Konrad, The Vienna Rules: A Commentary on

International Arbitration in Austria (The Hague: Kluwer Law International,

2009), 710.
113 See ICC Final Award No. 5759 (1989), in Albert Jan van den Berg (ed.),

Yearbook of Commercial Arbitration XVIII (The Hague: Kluwer Law International,

1993), 34. In that case, the tribunal awarded 75% of costs because claimant
prevailed on 75% of the claims raised. See also ICC Final Award No. 8786 (1997),
ASA Bulletin 20, no. 1 (2002): 67. That case saw a successful respondent awarded
only 80% of legal costs because the tribunal had dismissed an application for an
interim award on security. A similar result arose in ICC Case No. 5901 (1992),
ICC International Court of Arbitration Bulletin 4 (1993): 40, where the successful
respondent lost certain initial procedural defences. See also ICC Case No. 10274
(2003), in Albert Jan van den Berg (ed.), Yearbook of Commercial Arbitration
Volume XXIX (The Hague: Kluwer Law International, 2004), 107, where it was
stated that ‘in its decision over the allocation of costs the tribunal is guided by
the win/loss proportion in the case at hand’.
114 See, e.g., ICC Award No. 3493 (1983), in Pieter Sanders (ed.), Yearbook of

Commercial Arbitration Volume IX (The Hague: Kluwer Law International, 1984),

111.
115 John Y. Gotanda, ‘Awarding Costs and Attorneys’ Fees in International
Commercial Arbitrations', Michigan Journal of International Law 21, no. 1 (1999):
39, referring to the Welamson Doctrine further outlined in J. Gillis Wetter & Charl
Priem, ‘Costs and Their Allocation in International Commercial Arbitrations’,
American Review of International Arbitration 2 (1991): 273-274.
116 Michael Bühler, ‘Awarding Costs in International Commercial Arbitration: An

Overview’, ASA Bulletin 22, no. 2 (2004): 261, n. 78.

117 ICC Final Award No. 7047 (1994), in Albert Jan van den Berg (ed.), Yearbook

of Commercial Arbitration Volume XXI (The Hague: Kluwer Law International,

1996), 79. Judge Holtzmann offered suggestions about the approach to an issue-
by-issue basis in his separate opinion in the Sylvania case. See Sylvania Technical
Systems, Inc. et al. and the Government of the Islamic Republic of Iran (1985) 8
Iran-US Claims Tribunal Reports 329.
118 Michael Bühler, ‘Awarding Costs in International Commercial Arbitration: An
Overview’, ASA Bulletin 22, no. 2 (2004): 268.
119 M. Behler & T. Webster, Handbook of ICC Arbitration: Commentary,
Precedents, Materials, 2nd edn (London: Sweet & Maxwell, 2005), 188.
120 See, e.g., Final Award in ICC Case No. 8032 of 1995, in Albert Jan van den Berg

(ed.), Yearbook of Commercial Arbitration Volume XXI (The Hague: Kluwer Law
International, 1996), 121. Craig Park Paulsson also suggest that costs claims
should be subject to proof like any proof of damage. W. Laurence Craig, William
W. Park & Jan Paulsson, International Chamber of Commerce Arbitration, 3rd edn
(Dobbs Ferry: Oceana, 2000).
121 See, e.g., ICC Case No. 6673 (1992), ICC International Court of Arbitration

Bulletin 4 (1993): 47.

122 See, e.g., ICC Case No. 8332 (1996, unpublished) cited in Michael Bühler,

‘Awarding Costs in International Commercial Arbitration: An Overview’, ASA

Bulletin 22, no. 2 (2004): 266.
123 See, e.g., ICC Case No. 8486 (1996), in Albert Jan van den Berg (ed.), Yearbook

of Commercial Arbitration Volume XXIVa (The Hague: Kluwer Law International,

1999), 172. See also Liberiam Eastern Timber Corp v. Liberia, ICSID Case No.
ARB/83/2 (31 March 1986); Repsol YPF Equador SA & Empresa Estatal Petróleos
del Equador v. Republic of Equador, ICSID Case No. ARB/01/10, Decision on
Annulment (8 January 2007).
124 Near East Technological Services USA, Inc. v. Islamic Republic of Iran Air Force

Logistics and Command, Case No. 845, Award No. 406-845-1 of 16 December
1988.
125 See, e.g., Final Award in ICC Case No. 5731 of 1989, ICC International Court of

Arbitration Bulletin 3, no. 1 (1992): 18. Final Award in ICC Case No. 6959 of 1992,
ICC International Court of Arbitration Bulletin 4, no. 1 (1993): 48.
126 For a case where continual delaying tactics led to an entire costs award

against a defendant who successfully precluded some claims, see Award in ICC
Case No. 8486 of 1996, in Albert Jan van den Berg (ed.) Yearbook of Commercial
Arbitration Volume XXIVa (The Hague: Kluwer Law International, 1999), 162.
127 See, e.g., ICC Final Award No. 6527 (1991), in Albert Jan van den Berg (éd.),

Yearbook of Commercial Arbitration Volume XVIII (The Hague: Kluwer Law

International, 1993), 44. An older English case considered that reducing costs
because a claimant recovered significantly less than claimed was an error in
principle. Modern courts are less likely to interfere on this basis. See Channel
Island Ferries Ltd v. Cenar go Navigation Ltd [1994] 2 Lloyd's Reports 161 per
Phillips J at 170.
128 See, e.g., English Arbitration Act 1996 s. 53; UNCITRAL Rules 2010 Art. 40;

ICC Rules 2012 Art. 37; HKIAC Rules Art. 36.1; SCC Rules, Art. 44; Swiss Rules
2012 Art. 38; ACICA Rules Art. 39. See also Sylvania Technical SYS Inc v.
Government of the Islamic Republic of Iran (1985) 8 Iran-US Claims Tribunal
Reports 329.
129 John Y Gotanda, ‘Awarding Costs and Attorneys’ Fees in International
Commercial Arbitrations', Michigan Journal of International Law 21, no. 1 (1999):
41.
130 See, e.g., ICC Case No. 5726 (1992), ICC International Court of Arbitration

Bulletin 4 (1993): 35.

131 See, e.g., ICC Final Award No. 7006 (1992), in Albert Jan van den Berg (ed.),

Yearbook of Commercial Arbitration Volume XVIII (The Hague: Kluwer Law

International, 1993), 58.
132 See ICC Case No. 6914 (1992) ICC International Court of Arbitration Bulletin 4

(1993): 48. In that case, a costs order was reduced against an unsuccessful
claimant because of the respondent's failed attempt to object as to claimant's
capacity. Other examples of unnecessary failures to admit might include
information about corporate legal status and title to property.
133 See, e.g., ICC Case No. 6955 (1993), in Albert Jan van den Berg (éd.), Yearbook

of Commercial Arbitration Volume XXIV (The Hague: Kluwer Law International,

1999), 139.
134 Calderbank v. Calderbank [1975] 3 All ER 333. For domestic English litigation

such a procedure is now enshrined in Part 36 of the English Civil Procedure

Rules. It has been suggested that this procedure is also commonly used in
English domestic arbitration. See Bruce Harris,

Rowan Planterose & Jonathan Tecks, The Arbitration Act 1996: A Commentary,
4th edn (Oxford: Blackwell Publishing, 2000), para. 61G. As to sealed offers in
English arbitration see Donaldson J in Tramountana Armadora SA v. Atlantic
Shipping Co SA [1978] 1 Lloyd's Re page "391", 396.

135 See also Jonathan Wood, ‘Protection Against Adverse Costs’ Awards in
International Arbitration', Arbitration 74, no. 2 (2008): 142.
136 For a more critical perspective as to the factual findings see Jonathan Wood,

‘Protection against Adverse Costs' Awards in International Arbitration’,

Arbitration 74, no. 2 (2008): 139.
137 Michael Bühler, ‘Awarding Costs in International Commercial Arbitration: An

Overview’, ASA Bulletin 22, no. 2 (2004): 263.

138 SKI pic v. Metro-Cammell Hong Kong Ltd (1996) CILL 1162.
139 Michael O'Reilly, ‘Provisions on Costs and Appeals: An Assessment from an

International Perspective’, Arbitration 76 (2010): 710.

140 See, e.g., ICC Final Award No. 4629 (1989), Albert Jan van den Berg (ed.),

Yearbook of Commercial Arbitration Volume XVIII (The Hague: Kluwer Law

International, 1993), 11.
141 See, e.g., ICC Case No. 7453 of 1994, in Albert Jan van den Berg (ed.),

Yearbook of Commercial Arbitration (The Hague: Kluwer Law International,

1997); ICC Case No. 8486 (1996), Yearbook of Commercial Arbitration Volume
XXIV (The Hague: Kluwer Law International, 1999), 172.
142 Plama Consortium Ltd v. Republic of Bulgaria, ICSID Case No. ARB/03/24,

Award 27 August 2008 paras 321-323 available at

<http://ita.law.uvic.ca/documents/plamabulgariaawar...>; Europe Cement
Investment and Trade SA v. Turkey, ICSID Case No. ARB(AF)/07/2, Award 13
August 2009 available at
<http://ita.law.uvic.ca/documents/EuropeCementAward...>; Cementownia ‘nowa
huta’ SA v. Republic of Turkey, ICSID Case No. ARB(AF)/06/2, 17 September
2009.
143 See, e.g., ICC Case No. 6248 (1990), in Albert Jan van den Berg (ed.), Yearbook

of Commercial Arbitration Volume XIX (The Hague: Kluwer Law International,

1994), 124.
144 John Y Gotanda, ‘Awarding Costs and Attorneys' Fees in International
Commercial Arbitrations’, Michigan Journal of International Law 21, no. 1 (1999):
42.
145 Bernaidus Henricus Funnekatter and Others v. Republic of Zimbabwe, ICSID

Case No. ARB/ 05/6 (Netherlands/Zimbabwe), Award 22 April 2009, available at

<http://ita.law.uvic.ca/documents/ZimbabweAward.pdf...> Another case where
a tribunal chose not to award legal costs in part because of economic and
political circumstances in a developing country respondent was Himpurna v.
Indonesia, Final Award, in Albert Jan van den Berg (éd.), Yearbook of Commercial
Arbitration Volume XXV (The Hague: Kluwer Law International, 2000), 106.
146 Bernard Hanotiau, ‘The Parties' Costs of Arbitration’, in Evaluation of
Damages in International Arbitration, Dossier of the ICC Institute of World
Business Law, ed. Yves Derains & Richard H. Kreindler (Paris: ICC Publishing,
2006), 221.
147 George H. Aldrich, The Jurisprudence of the Iran-United States Claims Tribunal

(Oxford: Oxford University Press, 1996), 479.

148 Franz T. Schwarz & Christian W. Konrad, The Vienna Rules: A Commentary on

International Arbitration in Austria (The Hague: Kluwer Law International,

2009), 703.
149 CIETAC Rules Art. 50.
150 W. Laurence Craig, William W. Park & Jan Paulsson, International Chamber of

Commerce Arbitration, 3rd edn (Dobbs Ferry: Oceana, 2000), 395.

151 Charles Price & Yves Stans, ‘Using Costs as a Case Management Tool in

International Arbitration’, ASA Bulletin 25, no. 4 (2007): 714.

152 Ronald Bernstein, ‘Some Increased Responsibilities of Arbitrators and of

Those Who Appoint Them’, International Arbitration Law Review 1, no. 2 (1998):
52.
153 Daniel Wehrli, ‘Contingency Fees/Pactum de Talnario “Civil Law Approach”’,

ASA Bulletin 26, no. 2 (2008): 252.

154 However, some recent investment arbitrations have made use of early
procedures. ICSID Rules Art. 41.5, for example, permits parties to apply at an
early stage for a ruling that a claim is clearly without legal merit. See Global
Trading Resource Corp. v. Ukraine, ICSID Case No. ARB/09/11 (1 December
2010).
155 Sylvania Technical SYS Inc v. Government of the Islamic Republic of Iran

(1985) 8 Iran-US Claims Tribunal Reports 329. Judge Holtzmann's test has been
commonly cited in subsequent arbitration cases. See, e.g., Waguih Elie George
Siag and Clorinda Vecchi v. Egypt, ICSID Case No. ARB/05/15 (1 June 2009); ADC
Affiliate Ltd, ADC & ADMC Management Ltd v. The Republic of Hungary, ICSID Case
No. ARB/03/16 (2 October 2006).
156 Michael W. Bühler & Sigvard Jarvin, ‘ICC Arbitration Rules’, in Practitioners'

Handbook on International Arbitration, ed. S.B. Weigand & E.H. Betts (Oxford:
Oxford University Press, 2002), 299. See also Neil Kaplan, ‘Problems at Both
Ends’, in Liber Amicorum Eric Bergsten. International Arbitration and
International Commercial Law: Synergy, Convergence and Evolution, ed. S. Rröll et
al. (Alphen aan den Rijn: Walder Wyss, 2011).
157 Sylvania Technical SYS Inc v. Government of the Islamic Republic of Iran (1985)

8 Iran-US Claims Tribunal Reports, 323-324.

158 Ibid., 329.
159 See Chartered Institute of Arbitrators, ‘Guidelines for Arbitrators on Making
Orders Relating to the Costs of the Arbitration’, <www.ciarb.org/information-
and-resources/PracticeGuide-line9.pdf>.
160 Philip Yang, ‘The Organisation of International Arbitration Proceedings’, in

Asian Leading Arbitrators' Guide to International Arbitration, ed. Michael Pryles &
Michael Moser (Huntington: JurisNet, 2007), 222.
161 Daniel Wehrli, ‘Contingency Fees/Pactum de Talnario “Civil Law Approach”’,

ASA Bulletin 26, no. 2 (2008): 254.

162 W. Laurence Craig, William W. Park & Jan Paulsson, International Chamber of

Commerce Arbitration, 3rd edn (Dobbs Ferry: Oceana, 2000), 394.

163 See, e.g., ICC Final Award No. 8032 (1995), in Albert Jan van den Berg (éd.),

Yearbook of Commercial Arbitration Volume XXI (The Hague: KluwerLaw

International, 1996), 121.
164 This is a view taken by a Singaporean court in W v. VW [2008] SGHC 11, [53].
165 See, e.g., Neil Kaplan, ‘Problems at Both Ends’, in Liber Amicorum Eric
Bergsten. International Arbitration and International Commercial Law: Synergy,
Convergence and Evolution, ed. S. Kröll et al. (Alphen aan den Rijn: Walder Wyss,
2011), 284.
166 Bernard Hanotiau, ‘The Parties' Costs of Arbitration’, in Evaluation of
Damages in International Arbitration, Dossier of the ICC Institute of World
Business Law, ed. Yves Derains & Richard H. Kreindler (Paris: ICC Publishing,
2006), 218.
167 See, e.g., Australian International Arbitration Act 1974 s. 27.
168 As to this, see section 16.16.
169 Michael Bühler, ‘Awarding Costs in International Commercial Arbitration: An

Overview’, ASA Bulletin 22, no. 2 (2004): 249.

170 See UNCITRAL Rules 2010 Art. 40; HKIAC Rules Art. 36.1; ACICA Rules Art.

39. See also David D. Caron, Matti Pellopää & Lee M. Caplan, The UNCITRAL
Arbitration Rules: A Commentary, 2nd edn (Oxford: Oxford University Press,
2006), 932.
171 Attempts to collect prior to issuing proceedings were denied in Grommet

State Machine Company Co, Inc. and the Islamic Republic of Iran et al. (1981-
1982) Iran-US Claims Tribunal Reports 447.
172 Bernard Hanotiau, ‘The Parties' Costs of Arbitration’, in Evaluation of
Damages in International Arbitration, Dossier of the ICC Institute of World
Business Law, ed. Yves Derains & Richard H. Kreindler (Paris: ICC Publishing,
2006), 214. See also ICC Case No. 5896 (1992), ICC Court of Arbitration Bulletin, 4
(1993).
173 See, e.g., the views in ICC Case No. 5896 (1992), ICC International Court of

Arbitration Bulletin 4 (1993): 38.

174 See ICC, ‘Techniques for Controlling Time and Costs in Arbitration’,
<www.iccwbo.org/ uploadedFiles/TimeCost_E.pdf>, 2007. Klaus Sachs'
impressionistic views from his own arbitral experience were that legal costs
accounted for between 81% and 94% of total arbitration costs. See Klaus Sachs,
‘Time and Money’, in Pervasive Problems in International Arbitration, ed. Loukas
A. Mistelis & Julian D.M. Lew (The Hague: Kluwer Law International, 2006), 103.
175 For example, Thailand Arbitration Act s. 46.
176 Susan D. Franck, ‘Empirically Evaluating Claims about Investment Treaty

Arbitration’, North Carolina Law Review 86 (2007): 67, which found that legal
fees were only awarded in 13 of 54 final awards. See also footnotes 92-94 above
and related text.
177 Note, however, that s. 27 of the Australian International Arbitration Act
expressly empowers a tribunal to consider such principles.
178 It has been observed that some Canadian provincial courts award something

in the order of 30%-50% of legal costs to the winner while in England it tends to
be closer to 70%-80%. Murray L. Smith, ‘Costs in International Commercial
Arbitration’, Dispute Resolution Journal 56, no. 1 (2001): 31.
179 See, e.g., ICC Case No. 5008 (1992), ICC International Court of Arbitration

Bulletin, 4 (1993): 31.

180 See, e.g., ICC Case No. 5726 (1992), ICC International Court of Arbitration

Bulletin, 4 (1993): 36.

181 Such arrangements are sometimes described as success fees, conditional fees

or alternative fee arrangements. These are also known as ‘quota litis pactum de
talnario’ or the German streitanteil or erfolgshonorar. See Daniel Wehrli,
‘Contingency Fees/Pactum de Talnario “Civil Law Approach’”, ASA Bulletin 26,
no. 2 (2008): 241. Contingency fee arrangements have been accepted in the
United Kingdom in Thai Trading Co. v. Taylor [1998] 2 WLR 893 CA.
182 Michael Bühler, ‘Awarding Costs in International Commercial Arbitration: An

Overview’, ASA Bulletin 22, no. 2 (2004): 249, referring to UNIDROIT's Draft
Principles and Rules of Transnational Civil Procedures.
183 Daniel Wehrli, ‘Contingency Fees/Pactum de Talnario “Civil Law Approach”’,

ASA Bulletin 26, no. 2 (2008): 250.

184 Franz T. Schwarz & Christian W. Konrad, The Vienna Rules: A Commentary on

International Arbitration in Austria (The Hague: Kluwer Law International,

2009), 709.
185 For a review as to which civilian legal systems allow for success fee
arrangements see Daniel Wehrli, ‘Contingency Fees/Pactum de Talnario “Civil
Law Approach”’, ASA Bulletin 26, no. 2 (2008): 241.
186 Michael Bühler, ‘Awarding Costs in International Commercial Arbitration: An

Overview’, ASA Bulletin 22, no. 2 (2004), 271, n. 122.

187 Michael O'Reilly, Costs in Arbitration Proceedings (London: Lloyd's of London

Press, 1995), 67.

188 Daniel Wehrli, ‘Contingency Fees/Pactum de Talnario “Civil Law Approach”’,

ASA Bulletin 26, no. 2 (2008): 252.

189 The situation might be different if a party could show that they had to engage

in such an arrangement as the breach caused them severe cash-flow problems

such that they could not fund their own arbitration, but that would be a matter of
proof and causation under the applicable law.
190 Bernard Hanotiau, ‘The Parties' Costs of Arbitration’, in Evaluation of
Damages in International Arbitration, Dossier of the ICC Institute of World
Business Law, ed. Yves Derains & Richard H. Kreindler (Paris: ICC Publishing,
2006), 219.
191 Daniel Wehrli, ‘Contingency Fees/Pactum de Talnario “Civil Law Approach”’,

ASA Bulletin 26, no. 2 (2008): 254. Such duties are discussed in section 2.7.11.
192 Christopher Newmark, ‘Controlling Time and Costs in Arbitration’, in The

Leading Arbitrators' Guide to International Arbitration, ed. Lawrence W. Newman

& Richard D. Hill (Huntington: JurisPub, 2008), 95.
193 Jern-Fei Ng, ‘The Role of the Doctrines of Champerty and Maintenance in

Arbitration’, Arbitration 76 (2010): 208-209.

194 Kaplan J in Cannonway Consultants Ltd v. Kenworth Engineering Ltd (1994)
ADRLJ 95, held that it did not apply in arbitration. Similar obiter comments were
made by Steyn LJ in Giles v. Thompson [1992] 3 All ER 321. A contrary view was
expressed by Sir Richard Scott VC in Bevan Ashford v. Geoff Yeandle (Contractors)
Ltd (in liquidation) [1998] 3 WLR 172 and by the Singapore Court of Appeal in
Otech Pakistan TVT Ltd v. Clough Engineering Ltd [2006] SGCA 46. See also
Riberro CJ as part of a unanimous bench of the Hong Kong Court of Final Appeal
in Unruh v. Seeberger [2007] 2 HKLRD 414.
195 While this book is not concerned with a party's obligations to honour its

litigation funding contracts, it would also seem that champerty and maintenance
would never be needed, given that modern contract law principles such as
unconscionability, conflict of interest, duress and misrepresentation would cover
all legitimate policy concerns with litigation funding.
196 See Mark Kantor, ‘Third Party Funding in International Arbitration: An Essay

about New Developments’, ICSID Review-Foreign Investment Law Journal 24, no.
1 (2009): 44 and the articles in Transnational Dispute Management 8, no. 4
(2011).
197 ICC Final Award No. 7006 (1992), in Albert Jan van den Berg (éd.), Yearbook

of Commercial Arbitration Volume XVIII (The Hague: Kluwer Law International,

1993).
198 Cases concluding against this entitlement include ICC Case No. 5029 (1991),

ICC International Court of Arbitration Bulletin 4 (1993): 32; ICC Case No. 6293
(1990), ICC International Court of Arbitration Bulletin 4 (1993): 43; Anderson
Consulting Business Unit Member Firms v. Arthur Andersen Business Unit Member
Firms, ICC Case No. 9797 (2000), ASA Bulletin 18, no. 3 (2000): 514.
199 ICC Case No. 6564 (1993), ICC International Court of Arbitration Bulletin 4

(1993): 46 and ICC Case No. 8787 (1997), ASA Bulletin 20, no. 1 (2002): 68.
200 ICC Case No. 6293 (1990), ICC International Court of Arbitration Bulletin 4

(1993): 43.
201 John Y. Gotanda, ‘Awarding Costs and Attorneys' Fees in International
Commercial Arbitrations’, Michigan Journal of International Law 21, no. 1 (1999):
44, n. 194. See also Yves Derains & Eric A. Schwartz, Guide to the ICC Rules of
Arbitration, 2nd edn (The Hague: Kluwer Law International), 366.
202 For example it had previously been thought in Australia that there might be a

problem with the professional rules proscribing fee sharing arrangements with
non-lawyers if an in-house lawyer shared the recovered fee with his or her
employer. That is no longer the accepted view. See Commonwealth Bank of
Australia v. Hattersley & Anor (2001) 51 NSWLR 333.
203 See, e.g., ICC Case No. 6564 (1993), ICC International Court of Arbitration

Bulletin 4 (1993): 46.

204 Michael Bühler, ‘Awarding Costs in International Commercial Arbitration: An

Overview’, ASA Bulletin 22, no. 2 (2004): 275.

205 ICC Case No. 6564 of 1993, ICC International Court of Arbitration Bulletin 4

(1993): 46.
206 Gillian Lemaire referred to an unpublished ICC award of 2003 which adopted

this approach. See Gillian Lemaire, ‘Costs in International Commercial

Arbitration: The Case for Predictability’,
<www.iclg.co.uk/khadmin/Publications/pdf/2202.pdf>, 2008.
207 ICC Case No. 6564 of 1993, ICC International Court of Arbitration Bulletin 4
(1993): 46.
208 Neil Kaplan, ‘Problems at Both Ends’, in Liber Amicorum Eric Bergsten.

International Arbitration and International Commercial Law : Synergy,

Convergence and Evolution, ed. S. Kröll et al. (Alphen aan den Rijn: Walder Wyss,
2011), 286.
209 See, e.g., Ron Reisner v. Commonwealth (No 2) [2009] FCAFC 172 (Full Court

of the Federal Court of Australia).

210 Compare the UK position in London Scottish Benefit Society v. Chorley,
Crawford and Chester (1884) 13 QBD 872 and the Australian High Court's
decision in Cachia v. Hanes (1994) 179 CLR 403.
211 Separate opinion of Judge Holtzmann, Sylvania Technical Systems, Inc. et al.

and the Government of the Islamic Republic of Iran (1985) 8 Iran-US CTR 329,
331.
212 David D. Carón, Matti Pellopää & Lee M. Caplan, The UNCITRAL Arbitration

Rules: A Commentary, 2nd edn (Oxford: Oxford University Press, 2006), 930-931.
213 Piper Double Glazing Ltd v. Caulfield (1992) 64 BLR 32.
214 Peter Sheridan, Construction and Engineering Arbitration (London: Sweet &

Maxwell, 1999), 428.

215 Bernard Hanotiau, ‘The Parties' Costs of Arbitration’, in Evaluation of
Damages in International Arbitration, Dossier of the ICC Institute of World
Business Law, ed. Yves Derains & Richard H. Kreindler (Paris: ICC Publishing,
2006), 215.
216 See, UNCITRAL Rules 2010 Art. 40; HKIAC Rules Art. 36; ACICA Rules Art. 39;

SIAC Rules Art. 31.2.

217 ICC Case No. 5029 (1991), ICC International Court of Arbitration Bulletin, 4

(1993): 2; ICC Final Award No. 5759 (1989), in Albert Jan van den Berg (éd.),
Yearbook of Commercial Arbitration Volume XVIII (The Hague: Kluwer Law
International, 1993), 34.
218 See, e.g., ICC Case No. 6345 (1991) and No. 6959 (1992), ICC International

Court of Arbitration Bulletin 4 (1993): 45 and 49. Managerial time lost was
successfully allowed as damages in Tate & Lyle Feed and Distribution Ltd v. GLC
[1982] 1 WLR 149. See also R & v. Versicherung AG v. Risk Insurance and
Reinsurance Solutions & Ors [2006] EWHC 42 and 1705 Comm.
219 See, e.g., Alan Redfern & Martin Hunter, Redfern and Hunter on International

Arbitration, 5th edn (Oxford: Oxford University Press, 2009), 546.

220 M. Bühler an& T. Webster, Handbook of ICC Arbitration: Commentary,
Precedents, Materials, 2nd edn (London: Sweet & Maxwell, 2005), 187.
221 Lloyd-Jacob J in In re Mossen 's Letter Patent [1969] 1 WLR 638. See also ICC

Case No. 6293 (1990), ICC International Court of Arbitration Bulletin 4 (1993):
43.
222 See UNCITRAL Rules 2010 Art. 40; ACICA Rules Art. 39; Swiss Rules 2012 Art.

38.
223 The ICC website operates a costs calculator for ease of reference. See
International Court of Arbitration, Costs Calculator,
<www.iccwbo.org/court/arbitration/id4097/index.html>. The International
Court of Arbitration will set the final fee and will also consider questions such as
complexity, procedural stages undertaken, timeliness and general time spent.
224 See LCIA, ‘Schedule of Arbitration Costs’,
<www.lcia.org/Dispute_Resolution_Services/ LCIA_Arbitration_Costs.aspx> 8
July 2011.
225 ICSID, ‘Schedule of Fees’,
<http://icsid.worldbank.org/ICSID/FrontServlet?requestType=
CasesRH&actionVal=CaseScheduled>, 1 January 2008.
226 ICDR Rules Art. 32.
227 Emmanuel Gaillard & John Savage (eds), Fouchard, Gaillard, Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 626-627.
228 Permanent Court of Arbitration, ‘Procedural Guidelines for Requesting
Designation of an Appointing Authority’, <www.pca-
cpa.org/showpage.asp?pag_id=1062>.
229 For example, ACICA Rules Art. 39(b) and (d) expressly reference business

class airfares. Inferentially this would mean that first class airfares are
presumptively not reasonable.
230 Article 22.1(h).
231 Article 9.
232 See ICC Case No. 6268 of 1990, in Albert Jan van den Berg (éd.), Yearbook of

Commercial Arbitration XVI (The Hague: Kluwer Law International, 1991), 125.
There, the award noted the possibility of claiming legal costs in ancillary judicial
proceedings by way of damages, but the tribunal was not required to determine
the issue.
233 David D. Caron, Matti Pellopää & Lee M. Caplan, The UNCITRAL Arbitration

Rules: A Commentary, 2nd edn (Oxford: Oxford University Press, 2006), 933.
234 Bernard Hanotiau, ‘The Parties' Costs of Arbitration’, in Evaluation of
Damages in International Arbitration, Dossier of the ICC Institute of World
Business Law, ed. Yves Derains & Richard H. Kreindler (Paris: ICC Publishing,
2006), 215.
235 See, e.g., CME Czech Republic BV and the Czech Republic, Partial Award (ad hoc

UNCITRAL proceeding, 13 September 2001) available at

<http://italaw.com/documents/CME-2001PartialAward.p...>, paras 620-621.
236 Michael Bühler, ‘Awarding Costs in International Commercial Arbitration: An

Overview’, ASA Bulletin 22, no. 2 (2004): 269.

237 David D. Caron, Matti Pellopää & Lee M. Caplan, The UNCITRAL Arbitration

Rules: A Commentary, 2nd edn (Oxford: Oxford University Press, 2006), 945.
238 Bernard Hanotiau, ‘The Parties' Costs of Arbitration’, in Evaluation of
Damages in International Arbitration, Dossier of the ICC Institute of World
Business Law, ed. Yves Derains & Richard H. Kreindler (Paris: ICC Publishing,
2006), 214, citing ICC Case No. 5759 of 1989, in Albert Jan van den Berg,
Yearbook of Commercial Arbitration Volume XVIII (The Hague: Kluwer Law
International, 1993), 34.
239 For this reason, the Iran-US Claims Tribunal modified Art. 38 of the
UNCITRAL Rules 1976. See David D. Caron, Matti Pellopää & Lee M. Caplan, The
UNCITRAL Arbitration Rules: A Commentary, 2nd edn (Oxford: Oxford University
Press, 2006), 934-935.
240 David D Caron et al., Ibid., 953; Alan Redfern et al., Redfern and Hunter on

International Arbitration, 4th edn (London: Sweet & Maxwell, 2004), 406. This is
also ICJ practice notwithstanding the discretion to award costs in the Statute of
the International Court of Justice Art. 64.
241 David D. Caron, Matti Pellopää & Lee M. Caplan, The UNCITRAL Arbitration
Rules: A Commentary, 2nd edn (Oxford: Oxford University Press, 2006), 953-954.
242 Christoph Schreuer, The ICSID Convention, A Commentary, 2nd edn
(Cambridge: Cambridge University Press, 2009), 1229.
243 See John Y. Gotanda, ‘Attorneys' Fees Agonistes: The Implications of
Inconsistency in the Awarding of Fees and Costs in International Arbitration’,
The Villanova University School of Law Working Paper Series, Paper 144, 2009),
20.
244 Stephan W. Schill, ‘Arbitration Risk and Effective Compliance: Cost-Shifting in

Investment Treaty Arbitration’, Journal of World Investment & Trade (2006):

678-679.
245 See, e.g., International Thunderbird Gaming Corporation v. The United Mexican

States, UNCITRAL, Arbitral Award (26 January 2006) available at

<http://italaw.com/documents/ThunderbirdAward.pdf>; EDF (Services) Ltd v.
Romania, ICSID Case No. ARB/05/13, Award (8 October 2009), para. 327.
246 See, e.g., Glands Gold Ltd v. United States, UNCITRAL, Arbitral Award (8 June

2009) available at <www.state.gov/documents/organization/125798.pdf>.

247 See dissent of Arthur Rovine regarding costs in EDF (Services) Ltd v. Romania.

ICSID Case No. ARB/05/13, Award (8 October 2009), para. 9. See also Dissent of
Professor Thomas Walde, International Thunderbird Gaming Corporation v. The
United Mexican States,

UNCITRAL, Arbitral Award (26 January 2006) available at

<http://italaw.com/documents/ThunderbirdAward.pdfx> Another recent case
arguing that costs should follow the event in investment arbitration is Foresci et
al. v. Republic of South Africa, ICSID Case No. ARB(AS)/ 07/1 Award 4 August
2010.

