You are on page 1of 481

Fisnik 

Korenica

The EU Accession
to the ECHR
Between Luxembourg’s Search for
Autonomy and Strasbourg’s Credibility
on Human Rights Protection
The EU Accession to the ECHR
ThiS is a FM Blank Page
Fisnik Korenica

The EU Accession
to the ECHR
Between Luxembourg’s Search for Autonomy
and Strasbourg’s Credibility on Human Rights
Protection
Fisnik Korenica
Faculty of Law and Criminology
JURI Department
Vrije Universiteit Brussels
Brussels
Belgium

ISBN 978-3-319-21758-1 ISBN 978-3-319-21759-8 (eBook)


DOI 10.1007/978-3-319-21759-8

Library of Congress Control Number: 2015948896

Springer Cham Heidelberg New York Dordrecht London


© Springer International Publishing Switzerland 2015
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of
the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or
dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt
from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, express or implied, with respect to the material contained
herein or for any errors or omissions that may have been made.

Printed on acid-free paper

Springer International Publishing AG Switzerland is part of Springer Science+Business Media


(www.springer.com)
Preface

Let me start with this postulation: ‘The European Union is often considered and
portrayed as a complex institutional structure, on which it is difficult to put a label.’1
The EU’s relationship with international law is even more problematic—probably
something for which there does not exist any label at all. The EU’s relationship with
the European Convention on Human Rights continues to be at the forefront of this
debate. A first note is that the European Convention on Human Rights2 has been one
of the core international successes—probably the most successful—in international
human rights law and practice. Although originally designed to serve primarily as a
benchmark of law, the Convention soon became a benchmark not merely of law,
but also of practice. Due to its rapidly increasing legitimacy and mode of promo-
tional growth, the Convention went on to become a system of law, and is now the
most effective institutional framework for individual human rights protection in
Europe and probably the world.3 With such bifurcated growth taking place, the
Convention became a core instrument of democracy for most of the Western
European nations, and an enlightenment method for the largest part of the nations
that transformed from communism to democracy.
In its natural format, the Convention system was built to serve as an
international-European instrument of human rights law merely for state parties.
With Europe undergoing large reforms of common goals and institutional practices,
most of the western European nations formed and acceded to a more or less a
supranational organization, the European Union.4 As the latter undertook several
reforms which changed its nature from a pure economic organization to an

1
van Rossem (2009), p. 223.
2
European Convention on Human Rights and its Protocols 1950 (Rome)—entered into force on
3 September 1953—as amended by Protocols No. 11 and No. 14. Available at: http://conventions.
coe.int/Treaty/en/Treaties/Html/005.htm.
3
Accord: Helfer (1993), p. 133/4 (The Convention ‘is widely regarded as the most effective
international instrument for the protection of individual rights.’).
4
This being a designation for the latest constitutional name of this Union.

v
vi Preface

organized political system, it necessitated to make its authority covered by human


rights law limitations. While the EU gradually transformed into an organization of
human rights as a means to keep its supreme law steady in the view of its Member
States, the Convention system was somewhat being neglected and ruled out from
the possibility to review the EU’s human rights performance. Callewaert rightly
notes that the EU and ECHR systems have for a long time grown independently.5
The most rigid argument for this development was the legitimate fact that the EU
was not a contracting party to the Convention and was not therefore obliged to
submit to it, unless it saw itself bound by the functional succession6 of its Member
States’ obligations. In terms of effective human rights protection, this has been
especially problematic in some policy areas which have been an exclusive compe-
tence of the EU (e.g. competition policy) and where Member States have not
participated in the implementation of that law (which certainly resulted in abso-
lutely no external human rights control by Strasbourg).7
With the EU increasing its state-like competences and body of human rights
law,8 it became evident that there was an increasing need for the EU to accede to the
Convention for two basic reasons: first, the political reason, to strengthen the
Union’s legitimacy in terms of its international human rights obligations,9 and,
second, the practical reason of equality—to give equal-footing to persons falling
under the scope of jurisdiction of the EU to enjoy the same rights of standing before
the Convention system with those persons in the Convention’s EU Member
States,10 therefore making the European human rights landscape better unified.11

5
Callewaert (2014), p. 13.
6
An example of functional succession may be found at: Court of Justice of EU, International Fruit
Company v Produktschap voor Groenten en Fruit, Joined Cases 21 to 24/72 [1972] ECR 1219,
para. 18; or, see also: Court of Justice of EU, Defrenne v Sabena, Case 43/75 [1976] ECR
455, para. 20; A general account on state succession in international law may be read at: Brownlie
(2003), pp. 633 et seq.
7
Analysis (1997), p. 235.
8
See e.g.: Heringa and Verhey (2011), p. 31/2; See also: Tulkens (2013), p. 2 (‘As a result the
27 Member States of the Union, which, at the same time, are all parties to the Convention either
have lost altogether their capacity to control decisions, hitherto belonging to their jurisdiction, or at
least their jurisdictional freedom has been diminished.’).
9
Sera (1996), pp. 182 et seq; Although there is now a human rights instrument, the Charter of
Fundamental Rights, explicit in the EU Treaties. See on this: Garcı́a (2002), p. 500.
10
Groussot et al. (2011), p. 1/2; See also: Balfour (2007), p. 212; Odermatt (2014), p. 10; Gragl
(2013), p. 93; Contra: Jacobs (2007) (‘. . .while widely regarded as valuable for political and
symbolic reasons, will have rather limited concrete effects on the observance of human rights
standards. The effects will be limited because the ECHR is already accepted as the fundamental
standard of human rights protection in Europe. . .’).
11
Olsen (2009–2010), p. 65; Accord: Balfour (2007), p. 212, therefore removing the current
difference in the interpretation of human rights that currently exists between the two courts; On
the role that the Strasbourg Court has played in the ‘common understanding’ of human rights law
in Europe, see e.g.: Helfer (1993), p. 143; Cf.: Busby and Zahn argue that in the field of social
rights, there are rather huge discrepancies between the standards of Luxembourg and Strasbourg
Court. They argue that it would be really hard to reconcile these two orders, and they propose that
Preface vii

Callewaert rightly notes that ‘a legal system which rejected external supervision of
its compliance with human rights would be a legal order closed in on itself which,
with no input from outside, would be in danger of fossilisation.’12 While the
primary goal was to neutralize criticisms on EU human rights face, the accession
of the EU to the Convention became not merely a necessity, but also a complex task
to be properly addressed. Accession being the core intention,13 there was the
requirement to ensure that such accession will not hinder or impair any of the
core functions or characteristics of the Union,14 either in terms of its relationship
with the Member States’ legal orders or its relationship with international tribunals
that may produce constitutional consequences for the Union’s external features. In
addition, Gragl in this regard argues that accession will finally raise the question of
who will be the last fundamental rights court in Europe: the Luxembourg or
Strasbourg Court. Such question, in Gragl’s view, substantively demonstrates the
conflict that exists between the effective human rights protection and EU law
external autonomy.15
It is important to point out that the EU’s special nature as a more-or-less
supranational organization possessing internal obligations on human rights law—
something not common for international organizations—preconditions the acces-
sion procedure and fields of law that need be regulated through it with several
stipulations. Most of these stipulations would have to preserve the EU law’s
distinguished feature—its internal and external autonomy. Although the preserva-
tion of EU law autonomy remained the core concern,16 there were other decidedly
important problems that could raise tensions not only within the EU institutional

it is only accession may make that compromise possible. See: Busby and Zahn (2013), pp. 14 et
seq.
12
Callewaert (2014), p. 17.
13
Some suggest that instead of accession, a preliminary review procedure—wherein EU Court
would request an opinion from the Strasbourg Court when ECHR questions arise instead of
submitting EU law questions to its jurisdiction for review of compatibility with ECHR—would
better serve the communication between the two courts. See e.g.: Balfour (2007), p. 226 (‘This
mechanism would mean that if the ECJ is faced with a question on the interpretation of the
Convention in the absence of guidance from the Strasbourg Court, then the ECJ should stall the
proceedings and refer the matter for clarification to the ECtHR.’); See also: Joris and
Vandenberghe (2008–2009), pp. 3–4, which shows debates in the Council of Europe Parliamen-
tary Assembly in favour of accession of EU to ECHR as a key moment to enhance human rights
protection in Europe; On the latter, see also: Krüger (2002–2003), pp. 92/3; See also: Gragl (2013),
p. 5, who, referring to Kruger, argues that accession will ‘remove the increasing contradiction
between the human rights commitments requested from future EU Member States and the Union’s
lack of accountability vis-a-vis the ECtHR.’ Making the Convention a condition for potential
EU-membership candidate states seems moot if the Union itself does not accede there (Gragl 2013,
p. 5).
14
See e.g.: Lock (2010), p. 798.
15
Gragl (2013), p. 85.
16
Cf.: Gragl, referring to Lock, rightly notes that it would not be wise to maintain an ‘absolute
legal autonomy’ in face of human rights law and protection such as the Convention system, as that
would not be ‘desirable at all.’ See: Gragl (2013), p. 25.
viii Preface

balance but also in its internal and external constitutional relationships. One such
issue is the share of the burden between the EU and its Member States when they
have jointly contributed to a violation of the Convention, the establishment of such
joint responsibility by the Convention system being another major dilemma in
itself. Offering a favorable environment wherein the Strasbourg regime of law is
not given the chance to compete with the Luxembourg Court,17 whereas the former
does hold a normal external authority to protect human rights, was a further difficult
assignment to be addressed. No one, therefore, may dispute the notable fact that
accession will be a ‘highly exceptional development.’18 However, to put it in
Larsen’s words, ‘[t]he crucial question is who the “we” is in particular policy
areas and what the content, qualities and aims of this “we” are.’19 Therefore,
tackling content and qualities of the accession process remains a core objective
of this book.
The book is divided into four respective parts, altogether forming 12 chapters. In
Part I, the book starts with a brief justification of the research questions raised here
(Chap. 1), by delimiting not only the questions themselves but also the substance of
the issues that will be analyzed. A very short note on the methodology follows
afterwards, accompanied with a section on literature review. Chapter 2 tackles the
EU as a human rights organization—from its inception—and the gradual develop-
ment of its body of human rights law. This section includes an analysis of the
inception of the EU human rights, and how it became embedded into a body of
human rights law deriving not merely from its internal sources of law, but also from
the Convention system. This part also analyzes the initial interaction between the
two, the Strasbourg and Luxembourg regimes of law. Following this, the book
examines the external outlook of EU law, namely the relationship between the
Luxembourg Court and international courts—both from the perspective of their
harmony but also ongoing and natural competition. A specific chapter on EU law
autonomy follows the latter (Chap. 3). Chapter 4 introduces the Final Draft Acces-
sion Agreement (hereinafter referred to as DAA) of the EU to the European
Convention on Human Rights, and introduces the core mechanisms that it estab-
lishes. At this point, the book also questions the extent of EU treaty-making powers
in light of the accession to the Convention system, and examines the internal
consequences that this process produces. In addition, a conceptual explanation on
each of the key provisions of the Draft Accession Agreement is provided therein.
The latter is followed by Chap. 5, which covers the status of the ECHR and DAA
within the EU legal order. The co-respondent mechanism, its nature and legal
construction, and the means and basis on which the two courts are meant to
cooperate and compete are examined in Chap. 6. Chapter 7 examines the inter-
party complaint mechanism after accession, and questions how the EU will be

17
Cf.: Quirico (2010), p. 47, who instead proposes an informal dialogue between the two courts to
prevent such potential conflicts.
18
Odermatt (2014), p. 35–37.
19
Larsen (2009), p. 551.
Preface ix

settled into the new Convention environment for complaints between the
contracting parties inter se. Following this, Chap. 8 examines the prior involvement
review of the Luxembourg Court. This chapter also examines the implications in
terms of the remedies as well as the likelihood that parties will access the Luxem-
bourg and Strasbourg courts effectively. The latter is followed by Chap. 9 which
tests the functionality and sustainability of using the co-respondent mechanism
from the Strasbourg Court’s perspective. Chapter 10 examines the admissibility of
EU-originated applications and potential exceptional scenarios that may appear
from the Strasbourg Court’s point of view. Finally, in Chaps. 11 and 12 the book
first examines the Luxembourg Court’s Opinion 2/13 and, thereafter, concludes
with a summary of the core arguments put forth by presenting definite answers to
the questions raised. Throughout the book, the principle of dubia in meliorem
partem interpretari debent is applied as a means to offer reliable and consistent
arguments.

Brussels, Belgium Fisnik Korenica


May 2015

References

Analysis (1997) The European community cannot accede to the European Convention on Human
Rights. Eur Law Rev 1:235–249
Balfour ADJ (2007) Eliminating conflicting interpretations of the European Convention on Human
Rights by the European Court of Justice and the European Court of Human Rights: the Pdiq
system as a preventative solution. Intercult Hum Rights Law Rev 2:183–247
Brownlie I (2003) Principles of public international law, 6th edn. Oxford University Press, Oxford
Busby N, Zahn R (2013) The EU’s accession to the ECHR: conflict or convergence of social
rights? Paper to be presented at the Labour Law Research Network’s Inaugural Conference
Barcelona, 13–15th June 2013
Callewaert J (2014) The accession of the European Union to the European Convention on Human
Rights. Council of Europe, Strasbourg
Garcı́a RA (2002) The general provisions of the charter of fundamental rights of the European
Union. Eur Law J 8(4):492–514
Gragl P (2013) The accession of the European Union to the European Convention on Human
Rights. Hart Publishing, Oxford
Groussot X et al (2011) EU accession to the European Convention on Human Rights: a legal
assessment of the Draft Accession Agreement of 14th October 2011. Fondation Robert
Schuman/European Issues No 218, 7 November 2011
Helfer LR (1993) Consensus, coherence and the European Convention on Human Rights. Cornell
Int Law J 26:133–165
Heringa AW, Verhey L (2011) The EU charter: text and structure. Maastrich J Eur Comp Law 8
(1):11–32
Jacobs FG (2007) Accession of the European Union to the European Convention on Human
Rights. Hearing organised by the Committee on Legal Affairs and Human Rights in Paris on
11 September 2007. Available at: http://www.statewatch.org/news/2007/sep/jacobs-eu-echr.
pdf
x Preface

Joris T, Vandenberghe J (2008–2009) The Council of Europe and the European Union: natural
partners or uneasy bedfellows? Columbia J Eur Law 15:1–43
Krüger HC (2002–2003) Reflections concerning accession of the European Communities to the
European Convention on Human Rights. Pa State Int Law Rev 21(1):89–99
Larsen H (2009) A distinct FPA for Europe? Towards a comprehensive framework for analysing
the Foreign Policy of EU Member States. Eur J Int Relations 15(3):537–566
Lock T (2010) EU accession to the ECHR: implications for judicial review in Strasbourg. Eur Law
Rev 35(6):777–798
Odermatt J (2014) The EU’s accession to the European Convention on human rights: an interna-
tional law perspective. Leuven Center for Global Governance Studies. Working Paper No. 136
Olsen JB (2009–2010) Protecting fundamental rights and the evolving roles of the Court of Justice
of the European Union and the European Court of Human Rights; Europeanisation in Action at
the Supranational Level. Edinburg Student Law Rev 1(3):54–73
Quirico O (2010) Substantive and procedural issues raised by the accession of the EU to the
ECHR. Italian Yearb Int Law 20:31–53
Sera JM (1996) The case for accession by the European Union to the European Convention for the
protection of human rights. Boston Univ Int Law J 14:151–186
Tulkens F (2013) EU accession to the European Convention on Human Rights. Speech at: National
School of Judiciary and Public Prosecution (KSSIP) – Krakow (Poland), Friday 1 March 2013
van Rossem JW (2009) Interaction between EU law and international law in the light of Intertanko
and Kadi: the dilemma of norms binding the Member States but not the community. Nether-
lands Yearb Int Law 40:183–227
Acknowledgements

First and foremost, I would like to thank the Almighty GOD who gave me the
talent, mind, and audacity to become a scholar of international law of modern times.
I would like to thank my father from whom I inherited the profession of a modern
lawyer. Special thanks to my mother, who always kept educating me to become a
hard-working person that goes beyond borders. I must say, I would not have had the
motivation to read the Ph.D. degree without the 4-year guidance, teaching, and
careful inspiration which my supervisor Professor Paul De Hert provided to me. He
never stopped teaching me on how to read this degree with profundity and academic
maturity. Special thanks go to two other professors, members of the Ph.D. Com-
mittee, Professor Emmanuelle Bribosia and Professor Stefaan Smis, who dedicated
their personal time to guide and review my drafts consistently and with deep
academic analysis. I know I have changed rather often their personal agendas to
deal with my thesis and Committee meetings (I am sorry). My sisters (Elinda and
Genita, and my two nephews and two nieces) have all kept inspiring me to do this
project. I must specially thank my friends, Dren, Artan, Agon, Lorik, and Argjend
who kept encouraging me to become a Ph.D. in Law. Finally, I must also thank
Emma Founds, my colleague at GLPS, who proofread this book. This book is
essentially the Ph.D. dissertation which I have defended before the Jury at Vrije
Universiteit Brussel in March 2015.

xi
ThiS is a FM Blank Page
Contents

Part I The EU as a Sui Generis Human Rights Law Organization:


Situating the Roots of the Accession Question
1 Introduction to the Book . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.1 Delimitating the Questions of the Book and the Scope of
Substance Analyzed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.2 A Note on the Methodology . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.3 Setting the Scene for the Book’s Structure . . . . . . . . . . . . . . . 12
1.4 Introduction to the Individual Chapters of the Book . . . . . . . . 17
1.5 A Theoretical Survey on Competing International Jurisdictions
and Treaty Laws: Opening the Literature Box on the Wider
Topic of the Book . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
1.5.1 Proliferation of International Treaty Regimes and
International Courts: What About a ‘New-Fangled’
International Law? . . . . . . . . . . . . . . . . . . . . . . . . . . 22
1.5.2 Towards a Global Law Rule? . . . . . . . . . . . . . . . . . . 27
1.5.3 EU and ECHR (Under the Title of European Continent)
Slice in the Global Law Landscape? . . . . . . . . . . . . . 30
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
2 EU Becoming a Human Rights Law Organization: Starting from
Nowhere with a ‘gouvernment des juges’ . . . . . . . . . . . . . . . . . . . . . 35
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
2.2 The Origins of EU Becoming a Human Rights Law
Organization with Reference to the ECHR . . . . . . . . . . . . . . . 36
2.3 Strasbourg Looking Towards Luxembourg: What About
a Refined Legal Arrangement? . . . . . . . . . . . . . . . . . . . . . . . . 49
2.4 EU Charter on Fundamental Rights and Its Normative
Relationship with the ECHR . . . . . . . . . . . . . . . . . . . . . . . . . 57
2.5 Chapter’s Summary of Conclusions . . . . . . . . . . . . . . . . . . . . 65
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

xiii
xiv Contents

3 EU Law Autonomy: Where Does the Viewpoint for ‘Competition’ of


Luxembourg Start From? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
3.2 ‘EU Law Autonomy’: What Does It Actually Mean? . . . . . . . 72
3.3 Autonomy in the ‘European Way’: Tracing Its Origins and
Discussing Its Rationale . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
3.3.1 Costa ENEL: ‘EU Law Autonomy’ Where the Whole
Story Began . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
3.3.2 Opinion 1/91: CJEU Says No for Two Courts Under
the Same Roof . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
3.3.3 MOX Plant: When Jurisdictions Collide CJEU Should
Decide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
3.3.4 Kadi: CJEU Policing EU External Borders . . . . . . . . 78
3.4 Chapter’s Summary of Conclusions . . . . . . . . . . . . . . . . . . . . 82
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

Part II The Draft Accession Agreement of the EU Accession to the


ECHR: An Examination of the Central Mechanisms in Light of
EU Law Pecularities
4 A New Start for the Accession of the EU to the ECHR . . . . . . . . . . 89
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
4.2 Background on EU Accession to ECHR from a Treaty and
Human Rights Law Perspective . . . . . . . . . . . . . . . . . . . . . . . 90
4.3 Can EU Be a Master of Treaty in ECHR? . . . . . . . . . . . . . . . 95
4.4 ECHR Becomes a Hybrid and Complex Treaty System? . . . . . 98
4.5 Accession Enables the EU to Enjoy the Benefit of a primus inter
pares Position . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
4.6 The Accession Model and Possible Implications . . . . . . . . . . . 103
4.7 Outline of the Draft Accession Agreement of the EU to the
ECHR: What Substantial Issues Does It Address? . . . . . . . . . . 108
4.8 Chapter’s Summary of Conclusions . . . . . . . . . . . . . . . . . . . . 126
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
5 Status of ECHR and DAA in EU Legal Order . . . . . . . . . . . . . . . . 133
5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
5.2 Examining the Status of ECHR and DAA in EU Legal
Order . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
5.2.1 The Status of International Agreements in EU Legal
Order . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
5.2.2 Analyzing the Hierarchical Position of the Convention
and Accession Agreement in EU Legal Order . . . . . . 141
5.2.3 Five Specific Arguments: Shaping More Concretely
This Undeveloped Relationship . . . . . . . . . . . . . . . . . 145
5.3 Chapter’s Summary of Conclusions . . . . . . . . . . . . . . . . . . . . 157
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
Contents xv

6 Attribution of Liability Under the Co-respondent Mechanism . . . . 163


6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
6.2 A Background on the Complex Task of Attributing Liability in a
Post-accession Scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
6.3 EU Treaty-Based Provisions: Which Were the Initial
‘Peculiar’-Related Instructions for DAA? . . . . . . . . . . . . . . . . 168
6.4 The Differences Between Third-Party Interventions and the
Co-respondent Mechanism: Where Does the DAA
Go Blurry? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
6.5 The Core Function of the Co-respondent Mechanism: What Is It
Made for? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
6.6 A Comparison Note Between DAA’s Co-respondent Mechanism
and DARIO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
6.7 Friendly Settlements and Unilateral Declarations: Any
Interference in the EU Law Autonomy? . . . . . . . . . . . . . . . . . 192
6.8 EU and Member States as Co-respondents . . . . . . . . . . . . . . . 195
6.8.1 EU as Co-respondent: Examining Its Normative
Architecture and Potential Implicative Legal
Outcomes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195
6.8.2 Member State(s) as Co-respondent(s): Examining the
Normative Architecture and Potential Implicative Legal
Outcomes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
6.8.3 EU and Member State(s) as Joint Respondents: Still a
Possible Scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
6.9 Ambiguities in the Co-respondent Mechanism: Why Is It So
Blurred? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
6.9.1 Ambiguity 1: The Discretionary Nature of the
Co-respondent Mechanism . . . . . . . . . . . . . . . . . . . . 215
6.9.2 Ambiguity 2: Strasbourg Court’s Plausibility . . . . . . . 218
6.9.3 Ambiguity 3: Share of Burden Between the EU and
Member States When They Appear as (Co)respondents
Jointly . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220
6.9.4 Ambiguity 4: Lack of the Right Addressee—No
Answer—Political Consensus . . . . . . . . . . . . . . . . . . 226
6.10 Referral to the Grand Chamber: Is There Space for Divorce
Between (Co)-respondents? . . . . . . . . . . . . . . . . . . . . . . . . . . 227
6.11 A Comparison on the Model of Sharing the Liability Between
DAA, UNCLOS and UNCILDCSO: Which Are the Strengths
and Weaknesses of DAA? . . . . . . . . . . . . . . . . . . . . . . . . . . . 230
6.12 Chapter’s Summary of Conclusions . . . . . . . . . . . . . . . . . . . . 234
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
xvi Contents

7 Inter-Party Mechanism and the EU: Possible Implications from


the Strasbourg’s Jurisdiction? . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
7.2 A Background on the Inter-Party Complaint Mechanism with a
View to EU-ECHR Peculiar Context . . . . . . . . . . . . . . . . . . . 242
7.3 The Possible Impairment of the Luxembourg Court’s Exclusive
Jurisdiction by the DAA: A Perspective on the Inter-Party
Mechanism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246
7.4 Prohibition of Protocol 8(3) and Art 344 TFEU: Is There an
‘Intended’ Understanding? . . . . . . . . . . . . . . . . . . . . . . . . . . . 247
7.5 ‘Escaped’ Exclusivity in Inter-Party Complaints: Art 344 TFEU
in Light of Art 19(1) TEU and with Reference to Art 275 TFEU:
What Would This Entail in Terms of the DAA? . . . . . . . . . . . 250
7.6 Inter-Party Cases in Light of Mox Plant and Art 344 TFEU:
Which Standards Derive Thereof? . . . . . . . . . . . . . . . . . . . . . 253
7.6.1 The First Test: Mix Agreement or Not: Defining the
Attribution of Competence? . . . . . . . . . . . . . . . . . . . 255
7.6.2 The Second Test: Parallel (Mirrored) Provisions . . . . 259
7.6.3 The Third Test: It Is for the Luxembourg Court to
Delineate Its External Jurisdiction Borders in Each
Case . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260
7.6.4 The Fourth Test: Use of Assurances Not Allowed . . . 261
7.7 Beyond the Conventional Concept on Competing Jurisdictions
of the Two European Courts: ‘Reconciling’ Art 55 ECHR with
Art 344 TFEU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262
7.7.1 Conditions for the Special Agreement: What About
More Details? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265
7.7.2 Questioning Whether EU Treaties Could Serve as a
Special Agreement in Light of Art 55 ECHR? . . . . . . 267
7.7.3 Accession Agreement as a ‘Special Agreement’ Under
Art 55 ECHR? A Second Try. . . . . . . . . . . . . . . . . . . 270
7.7.4 The Effect of Art 5 (Second Clause) of the DAA to the
Relationship Between Art 344 TFEU and Art 55 ECHR:
Why Is the ‘Special Agreement’ Special? . . . . . . . . . 272
7.7.5 An Additional, More Hypothetical Explanation on the
Effect of Art 5 of the DAA on Art 344 TFEU (A Second,
More Constructive Scenario)? . . . . . . . . . . . . . . . . . . 276
7.8 Testing the Inter-Party Procedure to Ireland v. UK: What Does It
Mean in Practice? From Theoretical to a Practical Scenario . . . 282
7.8.1 Testing the Case . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284
7.9 Inter-Party Procedure as Room for the Strasbourg Court to
Interpret EU Law: Where Does the Risk for EU Law Autonomy
Stand in Exceptional Cases? . . . . . . . . . . . . . . . . . . . . . . . . . 287
7.10 Chapter’s Summary of Conclusions . . . . . . . . . . . . . . . . . . . . 290
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292
Contents xvii

8 EU Prior-Involvement Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295


8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
8.2 A Background on the Subsidiary Nature of the Convention
System from the Perspective of EU as a Party to the Latter . . . 297
8.2.1 Limited Access to Justice Under EU-Law Direct Actions:
In the Borders of Violation of the Right to Access the
Court Through an Effective Legal Remedy . . . . . . . . . 299
8.2.2 Whether Preliminary Reference Procedure Under
Art. 267 TFEU Complies with the ECHR Standard on
Access to Court Through Efficient Legal Remedies? . . . 303
8.3 An Examination of the Prior Involvement Mechanism and Its
Intended Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317
8.4 Conditions to Initiate the Prior Involvement: A Rather Complex
Task That May Inhibit Jurisdictional Allergies Between the
Two Courts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324
8.5 Prior Involvement of the Luxembourg Court: A Procedural
Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 330
8.6 Refraining from Undue Delay: How Can This Be Met? . . . . . . 337
8.7 Legal Effects of the Prior Involvement Procedure on the
Reviewed EU-Law Provisions . . . . . . . . . . . . . . . . . . . . . . . . 340
8.8 Is Prior Involvement a New Remedy: Defending the
Non-defendable? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 343
8.9 Does Prior Involvement Produce the Effect of a Hidden
Amendment to the Treaties? . . . . . . . . . . . . . . . . . . . . . . . . . 347
8.10 Chapter’s Summary of Conclusions . . . . . . . . . . . . . . . . . . . . 349
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 350

Part III A Strasbourg Perspective on Applications of EU-Law Origin


9 Testing the Co-respondent Mechanism from the Strasbourg Court’s
Perspective: Three Distinctive Cases with Three Distinctive
Scenarios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 357
9.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 357
9.2 A Starting Note to the Three Tests . . . . . . . . . . . . . . . . . . . . . 358
9.2.1 Testing Bosphorus with the Co-respondent
Mechanism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 358
9.2.2 Testing Mathews with the Co-respondent
Mechanism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 363
9.2.3 Testing Kokkelvisserij with the Co-respondent
Mechanism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 367
9.3 Chapter’s Summary of Conclusions . . . . . . . . . . . . . . . . . . . . 375
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 376
xviii Contents

10 Admissibility Before the Strasbourg Court: An Outlook on the


EU-Law-Originated Applications . . . . . . . . . . . . . . . . . . . . . . . . . . 379
10.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 379
10.2 A Notional Start: What Does One Need to Do to Reach the
Strasbourg Court? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 380
10.2.1 The Victim Status of the Claimant . . . . . . . . . . . . . . 381
10.3 Exceptions to the Rule on Exhaustion of Domestic Legal
Remedies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 383
10.3.1 Strasbourg’s Elastic Approach to the Rule on
Exhaustion: The EU System of Remedies May Become
Surpassed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384
10.3.2 There Are Several EU Remedies in Place: Which One
to Exhaust? A Question of Rationality and
Effectiveness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387
10.3.3 The Complaint Must Have Been Substantively Raised
Through the Domestic Legal Remedies: Does This
Apply to the Preliminary Reference Procedure? . . . . . 389
10.4 The Nature of the Strasbourg Court’s Rulings on EU-Related
Matters: Is There Space for the Supremacy of Strasbourg on
Luxembourg? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 392
10.5 Violations Originating in EU Primary Law: A Question of
Possibility to Challenge the Treaties at Strasbourg? . . . . . . . . 395
10.6 The New Life of Bosphorus Post-accession: A Rational
Viewpoint from the Strasbourg’s Lens . . . . . . . . . . . . . . . . . . 398
10.7 Chapter’s Summary of Conclusions . . . . . . . . . . . . . . . . . . . . 400
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 401

Part IV Approaching the Final ‘Station’


11 Before the Conclusion: Luxembourg Court’s Opinion 2/13 on the
DAA’s Compatibility with the EU Treaties . . . . . . . . . . . . . . . . . . . 407
11.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 407
11.2 Accession Shall (Should Not?!) Bring Significant Constitutional
Changes to the Treaty System . . . . . . . . . . . . . . . . . . . . . . . . 410
11.3 Art. 53 of the Charter (Un)coordinated with Art. 53 of the
Convention: Fighting for Internal Primacy? . . . . . . . . . . . . . . 411
11.4 Uniform Interpretation of EU Law and Mutual Trust Between
EU Member States May Not Be Jeopardized by the Convention
System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 414
11.5 Risks from the Application of Protocol 16 ECHR: Too Many
Doubts Being Raised? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 416
11.6 Art. 344 TFEU in Risk from the DAA: What About Excluding
Inter-Party Mechanism? . . . . . . . . . . . . . . . . . . . . . . . . . . . . 418
11.7 Two Small Threats from the Co-respondent Mechanism That
Need Be Addressed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 419
Contents xix

11.8 Prior Involvement Mechanism: The Need to Add Another Layer


of Safeguard . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 422
11.9 CFSP Measures Before the Strasbourg Court: Luxembourg
Court in a Panic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 423
11.10 The Technical Changes That Need Be Addressed in the Draft
Accession Agreement to Make It Compliant with Opinion 2/13
Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 424
11.11 A Closing Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 426
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 426
12 An Overall Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 427
12.1 A General Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 427
12.2 Conclusion on the Overall Functionality of the Accession
Agreement and Its Outputs . . . . . . . . . . . . . . . . . . . . . . . . . . 428
12.3 Searching for a Theoretical Model to Explicate the Accession
Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 434
12.4 Post-accession (Forthcoming) Perspectives: What About a New
Normative Order in Europe? . . . . . . . . . . . . . . . . . . . . . . . . . 435
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 438
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 439
Part I
The EU as a Sui Generis Human Rights
Law Organization: Situating the Roots
of the Accession Question
Chapter 1
Introduction to the Book

1.1 Delimitating the Questions of the Book and the Scope


of Substance Analyzed

This book presents a very specific and narrow approach to the core questions of the
EU accession to the ECHR (compare Fig. 1.1 to Fig. 1.2). First of all, it is important
to mention the fact that there is rather limited and mostly general literature—if a few
articles and two topic-specific books might be described as literature—covering the
Draft Accession Agreement of the EU accession to the ECHR, most of which have
been published some time ago to be relevant today. Therefore, as this topic is new
this book attempts to consult not merely every possible source on the issue, but also
intends to build upon them to produce a novel scientific result at the end of this
research project. One assumption nevertheless needs be made: the novelty of the
topic itself does not reduce the scientific quality that the arguments need to reflect.
Furthermore, the book—at some points and in a rather limited framework—takes on
board the task of examining not only how things stand at the theoretical level
regarding implications of EU accession to the ECHR, but also how they might
(de lege ferende) become practically exposed to the current and upcoming legal
implications on this field of law and practice. Therefore, central attention is given to
examining the factual problems and/or benefits that will result from EU accession to
the ECHR. This book aims to provide new, more developed knowledge in the field,
and assess concerns within advanced argumentative frameworks to elucidate the
mechanics and legal effects of EU accession to the ECHR.

© Springer International Publishing Switzerland 2015 3


F. Korenica, The EU Accession to the ECHR, DOI 10.1007/978-3-319-21759-8_1
4 1 Introduction to the Book

Fig. 1.1 The old jurisdictional picture on the relationship between the EU and ECHR systems

Fig. 1.2 The post-accession jurisdictional picture on the relationship between the EU and ECHR
systems

As this book tackles the notion of competition and cooperation between the
Luxembourg and Strasbourg regimes of law1 within the framework of EU accession

1
Cf.: Joris and Vandenberghe (2008–2009), p. 2, who asserts that EU and Council of Europe are
natural partners; On the argument that accession will reconcile the two courts, see: Balfour (2005),
p. 22; On the potential conflicts between the two courts, see also: Defeis (2000–2001), p. 317; See
also: Wetzel (2003), p. 2843, which takes Hoechst and Konstantinidis v. StadtAltensteig-
Standesanicases to demonstrate how Luxembourg Court and Strasbourg Court may prove diver-
gent at interpreting identical rights.
1.1 Delimitating the Questions of the Book and the Scope of Substance Analyzed 5

to the ECHR, the vast majority of the analysis encapsulates merely the central
issues and problems of this topic, while remaining cautious that existence of such
competition may erode trust in the rule of law in the EU and in Europe as a whole.2
Other, more minor issues are left aside and not included in the analysis. With this
being noted, the core research question that this book poses is: What is the legal
nature and scope of effect of the cooperation and/or competition between the
Strasbourg and Luxembourg regimes of law in the specific context of EU accession
to the ECHR (in the framework of the DAA)?3 The bigger research question,
therefore, seeks to solve the problem of how the ECHR would be able to accom-
modate a modus operandi whereby the EU does not become allergic in its relation-
ship with the Strasbourg Court, whereas the aim of human rights protection is not
compromised. To answer this question, this book will: (a) design and validate an
adequate doctrinal structure that provides a legal-positive examination on the core
issues relating to the DAA and accession process at large, (b) explore and appraise
the current and upcoming regulation of the relationship between the two regimes of
law, (c) present clear arguments in relation to the principles and guidelines which
may elucidate the understanding and positivist application of the DAA, (d) produce
a logically and theoretically validated comprehensive framework for identifying
problems and implicative outcomes that the two legal regimes may face once

2
See e.g.: Olsen (2009–2010), p. 56.
3
Paul Gragl’s book on this topic has a rather different research question, namely ‘whether and how
accession and the system of human rights protection under the Convention can be effectively
reconciled with the autonomy of European Union law.’ See: Gragl (2013), p. 8/9. Although Gragl
does not provide in his book each chapters’ specific research questions—namely, the subsidiary
research questions to the central research question—one may understand that he undertakes a
normative burden to show paths of reconciling both legal regimes. Contrary to this, my central
research question—and the specific/subsidiary research questions—have another purpose: that of
examining the effect of cooperation/competition between the two regimes of law in the context of
EU accession to ECHR (and specifically to the DAA), something that centrally covers also the
examination of a) autonomy of EU law in the context of the Convention’s credibility of human
rights protection, and, b) the functionality of the DAA mechanisms in light of the proclaimed
objectives of both legal orders and the DAA itself. My book, therefore, is not that centrally
concerned about the ‘reconciliation’ of EU law autonomy with the Convention’s human rights
protection, but rather with the examination of loopholes where that autonomy may become
encroached, in addition to the question of functionality of the DAA mechanisms (which not
always triggers the question of autonomy). The question of my book being more about the
examination of the nature and scope of ‘effect’ that will be produced as a result of cooperation/
competition between the two legal orders in the context of the DAA, one may rightly argue that it
is moderately different in many aspects with Gragl’s research question and intended outcome. In
terms of outcome, therefore, these two books come to rather different general conclusions: while
Gragl, on basis of his research question, finds way to reconcile and concludes that the DAA does
not interfere to EU law autonomy, my book concludes rather the opposite, showing where
loopholes remain both in terms of EU law autonomy concerns but also impaired-functionality
concerns. One final difference between the two books is the fact that Gragl looks at the DAA very
much from a micro perspective, while I also look at it from a macro perspective, taking account of
similar experiences and benchmarks from international law and courts (and global law) which
Gragl does not.
6 1 Introduction to the Book

accession becomes a reality, and, (e) offer theoretical and positivist solutions to
these implications with a view to sustaining the proclaimed objectives of the
accession process and project. This certainly leads to more detailed research
questions which will seek to expound on the theoretical and practical mechanics
that form the basis for cooperation and/or competition between the Strasbourg
Court and Luxembourg Court. These issues will be organized and functionally
established, so that the bigger picture regarding the cooperation and/or competition
concerned is scrutinized at its origin, and thoroughly considered when assessing the
consequence(s) that it produces. This definitely leads to more substantial—one may
also call subsidiary—research questions that this book raises in substance
(explained and separately written at the beginning of each chapter): whether and
how EU law autonomy will be preserved once the EU accedes to ECHR, and how
potential challenges stemming from Strasbourg on its autonomy may be neutralized
or counterbalanced? What is the nature of legal effects that the ECHR system will
produce upon the EU legal order, the latter’s court jurisdiction presumably being
immunized from external jurisdictional influence or attack? What is the position of
the ECHR and Accession Agreement of EU to ECHR in the EU legal order, and
what are the would-be mechanisms to maintain them ‘obedient’ to the Treaties
(if any)? What is the scope of self-restraint that the Strasbourg regime would accept
in order to keep Luxembourg’s autonomy protected, and the possible guarantees
which may assure passive jurisdiction of the former on the latter? How may the
distribution of burden on ECHR violations be shared between the EU and its
Member States, and what functional role may/should the Strasbourg Court play?
What is the level to which the subsidiary nature of the Strasbourg Court will be
maintained in the face of the EU? Will the mechanisms resulting from accession
assure the same degree of human rights protection for which the ECHR system has
been established and demonstrated to date? And, overall, how will the EU’s
external perspective change as a result of its accession to the ECHR, both within
the context of its attitudes toward ‘stateness’ and with regard to mandatory sub-
mission to international law?
Both the bigger/top research question and the subsidiary/subordinate research
questions aim to portray the systemic and functional outlook and changes that the
EU and ECHR will interdependently witness once the EU accedes to the ECHR,
with the scope of such effects questioned against the effectiveness of human rights
protection that the ECHR system ought to assure. Therefore, this book examines the
core components of the accession procedure and outcomes, that are: the position of
ECHR and the DAA in the EU legal order, the nature and effects of the DAA on the
EU and ECHR itself, the mechanisms provided for preserving the autonomy of EU
law in the face of the ECHR system, the means of burden sharing between the EU
and Member States in the face of Convention violations, and the mechanisms to
ensure that the Strasbourg Court does in fact remain a subsidiary court even in front
of the EU. Some of these substances are examined with deeper scrutiny—some with
more conventional analysis—as the primary aim is to provide for deeper assessment
in the parts wherein one can observe scarcer knowledge and literature on this topic.
1.1 Delimitating the Questions of the Book and the Scope of Substance Analyzed 7

It is also important to note that the book operates on three foundational hypoth-
eses and parameters: first, that the EU legal order will experience substantial
changes—at least in legal conceptual context—with its accession to the ECHR
and submission to the Strasbourg Court review, which Thym calls a “Trojan
Horse”4 to the EU legal order (Hypothesis 1—H1). This hypothesis finds the
support in the reasoning of Luxembourg Court’s Opinion 2/94, which had noted
that accession will be of significant constitutional impact to the Union’s constitu-
tional architecture5; second, that the EU’s growing submissive approach towards
the ECHR—which is a core international law instrument for Europe—implies its
increasing stateness attitude that reflects a better embodiment with sovereign acting
features (Hypothesis 2—H2). This view is supported by the fact that accession will
be a novel development in international law, as the EU is undertaking international
obligations in a field of law that was previously a state-reserved domain of law.
Interacting with international obligations at that level will push the EU towards
fortifying its ‘stateness’ identity in international law. This certainly implies the
EU’s increasing ‘stateness’ attitude; and third, that the EU’s accession to the ECHR
will provoke substantial challenges to the Luxembourg Court’s primary and exclu-
sively leading role in the EU hemisphere, and the increasing primacy of the
Strasbourg Court—which is approaching human rights law headship—regarding
fundamental rights jurisdiction in Europe and upon the EU as well (Hypothesis 3—
H3). This hypothesis, e.g., is regarded as a general attitude towards the accession by
Callewaert,6 who has generally argued that accession will position the Strasbourg
Court as a supreme court in relation to the Luxembourg Court, with the former
taking the leadership of human rights law jurisdiction in the European continent.
Luxembourg Court’s President, Judge Skouris, had supported this same proposition
in 2002 by arguing that accession will limit to certain extent EU law autonomy. He
has argued in that sense that ‘[r]egarding the Court of Justice in particular, it will
effectively lose its sole right to deliver a final ruling on the legality of Community
acts where a violation of a right guaranteed by the ECHR is at issue.’7 This book
therefore will test these three general hypotheses by answering the larger and
subsidiary research questions asserted above. In undertaking this research, the
book will tackle these topics with a rather exclusive ‘legal’ eye—a viewpoint that
will make the argument more credible and the research answers more reliable.