248 ICC Case No. 5759 of 1989, in Albert Jan van den Berg (éd.), Yearbook of
Commercial Arbitration Volume XVIII (The Hague: Kluwer Law International,
1993), 34; ICC Case No. 5896 (1992), ICC Court of Arbitration Bulletin, 4 (1993).
249 David D. Caron, Matti Pellopää & Lee M. Caplan, The UNCITRAL Arbitration

Rules: A Commentary, 2nd edn (Oxford: Oxford University Press, 2006), 934,
citing the report of the Secretary-General on the Revised Draft Set of Arbitration
Rules, UNCITRAL, 9th Session, Addendum 1. (Commentary), UN Doc.
A/CN.9/112/Add.l (1975).
250 See, e.g., SIAC Rules Art. 30; ICC Rules 2012 Art. 36; Swiss Rules 2012 Art. 41;

SCC Rules Art. 45.

251 See, e.g., ICC Rules 2012 Art. 37(3).
252 As noted above, the latter is complicated by the fact that the order for

production of documents may have been made after an application to the

tribunal.
253 M. Bühler & T. Webster, Handbook of ICC Arbitration: Commentary,
Precedents, Materials, 2nd edn (London: Sweet & Maxwell, 2005), 186.
254 Charles Price & Yves Stans, ‘Using Costs as a Case Management Tool in

International Arbitration’, ASA Bulletin 25, no. 4 (2007): 714.

255 M. Bühler & T. Webster, Handbook of ICC Arbitration: Commentary, Precdents,

Materials, 2nd edn (London: Sweet & Maxwell, 2005), 184.

256 Bernard Hanotiau, ‘The Parties' Costs of Arbitration’, in Evaluation of
Damages in International Arbitration, Dossier of the ICC Institute of World
Business Law, ed. Yves Derains & Richard H. Kreindler (Paris: ICC Publishing,
2006), 218.
257 Gerald Aksen, ‘International Arbitration Knowing the Practical Differences’,

in Liber Ami-corum Robert Briner: Global Reflections on International Law,

Commerce and Dispute Resolution, ed. Gerald Aksen et al. (Paris: ICC Publishing,
2005), 26.
258 Ibid.
259 Murray L. Smith, ‘Costs in International Commercial Arbitration’, Dispute

Resolution Journal 56, no. 1 (2001): 34. But again note that some lex arbitri
expressly allow this to be revisited, or would allow for an additional award.
260 Phillip Yang, ‘The Organisation of International Arbitration Proceedings’, in

The Asian Leading Arbitrators' Guide to International Arbitration, ed. Michael

Pryles & Michael J. Moser (Huntington: Juris Pub, 2007).
261 See Newfield Construction Ltd v. John Lawton Tomlinson [2004] EWHC 3051

(TCC).
262 This is expressly provided for in the Australian International Arbitration Act

1974 s. 27. Less clear is the situation in New Zealand. See Casata Ltd v. General
Distributors Ltd [2006] NZSC 8.

Part III: The Award, Chapter 16: The

Award

Jeff Waincymer,
16.1. Introduction

The ultimate mandate of a tribunal is to render a final and binding determination

of the dispute through a single award or through a series of awards. Such a
determination can also arise from an agreement between the parties where this
is given tribunal imprimatur as an award. This is discussed in section 16.3.7
below.

It is important to distinguish between awards and other decisions of a tribunal.

During the course of proceedings, a tribunal will make many determinations that
would not properly be seen as awards. These include many of the procedural
decisions or orders that frame the course of the process. Rights and obligations
in relation to such decisions differ significantly from those in relation to an
award. A tribunal needs to ensure that everything is done, both in process and
form, to lead to a final and enforceable result where awards are concerned.
There are a number of requirements. The first is that all members of a multi-
person tribunal must take part in the decision-making process and there must be
appropriate deliberations. Conversely, with many mere procedural matters, the
chairperson is entitled to make decisions alone. Where some procedural
decisions are concerned, such as determining whether confidentiality overrides
document production obligations, a tribunal is even permitted to invite an
independent expert to review the material so that confidentiality is not
inherently breached by the evaluation itself. That delegation would not be
possible if the decision was an award, as there is no entitlement to delegate an
adjudicatory function.

For an award to be effective, it must also follow certain processes and meet
certain form requirements, including signature, adequate reasoning and service.
The award may be invalid if these obligations are not met. Institutional scrutiny
may also be a requirement for validity. For example, in the ICC context, a
determination that is in reality an award but which has not been sent for
scrutiny to the page "1263" International Court of Arbitration, may be barred
from enforcement under Article V(1)(d) of the New York Convention by reason
of not following the agreed procedure as per the ICC Rules.

There are also important ramifications as well as requirements if a decision is by

way of an award. An award decision will give rise to res judicata effects as
between the parties. Entitlements to apply for annulment, requests for
interpretation, correction or additional awards, or challenges to enforcement,
depend on the prior determination being an award. Where a decision is an
award, parties also undertake to carry out the award without delay and waive
any rights to other forms of recourse insofar as such waiver can validly be made.
(1) Once there is a final award, the tribunal's powers have also come to an end,

save for any interpretation, correction or addition functions.

Decisions other than awards do not require reasons and would not need to be
scrutinised by an institution. Procedural orders are generally not challengeable
in the Seat of arbitration. Procedural orders can still be challenged at the
enforcement stage but not in and of themselves. This is only possible if they
involve procedural irregularities relating to something that truly is an award. (2)

16.2. What Is an Award?

Because of the abovementioned implications of finality, challenge and

enforceability, it is thus important to determine when a decision of a tribunal
truly constitutes an award. The important question is to distinguish between a
procedural ruling or determination on the one hand and an award on the other.
The description used by the tribunal will not be determinative. A suggestion that
a procedural decision would be an award if termed so by the tribunal was
rejected in the drafting of the Model Law. (3) The cases suggest that one should
look at the ‘function and … effect rather than the form or qualification …’ of the
tribunal's decision. A review court will thus look to the true nature of any
decision reached. (4)

page "1264"

16.2.1. Substance versus Procedure

While the importance of the distinction between awards and other decisions is
undoubted, there is no simple test that has been uniformly accepted. Attempts to
define the nature of an award during the Model Law drafting were rejected, as
there was no consensus on any particular definition. (5) Nor was a definition
included in the New York Convention. Nevertheless, while there is no recognised
definition, there is reasonable consensus as to the key distinguishing features.
This can be discerned from arbitral statutes, cases and commentary. A number of
Model Law countries have added definitions to their domestic arbitral statutes.
Some statutes define an award as ‘a decision of the arbitral tribunal on the
substance of the dispute and includes any interim, interlocutory or partial
award’. (6) Arbitral rules tend to refrain from defining awards, although the SIAC
Rules are an exception and indicate that an award is ‘a decision of the Tribunal
on the substance of the dispute …’. (7)

Commentaries and cases have also attempted definitions. ‘Any decision which
finally resolves a substantive issue affecting the rights and obligations of the
parties is an award’. (8) To truly be final it ‘must be final for all intents and
purposes’. (9) An award decides all or part of a dispute as opposed to deciding
how the proceedings are to be arranged. (10) Another test is to ask, does it settle a
substantive issue between the parties? (11) Redfern and Hunter consider that ‘the
term “award” should be reserved for decisions that finally determine the
substantive issues with which they deal’. The term ‘award’ is applied by arbitral
tribunals to characterise their determination on any matter that may require
judicial enforcement. (12) The Verbiese page "1265" case (13) considered a general
arbitral award to be one which finally determines, in whole or in part, the
dispute referred, be it on the merits, on jurisdiction or on a procedural defence
which might lead the tribunal to terminate the proceedings. These formulations
all distinguish between awards that concern issues on the one hand, and
procedural orders and directions that are concerned with conduct of the
arbitration on the other. (14) The essential difference is not as to timing but as to
nature. For example, a tribunal may make a preliminary award in relation to
jurisdiction or in relation to applicable law as one of its earliest determinations.
These are final determinations of one aspect of the dispute between the parties.
(15) Conversely, decisions whether to have an oral hearing or as to whether

written witness statements should be tendered are merely procedural.

While the reference to an award determining a substantive issue is commonly

alluded to, one of the problems in using this criterion relates to classification
uncertainties under conflict of laws. As noted throughout, different legal families
characterise certain issues differently as either substantive or procedural. This
can impact upon the way they describe or characterise awards, although ideally,
this should not be a determinative factor in any definition. (16) Poudret and
Besson conclude ‘that an award on procedural matters can be defined as a
decision ruling on an issue which may entail, depending on the solution adopted
by the arbitral tribunal, the complete or partial end of the arbitral proceedings’.
The authors suggest that ‘a procedural order deals with the conduct of the
procedure and evidentiary measures, and it has no impact on the possible
termination of the arbitral proceedings’. (17) When discussing procedural or
substantive awards, Poudret and Besson use the term ‘procedural’ award to
cover decisions as to admissibility or res judicata and substantive decisions in
relation to standing, statutes of limitation and preclusion, all of which the
authors believe should be seen as awards. (18) Born argues for an even broader
interpretation, suggesting that interlocutory decisions on disputed issues should
generally be regarded as awards where they finally dispose of a request for relief
by one of the parties by way of an application of legal rules to a factual record. (19)
On this basis, decisions for interim relief and disclosure should be treated as
awards. (20) He suggests that these decisions can be just as important as decisions
page "1266" on the merits of the substantive issues and should be deserving of
the New York Convention's protection. While the efficiency argument is an
important one, nevertheless, the New York Convention is unlikely to have been
intended to cover such matters. To see provisional measures as automatically
constituting awards would lead to some problematic results. For example, two
countries that had adopted the New York Convention and UNCITRAL Model Law,
where only one expressly accepted the interim measure modifications of the
2006 version of the Model Law, would both have provisional measures treated
the same way from an enforcement perspective unless a view was taken that the
measure imposed was beyond the tribunal's power.

As noted above, the description used by the tribunal is not determinative. In

Publicis, part of a claim in a joint-venture dispute was the failure to hand over
certain tax records. One tribunal member signed an ‘order’ which purported to
be ‘for and on behalf of the arbitrators’ calling for the tax information to be
handed over. The direction was not complied with and enforcement was sought,
requiring consideration as to whether it constituted an enforceable award. A US
Court of Appeal considered that the order was in fact a final award on a
particular issue and was hence enforceable. (21) There has been criticism of
Publicis by Pinsolle, (22) and by Poudret and Besson (23) but the distinguishing fact
is that the case was about a joint venture with a substantive right to the
documents. Decisions on substantive rights should presumptively be awards. In
Société Braspetro Oil Services Company (Brasoil), the Paris Court of Appeal had to
consider a tribunal's procedural order which held that allegations of fraud did
not justify an application to review a partial award. The court considered that
such a procedural order was in fact a partial award and should have been
submitted to the ICC International Court of Arbitration for scrutiny. Further, the
failure to follow the institution's rules was a violation of the tribunal's mandate
and/or violated due process rights under Article 1502(3) of the French New
Code of Civil Procedure which is to similar effect.

Some situations may be difficult to classify, in part because of differences in view

as to the essential nature of awards. For example, what about a challenge to the
independence of an arbitrator where the tribunal itself makes the
determination? Is a decision by a tribunal whether to extend the time period for
an award itself an award? Garry Born argues that a stay of arbitration should be
considered as an award. (24) In some cases, little may turn on how decisions on
the margin are classified.

page "1267"

16.2.2. Final Determinations

Even if the subject matter is one that is naturally dealt with in an award by
reason of being a matter of substance, the determination must be a final decision
on that issue if it is to be accepted as an award. (25) A final award is one which
settles the dispute between the parties and is enforceable in a court of law. (26)
Being final, it can be recognised, set aside, or refused recognition. (27) The concept
of finality also means that if the arbitration agreement provides for a two-tiered
or appellate mechanism, it might only be the second decision that has sufficient
finality to constitute an award. Alternatively, the initial decision could become
final once a time limit for triggering the second tier has expired. (28)

A number of courts in the US have taken differing views in both domestic and
international arbitration matters as to when an award is final. Most cases seem
to be consistent with those of supervisory courts in other jurisdictions, although
some decisions have held atypically that a complete answer to one aspect of a
damages claim was not final while there were still other damages elements
remaining. (29)

The fact that an award connotes finality does not mean that it cannot be
impacted upon by later decisions of the tribunal. In Gulf Petro Trading Co v
Nigerian National Petroleum Corp, (30) a tribunal with the Seat in Geneva
bifurcated proceedings and issued a partial award holding the respondent to be
liable, leaving quantum to be determined later. At a later stage, respondent
argued that claimant did not have standing. The final award accepted this
assertion, which in substance overrode the partial award. The final award was
confirmed by a Swiss supervisory court. The claimant nevertheless sought
enforcement of the partial award or alternatively sought to set aside or modify
the final award and assess damages or compel the respondent to arbitrate
damages. A US court refused to interfere with the Swiss court's findings. (31) The
better view is that standing and liability were separate albeit related issues.
When the tribunal ruled on liability it was at a time when standing was not in
issue, hence was final as to its terms. There will page "1268" often be situations
where a later award overrides the commercial effect of an earlier award. An
example would be a later determination of a set-off claim after a determination
of primary liability that goods purchased had not been paid for. If an arbitration
is being run concurrently or ancillary to other proceedings, it may also be the
case that it can finally settle some issue, but not resolve the dispute. This would
still be finality as to the issue within arbitral jurisdiction.

An award will also be enforceable, but it cannot simply be concluded that any
enforceable decisions are necessarily awards. For example, a number of arbitral
statutes allow for enforcement of provisional measures, (32) which many see as
distinct, although as noted above, some argue that these constitute awards in any
event. Some legal systems also allow for enforcement of disclosure orders. (33)

16.2.3. Decisions on Jurisdiction

Under the general test of what constitutes an award, namely dealing finally with
a matter of substance or an issue in dispute, it might at first be thought that any
decision on jurisdiction, both positive and negative, would naturally satisfy the
test. The situation is more complex and there are some arguments to the
contrary. It is important to distinguish between cases where a tribunal or an
institution merely agrees to proceed to a final award on the basis of a prima facie
belief that there is jurisdiction and those where actual decisions are made. For
example, where there is a challenge to jurisdiction in an ICC case, the court will
make a preliminary determination and will only allow a tribunal to be appointed
where the matter is arguable. Such a decision is obviously not a decision of the
tribunal itself and cannot be seen as an award. Similarly, if a tribunal considers
jurisdiction as a preliminary matter and merely concludes that there is a tenable
argument to proceed but leaves a final determination to a later date, it cannot be
an award because the tribunal is indicating that it is still to make the final
determination.

Even where a tribunal purports to make a final determination for or against

arbitral jurisdiction the situation has still been open to some debate. Where a
positive decision on jurisdiction is taken, most would see this as an award.
Nevertheless, Laurence Boo has noted the unique treatment under the
UNCITRAL Model Law which provides for judicial review under Article 16(3)
under specified time limits, rather than a right to challenge the decision as an
award under Article 34. (34) He, thus, questions whether it is properly treated as
an award. Redfern and Hunter still describe a partial decision on jurisdiction as
an award notwithstanding section 16 provisions. (35) Where a negative decision
on jurisdiction is made, Boo page "1269" asserts that such a ruling leaves all of
the substantive claims unresolved, a view accepted by the Singapore Court of
Appeal in Persero. (36) The Court held that such a decision cannot be an award,
because it is, after all, a decision not to determine the substance of the dispute.
In spite of these observations, it would be preferable to see both positive and
negative decisions on jurisdiction as awards. A negative decision on jurisdiction
deals with every substantive issue under the arbitration agreement. It is the final
decision to the effect that the claimant has no valid substantive claims in
arbitration (or that the respondent has no valid counterclaims or set-off rights in
arbitration) as there was never a valid agreement to arbitrate. While a positive
decision on jurisdiction may have more limited challenge rights under the
UNCITRAL Model Law or similar lex arbitri, it should still have all of the other
effects of a final award, including, applying as res judicata between the parties.
(37) Poudret and Besson consider awards on jurisdiction to be just that, but also

note their specific nature means that they can and must be challenged
immediately under most arbitral statutes. (38) Born sees both positive and
negative jurisdictional decisions as awards. (39)

16.2.4. Determinations other Than Awards

Any decision that is not intended to be binding would be contrary to the essential
nature of an award regardless of its title. Numerous rules make clear that an
award shall be final and binding as to an issue in dispute. Generally speaking,
procedural determinations such as to timing, whether there is to be a hearing,
number of witnesses or production of documents are not awards. A tribunal is
not dealing with an issue in dispute and may always change a procedural
determination where later circumstances suggest this is appropriate. While mere
procedural decisions, therefore, cannot be enforced for that reason, in some
cases a tribunal may have separate power to ensure compliance with such
determinations.

16.2.5. Classification and Due Process

As noted above, an important reason to distinguish between an award and other

determinations is to identify the impact and rights and obligations that flow from
page "1270" each kind of decision. The distinction is also important in terms of
the proper behaviour of the tribunal. If the tribunal has erred in its
understanding of the essence of what it is deciding, it may also have failed to
follow necessary processes in support of its intended decision.

If a tribunal wrongly treats what should have been an award as a mere

procedural determination, it may fail to follow necessary form and content
requirements of an award and hence the determination may be ineffective. For
example, if a chairperson alone made the determination of what should have
been treated as an award, there would be a failure of a multi-person tribunal to
engage in deliberations and take at least a majority decision. There may also be a
failure to provide an adequate opportunity for a party to argue the point, a
failure to provide adequate reasoning, a failure of a scrutiny obligation with an
institution and a failure of mandated form requirements.

16.3. Types of Award

Awards can be categorised in different ways although there is no clear consensus

as to typology. (40) Some civilian scholars see classification as important, although
it remains the case that to be properly treated as an award with all of the
implications that flow, it needs to meet the criteria discussed in section 16.2
above. Exact terminology thus is less important than understanding the real
nature of the determination made, which in turn can affect form and content
requirements, rights to challenge, future impacts of the decision on the parties
such as in relation to res judicata, and the ability of the tribunal or arbitral
institution to review the determination. (41) However, many statutes and rules
specifically refer to different terminology and these categories should be
separately considered for that reason alone.

16.3.1. Final Awards

The term ‘final award’ is generally used to describe an award that resolves any
outstanding issues within the reference to arbitration. This might occur in a
single award dealing with all issues or in the last of a series of awards when the
issues were dealt with sequentially. In part, it is an unfortunate expression
simply because the essential nature of any award involves a final determination
as to some issue at least, so all awards are final in that sense. Thus, in the current
context, the notion of a final award is not the same thing as an award that is final.
(42)

page "1271"

A final award is one intended by an arbitrator to be the complete determination

of every issue submitted. (43) In deciding whether an arbitral award is final, it has
been held that where a substantive task remained for the arbitrator to perform,
(44) where certain portions of the parties' dispute remained unresolved, (45) and

significant issues still need to be determined, (46) an award cannot be final in the
sense of being a complete award.

When used to refer to an award that is last in time, hence one that is dealing with
all outstanding issues, a number of implications flow. Subject to entitlements to
interpret, add to, or correct an award as identified in the arbitral rules or statute,
the final award completes the tribunal's mandate and renders it functus officio.
(47) Because of these implications, a tribunal should be wary of describing an

award as the Final Award if there are any outstanding issues, for example as to
costs or interest. Another issue with final awards is that if the tribunal does not
cover everything required, the award might be attacked for failing to complete
the mandate and thus being infra petita. (48) At times such an award may
nevertheless be validly remitted to the tribunal for completion.

Arbitral statutes referring to final awards tend to use the phrase in the context of
the last award. (49) Article 32(1) of the UNCITRAL Model Law indicates that ‘the
arbitral proceedings are terminated by the final award’. Some also refer to the
concept of finality as it pertains to awards generally. For example, Article 19B of
Singapore's International Arbitration Act contemplates the finality of all awards
by stating that ‘upon an award being made … the arbitral tribunal shall not vary,
amend, correct, review, add to or revoke the award’. (50) Rule 14.1 of the CPR
International Rules allows a tribunal to indicate in the award whether it views
the award as final for the purposes of any judicial proceedings. At times context
will show that a particular award is not final in the preclusive sense. For
example, the AAA Class Arbitration Rule 5 requires a tribunal to issue a ‘partial
final award’ as to the class but indicates in Rule 5(e) that such an award may be
altered or amended before a final award is rendered.

page "1272"

16.3.2. Partial or Separate Awards

This sub-section and the following sub-sections deal respectively with partial,
interim, interlocutory and provisional awards. Judges and commentators may
take differing views as to how broadly these phrases should be interpreted.
Some rules speak of ‘partial’ awards while others speak of ‘separate’ awards.
Many also refer to ‘interim’ awards, but these are discussed separately in section
16.3.3 below because they typically deal with the conceptually discrete topic of
provisional measures.

It is important to understand whether phrases such as ‘partial’, ‘separate’ or

‘interim’ cover all determinations such as in relation to jurisdiction, applicable
law, admissibility of a claim or counterclaim and liability in principle, as an
arbitral statute will typically indicate when court review may apply. A partial
award is an award that is a final determination of an issue between the parties
where there are still outstanding issues to be determined. Alonso defines a
partial award as ‘an award that settles disputes between the parties but does not
bring the arbitral proceedings to an end’. (51)

Separate or partial awards are expressly permitted in several arbitration

statutes (52) and also in various institutional rules. (53) The power to issue partial
or separate awards could also be given by agreement of the parties. As discussed
below, it is strongly arguable that it is an inherent power in any event. Partial or
separate awards may be made on a range of issues including jurisdiction
generally, questions of statute of limitations, applicable law, allegations of
breach, defences, quantum of damages or other relief, costs and interest.

Where the power to render partial or separate awards emanates from arbitral
rules or a statute, it will always be discretionary, subject to any agreement of the
parties. Hence, a tribunal can decide whether the circumstances make it
appropriate to do so. The power is certainly desirable as it gives the tribunal a
chance to try and handle the proceedings in the most efficient manner possible,
although there may be contrary considerations in some circumstances. (54) In
many cases, a partial award on one issue will make it easier for the parties to
settle the balance of their dispute. In other cases, a partial award will remove
uncertainty as to the way to best prepare the balance and will concentrate the
parties' attention on relevant matters. For example, once the applicable law is
known, counsel knows what law to research. Another example where a partial
award would be useful would be page "1273" where there is an exclusion clause
or a provision limiting the nature of damages available, which itself is in contest,
although that might at times lead to a final determination on the merits.

Redfern and Hunter also note some dangers in partial awards. Parties may alter
their cases during the course of proceedings and a final partial award may be
undesirably preclusive. A tribunal must be very clear that there is no possibility
that it would wish to reconsider a decision in a partial award. Other concerns are
the possibility of adversely affecting the confidence of the party against whom
the partial award is made. (55) There are also costs involved in rendering a partial
award. In some cases there might be duplicated costs flowing from separate
hearings for separate awards. Another disadvantage of partial or separate
awards is that because they are subject to annulment rights and may be subject
to enforcement proceedings, disputes about the validity of the partial award
could lead to delays in the arbitral proceedings. As a matter of principle,
annulment proceedings as to partial awards or enforcement challenges do not
require a halt in the arbitral proceedings, hence they do not mandate undue
delay. Nevertheless, in some circumstances, a tribunal might feel that an
adjournment is appropriate, in which case that might have been a factor to be
taken into account when first determining whether to render a partial award.

If there is neither agreement between the parties nor express or implied

provisions in the lex arbitri or arbitral rules, it has been doubted whether a
power exists. (56) A contrary argument is that unless there is an express reference
in an ad hoc agreement to rendering ‘an’ award in the singular, a grant of a
mandate to arbitrate a dispute could be argued to encompass a power to do so in
stages where that is thought to be fair and efficient. That could be seen as an
implied term of the agreement to arbitrate in any event and/or an inherent
adjudicatory power. Where the UNCITRAL Model Law applies, it does not
expressly refer to partial awards but the entitlement to render such awards
appears to have been intended by the drafters. (57) Lew, Mistelis and Kröll argue
that in the absence of an agreement, a partial award may still be rendered where
the tribunal considers it appropriate in the circumstances, citing a number of
arbitral decisions. (58) page "1274" Poudret and Besson suggest that there is no
doubt that when the laws are silent, arbitrators can make interim, preliminary or
partial awards unless the power has been excluded by the parties. (59) The
authors suggest that this is a result of the general power to organise the
proceedings. Nevertheless, because of the uncertainty, it is always better to
provide for an express power. (60) Attention should also be given to the scenarios
where the parties either demand a partial award on a particular issue or seek to
bar a tribunal from doing so. If only one party requests a partial award, the views
of each should be obtained before making a determination on this procedural
question. (61)

An important question in the face of either a positive or negative agreement by

the parties is whether a tribunal can ignore the parties' directions if it believes
that it would be fair and/or efficient to do so. This was discussed generally in
sections 2.10.3 and 6.2.4 above. In the specific context of partial or separate
awards, the natural concern is that a failure to follow an agreement of the parties
is a failure to follow the mandate and could hence lead to annulment of the
award. (62) However, if the parties' agreement comes after the tribunal has been
appointed, in appropriate circumstances the tribunal could legitimately say that
it would not be fair and reasonable to force it to make a partial determination. An
example would be where the tribunal concludes that subsequent evidence is
needed to properly determine the issue and that such evidence is not available
within the time frame that the parties have set for the partial award.

Redfern and Hunter suggest that where both parties agree that an interim award
should be made, this must be followed by the tribunal. (63) Subject to the above
qualification as to overriding autonomy, this makes sense, as at worst, the
agreement may needlessly add expense but in most cases, would be highly
unlikely to ever have problems from a fairness perspective. Poudret and Besson
see the parties as also having a power to exclude the making of partial or interim
awards. (64) Conversely, if the parties seek to bar the tribunal from rendering an
initial award, it is suggested that a tribunal may still render a partial award on
page "1275" jurisdiction or on the law applicable to the merits. (65) If the tribunal
made a negative decision on jurisdiction, this is obviously a final award in any
event. However, notwithstanding the above comments about overriding
autonomy there seems no practical reason to make a preliminary positive
decision on jurisdiction where the parties have asked the tribunal not to do so.

Where applicable law is concerned, even if the parties bar the tribunal from
making a determination, because it is desirable to indicate the tribunal's thinking
on issues that would affect efficiency, a tribunal that is convinced that a
particular law should apply could at least notify the parties of its thinking and
invite the parties to consider any reasons why that would not be so and make
submissions accordingly at the earliest opportunity. If the parties are not happy
with the tribunal's view they can always come to a binding agreement on
applicable law.

Party agreement as to partial awards could arise from the outset but more
typically this might occur under a submission agreement, terms of reference or
under an ad hoc agreement as to a provisional timetable. Like any agreement of
the parties, it can at times be ambiguous, which can give rise to challenges at a
later stage as to whether the tribunal has properly followed its mandate and if so,
what ramifications follow. For example, in the Sodifis case, a tribunal was asked
to successively determine jurisdiction, admissibility of claims and their merits.
After dealing with all in a sole award, a Paris Court of Appeal ruled that this was
improper, although that decision was itself overruled by the Cour de Cassation.
(66) A decision not to make a partial or separate award should not itself be

challengeable as an award. (67)

There is also a question as to whether certain forms of preliminary

determinations meet the criteria to constitute an award. Poudret and Besson
question whether an award simply determining liability in principle would be a
partial award or instead is merely a determination of a preliminary issue upon
which the outcome of the claim depends. (68) Born describes partial decisions as
‘arbitral decisions concerning preliminary substantive issues’ (69) and considers
that there must be some declaratory element to constitute an award. (70)
Differences in view can also be complicated by the fact that some authors are
only concerned with whether the term ‘partial’ is appropriate, and not whether
the determination is something other page "1276" than an award. (71) In some
cases, if a tribunal is thought to have failed to cover sufficient issues in a partial
award, it might be treated as less than a final determination and hence not be
enforceable. Even if the concern is not valid, the possibility of an enforcement
court taking a different view remains. The Philippines Supreme Court dealt with
a partial award that held that a call on a performance guarantee was unjustified.
Nevertheless, because the tribunal had not directed return of the proceeds, the
Court chose not to enforce the partial award. (72)

16.3.3. Interim or Interlocutory Awards and Provisional Measures

The term ‘interim award’ is particularly misleading, sometimes used

synonymously with partial awards, sometimes used to deal with a decision
dealing with a key issue leading up to the disposition of a claim without
disposing of the claim itself and sometimes referring to decisions in relation to
provisional measures. It is thus difficult to identify their proper status when the
phrase is used in such distinct ways. Partly for this reason, different courts have
taken varying views as to whether such determinations can be subject to court
application for recognition or annulment. It should not be the terminology that
matters in such circumstances.

This chapter discusses the different meanings separately, simply because of

these ambiguities and the potential for misclassification, including wrongly using
the term ‘award’ in a determination that is not final as to an issue in dispute. It
suggests that the phrase ‘interim award’ not be used unless mandated in an
applicable statute or rules. Interim measures should be described as such where
permitted. Other awards should best be described as ‘partial’ awards if not final
in time.

Nevertheless, some books, rules and cases appear to use the term ‘interim’
awards synonymously with partial or provisional awards. (73) Others concentrate
on the notion of finality. A comparison of comments shows how confusion has
arisen. The Working Group for the UNCITRAL Model Law considered that an
interim, interlocutory or provisional award was one that does not definitively
determine an issue that the tribunal has before it. (74) The JAMS International
Arbitration Rules allow a tribunal ‘to make interim, interlocutory, or partial final
awards’. (75) It does page "1277" not clarify whether the reference to ‘final’ is
limited to partial awards or is intended to relate to all awards of an interim,
interlocutory or partial nature. (76) The English Arbitration Act 1996 refrained
from using the phrase ‘interim award’ which had been included in the earlier
version of the legislation. (77)

The conceptual difference between partial and interim awards was also
addressed by an ICC Working Party. The Working Party suggested that the term
‘interlocutory award’ should not be used as it would cause confusion with
procedural directions. The Working Party noted that interim and partial awards
are often used interchangeably. (78) It suggested that no valid distinction could be
made but that, for practical purposes, it used ‘interim award’ to mean any award
made prior to the last award in the case, and ‘partial award’ to mean a binding
determination on one or more (but not all) of the substantive issues. (79) Craig,
Park and Paulsson also suggest that partial awards be used for those which
dispose of one or more substantive claims while interim awards would in
addition cover awards on jurisdiction or on prejudicial questions. (80) There
seems to be no reason to see this as a meaningful distinction. A partial award on
jurisdiction or on substantive law is still an award, so there would have to be
some merit in separately describing decisions on relief sought and decisions that
are necessary elements of such dispositive decisions. Furthermore, where
jurisdiction is concerned, a respondent's whole defence might only be as to
jurisdiction and hence a decision will be as to its claimed relief in any event.

It has also been suggested that for clarity, ‘the term interim award should be
limited to those awards which do not settle a separate part of the proceedings
finally’. (81) If that is the correct definition, it naturally raises the question of
whether such determinations are in fact awards. (82) Some commentators suggest
that a decision must deal with the claim itself and not merely a subset or gateway
issue to constitute an award. (83) The paradigm example would be an interim
measure granting certain relief pending the outcome of the hearing. The question
of whether page "1278" these are truly awards or not will typically come up in
the context of an attempt to enforce a determination of that nature. In some
countries, it is irrelevant whether a decision on an interim measure constitutes
an award or not as other provisions may indicate that they may be executed and
enforced as if they were final awards. (84)

Alonso questions whether determinations as to interim measures definitively

decide any dispute and hence whether they can be subject to annulment or
enforcement proceedings. Courts have differed on this question, in part because
of differences in facts as to the nature of the interim relief granted and in part,
differences in their willingness to re-characterise the determinations made. This
might also be impacted upon by the view a court takes as to whether it should be
consciously aiding the practical utility of arbitration, which might encourage a
greater wish to enforce interim measures.

The balance of this section looks at the proper status of interim measures and
whether these should be seen as awards. Born argues in favour of such
characterisation by reason that they are final in that they dispose of a request for
relief pending the conclusion of the arbitration; they should be complied with
and it is important for the efficacy of the arbitral process that they be treated as
awards. (85) There are contrary arguments that might prevail. While interim
measures may be final in relation to such requested relief, if the relief sought is
only temporary, then that argument should not be persuasive. Furthermore,
there are many final procedural determinations that satisfy the same test such as
in relation to requests for document production, but which are generally not
treated as awards. Similarly, the fact that interim decisions are meant to be
complied with puts them no higher than document production determinations.
Describing them as different to ‘interlocutory arbitral decisions’ encompassing
such things as liability and procedural timetables, brings together an unusual
array under that term. A decision on liability is somewhat fundamental to the
outcome between the parties. A timetable is not. The argument as to efficacy
seems an argument about ends justifying means. Even assuming that arbitration
requires meaningful interim measures of protection, the proper approach is to
have consensus on this or at least have particular jurisdictions indicate what
they wish to do on the issue. Modifications to the Model Law expanding on
interim measure powers could have dealt with enforceability directly if they
wished. Article 17H as included in the 2006 amendments to the UNCITRAL
Model Law allows for enforcement of interim measures at the Seat of arbitration.
(86) An individual jurisdiction can agree to enforce interim measure

determinations. page "1279" What is less desirable is an argument based on

general efficacy being the basis of inviting the New York Convention to apply to
determinations never intended.