4
Thym (2013), p. 1.
5
Cf.: Gragl, from a different perspective, comes to the conclusion that EU accession to ECHR
‘will have an unprecedented and enormous impact on the existing multi-level framework of human
rights protection in Europe [. . .]’. See: Gragl (2013), p. 278.
6
Callewaert (2014), p. 22.
7
Quoted from: Barbera (2012), p. 9.
8 1 Introduction to the Book

1.2 A Note on the Methodology

The book uses several methods to elucidate the research questions and convey the
issues into the framework of research. It mainly follows a legal positivist approach
to examining the problems and explaining the relationship between the two legal
orders, namely Luxembourg and Strasbourg. Therefore, the book carries out the
research mainly by examining the law as it is. One may legitimately ask why a legal
positivist approach has been primarily chosen in this case. Two basic reasons exist
for choosing this core approach: first, there is no dispute over the fact that the
relationship between the Luxembourg and Strasburg regimes of law within the
context of the DAA is so recent that nothing has changed in practice as of yet. A
foundational examination on this issue—but also a basic knowledge inquiry—
needs be made on a legal positivist basis first, in order to open ways for other
methodology works later on. Second, it would be too speculative at this stage of
knowledge on this issue to pull the research on the integrity of the regulation of this
relationship between the two regimes based on the DAA without there being an
empirical evidence of how both courts interact and form their own human rights
protection identity with post-accession case-law.
However, in some limited instances, this book also strives to deconstruct the
justification for certain rules’ existence, and their intended integrity output.
Although it is not the intention of this book to embark outside the positivist debate,
the author often offers arguments in relation to the justification of certain rules
provided in the DAA, in order to make the argument more plausible and to propose
an enhanced holistic approach to the arguments presented. It is agued here that an
absolute positivist approach to this topic would not make the overall picture of the
DAA complete. Two reasons exist for this: first, the DAA was construed in light of
certain political objectives of the EU,8 which needed to be reflected in view of the
mechanisms established by the Agreement itself and their intended output, and,
second, the DAA has left certain intentional gaps in order to leave certain legal
questions answerable to the political momentum of cooperation between the two
treaty orders. This said, this book, especially at the beginning, portrays and exam-
ines the rationale behind some of the core legal principles established by the DAA,
and reflects on their overall legitimizing effect towards the EU and the general
European pluralist human rights landscape. This approach has been applied in a
very limited context and only where the author thought that it is indispensible for
the uniqueness of the book to build in that direction as well.
In addition, this book often relies on the comparative method to contrast
comparable situations, norms and analytical results that relate to the EU-ECHR
topic. This book will—to that end—observe to what degree legal principles devel-
oped within one legal order may perhaps be of advantage to the other. In particular,
the comparative method has been regularly applied against some international

8
On the latter, see e.g.: White (2010), p. 435; See also: Jones (2012), p. 5; Odermatt (2014), p. 9.
1.2 A Note on the Methodology 9

instruments such as the American Convention on Human Rights, Statute of the ICJ,
the International Covenant on Civil and Political Rights, etc., and a rich body of
caselaw deriving from courts and tribunals established by those instruments. Com-
parative case-law of US Supreme Court has also been used to show how the federal
states comply with international human rights law obligations, and contrast them
with the EU’s supranational engagement in external relations. The selection of
these instruments and caselaw of their courts has been made on the basis of their
weight in the pool of international human rights law. The two Vienna conventions
on the law of treaties9 have been constantly used to make this even better fitted to
the international law debate. Likewise, two similar instruments have been consis-
tently used as comparative methods but also sources of international law that apply
in the EU-ECHR relationship, namely the Articles on the Responsibility of States
for Internationally Wrongful Acts10 and the Draft Articles on the Responsibility of
International Organizations.11 Without the comparative use of these instruments it
would have not been possible to root this book in a global law discourse. Finally, it
must be noted that while examining whether to choose a mixed-method approach,
the author consulted a non-exhaustive list of sources covering this topic and similar
courts and tribunals. A general conclusion was that there was almost no study that
was built merely on a one-method approach, hence this book reflects those expe-
riences and tries to build the methodology in the same light.
This book essentially pursues the theoretical observation of Neil MacCormick,
who argues that ‘[. . .] the most appropriate analysis of the relations of legal systems
is pluralistic rather than monistic, and interactive rather than hierarchical.’12 That
assumed, the book does not intend to argue for a certain hierarchical relationship
and for a one-sided approach to accession issues. It rather builds upon a pluralistic
and interactive landscape of legal understandings and arguments, in order to show a

9
Vienna Convention on the Law of Treaties between States and International Organizations or
between International Organizations. Done at Vienna on 21 March 1986. Not yet in force. Official
Records of the United Nations Conference on the Law of Treaties between States and International
Organizations or between International Organizations, vol. II (United Nations publication, Sales
No. E.94.V.5); and, Vienna Convention on the Law of Treaties. Signed at Vienna, on 23 May 1969
(UN Doc. No. 18232). Entered into force on 27 January 1980. Available at: https://treaties.un.org/
doc/Publication/UNTS/Volume%201155/volume-1155-I-18232-English.pdf.
10
ILC Responsibility of States for Internationally Wrongful Acts 2001 (ILC Articles on State
Responsibility). Text adopted by the Commission at its fifty-third session, in 2001, and submitted
to the General Assembly as a part of the Commission’s report covering the work of that session.
The report, which also contains commentaries on the draft articles, appears in Yearbook of the
International Law Commission, 2001, vol. II (Part Two). Text reproduced as it appears in the
annex to General Assembly resolution 56/83 of 12 December 2001, and corrected by document
A/56/49(Vol. I)/Corr.4.
11
ILC Draft Articles on Responsibility of International Organizations. Adopted by the Interna-
tional Law Commission at its sixty-third session, in 2011, and submitted to the General Assembly
as a part of the Commission’s report covering the work of that session (A/66/10, para. 87). The
report will appear in Yearbook of the International Law Commission, 2011, vol. II, Part Two.
12
Cormick (1995), p. 264.
10 1 Introduction to the Book

more dynamic map of the interaction between the two systems post-accession
internally and externally, as well as from the Member States’ legal orders perspec-
tive. Such dynamic framework contributes to making the understanding of func-
tionality of the DAA more suitable with regard to the practical interface that one
will witness between the two regimes of law post-accession. Methodologically, this
book endeavors to make abstraction where feasible, therefore providing the reader
not only with practical analysis, but also with higher-level conceptual accounts.
Seeing that EU accession to the ECHR is an essential indication of the concept of
legal pluralism in Europe and beyond, this book operates on a level of illustration
wherein the diversity of forms of law within each layer of governance are given
appropriate consideration. That being the standpoint, this book will operate through
clarifying areas of overlap, inconsistency and ambiguity in the architecture of
human rights law from the perspective of the interaction between the EU and
ECHR. In addition, this book takes into account the EU and ECHR’s distinct
approaches in strengthening or softening their response with regard to human rights
issues. Such tactics will be judged against the possibility to make the interaction
between these layers of human rights law consolidated and integrated from the
perspective of their normative development.
This book builds upon a non-exhaustive list of sources available, starting from
the core legal acts that establish the constitutional foundations of the two legal
regimes, the DAA ‘package of acts’, secondary-level legal acts of both organiza-
tions, case-law of both courts, and most importantly, a large scope of literature
covering the relationship between EU and Council of Europe, accession agreement,
the legal nature of the jurisdictional portrays of the two courts, but also literature on
the competing jurisdictions of international tribunals from a global law perspective.
This book makes no departure from the perspective of researching all cases
contained in the topic’s applicable time-frame, international relevant courts and
jurisdictional levels which are associated to the research questions. Political docu-
ments of the institutional bodies of both organizations have been used to examine
the rationale upon which the foundations of the regulatory framework on the
relationship between the two courts have been used. Several policy reports have
been considered to ensure that this book builds legal arguments from an informed
policy perspective, with the EU and Council of Europe being core policy-makers of
human rights law in Europe’s appealing legal-pluralism environment. All these
discussions have not been made merely in the body of the text, but also in the
footnotes attached to the main text. To note—finally—with the purpose of using
semi-structured interviews, the author has contacted a number of policy-makers
directly engaged in the negotiating process of the DAA and their institutions’ media
officers, and they have altogether refused to answer to our delicate questions. The
refusal to respond to our interview inquiries may be an indication of the sensitivity
of the issue, and their lack of willingness to open discussions on issues which may
seem problematic from an academic point of view.
It must be noted here that—methodologically—it is not the intention of this book
to introduce a new theory on the relationship between the two courts in light of the
DAA. Parenthetically, there is no single or authoritative theoretical model that
1.2 A Note on the Methodology 11

explains the past relationship on a safe ground. It is not the author’s intention to
innovate one such theory. However, provided that the DAA is so recent—even not
yet enforced legally—the author opined that there is a need to first examine the core
components of this agreement from a positivist point of view, combined with scores
of explanations on the rationale for certain mechanical choices that were made in
the agreement concerned. Having examined the core components of the Agreement
and their functional applicability, this book then examines the loopholes which may
seem to exist within the context of potential implications that may harm the
intended objectives that the EU and ECHR have proclaimed that they expect
from this process. However, taken as a whole, the conclusions of this book build
a slight theoretical layer of explaining the functionality of the DAA and the legal
nature of many of its mechanisms, therefore offering a solid theoretical model of
explaining and estimating the functioning of the relationship between the two
courts post-accession. It will—in many circumstances—also provide guidance on
the acceptable degree of divergence between the two legal orders, at the same time
as offering strategies to accommodate and concurrently manage the scale of
disparity between their values and mechanisms.
In addition to the previous issue, it must be noted that this book is a product of
several years’ research. As its relevance from the academic point of view is not
linked to the political development or success of the accession process, still the
author of this book has been advised by his PhD Committee (before which this book
was defended as a PhD thesis) to include a late-hour chapter (Chap. 12) on the latest
development on this field, the Luxembourg Court’s Opinion 2/13 on the compati-
bility of the DAA with the EU Treaties. The author was also advised to contrast
views—if there are differences—of this book’s findings with those of the Opinion
(where applicable and necessary). However, the author would like to note—as his
PhD Committee has also advised—that this book does not build upon this topic with
Opinion 2/13 serving as the ‘ground zero’. Rather, it takes the opinion as a
comparative tool to superficially flavor the arguments presented in this book. The
core arguments of this book—although the Opinion itself is not pragmatic in many
respects from the practical perspective of the Luxembourg Court—have been
merely supported by the Opinion 2/13 views; the latter being a fact that makes it
even more relevant for this book to cohabitate with the Opinion in place.
Finally, one needs to mention that there has been another PhD Dissertation and
Book written and published in roughly the same topic in 2013 (Gragl 2013).
Although the rough draft of this book has already been written before that book
was published, there was the need of academic weightiness to ensure that Gragl’s
book is considered and contrasted extensively with the findings of this book
(wherever possible). It must be mentioned, however, that Gragl’s book could
have not served again as ‘ground zero’ for this book, as it got published when
this book was almost finalized. However, in order to ensure that this new source of
literature has been extensively considered, this book addresses the work of Gragl
(2013) by: (a) contrasting the main arguments of Gragl with those of this book, and
enlightening on the differences and similarities in almost each of them (either in the
body of the text or in the footnotes) with regard to the three most innovative
12 1 Introduction to the Book

chapters: the chapter on attribution of responsibility, the chapter on the inter-party


complaints and the chapter on the prior involvement mechanism; (b) adding Gragl
as an additional source of knowledge wherever this book has cited someone else,
but Gragl says the same or roughly the same argument/detail/information; and,
(c) by building upon Gragl wherever this book disagrees with him, or, where this
book presents an even slightly different view. All these three actions have been
taken in order to assure the reader that this book consults extensively Gragl’s work
as well. By and large, lastly, this book drives this topic ahead of Gragl on these
perspectives: (a) this book goes beyond Gragl by extensively examining the impli-
cations and potential scenarios that may arise from the use of the co-respondent
mechanism, inter-party complaint mechanism and prior involvement mechanism;
(b) this book offers four extensive chapters on three issues which Gragl does not
examine: first, the macro impact of EU accession to ECHR from the perspective of
normative relationship between the two treaty regimes, second, the detailed test on
three landmark cases against the co-respondent mechanism standards, with ques-
tions of autonomy and functionality of this mechanism essentially tested in prac-
tice, three, the singular perspective of the procedure once an EU-related case comes
before the Strasbourg Court, and the implications therein, and, fourth, an examina-
tion of the latest Luxembourg’s Opinion 2/13 on the compatibility of DAA with the
EU Treaties, and its findings against the conclusions of this book. The theoretical
perspective of this book—building upon a global law dimension—is also an added
value compared to Gragl’s work, which has chosen merely a technical perspective
of examination of the macro impact of the accession project. This said, this book
pulls the original knowledge on this topic beyond Gragl’s work as well.

1.3 Setting the Scene for the Book’s Structure

This book is divided into four parts: each part has some chapters. The four parts—
and the chapters therein—follow a logical and theoretical flow which a book on the
EU accession to ECHR must essentially and unconditionally adhere to—especially
considering the current state of knowledge in the field. The four parts of the book
respond to three fundamental questions: how the EU and ECHR arranged their
communication before accession, with peculiarities of both systems examined and
an introduction to those perspectives presented in the global law literature (Part I);
how will EU accession to the ECHR work—considering mainly the Draft Acces-
sion Agreement and the relevant instruments of law (Part II); the functionality of
the co-respondent mechanism from the perspective of the Strasbourg Court and,
next, through what channels will an application originating in the EU reach the
Strasbourg Court post-accession (Part III); and, a concluding examination on the
Opinion 2/13 and a final conclusion on the entire research project (Part IV) With
these four parts, essentially, one may consider in principle the cycle on the
EU-ECHR accession topic soundly considered—at least from a sequential perspec-
tive. The book therefore will question constitutional and international law options
1.3 Setting the Scene for the Book’s Structure 13

to amalgamate these streams into an integrated system of cooperation and compe-


tition—at least in the context of institutionalization—of the two legal regimes in
Europe.
Let us now turn to the substantive rationale for the selected structure. Why is the
first part of the book structurally as it is? Let us start with the subchapter subsequent
to this. So, with regard to Part I, a subchapter introducing the global law literature
within the context of international competing jurisdictions has been provided at the
beginning—with EU accession to the ECHR reviewed from that perspective—as an
indispensable theoretical food for thought. That being logically justified, it is
important to mention that both the informed and uninformed reader needs to receive
some basic substantive knowledge and analysis of the path that shifted the EU and
ECHR into a synthesized topic of law and theory. That noted, it would not be true
that it is merely the uninformed reader who would need the latter information, and
this is attributed to two reasons: first, the path of the EU becoming a human rights
organization and the Strasbourg’s response to that development is not proverbial
knowledge: instead, it is a course that needs visionary analysis in order to serve as a
stepping stone to the deeper context of EU accession to the ECHR examination;
second, although the autonomy of EU law is a well known fact for most EU
lawyers, there is a distinct need to shape a brief analysis of the form that that
concept takes when EU accession to the ECHR becomes of contextual relevance.
Therefore, offering a springboard to the research project and its readers would have
been immature without covering—at least briefly—the analysis of the EU becom-
ing a human rights law organization, its interrelation to the Strasbourg regime of
law, and the latter’s response, in addition to a brief analysis of EU law autonomy
(both internal and external) which the EU carries on as its cornerstone value in all
treaty-law dealings. Most law scholars of the field do chase the same track13—at
least in terms of structure, if not of substance.
Coming to the more substantial and important part—Part II—the question
remains the same. To begin, Part II of the book is the most significant, relevant
and original segment of this book, because it tackles most essentially the major
research questions of the book. The structure of this part has not been devised or
invented by the author himself; rather, it is a logical derivative of the structure of
substantive issues which the negotiating teams of the EU and Council of Europe
have claimed as being of prime importance in the context of the accession and post-
accession process. Since the very first working meeting of the informal working
group on the Accession of the European Union to the ECHR with the European
Commission, a list of substantive issues—excluding institutional issues—have
been raised as having principal importance and needing special care within the
context of the preparation for the accession instruments; those issues being: the
attribution of ECHR responsibility upon the EU and its Member States (the
co-respondent mechanism), the inter-party mechanism and the submission of the
EU to it, the prior involvement mechanism and its normative construction, and the

13
See e.g.: Besselink (2013), p. 301.
14 1 Introduction to the Book

issue of exhaustion of domestic remedies in the case of the EU14 (each of the latter
topics being individual chapters in this book). It is not being argued here that the
opinion of this working group should be taken for granted—although it has also
been asserted by a large number of scholars in exactly the same format15—but
rather to illustrate that the two organizations (EU and CoU) have shown their
institutional concerns most acutely on these issues for very practical reasons:
each of them aiming to protect its own institutional interests and legal territorial
domains. The latter reasons being sufficient to argue that such principal concerns
need singular scientific thoughtfulness, as these issues will be the crux of the
competition or cooperation between these two organizations. One may continue
and ask why the other aspects that have been discussed and considered of a
concerning nature in the meetings of that working group have not been included
in Part II? The answer is rather short: the book does not engage with the institutional
questions of the accession, but rather with the jurisdictional human rights problems
between the two legal regimes post-accession. Issues regarding institutional prob-
lems and design have been only briefly analyzed in the first chapter of this part,
however they still have a marginal status in this book—adding that institutional
questions do not pose a significant challenge to the case for accession. Finally, it is
not the intention of this book to replicate the discussions of the working group or to
follow the arguments presented therein. This part does merely follow the working
group’s baseline format of issues which need to be considered: however, it analyzes
them from an absolutely academic, original, and more in-depth perspective com-
pared to the very technical and superficial discussions which the working group has
conducted on the issue. It also builds a normative model that will explain concrete
legal techniques—beyond those laid down in the DAA—that would both offer a
more sophisticated reconciling terrain to both legal orders while managing their
jurisdictional and substantive human rights law divergences.
Next, Part III is again considered. Confronted with the question of whether the
analysis on EU accession to ECHR would be complete if the process would not end
up in Strasbourg, we argue that it would not be complete if a book like this would
not tackle implications and possible effects which an EU-originated application will
face before the Strasbourg Court post-accession. Therefore, we chose to include
two chapters on how Strasbourg Court will jurisdictionally portray itself upon
applications of EU origin in cases when the co-respondent mechanism is utilized
by certain parties, and the nature of maneuvers that the Strasbourg Court may
choose to select admissibility tests in some extraordinary cases. Only with these
two chapters the circle is closed, and one may have a complete taste of the issues
regarding functionality and complete effects of the EU accession to the ECHR, both
from a competing/cooperation jurisdictional perspective and a human rights-law