Less clear are the implications of provisions in arbitral rules that expressly
indicate that an interim measure may be rendered in the form of an interim
award. (87) Just as erroneous classification by an arbitrator should not be
determinative, if there is an inherent set of features in the concept of an award,
the same may be argued if misclassification is contained in arbitral rules.
Nevertheless, there may be a difference in that if parties have consented to the
rules, they have consented to the descriptor, which should be an acceptance of
finality and res judicata and may even be a waiver of a challenge to enforcement
if that is required.

16.3.4. Provisional Awards and Determinations

As noted above, some arbitral statutes include reference to ‘interim’ or

‘interlocutory’ awards. Tribunals will often describe a determination in relation
to an application for interim measures as a provisional award. In this book, such
decisions are described as interim and interlocutory and were discussed in the
previous section. Note should also be taken of the English Arbitration Act 1996
which allows a tribunal to grant provisional relief if the parties agree and
generally refers to tribunal ‘orders’ as opposed to awards. (88) While the notion of
a provisional award is accepted under some lex arbitri and rules, it is again an
undesirable descriptor as it is likely to confuse in terms of its implications. If it is
not intended to be a final resolution of an issue it should not be described as an
award. (89) If it is intended to be final, the term ‘provisional’ is misleading as its
ordinary meaning conveys the impression that the decision will be revisited in
due course. Nevertheless, tribunals may continue to use such descriptions from
time to time, in which case the essential character of the decision may need to be
identified.

While the phrase ‘provisional award’ is seen as unfortunate, the notion of a

provisional determination capable of reconsideration can be perfectly
acceptable. For example, a tribunal might wish to make a provisional decision on
a key issue such as jurisdiction. In doing so a tribunal should carefully consider
why it is necessary to make any form of determination, what reasons to give for
the determination and what indication to give to the parties as to their
involvement in the ensuing reconsideration in due course. For example, if there
is a preliminary challenge on jurisdiction, a tribunal can merely determine to
consider the jurisdictional and merits issues together and leave an award until
that stage. It is not making a preliminary determination that it has jurisdiction,
but instead, is simply page "1280" deciding to hear both issues at the same time.
In many cases this will be sensible, for example, if there is a claim of lack of
formation of the contract containing the arbitration clause where the tribunal
will need to consider the nature of the contract for both jurisdictional and merits
purposes. Conversely, if jurisdiction can be finally determined at an earlier stage,
then it should normally be done through a separate award.

In some cases, a tribunal may form a view that it would be worth giving careful
consideration to jurisdiction prior to deciding whether to hear all evidence on
the merits. The parties may even have agreed to a preliminary determination on
jurisdiction. If in either circumstance the tribunal believes that there is prima
facie jurisdiction but wishes to revisit the issue after further evidence, the
reasoning provided should explain why the tribunal has formed that view, why
the tribunal is not able at that stage to form a definitive view, indicate what
further submissions if any on the issue are required from the parties and indicate
when a final determination will be made. If the tribunal provides too little in the
way of direction, the parties are not helped to know what needs to be done to
resolve the issue. If the tribunal provides too much in the way of the reasoning
behind a prima facie jurisdictional decision, it could appear to the ultimate loser
that it has prejudged the final determination.

16.3.5. Default Awards

Another potentially misleading category is that of a default award. This term is

used by some to describe an award, (whether partial or final), rendered in
proceedings where a party has failed to appear or otherwise failed to take part in
the proceedings. It is another misleading descriptor simply because it is very
different to default or summary judgments that are permissible in some
jurisdictions in litigation proceedings when a defendant does not challenge the
allegations made. It is certainly the case that a tribunal may render an award
notwithstanding the failure of appearance or involvement of a party. This is
made clear in virtually all rules (90) and statutes. (91) The power should also be
considered to be inherent. If the two parties agree to arbitrate and one
subsequently breaches its implied duty of cooperation, this should not interfere
with the mandate previously granted.

Describing it as a ‘default award’ is misleading because in such circumstances,

the tribunal must proceed to fully decide the case. There is no possibility of
summary judgment arising simply from the failure to appear and respond. The
tribunal must decide what evidence will be called for, the degree to which it
chooses to test page "1281" it, make any necessary decisions as to applicable law
or jurisdiction and apply the appropriate burden and standards of proof to the
material before it.

There are particular procedural requirements that would be sensible where a

tribunal intends to render an award in such circumstances. These are discussed
more fully in section 6.15. First, a tribunal must ensure that the non-attending
party is given adequate notice of every relevant stage and an adequate
opportunity to attend in the future, notwithstanding a failure to do so in the past.
As noted, the tribunal must also make a considered decision on the merits based
on the evidence and arguments before it. All would agree that it would be
improper to merely accept all of the contentions of the party in attendance,
although it is less certain how active a tribunal should be in challenging these. A
tribunal will have to consider the extent to which it will seek to challenge the
evidence in lieu of the non-represented party. While there are some differences
in view as to the degree of intervention of a tribunal in such circumstances, many
commentators suggest that a tribunal should always consider questions of
jurisdiction when there is no one there to raise them. (92) This is expressly dealt
with in the ICSID Arbitration Rules. (93)

Where the award itself is concerned, it should be as well reasoned as any other
award. In the context of potential challenges, the reasoning in such
circumstances may also help to show that appropriate procedures have been
followed.

16.3.6. Additional Awards

Most rules will allow a tribunal to provide an additional award to cover matters
not previously dealt with. (94) The term ‘additional award’ is applied to follow-up
awards after a final award has been rendered. The typical case is where a
tribunal has failed to address a key issue in its final award. Such awards are
discussed further in section 16.16 below, after discussion of correction and
interpretation powers. These raise similar questions as to entitlement, timing,
form, due process obligations and fee entitlements.

page "1282"

16.3.7. Settlement or Consent Agreements as Awards

If a settlement agreement is rendered as an award, it may be enforced as such. If

instead it is a mere contractual agreement, failure to comply would require an
action for breach of contract. (95) Settlement agreements rendered as awards can
also be described as consent awards. Many arbitral statutes and rules allow for
settlements to be recorded in the form of an award. (96)

Considering whether consent awards should be treated as awards mixes a

conceptual question with a question of practical utility. From the conceptual
perspective, one could be disinclined to see them as awards, not being a
reasoned decision by an adjudicator selecting between conflicting evidence and
arguments. The practical perspective would wish to see the resolution as having
res judicata effect and, in most circumstances, having the support of enforcement
mechanisms. Other advantages of consent awards are that official imprimatur
may make it easier for certain parties to honour the settlement agreement. This
may be so where a State is involved, although if the State can agree to the
settlement it can presumably agree to honour it. While these may be desirable
effects, as noted in the discussion of interim measures above, it is important that
the actions of the tribunal truly constitute an award in the sense of being an
appropriate determination of an issue. That is obviously difficult when the
agreement has been reached by the parties and not by a reasoned application of
law to contested facts by the tribunal itself. Another problem with seeing a
settlement agreement as an award may arise in States that require a dispute for
arbitration to be valid. If there is a settlement prior to any adjudicatory
determination, there may no longer be a dispute. (97)

The view that a purported award without an independent adjudication on the

merits by the arbitrator is not in reality an award is not the accepted view.
Nevertheless, the conceptual perspective suggests that an arbitrator should give
proper consideration to whether the settlement agreement can be justified as a
tenable conclusion of law in relation to the factual record as presented. Where
this may not be the case, a question then arises as to an arbitrator's powers in
such circumstances. This is discussed in the following sub-section.

Various rules and statutes allow a settlement agreement to be recorded as an

award made by consent. Many rules are to similar effect as Article 30(1) of the
page "1283" UNCITRAL Model Law, which indicates that if ‘the parties settle the
dispute, the arbitral tribunal shall terminate the proceedings and, if requested by
the parties and not objected to by the arbitral tribunal, record the settlement in
the form of an arbitral award on agreed terms’. Article 30(2) indicates that such
an award has the same status and effect as any other award on the merits of the
case. All such rules indicate that a tribunal has a right to decide whether to
render such an agreement as an award. The ICC Rules use the mandatory term
‘shall’ but still indicate that it is dependent on the tribunal agreeing to do so. (98)
Article 14 of the UNCITRAL Model Law on International Conciliation provides
that ‘if the parties conclude an agreement settling a dispute, that settlement
agreement is binding and enforceable….’ Nevertheless it includes bracketed
language to indicate that the enacting State may describe the method of
enforcement.

A consent award still must meet all form requirements, save for the more
controversial question as to the duty to give reasons, given that there is no
separate tribunal determination in a consent award. The rules will typically
indicate that an arbitral award in such circumstances does not require reasons.
(99) There might also be jurisdictional issues worthy of consideration to ensure

enforceability. For example, if the settlement goes beyond the ambit of the
original arbitration agreement or the request for arbitration, the settlement can
itself be argued to be an implied modification of that agreement allowing for a
broader consent award. When in doubt, the parties might need to expressly
authorise the broader jurisdiction. Where the parties purport to extend the
agreement they cannot bind the arbitrators to accept this. Redfern and Hunter
also make the very sensible observation that even if the parties do not wish to
have a settlement agreement rendered into the form of an award, it makes sense
to promptly advise the tribunal of the settlement as a failure to do so could
impact upon future fee and expense obligations and possible entitlements to
reimbursement. (100)
One impact of a settlement agreement is to bring to an end the tribunal's
mandate. Because the tribunal has a separate agreement with the parties,
including as to fees, the settlement agreement cannot adversely the affect the
arbitrators' contractual entitlements. In some cases this will be ambiguous and it
is sensible to expressly cover the entitlement in the event of settlement. (101) In
some cases there may be a dispute between the parties as to whether a
settlement agreement has in fact been reached. In such circumstances a tribunal
will need to rule on that question as part of a jurisdictional analysis if a party's
right to pursue arbitration page "1284" in the face of the alleged agreement is
challenged. (102) If a particular arbitration is settled and there are disputes as to
whether a party is validly complying with the terms of the settlement agreement,
a new arbitration will need to be commenced. This is because the settlement
agreement is itself a new agreement alleged to have been breached. There may
be a question as to whether it comes within the initial arbitration agreement. In
some cases the terms of the settlement agreement might be constructed in a way
to allow immediate enforcement of an award under the original arbitration,
although disputes about compliance with settlement may interfere with such a
process.

16.3.8. Refusal to Adopt Settlement Agreements as Awards

The previous section noted that a question arises as to whether an arbitrator

must always render a settlement agreement as an award when the parties invite
this outcome even if there is simply a discretionary power to this end in all or
most cases. The issue arises because of an arbitrator's duty to complete the
mandate, act commercially, and provide an award that is justifiable both legally,
factually and procedurally. Some assert that a tribunal should only refuse to
render a settlement agreement as an award in clearly problematic cases such as
fraud, corruption, and violation of statutory provisions, improper impact on
third party or public interests or unfair treatment of one of the parties. (103) This
response would obviously be recommended if the settlement agreement was
illegal or otherwise was contrary to international public policy. (104) However, an
arbitrator would be unlikely to refuse to adopt a settlement agreement on the
basis of international public policy except in the most extreme examples where
all potential enforcement countries would respect the relevant articulation of
public policy. Poudret and Besson raise an important example of a settlement
agreement that is itself anti-competitive. (105) page "1285" Another example is
where it would violate a mandatory provision of the law of the forum State. (106)

Hausmaninger suggests that the ‘ability of the arbitrator to oppose the recording
of a settlement in the form of an award arguably restricts the autonomy of the
parties in an unjustifiable manner’. Nevertheless, he also acknowledges that an
arbitrator arguably has a right to refuse if the settlement terms ‘are in conflict
with binding laws, public policy, fundamental notions of fairness and justice, or
in a case of suspected fraud or illicit settlement’. (107) This implies at the least that
a tribunal should consider whether the settlement reached is tenable on the
material already placed before the tribunal. Poudret and Besson refer to a
number of arbitral statutes and consider that they ‘do not oblige arbitrators to
review the contents of the settlement submitted for the record, but allow them to
not endorse abusive or illegal agreements’. (108)

The more challenging situation is where a tribunal simply cannot see any logical
basis for it to arrive at the result achieved by agreement if the dispute had
proceeded to a binding determination. For example, a respondent might concede
liability when the tribunal thinks the defence it raised is clearly applicable. This
might also occur where a respondent concedes that goods sold were not of
merchantable quality after an expert tribunal has inspected them and formed a
view to the contrary. Even these two situations could lead to differing outcomes.
A tribunal could form the view that in the first example, the respondent has
effectively waived its defence and the settlement is otherwise consistent with the
applicable law. In the second scenario the conclusion is contrary to the factual
situation as the tribunal sees it. Conversely, that situation could also be looked at
as one where the respondent has now conceded a fact to the point where it is
agreed and hence binding on the tribunal. Another example would be where the
damages claimed were excessive and without proof and where an ignorant
respondent agreed to pay too high a percentage on any reasonable view. There
are some good policy arguments in favour of an entitlement to refuse to render a
settlement agreement as an award in appropriate circumstances. Enforcement
under the New York Convention is supported because national legal systems are
prepared to give appropriate deference to an adjudicated determination by an
arbitrator. If the result simply flows from an agreement of the parties, this
justification is somewhat removed. However, policy issues in arbitration can
always be looked at from consent as opposed to jurisdictional perspectives. From
the perspective of consent, parties that have expressly selected rules allowing for
consent awards page "1286" might argue that they have agreed to settlements
being mandated as awards and hence being available for enforcement under the
New York Convention. Conversely, it can be argued that they consented to the
tribunal retaining discretion in such circumstances.

Importantly, an arbitrator who refuses is not interfering with the settlement. The
arbitrator is simply refusing to give it the imprimatur of an award with the
implications for enforcement or future res judicata arguments that this entails.
Where the parties have on-going business relations, a tribunal might also not
wish the decision to be treated as a guide to future arrangements. A settlement
agreement between the parties is nevertheless a binding agreement which could
be sued on in its own right and which could be shown to be a waiver of a right to
bring other proceedings. If the parties are truly intent on such a settlement, there
is no reason why voluntary compliance should not be the norm. When a tribunal
refuses to render a settlement agreement as an award, it is also not questioning
the reasonableness of the settlement, which could take many factors into
account, most importantly, the wish to find a mutually agreeable solution
without the transaction costs of an adjudicatory process. A tribunal should still
legitimately be able to say in some circumstances that it could not see any
reasoned way to come to that conclusion. Consider for example a case of an
arbitration agreement that allows an arbitrator to grant specific performance but
not damages. The proceedings are commenced and the parties settle for a sum of
money. The tribunal simply could not have rendered a reasoned award of this
nature in the event of the dispute proceeding and arguably could be entitled to
refuse to render the settlement agreement as an award for that reason alone,
although it is arguable as above that the agreement is an implied extension of the
tribunal's remedy mandate.

If a tribunal chooses to refuse to adopt a settlement agreement as an award there

should be no obligation to provide reasons. Giving reasons where the tribunal is
troubled by the settlement agreement could undermine the agreement itself.

Redfern and Hunter point to another issue that tribunals might need to be alert
to. This is where an arbitral process is being misused to hide a criminal purpose
such as money laundering. The authors refer to a situation where parties might
try and disguise an unlawful payment in the form of a consent award. The
authors sensibly point out that where a tribunal has any reasonable suspicion of
such a motivation, the parties should be given an opportunity to provide an
explanation, after which the tribunal might terminate the proceedings or refuse
to allow a consent award. The authors make the apt observation that if the
parties simply seek an order for payment of money, they could effect this
themselves with appropriate settlement agreements and do not need an
enforceable award. (109)

An alternative to a consent award is an express term in the settlement agreement

that if a debtor defaults in required payments, the arbitration can be proceeded
with under an expedited time frame.

page "1287"

16.3.9. Foreign Awards

It is important to consider the distinction between notions of foreign,

international or domestic awards. For example, the UNCITRAL Model Law is
concerned with ‘international’ commercial arbitration. The New York
Convention is concerned with awards other than domestic awards. (110) Some
arbitral statutes seek to define the notion of a foreign award. (111) Depending on
the wording of enforcement provisions, an award must be foreign, non-domestic
or international for enforcement purposes. The key implication is that the New
York Convention is not available to support enforcement of awards in the Seat
where rendered. Under the Model Law, annulment and enforcement can be
sought in that Seat but only for awards in relation to ‘international’ commercial
arbitration.

Because these questions impact upon enforceability, it is not surprising that

some domestic courts have taken different views as to the nature of international
or foreign awards. (112) The key test is to consider the place where the award is
made and compare it to the place where enforcement is sought, although some
domestic courts have taken the view that an overseas award dealing with two
domestic parties on a local issue might not be properly seen as foreign.

16.3.10. Termination of Proceedings without an Award

In some cases, arbitral proceedings may be terminated without an award. For
example, Article 32 of the UNCITRAL Model Law provides for the tribunal to
issue an order terminating proceedings when the claimant withdraws the claim
unless the respondent objects ‘and the arbitral tribunal recognises a legitimate
interest on his part in obtaining a final settlement of the dispute’; the parties
agree on termination; or the tribunal finds that continuation ‘has for any other
reason become unnecessary or impossible …’ (113) This is also reflected in some
arbitral rules. (114) The power to terminate should be seen as inherent, regardless
of any express provision in the rules. (115)

Where a claimant withdraws a claim, a respondent can typically ask for the
proceedings to continue in the hope of receiving an award in its favour, including
a cost determination. Where party agreement is concerned, this also covers cases
where a tribunal refuses to render a settlement agreement as a consent award.

page "1288"

16.4. Truncated Tribunals

As noted in Chapter 5, it is important to consider the impact on award powers if

an arbitrator in a multi-person tribunal resigns, dies or refuses to cooperate. If
an arbitrator resigns improperly in order to frustrate the process, the first
question is whether the relevant rules indicate that a truncated tribunal may
proceed or whether the arbitrator must be replaced. If there is no guidance in the
lex arbitri or the rules, there is a difference in view between those who argue that
the remaining arbitrators may continue as a truncated tribunal and those who
argue that the arbitrator who has resigned must be replaced. (116) This is
discussed in section 5.16. There is also the question of majority deliberations and
voting if an arbitrator refuses to either cooperate or resign. This is dealt with in
section 16.7 below.

16.5. Time Limits

A number of European systems provide for time limits in their lex arbitri as to
when an award must be rendered. (117) Most other arbitral statutes generally do
not impose time limits on the rendering of awards. Where time limits are
contained in arbitration statutes, these are not considered mandatory rules, so
they can be waived or modified by the parties. An agreement to rules that allow
for extensions will ordinarily be seen as an implied variation or waiver. (118) In
some cases the parties may impose time limits in their arbitration agreement,
although this is not common, particularly as it would be hard to define an
optimal time for all future disputes. If the parties impose time limits in their
arbitration agreement, a tribunal should obviously comply or should seek an
agreed extension where it would be impossible or undesirable to meet the
deadline.

There are also no time limits in most institutional rules. One reason why most
rules do not impose limits is to support the right of a tribunal to withhold the
award pending payment of the fees. (119) Another is the concern that a failure to
meet the time limit can have problematic ramifications for jurisdiction. The ICC
Rules are one of a number of exceptions. (120) Article 30 of the ICC Rules 2012
indicates that the tribunal must render its final award within six months from
the date of the last signature by the tribunal or the parties of the terms of
reference or, where Article 23(3) applies, the date of the Secretariat's
notification to the tribunal of page "1289" approval of those terms by the Court.
The award is considered ‘rendered’ on the date the arbitrators or a majority of
them sign the award. That time period must also include scrutiny and approval
of the draft award by the Court. Some rules include special time limits in their
expedited procedure rules. (121) The ICC Court may extend the time limit pursuant
to a reasonable request from the tribunal or on its own initiative if it is necessary
to do so. (122) Notwithstanding the strict language of the ICC Rules, the time limit
is typically extended. The ICC will typically grant three-month extensions, six-
months where there are convincing reasons presented by the arbitrator and
twelve months in exceptional cases where the arbitrators and parties concur
with reasons. (123) Where ICC arbitration is concerned, a tribunal should always
be aware of the time limits and seek necessary extensions, notwithstanding
Secretariat and court practice to manage this and grant extensions without a
specific request of the tribunal.

There are advantages and disadvantages of either institutional rules or the

parties imposing time limits on a tribunal. Where these are realistic and known
in advance, only arbitrators able to comply should accept appointment. The
parties can plan accordingly where there are known time limits. This can be
particularly important when one or both parties would have significant cash flow
concerns if a prompt decision is not rendered. Uncertainty in significant cases
can also impact on other commercial activities or attitudes of investors on public
stock exchanges. Conversely, too tight a deadline can impact on the quality of the
submissions and deliberations. If they can be extended, this raises other policy
questions as to when this should occur and how easy it should be for a successful
request to be made. Time limits that are habitually extended may have little
value. As Alonso suggests, ‘there is no better target than one that can really be
met’. (124) There is also a significant difference in practice between an overall time
limit from commencement of the process or from some later stage such as
drafting terms of reference and instead, a discrete time limit for the rendering of
an award after the close of proceedings. Where a total time period involves the
proceedings themselves, there is then a need to consider how to deal with
behaviour by a party that may appear to frustrate the process. Extreme cases are
easy to deal with, but in many instances, a party will raise tenable arguments as
to why it needs more time than has been allocated.

Whether via statute rules or party agreement, if a tribunal fails to meet a

designated deadline there is then the question as to whether it is still entitled to
render a binding award. A failure to comply or to have an agreed extension could
pose challenges for enforcement. Some courts might hold that consent was
limited page "1290" to being bound by an award rendered within the time limit
and hence no later award is possible, the tribunal being functus officio. Other
courts might hold that the obligation was not mandatory but merely hortatory.
(125) Because a tribunal cannot know in advance, the scenario should not be
allowed to arise without some agreed mechanism.

If an arbitrator is approached to take on a case where there is an express time

limit without an extension mechanism, this should be raised with the parties at
the appointment stage, ideally granting the tribunal a power to extend. Another
possibility would be for the tribunal to clarify that the declaratory award itself
could be rendered within the time period but the reasoned award provided at a
later stage. Such options should always occur with the consent of the parties.

16.6. Process of Deliberations and Decision-Making

16.6.1. The Requirements to Deliberate for Multi-member Tribunals

A reasoned award must be a considered one. Where a multi-member tribunal is

concerned, this involves the need for deliberations between the tribunal
members. The duty to complete the mandate also includes a duty to participate
in the deliberations. This should apply equally to partial awards as well as to
final awards. National legal systems might expressly require all the arbitrators to
take an active part in the deliberations. Even without such a requirement in the
national law, the Swiss Federal Supreme Court has stated that the deliberations
are ‘essential’, as it is an ‘unwritten rule of international public policy applicable
in all international arbitrations’. (126) Alonso suggests there are four subsidiary
obligations in the requirement to deliberate: an obligation to fix a date to
commence deliberations, an obligation to organise them, an obligation to
participate and an obligation to deliberate on all key questions. (127)

While there is a duty, consideration needs to be given to cases where that duty is
breached. If an arbitrator refuses to take part, the better view is that the majority
page "1291" can render a valid award. (128) Most institutional rules other than the
ICC Rules expressly allow the majority to continue notwithstanding a recalcitrant
arbitrator. This is discussed further in section 16.7.

16.6.2. Timing of Deliberations

Fixing a time to commence deliberations and determining the appropriate time

frame for their completion involves a balance between the need for expediency
and the need for an appropriate consideration of all of the issues. (129) That will in
turn depend on such factors as complexity, geographical location of arbitrators,
how familiar the tribunal was with all materials at the time of the hearings and
how important review of transcripts may be to the ultimate determinations.
Timing will also be affected if the tribunal calls for post-hearing submissions. To
the extent that an arbitrator has any doubt as to the nature of the arguments or
evidence, there would also be a duty to clarify this prior to the close of
proceedings. (130)

It is desirable that deliberations as to witness veracity occur as soon as possible

after the hearing while recollections are vivid. To the extent that there will be
any delay, arbitrators need to make detailed file notes that will help jog their
memory. Karl-Heinz Böckstiegel has suggested that witness photographs and
even videotaping of witness testimony may be desirable in appropriate cases.
Problems can typically arise in larger cases not only because of the fact that
leading arbitrators have busy schedules but also because parties typically want
one or two rounds of post-hearing briefs. David Rivkin argues strongly that
tribunals should take no longer than three months to issue an award given that it
is appropriate to do so as close as possible to when they have a full memory of
the hearing and understanding of the material they reviewed prior to that event.
If counsel had to work to tight schedules, so too should the tribunal.

16.6.3. The Deliberative Process and Control by the Chair

There is as much a need for proactive management of the award stage as there is
of the pre-award procedures. (131) The chair should manage the stages of
deliberation, page "1292" preparation of drafts, reviewing drafts, revisions,
finalisation, signature and service. It makes sense to allocate time periods where
all tribunal members are needed for any of these steps at the same time as
designating hearing times otherwise delays are only to be anticipated for busy
arbitrators trying to coordinate their spare time.

Where a multi-member tribunal is concerned, the tribunal chair will typically

manage the process of deliberations, although it is also suggested that decisions
as to how deliberations will occur can be taken by a majority of arbitrators. In
each case this would obviously be subject to obligations of reasonableness and
fair treatment of all of the arbitrators concerned. (132) A chairperson might
distribute an agenda or list of topics or questions for discussion prior to the
commencement of the deliberations. The chairperson might also draft a short
note of the issues, or circulate a draft or drafts of an award as prepared by
various members. Poudret and Besson have suggested as one option, separating
consideration of the facts and legal issues, with submission of a written draft of
the facts, then oral discussion of these, then oral deliberation of the issues in
dispute and circulation of a draft reflecting such deliberations. (133)

The early discussion might indicate which matters the members agree upon,
where there is disagreement and the key reasons for such disagreement. It
would then be important to consider the proper order of dealing with the
various matters. An experienced chair might be concerned to order the issues to
promote efficiency so that a unanimous view on one matter may have necessary
implications and save time for other outstanding issues. An alternative approach
is to try and promote the greatest harmony and willingness to work towards a
majority view and select an order that is thought to best promote this. It is
sometimes desirable to change the order of discussion of each topic.
Psychological studies show that the last person in a group who has a different
view to preceding speakers will often feel intimidated and reluctant to speak out
about their honest belief. Shifting the order so that they can sometimes speak
first ensures that tribunal members feel that their views are equally important.
In many cases, a chairperson might prefer to hear the views of other members
before venturing their own opinion. Other chairpersons may prefer to have
greater control and present their opinions at the outset, seeing where the fellow
arbitrators agree or disagree.

An important question is the appropriate time to be devoted to deliberations and

the willingness of a majority to accommodate minority arbitrators' reasonable
time frames for debate. There is a range of permutations and the only overriding
criterion is that arbitrators must behave reasonably. Each must be given an
appropriate opportunity to engage in meaningful deliberations. A minority
arbitrator must be heard by a majority predisposed to a particular view and
given an opportunity to try to change their minds. However, a minority
arbitrator cannot unduly page "1293" delay finalisation of the award by
asserting an on-going entitlement to do so. Where the tribunal are not all in clear
agreement on key issues, particular skills are needed by the chair in balancing
fairness and efficiency, fairness in terms of giving each member an opportunity
to present their arguments to the best of their ability and to alter the views of
other members, and efficiency in ensuring that no more time is devoted than is
reasonably appropriate. The time to call it a day is when further discussion is
fruitless. The position is particularly challenging as the chairperson has a
legitimate entitlement to present his or her own view to the best of their ability
and at the same time, a duty to make other members feel comfortable in
presenting conflicting views.

At times the chair may need to consider how to deal with pathological behaviour
of a co-arbitrator. One pathological tactic is to seek to resign at a late stage to
frustrate the deliberations. Most rules allow for truncated tribunals in such
circumstances. At times a tribunal chair will also need to deal with unreasonable
advocacy by a party-appointed arbitrator. Another problem arises if a member
simply refuses to take part in the deliberations with a view to deliberately
delaying the decision-making process. As long as the chair sets a reasonable time
for deliberation, a majority should be able to proceed in such circumstances. If
the remaining arbitrators intend to deliberate in the absence of the recalcitrant
member, it is preferable that the parties be advised so that the person might be
removed as a member or otherwise encouraged to participate. (134) Even if a
party-appointed arbitrator displays no willingness to properly participate, every
opportunity should be given for that person to do so including proper invitation
to every meeting, offers to accommodate schedules to their reasonable request
and provision of draft documents including the award. The important thing is to
show that the person had the opportunity to participate even if this was not
availed of. It also makes sense to keep careful records of the invitations made to
the recalcitrant arbitrator to participate. This may help in the event of future
challenges.

If the tribunal relationship is suboptimal, a chair might best conduct

deliberations in a formal manner with written records. (135) A particularly
sensitive situation arises if one arbitrator believes there is some unethical
behaviour on the part of a fellow arbitrator. The better view is that such matters
should first be raised amongst the arbitrators themselves and if necessary with
an applicable institution. Only as a last resort should an arbitrator communicate
with one of the parties to raise such misgivings. (136)
page "1294"

16.6.4. Must There Be Oral Deliberations?

While the norm in multi-member tribunals is to at least begin to deliberate orally

after the conclusion of the hearing where one is held, this is not a requirement.
Furthermore, even where a tribunal sets aside some time after the conclusion of
the hearing to discuss the matter, often there will be no clear agreement at that
stage.

The Swiss Federal Supreme Court has considered that all that is required is that
each arbitrator be allowed to submit his or her opinion and comment on the
views of others. Deliberations can simply involve consideration of written award
proposals presented by other members. (137) Deliberations must involve
consideration of the reasons as well as the ultimate decision and relief awarded.
In many cases each member is invited to draft a particular part of the award or
present suggestions for consideration by fellow members.

There are many reasons why some meaningful oral deliberations are to be
preferred, however. It is generally desirable to have a discussion, as debating
differences in view is much more efficiently conducted orally than in an
exchange of emails. It is much easier to develop an emerging view when persons
can engage in a dynamic discussion. It is also easier to present preliminary
thoughts absent a definitive conclusion as a first step, then invite each to reflect
again on the plurality of views in indicating their likely ultimate conclusions. In
this way it is easier for one member to question another as to the reasons for
differences or pose questions the answers to which would help clarify the views
of an undecided member. However, oral discussions mean that the person
speaking first has to present their views without hearing the opinions of others.
This would not be the case where tribunal members simultaneously present
their views in writing.

In complex matters, the norm is to combine some oral deliberations with an on-
going written procedure, perhaps with a final teleconference to resolve any
outstanding matters. Often the deliberations will simply flow from prior
discussions between tribunal members as the hearing evolves. The members
might have informally discussed what they felt the key issues were and which
testimony would prove to be crucial and which questions needed particular
answers. Such informal discussions can be the background to directions to the
parties towards the matters of greatest remaining uncertainty for the tribunal. At
other times the interpersonal discussions between tribunal members can be
highly influential in terms of the ultimate deliberations.

page "1295"

16.6.5. Place of Deliberations

There is no requirement that deliberations occur in the Seat of the arbitration. At

times, oral deliberations may occur wherever tribunal members are meeting on
another occasion, such as at an international conference. With the ability of video
conferencing or telephone conferences, oral discussions can now also occur in a
number of forms.

16.6.6. Deliberations, Queries and New Evidence

While various statutes and rules indicate that once proceedings are closed,
parties cannot introduce new material or arguments, the tribunal itself is
generally not limited and may even seek further information or clarification from
the parties during its deliberations. Due process should suggest that it should
seek to do so where it is unclear as to any key matter, but it should also consider
whether further submissions from each party would be necessary. (138)

16.6.7. Confidentiality of Deliberations

While the full ambit of confidentiality norms in arbitration is controversial and

has not seen uniform consensus amongst a range of jurisdictions, virtually all
agree that the tribunal's deliberations should remain confidential. (139) This is
enshrined in Rule 9 of the IBA Rules of Ethics for International Arbitrators:

The deliberations of the arbitral tribunal, and the contents of the award itself,
remain confidential in perpetuity unless the parties release the arbitrators from
this obligation. An arbitrator should not participate in, or give any information
for the purpose of assistance in, any proceedings to consider the award unless,
exceptionally, he considers it his duty to disclose any material misconduct or
fraud on the part of his fellow arbitrators.

Obviously the Ethical Rule cannot override a mandatory provision of some court
procedure that makes an arbitrator compellable in challenge proceedings.

The justification for the prima facie norm of confidentiality is that this is an
inherent aspect of the adjudicatory function and one that supports its optimal
exercise. It has been suggested, ‘that secrecy of deliberations is essential if the
deliberation is to produce a true discussion and argument and not become a
mere exchange of cautiously expressed and selected views’. (140) An additional
reason is page "1296" that such confidentiality ‘is intended to ensure that each
arbitrator is able to exercise his or her independent judgment in a collegial
context free of any outside influence’. (141) Karrer suggests that confidentiality
over deliberations aims to protect the serenity of the tribunal and the freedom of
expression. (142) Secrecy is important to allow for a robust exchange of views,
changes of opinion and compromise to form a majority. Lack of secrecy would be
a strong incentive for party-appointed arbitrators to adopt a more persuasive
approach in favour of their appointing party lest that party feels aggrieved. A
further reason to support confidentiality of deliberations is to minimise spurious
annulment or enforcement challenges based on matters raised in deliberations
or differences between the deliberations and the final award. Improper
disclosure of the nature of the deliberations could be a basis for a challenge as to
independence. (143)
Confidentiality has implications for the right to dissent as a dissenting opinion
can naturally divulge differences. Some even assert that dissenting opinions
should not be permitted for this very reason. At the very least, it is important
that any permitted dissent only deals with the express reasoning in the majority
award and articulates contrary views without divulging the deliberations
themselves. Dissenting opinions are discussed further in section 16.7.1.

Confidentiality also has implications for third parties from whom the tribunal
seeks assistance. Because the deliberations are private and confidential, no other
person should be allowed to be present. This should be so even if the parties
have agreed on a tribunal secretary unless this is expressly agreed to by the
parties. This principle should not prevent the tribunal from asking the secretary
to undertake certain forms of assistance or for the tribunal to ask further
questions of the parties or seek advice from a tribunal-appointed expert. As
noted in section 12.13.3, inviting a tribunal-appointed expert to vet a draft award
to check for errors in scientific terminology does not offend against the
obligation to keep deliberations to tribunal members alone. Scrutiny of the draft
award is not the same as scrutiny of deliberations.

A controversial recent decision in the Amsterdam Court of Appeal arose in the

Knowsley case where the court directed that the tribunal secretary's notes should
be disclosed. The decision was overturned by the Supreme Court. (144) The case
dealt with a particular provision in the Dutch Code of Civil Procedure entitling
persons with a legitimate interest to request disclosure of documents or records
relating to page "1297" legal relationships in which they are a party, in this case
the argued relationship between a party and the tribunal. The case was
concerned with notes generally and not particularly with notes as to
deliberations.

16.6.8. Institutional Scrutiny and Further Deliberations

Where there is institutional scrutiny, such as is the case with the ICC
International Court of Arbitration, substantive comments might expressly or
impliedly call for further deliberation by the tribunal. (145) This would be the case
with matters of substance, although the chairperson alone could no doubt
correct a mere typographical error. (146) The same may be so where a draft
dissenting opinion causes further reflection by the majority.

16.6.9. Challenges as to Adequacy of Deliberations

While the deliberations are confidential, at times a tribunal member may breach
this obligation and warn a party of the direction of the discussions. In other
cases, an arbitrator may allege that he or she was denied an adequate
opportunity to engage in the deliberations. Each arbitrator must be given a
reasonable opportunity to have their views heard and considered by fellow
arbitrators. A failure to afford such an opportunity would be a fundamental
procedural error.

Supervising courts will be loath to entertain or at least accept such arguments,

although such claims are always possible. (147) While deliberations are secret, if a
challenge is made and the arbitrators are asked to give evidence, confidentiality
is obviously breached. (148) The issues are not without controversy. In the CMEA
case, all arbitrators were forced to give evidence in the Swedish court
proceedings. (149) Poudret and Besson criticise the view of the Svea Court of
Appeal that arbitrators should be treated as witnesses of fact and made to testify
about their arbitral proceedings. (150) Whether they may be compelled to give
evidence will depend on the law of the supervising court and the extra-territorial
reach of that law. How an arbitrator will choose to behave may depend on an
amalgam of express page "1298" provisions in the law and general principles of
confidentiality. If the supervising court seeking tribunal evidence is in a Seat
selected by the parties, the duty to divulge can be traced to both party consent
and the jurisdictional basis of the tribunals powers and duties. While an
arbitrator should be reluctant to divulge information without compulsion,
extreme situations are more complex. Most agree that there would be a duty to
disclose gross breaches of propriety by fellow arbitrators. Yet an arbitrator
wishing to disrupt the process can easily abuse such a power with false
accusations if it was too readily afforded.

16.7. Majority Decisions

It is important for arbitral statutes and rules to give guidance as to whether

unanimity is required and if not, how to deal with a range of situations where a
multi-person tribunal does not reach consensus. While majority voting is
generally allowed, this is separate to the obligation to first deliberate and
attempt a consensus position. (151) Arbitral statutes and rules will generally allow
for majority awards. (152) A majority opinion is also generally permitted if one
arbitrator simply refuses to take part in the deliberations or in any vote. (153) Such
provisions are generally not mandatory, hence parties could in theory agree to a
requirement of a unanimous decision. In most cases this would be undesirable,
as it would grant each arbitrator in a multi-person tribunal an effective veto
power. Nevertheless, in some cases parties might wish to maintain the
commercial status quo unless there was a unanimous view to the contrary.

Some situations are particularly complex even if majority voting is permitted and
the chair is not given a determinative vote as is sometimes the case. One complex
situation is where three tribunal members all have distinct views. This will
typically occur on quantum questions. It is entirely conceivable that there may be
no majority view in favour of any form of relief between a three-person tribunal.
Consider, for example, a claim in damages for failure to pay for services. The
claimant argues that there was a valid contract concluded or, alternatively, that
there is an entitlement based on a restitutionary principle such as unjust
enrichment or quantum meruit. One tribunal member might uphold the
claimant's findings on contract alone. A second member might uphold the
claimant's findings page "1299" on quantum meruit alone. The third member
might reject the claimant's submissions entirely. A majority has found in favour
of the claimant but there is a majority against each legal argument. Does the
claimant deserve to win? Is there any obligation on the tribunal to reconsider its
position? Legal systems may differ on these questions and arbitral rules and
statutes give no guidance. Alonso notes the position in relation to Article 16 of
the ICSID Arbitration Rules that if a proposal fails to achieve a majority it should
fail. (154) This still depends on what is considered to constitute a ‘proposal’.
Absent any directive, the better view should be that if the majority considered
that the claimant deserved to succeed, albeit for differing reasons, then this
should arise if there is a consensus on quantum, even if only a consensus as to
the minimum entitlement.

Some statutes and rules allow for a definitive award by the chairperson in
circumstances where no majority exists. (155) Where the chairperson has the right
to make a determination absent a majority view, it is thus possible that both co-
arbitrators would dissent. (156) A further alternative approach is to require the
chair to select among the differing positions of co-arbitrators. This may be a
problem in that each of those views may have been disapproved of by the
chairperson. (157) Fouchard suggests that there may be a transnational norm of
public policy supporting such a chairperson's casting vote, (158) but Poudret and
Besson point to the diversity of rules as a countering argument. (159) Alonso
suggests that allowing the chairperson to make the decision if a majority is not
obtained prevents page "1300" deadlocks and protects the effectiveness of the
arbitration proceedings. (160) While such a rule ensures that there will be no
deadlock, it has important implications for the power of the chairperson and
their ability to sway others to their views in deliberations. (161) A unilateral
deadlock power may impact even sub-consciously upon the entire deliberations
and not simply operate as a residual power.

One interpretative question concerns the ambit of a chairperson's powers if

there is an agreement on some matters but not others. (162) Poudret and Besson
suggest that unless the questions are necessarily linked, there ought to be a vote
on each separate issue and that the chairperson should only exercise a casting
vote when it is unavoidable. (163) In other cases, statutes and rules might require a
majority decision. An abstention should be seen as a negative vote. (164) If there is
no apparent majority position where a majority is required, the preferable view
is that the duty to complete the mandate involves a duty to continue
deliberations to try to form a majority. (165) Hausmaninger thus suggests that
where there are three different legal opinions, arbitrators might arguably be
under a duty to find a compromise. (166) A contrary argument would be that they
have deliberated fully and the only question is where the majority opinion is in
such circumstances. Once again there are advantages and disadvantages of any
such obligation. The disadvantage is that some arbitrators may feel a need to
reach a compromise that in their view could simply not be reached by any
reasonable adjudicatory thought process. If they then provide honest reasons for
their concurrence with the majority, that might even lead to challenges to the
award on the basis that there is no true majority for a particular declaratory
finding. (167)

There also needs to be some mechanism where there is an even number of

arbitrators. This will usually be a casting vote of the chair. If there is no
indication, page "1301" a failure to achieve a majority in favour of any form of
relief is again arguably rejection of that claim.

16.7.1. Dissenting Opinions

16.7.1.1. Permissibility and Utility of Dissent

Most arbitral statutes or rules do not expressly refer to dissenting opinions in

arbitral awards although there are exceptions. It is widely accepted that such
opinions are not part of the final award and should not be treated as a distinct
award. Civil law jurisdictions have had the greatest concern with dissenting
opinions, seeing them as being contrary to the collegial nature of the courts, (168)
although a number of civil law jurisdictions have now expressly allowed for
dissenting opinions. (169)

The divergent historical views meant that no specific provisions could be

included in the UNCITRAL Model Law or the UNCITRAL Arbitration Rules. (170)
The UNCITRAL Secretariat commented that ‘the Model Law neither requires nor
prohibits ‘dissenting opinions’. (171) The Iran-US Claims Tribunal incorporated an
amendment to the UNCITRAL Rules to allow for dissenting opinions. (172)
Dissenting opinions are expressly allowed for under the ICSID Convention. (173)
The express reference in the ICSID Rules and the modification by the Iran-US
Claims Tribunal are based on the view that dissenting opinions are seen as
particularly suitable where a State is a party. (174) In other cases, dissents are now
expressly allowed for in some national laws or arbitral rules. (175) Where arbitral
rules are silent as to the possibility of dissenting opinions, there has been debate
about whether there is a general entitlement or not. The 2007 SCC Rules
removed the reference to dissenting opinions but this was not intended as a
change in page "1302" entitlement. (176) More generally, Karrer suggests that
once the decision has been reached and the award issued, an arbitrator may
rightly distance himself from a majority decision and voice a dissent. (177) Born
argues that the right to provide a dissenting or separate opinion is a natural
corollary of the adjudicative function and the obligation to provide a reasoned
award. (178) That is certainly the better view although the argument is circular in
that the contrary assertion would simply be that the duty is only to reach a
majority reasoned award as this is the only outcome with dispositive effect.

Redfern presents a strong argument against the prevalence of dissents in

arbitration although he accepts the reality that dissents are here to stay. (179)
Arguments against the use of dissenting opinions include that it would disclose
too much about the deliberative process, would identify a dissenter with the
appointing party and would identify grounds for challenging the award. (180) They
may undermine respect for the majority conclusions. Dissenting opinions can
also add costs and delay the process. There are thus both theoretical and
practical elements of the policy debate. Because it is widely accepted that the
deliberations of the tribunal are confidential, some have questioned whether
dissenting opinions are even permissible given that they would at least point to
some disagreement emanating from those deliberations. While some eminent
scholars support the view that any dissent breaches secrecy, (181) as noted, some
rules take a different approach in expressly supporting the right to dissent at the
same time as requiring secrecy of deliberations. (182)

The better view is that dissents do not breach confidentiality in any problematic
way. Confidentiality is as to the deliberations, not the reasons for the decisions.
Transparency as to reasons is a requirement itself. The duty to complete the
mandate is an individual one, albeit encompassing duties of collegiality and
deliberation. Each arbitrator should ultimately make a determination as to the
way the applicable law applies to the facts as found and present that view page
"1303" within the deliberations. The common practice of having an odd number
of arbitrators implies that it can readily be contemplated that differing
arbitrators may come to different conclusions. An odd number at least means
that a majority decision can be identifiable in all but exceptional cases. (183) A
dissenting opinion that merely presents the reasoning that was found
compellable by that arbitrator is not divulging the nature of the deliberations.
The contrary view often fails to distinguish between the deliberations
themselves and the reasons why an individual arbitrator cannot concur with the
majority's published reasoning. The only tenable aspect of the breach of secrecy
argument is as to the voting process itself. To some, even a mere indication that
there was a majority decision divulges an aspect of the vote and should not be
allowed. The concern is even stronger when the identity of the dissenter is
known. While it cannot be denied that disclosure of dissent shows a lack of
consensus in deliberations, in essence, it simply shows that the reason for the
decision is that a majority preferred a set of arguments over those preferred by
the dissenter. That is best seen as an honest articulation of required reasons,
with the corollary that inferences as to deliberations must be allowed to that
extent at least. Even if a dissenting opinion is an improper breach of
confidentiality, there is no reason why it should invalidate the award as the
majority has made the decision and the dissent merely presents a different view.
The ICC Working Party did not identify any country where an award might be
challenged simply because of the inclusion of a dissenting opinion. (184)

Other arguments operate on a more practical level. There is a difference between

what is permissible and what is desirable. While it may make sense to expressly
indicate that dissenting opinions are permissible to avoid doubt, it has been
observed that this may be undesirable in unduly encouraging such opinions. (185)
An ICC Commission Report concluded that no active steps should be taken to
encourage dissenting opinions. (186) There is some concern that dissents may
increase the likelihood of challenge by providing criticisms of the majority from
within. Whether a dissent encourages challenge depends on the matters
addressed. If the dissenting arbitrator has taken a different view on a substantive
issue whether as to fact or law, it ought not to justify any challenge to the award.
The more difficult situation is where the dissent is based upon a difference in
view as to procedure. The aggrieved party, regardless of whether there is a
dissent, could page "1304" address such irregularities in any event and should
have done so during the course of proceedings. The same is so if only the
arbitrators know the irregularity.

Where an institution is involved, it might have been invited to intervene as to the

irregularity. In an extreme case, an individual arbitrator might justify resignation
on the basis of a procedural impropriety. To continue with the hearing and
allude to procedural improprieties in a dissenting opinion is decidedly
suboptimal as it gives grounds for challenging the award but no ideal evidentiary
basis for making a determination. Nor is it desirable to call arbitrators as
witnesses in dealing with these issues.

There may be a range of factual permutations that suggest differing solutions on

a case-by-case basis. In some cases a dissent on procedural grounds might even
make it easier to dispose of a challenge. This would be so where comments about
procedural error in a dissent could be analysed as compared to the behaviour of
the aggrieved party and the arbitrator prior to the award. In some cases this may
bar a successful challenge by showing consent or waiver at an earlier stage. In
one case, the contents of a dissenting opinion even allowed a supervising court to
conclude that the majority had not overlooked a witness statement even though
it was not referred to by them. The statement was referred to in the dissenting
opinion and the court noted that such a written opinion could not contain
anything not presented by the dissenter to the co-arbitrators. (187) However, an
assertion in a dissenting opinion where there is no contemporaneous evidence
may support some challenges that could not otherwise be successfully made.
There is also the question of whether the dissenter could provide the
information in other ways, such as by affidavit to be tendered in court.

When analysing the utility of dissents, it is important to understand the

difference between dissenting opinions in courts and dissenting opinions in
arbitrations. Where common law courts are concerned, a dissent provides a
counterbalancing argument to the majority for the benefit of an appellate court.
Other courts, for which the decision is merely persuasive, are given competing
arguments through the dissent. Because arbitration does not have an appellate
process and because there is no doctrine of precedent and only modest reporting
of awards, a dissenting opinion must be justified on other grounds. While there is
no doctrine of stare decisis in arbitration, nevertheless certain opinions are
influential, particularly in the growing field of investment arbitration. Here we
see many experienced commercial arbitrators coming to grips with complex and
at times unfamiliar concepts such as national treatment, fair and equitable
treatment, nondiscrimination, most favoured nation clauses and expropriation.
Where such decisions are published, critically analysed and cited in other cases,
they will obviously be influential. Having said that, the dissenter is not under the
same duty as a publicly appointed judge in a common law jurisdiction to use
individual cases as a mechanism to help develop the law for the benefit of a
broader society or group.

page "1305"

Some criticisms are based on presumed impact on behaviour. While some argue
that the potential for a dissenting opinion will encourage party-appointed
arbitrators to feel pressured to support their appointer, the same situation
would arise with a simple refusal to sign an award. Even if dissents are
invariably provided in favour of the party that appointed the dissenter, this does
not necessarily show bias as the parties may have selected arbitrators with a
particular approach to questions of law and fact that would be favourable to
their case strategy. Alonso makes the observation that preventing dissenting
opinions because a party-appointed arbitrator may wish to frustrate the process
does not stop other forms of frustrating behaviour. Similarly, some of the key
problems with dissenting opinions might arise in any event. An example would
be a breach of the confidentiality in the deliberative discussions other than
through an express written opinion. (188) There is even a contrary logic that the
ability to provide a written dissent might even discourage the dissenter from
continuing to press their views in deliberations, knowing that its distinct
position can at least be identified to the appointing party. (189) If so, that might be
a distinct problem arising from the entitlement to dissent.

One argument in favour is that the very essence of adjudication is to make the
decision on the issues before the tribunal. If a particular tribunal member has
made a decision, albeit a minority one, he or she ought to be able to reflect that in
the award. To ask them to sign on to reasoning of the majority, to which they do
not agree, is contrary to the decision they actually reached. (190) As noted, some
argue that the right to dissent is an inherent power of the adjudicatory process.
Such opinions may be the last resort to identify flaws in the majority and even
fundamental procedural irregularities, although as to the latter, it has been
observed that there a better avenues earlier on in the proceedings in all but
extreme cases. Publication of dissents will also help scholarly analysis.

From the perspective of the parties, an important justification for reasoned

awards is to explain to the loser why the particular result was arrived at. If there
is dissent, the real reason for the decision is that the majority took a particular
view contrary to the third tribunal member. An honest and comprehensive
articulation of the reasoning would indicate contrary arguments that are tenable,
albeit ones that have been rejected by the majority. If the reasoning is complete,
the losing party not only benefits from knowing why it lost in the instant case,
but in many circumstances of modern commerce, is then better able to modify its
commercial transactions in the future to better protect its position.

Dissenting opinions may also assist in settlement negotiations and clarification

of entitlements in future transactions. A very significant percentage of modern
international trade is between parties with on-going commercial relationships.
page "1306" A dissenting opinion might show that there is at least some
uncertainty in the terms and conditions of the parties' dealings that might
require greater clarity to avoid future disputes. They also allow an arbitrator to
maintain his or her reputation if the majority reasoning is thought to be of an
inferior standard. (191) Another advantage is that a draft dissent that is well
argued may alter the views of the majority. (192) A draft dissent is no different to a
well-argued position communicated between arbitrators during deliberations. It
may be more persuasive when articulated in writing, showing the majority that
they will have to reject a powerful set of reasons to reach a contrary conclusion.
(193) A dissent may also be useful for an ICSID annulment committee or to assist

ICC Court scrutiny. While dissenting opinions are not scrutinised as such by the
ICC Court, their very existence might call for greater scrutiny of the majority's
view. (194)

While there are thus some potential benefits, given the potential problems,
consideration should also be given to whether the suggested benefits could be
achieved in less contentious ways. For example, the concern for robust analysis
should arise with open and frank deliberations and can even have the
dissentient's arguments included in the award as well-expressed alternative
submissions that were nonetheless rejected, without needing to identify the
voting pattern.

16.7.1.2. Process for and Treatment of Dissenting Opinions

The debate has generally moved from an initially polarised discussion as to

whether dissent should be permitted and/or is an inherent right of an
adjudicator, to consider the way dissenting opinions should be handled in
practice to minimise disruption and promote any perceived advantages. Arroyo
identifies a range of possibilities for the treatment of a dissenting opinion. If the
majority agree on incorporation of the dissenting opinion in the award, they
could include it in full; include a summary of the dissentient's views or
intersperse particular aspects of the dissent throughout the majority reasoning.
Where reference is made to the dissent, the majority can then decide whether to
give reasons why they disagree with the dissenter, or simply provide a brief
summary of those reasons or give no indication of its content.

page "1307"

Where the rules are silent, it is left to the majority or agreement of the parties to
determine whether a dissenting opinion may be joined to the award. (195) Poudret
and Besson support this power, although they acknowledge the potential conflict
in such a power and sensibly recommend that such a decision be taken at the
opening of deliberations before divisions in views are known. (196) However, even
raising it for discussion may encourage greater use of the possibility. If an
arbitrator indicates an intention to provide a dissenting opinion, the chair should
call for its submission before the final award so its contents can be adequately
considered. (197) If there are two dissents, there will also need to be an
opportunity for each to comment on the other's views. (198) In cases where a well-
argued dissent might conceivably cause reflection by the majority, some time
ought to be granted to allow this to occur. This would not be necessary in a
simple case where there was a dispute of fact between conflicting witnesses and
different arbitrators simply took a different view as to which witness to prefer.

The Working Session of the International Court of the ICC adopted the following
approach to dissents in September 2000:

… the dissenting arbitrator should be invited to indicate if his document

constitutes a dissenting opinion which he wants to have communicated to the
parties or just comments for the benefit of the Secretariat and the Court. The
majority should then be invited to consider whether in view of the dissenting
opinion, they want to change anything in their award. At the same time the
arbitral tribunal shall be informed that the dissenting opinion will be
communicated to the parties when notifying a signed award. They should not be
asked whether they agree or disagree with the communication of the dissenting
opinion. If the arbitrators take the initiative and request that the dissenting
opinion not be communicated by the Secretariat, they are to be invited to state
good reasons why the communication could endanger the validity and execution
of the majority award … All comments, etc. will be submitted to the Court Session
when approving the award. In the absence of valid reasons the opinion will be
sent to the parties. (199)

The ICC Court will also consider whether to attach the individual opinion to the
award and pursuant to Article 41 of the ICC Rules 2012 will seek to ensure
enforceability in making such determinations. (200)

page "1308"

Some take the view that a dissenting opinion can only explain the reasons for
disagreement with the majority award and hence cannot be circulated prior to
the majority award being known. (201) The better view is that prior dissemination
is of a draft dissenting opinion. As noted, the majority might incorporate the
views or even change their views, making publication of the dissent unnecessary.
Conversely, if the dissenter feels that the majority has attempted to cover the
minority views in its final award but has done so inadequately, a revised final
dissenting opinion might be utilised. The key issue is whether to simply include
the dissenting reasons in a unified award. If the majority award includes the
arguments of the dissentient, this can ensure that the views are expressed
objectively and without rancour. This also helps show that all arguments were
properly considered. Where the dissent is based on a tenable difference in view,
the position might well have been covered in any event in a unanimous decision,
given that the losing party would have presented the arguments ultimately
rejected. A well-reasoned award would explain why those arguments were
rejected.

Where all signatures are required for award validity, it would be far preferable to
include the dissentient's view in the actual award to ensure that there is no valid
ground to refuse to sign. (202) In some cases this may be impractical if the
dissenter wishes to have the primary role in articulating his or her arguments
but cannot meet a reasonable deadline for completion of the draft award. There
may also be problems if the dissenter takes issue with the way the majority deals
with the minority view in the draft award.

There is also the question whether the dissenting arbitrator can demand that his
or her opinion be attached to the award regardless of the views of the majority.
As noted, in the absence of an express direction, one view is that a dissenting
opinion might only be permitted if the majority of the tribunal allows. (203) Article
53 of the Chinese Arbitration Law indicates that the opinion of the minority
arbitrator shall be recorded in writing. A refusal to allow a dissenter to attach a
dissenting opinion should not lead to challenge to the award unless such an
entitlement is expressed in party agreement, the lex arbitri or the arbitral rules.
(204) Evenif there was a right to include a dissenting opinion, this is only a matter
of form and should not affect the enforceability of the award itself, particularly as
the minority arbitrator can notify the dissenting opinion to the parties
separately. While most believe that the majority can determine whether a
dissenting opinion will be attached to the page "1309" award, the majority
cannot bar the dissenter from separately presenting his or her opinion to the
parties and any relevant institution. (205)

While the better view is that dissents do not breach principles of confidentiality
over deliberations, and in any event, should not ground successful enforcement
challenges, if the country where the arbitration is held took a different view, a
dissent should not be included with the award. A tribunal might consider a
similar response if a known enforcement country also would have concerns with
dissenting opinions. Care should also be taken in the contrary situation where a
failure to include a dissent would be seen by an enforcement court as improper.

If the award does not contain the dissent, the tribunal may choose to organise
service of the dissent simultaneously with the award. Arroyo suggests that this
should occur through separate mail. (206) If the dissent is separately served, the
tribunal will then need to decide whether to refer to the dissent in the award. A
tribunal might also choose to indicate that the dissenting opinion is not part of
the award. Most institutions will distribute dissenting opinions to the parties
where they are provided to the institution. (207) Institutions will tend to have the
dissenting opinion sent to the parties simultaneously with the award if not
already included, but would justifiably not wish to hold up delivery of the
majority award if the dissenting opinion is delayed. Where the ICC invites the
majority to consider whether the dissenting opinion should be communicated,
arguments presented against might relate to the invalidity of a dissent under the
applicable law or a potential adverse impact on enforceability. (208) Another
question is whether to disclose the name of the dissenting arbitrator. The ICC
Working Party recommended that dissenting views be presented on an
anonymous basis. (209)

It has been suggested that if an award is annulled, a dissenting opinion might

provide grounds for challenging the dissenting arbitrator on the basis of
independence (210) although if an award is annulled it is the majority view that
has been overturned. The dissenter might be challenged on the basis that the
opinion is so unreasonable that it shows bias, but this would be rare and would
require extreme facts. If the award is annulled, a priori the majority view should
at least be equally open to such scrutiny.

page "1310"

16.7.1.3. Content of Dissenting Opinions

The IBA Rules of Ethics for International Arbitrators do not address the issue of
dissenting opinions. (211) Lévy suggests a code of ethics covering the need to warn
colleagues of dissenting intentions, submissions of a draft dissent, limiting the
statement to appropriate matters and not unduly delaying the award. The ethical
position of the dissenter should ensure that the aim is to promote reflection by
the majority and if no changes are made, to promote the optimal articulation of
the conflicting arguments. If included in any form, a dissent should not be
described as an award. It should be set out in a way that is clear to the reader
that it is a separate opinion and not dispositive in any manner. Arroyo suggests
that it might appear in italics or be added after the signatures under the heading
‘dissent’. (212)

Even where a dissent is permissible, an arbitrator should not use this power to
unduly criticise the majority opinion and undermine the award. Just as the
parties have agreed to be bound by a majority decision, so too has an arbitrator
in the minority agreed to work within that framework and respect the decision-
making procedures leading to a binding award. The dissenting arbitrator should
never use the process to try and demonstrate why the dissent is the only position
that is reasonable except in the most extreme cases where the majority decision
itself could not be reached on reasonable analysis. As James Carter has argued
cogently ‘(t)he arbitrator's right to dissent should be exercised with the due
respect for the integrity of the process’. (213) A good faith dissent merely explains
the reasons for the difference in view. On this basis, it should not deal with
arguments that were not presented in the deliberations. A dissenting opinion can
address the contrary reasoning in the majority and explain why a different view
is held but should ensure that this is done respectfully, objectively and
impartially. At times, an arbitrator may simply wish to indicate dissenting
reasons but nevertheless indicate a willingness to concur in the conclusion. (214) A
dissent should not include personal attacks on co-arbitrators, breach of
confidential aspects of deliberations or a mechanism to assert grounds for
annulment or challenges to enforcement. If there were such legitimate grounds,
an individual arbitrator should already have raised them, or have even resigned.

Some Swiss commentators have argued that a dissent may not even be a critique
of the award and should be limited to questions of law. There is no reason it
should be narrowed in this way or why a respectful explanation of the differing
reasons and conclusions from that of the majority, which by definition is a
critique, page "1311" cannot occur. (215) It may be more a question of semantics
as some commentators suggesting that dissenting opinions should be limited to
questions of law nevertheless include rules of evidence and even appreciation of
facts within that concept. (216)

It may also be that the opinion is not truly a dissent. For example, section 16.7
looked at the situation of differing views of arbitrators where it may be uncertain
whether there is a majority for any proposition. What about a separate opinion
that merely has differing reasons for the same relief as granted by the majority?
In such circumstances there is in fact a unanimous view of relief and the better
view would be that properly articulating the reasoning of all members should
indicate the alternative views.

16.8. Content and Formalities of the Award

As noted in section 2.7.14.1 above, a number of rules require a tribunal to make

every effort to ensure that the award is enforceable at law. A contentious
question is whether an award needs to comply with each and every special
requirement for enforceability in particular jurisdictions that may be thought to
be relevant. There is no automatic obligation to this effect but the more the
parties bring the requirements to the attention of the tribunal and the less
onerous they are, the more sense it makes to comply.

A number of national legal systems impose strict form requirements on awards.

These may require awards to be in writing, indicate the date and place of the
award, name the parties and arbitrators and be signed by the majority. Some
rules and statutes require particular matters to be dealt with in the award. For
example, ICC Rules 2012 Article 37(4) states that the final award shall fix the
costs of the arbitration and decide which of the parties shall bear them or in
what proportion they shall be borne by the parties. The Swiss PIL indicates than
an award must be signed and dated and indicate if unanimity was not achieved.
(217) Conversely, Section 52(1) of the English Arbitration Act 1996 indicates that

the parties are free to agree on the form of the award. Content, form and
structure are addressed in the following sections.

page "1312"

16.8.1. Awards in Writing

While conventions do not expressly call for awards to be in writing, this is

implied, for example, through Article IV(1)(a) of the New York Convention
requiring a ‘duly authenticated original award or a duly certified copy thereof for
the purposes of recognition. Most arbitral statutes require an award to be in
writing. (218) Article 31(1) of the UNCITRAL Model Law requires the award to be
in writing, which is said to be mandatory. (219) If it is not mandatory, there seems
no reason why parties could not agree to an oral award, perhaps with written
reasons presented at a later stage. While this is conceivable, there could be a
range of problems including the process of enforcement under the New York
Convention, and the practicality of challenges against an award with no
documented reasons. Other rules simply indicate that the arbitrators must sign
the award, which implies a writing requirement.

Just as the writing requirement for arbitration agreements has been expanded to
include electronic forms, it may be that over time, awards may also be treated as
being in writing when submitted electronically, although at present this should
only be so where expressly allowed. (220) In such circumstances, particular laws
dealing with electronic signature would need to be complied with.

16.8.2. Who Drafts the Award and When?

There is no express rule as to the way awards should be drafted. It is generally

accepted that the chairperson is in charge of the drafting process. Some chairs
prefer to draft the entire award themselves and then invite comments from co-
arbitrators. At other times, a chair might delegate different sections to each
arbitrator based on their preferences or particular expertise. In a more complex
matter and if the working relationship between tribunal members is at a
sufficient level, various parts may be allocated to each for preliminary drafts,
particularly where one tribunal member has special expertise. (221) For example,
in a construction dispute, an engineer arbitrator may be allocated the task of
synthesising the expert evidence while a lawyer may deal with questions of
applicable law or contractual standards for applicable remedies. Such division of
labour should not entail delegation of decision-making but should follow full
deliberations by the arbitrators on each issue. If one has particular expertise,
then they may have an educative function with their co-arbitrators during such
deliberations to allow the others to page "1313" truly have an independent view
notwithstanding their relative lack of expertise. Each arbitrator must by the end
become sufficiently expert in any key area to make the decisions of law and fact
that are required in an adjudicatory process. If different parts are delegated to
different members of the tribunal, the tribunal might first agree on the key
principles and decision, leaving it to individual members to simply draft the
reasoning behind it. (222)

As to timing, there is a need to comply with time limits overall, which in the
extreme could render the tribunal without power if not complied with. This was
discussed in section 16.5. Another issue is whether the drafting must only begin
after proceedings are closed. While a tribunal must keep an open mind until after
the close of proceedings and then evaluate the conflicting evidence and
arguments, there is nothing improper in preparing parts of the award in draft on
an ongoing basis, particularly in relation to the description of the parties,
counsel, appointment of the tribunal, procedural history, outline of the written
submissions of the parties and summaries of documents and witness statements
tendered. In many cases, setting out a succinct outline of the written submissions
prior to the oral hearings in draft form will be an excellent way to ensure
sufficient familiarity with the key issues so that the tribunal is best able to
understand and manage the oral stages.