14
See: Council of Europe, doc. CDDH-UE(2010)06rev (1st Working Meeting of the CDDH
Informal Working Group on the Accession of the European Union to the European Convention
on Human Rights (CDDH-UE) with the European Commission).
15
See e.g.: Callewaert (2014), p. 7.
1.3 Setting the Scene for the Book’s Structure 15

safeguard perspective. Part IV, finally, addresses the latest policy and legal devel-
opment in this regard, the Luxembourg Court’s Opinion 2/13, and, on basis of it,
provides a concluding account on the book.
A more interesting and important question to be analyzed here, then, is why this
sequence of chapters has been followed—if one would take for granted that the
selection of chapters is of a reasoned quality. This second question is of an
equivalent weight to the first question. First, in regards to the first part, the book
starts with an introduction and the general research methodology issues which need
to be normally presented in every book of this kind. It then begins with a chapter on
the EU becoming a human rights law organization and its relationship with the
Strasbourg regime of law, followed by a chapter on EU law autonomy. As to the first
chapter, it is argued that there was a definite need to start with an examination of the
EU becoming a human rights law organization in order to deconstruct the theoretical
and practical steps which pressured the EU, upon which the latter projected its path,
to incrementally build its human rights domain of law. That is then followed with an
examination of the EU’s gradual expansion of relationship with the Strasbourg
regime of law, and the latter’s reaction to the growth of EU human rights law.
Without a chapter on this matter, the book would not be able to discuss the issue of
EU accession to the ECHR, as the accession per se has been incentivized, developed
as an idea, and finally, literally managed through a certain enduring process. It
would have been immature to leave such a course outside of this analysis; the latter
serving as a stepping stone to the core topic of this book. E.g. Besselink takes the
same view.16 Chapter 2 then analyzes the autonomy of EU law: why should this
have been sequenced as the second chapter? Of overriding importance to the
relationship between EU law and other international treaty regimes is the nature
of internal prohibitions which EU law has established related to the scope and mode
of submission to such treaty regimes. Applied in the EU accession to ECHR context,
the most important problem from that perspective is the issue of EU law autonomy,
the latter being the viewpoint whereupon the EU carefully looks and engages with
the accession process. The second chapter is therefore EU specific, but also the most
delicate point which has shaped and formatted the accession process quite thor-
oughly. The accession negotiations evidently reveal this. Luxembourg Court’s
Opinion 2/13 does also confirm the same. That being the case, it was logically
decided to proceed with a basic analysis of EU law autonomy in Chap. 2 in order to
open the way for the focused examination of accession, Chap. 2 serving to allow the
debate to develop in its natural and gradual progression.
As relating to Part II, it is worth noting that this part is dedicated to the DAA and
exclusively to the accession implications and outcomes deriving therefrom. It
tackles all accession-related problems and issues from a basis that is reflective of
the DAA.17 This part—of course—undertakes to develop the thesis of functionality

16
Besselink (2013), p. 301.
17
Accord: Callewaert (2014), p. 7, who argues in the same vein that the DAA ‘provides, at this
stage, a sufficiently solid and stable basis to warrant discussion of its contents.’
16 1 Introduction to the Book

and nature of legal effect in the context of cooperation and competition between the
two courts that accession will produce, and goes well beyond the Part I. This part
begins with a chapter debating the accession route, its problems, policy solutions
and supposed outcomes, and finally an examination of each of the DAA articles. It
is argued here that it would have been inadequate and formless to start discussing
the DAA’s individual mechanisms (e.g. co-respondent mechanism, prior involve-
ment, etc.) without making an overall examination of the accession process and its
impact on the systemic picture, and introducing through a profound examination all
articles of the DAA—including those which are not directly relevant for the core
research questions of this book. No reader would have been able to delve into this
analysis without first being provided a systemic analysis of the DAA and its main
outcomes from a constitutional and international law perspective. Only from this
standpoint is one then capable of better judging the individual Accession Agree-
ment mechanisms. The next chapter is one which examines the position of the
Convention and the DAA in EU law post-accession. Why is this situated as the
second chapter of Part II? It has been argued by scholars for a long time that18—
although not relevant from the point of view of international law—the position of
the Convention and the DAA in EU law have a positive preconditioning effect on
the functioning and relationship between Strasbourg and Luxembourg regimes of
law post-accession. It is of paramount importance to consider how the Convention
and Accession Agreement ‘swim’ in the pool of EU law, as a prerequisite to embark
on an examination of the DAA’s outcomes and functionality post-accession. The
latter is supported by the fact that the manner in which the Convention and the DAA
are positioned in EU law, and their relationship with the EU Treaties, directly
condition the mode and scope of application of the DAA’s individual mechanisms.
This is why one may not jump on the issue of specific mechanisms of the DAA
without looking once at the issue of the position of the Convention and Accession
Agreement in EU law, and the conditionality which such position may establish as a
foundation for the functioning of the DAA’s mechanisms. The four remaining
chapters of this part follow the normal path which reflects the structure of issues
and their sequence discussed in the accession negotiations19 and their order in the
DAA. Therefore, the latter structure of Part II is taken for granted, as follows: first, a
chapter on the attribution of responsibility when the EU and its Member States
appear as joint (co-)respondents, second, a chapter on the very important inter-party
mechanism, and, third, a chapter on the prior involvement mechanism. Regarding
the latter sequencing, one may legitimately ask why the issue of exhaustion of legal
remedies has been examined within the chapter on prior involvement mechanism.
This is so because we follow the normal circle that an application originating in the
EU follows: it first needs to be exhausted. If it is not exhausted before EU courts and

18
See e.g.: ECtHR, Scordino v. Italy (No. 1), Application No. 36813/97, 29 March 2006, para.
191; See also a similar logic at: ECtHR, Paksas v. Lithuania, Application No. 34932/04, 6 January
2011; See also: See: Gragl (2013), p. 97.
19
See e.g. on this: Jones (2012), p. 2.
1.4 Introduction to the Individual Chapters of the Book 17

it reaches the Strasbourg Court, then the EU activates the prior involvement
mechanism. The prior involvement mechanism comes successively only after
there has been an impossibility to exhaust domestic legal remedies. The question
could remain, then, why the issue of exhaustion of domestic legal remedies has not
been positioned as the first chapter of Part II? There are two reasons for having
chosen to put this chapter in the current sequence: first, because it would have been
too odd to start the debate on accession and the DAA with a chapter on the
exhaustion of legal remedies, as this issue comes—at least theoretically—at the
side-line of the core research question, and it merely has a subsidiary nature to the
main effects of the accession (although it is a condicio sine qua non chapter);
second, the current sequence follows the structure of issues which the negotiators,
and also the meetings of all participants working on the DAA, have laid down at this
part of the chain. The author has not therefore innovated the concerned chapter’s
sequence.
The third part has two chapters, both of them completing the discussion of issues
that the topic of EU accession to the ECHR should cover. These two chapters—and
Part III, respectively—tackle the scheme and admissibility problems that may
appear once EU applications appear before the Strasbourg Court, starting with the
practical testing of the functionality of the co-respondent mechanism. With this
chapter, the topic on accession becomes clarified more fully from the perspective of
an outsider, who would like to more fully understand the process. Part IV, tackles
Opinion 2/13 as the latest legal and policy development in the EU law, and, then,
draws upon a concluding perspective on the accession project and its legal nature
through Chap. 12. Let us now jump to the subchapter briefly explaining the contents
of each chapter of this book.

1.4 Introduction to the Individual Chapters of the Book

In this part, a more thorough explanation of the structure and contents of specific
chapters of this book will be presented, showing not only the general issues covered
by the latter but also the specific research questions addressed and pointed in each
individual chapter. We start from Chap. 2 onwards. The remaining part of
Chap. 1—which is not explained in this section—will dedicate a theoretical review
on the topic of international competing jurisdictions from a global law perspective
and the EU-ECHR’s slice in that picture.
Part I
Chapter 2, entitled ‘EU becoming a human rights law organization: starting from
nowhere with a “gouvernment des juges”’, offers a grounded outlook for the EU’s
human rights law growth, examining the nature and development of the main
course of its incremental human rights identity within the context of its relationship
with the ECHR and the Strasbourg Court as well. In this chapter, the book addresses
the question: ‘How did the EU become a human rights law organization and how
18 1 Introduction to the Book

has its relationship with the ECHR system developed?’ The chapter first addresses
the landmark cases of then ECJ jurisprudence which initially construed the human
rights skeleton of EC, analyzing specifically Stork, Geitling, Van Gend en Loos,
Stauder, Internationale Handelgesellschaft, Nold KG, and many others. These
cases demonstrate how the Luxembourg Court’s jurisprudence gradually improved
both procedural and substantial guarantees for human rights protection in the EC,
certainly producing a bulk of human rights based on a gouvernment des juges
prerogative. Besides presenting this analysis, the chapter examines the incremental
relaxation of the Luxembourg Court towards the ECHR and Strasbourg Court, but
also international law at-large. Through the lens of the Luxembourg Court, this
chapter attempts to argue the rationality behind the force that pushed the latter to
open gates to the ECHR, certainly offering space for cooperation more than
conflict. The second part of this chapter examines how the Strasbourg Court
received Luxembourg’s approach to the ECHR, offering examples (such as
Mathews and Bosphorus) which demonstrate the struggle facing the Strasbourg
Court, which required it to build a pragmatic relationship with the Luxembourg
Court to keep its function played upon EC flexibly, as a minimum. The chapter
concludes by arguing that the relationship between the two regimes of human rights
law in Europe has been rather tense but well managed and pragmatic, with acces-
sion being a condition for this relationship to become institutionalized.
Chapter 3, entitled ‘EU Law Autonomy: Where does the viewpoint for ‘compe-
tition’ of Luxembourg start from?’ offers a sound analysis of the core legal concepts
that prescribe EU law autonomy, paying special attention to the latter’s external
autonomy. In this chapter, the book addresses the question: ‘Which are the land-
mark legal principles that prescribe EU law’s external autonomy, especially with
regard to an external tribunal such as the Strasbourg Court?’ The chapter starts by
analyzing the foundational steps of EU law autonomy, starting from Costa ENEL,
Opinion 1/91, Mox Plant and ending with the landmark Kadi case. The chapter
carefully introduces and examines the Luxembourg Court’s autonomy from a
tribunal like the Strasbourg Court, both in terms of jurisdiction but also substantive
law. The chapter also portrays the means by which Luxembourg tries to guard its
external borders when certain segments of EU law become interpreted by external
tribunals. A clear jurisdictional examination of the position of international agree-
ments in the EU legal order is provided, trying to demarcate the borders which
Luxembourg Court will draw against the Strasbourg Court even when the EU
accedes to the ECHR (something maintained strongly with the very recent Opinion
2/13 as well). The chapter concludes that the Luxembourg Court carefully considers
its relationship with an external tribunal like the Strasbourg Court, and draws rather
strong lines of ensuring that EU law autonomy is exclusively interpreted within its
boundaries.
Part II
Chapter 4, entitled ‘A new start for the accession of the EU to ECHR’ offers an
examination of the initial steps that prompted accession, the debates that
surrounded the articulation of the need for accession, and the main structural
1.4 Introduction to the Individual Chapters of the Book 19

changes that this process will bring about in relation to the EU’s external image. In
this chapter, the book addresses the question ‘What are the main changes that the
DAA will bring to the EU & ECHR’s external image, and the core issues settled in
the DAA?’ This chapter starts by introducing how the accession debate became
institutionalized within the EU and Council of Europe, and the main arguments that
reflect that debate. It then turns to the question of the main changes that the treaty
system will see with the EU positioned in a state-like contracting position within the
ECHR, and how this will affect the external treaty-making picture of the EU. All
this is analyzed in the context of a multilayered European, human rights law debate,
with the EU fostering its federal attributes with accession to the ECHR. The chapter
then analyzes whether the EU can become a Master of a Treaty made for states only
(ECHR), referring to the Vienna Convention on the Law of Treaties’ principles and
current understandings. Another layer of analysis is provided on the issue of the
ECHR becoming a hybrid treaty system once the EU has acceded, while analyzing
how accession will portray the EU as primus in face of Member States when it
comes to representation before the Strasbourg regime of human rights law, arguing
that this will further change the nature of the EU law primacy principle. The chapter
also discusses the accession model and potential implications, and outlines analyt-
ically each article of the DAA, discussing every potential legal implication that it
may introduce in practice and in theory. The chapter concludes that the DAA will
positively affect the EU’s external image, but will also bring major changes in the
structure and nature of the ECHR system of law.
Chapter 5, entitled ‘Status of ECHR and DAA in EU legal order: the preliminary
question on the effects of accession on EU law’ offers an examination of the legal
status of the ECHR and DAA within the EU legal order post-accession. In this
chapter, the book addresses the question ‘What is the legal status of ECHR and
DAA in EU legal order post-accession’. The chapter starts first by examining the
status of international agreements and customary international law in the EU legal
order, and analyzing the seminal cases of ECJ that have construed this relationship.
The chapter then discusses the reach of the Luxembourg Court when interpreting its
jurisdiction on international agreements, having special reference to the ECHR and
DAA. An in-depth analysis is then provided on the specific status of the ECHR and
DAA post-accession, examining both their rank and implied position in a multi-
layered system of human rights. The chapter then concludes that the ECHR and
DAA will enjoy a rather distinguished position within EU law, something that may
serve as ground for a new understanding of EU Treaties’ setting.
Chapter 6, entitled: ‘Attribution of Liability under the Co-respondent Mecha-
nism’, offers an in-depth analysis of the model of liability sharing between the EU
and Member States that DAA installs within the context of EU Treaties’ prescrip-
tions and limitations in this field. In this chapter, the book addresses the question
‘How may the distribution of burden on ECHR violations be shared between the EU
and Member States under the co-respondent mechanism, and where do the loop-
holes remain?’ The chapter starts by analyzing the EU Treaties’ specific guarantees
made to ensure that the division of liability between the EU and Member States
before the Strasbourg Court respect EU law autonomy. It then compares the nature
20 1 Introduction to the Book

of the co-respondent mechanism with a third party intervention mechanism, draw-


ing the core differences and demonstrating that the former is far more complex and
has a far different intention. The chapter also tackles in detail the functional
mechanics of the co-respondent mechanism, paying special attention to implica-
tions that it may pose once an admissibility test is performed before the Strasbourg
Court. The co-respondent mechanism is then tested against DARIO standards, and
evaluated through the latter’s lens. A deep analysis of the means by which friendly
settlements will be processed within the co-respondent process is provided therein,
hinting at the possible implications and controversies that this process may lead
to. The chapter then carefully and substantively analyzes situations in which the EU
and Member States may jointly appear as co-respondents before the Strasbourg
Court, and the main loopholes that DAA leaves in this process. The chapter ends
with an analysis of the core ambiguities identified by this book on the co-respondent
mechanism, putting special emphasis on situations which may erode EU law
autonomy from the perspective of the jurisdiction of the Strasbourg Court. The
chapter concludes that the co-respondent mechanism is an internationally appreci-
ated innovation of the DAA, however it will lead to many uncertainties that may
provoke the erosion of EU law autonomy and/or impair the functionality of the
ECHR system and the EU’s responsibility towards the latter.
Chapter 7, entitled ‘Inter-Party Mechanism and the EU: Possible Implications
from the Strasbourg’s Jurisdiction?’ offers an in-depth examination of the inter-
party mechanism of the ECHR within the context of the EU as a High Contracting
Party. The chapter addresses the question ‘Whether and, if yes, how, can ECHR and
EU Law reconcile to allow for EU and its Member States to use the inter-party
mechanism of the ECHR?’ The chapter then examines carefully the EU law
limitations on the use of external tribunals for purposes of interpreting segments
of EU law, under the note that the ECHR has already become part of EU law post-
accession. For this purpose, the chapter examines the extent to which the EU
Treaties’ limitations on external tribunals’ jurisdiction should be interpreted, and
examines ways which may reconcile it with ECHR provisions on inter-party
jurisdiction of the Strasbourg Court. Mox Plant and Ireland v. UK are presented
and thoroughly examined against the inter-party mechanism and mechanics pro-
vided for by DAA in order to test the applicability, integrity and functionality of its
use with the EU as one of the parties. The chapter concludes by arguing that the
inter-party mechanism under the DAA and the ECHR may allow space for recon-
ciliation with EU law, with the EU being a party that may file and become attacked
under this mechanism before the Strasbourg Court (certainly EU law autonomy
questions are implicitly raised there as well).
Chapter 8, entitled ‘EU-prior-involvement review’ offers a systemic and origi-
nal examination of the prior involvement procedure under DAA, analyzed against a
complex background of EU law autonomy, EU peculiar institutional features, and
the Strasbourg Court’s natural jurisdictional function. The chapter starts by
discussing the main deficiencies of the system of EU legal remedies and its
reflection on the Strasbourg Court’s standards on this issue, placing doubt on the
integrity of the EU to stand as a legitimate system for the application of the rule on
1.4 Introduction to the Individual Chapters of the Book 21

exhaustion. The chapter, then, addresses the question ‘What are the legal implica-
tions that the use of the prior-involvement mechanism may reflect, and how can one
deconstruct its mechanics to make it decipherable to the individual claimant?’ The
chapter provides an in-depth examination of the prior-involvement provisions of
the DAA and conditions to utilize it, paying special attention to the effect on norms
that may become reviewed and jurisdictional means that may become utilized under
this procedure both at the EU law and Strasbourg levels. The chapter also analyzes
the integrity of this procedure from the perspective of the current EU legal remedies
system, recognizing that this new procedure may introduce certain implicit Treaty
changes in EU law and its institutional structure. The chapter concludes that the
prior-involvement procedure may implicate certain aspects of the EU legal reme-
dies’ system, but appears to be a very good instrument to protect the EU Court’s
exclusive jurisdiction.
Part III
Chapter 9, entitled ‘Testing the co-respondent mechanism: three distinctive cases
with three distinctive scenarios’ offers a genuine test of the co-respondent mecha-
nism in three seminal cases applied in a post-accession scenario. The chapter
addresses the question ‘How functional and legally certain is the co-respondent
mechanism if applied on individual and real cases?’ The chapter carefully examines
means, procedures and potential implications that the application of the
co-respondent mechanism in Mathews, Bosphorus, and Kokkelvisserij cases may
produce, certainly offering an original and provocative analysis on the loopholes
that this mechanism presents. The chapter also questions how the Strasbourg Court
would typically perform the co-respondent procedures in these cases, investigate
the origins of the violation, and apportion the burden. Of course, a specific analysis
of these three cases identifies practical loopholes within the mechanics of the
co-respondent procedures wherein the Strasbourg Court may touch upon EU law
autonomy. The chapter concludes that the test upon these three seminal cases
demonstrates both the lack of full functionality and the lack of achievement of
the intended aims of the co-respondent mechanism drawing also from the perspec-
tive of the legal implications that may arise therefrom.
Chapter 10, entitled ‘Admissibility before the Strasbourg Court: An Outlook on
the EU-law-originated applications’ offers a final examination of the procedure,
tests and potential implications that may be faced once an application against
Luxembourg is filed before the Strasbourg Court in a post-accession scenario.
The chapter addresses the question ‘How should the Strasbourg Court apply its
admissibility jurisdiction in Luxembourg/EU-related cases, with an eye also to the
exceptional admissibility standards? How may these decisions of the Strasbourg
Court look into EU legal order?’ The chapter carefully examines Strasbourg’s
admissibility standards and applies them to EU-related cases, considering the
flexibility clauses which the former may apply to EU-originated cases, and offering
perspectives on such scenarios. The chapter also addresses whether and how EU
primary law violations may be filed at Strasbourg, in addition to a special analysis
of the effect of the latter’s decision within the EU legal order. It also dedicates a
22 1 Introduction to the Book

sub-chapter on the future of Bosphorus post-accession from the perspective of the


Strasbourg Court. The chapter concludes with the remark that the Strasbourg Court
may certainly skip some EU-law-autonomy safeguards and find cases admissible
using numerous flexibility standards, which may unquestionably be seen as inter-
fering to the EU law autonomy. The internal reception of Strasbourg decisions in
EU law follows the same logic.
Part IV
Chapter 11 examines the Luxembourg Court’s latest opinion, the Opinion 2/13 and
its meaning and implications for the accession project. Then, Chap. 12 concludes
with a thorough final summary on the book’s findings and overall arguments, also
offering a hypothetical outlook on the impact of accession to the broader context of
EU and ECHR in the multilayered European human rights law architecture
Let us now jump to the subchapter on literature review.

1.5 A Theoretical Survey on Competing International


Jurisdictions and Treaty Laws: Opening the Literature
Box on the Wider Topic of the Book

1.5.1 Proliferation of International Treaty Regimes


and International Courts: What About a ‘New-
Fangled’ International Law?