A chair may need to deal with circumstances where the majority is happy with
the draft award but the third arbitrator demands numerous changes before he or
she would be willing to sign. The changes might be agreed but the arbitrator then
refuses to sign. For those changes that the majority did not feel were truly
warranted but were still seen as a legitimate compromise in the hope of reaching
unanimity, it is desirable that it is made clear at the outset which changes are
accepted as such and which are only conditional on the third arbitrator signing
the award and not dissenting. Proposed changes could be in tracked form and
identified as such. (223)

16.8.3. Structure of the Award

The award should outline a number of key elements, although there is no strict
ordained order accepted by all. Typically an award will begin with an outline of
the parties, counsel and the tribunal. Reasons will need to be given in a later
section if the parties subject to any order as to relief are not identical to the
parties in the arbitration agreement, perhaps as a result of claims of extension to
non-signatories, such as via agency or application of the group of companies
notion.

The award should contain particulars of the arbitration agreement and of the
relevant commercial contracts or other activities in relation to which the dispute
arises. If the arbitration agreement contains preconditions, the award should
page "1314" indicate whether these have been met or waived. (224) If terms of
reference are utilised, such as in ICC arbitration, these should be set out. In some
cases there may have been a revised list of issues, in which case that might best
be set out and frame the structure of the award.

An award will separately outline the procedural history and the facts of the case.
The procedural history should aim to show that an adequate process occurred
with input from all parties. Because the grounds for challenge are based on lack
of procedural due process, it makes sense for an award to go to sufficient lengths
to adequately set out the procedural background. Providing the reasoning for
any contentious procedural orders previously made is also useful, as any
reviewing court will typically look at the award as a self-contained document.
(225)

As to the factual background, if the case is not too complicated, the introduction
can set out what the case is essentially about. (226) If a summary of facts is
contained at the outset by way of background, it is important that this does not
distort the position, pre-empt later factual determinations or lead the losing
party to believe that some prejudgment has occurred. Where preliminary factual
matters are concerned, the aim is to set out the facts in a non-controversial way
as background to the dispute rather than give an indication of the factual
determinations made as part of the award. At times it may nevertheless be
desirable to articulate a finding that has been made in the award even at this
early point, if this is needed in order to promote a better understanding of the
award's reasoning and structure.

An award will then outline the key claims, defences and counterclaims, including
the key arguments of each of the parties. It will identify the relief sought by each
party. These issues can be separated into claim, response, and reply stages or can
be grouped by issue. The award will then contain the evaluation by the tribunal.
This will outline the key evidence presented, the factual decisions made by the
tribunal and the views on legal questions. It must provide a sufficiently reasoned
enunciation of the conclusions reached. Where an award summarises arguments
and evidence, it is important that the right balance is struck in terms of the detail
and coverage. As with most forms of writing, it is important that the award is
neither too long nor too short. If too short, there is likely to be an inadequacy of
reasons. If too long, business persons concerned may find it too difficult to
comprehend the gist of the decision. (227) Where key factual matters are page
"1315" concerned, it is appropriate to summarise the evidence and separately
indicate why the tribunal came to particular conclusions about disputed facts.

Finally, the award must contain the directions, orders and other remedies that
have been determined to apply. An award must include a section clearly
indicating the relief granted by the tribunal.

The duty to complete the mandate requires an arbitrator to deal with all claims,
counterclaims and defences in the award, not make findings beyond the matters
raised and contain a definitive dispositive part. (228) A related concern is not to
engage in obiter dicta for educational or diplomacy reasons. (229) Article 47 of the
ICSID Rules is an example of a highly prescriptive list of formal requirements for
an award. Even where this does not apply, it forms a useful guide to the matters
that might be covered. Article 47 states that:

1. The award shall be in writing and shall contain:

a. A precise designation of each party;
b. A statement that the Tribunal was established under the Convention, and
a description of the methods of its constitution;
c. The name of each member of the Tribunal, and an identification of the
appointing authority of each;
d. The names of the agents, counsel and advocates of the parties;
e. The dates and place of the sittings of the Tribunal;
f. A summary of the proceedings;
g. A statement of the facts as found by the Tribunal;
h. The submissions of the parties;
i. The decision of the Tribunal on every question submitted to it, together
with the reasons upon which the decision is based; and
j. Any decision of the Tribunal regarding the cost of the proceeding.
2. The award shall be signed by the members of the Tribunal who voted for it;
the date of each signature shall be indicated.
3. Any member of the Tribunal may attach his individual opinion to the award,
whether he dissents from the majority or not, or a statement of his dissent.

16.8.4. Language of the Award

Selection of the language of the arbitration generally extends to the appropriate

language for the award itself. Some national laws have at times required awards
to be rendered in the official language of the Seat, although this would be rare in
modern times. The parties can agree on a different language for the award to that
of page "1316" the arbitration. Alonso also suggests that the tribunal can decide
whether a more appropriate language is to be utilised. (230)

Where recognition and enforcement is concerned, an award needs to be

translated into the language of the relevant jurisdiction where recognition and
enforcement is sought under the terms of the New York Convention. (231)

16.9. Reasoning in the Award

16.9.1. The Requirement for Reasons

Two issues arise in relation to reasons. The first is what level of reasons should
be contained in an award. The second is what challenges can be made where
reasoning is asserted to be inadequate. Section 16.7.1.2 above dealt with the
separate question as to whether dissenting reasons can or should be included.
Historically, arbitrators in common law countries refrained from providing
reasons at a time when courts were more likely to intervene in response to
reasoning thought to be defective, while some civilian scholars considered the
duty to give reasons to be a matter of public policy.

The requirement to provide adequate reasons is a disincentive to arbitrariness,

allows participants to feel that justice is seen to be done and in some cases even
leads the adjudicator to conclude that the initial inclination simply cannot be
justified on any reasoned basis. (232) The provision of reasons also supports the
entitlement to call for corrections, interpretations and additional awards. (233)
Many arbitral rules and statutes indicate that for other than consent awards,
reasons must be given subject to contrary agreement of the parties. (234) An
agreement that reasons are not necessary might be taken for cost-saving
purposes. It would be a waiver of a right to challenge the award on that ground
alone, although it ought still to be possible to raise other challenges. Waiver may
be controversial in that the entitlement to reasons may also be affected by other
laws such as Article 6.1 of the European Convention on Human Rights which
could be page "1317" asserted to be mandatory, (235) and by the civilian view that
it is a matter of public policy.

As to the ramifications for lack of sufficient reasoning, a distinction needs to be

made between the general requirements in arbitration and specific cases where
a certain degree of reasoning is specifically required under the relevant rules. In
the latter case, a lack of reasons can be argued to be a failure to follow agreed
procedures and hence be more susceptible to challenge. For example, awards
rendered under the English Arbitration Act 1996 can only be challenged for
‘serious irregularity’. (236) Section 52(4) provides a general obligation to give
reasons and section 70(4) allows a court to remit an award where reasons are
insufficient. Section 70(4) allows a party seeking to appeal on a question of law
or set aside an award to request an order from the court calling for further
reasons from the tribunal so that the court can properly consider the application.
The English Court of Appeal considered that the Arbitration Act 1996 allowed
confidential reasons to be provided which were not part of the award but the
court considered that these could be examined in relation to serious
irregularities. (237) One English case considered that an ambiguity can arise as a
result of ‘inadequate rationale or incomplete reasons’. (238)

In most jurisdictions, a failure to provide reasons when required may allow the
award to be set aside on public policy grounds. (239) There is no express provision
in the New York Convention in relation to reasons underlying awards, although
this could also come under the public policy exception in Article V(2)(b). (240)

16.9.2. Adequacy of Reasons

A number of cases have considered the adequacy of reasons in the context of

arbitral awards. It is desirable to consider the policy issues in determining the
nature and ambit of reasons that ought to be provided. The value of reasoned
argument was touched on in the previous section. The benefits are thought to be
to ensure that the adjudicator gives appropriate consideration to reasoning,
provides a transparent mechanism for the parties to consider whether all
relevant factors were taken into account and all irrelevant factors were excluded,
to page "1318" similar effect ensure that the parties can see that their
contentions were properly considered even if the result was not in their favour,
helps the parties determine whether there are any errors or gaps needing
correction, interpretation or additional awards, allows for a consideration as to
whether annulment proceedings should commence and gives guidance to the
parties in relation to their future behaviour.

These objectives then provide guidance as to the degree of reasoning required.

Reasoning must be sufficient as to breadth, covering all claims, and depth,
indicating the logic behind each decision. Conventional wisdom suggests ‘that a
losing party should by the terms of an award understand clearly and
unequivocally why its case was not accepted by an arbitral tribunal’. (241) Born
suggests that ‘(t)he essential requirement is that the tribunal identify the issues
that were dispositive in the dispute and explain, concisely, the thought-process
underlying its decision’. (242) An award should thus indicate each step in the
reasoning and the reasons why individual issues were determined, such as why a
particular choice of law was made. (243) The award should also indicate why
alternative submissions have been rejected. Reasons must be able to be
understood, be sufficient and deal with the key points. They must also not be
internally inconsistent. (244) However, the requirement for a reasoned award is
not a requirement for logical or well-argued awards, albeit that these are the
ideal. The important distinction is that a party should not be able to challenge an
award for lack of adequate reasons simply because it finds some logical flaw in
the reasoning itself. If it could do so, that would be tantamount to allowing for
appeals. In some cases, tribunals have used contradictory language in describing
this aspect of the adequacy of reasons. (245) In Klöckner, (246) and ICSID tribunal
considered that statement of reasons needed to be ‘sufficiently relevant’. (247)
There is a fine line distinguishing between an irrelevant reason which is illogical
per se and tantamount to an appeal, and an irrelevancy that means that there is
no reason whatever in relation to the decision as reached. There remains a page
"1319" live debate as to whether challenges should be limited to form or
conversely, the extent to which they can enter into matters of substance. (248)

Karrer suggests that it will often be good practice to include subsidiary reasoning
supporting the same conclusion if the primary reasoning is incorrect. (249) Of
course such reasoning should be independently valid and should not be strained
simply so as to diminish the success of attacks against primary reasoning. One
advantage of valid subsidiary reasoning is to obviate the need for a remittance
back to a tribunal or worse, disputes about the availability of a residual forum if
recourse cannot be had to the original tribunal. (250) In jurisdictions where a court
may hear appeals on questions of law, it may be appropriate for an arbitrator to
indicate what view would have been taken of the facts if a different legal position
pertained, again so that costly remittance is not required. (251)

While due process suggests that each of the parties' arguments ought to be
considered, there is no commensurate duty to address each argument in the final
award. On one view, reasons are only needed for the decision in relation to each
and every claim and counterclaim, which could mean that there is no need to
address the weakest arguments. However, the ultimate decision was dependent
on rejecting those weak arguments so the reasons for the decision should
logically cover the latter as well, albeit briefly. In a practical sense, investment
arbitrations tend to address such issues because of a concern for deference to a
sovereign nation State as a party. If the sovereign raises an argument, an
arbitrator will naturally be disposed to at least address it. Where private parties
are concerned, an arbitrator might feel more comfortable in simply ignoring a
specious argument. Nevertheless, Karrer even suggests that attention be given to
arguably irrelevant submissions. Counsel has typically presented the results of
expensive research in support of such arguments. Failing to give them any
careful attention can simply increase the chances of challenges to the award. (252)
In terms of guidance for the future, it may help if the tribunal explains why the
arguments were without merit.

In some instances reasoning is difficult to articulate. It can often be particularly

difficult to provide clear reasons as to why certain factual assertions are
preferred. An expert trade arbitrator asked to evaluate whether a commodity is
of appropriate quality or not can rarely provide much in the way of meaningful
reasoning, save to say that a particular view was formed on inspection. (253) At
other page "1320" times a tribunal will simply indicate which of conflicting
witnesses was found to be more reliable. It is difficult to provide objective
reasons for such a view if there is no contemporaneous evidence showing that
one witness is untruthful, for example, where statements in documents conflict
with the testimony and where there is no adequate explanation for this. Where
expert evidence is concerned, the tribunal's reasoning may need to enter into
complex theoretical and factual elements.

Often an award will contain little in the way of reasoning on the level of interest
chosen, whether simple or compound interest is applied or the reasons for the
allocation of costs. To a lesser extent, reasons tend to be suboptimal in many
quantum assessments. In each case, clear reasoning is still to be preferred.
Another question is whether arbitrators who came to a unanimous view as to
relief but for differing reasons should state this clearly in the award. All other
things being equal, a duty to provide reasons should be a duty to provide the true
reasons, including the differing reasons in such circumstances. The only contrary
argument is in relation to confidentiality of deliberations. That should not be a
concern. If after those deliberations, different arbitrators truly have different
reasons for their conclusions, this could be articulated but without identifying
the individuals in the award.

Landau argued that investment arbitration calls for more carefully and fully
reasoned awards in view of the inherent public interest in most investment
arbitrations and the ongoing ramifications through the interpretation of complex
international investment law norms. (254) Yet many senior commercial arbitrators
may show a decided reluctance to make definitive comments about complex
public international law norms. (255)

16.9.3. Challenges to Adequacy of Reasoning

It is important to consider the likely result of a challenge as well as any

articulation of principle as to adequacy suggested by tribunals, reviewing courts
or scholars. Some courts hold that a failure to give reasons, while breaching a
form requirement, does not constitute sufficient grounds for annulment. (256)
Courts in France will not support annulment on the basis of inadequate reasons.
(257) From a policy perspective, if reasons are inadequate it is better for a

supervisory court to remit the page "1321" matter back to the tribunal rather
than set aside the award. (258) It should only be where there are separate grounds
for annulment that this should arise. A failure to give adequate reasons was seen
as one factor in an American court setting aside an award for ‘manifest disregard
of the law’, (259) a uniquely American criterion.

Reviewing courts will generally limit themselves to determining whether there is

reasoning and whether the reasoning is adequate. They will not concern
themselves with the correctness of the decision although if the reasons are
internally contradictory, some challenges have been upheld. Where reasoning is
internally contradictory, not only is it illogical but it can become impossible to
know just what the reasons were. Hence, in substance it is as if there was no
reasoning provided. As to adequacy, a distinction is made between coverage of
issues and coverage of all of the arguments pertaining to the issues. A reasoned
decision is required on each issue but the tribunal is not bound to address every
argument raised by each party.

In a widely criticised decision, an Australian court in BHP Petroleum held that an

arbitrator must give reasons commensurate to that provided by judges in their
determinations. This was a case dealing with the domestic Arbitration Act where
historically there were broader grounds for judicial review than contained in the
UNCITRAL Model Law. The judgment considered that the degree of reasons
would vary depending on the facts and nature of the dispute, the procedures
adopted, the conduct of the parties and the qualifications and experience of the
tribunal. In a large-scale matter with a retired judicial officer as arbitrator, a
judge considered that a tribunal was ‘under a duty to give reasons of a standard
which was equivalent to the reasons to be expected from a judge deciding a
commercial case’. (260) The view has not been followed in subsequent decisions.
(261) It is unhelpful to define principles by reference to judicial standards,

particularly as there is no common approach worldwide that would be a

yardstick for international arbitration.

16.10. Signature

It is normally the case that all arbitrators must sign the award, although this
should not allow one arbitrator to delay its completion. If one arbitrator is
refusing to sign a common award, most rule systems allow for signature by a
majority or even by the chair alone. (262) If the chairperson refuses to sign, the
majority will page "1322" suffice. (263) If an arbitrator refuses to sign, many rules
also indicate that reasons must be given. (264) Care needs to be taken to deal with
cases where there is a difference between the lex arbitri and selected rules as
there may be arguments as to which applies. In most cases this will not be a
problem but in some, one or other has a unique requirement in such
circumstances. Normally the rules would prevail but it might be arguable that the
lex arbitri provision, if more exacting, is mandatory. This would require analysis
and interpretation of the particular law's intent.

If that arbitrator is dissenting, the person's view is not part of the majority and
hence they are not signing the majority award in any event, although they are
entitled to do so. A dissenter should be prepared to sign an award in any case, as
signature by a dissenter does not constitute consent to the majority reasoning
but merely shows that all tribunal members were involved in the process. If a
statute or rule required all members of the tribunal to sign an award, each
arbitrator is contractually bound to do so. (265) Signature by all is apparently still
required under Article 41 Saudi Arabian Law on Arbitration. (266) If the applicable
law requires the signature of all tribunal members, a dissenting arbitrator cannot
refuse to sign simply because a majority refused to allow inclusion of the
dissenting opinion. (267) Born suggests that a refusal to sign in such circumstances
would allow for challenge and removal of the arbitrator. (268) It might also be that
there would be personal liability and any immunity would not cover a wilful
refusal to comply with the mandate. In a converse scenario, there is even the
potential that in a multi-member tribunal subject to overriding liability, one
arbitrator might seek an indemnity from the majority. (269)

If a draft award is submitted to the ICC for scrutiny, it may be signed to ensure
that arbitrators will not change their view although they may need to do so
depending on the court's recommendations. (270) It is also necessary to consider
whether the particular jurisdiction requires the signature to be witnessed and if
so, whether page "1323" there are any particular qualifications required of the
witness. In some jurisdictions, the award may need to be notarised. (271)

16.10.1. Place and Date of Signature

Arbitrators need not all sign the award in the same place or at the same time. (272)
Many arbitral statutes and rules indicate that an award is deemed made at the
place of arbitration. (273) Previously, there had been some concern that the
physical place of signature could affect the notion of the place where the award
was ‘made’ for New York Convention purposes. Older and contrary decisions (274)
holding that an award was made where an arbitrator physically signed it have
been widely criticised and modern statutes and rules have sought to prevent this
from applying. Section 53 of the English Arbitration Act 1996 was enacted to
overcome the judgment of the House of Lords in Hiscox v Outhwaite (275) which
had held that an award was foreign simply because it had been signed in Paris
when the dispute was between English parties with an English arbitrator and a
Seat in London.
Legislative provisions and arbitral rules commonly require that an award be
dated. (276) Because the date is important for commencement of time limits for
confirmation, annulment, correction or interpretation, it is important that this is
an accurate date and that the award be promptly communicated to all relevant
parties. If there is more than one arbitrator signing, there is a question as to
whether each should date when they sign, or whether the date should be that of
the last signature, as this is the only time when the tribunal as a whole has
complied with signature obligations. Article 47 of the ICSID Rules indicates that
each signatory should indicate the date of the signature. (277) More normally, an
award should be considered rendered on the date of the last signature. Typically
the chairperson will sign last. In any event, presumably it is the last date of the
majority that is page "1324" determinative for commencing time periods for
correction, interpretation, additional awards, annulment and enforcement.

Arbitrators are not obliged to all sign the same copy of the award. (278) One
reason to allow arbitrators to sign different copies is so that a recalcitrant
arbitrator cannot frustrate the process by holding on to the originals and not
signing them or forwarding them on to co-arbitrators.

16.10.2. Non-compliance with Form Requirements

If there are such form requirements as to date, signature and designation of

place of the award, a question arises as to what is the effect of the award if there
has been some failure to comply by the tribunal. A failure to follow a form
requirement may not necessarily invalidate the award. There is a range of
permutations depending on whether the form requirements are in the lex arbitri,
the arbitral rules, a stipulation of the parties or are required by an enforcement
court. The first question should always be whether the form requirements are
mandatory. Even where this is so, statutes and rules will typically allow for
correction.

In other circumstances it is necessary to determine whether annulment is

possible on the grounds of form defects. (279) Even where challenges are possible,
another question is whether a court will allow for substantial compliance or
would differentiate between serious or minor form defects. This may also
depend on whether a late challenge could be barred on the grounds of estoppel
or waiver. As noted, a failure to provide adequate reasons has been seen as
grounds to block enforcement in some jurisdictions. Some national systems seek
to provide a rule of reason test. For example, section 58(2)(h) of the English
Arbitration Act 1996 allows for enforcement to be blocked where form errors in
the award cause or will cause substantial injustice to an applicant.

16.11. Scrutiny of Awards

A number of institutional rules call for scrutiny by the institution's Secretariat.

ICC Rules call for scrutiny by the Court of Arbitration. (280) Such scrutiny is
limited, in particular to ensure that there are no demonstrable errors or gaps in
coverage. In practice, the ICC Court is seen as reasonably intrusive, concerned to
ensure the highest quality of awards rendered under its control. (281) Derains and
Schwartz noted that some 30% of awards reviewed by the ICC Court in 1997
were returned page "1325" with comments as to either form or substance. (282)
The ICC Court does not provide reasons when reviewing awards. (283) The ICC
supports this approach on the basis that it aims to reach a consensus and that
articulation of reasons would encourage court challenges. (284)

Where the ICC is concerned it has also been noted above that it does not
scrutinise or approve a dissenting opinion as a required element vis-à-vis the
award itself, but does take the contents into account in scrutinising the draft
majority award. There is also a difference between the matters that the ICA Court
can require and instead the matters of substance that can simply be drawn to the
attention of arbitrators. A question then arises if the tribunal does not wish to
adopt the suggestions of the Court.

At times parties themselves will include the possibility of some form of scrutiny
in the arbitration agreement itself, either a second tier arbitral tribunal or
purport to call for judicial review of arbitral awards. In most cases they would
not truly be matters of scrutiny.

16.12. Service and Notice

16.12.1. Communication Requirements

It is important to consider what notice obligations there are in relation to an

award and what powers a tribunal has to modify an award up until such time.
Proper communication has a number of ramifications. An award can only be said
to be binding when it is brought to the attention of the relevant parties. Subject
to rights to correct, clarify, interpret or produce additional awards, proper notice
is the final act in the arbitrator's mandate and may thus render the tribunal
functus officio. Time periods also begin in terms of challenges to the award,
although that is not always so. Time limits for correction, clarification or
additional awards may also be dependent on the date of communication. These
powers are considered in sections 16.13 to 16.17.

A tribunal generally has a duty to deliver the award to the parties or at least
notify the parties of its completion. (285) Different arbitral statutes and rules use
different terminology in terms of notification or delivery or the like. Different
statutes refer to the award being communicated, notified, delivered, rendered or
served. In some cases it may even suffice that the parties are notified that the
award page "1326" is available for collection. Generally speaking, the award
must be delivered promptly or without delay after it is made. (286) Another issue
is whether delivery should be directly to the parties or may occur via counsel. It
is also important to determine whether relevant time periods begin on receipt by
counsel where that is considered appropriate. Attention would need to be given
to the lex arbitri and to any agreement of the parties, which may be included in
notice provisions contained within terms of reference or articulated in
procedural directions. (287)

In ad hoc arbitration, the tribunal or its chairperson generally makes notification.

Where an institution is involved, the tribunal may have to submit the award to
the institution which then notifies the parties. (288) In some cases the tribunal is
itself directed to distribute the award. (289) The award itself should not be
distributed without permission of the parties. (290) There is no longer a general
obligation to deposit an award with a judicial authority in the Seat to allow for
enforcement. The New York Convention changed this historical position.

16.12.2. Form of Delivery

Some laws and rules indicate that an original of the award must be sent to each
party. (291) In these circumstances, enough originals must be signed so that all can
receive an appropriate document. (292) Others simply provide that copies are to
be delivered. (293) Some are silent, in which case it would be better to provide
originals.

As to the method of delivery, attention should again be given to any special

requirements in the lex arbitri, rules or party agreement. In some cases
intergovernmental agreements on service of judicial and extrajudicial documents
may apply. (294) Van Houtte argues that this should apply to arbitral awards
notwithstanding that these are not expressly mentioned because they should be
seen as coming within the phrases ‘extra-judicial document’ and ‘civil and
commercial matters …’ It is also arguable that an award is a ‘judicial document’
for the purpose of the regulation. Germany has taken this view, but not other EU
page "1327" Member States. (295) Arbitration institutions are not transmitting
agencies for the purpose of the regulations.

Some legal systems might require an award to be translated in the language of

the country where the award is served or a language understood by the
addressee. (296) Care should be taken where the parties agree on a different
manner of delivery than is required under the arbitration statute as it could have
implications for its legal effect, although there may be principles of waiver were
a party to try and rely unfairly on such an agreement. (297) Even if informal
delivery is agreed upon, a tribunal should ensure that it complies with any
delivery obligations that commence relevant time periods for challenge. (298) In
some cases a formal time period will simply not run if a required mode of
delivery has not been adopted. An example would be where the parties agreed to
receive the award electronically where the arbitral statute does not contemplate
this mode of delivery.

Attention also needs to be given to guerrilla tactics and the possibility that the
losing party may seek to evade notification or at least argue that notification
never occurred. In such circumstances it would be appropriate to ensure the best
possible evidentiary record such as registered mail.

16.12.3. Time for Delivery

As noted, the time of delivery is important in establishing starting time periods

for requests to correct or interpret, requests for additional awards or annulment
proceedings. (299) Delivery of the final award is also generally the time at which a
tribunal becomes functus officio. (300) An award should be delivered to all parties,
including the institution where one is involved. It is then important to determine
the exact date on which this can be said to have occurred.

As is often the case with time periods that have serious repercussions, attention
needs to be given to concepts such as communication and delivery. It is
necessary to indicate how a factual determination will be made in the event of a
dispute and whether any deeming provisions apply. Older rules simply refer to
the notion of receipt without any further guidance. That is not the case with
UNCITRAL Rules 2010 Article 2 and ICC Rules 2012 Article 3 which have detailed
tests. It should be understood that because of the possibility of differing times of
delivery page "1328" and receipt, different time periods may apply to different
parties. (301) There may also be conflicts questions as to the timing of receipt.
While the arbitral statute in the Seat may specify a term such as ‘receipt’, it does
not necessarily purport to apply domestic concepts in the Seat. (302) Receipt
should be actual receipt, not deemed receipt, unless there are specific statutory
or rule provisions in that regard.

16.12.4. Advance Notice

In some circumstances, the parties may wish to know the outcome before a
reasoned award is provided. Given the prevalence of emails and faxes, tribunals
will often provide advance notice of an award prior to formal delivery of a signed
copy. That is an accepted practice but should not trigger the time periods for
correction, interpretation or annulment. This could be made clear. Advance
notice of the decision should not be seen as breaching the confidentiality of the
deliberations. (303)

Any such announcement should only come after the deliberations are complete
and the tribunal has identified the reasoning to be employed in the written
award. The ICC takes the view that because of the required scrutiny by the
International Court of Arbitration, prior notification should only be given with
the consent of both parties. (304)

16.12.5. Publication of Awards

There is a difference in approach between various institutions as to whether to

assist in the publication of awards. The primary rule is that these are confidential
and may only be published with the permission of the parties. The ICC does
publish redacted versions of some awards. Article 48(5) of the ICSID Convention
indicates that awards may not be published unless both parties agree. (305)

16.12.6. Registration

While it remains rare, some statutes require arbitration awards rendered within
the Seat to be deposited or registered with a notary or local court. (306)

page "1329"
16.12.7. Fees as Conditions of Service

Once an award is completed and available an arbitrator may be entitled to a lien

on the award to secure payment of fees or costs of the arbitration. Poudret and
Besson suggest that notification may only be withheld pending payment of fees
‘if recognised by law, the arbitration rules or the parties' agreement’. The
authors do not consider that the contract with the arbitrator implies
simultaneous performance. (307) That may well be so but that does not indicate
whether one could imply a term granting arbitrator discretion to withhold the
award in preference to calling for the arbitrator to sue on the contract. It would
be better to specify this right in any contract. Some arbitral rules expressly deal
with whether the tribunal can withhold delivery of the award until payment of
fees has been effected. An institution may generally withhold notification until
such time as its costs have been provided.

16.13. Further Actions, Correction, Interpretation and Additional Awards

16.13.1. Introduction and General Principles

While rules typically distinguish between corrections, interpretations and

additional awards and while there is a history of significant debate between
scholars and practitioners from differing legal families on these matters, the
policy issues are largely identical. Hence, this section deals with some general
issues pertaining to all forms of further action. Subsequent sections add specific
commentary as to each power. There are questions as to whether these are
inherent powers or are dependent on an agreement of the parties or powers to
that effect in the statute or rules. There may also be an issue as to whether the
tribunal has lost its jurisdictional mandate through some statutory or
contractual time limit over its powers. This may be more of a concern with
additional awards as these have added dispositive effects. The concern may not
as readily apply to corrections and interpretations which merely explain or
rectify decisions already made within the time period and do not add further
items of relief. (308)

An arbitrator's duty to complete the mandate and act with due care suggests that
if a particular claim was not addressed, an application should be permissible to
invite the tribunal to do so. If an obvious clerical error has been made, the duty of
care has not been satisfied and it would be better to have this rectified efficiently
rather than undermine the entire award or invite a court not privy to the
proceedings to take action in the circumstances. The same is true where
interpretation is concerned. Hence, the emerging view is to provide such powers
within arbitral page "1330" statutes and rules. Key policy issues in the drafting of
such provisions include whether the right can be varied by agreement, the extent
to which a tribunal can undertake such actions on its own volition, applicable
time limits, whether the decisions both positive and negative constitute awards
and can be separately challenged, whether the tribunal can charge further fees
and the ramifications if a tribunal does not accede to the request.
It is entirely possible that an application for correction, interpretation or
addition may flow from a corrective decision itself. In principle there is no
reason why a corrected or interpreted award or additional award should not also
be subject to further requests for correction, interpretation or addition. (309)
Because a correction, interpretation or addition is a separate determination by
the tribunal, it is conceivable that it could be separately challenged by way of
annulment or by way of challenging enforcement.

If there is a multi-person tribunal, the decision whether to make a correction,

interpretation or addition should be based on the same deliberation obligations
and the same majority voting powers as apply generally. There is also the
question of the appropriate time limits for completion of these tribunal tasks of
correction, interpretation or addition. Poudret and Besson suggest that the time
limits tend to be too short given that there is often a need for adversarial
proceedings and proper notification to the parties affording them an opportunity
of presenting their views. (310) Some national committees objected to the
tightness of the time limits when Article 29 was first introduced into the ICC
Rules 1998. (311)

There is also the question of whether the time limits are mandatory or can be
extended or overlooked. It seems reasonable that if a time limit for correction
has elapsed, either in terms of a late application or tribunal determination, a
competent court should, if necessary, be able to rectify errors that are obvious on
the face of the award. Given that many commentators are of the view that the
power to correct is inherent in any event, they might suggest that the inclusion of
an express power in a statute or rules together with a time limit, should see the
latter as directory, without aiming to set the outer limits of competence. It is of
course always preferable to comply or seek agreed extension.

Rules as to correction, interpretation or addition simply refer to awards and do

not seek to distinguish between interim, partial or final awards. Nevertheless, if
it is truly an award, an interim or partial determination should be subject to such
applications. This also means that the time limits should run from that stage,
although that may depend on the wording of particular rules. From a policy
perspective, it would be inappropriate for there to be a mismatch between the
page "1331" entitlement to challenge the partial award and the entitlement to
seek its correction or interpretation. (312)

16.14. Correction or Rectification of Awards

16.14.1. The Power to Correct

Virtually all legal systems allow for corrections of typographical or

computational errors in awards. The power to correct awards would first be
determined under the lex arbitri. (313) Almost all procedural rules also contain
clear provisions allowing for corrections of an award. (314) To the extent that
provisions in rules vary from the lex arbitri, they ought to take precedence unless
the latter is seen as containing a mandatory norm. If the institutional rules give
the due process right to correct, then the better view is that they maintain the
entitlement and at most vary the timelines. Care would need to be taken where
there is a mismatch between time periods. It obviously makes sense to be safe
and comply with the shorter of the two time limits, whether that is in the
institutional rules or the lex arbitri.

Article 33 of the UNCITRAL Model Law provides for a thirty-day time period for
such a request for correction. The time runs from the receipt of the award and
applies unless the parties have agreed upon another period of time. It goes on to
state that the interpretation shall form part of the award. Errors can be fixed on
the tribunal's own initiative. (315) Article 38 of the UNCITRAL Rules 2010 also
provides a thirty-day time limit after receipt of the award for a party to request a
correction, with a further forty-five days for the tribunal to comply if it believes
the request is justified. Alternatively, the tribunal itself may make a correction
within thirty days after ‘communication of the award’. Care should be taken with
the possible differences in timing between ‘receipt of the award’ by a party and
‘communication of the award’ by the tribunal, although these could be
interpreted to be identical. (316)

page "1332"

Even where the lex arbitri and rules are silent, it would also seem reasonable to
imply consent to changes of obvious errors as an inherent element of the parties'
consent to arbitration, or arising via the duty to complete the mandate and/or
the duty to act with due care. There has been a lively debate amongst scholars on
this point. Poudret and Besson suggest that if the rules are silent the power can
be implied from the will of the parties to exclude the jurisdiction of the courts on
such matters. (317) The authors also believe that such a power is a general
principle of civil procedure. (318) The power to correct is described by Moreau as
a ‘residual power’. (319)

The view that this is an inherent right is supported by the approach in

Switzerland where there is no express power in the Swiss Private International
Law but this has occurred in a number of cases and Swiss courts and scholars
have accepted that the power should be seen as inherent. (320) There are
differences in view in various legal systems as to whether there is statutory
power to refer matters back to a tribunal where the award is infra petita. (321) The
inherent entitlement was also supported by ICC practice prior to an express
power of correction being included in the 1998 Rules. (322) While that should be
so, it is preferable for a national statute to set out appropriate time limits as
specific limits would not be easy to discern under broad implied consent
analysis.