International law has undergone huge development during the last century. An
increasing number of treaties have seen light particularly in the last two decades.
The latter fact has paralleled the proliferation of international courts and tribunals,20
but also the explosion of applications being submitted to international courts—such
an increase in applications—as argued by Han—basically being a signal of an
amplified confidence in international law.21 ICJ Judge Stephen Schwebel, to illus-
trate this trend, has noted that the ICJ is ‘busier than ever before’.22 Evidently,
construing specialized treaty regimes and international courts is permissible from a
UN Charter point of view, which in Art. 95 asserts that:

20
To mention only a few: International Criminal Court, the ad hoc International Criminal Tribunal
for Yugoslavia, International Tribunal for the Law of the Sea, the European Court of Human
Rights, the Court of Justice of the European Union, the Inter-American Court of Human Rights,
the African Court of Human and Peoples’ Rights, the Andean Court of Justice, Central American
Court of Justice, the ad hoc International Criminal Tribunal for Rwanda, etc.
21
Han (2006), p. 103.
22
Alford (2000), p. 160.
1.5 A Theoretical Survey on Competing International Jurisdictions and Treaty. . . 23

Nothing in the present Charter shall prevent Members of the United Nations from
entrusting the solution of their differences to other tribunals by virtue of agreements already
in existence or which may be concluded in the future.

One may as a result argue that it is the UN Charter itself which had opened the
system for particularistic international treaty regimes and international courts
which we witness today. Coming back to the central issue, Roger observes that
‘more than fifty international courts and tribunals are now in existence, with more
than thirty of these established in the past twenty years.’23 Both Lavranos and
Brown persuasively replicate this finding.24 This ‘rate of growth has been so furious
that government leaders now express concern over “tribunal fatigue.”’25 Alford, in
this regard, observes that ‘[t]he proliferation of international courts and tribunals
represents a profound change in international law and international relations.’26
Brown, on the other hand, argues that the proliferation of international courts and
tribunals demonstrates the growing interest of the international community to
resolve issues judicially in the international plane.27 Such interest may perhaps be
related to both the need for a safer grounding of international relations in the rule of
law, and the development of an attitude towards legalization of relations between
states and organizations internationally.28 As to the very origin of such an interest,
Helfer & Slaughter basically argue that ‘as a first order of regime design, states
choose independent tribunals over dependent ones when they face multilateral, as
opposed to bilateral, cooperation problems.’29 All this is proclaimed by Sands as the
trend towards ‘judicialization’ of international practice.30 The latter observation is
illustrated by Buergenthal as leading towards ‘the development and application of
international law and its increased relevance in international relations.’31 As one
can witness nowadays a ‘growing number of international courts have been
invested with compulsory jurisdiction [. . .and a. . .] considerable progress has
been made [. . .] to institutionalize [. . .]’ them.32 Such institutionalization will likely
propel ‘the advancement of international law into higher levels of effectiveness.’33
In this regard, Lavranos rightly emphasizes that the increasing number of treaty
regimes and courts ‘inevitably leads to a higher density of international law, which

23
Alford (2000), p. 160.
24
Brown (2002), pp. 4 et seq; Lavranos (2009), p. 4.
25
Alford (2000), p. 160.
26
Alford (2000), p. 165.
27
Brown (2002), p. 6.
28
Cf.: Lavranos (2009), p. 56 (‘The main problem with this proliferation is the fact that it takes
place in an uncoordinated fashion, without clearly formally regulating the jurisdictional relation-
ship between all the various international courts and tribunals.’).
29
Helfer and Slaughter (2005), p. 33.
30
Sands, cited at: Brown (2002), p. 6.
31
Buergenthal (2001), p. 272.
32
Letelier (2005), p. 126.
33
Letelier (2005), p. 126.
24 1 Introduction to the Book

in turn forms the basis for its greater institutionalization or even constitutionaliza-
tion.’34 Whether such a ‘basis’ is safely and literally of a constitutional quality it
remains to be seen from the future practice.
The proliferation of international courts and tribunals is in principle a positive
development, as argued by Brown. However, the problem appears to be the fact that
there is no established ‘overarching framework’ to make the interaction between
these courts and tribunals synchronized and coordinated.35 ICJ Judge Guillaume
rightly argues that the issue of competing and overlapping jurisdictions of interna-
tional courts amplifies ‘the risk of conflicting judgments, as a given issue may be
submitted to two courts at the same time and they may hand down inconsistent
judgments.’ He justifiably points to the deficient nature of international law to
police this trouble.36 In literature, nevertheless, it is noted that the proliferation of
international courts and tribunals essentially parallels the fragmentation of interna-
tional law, the latter product possibly harming the substantive development of
international law.37 A striking example of competing jurisdictions is Southern
Bluefin Tuna which—from a substantive point of view—could have been raised
under three instruments of international law and courts/bodies: first, ICJ, second,
ITLOS acting under UNCLOS, and third, the Arbitral Tribunal under Convention
for the Conservation of Southern Bluefin Tuna.38 Another striking example is La
Grand which, while pending before ICJ, was heard and given an in-depth advisory
opinion by the Inter-American Court of Human Rights.39 These two courts then had
to deal with the same material from different perspectives. A third illustration is the
Swordfish Dispute which fell within both the jurisdiction of ITLOS and that of the
WTO Dispute Settlement Body.40 To make this discussion of an in-depth nature, it
is worth quoting the Prosecutor v Dusko Tadic case, where the ICTY had compel-
lingly argued that:
International law, because it lacks a centralized structure, does not provide for an integrated
judicial system operating an orderly division of labour among a number of tribunals, where

34
Lavranos (2009), p. 5.
35
Brown (2002), p. 7.
36
Statements by the President, 26 October 2000. Address by H.E. Judge Gilbert Guillaume,
President of the International Court of Justice to the United Nations General Assembly, Available
at: http://www.icj-cij.org/court/index.php?pr¼84&pt¼3&p1¼1&p2¼3&p3¼1; Cf.: Berman
(2002), pp. 311–533, p. 318 (Berman says that in the legal sense, jurisdiction refers to ‘the
circumstances under which a juridical body can assert authority to adjudicate or apply its legal
norms to a dispute.’).
37
See e.g.: Koroma (1996), p. 154 et seq; Professor Charney disagrees with this conclusion. See
more at: Lavranos (2009), p. 6; International Law Commission has also studied the issue of
fragmentation of international law, and has come out with a study. See: ‘Report of the Study
Group on Fragmentation of International law: Difficulties arising from the Diversification and
Expansion of International Law’. Session of 18 July 2003, ILC, 55th A/CN.4/L.644.
38
Brown (2002), p. 13.
39
Buergenthal (2001), p. 272.
40
Han (2006), p. 114.
1.5 A Theoretical Survey on Competing International Jurisdictions and Treaty. . . 25

certain aspects or components of jurisdiction as a power could be centralized or vested in


one of them but not the others. In international law, every tribunal is a self-contained system
(unless otherwise provided).41

The abovementioned statement pronounced by the ICTY noticeably portrays the


current uncoordinated situation pertaining to autonomous regimes which operate
within unfettered international law structures. Such self-contained regimes of law
within international law—such as the EU and ECHR, the former being rather fully
self-contained whereas the latter being more liberal in that sense—compete and
cooperate under no certain umbrella of rules and orders. This leads to gaps in their
relationships which may either make their functions contentious and therefore lead
to the fragmentation of international law, or harm their own integrity and effec-
tiveness as to their individual adjudicating jurisdictions. These two developments
may set in motion—what Manley Hudson calls—‘“the danger of a particularistic
development of international law”’.42 This particularistic approach of self-
contained regimes of international law—Buergenthal argues—may seriously
erode the ‘unity of international law, [and] lead to the development of conflicting
or mutually exclusive legal doctrines, and thus eventually threaten the universality
of international law.’43 Buergenthal’s point on the universality of international
law,44 one may legitimately argue, should be the point under which one vigilantly
considers the issue of competing jurisdictions as a global law modern phenomenon
that may seriously harm the nature and nucleus aim of the law of nations. On the
other hand, Lavranos rightly argues that the proliferation of international courts
plays a role in the institutionalization of international law, while simultaneously
increasing its fragmentation. ‘This is particularly the case because of the lack of a
binding legal hierarchy between the norms of international law, and between the
various international courts and tribunals.’45
To provide a very original example of the underlying rationale for this judicial
deference of one court to another—and the mutual respect—one should bring the
landmark case of the US Supreme Court, Hilton v. Guyot, which asserts that:
“[C]omity”, in the legal sense, is neither a matter of absolute obligation, on the one hand,
nor of mere courtesy and good will, upon the other. But it is the recognition which one
nation allows within its territory to the legislative, executive or judicial acts of another
nation.46

41
ICTY Appeals Chamber, Prosecutor v Dusko Tadic. Decision on the Defence Motion for
Interlocutory Appeal on Jurisdiction, Judgment of the Appeals Chamber. Case No. IT-94-1-
AR72 (2 October 1995), para. 11 [emphasis added].
42
Brown (2002), p. 8.
43
Buergenthal (2001), p. 272.
44
Cf.: Shapiro (1993), p. 37 (‘Nor can we even confidently claim that law has become global or
universal in the sense that everyone on the planet can be sure that wherever he or she goes on the
planet, human relationships will be governed by some law, even if not by a law that is everywhere
the same.’).
45
Lavranos (2009), p. 9.
46
US Supreme Court, Hilton v. Guyot, 159 U.S. 113, 163/4 (1895).
26 1 Introduction to the Book

US Supreme Court seems to say that judicial comity is not an issue of a


respectful relationship that courts should build themselves, but rather an issue of
sovereign recognition which remains within the full discretion of a certain treaty
regime to decide upon. Although one may observe an aversion or hesitation in the
approach to judicial comity in this US example, if this rationale would have been
the tool used by international courts, there would have been little tendency towards
unity in international law. Hence, a more liberal stance or a wider view should be
developed to allow international law to grow on its own right and according to its
own aims. In this regard, one should note a statement from the Preamble of the
Vienna Convention on the Law of Treaties (VCLT) of 1969 which explicitly stated:
‘Affirming that disputes concerning treaties, like other international disputes,
should be settled by peaceful means and in conformity with the principles of justice
and international law’.47 Which are those rules of justice and international law that
are supposed to serve in resolving muddled waters between competing treaty
regimes and international courts? In principle—although there is no set of formally
proclaimed rules—there are two basic principles which undertake to regulate the
competing jurisdictions of international courts and tribunals, namely, first, the lis
alibi pendens principle (and forum non conveniens), and, second, res judicata.48
The first rule basically asserts that as long as there is a case pending before another
international court, it will be refused jurisdiction at any other international court.
The second (res judicata) asserts that once an issue has been judged by an
international court, it may not be judged again either in the same court or in another
court of law. Although Vaughan Lowe notes that there is little support in interna-
tional law for forum non conveniens,49 it still plays a core function in regulating the
competing jurisdictions of international courts along with the res judicata principle.
Will these two rules serve sufficiently the competing jurisdictions problem in
international law? Of course not. These two rules merely provide a framework on
the basis of which this problem may be laid down if the conflicting courts of law
regimes voluntarily submit themselves to such rules. If the contrary applies—e.g.
like the EU law which strictly and fully autonomously observes its internal and
external borders—then these rules become almost entirely irrelevant.
To look at this issue in detail, ICJ Judge Guillaume, alternatively, argues that
forum shopping in this regard ‘may indeed stimulate the judicial imagination, but it
can also generate unwanted confusion. Above all, it can distort the operation of
justice, which, in my view, should not be made subject to the law of the market-
place.’50 Essentially, Judge Guillaume argues that there is a market of disorder of
international tribunals, thereby no legal certainty may appear as long as this field of

47
Emphasis added.
48
E.g.: Lavranos (2009), p. 45.
49
Brown (2002), p. 14.
50
Statements by the President, 26 October 2000. Address by H.E. Judge Gilbert Guillaume,
President of the International Court of Justice to the United Nations General Assembly, Available
at: http://www.icj-cij.org/court/index.php?pr¼84&pt¼3&p1¼1&p2¼3&p3¼1, note 54.
1.5 A Theoretical Survey on Competing International Jurisdictions and Treaty. . . 27

troubles has been marketized and not regulated. Even more problematic in this
regard—as Buergenthal argues—‘is the regrettable tendency of some international
tribunals to cite only their own decisions as if other courts did not exist.’51 This is
clearly the core characteristic of aggressively self-contained regimes of law like the
EU Court, which has developed a very strong concept of internal and external
autonomy. Conversely, Han argues that ‘forum shopping in a horizontal system of
international law may not necessarily have a negative impact as long as the relevant
judicial bodies keep their views on law coherent and intact.’52 The issue of a
divergent rather than coherent voice is therefore the nucleus problem.
Han also notes that ‘[t]he fundamental question that underlies the subject of
international law is whether the international legal order is a legal system, a loose
agglomeration or anarchy.’ He struggles to argue that this is—or should be—a
relatively organized system.53 Lavranos—to this point—therefore notes that ‘there
are currently no formal mechanisms in place [. . .] which could ensure that the
on-going institutionalization of international law is accompanied by a hierarchiza-
tion, or at least co-ordination, between the various international courts and tribunals
and their decisions.’54 Letelier—to this end—questions whether it would be advis-
able to regulate more in-depth the jurisdictional competition between international
courts and treaty regimes. The answer is clearly yes, ‘as long as coherence of
international law is strengthened and not undermined.’55 The strengthened coher-
ence of international law would lead ‘to increasing the reputation and legitimacy of
international law, and encourage [. . .] a better level of compliance with its
norms.’56 Most practically, this would have been realized through vesting a certain
international court with the authority of Kompetenz-Kompetenz over the issue of
definition of competences between treaty regimes and international courts, a role
mirroring the national constitutional courts’ functions. Letelier proposes that such a
function in international law could be assigned to the ICJ—in the shape of a
‘mandatory universal jurisdiction to arbitrate over jurisdictional disputes between
competing jurisdictions.’57

1.5.2 Towards a Global Law Rule?

These huge developments in international law relate to the rising phenomenon of


the globalization of law or global law. From a fundamental notional perspective,

51
Buergenthal (2001), p. 274.
52
Han (2006), p. 111.
53
Han (2006), p. 102.
54
Lavranos (2009), p. 9.
55
Letelier (2005), p. 129.
56
Letelier (2005), p. 129.
57
Letelier (2005), p. 132.
28 1 Introduction to the Book

Shapiro asserts that the globalization of law ‘refer[s] to the degree to which the
whole world lives under a single set of legal rules.’58 Jayasuriya, in this regard,
argues that—as a first condition—one needs to understand the changed role of
sovereignty under globalization perspectives. She argues that ‘[u]nderstanding the
fact that the form of sovereignty is transformed by underlying changes in global-
ization of economic relations is a first and important step in inaugurating a new and
innovative research agenda for the relationship between globalization and law.’59
Maurice Strong for this purpose observes that it is not possible any longer in this
globalized world to act unilaterally; therefore she proposes that this new reality
needs to be accredited.60 Shapiro rightly argues that ‘it may be of some help to
appreciate that certain phenomena are now globally common and generate globally
parallel legal responses.’61 Competing jurisdictions phenomenon seems to stand
exactly in that line of argument. To this point, Law plausibly argues that it is a
mistake for globalization to be disregarded in constitutional studies.62 Hence, he
further notes that globalization will definitely influence the growth and develop-
ment of constitutional law.63 This judicial utilization of foreign law and jurisdic-
tions—goes the argument—is forced by the growing international economic,
cultural and political relationships.64 It is an inherent feature of the globalization
of law ‘that territorial borders might no longer be as significant as they once
were.’65 This phenomenon is seen today in almost every caselaw of international
courts. Luxembourg and Strasbourg courts comprise the prime examples of this
fashion. As far as this is concerned, Law crafts the argument that countless internal
constitutional developments today are grounded in global values and instruments of
law,66 with the Luxembourg and Strasbourg regimes of law needing to think in that
direction as well. In principle, such a tendency towards the globalization of law
should not be ruled out as being underdeveloped and destructive. Charles Sampford
rightly argues that globalization should aim to civilize intolerant international
relations,67 the same goal being the aim of global law as well. In this context, the
globalization of law phenomenon should instead be seen as civilizing the relation-
ship between self-contained regimes of international law, and making their func-
tions integrated under certain global objectives and realities. In this context, the
globalization of law is quickly leading towards the phenomenon of cross-

58
Shapiro (1993), p. 37.
59
Jayasuriya (1999), p. 455.
60
Barber (1993), p. 120.
61
Shapiro (1993), p. 64.
62
Law (2008), p. 1281.
63
Shapiro (1993), p. 64 (‘Studies of globalization of law will depend as much on a subtle
appreciation of differences among peoples of the globe as on similarities.’).
64
Law (2008), p. 1279–1280.
65
Berman (2002), pp. 311–533, p. 315.
66
Law (2008), p. 1285 et seq.
67
Sampford (2005), p. 10.
1.5 A Theoretical Survey on Competing International Jurisdictions and Treaty. . . 29

fertilization of treaty regimes, the EU and ECHR being prime examples. Law—in
the same vein—points that this has been ‘remarked specifically upon the growth of
legal “cross-fertilization” in the areas of human rights and constitutional law.’68
It is also important to mention that the globalization of law in the field of
international competing jurisdictions is transforming the model of competition
between treaty regimes and international courts. Vogel has legitimately warned
that globalization may lead to a type of development on the basis of a ‘“race-to-the-
top” or competition in strictness.’69 This may seriously challenge the nation-based
balance of legal authority managed thus far in Europe, and place Strasbourg and
Luxembourg at the forefront of this battle, making national courts irrelevant
players. This phenomenon goes in line with Law’s argument that ‘[g]lobalization
rewrites the rules of transborder interaction in ways that render certain constitu-
tional practices advantageous and others disadvantageous.’70 In this milieu, espe-
cially in the context of the relationship between the EU and ECHR, the
constitutional practice of shifting the level of judicial policy-making from national
courts to the two constitutional courts of Europe—namely Luxembourg and Stras-
bourg—seems to be a reality stemming from the globalization of law. Barber stands
in the same framework of argument by asserting that ‘law has played a powerfully
integrating role in Europe’.71 This assumed, it is of utmost value to note that—as
Lavranos puts it—‘[t]he proliferation of international courts and tribunals, and with
it their increasing power to shape these complex relationships, is just one aspect of
the globalization and legalization of international relations.’72
Placing the Luxembourg and Strasbourg courts at the head of Europe’s global
law development, Helfer & Anne-Marie Slaughter argue:
Yet the nations of Europe have somehow managed to establish not one but two suprana-
tional courts—the ECJ and the ECHR—with active dockets, extensive and well-reasoned
case law, and, most importantly, judgments with which governments have habitually
complied. If the factors that contributed to the success of the ECJ and ECHR could be
isolated and replicated in other parts of the globe, they could significantly enhance the
compliance opportunities for international law in general and for international adjudication
in particular.73

Therefore, one needs to tackle this from a more concrete and focused perspec-
tive, as the Luxembourg and Strasbourg regimes of law remain the universal
examples of development of competing jurisdictions under a mediated dialogue
that further develops the shape of global law.

68
Law (2008), p. 1286.
69
Quoted from: Kelemen and Sibbitt (2004), p. 107.
70
Law (2008), p. 1288.
71
Barber (1993), p. 137.
72
Lavranos (2009), p. 56.
73
Helfer and Slaughter (2005), p. 8.
30 1 Introduction to the Book

1.5.3 EU and ECHR (Under the Title of European


Continent) Slice in the Global Law Landscape?