There is also some debate as to whether parties are entitled to exclude a

tribunal's power to correct errors in an award. General party autonomy might
suggest that this should be possible. (323) A contrary argument might be that the
parties have awarded a tribunal a mandate that comes with an inherent right to
correct obvious errors. If the parties sought to prevent a tribunal after the
appointment stage, the tribunal could also form the view that a subsequent
agreement page "1333" cannot change the original mandate. Conversely, if the
tribunal accepted the appointment knowing that the parties did not want any
corrections, party autonomy ought to prevail. Born suggests that party
agreement to bar corrections should be given effect, but also argues for an
inherent power on the basis that ‘it is contrary to basic conceptions of
procedural fairness for a mistaken award to be given binding effect’. (324) To the
extent that the latter proposition is valid, it would seem to be a mandatory norm
of procedure that overrides the first proposition.

16.14.2. The Ambit of the Power to Correct

It is important to limit the matters that can be addressed in a correction request

to stop parties seeking to effectively make an appeal application calling for a
change of reasoning and result. Carefully circumscribing the nature of permitted
corrections could address the type of corrections or their effect. The Model Law
chooses to describe the type of corrections. Article 33 limits the right to requests
‘to correct in the award any errors in computation, any clerical or typographical
errors or any errors of similar nature …’ Another approach which has been
suggested is to limit corrections to matters that do not alter the meaning of the
award. That would be an undesirable approach as it would be virtually
impossible to not see this impacting upon any corrected error. (325)

Courts have uniformly interpreted the basis for correction narrowly to prevent
aggrieved persons trying to re-argue key elements. (326) In Panacaviar, SA v
Islamic Republic of Iran, (327) the aggrieved party argued that the tribunal had
mischaracterised the nature of the underlying dispute and sought a ‘correction’
in relation to such alleged mischaracterisation. The tribunal rejected the request,
considering that the respondent was instead seeking to re-argue the case and
disagree with the conclusions. On this basis, if for example the tribunal
completely misread the terms of the parties' contract, this cannot be corrected.
The only possibility is to seek to annul the award or block enforcement but this
can only occur on limited grounds that would not cover an error in analytical
reasoning.

page "1334"

Typical examples of simple errors include putting the decimal point in the wrong
place, inverting the description of parties (describing claimant as respondent), or
an error of calculation where the methodology is properly articulated (e.g.,
identifying the percentage profit share but making an inappropriate calculation
into applicable currency of the award). (328)

One question is whether the power to correct awards could deal with form
omissions. Using the UNCITRAL Rules 2010 as an example, this raises the
ejusdem generis principle because of the reference to specific types of errors and
then ‘any error of a similar nature’. Peter Sanders makes the point that the
generic phrase is difficult to understand as the three stipulated areas of errors in
computation and clerical and typographical errors are not of a similar nature.
(329) It is only fair and efficient to see a correction power as covering an error of

form that would otherwise render the award subject to challenge where
correction does not alter the true dispositive intent. The English Arbitration Act
1996 refers to the power on application or on a tribunal's own initiative, to
‘remove any clerical mistake or error arising from an accidental slip or omission
or clarify or remove any ambiguity in the award.’ (330) The time limits are twenty-
eight days for the application and for the correction. It is not clear what is
intended by the phrase ‘accidental slip or omission’ and whether this goes
beyond mere clerical errors to matters of analytical error. (331)

16.14.3. Due Process and Corrections

Unless the correction is clear and obvious, the tribunal should give the parties an
opportunity to have an input. An example might be complex calculations of
damages where one party has noted an arithmetical error. The tribunal might
revise the computations but it still could be the case that the revised figure is
itself inaccurate or would at least benefit from party input. If a request for a
correction is made, it seems appropriate to hear the other party's view as is the
case with most unilateral applications for tribunal orders and awards. (332) While
the other party's view should be sought, it would generally be unnecessary to
hold a hearing for such a purpose (333) page "1335" and such a request should not
allow a recalcitrant party to delay the process. Situations might be more complex
if a tribunal has had to be reconstituted between the time of an award and the
time of a requested correction.

16.14.4. Form Requirements and Challenges

Article 33(5) of the Model Law indicates that the form and content obligations
under Article 31 apply to a correction of an award. This should also be so in
relation to any clerical errors. The English Arbitration Act 1996 calls for an
application for rectification to first be made to the tribunal and only allows for
judicial challenge after the section 57 options have been exhausted. (334) A court
has held that a direct court challenge can be brought on matters that the tribunal
could not have corrected. (335)

A positive decision on rectification or correction is itself an award open to

challenge. (336) A refusal to rectify an award is considered by French courts to be a
decision appropriate for challenge. (337) The Note by the Secretariat of the ICC on
Correction and Interpretation of Awards considers that for corrections, where a
request is acceded to, it should take the form of an Addendum while a refusal
should be taken as a Decision. (338) Both addenda and decisions need to have
reasons and be presented for scrutiny to the ICC Court.

It is not immediately clear whether a request for a correction suspends the time
limit for a challenge to the award. The Italian Code of Civil Procedure Article
828(3) expressly holds that it does not suspend the time limit. (339) The same
view was taken by the Swiss Federal Supreme Court in relation to a request for a
correction under Article 29 of the ICC Rules 1998. (340) On this view, a separate
challenge could be made against a rectified award but only against the corrected
parts. A well-drafted lex arbitri will make it clear that any time limit for setting
aside the award does not run pending a request for correction, interpretation or
an additional award. That is made clear under Article 34(3) of the UNCITRAL
Model Law. Poudret and Besson wisely suggest that if the law is not clear, it
would be advisable to simultaneously file a request for correction and an
application to set aside the award within the stipulated time frames. (341)

page "1336"

16.14.5. Timing of Corrections

Time limits in governing statutes are typically twenty-eight or thirty days. (342)
Where there is no express provision, some commentators have recommended a
particular limit within which to apply for correction, although it would be
difficult to set a hard and fast rule. (343) From a policy perspective, some deadline
supports finality and supports the view that a party should be vigilant in
analysing an award to look for errors that apply against its interests. However, it
might be argued that any error identified up until the time the award must be
honoured should be allowed on the basis that no harm is caused in such
circumstances. If the entitlement to correct is under a general inherent power,
then the tribunal ought to be able to look at all relevant circumstances and
decide on a case-by-case basis.

16.14.6. Other Modes of Correction

Another approach to correction is through scrutiny by institutions. Scrutiny

under the ICC Rules is much broader than the correction powers, as Article 33
allows for points of substance to be drawn to the attention of the tribunal. Other
possibilities for correction might include remittance by a court (344) or correction
by the court itself when the tribunal cannot be reconstituted. (345) Some US courts
have also held that the power to correct errors is inherent. (346) The US situation
may be problematic in that it may deny arbitrators their own entitlement to
correct the award. Born criticises the breadth of the section 11 criteria,
suggesting that this invites inappropriate applications and has allowed some
courts to correct matters of substance. He cites as an example Eljer Mfg Inc v
Kowin Dev Corp, (347) but the facts may not be controversial.

page "1337"

16.15. Interpretation of Awards

16.15.1. Power to Interpret

Some national statutes expressly allow for interpretation. (348) Where national
legislation is silent, courts will typically see the power to clarify or interpret as
inherent, may allow a court to refer back to the tribunal or in some cases may
allow the court itself to resolve the ambiguity. (349)

If the arbitral statute and rules are silent on the question of interpretation, the
better view is that this is either an inherent adjudicator's power or an implied
element of a duty to complete the mandate, although not all support this view.
(350)The argument in favour is that a tribunal has to provide dispositive
directions that are capable of being implemented and enforced. To the extent
that they are so ambiguous as to lead to disputes about the implications of the
award, the tribunal should simply be able to explain what it meant. If an
ambiguity is such that the parties cannot understand how to implement the
award, the better view is that a tribunal nonetheless has an inherent power to
resolve the ambiguity. (351) ICC practice was to allow for interpretation of awards
even prior to an express power to do so being included in the 1998 Rules. (352)
Lew, Mistelis and Kröll also seem to support an inherent power. (353) The inherent
right to interpret an award has been accepted by an arbitral tribunal in the
Delimitation of the Continental Shelf (United Kingdom v France). (354)

Some have still debated whether there is an inherent power to interpret awards
given that the award already has res judicata effect. (355) Poudret and Besson
point out that judges are entitled to interpret and arbitrators should have no
different powers. The key issue is not whether the award has res judicata effect
but who is page "1338" best placed to determine just what effect it has when its
terms are inherently ambiguous.

Some rules may also be limiting. The Model Law limits interpretation to cases
where the parties have agreed, (356) Article 33(1)(b) indicating that ‘if so agreed
by the parties, a party, with notice to the other party, may request the arbitral
tribunal to give an interpretation of a specific point or part of the award’. Even
this articulation does not indicate whether the tribunal itself may provide an
interpretation absent a party request. The English Arbitration Act 1996 does not
refer to interpretation but allows for clarification of ‘ambiguities’ which should
amount to the same thing. (357) Section 57(3)(a) does not call for consent of the
other party before a tribunal can be asked for clarification.

16.15.2. The Ambit of the Interpretation Power and Due Process Obligations

Poudret and Besson suggest that interpretation ‘means to restore the true
meaning of the original decision where it was badly expressed in the holdings, be
it because the latter seem to be at odds with the reasons or contain obscurities or
ambiguities. Interpretation does not modify or supplement the original decision’.
(358) Merkin goes further and considers that the power would allow a tribunal ‘to

rethink inconsistencies in the award’. (359)

In any event, interpretation should be limited to ambiguities in the rulings or the

means of execution, (360) and not simply address the reasoning underlying the
final dispositive provisions. (361) On this view the notion of an interpretation is
instead more akin to a clarification. (362) In some cases a final award might also
interpret a prior ambiguous partial award. (363) On this approach it does not seem
that a tribunal would need to afford the parties an opportunity of being heard on
the page "1339" matter other than simply articulating what aspects of the award
they find ambiguous in the context of implementation. Unlike a correction, which
might require actual recalculations by a tribunal, an interpretation is merely a
better explanation of what the historical decision actually was and does not seek
to change the nature of the remedies awarded.

However, in some cases the remedy is so unclear that the interpretation

nevertheless has to select between two conflicting views. In such cases, requests
for interpretation are really more akin to requests for completion of the
mandate. For example, if a tribunal awards interest but does not indicate
whether it is simple or compound interest, it is not clear on the face of the award
that the tribunal actually turned its mind to this question in the first instance.
Similar issues arise where there might be an award of royalties without an
indication of the time period for calculation or geographical coverage. (364) If this
is the case, an opportunity to be heard may be necessary.

In some cases there will be an important question of classification as to whether

the award simply needs interpretation or whether there is a gap. In the latter
event, that can be a ground for setting aside the award where the tribunal has
not completed the mandate. The same may be so where it is a question of
interpretation as to whether the tribunal has gone beyond its mandate. Such a
situation might be dealt with via remittance to the tribunal or instead, under a
presumption that a tribunal has not purported to exceed its mandate and hence
annulment and enforcement courts should read down the dispositive parts
under that presumption.

A request for an interpretation will typically be rejected if it is not necessary to

implementation or enforcement of the award. A tribunal also needs to ensure
that such requests are not used to unduly delay implementation or enforcement.
Peter Sanders recommended that the UNCITRAL Rules be amended so that
requests for interpretation cannot be used for delaying purposes or to obtain
revisions under the guise of requests for interpretation. (365)

16.15.3. Form of Interpretation

Again the interpretation shall form part of the award and the form requirements
of provisions such as Article 31 of the UNCITRAL Model Law must be met. The
Note by the Secretariat of the ICC on Correction and Interpretation of Awards
considers that for interpretations as well as corrections, where a request is
acceded to, it should take the form of an Addendum while a refusal should be
taken as a Decision. (366) Both addenda and decisions need to have reasons and be
presented for page "1340" scrutiny to the ICC Court. As with corrections,
situations might be more complex if a tribunal has had to be reconstituted
between the time of an award and the time of a requested interpretation.

16.15.4. Time Limits for Interpretation

There is a difference in time limits as to interpretation of an award between

various rules and statutes. Under Article 38 of the UNCITRAL Rules 2010 and
Article 33 of the Model Law the tribunal has a thirty-day time period, although
under Article 33(4) of the latter, the tribunal may extend if necessary the period
of time for the making of a correction, interpretation or an additional award.
Once again, it may be sensible to comply with the shorter period for safety
purposes, although provisions in the rules selected would be expected to prevail.

16.15.5. Challenge

An English case has held that a failure to respond to a requirement to interpret

an award can constitute a serious irregularity for challenge purposes. (367) That
would obviously depend on the circumstances.

16.16. Additional Awards

Because additional awards must have added dispositive effect, they need to be
rendered within any applicable time period that limits the tribunal's powers. (368)
They also need to be promulgated subject to the due process rights of the parties
to be heard and the deliberation obligations on the tribunal. Article 33(3) of the
UNCITRAL Model Law indicates that unless otherwise agreed by the parties, a
party may with notice to the other, request an additional award as to presented
claims that were omitted from the award. (369) There is a thirty-day time limit for
the request and if the tribunal considers it justified, sixty days for the making of
the additional award. Article 39 of the UNCITRAL Rules 2010 allows for a request
for awards on claims presented but not decided within thirty days of receipt. (370)
If the tribunal considers the request justified it has sixty days to comply but may
extend if necessary.

page "1341"

Neither the 1998 nor the 2012 ICC Rules include such a provision. However, the
ICC accepts that this is at least permissible where it is provided for by the lex
arbitri. (371) A specific rule for additional awards was not thought appropriate to
the ICC Rules simply because of the court scrutiny, which would hopefully
identify such matters in any event. (372) While the hope of the ICC Working Party
was that scrutiny would obviate the need for additional awards, no one is
infallible and in any event the Secretariat and the court have not had the benefit
of being at the proceedings. The parties may have orally agreed to clarification of
the ambit of a term of reference, which might then have not been adequately
addressed in the award. This omission would not be obvious to someone
comparing the terms of reference with the actual award to determine whether all
matters were adequately covered. There is also the possibility of human error. It
would be unfortunate if the result was that a need for an additional award not
picked up by the court would force an annulment application. That would be
particularly problematic in terms of those aspects of the award which were
properly concluded.

There is generally no express provision in rules for the tribunal to provide an

additional award of its own volition. If the tribunal has simply forgotten to cover
something that should have been dealt with in an award it declares to be final,
one view is that it would be. functus officio under a provision such as that of the
Model Law. That should not be so where the rules or statute expressly indicate
that an additional award may be rendered after an award purporting to be a final
award. A contrary view is that it should be an inherent power or an implied term
in its duty to complete its mandate. In any event, it is subject to contrary
agreement by the parties who may, therefore, bar the entitlement to additional
awards. Even a later agreement by the parties should not be a problem in terms
of imposing on the tribunal as it would only be reducing the potential workload
of the tribunal not adding to it and hence could not be seen as an improper
interference with contractual rights agreed to on appointment. (373)

Civilian jurisdictions have tended to allow tribunals to subsequently deal with

omissions without any express basis in the statute or rules, while some common
law jurisdictions have considered the tribunal to be. functus officio. The better
view should be to see the functus officio doctrine as having exceptions to correct
mistakes, clarify ambiguities and decide matters which have been inadvertently
omitted. (374) If a tribunal itself cannot render an additional award unless
requested to do so by a party, this may unfortunately mean that the award might
be set aside if a party preferred that approach, at least in a jurisdiction that does
not require a request for an additional award as a gateway to annulment.

page "1342"

A further policy issue is the implication for any annulment or challenge

proceedings where such rectification or interpretation applications have been
made or can be made. It would be wasteful to force parties to take appeal
proceedings where one would obviate the need for the latter. Section 70(2)(b) of
the English Arbitration Act 1996 indicates that a challenge or appeal cannot be
brought until remedies provided under section 57 as to correction or additional
awards have first been exhausted. (375) Nevertheless in legal systems where there
is no such express provision, the parties may need to simultaneously file to
ensure that their rights are clearly protected. (376) In France a failure to decide on
an aspect of the claim does not constitute a breach of the mandate or a ground
for setting aside, hence the proper remedy is to require an additional award. (377)

There is a difference between situations where a tribunal has simply failed to

address a matter that should have been dealt with and instead a situation where
a tribunal has impliedly rejected a claim but not addressed it in the award. As to
the latter, because of the requirement to provide reasons, a tribunal should
always address rejected claims and explain the reason for the rejection. Due
process and natural justice issues should commonly arise with additional awards
if the tribunal has not already ensured that each party has had an adequate and
equal opportunity to present its argument on the matters not covered in what
purported to be the final award. If a tribunal needs to afford parties an
opportunity to make submissions in relation to an additional award, it would
seem reasonable to impose strict time limits so that the tribunal itself can meet
its required deadlines, although it is equally the case that where the tribunal has
a power to extend a time period, it should ensure that both the tribunal and the
parties have an adequate opportunity to present relevant material. (378)

Additional awards are not merged with previous awards and may thus be the
subject of discrete annulment or enforcement applications. They again must
comply with all of the form requirements for an award.

page "1343"

16.17. Remission of Awards

Another mechanism for follow-up action by a tribunal is where an annulment

application has been made and the relevant court has a power to remit the
award to the tribunal. Article 34(4) of the UNCITRAL Model Law indicates that:

The court, when asked to set aside an award, may, where appropriate and so
requested by a party, suspend the setting aside proceedings for a period of time
determined by it in order to give the arbitral tribunal an opportunity to resume
the arbitral proceedings or to take such other action in the arbitral tribunal's
opinion will eliminate the grounds for setting aside.

This provisions does not establish a typical remittance power where a court
might direct a tribunal to behave in a particular way. Instead it is merely a power
to suspend setting aside proceedings and must be based on a presumption that
the tribunal has a separate entitlement to take some further action that might
eliminate the grounds for setting aside.

16.18. Fees and Expenses for Corrections, Interpretations or Additional

Awards

Another question is whether a tribunal may seek fees or expenses where

correction, interpretation or additional awards are involved. The policy issue of
concern is that the need for correction, interpretation or additional awards
might often arise because of tribunal error.

Where an institution sets the final fee it might well take this into account in any
event. Where the parties have agreed on an hourly or daily rate, and the tribunal
has no indemnity or exclusion for negligent behaviour, if the error is clearly that
of the tribunal, seeking a fee might simply lead to a claim for reimbursement
under the contract with the parties. Some laws expressly indicate that the
tribunal is not entitled to receive fees for such purposes. If a case is clearly one of
tribunal error, one would normally expect that a tribunal would not seek
payment for that purpose. More problematic are cases where the tribunal
believes the award is sufficient but the parties have sought clarification
nonetheless. It will obviously be a question of judgment on a case-by-case basis
whether a tribunal could justify fees in such circumstances. Article 40(3) of the
UNCITRAL Arbitration Rules 2010 makes clear that where interpretation,
correction or completion arises the tribunal may charge specified costs but no
additional fees. A tribunal's broad powers to award costs as between parties
should also apply to any contest in relation to correction applications. While
there is a strong argument to the effect that a tribunal ought not to be entitled to
fix errors or ambiguities and receive fees as well, (379) nevertheless Article 2(10)
of page "1344" Appendix III to the ICC Rules 2012 allows an advance to cover
additional fees and expenses where appropriate in relation to Article 29(2)
applications.

It certainly seems reasonable that a tribunal should be entitled to fees and

expenses where it legitimately rejects such an application. The difficulty arises
where the tribunal accepts that it has made a typographical or computational
mistake or has failed to clarify an ambiguous matter or has failed to deal with a
claim. A blanket rule either way would be problematic. For example, where
additional awards are concerned, if the tribunal should have dealt with a matter
and did not, the original fee may be less than it ought to have been. Hence, when
the tribunal completes the mandate, some entitlement to extra fees seems
appropriate. The same may be true with certain interpretations that in reality
invite a tribunal to address a particular issue such as whether interest awarded
should be on a compound or simple basis. Even here there may need to be a
legitimate discount for the extra costs occasioned to the parties by reason of the
tribunal's oversight.

16.19. Effect of an Award

16.19.1. Impact of the Award on an Arbitrator's Powers

As noted at the outset, the key effects of an award are that it is enforceable as
between the parties, it has res judicata effect in relation to future actions, parties
themselves have duties to comply and not to act in a contrary manner, and
potential rights to seek correction, interpretation, annulment or enforcement are
triggered.

Once a final award is rendered, another important question is as to the impact on

the tribunal's own powers. Generally speaking, the rendering of a final award
renders a tribunal functus officio, save for express correction and interpretation
powers, although the concept is to some degree subject to debate and is
discussed further in the following section. The functus officio concept may have
important practical implications where a tribunal omits to cover certain aspects
that ought to have been dealt with. An example would be interest and costs. Save
for express powers to provide additional or corrective awards, the possibility
exists that a tribunal which purports to render a final award not covering all
matters can no longer repair the oversight. A converse argument is also highly
tenable. A tribunal has a duty to complete the mandate. The parties never
consented to an award that was infra petita. Consequently, once the oversight is
noticed, the tribunal should be seen as having power to complete that which it
was originally asked to do. There are thus desirable exceptions to the functus
officio principle which can be provided for in a range of ways. In addition to the
inherent powers argument, most rules allow for correction, interpretation and at
times revision of an award. Second, an annulment or enforcement court might
remit the matter back to a tribunal for reconsideration. Third, the tribunal's own
award might provide the parties with liberty to apply for variations of the orders,
for example where one party fails to respect a particular direction in the award.

page "1345"
Where arbitral statutes address the issue, they typically follow the structure of
Article 32 of the UNCITRAL Model Law. This provides that ‘the arbitral
proceedings are terminated by the final award’ subject to the provisions relating
to correction, interpretation and additional awards. Article 32(3) indicates that
‘(t)he mandate of the arbitral tribunal terminates with the termination of the
arbitral proceedings’. (380) Article 32(2) also indicates that the proceedings may
be terminated by order of the tribunal without necessity for a final award where
the claimant withdraws the claim unless the respondent objects and the tribunal
agrees that the respondent is entitled to an award, the parties agree on
termination, or the tribunal finds a continuation is unnecessary or impossible.

Where ICSID arbitration is concerned, an ad hoc committee may consider a

request for nullification under Article 52(1) of the ICSID Convention. If the award
is annulled, either party can request that the matter be brought to a new
tribunal. (381) Article 51(1) of the ICSID Convention allows reconsideration of an
award based on subsequently discovered evidence ‘of such a nature as to
decisively affect the award’. An UNCITRAL tribunal under Judge Schwebel
considered that this was an inherent element of a tribunal's power. (382)

While an arbitrator would generally be seen as functus officio after rendering an

award and after the appropriate time periods for rectification are concluded, this
only relates to the adjudicatory function itself. An arbitrator certainly has
ongoing duties of confidentiality. A related question is whether there are then
any commensurate rights or powers in relation to any matters over which there
is a duty. Schwartz alludes to a complex dispute as to whether an arbitrator was
entitled to release material in its possession at the request of a party. (383)

In addition to duties of confidentiality, there may be duties to conserve

documents, assisting the parties in complying with further formalities, (384) the
duty to correct errors, interpret where necessary and at times supplement an
award and the duty not to interfere with or adversely affect enforcement of an
award. Most do not see any positive obligation to actively support enforcement,
although this has been argued by Bedjaoui in the context of helping sovereign
States to come to terms with page "1346" adverse determinations in investment
arbitrations. (385) Identifying such duties after the award is completed is further
complicated where there are multi-member tribunals. Added to the conceptual
question there is also the problem of what remedy could possibly apply if an
arbitrator breaches post-award obligations. Removal or loss of fees cannot be
applicable. Damages would be difficult to prove even where immunities do not
apply.

One residual uncertainty is the degree of control a tribunal may have over
certain forms of relief. In some cases, a tribunal might see valuable practical
reasons to try and maintain such an ongoing function but would need to consider
whether it is unable to do so, being functus officio. This is discussed further in the
following section.
16.19.2. The Nature of the Functus Officio Concept and Implied Exceptions

The logic behind the concept of functus officio is that if parties consent to a
tribunal having a mandate to do certain things, once they are completed the
jurisdictional authority no longer exists. The notion that a tribunal is functus
officio has a strong relationship to other concepts clearly applicable, namely that
an award gives rise to res judicata effects between the parties and that awards
are binding. The notion of functus officio in common law is simply to the effect
that an arbitrator may not revisit the merits of an award once it has been issued.
(386) Because of this, the concept must be looked at alongside the doctrine of

finality. (387) The functus officio concept would also mean that even where partial
awards are rendered, that is the final word of the tribunal on that matter and the
tribunal cannot change its view even if the balance of the proceedings are
ensuing. As noted previously in relation to Gulf Petro Trading Co v Nigerian
National Petroleum Corp, (388) later stages can still impact on earlier ones.

Jurisdictions such as the US that do not expressly refer to anything akin to a

functus officio doctrine, still support it under common law principles, with
similar exceptions to deal with mistakes and ambiguities. (389) All legal systems
would agree that while a tribunal may correct obvious errors, fill gaps and
resolve ambiguities, page "1347" the functions are not there to support a change
in view that a tribunal might have come to on further reflection. The policy
reasons against this are that the parties wanted a speedy and final resolution of
their dispute, there would be too much concern with the possibility of ex parte
submissions after the award trying to have the tribunal change its mind and the
concern that if the tribunal felt disposed to change its mind on one occasion, why
might it not do so again on even further representations being made.

There is a debate as to whether a tribunal nonetheless has an inherent power to

revise an award in cases of fraudulent or similar behaviour. A number of
tribunals and commentators have considered the possibility and not rejected it
outright, although language used is very mindful of the question of whether the
tribunal still has potential jurisdiction. The French Cour de Cassation considered
that this would be possible ‘if the arbitral tribunal remains constituted after the
rendering of the award (or can be reconstituted)’. (390) If the tribunal is aware of
fraud prior to becoming functus officio, it seems reasonable to allow it to reopen
its proceedings. All arbitration agreement are properly presumed to be agreed to
in good faith. This would suggest an implied term that if one party acts
fraudulently and misleads the tribunal, at least as long as the tribunal has some
remaining duties, these can include correction of determinations induced by
fraud. (391) Brower supports the view that the tribunal itself ought to be able to
review such awards. (392)

page "1348"

1 For example, ICC Rules 2012 Art. 34(6).

2 If a party is aggrieved by a mere procedural order or determination, a protest
should be made at the time as this may be considered relevant by an
enforcement court in exercising its discretion: Nigel Blackaby et al., Redfern and
Hunter on International Arbitration, 5th edn (Oxford: Oxford University Press,
2009), 516, n. 12.
3 Aron Broches, ‘Recourse against the Award; Recognition and Enforcement of

the Award’, in UNCITRAL's Project for a Model Law on International Commercial

Arbitration, ICCA Congress Series No. 2, ed. Pieter Sanders (The Hague: Kluwer
Law International, 1984), 208–209.
4 See, e.g., Judgment of 13 April 2010, Swiss Federal Supreme Court, ASA Bulletin

28, no. 3 (2010); Judgment of 1 July 1999, Brasoil v. GMRA, Cour d'appel de Paris,
Revue de l'Arbitrage, (1999), no 3: 834; Publicis v. True North, 206 F. 3d 725. See
also Andrea Carlevaris, ‘La qualificazione delle sentenze arbitrali’, Rivista
dell'arbitrato 12, no. 3 (2002).
5 Howard M. Holtzmann & Joseph E. Neuhaus, A Guide to the Uncitral Model Law

on International Commercial Arbitration: Legislative History and Commentary

(The Hague: Kluwer Law International, 1995), 153–154 and 867–868.
6 Singaporean International Arbitration Act s. 2(1). See also Malaysian
Arbitration Act 2005 s. 2(1); New Zealand Arbitration Act s. 2(1).
7 SIAC Rules Art. 1.3.
8 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröil, Comparative International

Commercial Arbitration (The Hague: Kluwer Law International, 2003), 631.

9 James M. Gaitis, ‘The Federal Arbitration Act: Risks And Incongruities Relating

to the Issuance of Interim and Partial Awards in Domestic and International

Arbitrations’, The American Review of International Arbitration 16 (2005): 70.
10 See Judgment of 17 May 1990, Commissariat à l'énergie atomique v. Iran, Swiss

Federal Supreme Court, ASA Bulletin 8, no. 3 (1990): 288.

11 Judgment of 1 July 1999, Brasoil v. GMRA, Cour d'appelde Paris, Revue de

l'Arbitrage (1999): 834. For other cases where designation by the tribunal was
not determinative see Resort Condominiums International Inc. v. Bolwell (1993)
118 ALR 655, in Albert Jan van den Berg (ed.), Yearbook Book of Commercial
Arbitration Volume XX (The Hague: Kluwer Law International, 1995), 628, para.
40; PT Asuransi Jasa Indonesia (Persero) v. Dexia Bank SA [2006] SGCA 41, para.
70; Tang Boon Jek Jeffrey v. Tan Poh Leng Stanley [2001] 3 SLR 237.
12 Custodio O. Parlade, ‘Arbitral Awards’, in The Asian Leading Arbitrators' Guide

to International Arbitration, ed. Michael Pryles & Michael Moser (Huntington:

JurisNet LLC, 2007), 377.
13 Verbiese v. SEE, Judgment of 10 November 1995, Cour d'appel de Paris (Paris

Court of Appeal), Revue de l'Arbitrage (1997): 583-595.

14 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 516.

15 There is some debate on this, discussed below.
16 The Philippines legislation uses a test of whether it resolves an ‘issue in

controversy’, which at least avoids the use of the term ‘substantive’ with its
comparative law overtones, but which may be interpreted too broadly, as on
plain meaning it could be asserted to apply to some procedural disputes:
Philippines Alternative Dispute Resolution Act 2004 s. 3(f).
17 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 642.

18 Ibid., 645.
19 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law
International, 2009), 2357.
20 Ibid., 2358.
21 It also rejected the defence on the basis that signature by one arbitrator alone

was insufficient. Publicis Communication & Publicis SA v. True North

Communications Inc., 206 F. 3d 725 (7th Cir, 14 March 2000).
22 Philippe Pinsolle, ‘La Convention de New York peut-elle permettre l'exequatur

d'un order par lequel un Tribunal arbitral enjoint a une partie de produire des
pieces’, Revue de l'arbitrage (2000): 657.
23 Jean-François Poudret & Sébastien Besson, Comparative law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 643.

24 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2358.

25 Note, however, that the New York Convention 1958 does not expressly
require an award to be final, but instead that it be binding.
26 Custodio O. Parlade, ‘Arbitral Awards’, in The Asian Leading Arbitrators' Guide

to International Arbitration, ed. Michael Pryles & Michael Moser (Huntington:

JurisNet LLC, 2007), 378.
27 UNCITRAL Model Law Arts 35 and 36.
28 See, e.g., Judgment of 7 July 1995, Paris Cour d'appel, Revue de l'Arbitrage

(1996): 270.
29 For a detailed discussion see James M. Gaitis, ‘The Federal Arbitration Act:

Risks and Incongruities Relating to the Issuance of Interim and Partial Awards in
Domestic and International Arbitrations’, The American Review of International
Arbitration 16 (2005): 1.
30 Judgment of 23 October 2003, United States District Court, Northern District

of Texas, Dallas Division, Civil Action No. 3:03-CV-0406-G in Albert Jan van den
Berg (ed.), Yearbook Commercial Arbitration Volume XXIX (The Hague: Kluwer
Law International, 2004): 1114–1128.
31 Rau questions what would have happened if the partial award had been first

presented for confirmation in the US. Alan Scott Rau, ‘Provisional Relief in
Arbitration: How Things Stand in the United States’, Journal of International
Arbitration 22 (2005): 59, n. 264.
32 English Arbitration Act 1996 ss 38, 39 and 42(1); German Code of Civil

Procedure Art. 1041(2).

33 English Arbitration Act 1996 ss 42, 138; Swiss Private International Law Art.

184(2).
34 Laurence Boo, ‘Ruling on Arbitral Jurisdiction – Is That an Award?’, Asian

International Arbitration Journal 3, no 2. (2007): 125.

35 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 516.

36 PT Asuransi Jasa Indonesia (Persero) v. Dexia Bank SA [2006] SGCA 41 at para.

66.
37 The fact that the UNCITRAL Model Law treats certain situations differently

ought not to be determinative. For example, if a respondent, after losing an

arbitration commenced legal proceedings in a Model Law country, the arbitral
claimant would seek to invigorate Art. 8 of the Model Law, which should
naturally apply even if some courts might take a different view to that of the
tribunal outside of the context of enforcement proceedings.
38 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 644.

39 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2359.

40 José María Alonso, ‘Deliberation and Drafting Awards in International
Arbitration’, in Liber Amicorum Bernardo Cremades, ed. M.Á. Fernández-
Ballesteros & David Arias (Madrid: La Ley, 2010), 145.
41 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 644.

42 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2429.

43 Ameritech Services, Inc. v. Local Union No. 336, NO. 95 C-5897 WL 222439 (N.D

Ill., 30 April 1957).

44 Ibid.
45 Olsen v. Wexford Cleaning Services Corp. (7th Cir, 3 February 2005).
46 Local 36, Sheet Metal Workers Int'l Assoc, v. Pevely Sheet Metal Co., 951 F.2d

417 (8th Cir 1993).

47 See sections 16.13–17 and 16.19 below.
48 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2430.

49 For example, UNCITRAL Model Law Art. 32(1).
50 There are however many exceptions to the notion that a tribunal is functus

officio after rendering a final award. Jean-François Poudret & Sébastien Besson,
Comparative Law of International Arbitration, 2nd edn (London: Sweet &
Maxwell, 2007), 687–697.
51 José María Alonso, ‘Deliberation and Drafting Awards in International
Arbitration’, in Liber Amicorum Bernardo Cremades, ed. M.Á. Fernández-
Ballesteros & David Arias (Madrid: La Ley, 2010), 146.
52 English Arbitration Act 1996 s. 47; Swiss Private International Law Art. 188;

Swedish Arbitration Act s. 29.

53 LCIA Rules Art. 26.7; SIAC Rules Art. 1.3; Swiss Rules 2012 Art. 32.1; SCC Rules

Art. 38; ACICA Rules Art. 33.1; HKIAC Rules Art. 30.1; UNCITRAL Rules 2010 Art.
34.1; ICDR Rules Art. 27.7.
54 Previously, Art. 21.4 UNCITRAL Rules 1976 indicated that a tribunal should

generally rule on jurisdiction as a preliminary matter. This presumption was not

repeated in the 2010 Rules.
55 W. Laurence Craig, William W. Park & Jan Paulsson, International Chamber of

Commerce Arbitration, 3rd edn (New York: Oceana Publications, 2000), 362–364;
Emmanuel Gaillard & John Savage (eds), Fouchard Gaillard Goldman on
International Commercial Arbitration (The Hague: Kluwer Law International,
1999), 743.
56 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 522.

57 Howard M. Holtzmann & Joseph E. Neuhaus, A Guide to the UNCITRAL Model

Law on International Commercial Arbitration: Legislative History and

Commentary (Deventer: Kluwer Law and Taxation Publishers, 1989); 868.
58 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 633; ICC Case No. 3790 (1983), in Albert Jan van den Berg (ed.), Yearbook
of Commercial Arbitration Volume XI (The Hague: Kluwer Law International,
1986); Wintershall AG v. Government of Qatar, in Albert Jan van den Berg (ed.),
Yearbook of Commercial Arbitration Volume XV (The Hague: Kluwer Law
International, 1990).
59 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 632. Other leading
commentators agree. See Gary B. Born, International Commercial Arbitration
(The Hague: Kluwer Law International, 2009), 431; Emmanuel Gaillard & John
Savage (eds), Fouchard Gaillard Goldman on International Commercial
Arbitration (The Hague: Kluwer Law International, 1999), 743.
60 Poudret and Besson also suggest the advisability of consulting the parties:

Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 636.
61 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 523.

62 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 634.
63 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 523.

64 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 635.

65 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröil, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 634.
66 Sodifis Case, Judgment of 8 March 1988, Sociétés Sodifis v. Oiaeti, Cour de

cassation, Revue de l'Arbitrage (1989): 481 with a note by Jarrosson.

67 The Angelic Grace [1980] 1 Lloyd's Rep 288 (CA). See however Trade Fortitude

[1992] 1 Lloyd's Rep 169.

68 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 633. Conversely, see
Klaus Peter Berger, International Economic Arbitration (The Hague: Kluwer Law
and Taxation Publishers, 1993), 691.
69 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2359.

70 Ibid., 2360.
71 See, e.g., Jean-François Poudret & Sébastien Besson, Comparative Law of

International Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 640, fn. 52.
72 Transfield Philippines v. Luzon Hydro Corporation, et al., 19 May 2006,
Supreme Court, Special Second Division, G.R No. 146717.
73 See the cases cited in Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll,

Comparative International Commercial Arbitration (The Hague: Kluwer Law

International, 2003), 635, n. 29.
74 Howard M. Holtzmann & Joseph E. Neuhaus, A Guide to the UNCITRAL Model

Law on International Commercial Arbitration: Legislative History and

Commentary (Deventer: Kluwer Law and Taxation Publishers, 1989), 867.
75 JAMS International Arbitration Rules Art. 30.3.
76 James M. Gaitis, ‘The Federal Arbitration Act: Risks And Incongruities Relating
to the Issuance of Interim and Partial Awards in Domestic and International
Arbitrations’, The American Review of International Arbitration 16 (2005): 68.
Gaitis sees it in the first manner.
77 Section 14 of the English Arbitration Act.
78 Working Party on Dissenting Opinions and Interim and Partial Awards of the

ICC Commission on International Arbitration, ‘Final Report on Interim and

Partial Awards’, ICC International Court of Arbitration Bulletin 2 (1990): para. 5.
79 Ibid., para. 11.
80 W. Laurence Craig, William W. Park & Jan Paulsson, International Chamber of

Commerce Arbitration, 3rd edn (New York: Oceana Publications, 2000), 359–360.
81 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 635.
82 There is a debate as to whether interim awards are enforceable under the

New York Convention. See UNCITRAL Working Group II, ‘Preparation of uniform
provisions on interim measures of protection’, UN Doc. A/CN.9/WG 119.
83 Judgment of 10 May 2007, Oberlandesgericht Frankfurt (Higher Regional

Court of Frankfurt), Schieds VZ (2007): 278. See also Gary B. Born, International
Commercial Arbitration (The Hague: Kluwer Law International, 2009), 2360.
84 See, e.g., the Spanish Arbitration Act 2003 Art. 23. In Polydeskis Corp. v.

Transcontinental Fertiliser Co. a US District Court considered it could confirm an

order directing deposit of a sum into escrow by way of an interim measure:
(1996) WL 683629, No Civ. A. 95-0242 (E.D.Pa. 26 November 1996) cited in
James M. Gaitis, ‘The Federal Arbitration Act: Risks and Incongruities Relating to
the Issuance of Interim and Partial Awards in Domestic and International
Arbitrations’, The American Review of International Arbitration 16 (2005): 62.
85 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2023.

86 Article 17H has been incorporated in Schedule 1 to New Zealand Arbitration

Act 1996. The Hong Kong Arbitration Ordinance (2011) also provides for the
enforcement of interim measures.
87 See, e.g., ICC Rules 2012 Art. 28(1); SIAC Rules Art. 26.1; HKIAC Rules Art.

24.2; ACICA Rules Art. 28.1; SCC Rules Art. 32(3); Swiss Rules 2012 Art. 26.3.
88 English Arbitration Act 1996 s. 39.
89 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 516, n. 11.

90 See, e.g., ICC Rules 2012 Arts 6(3) and 26(2); LCIA Rules Art. 15.8; ICDR Rules

Art. 23; SIAC Rules Art. 21.3; HKIAC Rules Art. 26.2; SCC Rules Art. 30(2); Swiss
Rules 2012 Art. 28; ACICA Rules Art. 29.3; UNCITRAL Rules 2010 Art. 30.2 and
30.3.
91 See, e.g., UNCITRAL Model Law Art. 25; English Arbitration Act s. 41(4);

Swedish Arbitration Act Art. 24.

92 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 524.

93 ICSID Arbitration Rules 42(f)).
94 See, e.g., ICDR Rules Art. 30; LCIA Rules Art. 27.3; SCC Rules Art. 42; SIAC

Rules Art 29.3; HKIAC Rules Art. 35; UNCITRAL Rules 2010 Art. 39; Swiss Rules
2012 Art. 36; ACICA Rules Art. 38. For statutes expressly referring to additional
awards, see UNCITRAL Model Law Art. 33; English Arbitration Act s. 57; Swedish
Arbitration Act s. 32; Austrian Code of Civil Procedure Art. 610; German Code of
Civil Procedure Art. 1058(1). While both the 1998 and 2012 versions of the ICC
Rules allow for corrections for clerical errors, they do not expressly allow for an
additional award for omissions.
95 There may even be conflict of law issues with such a contract if it has a

different applicable law to the underlying dispute.

96 For statutes, see UNCITRAL Model Law Art. 30(1); English Arbitration Act s.

51; Austrian Code of Civil Procedure Art. 605; German Code of Civil Procedure
Art. 1053; Swedish Arbitration Act s. 27. For arbitral rules, see ICC Rules 2012
Art. 32; LCIA Rules 1998 Art. 26.8; ICSID Arbitration Rules Art. 43(2); ICDR Rules
Art. 29.1; ACICA Rules Art. 35.1; HKIAC Rules Art. 32.1; SIAC Rules Art. 28.8; SCC
Rules Art. 39(1); Swiss Rules 2012 Art. 34.1; UNCITRAL Rules 2010 Art. 36.1.
97 James H. Carter, ‘Part I – Issues Arising from Integrated Dispute Resolution

Clauses’, in New Horizons in International Commercial Arbitration and Beyond,

ICCA International Arbitration Congress Series No. 12, ed. Albert Jan van den
Berg (The Hague: Kluwer Law International, 2004), 446, 468.
98 ICC Rules 2012 Art. 32. See also ICDR Rules Art. 29(1); WIPO Rules Art. 65(c).
99 UNCITRAL Model Law Art. 30(1); English Arbitration Act s. 51(2); LCIA Rules

Art. 26.8; UNCITRAL Rules 2010 Art. 36.1; ACICA Rules Art. 35.1; Swiss Rules
2012 Art. 34.1; HKIAC Rules Art. 32.1; ICDR Rules Art. 29.1. The LCIA Arbitration
Rules requires that the award contain an express statement that it is an award
made by the parties' consent: Art. 26.8.
100 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 526.

101 There may be a question as to whether a cancellation fee is excessive and

abuse of a tribunal's ethical obligations. See section 5.17.12.

102 See, e.g., the Singapore High Court decision in Doshion Ltd v. Sembawang

Engineers and Constructors Pte Ltd [2011] SGHC 46. See also Fiona Trust &
Holding Corp. v. Privalov [2007] UKHL 40.
103 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2437–2438; David D. Caron, Matti Pellonpää & Lee M.

Caplan, The UNCITRAL Arbitration Rules: A Commentary, 2nd edn (Oxford: Oxford
University Press, 2006), 859. See also Islamic Republic of Iran v. United States of
America (1981–1982) 1 Iran-US Claims Tribunal 144, 153. This case suggested
that an arbitrator ‘should not attempt to review the reasonableness of the
settlement in the place of the arbitrating parties’. Fouchard presents the
strongest argument that an arbitrator must generally render such an award:
Emmanuel Gail-lard & John Savage (eds), Fouchard Gaillard Goldman on
International Commercial Arbitration (The Hague: Kluwer Law International,
1999), 1365.
104 Simon Greenberg, Christopher Kee & J. Romesh Weeramantry, International

Commercial Arbitration: An Asia-Pacific Perspective (Cambridge: Cambridge

University Press, 2011), 399.
105 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 637.

106 Custodio O. Parlade, ‘Arbitral Awards’, in The Asian Leading Arbitrators' Guide

to International Arbitration, ed. Michael Pryles & Michael Moser (Huntington:

JurisNet LLC, 2007), 381.
107 Christian Hausmaninger, ‘Rights and Obligations of the Arbitrator with
Regard to the Parties and the Arbitral Institution – A Civil Law Viewpoint’, in The
Status of the Arbitrator, ICC International Court of Arbitration Bulletin: 1995
Special Supplement, ed. Jean-François Bourque (Paris: ICC Publishing, 1995), 44.
108 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 637.

109 Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th edn

(Oxford: Oxford University Press, 2009), 338.

110 New York Convention Art. I(1).
111 See, e.g., Australian International Arbitration Act s. 3.
112 See, e.g., Ascom Electro AG v.PT Manggala Mandiri Sentosa cited in Michael

Hwang & Shaun Lee, ‘Survey of South East Asian Nations on the Application of
the New York Convention’, Journal of International Arbitration 25 (2008): 876.
113 Similar provisions are contained in UNCITRAL Rules 2010 Art. 36.
114 UNCITRAL Rules 2010 Art. 36; ICDR Rules Art. 29; ACICA Rules Art. 35;

HKIAC Rules Art. 32; SCC Rules Art. 30.

115 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2442.

116 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 657.

117 Article 1456 French Code of Civil Procedure; Art. 820 Italian Code of Civil

Procedure; Art. 37 Spanish Arbitration Act 2003.

118 Judgment of 16 June 1976, Dame Krebs et autre v. Milton Stern et autre, Cour

de cassation, Revue de l'Arbitrage (1977): 272–275.

119 Howard M. Holtzmann & Joseph E. Neuhaus, A Guide to the UNCITRAL Model

Law on International Commercial Arbitration: Legislative History and

Commentary (Deventer: Kluwer Law and Taxation Publishers, 1989), 841.
120 The KCAB Rules Art. 33, fix the shortest period, just forty-five days. See also

SIAC Rules Art. 28.1; the Milan Rules Art. 21(1); SCC Rules Art. 37; KLRCA Rules
Art. 6.
121 See HKIAC Rules 38.2(d); ACICA Expedited Arbitration Rules Art. 27; SIAC

Rules Art. 5.2(d); Swiss Rules 2012 Art. 42.1(d).

122 ICC Rules 2012 Art. 30(2).
123 W. Laurence Craig, William W. Park & Jan Paulsson, International Chamber of

Commerce Arbitration, 3rd edn (New York: Oceana Publications, 2000), 356-357.
124 José María Alonso, ‘Deliberation and Drafting Awards in International
Arbitration’, in Liber Amicorum Bernardo Cremades, ed. M.Á. Fernández-
Ballesteros & David Arias (Madrid: La Ley, 2010), 157.
125 In Hasbro, Inc. v.Catalyst USA Inc. (7th Cir, 10 May 2004) Docket No. 02-4301,

the losing party sought to vacate the award on the ground that the arbitrators
exceeded their powers by issuing an award beyond the period for the making
and release of the award. The district court vacated the award, however their
decision was reversed on appeal. The US Federal Circuit Court of Appeals found
that the arbitrators did not exceed their authority by issuing an untimely award:
time was not of the essence in the arbitration, despite the fact that the AAA Rules
that governed the arbitration specify a thirty-day deadline for the making of the
arbitral award.
126 Judgment of 16 October 2003, X S.A.L et al. v. Z Sàri, Swiss Federal Supreme

Court, ASA Bulletin 22, no. 2: 372.

127 José María Alonso, ‘Deliberation and Drafting Awards in International
Arbitration’, in Liber Amicorum Bernardo Cremades, ed. M.Á. Fernández-
Ballesteros & David Arias (Madrid: La Ley, 2010), 157.
128 V.V. Veeder, ‘Laws and Court Decisions in Common Law Countries and the

UNCITRAL Model Law’, in Preventing Delay and Disruption of Arbitration, ICCA

Congress Series No. 5 (Stockholm, 1990), ed. Albert Jan van den Berg (The
Hague: Kluwer Law International, 1991), 277–278, 321–322; Stephen N.
Schwebel, International Arbitration: Three Salient Problems (Cambridge:
Cambridge University Press, 1987).
129 Article 1468 of the French New Code of Civil Procedure provides that the

arbitrator is to set the date on which the case will be adjourned for deliberation.
130 Allan Philip, ‘The Duties of an Arbitrator’, in The Leading Arbitrators' Guide to

International Arbitration, ed. Lawrence W. Newman & Richard D. Hill (New York:
Juris Publishing, Inc., 2008), 79.
131 Professor Doug Jones, ‘International Dispute Resolution in the Global
Financial Crisis’, The Arbitrator and Mediator, October 2009, 49.
132 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 651.

133 Ibid., 652.
134 This is mandated by para. 1052(2) of the German Code of Civil Procedure.
135 Karl-Heinz Böckstiegel, ‘Case Management by Arbitrators: Experiences and

Suggestions’, in Global Reflections on International Law, Commerce and Dispute

Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al.
(Paris: ICC Publishing, 2005), 126.
136 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 1623.

137 Swiss Federal Court in Judgment of 23 October 1985, SEFRI SA v. Komgrap,

Swiss Federal Supreme Court, ASA Bulletin 4, no. 2 (1986): 81. This was in the
context of one arbitrator being unable to personally attend meetings due to poor
health, but this did not seem to affect the Court's reasoning. See also Judgment of
16 October 2003, X S.A.L et al. v. Z. Sàrl, Swiss Federal Supreme Court, ASA
Bulletin 22, no. 2 (2004): 372.
138 See, e.g., ICC Rules 2012 Art. 27; UNCITRAL Rules 2010 Art. 31.2; ICDR Rules

Art. 24.2; SCC Rules Art. 34; SIAC Rules Art. 28.1; HKIAC Rules Art. 27; ACICA
Rules Art. 30.
139 See Noble China Inc. v. Lei 42 OR (3d) 69, which rejected a dissenting

arbitrator's affidavit as to the deliberations.

140 Decision of the Appointing Authority to the Iran-US Claims Tribunal, 7 May

2001, Mealeys International Arbitration Report 16, no. 5 (2001).

141 Himpurna Calif Energy Ltd v. Republic of Indonesia, Interim ad hoc Award, 26

September 1999, in Albert Jan van den Berg (ed.), Yearbook of Commercial
Arbitration Volume XXV (The Hague: Kluwer Law International, 2000), 152.
142 Pierre Karrer, ‘Responsibility of Arbitrators and Arbitral Institutions – Let

Sleeping Dogs Lie’, in The Leading Arbitrators' Guide to International Arbitration,

ed. Lawrence W. Newman & Richard D. Hill, 2nd edn (New York: Juris Publishing,
Inc., 2008), 618.
143 Decision of the Appointing Authority to the Iran-US Claims Tribunal, 7 May

2001, Mealeys International Arbitration Report 16, no. 5 (2001).

144 Supreme Court decision of 29 January 2010, LJN BK 2007, discussed by Eelco

Meerdink & Niels Dekker, ‘Supreme Court Rules Arbitral Tribunal Not Required
to Disclose Hearing Notes’, 15, no. 1 IBA Arb News 131.
145 Jean-François Poudret & Sébastien Besson, Comparative Law of International
Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 650.
146 Ibid.
147 The Svea Court of Appeal considered the need to balance fairness and

efficiency and rejected such an argument in Czech Republic v. CME Czech Republic
BV, Svea Court of Appeal, Case No. T8735-01 (2003).
148 Alan Redfern, ‘The 2003 Freshfields Lecture: Dissenting Opinions in
International Commercial Arbitration: The Good, the Bad and the Ugly’,
Arbitration International 20, no. 3 (2004): 238.
149 Ibid., 229.
150 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 654.

151 Ibid., 649. The DIS (German Institution for Arbitration) Arbitration Rules s.

33.4, provide that where an arbitrator refuses to take part in the vote on a
decision the remaining arbitrators may decide without him and decide by
majority vote. However, the parties need to be given advance notice of the
intention to make an award without the arbitrator who refuses to participate in
the vote.
152 UNCITRAL Model Law Art. 29; German Code of Civil Procedure Art. 1052;

UNCITRAL Rules 2010 Art. 33.1 and 34.4 LCIA Rules Art. 26.3 and 26.4; Austrian
Code of Civil Procedure Art. 604; Swiss Private International Law Art. 189(2);
English Arbitration Act s. 52(3); French New Code of Civil Procedure Art. 1480.
Some Latin American countries previously required unanimity.
153 See section 5.16 which deals with truncated tribunals in more detail.
154 José María Alonso, ‘Deliberation and Drafting Awards in International
Arbitration’, in Liber Amicorum Bernardo Cremades, ed. M.Á. Fernández-
Ballesteros & David Arias (Madrid: La Ley, 2010), 139.
155 English Arbitration Act s. 20(4); Swiss Private International Law Art. 189(2);

ICC Rules 2012 Art. 31(1); LCIA Rules Art. 26.3. Arts 30-31 of Swedish
Arbitration Act of 1999 Art. 176a(2) and Art. 189(2) of the New Swiss
International Arbitration Law provide a casting vote to the chair. Similarly, see
LCIA Arbitration Rules 1998 Art. 26.3 and 26.4; Milan Chamber of Commercial
International Arbitration Rules Art. 18.1; WIPO Arbitration Rules Art. 61; Rules
of Procedure of the Inter-American Commercial Arbitration Commission Art. 28;
Singapore International Arbitration Centre Arbitration Rules, rule 28.3. Note that
under the SCC Arbitration Rules Art. 36.3 the chairman may sign the award
where there is no majority.
156 See, e.g., ICC Case No. 7365, Mealey's International Arbitration Report, 13

(1998): G-5; ICC Case No. 3881 (1984), in Collection of ICC Arbitral Awards, 1986–
1990, ed. Sigvard Jarvin, Yves Derains & Jean-Jacques Arnaldes (Deventer:
Kluwer Law, 1994), 257; ICC Case No. 4567 in Albert Jan van den Berg (ed.),
Yearbook Book of Commercial Arbitration Volume XI (The Hague: Kluwer Law
International, 1986), 143–147; Francis Patrick Donovan, ‘Dissenting Opinions’,
ICC International Court of Arbitration Bulletin 7, no. 2 (1996): 77.
157 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 660. The Swedish
Arbitration Act s. 30 and Swiss Private International Law Art. 189(2) provide a
casting vote to the chair. Section 20(4) of the English Arbitration Act follows the
ICC model.
158 Emmanuel Gaillard & John Savage (eds), Fouchard Gaillard Goldman on
International Commercial Arbitration (The Hague: Kluwer Law International,
1999), 747.
159 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 661. UNCITRAL Model
Law Art. 29 only deals with issues of procedure. Likewise, the German Code of
Civil Procedure Art. 1052(3) does not provide a mechanism for dealing with
deadlocks.
160 José María Alonso, ‘Deliberation and Drafting Awards in International
Arbitration’, in Liber Amicorum Bernardo Cremades, ed. M Á. Fernández-
Ballesteros & David Arias (Madrid: La Ley, 2010), 139.
161 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2461.

162 See Nigel Blackaby et al., Redfern and Hunter on International Arbitration, 5th

edn (Oxford: Oxford University Press, 2009), 371. The authors suggest that the
chair might decide alone on all issues.
163 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 661.

164 Ibid., 659; ICSID Rules Art. 15.1.
165 See, e.g., Report of the Secretary-General on the Revised Draft Set of
Arbitration Rules, UNCITRAL, 9th Session, UN Doc. A/CN.9/112/Add.1, VII
UNCITRAL Yearbook 166, 178 (1976); Gary B. Born, International Commercial
Arbitration (The Hague: Kluwer Law International, 2009), 2461. See the Notes to
Art. 47 of the ICSID Arbitration Rules 2006 and the possibility of sequential
voting systems.
166 Christian Hausmaninger, ‘Rights and Obligations of the Arbitrator with
Regard to the Parties and the Arbitral Institution – A Civil Law Viewpoint’, in The
Status of the Arbitrator, ICC International Court of Arbitration Bulletin: 1995
Special Supplement, ed. Jean-François Bourque (Paris: ICC Publishing, 1995), 46.
167 See generally Stephen Schwebel, ‘May the Majority Vote of an International

Arbitral Tribunal Be Impeached?’, Arbitration International 13, no. 2 (1997).

168 Laurent Lévy, ‘Dissenting Opinions in International Arbitration in
Switzerland’, Arbitration International 5, no. 1 (1989): 35.
169 See the list in Manuel Arroyo, ‘Dealing with Dissenting Opinions in the

Award: Some Options for the Tribunal’, ASA Bulletin 26, no. 3 (2008): 442.
170 Howard M. Holtzmann & Joseph E. Neuhaus, A Guide to the UNCITRAL Model

Law on International Commercial Arbitration: Legislative History and

Commentary (Deventer: Kluwer Law and Taxation Publishers, 1989), 837.
171 UNCITRAL Secretariat, ‘Explanatory Note by the UNCITRAL Secretariat on

the Model Law on International Commercial Arbitration as Amended in 2006’,

<http://www.uncitral.org/pdf/english/texts/arbitrat...>.
172 Iran-US Claims Tribunal, Rules of Procedure Art. 32.
173 Article 48(4).
174 Emmanuel Gaillard & John Savage (eds), Fouchard Gaillard Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 766; Manuel Arroyo, ‘Dealing with Dissenting Opinions in the Award:
Some Options for the Tribunal’, ASA Bulletin 26 (2008): 443.
175 CIETAC Rules Art. 47.5.
176 Annette Magnusson & Patricia Shaughnessy, ‘The 2007 Arbitration Rules of
the Arbitration Institute of the Stockholm Chamber of Commerce’, Stockholm
International Arbitration Review (2006): 33-66, available at
<http://www.sccinstitute.com/filearchive/2/21223/20...>.
177 Pierre Karrer, ‘Responsibility of Arbitrators and Arbitral Institutions – Let

Sleeping Dogs Lie’, in The Leading Arbitrators' Guide to International Arbitration,

ed. Lawrence W. Newman & Richard D. Hill, 2nd edn (New York: Juris Publishing,
Inc, 2008), 618.
178 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2466.

179 See Alan Redfern, ‘The 2003 Freshfields Lecture: Dissenting Opinions in

International Commercial Arbitration: The Good, the Bad and the Ugly’.
Arbitration International 20, no. 3 (2004).
180 James H. Carter, ‘The Rights and Duties of the Arbitrator: Six Aspects of the

Rule of Reasonableness’, in The Status of the Arbitrator, ICC International Court of

Arbitration Bulletin: 1995 Special Supplement, ed. Jean-François Bourque (Paris:
ICC Publishing, 1995), 32.
181 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 675.

182 ICSID Rules Arts 15(1) and 47(3); Peter J. Rees & Patrick Rohn, ‘Dissenting

Opinions: Can They Fulfil a Beneficial Role?’, Arbitration International 25, no. 3
(2009): 337.
183 Section 16.7 dealt with complex situations where it is hard to identify any

majority position as a result of differing reasons and quantum amounts.

184 ICC Commission on International Arbitration, Working Party on Dissenting

Opinions and Interim and Partial Awards ‘Final Report on Dissenting and
Separate Opinions’, ICC International Court of Arbitration Bulletin 2, no. 1 (1991):
36.
185 Pieter Sanders, ‘Has the Moment Come to Revise the Arbitration Rules of

UNCITRAL?’, Arbitration International 20, no. 3 (2004): 253.

186 ICC Commission on International Arbitration, Working Party on Dissenting

Opinions and Interim and Partial Awards ‘Final Report on Dissenting and
Separate Opinions’, ICC International Court of Arbitration Bulletin 2, no. 1 (1991):
32.
187 Decision of 19 June 2006, X v. Z, Swiss Federal Supreme Court, ASA Bulletin

24, no. 4 (2006): 761.

188 Peter J. Rees & Patrick Rohn, ‘Dissenting Opinions: Can They Fulfil a
Beneficial Role?’, Arbitration International 25 (2009): 338.
189 Ibid., 336.
190 For an argument in favour of the right to dissenting opinions see Jacques

Werner, ‘Dissenting Opinions: Beyond Fears’, Journal of International Arbitration

9, no. 4 (1992): 29.
191 Similarly, others will be warned about an arbitrator whose dissent is
unreasonable.
192 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2469.

193 This is a corollary of Lord Bingham's observation that, at times when a judge

sits down to write a reasoned argument in favour of a decision already reached,

the process leads to the inevitable conclusion that a contrary decision is the only
legitimate one. Lord Bingham, ‘Reasons and Reasons for Reasons’, Arbitration
International 4, no. 2 (1988): 143.
194 ICC, ‘Final Report on Dissenting and Separate Opinions’, ICC International

Court of Arbitration Bulletin 2, no. 1 (1991): 32; José María Alonso, ‘Deliberation
and Drafting Awards in International Arbitration’, in Liber Amicorum Bernardo
Cremades, ed. M.Á. Fernández-Ballesteros & David Arias (Madrid: La Ley, 2010),
141.
195 Cargill International v. Sociedad Iverica de Molturacion [1998] 1 Lloyd's Rep

489; Judgment of 11 May 1992, Swiss Federal Supreme Court, ASA Bulletin 10,
no. 3 (1992): 381.
196 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 675.

197 Manuel Arroyo, ‘Dealing with Dissenting Opinions in the Award: Some
Options for the Tribunal’, ASA Bulletin 26, no. 3 (2008): 462.
198 Ibid.
199 See Alan Redfern, ‘The 2003 Freshfields Lecture: Dissenting Opinions in

International Commercial Arbitration: The Good, the Bad and the Ugly’,
Arbitration International 20, no. 3 (2004): 241.
200 José María Alonso, ‘Deliberation and Drafting Awards in International
Arbitration’, in Liber Amicorum Bernardo Cremades, ed. M.Á. Fernández-
Ballesteros & David Arias (Madrid: La Ley, 2010), 141.
201 For example, Emmanuel Gaillard & John Savage (eds), Fouchard Gaillard

Goldman on International Commercial Arbitration (The Hague: Kluwer Law

International, 1999), 768.
202 Inclusion in the majority award is the preferred option of Peter J. Rees &

Patrick Rohn, ‘Dissenting Opinions: Can they Fulfil a Beneficial Role?’, Arbitration
International, 25 (2009): 342.
203 Christian Hausmaninger, ‘Rights and Obligations of the Arbitrator with
Regard to the Parties and the Arbitral Institution – A Civil Law Viewpoint’, in The
Status of the Arbitrator, ICC International Court of Arbitration Bulletin: 1995
Special Supplement, ed. Jean-François Bourque (Paris: ICC Publishing, 1995), 47.
204 See, e.g., Judgment of 11 May 1992, Swiss Federal Supreme Court, ASA

Bulletin 10, no. 3 (1992): 381.

205 Manuel Arroyo, ‘Dealing with Dissenting Opinions in the Award: Some
Options for the Tribunal’, ASA Bulletin 26, no. 3 (2008): 452.
206 Ibid., 462.
207 Peter J. Rees & Patrick Rohn, ‘Dissenting Opinions: Can they Fulfil a Beneficial

Role?’, Arbitration International 25 (2009): 323.

208 Ibid., 334.
209 ICC Commission on International Arbitration, Working Party on Dissenting

Opinions and Interim and Partial Awards, ‘Final Report on Dissenting and
Separate Opinions’, ICC International Court of Arbitration Bulletin 2, no. 1 (1991):
35.
210 Laurent Lévy, ‘Dissenting Opinions in International Arbitration in
Switzerland’, Arbitration International 5, no. 1 (1989): 40.
211 An early draft did allow for the right to present dissenting opinions. Laurent

Lévy, ibid., 42.