To consider this topic from another technical standpoint, the issue of competing
jurisdictions of international courts may be seen from two perspectives, to be exact
the horizontal competing jurisdictions and vertical competing jurisdictions. The
former appears when two treaties and two international courts respectively compete
on the same level, whereas the latter refers to the situation when two courts or
treaties—although one subordinated to the other—compete for their own hierar-
chized dominion. Lavranos argues that between the Strasbourg and Luxembourg
courts there appears a ‘horizontal jurisdictional competition’.74 Bosphorus judg-
ment of the Strasbourg Court seems to be the principle to regulate this horizontal
relationship, which puts in motion a doctrine of equivalent protection to keep both
courts synchronized and peaceful. On the other hand, the competition between the
national courts of EU Member States and the Luxembourg Court is of a vertical
jurisdictional fight. The BVerfG’s Solange has been promotionally designed as the
tool to regulate the vertical jurisdictional competition between Luxembourg and
national courts, ‘which allows [. . .BverfG. . .] to accept or reject the exclusive
jurisdictional authority of the ECJ on a flexible basis.’75 The latter, substantively,
is precisely the same with Bosphorus. The Luxembourg Court on its side has been
viewed as rather activist in regulating selfishly its relationship especially with
regard to the international regimes of law and international courts. It has ruled in
Kadi with a somewhat reactive tone that:
It is also to be recalled that an international agreement cannot affect the allocation of
powers fixed by the Treaties or, consequently, the autonomy of the Community legal system,
observance of which is ensured by the Court by virtue of the exclusive jurisdiction
conferred on it by Article 220 EC, jurisdiction that the Court has, moreover, already held
to form part of the very foundations of the Community.76

The Luxembourg Court seems rather harsh in rejecting the reception of any
international treaty or international courts’ jurisdiction if it produces internal effects
to the authority of EU institutions, including its own exclusive jurisdiction. Such
statement seems relevant for both horizontal and vertical jurisdictional competi-
tions. A basic argument in this regard would be that the Luxembourg Court seems a
rather heavily self-contained regime in international law, refusing to swim in the
pool of courteous judicial comity in the international law arena. As Buergenthal
would have argued, it would be important for the Luxembourg Court ‘to recognize
that [. . .it and all other international courts. . .] are all part of the same legal system

74
Lavranos (2009), p. 2.
75
Lavranos (2009), p. 1.
76
Court of Justice of EU, Joined Cases C-402/05 P and C-415/05 P, Yassin Abdullah Kadi and Al
Barakaat International Foundation v Council of the European Union and Commission of the
European Communities, [2008] ECR I-06351, para. 282 [emphasis added]; See also: Court of
Justice of EU, Opinion 1/91 [1991] ECR I-6079, paras 35 & 71.
1.5 A Theoretical Survey on Competing International Jurisdictions and Treaty. . . 31

and that this fact imposes certain obligations.’77 Buergenthal further makes an
appealing note in this regard that:
This means, among other things, that each tribunal has an obligation to respect the general
and special competence of the other judicial and quasi-judicial institutions which comprise
the system, to recognize that it has an obligation, when rendering judgments, to take
account of the case-law of other judicial institutions that have pronounced on the same
subject and, most importantly, to promote and be open to jurisprudential interaction or
cross-fertilization.78

The Luxembourg Court seems careless about adopting a respectful model of


cohabitation in international law, and clearly reacts negatively towards the idea of
cross-fertilization with international treaty regimes if they appear to produce
internal effects on the autonomy of EU law. Making the latter a precondition on
engaging with international tribunals seems in actual fact refusing to accept that
common rules should be respected in view of promoting harmony within the global
law arena. This phenomenon—similarly borrowed from Kinley—may be described
as ‘“globalized localism”, which occurs when essentially local phenomena are
exported or propagated globally.’79 In Kinley’s model of theoretical remark, the
EU should be likewise considered as trying to force the world of nations implant its
own (in-house) resistant model of relating with international courts and treaty laws.
Coming to the more regional question of interaction between the EU and ECHR,
Lavranos notes that:
Indeed, a wellknown example is the occasional divergent jurisprudence that exists between
the ECJ and the European Court of Human Rights (ECrtHR) regarding the interpretation
and application of fundamental rights as protected by the European Convention on Human
Rights (ECHR). The lack of a clear explicit hierarchical determination of which court
should have the last word regarding European fundamental rights has lead to several
divergent judgments by both courts on similar issues.80

Lavranos’ position on the often divergent substantive approaches—but also


jurisdictional competition between the Luxembourg and Strasbourg courts—exem-
plifies the European slice in global law from the perspective of competing juris-
dictions. Callewaert—in relation to the competing language between the two
European courts—argues that ‘whether they are real or only apparent, such centrif-
ugal tendencies need to be curbed.’81 The horizontal relationship between the
Luxembourg and Strasbourg courts needs certain management to become not
only reconciled but also legally certain under the rule of global law. In this regard,
in the words of Alford, ‘[t]he effectiveness of a particular international court
depends in large measure on the theoretical rationale for why nations obey its
mandates.’82 There is no doubt that the Luxembourg and Strasbourg courts have

77
Buergenthal (2001), p. 274.
78
Buergenthal (2001), p. 274.
79
Kinley (2005), p. 97.
80
Lavranos (2009), p. 8.
81
Callewaert (2014), p. 11.
82
Alford (2000), p. 164.
32 1 Introduction to the Book

had rather different rationales for their business: the former being a constitutional
court for the EU, and the latter being a human rights specialized court of European
international law. Although this stands rather clearly delineated, the multilevel
approach to the Convention and the law of the EU often triggers competing
approaches in the relationship between the two regimes. Such competition will
become institutionalized or constitutionalized with the DAA of the EU to the
ECHR; such an agreement making the competition between the two courts and
regimes vertical in nature83—as opposed to the horizontal scheme that has existed
until now. This will become a novel development inside and outside Europe from a
global law perspective. Han—for instance—argues that with the increase of inter-
national jurisdictions it ‘is necessary to cope with the increasing trend of decen-
tralization and the growing complexity of an international society. As international
society grows more complex, international obligations are becoming more burden-
some.’84 The DAA seems to stand in the center of this growing decentralization
from a global law perspective, wherein the European continent builds a consistent
human rights landscape with two courts functioning in a coordinated manner,
thereby better channeling the rising of multilayered obligations of human rights
for both the EU and its Member States.
Therefore, there is a rather well established theoretical observation that the
European agenda of these European courts stands at the top of this global law
development. To prove this proposition, Posner and Yoo have empirically found
that Luxembourg and Strasbourg courts rank at the top, from a global measurement
map, when it comes to the dimensions of effectiveness and independence.85 It is not
without a grounded raison d’être that EU accession to the ECHR will be of a central
impact in the context of global law studies and development. In this regard, Helfer
& Slaughter eloquently suggest that:
The conditions under which international tribunals are more or less effective, and the range
of choices open to national decision makers in creating or reforming such tribunals, lie at
the heart of global governance.86

It is exactly this shape of argument which needs be considered from the


perspective of Accession Agreement’s mechanisms and their functionality that
makes the context of global law centrally relevant regarding the accession process
and the practical interface between Luxembourg and Strasbourg post-accession.

83
Contra: Interestingly enough, Gragl views the relationship between the Luxembourg Court and
Strasbourg Court of a horizontal level. See: See: Gragl (2013), p. 9. Contrary to Gragl, it is argued
here that such relationship will be of a vertical relationship after EU accedes to ECHR, as EU will
become a high contracting party in the same par with state contracting parties to the Convention
system. Such status will definitely make the Luxembourg Court a court of domestic EU law and
the Strasbourg Court a court of international law in face of the former. It is therefore argued that
the new relationship between the Luxembourg and Strasbourg will be of a vertical relationship,
and there seems nothing to argue the contrary.
84
Han (2006), p. 108.
85
Helfer and Slaughter (2005), p. 5.
86
Helfer and Slaughter (2005), p. 57.
References 33

Han, correctly—as a general rule—argues that settling the competing positions of


international treaty regimes and international courts is ‘an immensely difficult task
because it requires resolution of many tensions—between a general regime and a
special regime, universalism and particularism, and general international law rule
and special international law rule.’87 The DAA is the tool that will likely establish
the roles between the EU and ECHR on the one hand, and Luxembourg and
Strasbourg courts, on the other hand, with regard to their seating in a general rather
than special regime, universalism rather than particularism, and general rules rather
than special rules of international law. That model of institutionalization of the
European courts’ relationships—if not building a permanent peace between them—
will certainly make the following observation worthless post-EU accession to the
ECHR: ‘What Gunter Grass said of Germany—“unified, the Germans were more
disunited than ever”—applies in spades to Europe: integrated, it is more disintegral
than ever.’88 In this regard, it is worth mentioning as a final point, Callewaert rightly
argues that the Lisbon’s duty to accede the EU to the ECHR aims to prevent a
situation where ‘there are [. . .] two “worlds” of fundamental rights based on two
different types of fundamental rights in Europe, with the same rights possibly
having a different substance depending on whether EU law applies or not.’89 Let
us now jump on the next chapter relating to the EU’s path to become a human rights
law organization, and its relationship with the Strasbourg regime of law.

References

Alford RP (2000) The proliferation of international courts and tribunals: international adjudication
in ascendance. Scholarly Works. Paper 9. Available at: http://scholarship.law.nd.edu/law_
faculty_scholarship/9, pp 160–165
Balfour ADJ (2005) Application of the European Convention on Human Rights by the European
Court of Justice. Harvard Law School Student Scholarship Series. Paper 4, pp 1–57
Barber BR (1993) Global democracy or global law: which comes first? Indiana J Global Leg Stud
1(1):119–137
Barbera M (2012) Diritti sociali e crisi del costituzionalismo europeo. WP CSDLE “Massimo
D’Antona”. INT – 95/2012
Berman PS (2002) The globalization of jurisdiction. University of Connecticut School of Law
Articles and Working Papers. Paper 13. Available at: http://lsr.nellco.org/uconn_wps/13
Besselink LFM (2013) Should the European Union ratify the European Convention on Human
Rights? Some remarks on the relations between the European Court of Human Rights and the
European Court of Justice. In: Føllesdal A et al (eds) Constituting Europe: the European Court
of Human Rights in a national, European and global context. Cambridge University Press,
Cambridge

87
Han (2006), p. 121.
88
Barber (1993), p. 124.
89
Callewaert (2014), p. 10.
34 1 Introduction to the Book

Brown C (2002) The proliferation of international courts and tribunals: finding your way through
the maze. Review essay. Melb J Int Law 3(2). Available at: http://www.law.unimelb.edu.au/
files/dmfile/downloadb15e1.pdf
Buergenthal T (2001) Proliferation of international courts and tribunals: is it good or bad? Leiden J
Int Law 14(2):267–275
Callewaert J (2014) The accession of the European Union to the European Convention on Human
Rights. Council of Europe, Strasbourg
Cormick NM (1995) The Maastricht Urteil: Sovereignty Now. Eur Law J 1(3):259–266
Defeis EF (2000–2001) Human rights and the European Union: who decides? Possible conflicts
between the European Court of Justice and the European Court of Human Rights. Dickinson J
Int Law 19(2):301–331
Gragl P (2013) The accession of the European Union to the European Convention on Human
Rights. Hart Publishing, Oxford
Han SWD (2006) Decentralized proliferation of international judicial bodies. J Transnat Law Pol
16(1):101–122
Helfer LR, Slaughter A-M (2005) Why states create international tribunals: a response to Pro-
fessors Posner and Yoo. Calif Law Rev 93:1–58
Jayasuriya K (1999) Globalization, law, and the transformation of sovereignty: the emergence of
global regulatory governance. Indiana J Global Leg Stud 6(2):425–455
Jones C (2012) StateWatch Analysis: ‘The EU’s accession to the European Convention on Human
Rights: a cause for celebration or concern?’ Available at: http://www.statewatch.org/analyses/
no-187-echr.pdf
Joris T, Vandenberghe J (2008–2009) The Council of Europe and the European Union: natural
partners or uneasy bedfellows? Columbia J Eur Law 15:1–43
Kelemen RD, Sibbitt EC (2004) The globalization of American law. Int Organ 58(1):103–136
Kinley D (2005) The universalizing of human rights and economic globalization. What roles for
the rule of law? In: Zifcak S (ed) Globalization and the rule of law. Routledge, London
Koroma JA (1996) The interpretation of international law. In: Contemporary International Law
Issues: Proceedings of the Third ASIL/NVIR Joint Conference
Lavranos N (2009) On the need to regulate competing jurisdictions between international courts
and tribunals. EUI Working Paper MWP 2009/14
Law DS (2008) Globalization and the future of constitutional rights. Northwest Univ Law Rev 102
(3):1278–1349
Letelier R (2005) The competing jurisdictions of international courts and tribunals. Revista De
Estudios Internacionales 38(150):125–133
Odermatt J (2014) The EU’s accession to the European Convention on Human Rights: an
international law perspective. Leuven Center for Global Governance Studies. Working Paper
No. 136
Olsen JB (2009–2010) Protecting fundamental rights and the evolving roles of the court of justice
of the European Union and the European Court of human rights; Europeanisation in action at
the supranational level. Edinburgh Student Law Rev 1(54):54–73
Sampford C (2005) Reconceiving the rule of law for a globalizing world. In: Zifca S
(ed) Globalization and the rule of law. Routledge, London
Shapiro M (1993) The globalization of law. Indiana J Global Leg Stud. Symposium: The
Globalization of Law, Politics, and Markets: Implications for Domestic Law Reform 1
(1):37–64
Thym D (2013) ‘A Trojan Horse? Challenges to the Primacy of EU Law in the Draft Agreement on
Accession to the ECHR’, VerfBlog, 2013/9/11, http://www.verfassungsblog.de/en/atrojan-
horse-challenges-to-the-primacy-of-eu-law-in-the-draft-agreement-on-accession-to-the-echr/
Wetzel JR (2003) Improving fundamental rights protection in the European Union: resolving the
conflict and confusion between the Luxembourg and Strasbourg Courts. Fordham Law Rev
71:2823–2862
White S (2010) The EU’s accession to the convention on human rights: a new era of closer
cooperation between the Council of Europe and the EU? New J Eur Crim Law 1(4):433–446
Chapter 2
EU Becoming a Human Rights Law
Organization: Starting from Nowhere
with a ‘gouvernment des juges’

2.1 Introduction

This chapter examines human rights law development of the Union with reference
to the Convention and Strasbourg regime of law at-large. The chapter evolves
around the research question ‘How did the EU become a human rights law organi-
zation and the gradual development of its relationship with the ECHR system?’
Two perspectives of analysis are provided below: first, an EU-based perspective of
human rights law development, focusing on the Convention and Strasbourg regime
of law, and, second, a Strasbourg-based view of EU human rights development, and
gradual recognition of the EU as a human rights liable entity and the means via
which such relationship became construed. Overall, the chapter offers a much
grounded outlook of the relationship between the Luxembourg and Strasbourg
regimes of law, one that is needed to start shedding light on the preconditions
that EU accession to the ECHR will likely materialize on these two legal orders.
In the first part, the chapter starts with the EU’s initial position on human rights
law, wherein the insistence of the Luxembourg Court was that the European
Community did not have human rights competence therefore dismissing as inad-
missible all human rights-related applications. Two seminal cases are deconstructed
to give this flavor, namely: Stork and Geitling. The chapter then departs on the more
dynamic approach that the Luxembourg Court started to demonstrate with regard to
human rights law, thus gradually building the human rights law face of the
EC. Luxembourg’s seminal cases of Stork, Geitling, Van Gend en Loos, Stauder
are carefully analyzed and contrasted with the gouvernment des juges prerogative
that the Court incrementally attached to itself. Afterwards, the chapter tackles the
seminal cases of Internationale Handelgesellschaft, Nold KG, Rutili, Hauer, ERT
and SPUCI, whose examination not only demonstrates the gradual development of
EU human rights law identity but also a clear and significant deference to the
Convention and Strasbourg regime of law and international law at-large. The
examination of these cases makes the point that a rather vibrant judicial activism

© Springer International Publishing Switzerland 2015 35


F. Korenica, The EU Accession to the ECHR, DOI 10.1007/978-3-319-21759-8_2
36 2 EU Becoming a Human Rights Law Organization: Starting from Nowhere with a. . .

by the Luxembourg Court mixed with pressure both from Member States and
claimants has pushed the EU to the borders of the Convention system. Not only
may the latter have been interpreted as a friendly order, but can also be seen as an
expression of aggression from a competitor.
In the second part, the chapter embarks on the Strasbourg Court’s view on the
EU human rights law and performance. The chapter carefully examines the initial
and then more advanced stances of the Strasbourg Court on EU human rights and
the Luxembourg Court in general, with a focus on the significant steps in the
development of a flexible relationship with Luxembourg which Strasbourg opted
for. Strasbourg’s seminal cases of Mathews and Bosphorus are carefully examined
to demonstrate how the Strasbourg Court construed its jurisdiction with regard to
EU Treaties and how it refrained in principle from engaging with the EU secondary
law based on the doctrine of equivalent protection, in the absence of a manifest
deficiency. This part then argues how the tension between the two orders has been
managed especially by the Strasbourg Court, specifically the Court’s attempt to
keep the language of human rights theoretical at times (although not withdrawing
from that jurisdiction) in order to escape from potential conflicts with EU law.
Finally, the third part deconstructs the EU Charter of Fundamental Rights and
ECHR from the perspective of their cohabitation within the context of EU
internal law.
The chapter concludes by arguing that the relationship between the two regimes
of human rights law in Europe has been rather tense, but well managed and
pragmatically driven by the two courts, with accession being a condition for this
relationship to become institutionalized.

2.2 The Origins of EU Becoming a Human Rights Law


Organization with Reference to the ECHR

In its original composition, the EEC establishing treaties had no human rights
substance, and accordingly human rights were neither enshrined nor proclaimed
by the establishing treaties of the then-EEC.1 Made in very specific historical and
political circumstances, there were numerous reasons for the ‘fathers’ of the
establishing treaties to refrain from including a list of human rights. The
establishing treaties, in that form, provided no human rights limitations to the
authority of the polity being established,2 thereby leaving many concerns as to
the legitimacy of such a treaty system. Designed in that mode, the EEC treaty

1
See e.g.: Schimmelfennig (2006), p. 1247; See also: Landau (2008), p. 557; Franklin (2010–
2011), p. 139; Škugor (2012), p. 106.
2
Cf.: Olsen (2009–2010), p. 59, who argues that EU grew up from an economic community to a
constitutional system nowadays; EU accession to ECHR, one would argue, being one of these core
layers of its constitutional consolidation not only in internal matters but also in external action.
2.2 The Origins of EU Becoming a Human Rights Law Organization with Reference. . . 37

system provoked numerous claims—both academic and practical—as to the polit-


ical accountability and constitutional legitimacy. The raison d’etre of the initial
construction of the EEC was to provide for economic integration of then member
states,3 therefore, went the argument, there was no need to surpass into the territory
of political issues, such as human rights law.4
In its original shape, EEC proved of little relevance to the individual human
rights regime, hence illustrating its strict economic nature. Although the EEC had
an almost pure economic nature,5 this did not provide a reason for individuals to
refrain from seeking the application of human rights standards to the EEC legal
measures. To note, ‘[. . .] in both Community and English law, both systems are still
to a large extent – as is the European Convention – a case-law system, in which the
decisions of the courts play a leading role.’6 A suitable argument therefore is that
‘the lack of comprehensive provisions for the protection of human rights has not
meant the absence of any protection in the EEC legal order.’7 In this regard, several
cases were filed before the Luxembourg Court,8 requesting that certain legal acts of
the EEC be checked against the fundamental rights standards.9 The Stork10 and
Geitling11 provide the two, most representative examples wherein Luxembourg
refrains from entering into the human rights ‘box’.12 In Stork, the Court fully
opposed the applicant’s argument that the ‘misuse of powers or an infringement
of the Treaty sufficient to justify the annulment of the contested decision is to be

3
See: Gragl (2013), p. 3/4, who notes that the initial European integration project focused on
‘economic integration and welfare of the Member States.’ The Luxembourg Court, on the other
hand, exercises a broad jurisdiction on any field of law as opposed to the Strasbourg Court which is
fundamental-right focused. This is a reason why Gragl, referring to Quinn, calls the EU and
Council of Europe as ‘twins separated at birth’ (Ibid., p. 4).
4
Cf. with a new understanding on this: Court of Justice of EU, Case C-319/06 Commission v
Luxembourg [2008] ECR 1-4323, para. 43, which states that: ‘the Court has already held that the
freedom to provide services, as one of the fundamental principles of the Treaty, may be restricted
only by rules justified by overriding requirements relating to the public interest. . .’; See also:
Shelton (2003), pp. 110 et seq.
5
On the economic nature of EC, see: Defeis (2007–2008), p. 1; See also: Defeis (2000–
2001), p. 302.
6
Jacobs (2007), p. 13.
7
Groussot et al. (2011), p. 1.
8
I refer here to the Court of Justice of the European Communities, whereas throughout the book
the terms Luxembourg Court, EU Court, ECJ, and Court of Justice of EU are used interchangeably.
9
On the reasons for requesting the introduction of human rights in the EC law, see: Coppel and
O’neill (1992).
10
Court of Justice of EU, Case 1/58, Friedrich Stork & Co., Kohlengroßhandlung vs. High
Authority Of The European Coal And Steel Community, Court of Justice of the European
Communities, Judgment.
11
Court of Justice of EU, Case 16/59, Geitling Ruhrkolhlen-Verkaufsgesellschaft Mbh vs. High
Authority of the European Coal and Steel Community, Judgment, Court of Justice of the European
Communities.
12
See e.g.: Schimmelfennig (2006), p. 1252.
38 2 EU Becoming a Human Rights Law Organization: Starting from Nowhere with a. . .

found in that the High Authority wrongly failed to take account of the fact that the
decisions in question had to be assessed from the point of view of German law, by
virtue of which they were void.’13 It was obvious that the applicant requested that
the Court ‘transpose’ a domestic German constitutional right into a community
standard for judging the legality of an act under the EEC establishing treaties. The
Court’s argument, more explicitly, was merely to confirm that under the ‘[t]reaty
the High Authority is only required to apply Community law. It is not competent to
apply the national law of the Member States.’14 Advocate-General Lagrange further
argued that the High Authority had no competence to check the EEC acts’ compli-
ance with the national legislation of a member state, therefore, he argued, illegality
in that sense would not be important at all.15 In view of Stork, the Luxembourg
Court simply reiterated that it sees the community legal order as autonomous from
those of the member states, thereby refusing to borrow any human rights materie
from there. In both Stork and Geitling (but also Sgarlata16), Luxembourg refrained
from annulling EEC legal acts being based on member states’ set of fundamental
rights, which noticeably exhibited its ‘self-limiting’ character in the face of human
rights law.17 The lack of a human rights legal structure—although citizens may
have benefited from the common market aims of the Community—led to Williams’
argument that the EC had insufficient legitimacy to exercise constitutional authority
over its citizens.18 The author of this book expands this argument by making the
point that without a human rights law foundation, the Community even lacked a
basic democratic standard which it had persistently proclaimed.
The inactive behavior of the Luxembourg Court, however, started to change19
some years after Stork. Not only the practical circumstances, but also the self-
pressure to sustain the community construction pushed the Court to change its view
on the human rights context of the community. Member States’ courts exerted
intense pressure, resulting in part from the necessity of the courts to choose between
guaranteeing their own constitutional human rights to their citizens or threatening
the supremacy of EC law.20 A turning point was consequently made in Van Gend en
Loos, where the Court ruled that:

13
Court of Justice of EU, Case 1/58, Friedrich Stork & Co., Kohlengroßhandlung vs. High
Authority Of The European Coal And Steel Community, Court of Justice of the European
Communities, Judgment, note 10, p. 26.
14
Court of Justice of EU, Case 1/58, Friedrich Stork & Co., Kohlengroßhandlung vs. High
Authority Of The European Coal And Steel Community, Court of Justice of the European
Communities, Judgment, note 10, p. 26.
15
Opinion of Advocate-General Lagrange, in Case 1/58 Stork & Co., p. 38.
16
Court of Justice of EU, Sgarlata v Commission, Case 40/64 [1965] ECR 215.
17
Weiler (1986), p. 1114.
18
Williams (2004), p. 139.
19
And this was mainly a result of pressures from national constitutional courts. See e.g.: Eckes
(2013), p. 257.
20
Scheuner (1975), p. 173/4.
2.2 The Origins of EU Becoming a Human Rights Law Organization with Reference. . . 39

Independently of the legislation of Member States, Community law not only imposes
obligations on individuals but is also intended to confer upon them rights which become
part of their legal heritage. These rights arise not only where they are expressly granted by
the Treaty but also by reason of obligations which the Treaty imposes in a clearly defined
way upon individuals as well as upon the Member States and upon the institutions of the
Community.21

The Van Gend en Loos landmark ruling not only proclaimed the legal person-
ality under international law of the EEC—which in Timmermans’ view meant
building the basis for the ‘constitutionalization’ of the Community legal
order22—but also marked a fundamental step in terms of human rights law for the
Community. This was an important step forward compared to Stork, clearly
confirming the Community’s human rights construction.23 More specifically, Van
Gend en Loos seemed to proclaim that the foundation of the human rights law
derives from the establishing treaties themselves, separate from that of the member
states’ laws, thereby installing a sort of Community-originated law of rights while
further advocating the autonomous nature of Community law.24 Nevertheless, Van
Gend en Loos still struggled to repair the lack of human rights laws in the treaties
themselves, hence it ruled that rights in the Community arise by reason of obliga-
tions which the Treaty places upon persons, member states and institutions. In
general, Van Gend en Loos proved to be the first step of the Community to bind it
with human rights law, therefore constituting the earliest basis of human rights law
in the history of the EEC. Following the striking decision in Van Gend en Loos, the
Luxembourg Court further innovated the context of human rights law in Stauder.25
Examining a request for a preliminary ruling, the Court decisively affirmed that
fundamental rights are enshrined by Community law supplemented by the ‘unwrit-
ten Community law, derived from the general principles of law in force in Member

21
Court of Justice of EU, Van Gend en Loos v. the Netherlands, Case 26/62, Judgment, CJEC, p. 2;
See also a very wealthy analysis on the main conceptual outcome of this, at: Weiler (1981),
pp. 275 et seq.
22
Timmermans (2001), pp. 2 et seq.
23
It went even further in: Court of Justice of EU, Parti écologiste ‘Les Verts’ v European
Parliament, Case 294/83 [1986] ECR 1339, para. 23, by acknowledging the basic municipal
constitutional nature of the treaty, which plays then an essential role in the interpretation of human
rights law as well; See also: Court of Justice of EU, Weber v European Parliament, Case C-314/91
[1993] ECR I-1093, para. 8.
24
As to this issue, AG Karl Roemer argued: ‘The Community has its own institutions, independent
of the Member States, endowed with the power to take administrative measures and to make rules
of law which directly make rights in favour of and impose duties on Member States as well as their
authorities and citizens’ (Opinion of Mr Advocate-General Karl Roemer in Case 26/62, delivered
on 12 December 1962, p. 20.).
25
See e.g.: Miiller (2007), p. 34; See also: Defeis (2000–2001), p. 310; See also the predecessor of
Stauder, Van Eick case. See: Court of Justice of EU, Van Eick v Commission, Case 35/67 [1968]
ECR 329.
40 2 EU Becoming a Human Rights Law Organization: Starting from Nowhere with a. . .

States.’26 Stauder, remarkably, innovates the general ‘principles of law’ deriving


from the Member States’ legal orders as a basis for the Court to interpret and
declare human rights in the EEC. Certainly, Stauder bound the Community with the
member states’ general principles of law, thereby converting the latter’s fundamen-
tal rights into a set of rights to be observed/assured in the Community.27 In this
regard, the Court also pointed the fact that it has a monopoly on assuring that the
rights in the Community are respected, whereas the construction of rights under the
general principles of law in the Member States was ‘implicitly’ proclaimed to be
under the authority of the Luxembourg Court as well. With Stauder (but also Van
Eick28 previously), most appreciably, the Luxembourg Court authorized itself to
observe and assure human rights at the Community level inter alia on basis of
general principles of law deriving from the Member States. This proved to be the
first ‘big’ step towards a more secured Community rights system based on the
constitutional traditions of Member States.
One year later, the Court took a more sophisticated step with respect to human
rights formation in the Community. In Internationale Handelgesellschaft, it ruled
that finding ways based on the national law to assess the legality of a Community
legal act would harm the uniformity and efficacy of Community law.29 The Court
added that the validity of Community measures could only be judged within the
context of Community law, and the latter could not be overridden or annulled by
means of national law,30 even where constitutional rights have been violated. Most
strikingly, the Court argued: ‘In fact, respect for fundamental rights forms an
integral part of the general principles of law protected by the Court of Justice.
The protection of such rights, whilst inspired by the constitutional traditions
common to the Member States, must be ensured within the framework of the
structure and objectives of the Community.’31 In other words, human rights are
an obligation for all Community institutions, without exception. However, the
Community could protect those rights as long as such rights are concerned with a
Community measure, meaning that the Community could not undertake to protect
rights which are naturally bound to the Member States’ competences.32 The
Internationale Handelgesellschaft, on the one hand, proved the general

26
Court of Justice of EU, Stauder v. City of Ulm, Case 29/69 [1969] ECR 419, request for
preliminary ruling, p. 422; See: Schimmelfennig (2006), p. 1248; Shelton (2003), p. 111.
27
See also: Morano-Foadi and Andreadakis (2011), p. 597.
28
Court of Justice of EU, Van Eick v Commission, Case 35/67 [1968] ECR 329.
29
Cf.: Besselink and Reestman (2008), p. 199.
30
See e.g.: Wetzel (2003), p. 2836.
31
Court of Justice of EU, Internationale Handelgesellschaft mbh v. Einfuhr und Vorratsstelle f€
ur
Getreide und Futtermittel, Case 11/70, 1970, ECR 1125, p. 1134; See also: Lawson (2005), p. 28,
calling this the self-restrained mood on human rights; See also: Landau (2008), p. 557/8; Sera
(1996), pp. 160 et seq; Škugor (2012), p. 111.
32
See, for instance: Opinion of Advocate General Capotorti, Case 149/77, Delivered On 30 May
1978, p. 1385.
2.2 The Origins of EU Becoming a Human Rights Law Organization with Reference. . . 41

presumption that the Luxembourg Court does not wish to see the application of a
Community measure rejected on the grounds that national law standards are
invoked. On the other hand, the Internationale Handelgesellschaft ruled explicitly
that the protection of human rights is a task of the Community, and should be
ensured within its institutional structure, being inspired by the member states’ legal
orders.33 With regard to the Member States’ human rights orders, AG De Lamothe
had compellingly argued that those ‘contribute to forming that philosophical,
political and legal substratum common to the Member States from which through
the case-law an unwritten Community law emerges, one of the essential aims of
which is precisely to ensure the respect for the fundamental rights of the individ-
ual.’34 It is precisely AG De Lamothe’s argument which illustrates that
Internationale Handelgesellschaft built the concept that the Community proclaims
nothing less than member states in terms of human rights law and observance,
therefore the arguments to ‘challenge’ the applicability of its measures on the basis
of national human rights would have no rationality. Seen from a more historical
perspective, it seems clear that Internationale Handelgesellschaft proves
Luxembourg’s concern that the Community measures could be challenged by
means of domestic law.35 It is therefore argued that Luxembourg’s decision on
this case is at worse a response to save the Community law from being challenged,
and at best, a response to support the further assurance of human rights at the
Community level.36 After this—having considered the EU Court’s tendency to use
a language that is rather formalistic and of a proclamatory nature as opposed to
substantive human rights observance (which e.g. Besselink says has been criticized
for not ‘taking rights seriously’)37—two Member States’ constitutional courts—
namely, the German one in Solange I38 and the Italian one in Frontini v Ministero
delle Finanze39—threatened the supremacy of EC laws. These national courts had
argued that if the latter fails to respect national constitutional human rights, then it

33
See generally: Morano-Foadi and Andreadakis (2011), p. 597.
34
Opinion of Mr Advocate-General Dutheillet De Lamothe in Case 11/70. Delivered On
2 December 1970, p. 1146.
35
Some member states’ constitutional courts had challenged the Community law application with
the question of its national human rights’ compliance. As an example for Germany, see:
Bundesverfassungsgericht [BVerfG] [Federal Constitutional Court] Oct. 18, 1967,
22 Entscheidungen des Bundesverfassungsgerchts [BVerfGE] 223 (1967) (F.R.G).
36
Although the Court referred to the human rights law in Internationale Handelgesellschaft, it
nevertheless found no human right violation in that individual case. This seriously upset the
German Federal Constitutional Court, which revenged to Luxembourg by issuing Solange I; On
Solange I, see also the interaction between the German court and ECJ which forced the latter to
better tackle the issue of rights in order to save the primacy of the EU law, at: Franklin (2010–
2011), p. 145.
37
Besselink (2013), p. 304.
38
BVerfGE 37, 271 2 BvL 52/71 Solange I-Beschluß (Internationale Handelsgesellschaft v
Einfuhr- und Vorratstelle f€ ur Getreide und Futtermittel); See also: Škugor (2012), p. 112;
Besselink (2013), p. 302/3.
39
Italian Constitutional Court, Frontini v Ministero delle Finanze, Case 183/73.
42 2 EU Becoming a Human Rights Law Organization: Starting from Nowhere with a. . .

will be invalid in their constitutional orders. Alarmed with German Solange I and
Italian Frontini which had seriously undermined the principle of EC law suprem-
acy, the EU Court started to dramatically change its approach to human rights law
from mere proclamations to effective and substantive human rights protection.
It was time to take rights more seriously. Although the Luxembourg Court had
embarked into the human rights sphere—at least formally—it had not yet made any
reference to any international treaty on human rights, neither had it ruled on the
responsibility of Community law towards international human rights law. To ensure
that it is assigned with the task of observing the enjoyment of rights at the
Community level, the Court consolidated rather well its human rights approach
with Nold KG,40 where it reaffirmed that fundamental rights form part of the
general principles of law and ‘it cannot therefore uphold measures which are
incompatible with fundamental rights recognized and protected by the Constitu-
tions of those States.’41 Besides reiterating its status as an observer of rights in the
Community,42 the Court in Nold KG made a pyrrhic step forward, by asserting that:
[i]nternational treaties for the protection of human rights on which the Member States have
collaborated or of which they are signatories, can supply guidelines which should be
followed within the framework of Community law43

Nold KG, independently, exemplified the first step of the Luxembourg Court to
engage with obligations under international treaties,44 and to form the path towards
assuming the responsibility for the respect of fundamental rights deriving from such
treaties. In a more concise context, Nold KG attributed to the Community the
international law component on the protection of human rights, providing that
persons subject to Community law claim rights and freedoms deriving from the
international treaties where Community Member States are contracting parties.
This understood, the Court not only fortified its human rights framework, but it
also advanced its legitimacy to observe and protect human rights, as the interna-
tional human rights law component was explicitly made binding internally. On the
other side, Nold KG simply opened the way for both the emergence of a Community
human rights order and the ECHR, keeping in mind that the most binding treaty for
Community Member States at that time was the ECHR.45 To this extent, Nold KG
served as the implicit step wherein the Court designed a soft relationship with the

40
Court of Justice of EU, Nold KG v. Commission, Case 4/73 [1974] ECR 491, judgment.
41
Nold KG v. Commission, Case 4/73 [1974] ECR 491, judgment, note 40, p. 507 (paragraph 13);
See also: Schimmelfennig (2006), p. 1248.
42
See generally: Morano-Foadi and Andreadakis (2011), p. 598.
43
Nold KG v. Commission, Case 4/73 [1974] ECR 491, judgment, note 40, p. 507 (paragraph 13).
44
On the international treaties’ route opened with Nold case, see also: Franklin (2010–2011),
p. 152.
45
See e.g.: Wetzel (2003), p. 2838.
2.2 The Origins of EU Becoming a Human Rights Law Organization with Reference. . . 43

ECHR, thereby expressing willingness to submit to the Member States’ tradition as


to the protection of human rights from an international perspective.46
Although Nold KG opened way for the international treaties to become embed-
ded in the Community human rights skeleton,47 the Court, in answering to a
reference for preliminary ruling, more profoundly, for the first time, overtly
referenced the ECHR in Rutili.48 By adjudicating the extent of the limitation of
rights, the Court in Rutili49 referred to the ECHR and its protocols, thereby judging
the application’s substantive merit inter alia on the basis of it. The logical reasoning
that the Court set forward seemed to defer rigidly to the ECHR, not only on basis of
a Community obligation but also on the basis of the general interest of human rights
in Europe’s public order. In light of Rutili, one can argue that the Court essentially
acknowledged the application of the ECHR at the Community level as a duty
deriving from the general principles of law which all Member States have adhered
to, therefore proving the distinguishing nature and relevance of the ECHR for the
Community’s human rights law. Nold KG and Rutili combined, for this reason,
paved the way for the domestic application of the ECHR at the Community level,
surely proving the intention of the Court to ensure that international human rights
law serves as guidelines that should be followed.
Following Nold KG and Rutili, the issue of fundamental rights became more
thoroughly substantiated in the minds of the EU policymakers. Although the
fundamental rights ‘context’ was fashioned by the Luxembourg Court, the three
political institutions of the EU started to become visible in this policy area. In 1977,
in a Joint Declaration—although legally not binding—the European Parliament, the
Commission and the Council for the first time officially articulated the scope of
human rights law at the Community level. Following the Court, the three institu-
tions loudly ‘stressed the prime importance they attach to the protection of funda-
mental rights, as derived in particular from the constitutions of the Member States
and the European Convention for the Protection of Human Rights and Fundamental
Freedoms.’50 The Joint Declaration proved to be the first political expression on the
issue of human rights law, thereby giving political support to the human rights
orientation of the Luxembourg Court’s case-law. Of prime importance is the fact

46
See also in this regard AG Trabucchi’s argument, asserting that: ‘As emerges from the case law
of this Court, the fundamental rights generally recognized by the Member States form an integral
part of our Community system, which, by drawing inspiration from the common traditions of the
Member States, guarantees respect for these rights within the limits of the powers conferred on the
Community and in accordance with the objectives assigned to it.’ (Opinion of Advocate-General
Trabucchi Delivered On 28 March 1974, P. 513, in Case. 4/73, Nold KG).
47
E.g.: Landau (2008), p. 558.
48
Court of Justice of EU, Roland Rutili v. The Minister for the Interior, Case 36/75, Reference for
Preliminary Ruling; Defeis (2000–2001), p. 311; See also how the Court started to treat ECHR as
inspiration for the growing human rights law: Lenaerts and de Smijter (2001), p. 90/1.
49
Court of Justice of EU, Roland Rutili, Case 36/75, Reference for Preliminary Ruling, p. 1232,
para. 32; See also: Jacque (2011), p. 999.
50
Joint Declaration, 27. 4. 77 Official Journal of the European Communities No C 103/1, p. 1.
44 2 EU Becoming a Human Rights Law Organization: Starting from Nowhere with a. . .

that the Joint Declaration gave principal significance to the ECHR, positioning it
distinctively in the context of Community human rights order. Not only did the
Joint Declaration give primary significance to the ECHR, but it is also important
with regard to the latter’s position in the human rights spectrum within the Com-
munity. This development can also be seen as representing a change in the political
‘direction’ of the Luxembourg Court to reflect on the ECHR with such weight.
Building upon its own case-law and the Joint Declaration, the Luxembourg
Court went further with Hauer 197951 in building on its relationship with the
ECHR, and setting precedence to judge on the basis of it. In this case, the Court
for the first time used the ECHR as a basis on which to answer the request for a
preliminary ruling, and judged the merits of the case primarily on the rights
deriving from the ECHR, which, in the Opinion of AG Capotorti, ‘must be
considered to have been incorporated into the Community legal order.’52 With
Hauer, the Court clearly attached itself to the ECHR’s rights and proved its primary
significance in the entire Community human rights system, even attributing to it a
higher position than the constitutional traditions/rights of the Member States. In
doing so, Hauer substantively switched the ECHR from an instrument of Member
States’ tradition to a more or less self-reliant Community general principle. Within
the broader perspective, Hauer proved important for human rights law being
established in the Community, as it implicitly gave direction on considering the
ECHR as the primary resource for such law, while recognizing the ECHR as a key
general principle of Community law.
It is therefore important to point out that the path followed by the Court to build
upon the human rights law framework in the Community followed a gradual
expansion. From a position where the Court engaged with human rights law from
its own perspective, it transformed itself into a more coherent, forward looking and
widely applicable system. With ERT53and SPUCI,54 the Court implicitly ruled that
the Member States’ national courts should also observe fundamental rights when
their respective governments apply Community law. In other words, human rights
law, especially ECHR, is to be applied by the Member States’ national courts when
considering the validity of national measures implementing the Community law.
Such a requirement, it is obvious, meant that national courts are obliged under the
Community law as an autonomous source of law to observe the respect for
fundamental rights when assessing their governments’ measures applying the
Community law. However, the Court implicitly repeated that it is the sole final