212 Manuel Arroyo, ‘Dealing with Dissenting Opinions in the Award: Some
Options for the Tribunal’, ASA Bulletin 26, no. 3 (2008): 460.
213 James H. Carter, ‘The Rights and Duties of the Arbitrator: Six Aspects of the
Rule of Reasonableness’, in The Status of the Arbitrator, ICC International Court of
Arbitration Bulletin: 1995 Special Supplement, ed. Jean-François Bourque (Paris:
ICC Publishing, 1995), 33.
214 Government of Kuwait v. American Independent Oil Company (Aminoil), 24

March 1982, in Albert Jan van den Berg (ed.), Yearbook Book of Commercial
Arbitration Volume IX (The Hague: Kluwer Law International, 1984), 71.
215 See Manuel Arroyo, ‘Dealing with Dissenting Opinions in the Award: Some

Options for the Tribunal’, ASA Bulletin 26, no. 3 (2008): 456 and the authors
cited.
216 Laurent Lévy, ‘Dissenting Opinions in International Arbitration in
Switzerland’ Arbitration International 5, no. 1 (1989): 39. La prestation
caractéristique en droit international privé des contrats: l'influence de la Suisse,
in Essays on the occasion of the 75th Anniversary of the Swiss Association for
International Law, Zurich 1989.
217 Article 189.2. It seems that in the Philippines arbitrators who make an award

must acknowledge or verify the award (by confirming its correctness, truth or
authenticity) as a separate juridical act. Grogun, Inc. v. National Power
Corporation, G.R. No. 156259, Philippines Supreme Court, 18 September 2003.
218 See, e.g., UNCITRAL Model Law Art. 31(1).
219 Holtzmann & Joseph E. Neuhaus, A Guide to the UNCITRAL Model Law on

International Commercial Arbitration: Legislative History and Commentary

(Deventer: Kluwer Law and Taxation Publishers, 1989), 827.
220 That is expressly covered in Art. 37.3 of the Spanish Arbitration Act.
221 Karl-Heinz Böckstiegel, ‘Case Management by Arbitrators: Experiences and

Suggestions’, in Global Reflections on International Law, Commerce and Dispute

Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al.
(Paris: ICC Publishing, 2005), 126.
222 Neil Kaplan & Karen Mills, ‘The Role of the Chair in International Commercial

Arbitration’, in The Asian Leading Arbitrators' Guide to International Arbitration,

ed. Michael Pryles & Michael Moser (Huntington: JurisNet LLC, 2007), 142.
223 Ibid., 143.
224 Geoffrey M. Beresford Hartwell, ‘The Reasoned Award in International
Arbitration’ available at <http://www.hartwell.demon.co.uk.intaward.htm>.
225 Pierre Karrer, ‘Must an Arbitral Tribunal Really Ensure that its Award is

Enforceable?’, in Global Reflections on International Law, Commerce and Dispute

Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al.
(Paris: ICC Publishing, 2005), 431.
226 Lord Bingham, ‘Reasons and Reasons for Reasons’, Arbitration International

4, no. 2 (1988): 143.

227 Judge Howard Holtzmann criticised the tendency to length in awards in his

Dissenting and Concurring opinion in Mohsen Asgari Nazari v. Islamic Republic of

Iran (24 August 1994) cited in Gary B. Born, International Commercial
Arbitration (The Hague: Kluwer Law International, 2009), 2454, n. 168.
228 Robert M. Merkin & Louis Flannery, Arbitration Act 1996, 3rd edn (London:

Informa, 2005), 139.

229 Pierre Lalive, ‘On the Reasoning of International Arbitral Awards’, JIDS 1, no.

1 (2010): 55, 57.

230 José María Alonso, ‘Deliberation and Drafting Awards in International
Arbitration’, in Liber Amicorum Bernardo Cremades, ed. M.Á. Fernández-
Ballesteros & David Arias (Madrid: La Ley, 2010), 152.
231 New York Convention Art. IV.2.
232 Lord Bingham, ‘Reasons and Reasons for Reasons’, Arbitration International

4, no. 2 (1988): 142–143.

233 For arguments in favour of reasons, see Thomas E. Carbonneau, ‘Rendering

Arbitral Awards with Reasons: The Elaboration of a Common Law of

International Transactions’, Columbia Journal of Transnational Law 23, no. 3
(1985): 579–614.
234 See, e.g., UNCITRAL Model Law Art. 31(2); English Arbitration Act s. 52(4);

Austrian Code of Civil Procedure Art. 606(2); German Code of Civil Procedure
Art. 1054; Swiss Private International Law Art. 189(2); ICC Rules 2012 Art.
31(2); UNCITRAL Rules 2010 Art. 34.3; LCIA Rules Art. 26.1; ICDR Rules Art.
27.2; SCC Rules Art. 31.2; ACICA Rules Art. 33.3; ICSID Convention Art. 48(3).
235 José María Alonso, ‘Deliberation and Drafting Awards in International
Arbitration’, in Liber Amicorum Bernardo Cremades, ed. M.Á. Fernández-
Ballesteros & David Arias (Madrid: La Ley, 2010), 156.
236 English Arbitration Act s. 68. See also World Trade Corp. Ltd v. C Czarnikow

Sugar Ltd [2004] EWHC 2332 at [20]; Margulead Ltd v. Exide Technology [2004]
EWHC 1019 at [42].
237 The Easy Rider [2004] 2 Lloyd's Rep 626 (QB).
238 Al-Hadha Trading Co. v. Tradigrain SA [2002] 2 Lloyd's Rep 512, 526.
239 Christian Hausmaninger, ‘Rights and Obligations of the Arbitrator with
Regard to the Parties and the Arbitral Institution – A Civil Law Viewpoint’, in The
Status of the Arbitrator, ICC International Court of Arbitration Bulletin: 1995
Special Supplement, ed. Jean-François Bourque (Paris: ICC Publishing, 1995), 47.
240 For example, Mut Shipping Corp. v. Bayshore Shipping Co, The Montan [1985]

1 Lloyd's Rep 189.

241 A.A. de Fina, ‘The Party Appointed Arbitrator in International Arbitrations –

Role and Selection’, Arbitration International 15, no. 4 (1999): 387.

242 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2454.

243 Pierre Karrer, ‘Must an Arbitral Tribunal Really Ensure That Its Award Is

Enforceable?’, in Global Reflections on International Law, Commerce and Dispute

Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al.
(Paris: ICC Publishing, 2005), 432.
244 Mahon v. Air New Zealand Ltd [1984] AC 808, 821 per Lord Diplock; Minister

for Immigration & Ethnic Affairs v. Pochi (1980) 44 FLR 41, 67–68 per Deane J;
Westminster City Council v. Great Portland Street Estates PLC [1985] AC 661, 673
per Lord Scarman.
245 Pierre Lalive points to the inconsistency between the ad hoc committee in

MINE v. Guinea stating on the one hand that ‘the adequacy of the reasoning is not
an appropriate standard of review’, but also saying that the minimum
requirement for adequate reasons ‘is in particular not satisfied by either
contradictory or frivolous reasons’. Pierre Lalive, ‘On the Reasoning of
International Arbitral Awards’, JIDS 1, no. 1 (2010): 55, 65.
246 Klöckner v. Cameroon case (3 May 1985) 2 ICSID Rep 95 (1994).
247 ICSID No. ARB/82/1, Decision on Annulment, 3 May 1985, ICSID Review

(1986), 126.
248 Pierre Lalive, ‘On the Reasoning of International Arbitral Awards’, JIDS 1, no.
1 (2010): 55, 65.
249 Pierre Karrer, ‘Must an Arbitral Tribunal Really Ensure that Its Award Is

Enforceable?’, in Global Reflections on International Law, Commerce and Dispute

Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al.
(Paris: ICC Publishing, 2005), 432.
250 Ibid.
251 Michael J. Mustill & Steward C. Boyd, The Law and Practice of Commercial

Arbitration in England (London: Butterworths, 1982), 548.

252 Pierre Karrer, ‘Must an Arbitral Tribunal Really Ensure that its Award is

Enforceable?’, in Global Reflections on International Law, Commerce and Dispute

Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al.
(Paris: ICC Publishing, 2005), 433.
253 Lord Bingham, ‘Reasons and Reasons for Reasons’, Arbitration International 4

(1988): 145.
254 T.T. Landau, ‘Reasons for Reasons: The Tribunal's Duty in Investor-State

Arbitration’, ICCA Congress Series No. 14, Dublin Conference, 2008 (The Hague:
Kluwer, 2009), 187; see also G.-A. Alvarez & W.M. Reisman (eds), The Reasons
Requirement in International Investment Arbitration (Leiden: M Nijhoff 2008).
255 Pierre Lalive also suggest that leading arbitrators too often take on too much

work and provide for superficial reading in complex cases of this nature. Pierre
Lalive, ‘On the Reasoning of International Arbitral Awards’ JIDS 1, no. 1 (2010):
55, 58.
256 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2456.

257 Judgment of 14 June 2000, Inter Arab Investment Guarantee Corp. v. Banque

Arabe et Internationale d'Investissenent, Cour de cassation, 2000 IR 195.

258 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 671.

259 Halligan v. Piper Jaffray, 148 F. 3d 197 (2d Cir 1998), Cert denied, 526 US

1034 (1999).
260 BHP Petroleum Pty Ltd v. Oil Basins Ltd [2006] VSC 402 at [23].
261 Westport Insurance Corporation v. Gordian Runoff Ltd [2011] HCA 37.
262 UNCITRAL Model Law Art. 32(4); English Arbitration Act s. 52(3); Austrian

Code of Civil Procedure Art. 606(1); German Code of Civil Procedure Art.
1054(1); French New Code of Civil Procedure Art. 1480; Swiss Private
International Law Art. 189; Swedish Arbitration Act s. 30. See also ICDR Rules
Art. 27.
263 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 681.

264 UNCITRAL Model Law Art. 32(4). It is important to ensure that this is not

abused to allow a potential dissenter to provide reasons which might aim to

undermine the award itself. See David D. Caron, Matti Pellonpää & Lee M. Caplan,
The UNCITRAL Arbitration Rules: A Commentary, 2nd edn (Oxford: Oxford
University Press, 2006), 826 et seq.
265 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2460.

266 Cited in José María Alonso, ‘Deliberation and Drafting Awards in
International Arbitration’, in Liber Amicorum Bernardo Cremades, ed. M.Á.
Fernández-Ballesteros & David Arias (Madrid: La Ley, 2010), 154.
267 Cargill International v. Sociedad Iverica de Molturacion [1998] 1 Lloyd's Rep

489.
268 Ibid.
269 Pierre Karrer, ‘Responsibility of Arbitrators and Arbitral Institutions – Let

Sleeping Dogs Lie’, in The Leading Arbitrators' Guide to International Arbitration,

ed. Lawrence W. Newman & Richard D. Hill, 2nd edn (New York: Juris Publishing,
Inc., 2008), 619.
270 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 682.

271 In centrally planned economies there may be a need to have a witness in an

official position affixing an appropriate stamp or authorisation. Geoffrey M.

Beresford Hartwell, ‘The Reasoned Award in International Arbitration’ available
at <http://www.hartwell.demon.co.uk.intaward.htm>.
272 Pieter Sanders, ‘Has the Moment Come to Revise the Arbitration Rules of

UNCITRAL?’, Arbitration International 20, no. 3 (2004): 253. See also

Explanatory note by the UNCITRAL Secretariat on the Model Law on
International Commercial Arbitration.
273 UNCITRAL Model Law Art. 31(3); English Arbitration Act s. 53; German Code

of Civil Procedure Art. 1054(3); ICC Rules 2012 Art. 31(3); LCIA Rules Art. 26.1;
ICDR Rules Art. 27.3; SCC Rules Art. 20.3; ACICA Rules Art. 19.4.
274 Hiscox v. Outhwaite [1991] 2 WLR 1321.
275 [1991] 2 Lloyd's Rep 435, HL.
276 See UNCITRAL Model Law Art. 34(3); English Arbitration Act s. 70(3); French

New Code of Civil Procedure Art. 1481; Swedish Arbitration Act s. 31; UNCITRAL
Rules Art. 34.4; LCIA Rules Art. 26.1; ICDR Rules Art. 27.3; HKIAC Rules Art. 30.4;
ACICA Rules Art. 33.4.
277 Individual dates of signature for each arbitrator are still required under ICCP

2006 Art. 823(2) No. 8. However, an award signed only by the majority is still
valid if all arbitrators took part in the deliberation, and one arbitrator was
unable or unwilling to sign.
278 José María Alonso, ‘Deliberation and Drafting Awards in International
Arbitration’, in Liber Amicorum Bernardo Cremades, ed. M.Á. Fernández-
Ballesteros & David Arias (Madrid: La Ley, 2010), 154.
279 See, e.g., UNCITRAL Model Law Art. 34(2).
280 ICC Rules 2012 Art. 33.
281 Günther J. Horvath, ‘The Duty of the Tribunal to Render an Enforceable

Award’, Journal of International Arbitration 18, no. 2 (2001): 149.

282 Yves Derains & Eric A. Schwartz, A Guide to the ICC Rules of Arbitration (The

Hague: Kluwer Law International, 1998), 290.

283 Katherine Gonzalez Arrocha, Director, ICC Dispute Resolution Services, Latin

America, Welcoming Remarks at International Commercial Arbitration in Latin

America: The ICC Perspective Conference (5 November 2007) cited in Nana
Japaridze, ‘Fair Enough? Reconciling the Pursuit of Fairness and Justice with
Preserving the Nature of International Commercial Arbitration’, Hofstra Law
Review 36, no. 4 (2008): 1428.
284 Arrocha, ibid.
285 See, e.g., German Code of Civil Procedure Art. 1054(4); Swedish Arbitration

Act s. 31(3).
286 For example, English Arbitration Act s. 55(2).
287 Hans van Houtte, ‘The Delivery of Awards to the Parties’, Arbitration
International 21, no. 2 (2005): 181.
288 See, e.g., ICC Rules 2012 Art. 34(1); ICDR Rules Art. 27; LCIA Rules Art. 26.5.
289 Swiss Rules 2012 Art. 32; SCC Rules Art. 36.4.
290 ICC Rules 2012 Art. 34(2); ICDR Rules Arts 27.4 and 34; LCIA Rules Art. 30.1;

ICSID Rules Art. 48(4); Swiss Rules 2012 Art. 44.1.

291 Article 32.6 of the Swiss Rules provides for originals to be communicated to

the parties. Art. 34.4 of the ICC Rules 2012 requires originals to be deposited
with the Secretariat.
292 See, e.g., German Code of Civil Procedure Art. 1054(4); DIS Rules 1998 Art.

36; Swiss Rules 2012 Art. 32.6.

293 See, e.g., English Arbitration Act s. 55; Swedish Arbitration Act s. 31;
UNCITRAL Rules 2010 Art. 34.6; LCIA Rules Art. 26.5.
294 An example is EC Council Regulation 1348/2000 [2000] OJ L 160/37; Hans

van Houtte, ‘The Delivery of Awards to the Parties’, Arbitration International 21,
no. 2 (2005): 182.
295 Hans van Houtte, ibid., 183. One of the most important such agreement is the

1965 Hague Convention on the Service Abroad of Judicial and Extrajudicial

Documents in Civil or Commercial Matters, which has sixty-four contracting
States. There is also the 1954 Hague Convention on Civil Procedure, which has
forty-seven contracting States.
296 Hans van Houtte, ‘The Delivery of Awards to the Parties’. Arbitration
International 21, no. 2 (2005): 184.
297 Ibid., 180.
298 Ibid., 182.
299 See, e.g., UNCITRAL Model Law Art. 33(1). This provides a time limit of thirty

days to apply for correction, interpretation or an additional award.

300 UNCITRAL Model Law Art. 32(1).
301 Hans van Houtte, ‘The Delivery of Awards to the Parties’, Arbitration
International 21, no. 2 (2005): 182.
302 Ibid.
303 Judgment of 10 November 2005, Swiss Federal Supreme Court, Swiss Federal

Supreme Court, ASA Bulletin 24, no. 1 (2006): 113.

304 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2353, n. 135.

305 Individual countries may vary this. For example, the US and Canada are given

the right to unilaterally publish ICSID awards between themselves under NAFTA.
306 See, e.g., Netherlands Code of Civil Procedure Art. 1058(2).
307 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 683.

308 Ibid., 697.
309 Brooks W. Daly, ‘Correction and Interpretation of Arbitral Awards under the

ICC Rules of Arbitration’, ICC International Court of Arbitration Bulletin 13, no. 1
(2002): 61.
310 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 697.

311 Carlos Henrique de C. Fróes, ‘Correction and Interpretation of Arbitral
Awards’, in Global Reflections on International Law, Commerce and Dispute
Resolution: Liber Amicorum in Honour of Robert Briner, ed. Gerald Aksen et al.
(Paris: ICC Publishing, 2005), 288.
312 James M. Gaitis, ‘The Federal Arbitration Act: Risks And Incongruities
Relating to the Issuance of Interim and Partial Awards in Domestic and
International Arbitrations’, The American Review of International Arbitration 16
(2005): 87.
313 Horacio Grigera Naón, ‘Choice of Law Problems in International Commercial

Arbitration’, Recueil des Cours (2001) 289: 166–167.

314 See ICC Rules 2012 Art. 35; ICDR Rules Art. 30; Swiss Rules 2012 Art. 36;

LCIA Rules, 27; CIETAC Rules 2005 Art. 51; HKIAC Rules Art. 34; SIAC Rules Art.
29; SCC Rules Art. 41; ACICA Rules Art. 37.
315 UNCITRAL Model Law Art. 33(2).
316 Other similar institutional rules include Art. 35 Swiss Rules 2012; Rule 50(1)

ICSID Arbitration Rules. Art. 56 of the ICSID Additional Facility Rules Schedule C,
allows a party to request the Secretary-General to obtain a correction of clerical,
arithmetical or similar errors from the tribunal within forty-five days of the
award. A tribunal may also make such changes on its own initiative. The ICC
Rules (Art. 35.1) provide for a thirty-day limit and also require scrutiny by the
ICC Secretariat and the International Court of Arbitration. The ICC Secretariat
has also published a Note regarding Correction and Interpretation of Arbitral
Awards in ICC, International Court of Arbitration Bulletin, 10, no. 2 (1999): 4. The
Note makes clear that even a decision to reject a correction request should itself
be subject to scrutiny by the Secretariat and the International Court of
Arbitration.
317 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 688.

318 Ibid., 686, but compare Klaus Peter Berger, International Economic
Arbitration (The Hague: Kluwer Law and Taxation Publishers, 1993), 643.
319 See Jean-François Poudret & Sébastien Besson, Comparative Law of
International Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 696.
320 See Judgment of 2 November 2000, Swiss Federal Supreme Court, ASA

Bulletin 19, no. 1 (2001): 88. See generally Gabrielle Kaufmann-Kohler & Antonio
Rigozzi, ‘Correction and Interpretation of Awards in International Arbitrations
Held in Switzerland’, Mealey's International Arbitration Report 16, no. 4 (2001):
25. The Swiss PILA does not include an equivalent provision to Art. 39 of the
Concordat which allowed awards to be remitted to the tribunal to rectify errors
of fill gaps.
321 Compare Eco v. Eurotech SA Paris Court of Appeal, 1re Ch. C 12 September

2002 and Hussman (Europe) Ltd v. Ahmed Pharaon [2003] EWCA Civ 266.
322 Brooks W. Daly, ‘Correction and Interpretation of Arbitral Awards under the

ICC Rules of Arbitration’, ICC International Court of Arbitration Bulletin 13, no. 1
(2002): 61. A possible drafting gap in the French New Code of Civil Procedure
appears to have been filled by the January 2011 amendments. Art. 1500 now
makes it clear that the power to interpret and correct which exists in domestic
arbitration extends to international arbitration.
323 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2524. See also English Arbitration Act s. 57(1).

324 Ibid., 2531. See also Draft on Arbitral Procedure prepared by the
International Law Commission at Its Fourth Session, 1952, UN Doc. A/CN.4/59,
in Yearbook of the International Law Commission II (1952), Art. 26.
325 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2523, n. 59 criticising David Williams & Amy Buchanan,

‘Corrections and Interpretations of Awards under Article 33 of the Model Law’,
International Arbitration Law Review (2001): 121 and Emmanuel Gaillard & John
Savage (eds), Fouchard Gaillard Goldman on International Commercial
Arbitration (The Hague: Kluwer Law International, 1999), 1416. The broader
formulation in Caron et al. is also less helpful. The authors state ‘Article 36
permits correction of errors in the award that the arbitral tribunal made
unintentionally or heedlessly’: David D. Caron, Matti Pellonpää & Lee M. Caplan,
The UNCITRAL Arbitration Rules: A Commentary, 2nd edn (Oxford: Oxford
University Press, 2006), 895.
326 For cases seeking correction see ICC, ‘Extracts from ICC Addenda and
Decisions Rendered under Article 29 of the ICC Rules of Arbitration’, ICC
International Court of Arbitration Bulletin 13, no. 1 (2002): 72.
327 (1987) 14 Iran-US CTR 100, 101.
328 For cases seeking correction see ICC, ‘Extracts from ICC Addenda and
Decisions Rendered under Article 29 of the ICC Rules of Arbitration’, ICC
International Court of Arbitration Bulletin 13, no. 1 (2002): 72.
329 Pieter Sanders, ‘Has the Moment Come to Revise the Arbitration Rules of

UNCITRAL?’, Arbitration International 20, no. 3 (2004): 253.

330 English Arbitration Act s. 57.
331 Al Hadha Trading Co. v. Tradigrain SA [2002] 2 Lloyd's Rep 512 (QB); Torch

Offshore LLC v. Cable Shipping Inc. [2004] 2 All ER (Comm) 365 (QB).
332 This seems to be the intent of the Working Party to the UNCITRAL Model

Law. See Howard M. Holtzmann & Joseph E. Neuhaus, A Guide to the UNCITRAL
Model Law on International Commercial Arbitration: Legislative History and
Commentary (Deventer: Kluwer Law and Taxation Publishers, 1989), 889.
333 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2525. See also Judgment of 17 May 2004,

Oberlandesgericht Frankfurt (Higher Regional Court of Frankfurt) SchiedsVZ
(2005): 311.
334 English Arbitration Act s. 72(b).
335 Gbangbola v. Smith & Sheriff [1998] 3 All ER 730, QB.
336 See Jean-François Poudret & Sébastien Besson, Comparative Law of
International Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 692.
337 See ibid.
338 Note of the Secretariat of the International Court of Arbitration of the
International Chamber of Commerce Regarding Correction and Interpretation of
Arbitral Awards in ICC, International Court of Arbitration Bulletin 10, no. 2
(1999): 4.
339 See also Dutch Code of Civil Procedure Art. 1060(7).
340 Judgment of 12 January 2005, Swiss Federal Supreme Court, ASA Bulletin 23,

no. 2 (2005): 356.

341 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 693.

342 English Arbitration Act s. 57; UNCITRAL Model Law Art. 33; German Code of

Civil Procedure Art. 1057(2); Austrian Code of Civil Procedure Art. 610(1);
Chinese Arbitration Law Art. 56; Singapore Arbitration Act s. 43(1); Swedish
Arbitration Act s. 32.
343 Suggestions have ranged from thirty days to a year. See Gary B. Born,

International Commercial Arbitration (The Hague: Kluwer Law International,

2009), 2527, n. 79. See also B. Berger & F Kellerhals, Internationale und interne
Schiedsgerichtbarkeit in der Schweiz (Bern: Stämpfli, 2006), para. 1406.
344 See, e.g., English Arbitration Act s. 56.
345 French New Code of Civil Procedure Art. 1485(3). See also US Federal

Arbitration Act, 9 USC § 11 which allows the court rather than the tribunal to
modify or correct the award if ‘there was an evident material miscalculation of
figures or an evident material mistake in the description of any person, thing or
property referred to in the award’ or the award ‘is imperfect in matter or form
not affecting the merits of the controversy’.
346 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2529, n. 89.

347 14 F. 3d 1250 (7th Cir 1994).
348 Singapore Arbitration Act s. 43; French New Code of Civil Procedure Art.

1485; German Code of Civil Procedure Art. 1058(1); Swedish Arbitration Act s.
32.
349 For Switzerland, see Judgment of 2 November 2000, Swiss Federal Supreme

Court, ASA Bulletin 19, no. 1 (2001): 88. For the US, see US Energy Corp. v. Nukem
Inc., 400 F. 3d 822 (10th Cir 2005).
350 In Norway, the arbitration statute is silent, and arbitrators presented with a

request for interpretation do not have to reply. See Anders Ryssdal & Kristian S.
Myrbakk, ‘National Report for Norway’, in International Handbook on
Commercial Arbitration, ed. Jan Paulsson (The Hague: Kluwer Law International,
2009), 29. Likewise, the Dutch Code of Civil Procedure provides no mechanism
for interpretation.
351 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2537–2538.

352 Wolfgang Kühn, ‘Rectification and Interpretation of Arbitral Awards’, ICC

International Court of Arbitration Bulletin 7, no. 2 (1996): 78.

353 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 659.
354 See Decision of 14 March 1978, Interpretive Decision, International Law

Reports 54 (1979).
355 See, e.g., Nigel Blackaby et al., Redfern and Hunter on International
Arbitration, 5th edn (Oxford: Oxford University Press, 2009), 400–401; Klaus
Peter Berger, International Economic Arbitration (Deventer: Kluwer Law and
Taxation Publishers, 1993), 640–644.
356 Howard M. Holtzmann & Joseph E. Neuhaus, A Guide to the UNCITRAL Model

Law on International Commercial Arbitration: Legislative History and

Commentary (Deventer: Kluwer Law and Taxation Publishers, 1989), 890–891.
357 English Arbitration Act, 57.
358 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 693–694.

359 Robert M. Merkin & Louis Flannery, Arbitration Act 1996, 3rd edn (London:

Informa, 2005), 126.

360 Julian D.M. Lew, Loukas A. Mistelis & Stefan M. Kröll, Comparative
International Commercial Arbitration (The Hague: Kluwer Law International,
2003), 658.
361 Emmanuel Gaillard & John Savage (eds), Fouchard Gaillard Goldman on

International Commercial Arbitration (The Hague: Kluwer Law International,

1999), 776. For a contrary view see Brooks W. Daly, ‘Correction and
Interpretation of Arbitral Awards under the ICC Rules of Arbitration’, ICC
International Court of Arbitration Bulletin 13, no. 1 (2002): 61.
362 See Pepsi Co., Inc. v. Islamic Republic of Iran (1986) 13 Iran-US CTR 328, 329–

330.
363 See, e.g., Wintershall AG v. Government of Qatar (5 February 1988 and 31 May

1988), in Albert Jan van den Berg (ed.), Yearbook of Commercial Arbitration
Volume XV (The Hague: Kluwer Law International, 1990), 30.
364 See, e.g., ICC Addendum in Case No. 10189, ICC Bulletin, 15, no. 1 (2002): 81.
365 Pieter Sanders, ‘Has the Moment Come to Revise the Arbitration Rules of

UNCITRAL?’, Arbitration International 20, no. 3 (2004): 255.

366 Note of the Secretariat of the International Court of Arbitration of the
International Chamber of Commerce Regarding Correction and Interpretation of
Arbitral Awards, ICC International Court of Arbitration Bulletin 10, no. 2 (1999):
4.
367 Novel Assurance Co. v. Gerling-Konzern General Insurance Co. [2007] EWHC

253 at [46].
368 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 697.

369 For similar statutes to the UNCITRAL Model Law, see English Arbitration Act

1996 s. 57(3)(b); French New Code of Civil Procedure Art. 1485; German Code of
Civil Procedure Art .1058(1) and (3); Singapore Arbitration Act s. 43(4); Swedish
Arbitration Act s. 32.
370 To similar effect are ICSID Rules Art. 49; ICDR Rules Art. 30.1; LCIA Art. 27.3.
371 Note of the Secretariat of the ICC, ‘Correction and Interpretation of Arbitral

Awards’ (1999); Yves Derains & Eric A. Schwartz, Guide to the ICC Rules of
Arbitration, 2nd edn (The Hague: Kluwer Law International, 2005), 323–324.
372 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 695.

373 Even that is debatable as an arbitrator might argue that it always expected to

be able to repair an oversight in the interests of justice and the arbitrator's

reputation.
374 See, e.g., Brown v. Witco Corp., 340 F. 3d 209, 219–220 (5th Cir 2003).
375 See, e.g., Torch Offshore v. Cable Shipping [2004] 2 Lloyd's Rep 446 QB.
376 Jean-François Poudret & Sébastien Besson, Comparative Law of International

Arbitration, 2nd edn (London: Sweet & Maxwell, 2007), 696.

377 Ibid.
378 See, e.g., Judgment of 20 December 2006, Oberlandesgericht München
(Higher Regional Court of Munich) SchiedsVZ 34 (2007) which supported the
imposition of a three week limit so that the tribunal could comply with a sixty-
day deadline. Peter Sanders appeared to suggest that former Art. 37(2) of the
UNCITRAL Rules limited the tribunal from having further hearings on evidence if
an additional award is called for because of the incorporation of the words
‘without further hearings or evidence’: Pieter Sanders, ‘Has the Moment Come to
Revise the Arbitration Rules of UNCITRAL?’, Arbitration International 20, no. 3
(2004): 256. A preferable reading of Art. 37(2) is that it simply put a time limit
on the tribunal in cases where further hearings on evidence will not be called for.
Where they are, it would be difficult to impose a time limit without knowing the
availability of witnesses and the like. In any event, Art. 39 UNCITRAL Arbitration
Rules 2010 does not.
379 Note the ICC Working Party's view to that effect, Yves Derains & Eric A.

Schwartz, A Guide to the ICC Rules of Arbitration (The Hague: Kluwer Law
International, 1998), 301.
380 For similar provisions to the UNCITRAL Model Law, see French New Code of

Civil Procedure Art. 1485; German Code of Civil Procedure Art. 1056.
381 ICSID Convention Art. 52(6).
382 Antoine Biloune and Marine Drive Complex Ltd v. Ghana Investments Centre

and the Government of Ghana, Award of 30 June 1990, in Albert Jan van den Berg
(ed.), Yearbook of Commercial Arbitration Volume XIX (The Hague: Kluwer Law
International, 1994), paras 59– 62 filed in Antoine Biloune et al. v. Ghana
Investments Centre, et al., Case No. 90–2109, United Stated District Court for the
District of Columbia. See also Charles H. Brower, ‘Correction and Completion of
Awards; Enforcement of Partial and Final Awards; Collaboration by Courts for an
Award to Be Effective; Impact of “International Public Policy” on Arbitration’, in
International Arbitration in a Changing World, ICCA Congress Series No. 6, ed.
Albert Jan van den Berg (Deventer: Kluwer Law and Taxation Publishers, 1994),
214.
383 Eric A. Schwartz, ‘The Rights and Duties of ICC Arbitrators’, in The Status of

the Arbitrator, ICC International Court of Arbitration Bulletin: 1995 Special

Supplement, ed. Jean-François Bourque (Paris: ICC Publishing, 1995).
384 See, e.g., ICC Rules 2012 Art. 34(5).
385 Mohammed Bedjaoui, ‘The Arbitrator: One Man-Three Roles: Some
Independent Comments on the Ethical and Legal Obligations of an Arbitrator’,
Journal of International Arbitration 5, no. 1 (1988): 18–19.
386 James M. Gaitis, ‘The Federal Arbitration Act: Risks and Incongruities
Relating to the Issuance of Interim and Partial Awards in Domestic and
International Arbitrations’, The American Review of International Arbitration 16
(2005): 7083 citing Office & Professional Employees Int'l Union v. Brownsville Gen
Hosp 186 F. 3d 326, 331 (3d Cir 1991).
387 James M. Gaitis, ‘The Federal Arbitration Act: Risks And Incongruities
Relating to the Issuance of Interim and Partial Awards in Domestic and
International Arbitrations’, The American Review of International Arbitration 16
(2005): 7083.
388 288 F. Supp. 2d 783 (N.D. Tex. 2003).
389 Gary B. Born, International Commercial Arbitration (The Hague: Kluwer Law

International, 2009), 2516–2517; see also Judgment of 2 November 2000, Swiss

Federal Supreme Court, ASA Bulletin 19, no. 1 (2001): 88, allowing a further
tribunal award to explain that interest awarded meant compound interest.
390 Judgment of 25 May 1992, Fougerolle v. Procofrance, Cour de cassation, Revue

de l'Arbitrage (1993), no. 1: 91–92. See also Emmanuel Gaillard & John Savage
(eds), Fouchard Gaillard Goldman on International Commercial Arbitration (The
Hague: Kluwer Law International, 1999), 919; Jean-François Poudret & Sébastien
Besson, Comparative Law of International Arbitration, 2nd edn (London: Sweet &
Maxwell, 2007), para. 846. Both criticise the decision.
391 See, e.g., Antoine Biloune and Marine Drive Complex Ltd v. Ghana Investments
Centre and the Government of Ghana, Award on Jurisdiction and Liability (27
October 1989) in Albert Jan van den Berg (ed.), Yearbook of Commercial
Arbitration Volume XIX (The Hague: Kluwer Law International, 1994), 21.
392 Charles H. Brower, ‘Correction and Completion of Awards; Enforcement of

Partial and Final Awards; Collaboration by Courts for an Award to Be Effective;

Impact of “International Public Policy” on Arbitration’, in International
Arbitration in a Changing World, ICCA Congress Series No. 6, ed. Albert Jan van
den Berg (Deventer: Kluwer Law and Taxation Publishers, 1994), 215.