51
Court of Justice of EU, Hauer v. Land Rheinland-Pfalz, Case 44/79, Judgment of the Court of
13 December 1979, [1979] ECR 3727; See also: Landau (2008), p. 558; Sera (1996), p. 164;
Jacque (2011), p. 999.
52
Opinion of Mr Advocate General Capotorti in Case 44/79 Hauer, Delivered On 8 November
1979, p. 3760.
53
Court of Justice of EU, ERT Case, C-260/89, Reference for Preliminary Ruling, Court of
Justice, p. 2964.
54
Court of Justice of EU, Society for the Protection of Unborn Children Ireland, Case C-159/90
[1991] ECR I-4685, para. 31.
2.2 The Origins of EU Becoming a Human Rights Law Organization with Reference. . . 45

arbiter with regard to questions of fundamental rights in the Community, be it raised


within the Community or at the level of Member States implementing Community
law. This clearly obliges the national courts to refer preliminary ruling requests to
the Court when dealing with such cases. To ensure that it does not emerge beyond
Community law border, the Court e.g. in SPUCI55 also reiterated that it does not
have authority to judge the validity of a national rule regarding fundamental rights,
as this rests outside its competence.
However, a major issue in this regard is the extent to which an issue belongs to
Community law, and must therefore be checked against Community human rights
standards by the Court. Generally speaking, the Community encapsulated very
broad policy areas, therefore, one can say that there remains almost nothing to be
excluded from Community human rights law. In an effort to provide an answer to
request for preliminary ruling, the Court, in Kremzow,56 defined rather well the
borders between the Community human rights order and that of the Member
States.57 In drawing the external boundaries of ERT58 and SPUCI,59 the Court
was asked to judge the validity of a national criminal judgment in light of the
ECHR, as, the applicant, considered that the national judgment enters in the field of
the freedom of movement (a Community policy area) and, he argued, that the
Luxembourg Court must therefore determine whether the decision complies with
the ECHR. The Luxembourg Court, in Kremzow, responded concisely by ruling
that although the ‘deprivation of liberty may impede the person concerned from
exercising his right to free movement, the Court has held that a purely hypothetical
prospect of exercising that right does not establish a sufficient connection with
Community law to justify the application of Community provisions.’60 ‘The Com-
munity legal order is not all-embracing’61 when it comes to the human rights law
applied therein. In this way, the Court built the external border of its growing
human rights law, clearly omitting the consideration of supposed linkages between
national measures and the Community law as insufficient for a case to become
under the auspices of Community human rights law. Therefore the Court refused to
engage with such questions regarding the compliance of national issues with EC

55
Court of Justice of EU, Society for the Protection of Unborn Children Ireland, Case C-159/90
[1991] ECR I-4685, note 54, para. 31.
56
Court of Justice of EU, Kremzow, Case 299/95, preliminary ruling, 1997.
57
See e.g.: Defeis (2000–2001), p. 315/6.
58
Court of Justice of EU, ERT Case, C-260/89, Reference for Preliminary Ruling, Court of Justice,
note 53, p. 2964.
59
Court of Justice of EU, Society for the Protection of Unborn Children Ireland, Case C-159/90
[1991] ECR I-4685, note 54, para. 31.
60
Court of Justice of EU, Kremzow, Case 299/95, preliminary ruling, 1997, note 56, p. 2645 (para.
16).
61
Opinion of Advocate General La Pergola in Case. 299/95, delivered on 6 February
1997, p. 2635.
46 2 EU Becoming a Human Rights Law Organization: Starting from Nowhere with a. . .

human rights standards. By this time, the national constitutional courts had also
softened their ‘threats’ to the EC law, with Solange II62 of the German Constitu-
tional Court being the prime example of a standard which basically said that ‘as
long as’ human rights in the EC are equivalent to those in the German constitutional
order the EC law will not be reviewed for its validity at the national level. Certainly,
the Solange II doctrine—not only reversed Solange I standard—but also offered a
friendly and legitimizing support to the EC human rights law’s constitutional and
moral authority, contributing to its further development.
Three important issues are considered with regard to the changing attitude of
national constitutional courts towards EC law. First, that the issue of human rights
‘corpus juris’63 was established by the Luxembourg Court,64 whereas political
institutions had little influence in building a stable basis for this body of law.
Second, that the human rights law until the Single European Act 1987 were merely
a product of case-law—the Court ‘innovated’ those human rights law obligations
without any specific treaty basis—proving the concept of gouvernment desjuges65
wherein rights become ‘written’ and observed by the Court alone.66 And third, that
the case-law of the Court reflected nothing less than a general progressive attitude
towards the recognition of fundamental rights within the Community. This often
included reference to the ‘special significance’ of the Convention as well,67 which

62
BverfGE, Solange II – Wünsche Handelsgesellschaft decision of 22 October 1986, BVerfGE
73, 339, case number: 2 BvR 197/83, Europäische Grundrechte-Zeitschrift; See also: Škugor
(2012), p. 112.
63
Landau (2008), p. 561.
64
E.g.: de Búrca (2011), p. 650; See also: Landau (2008), p. 557; In making a comparison between
the past and now, Gragl argues that nowadays the Luxembourg Court seems less worried in
referring to the case-law of the Strasbourg Court. In this regard, Gragl has found out that the
Luxembourg Court now uses more often the standard formula of ‘as interpreted’ by ECtHR when
interpreting certain rights under EU law (Gragl 2013, p. 54).
65
Weiler (1986), p. 1115. Cf. Wetzel (2003), p. 2823, who argues that this approach of the EU
Court was a preventive measure from member states’ courts refusing the primacy of Community
law with the reason that it did not comply with national constitutional human rights’ standards; On
the latter, see also: Von Bogdandy et al. (2012), p. 495.
66
Which is both praised and sometimes criticized. See: Balfour (2007), pp. 185 et seq; On the
latter, see also: Phelps (2006), p. 276.
67
Court of Justice of EU, Opinion 2/94[1996] ECR I-1759, para. 33; See also: Balfour (2007),
p. 190; On the special significance of ECHR, see also: Franklin (2010–2011), p. 152; On specific
cases, see e.g.: Court of Justice of EU, Booker Aquacultur Ltd. v. The Scottish Ministers, Joined
Cases C-20/00 & C-64/00, [2003] E.C.R. 1-7411, para. 65; On modes of deference to ECtHR, see
also these two cases: Court of Justice of EU, Bauer Verlag, C-368/95 [1997] ECR I-3689, para.
26, and Court of Justice of EU, Grant v South West Trains, C-249/96 [1998] ECR I-621, para. 34;
Accord: Gragl argues that the Luxembourg Court has used ECHR as a very significant interpre-
tative tool, adding that Luxembourg has never shown more deference to any other international
tribunal than the Strasbourg Court (referring to literature from Douglas-Scott as well). See: Gragl
(2013), p. 54.
2.2 The Origins of EU Becoming a Human Rights Law Organization with Reference. . . 47

has recently been advanced as a direct reference to the Convention as a source of


law.68 The Court later also referred to a few other international human rights
instruments, such as the International Covenant on Civil and Political Rights in
Orkem v Commission69 and the UN Convention on the Rights of the Child in
Parliament v Council70—noticeably seeking more legal certainty by means of
enshrining rights on a treaty basis, partly as a means to justify the primacy of EU
law.71 On the latter point, many argue (e.g. Coppel & O’neill) that the rhetoric of
rights in the EU Court’s jurisprudence was packed with the aim of persuading
national courts as a tool to save the principle of supremacy before them therefore
completing the market project—rather than pursuing the human rights aims.72
Weiler & Lockhard promote the concept that Member States were reluctant to
engage in rigid control of EC human rights law as they were primarily interested in
the common market project, therefore showing satisfaction with a solid culture of
rights introduced by the EU Court. The author agrees with both Coppel & O’neill
and Weiler & Lockhard in this regard, however—building upon this argument—
one needs to say that the EU Court was also interested in improving its image before
the Member States and international community, in order to support its growing
constitutional authority with a human rights face and legitimacy. Therefore, the
more correct view would be to argue that the EU Court engaged in human rights law
cases due to pressures from Member States, therefore strategically trading their
courts’ pressure for maintaining the primacy of EU law. However the EU Court also
needed the human rights dimension to support its growing image as a democratic
institutional structure that offered not only governance but also rights to those
governed. Democratic intentions—either for sustaining its growing constitutional
authority or merely improving its image—may certainly have been at the center of
the motives of the Court as well.
The first step to provide for a treaty-based assertion on the human rights law was
undertaken by the Single European Act in 1987, which gave a special position to the
ECHR in its preamble,73 hence confirming the treaty-status of the ECHR as a
general principle of law in the Community independently of Member States’
legal orders. Despite the fact that the Single European Act proved the first treaty-
based devotion to human rights as general principles of law, especially to the

68
See e.g.: Court of Justice of EU, Baumbast, Case C-413/99 [2002] ECR I-7091, para. 72; See
also: Court of Justice of EU, Tielorassi v. Pliroforissis, Case C-260/89, [1991] E.C.R. 1-2925,
para. 41; Jacque (2011), p. 1000; On such judicial dialogue from the part of Luxembourg Court
with the Strasbourg Court and Convention system at-large, see: O’Meara (2011), p. 1815; Cf.
Callewaert (2014), p. 32/3, who argues that this practically meant that the Convention was treated
by the EU Court as legally applicable.
69
Court of Justice of EU, Orkem v Commission, Case 374/87 [1989] ECR 3283.
70
Court of Justice of EU, Parliament v Council, Case C-540/03 [2006] ECR I-5769.
71
E.g.: Schimmelfennig (2006), p. 1262.
72
See e.g.: Coppel and O’neill (1992), pp. 669 et seq; See also a general debate on this reluctance,
at: Weiler and Lockhart (1995a, b).
73
Single European Act, Official Journal of European Communities. N L 169/2, d. 29.06.1987, p. 2.
48 2 EU Becoming a Human Rights Law Organization: Starting from Nowhere with a. . .

ECHR, it nonetheless failed to provide a set of specific human rights to be observed.


The more rigorous treaty-based development in this regard was accomplished with
the Maastricht Treaty. This treaty formally asserted in its ordinary part that the
‘Union shall respect fundamental rights, as guaranteed by the European Convention
for the Protection of Human Rights and Fundamental Freedoms signed in Rome on
4 November 1950 and as they result from the constitutional traditions common to
the Member States, as general principles of Community law.’74 The same was
reiterated more thoroughly in the Amsterdam Treaty 1997.75 Taken together, the
Maastricht and the Amsterdam Treaty instituted the treaty basis for the human
rights law in the Union, giving the Court a written basis to build upon its human
rights law, thereby allowing that a more legally certain system of human rights
protection be devised in the Union. This not only provided a treaty basis for the
Court to strike down EU secondary law when it was found to conflict with the
ECHR and Member States’ human rights traditions,76 but it also allowed for the
interpretation of EU primary law in light of those standards.77 This significant
development is certainly in line with Forrest Martin’s concept on ‘International
Legal Constructionism’,78 which states that national constitutional frameworks
should be construed in line with the international law obligations. This argument
goes hand-in-hand with the observation of Thomas M. Franck who notes that ‘[t]he
surprising thing about international law is that nations ever obey its strictures or
carry out its mandates.’79 The EU seems to have not been able to escape from this
moral practice. It is therefore argued by the author that the EU has followed an
incremental path of liberalizing its borders to the influence of international law, and
especially to international human rights law such as the Convention. This approach
has certainly aligned the EU with—also in the lines of its constructive reception
of—a multilevel-Europe under the terms of globalization of law.
Coming back to the normative questions, the first complete set of human rights
in EU was introduced in 2000 by the European Parliament, the Council of Ministers
and the European Commission, titled the Charter of Fundamental Rights of the
European Union.80 The Charter, however, had no legal effect but was nevertheless a

74
Maastricht Treaty, Title I, Article F, para. 2, Official Journal of the European Communities, OJ
C 191 of 29.07.1992; See also: Miiller (2007), p. 34; Landau (2008), p. 558.
75
Amsterdam Treaty, Official Journal of the European Communities, OJ C 340 of 10.11.1997; See
also: Landau (2008), p. 560; Defeis (2000–2001), p. 302; Škugor (2012), p. 107.
76
On how the Luxembourg Court drew on ECHR as a comparative law reference, see: Lenaerts
(2003), pp. 878 et seq; See also: Court of Justice of EU, Spain v. UK and Northern Ireland, Case
145/04, Judgment of 12 September 2006, ECR, 2006, p. 1, specifically paras. 60 and 90.
77
In relation to ECHR, see e.g.: Martinico (2012), p. 402.
78
Martin (2004), pp. 272 et seq.
79
Franck (1988), p. 705.
80
See a very in-depth critique of the EU’s constitutional philosophy foundation, at: De Hert
(2005), pp. 215–253.
2.3 Strasbourg Looking Towards Luxembourg: What About a Refined Legal. . . 49

proclamation of rights that were to be followed by the Union institutions.81 The


Charter contains a set of written rights, pretty similar to those of national state
constitutions, and encapsulates—in contrast to the ECHR—not only civil and
political rights, but also economic and social rights. The Charter was foreseen to
take full legal effect with the Treaty Establishing a Constitution for Europe, which
failed to be ratified by all EU Member States. Generally speaking, the Charter
opened up the route for instituting a written set of rights for the Union, whereby
providing the first document of fundamental rights tracking the Luxembourg
Court’s route. With the Lisbon Treaty, which marks the most energetic period of
EU human rights law,82 as seen infra, the Charter became part of the EU primary
law and legally effective with that level of dominance. Human rights law in the
Union is therefore becoming increasingly important.83 ‘For reasons of legal clarity
and legal certainty, accession to the European Convention on Human Rights would
be a logical and sensible addition to the Charter.’84 This would further fortify the
concept on ‘International Legal Constructionism’ from the EU’s perspective. In the
words of De Vries, with accession ‘a new and possibly final stage in the remarkable
development of fundamental rights in EU law has begun.’85

2.3 Strasbourg Looking Towards Luxembourg: What


About a Refined Legal Arrangement?

On its side, Strasbourg Court has not stood unconscious and unresponsive of the
rising EU human rights law, which originally derived from the EU Member States
transfer of competences to the EU but was not in principle submitted to the
jurisdiction of the former. Callewaert in this regard rightly argues that ‘every
transfer of powers by the member states to the European Union has had the
concomitant effect of removing the exercise of the powers in question from the
[Strasbourg] Court’s scrutiny, although that is not provided for in the Convention
and, indeed, is not in keeping with its spirit.’86 Taking note on this situation, rather
early, in 1958, the European Commission on Human Rights—referred to here as the
Strasbourg Court—built a mechanism to ensure that the transferred competences
from Member States to the EU do not remain immunized from its review, by
establishing that:

81
See, for instance: ‘Chapter 11: Human rights in the EU’, at Craig and de Búrca (2007), p. 15.
82
E.g.: de Búrca (2011), p. 649.
83
Von Bogdandy (2000), p. 1338.
84
Krüger (2002–2003), p. 93.
85
de Vries (2013), p. 78.
86
Callewaert (2014), p. 15.
50 2 EU Becoming a Human Rights Law Organization: Starting from Nowhere with a. . .

If a State contracts treaty obligations and subsequently concludes another international


agreement which disables it from performing its obligations under the first treaty it will be
answerable for any resulting breach of its obligations under the earlier treaty.87

It was made clear by the Strasbourg Court since 1958 that the transferred
competences from the Member States to the EC/EU (at that time) may not discharge
those Member States from their obligations under the Convention law. Breuer calls
this doctrine in German: ‘keine Flucht ins V€olker-/Europarecht’.88 Virtually, the
Strasbourg Court produced the concept that although there could be a parallel EC
integration process taking place—with the EC not being a contracting party to the
Convention—the original Convention contracting parties maintain responsibility
for the violations of the EC as well.89 Not being plainly an act of hindering the
nature of EC integration from the context of previously undertaken obligations
under the Convention by state contracting parties, the Strasbourg Court however
overtly portrayed the Member States as the mere responsible parties for acts or
omissions which the EC might have undertaken in violation of the Convention; this
basically meant that there was no consensus from the Strasbourg Court to consider
the EC developing human rights law as being unreviewable from the Convention
system. The Strasbourg Court—in this period—and for a long time, however, kept
responding negatively to locus standi for parties which challenged EC/EU mea-
sures at the Strasbourg Court (e.g. Confederation Francaise Democratique du
Travail).90 Years later, the Strasbourg Court relaxed its tone and tried to establish
a rather conflict-free position91 towards the Luxembourg Court’s regime of human
rights law. In 1990, in M. & Co., the Strasbourg Court established that:
[t]he transfer of powers to an international organization is not incompatible with the
Convention provided that within that organization fundamental rights will receive an
equivalent protection. [. . .]The Commission notes that the legal system of the European

87
ECtHR (European Commission of Human Rights), N 235/56, Dec. 10.6.1958, Yearbook
2 p. 256 (300); See also: Škugor (2012), p. 114.
88
Breuer (2013), p. 1.
89
See also a discussion on this concept, at: Eckes (2013), p. 260.
90
ECtHR, Confederation Francaise Democratique du Travail v. the European Communities,
Application no. 8030/77 (D. 10.07.1978); See also: See: Gragl (2013), p. 65; Cf.: The Human
Rights Committee—in H. v. d.P. v. Netherlands—has ruled that ‘it can only receive and consider
communications in respect of claims that come under the jurisdiction of a State party to the
Covenant. The author’s grievances, however, concern the recruitment policies of an international
organisation, which cannot, in any way, be construed as coming within the jurisdiction of the
Netherlands or of any other State party to the International Covenant on Civil and Political Rights
[. . .].’ (Human Rights Committee, H. v. The Netherlands, Communication No. 217/1986,
U.N. Doc. CCPR/C/OP/2 at 70 (1990)). This stands quite in contrast to the practice of Strasbourg
Court, which somehow recognizes its jurisdiction on EU, although the latter is not a party to the
Convention.
91
Cf. Toth (1997), p. 499, who argues that as both courts have different aims—one to protect the
ECHR and the other to protect Community law—their conflicting interpretations are rather
normal; On this tension-free mood of Strasbourg Court, see also: Olsen (2009–2010), p. 64.
2.3 Strasbourg Looking Towards Luxembourg: What About a Refined Legal. . . 51

Communities not only secures fundamental rights but also provides for control of their
observance.92

M. & Co. seemed to provide for a negative legislating approach to the relation-
ship between Strasbourg and EC/EU regime of human rights, by establishing a
Solange II93 model of attributing immunity to the EC/EU if that is encountered with
the Convention system.94 By setting this standard, the Strasbourg Court did not
escape from its principle set in 1958, as explained above, but rather interpreted it in
a manner that does not open conflicts with the Luxembourg Court. On a substantive
level, the M. & Co. seems not to be intending to offer real human rights protection
towards the EC/EU, but rather serves the aim of reconciling the two orders. Such
management of possible conflicts between the two orders, nevertheless, renders it
impossible for the Convention to subsist at the level at which it functioned towards
the state contracting parties. Such immunization of the EC/EU and the Luxembourg
Court with the M. & Co. served as the first step towards institutionalizing the
conflict-free relationship between the Luxembourg and Strasbourg courts in a
way in which the Convention is still applicable to the EC/EU, as long as the said
presumption is valid. Six years after M. & Co., the Strasbourg Court went even
further to ensure that it still has some authority over the EC/EU regime of law, such
as in Cantoni,95 wherein the Court established that it has jurisdiction to review
EU-law implementing measures of Member States.96 Such behavior expressed by
the Strasbourg Court in Cantoni opened way to direct its ‘eyes’ in a straight line on
the EU law for possible review. Two landmark rulings of the Strasbourg Court
came after Cantoni, namely the Matthews and Bosphorus Airways cases.
In Matthews, explained infra,97 the Strasbourg Court still kept the tone towards
the Luxembourg Court relaxed, by establishing that acts of EC/EU (secondary law)
may not be challenged before the Strasbourg Court as long as the EC/EU is not a
party to the Convention. However, the Court ruled that this rule does not apply to
the EC/EU primary law, by establishing that:

92
ECtHR, M. & Co. v. Federal Republic of Germany, Application No. 13258/87, Dec.
9 February 1990.
93
The origin deriving from the Solange II (Solange II – Wünsche Handelsgesellschaft decision of
22 October 1986, BVerfGE 73, 339, case number: 2 BvR 197/83, Europäische Grundrechte-
Zeitschrift); See also the same explanation on Bosphorus, at: Miiller (2007), p. 37; On Solange II
of the German Federal Constitutional Court, see also: Franklin (2010–2011), p. 146 (‘[. . .] BverfG
clearly felt that a clash between EC and German national law over fundamental rights issues would
be less likely to occur. That did not mean, however, that it would abandon its jurisdiction to review
EC legislation in light of the German constitution entirely – it merely rendered such jurisdiction
dormant to the extent that the Community continued to prove itself capable of providing a level of
protection for fundamental rights substantially similar to that provided for under the German
constitution in the future.’).
94
See also: Quirico (2010), p. 41; Gragl (2013), p. 66.
95
ECtHR, Cantoni v. France, Application No. 17862/91, decision of 15/11/1996 (Grand
Chamber).
96
The same was established in Connolli, at: Gragl (2011), p. 70.
97
See the Chap. 10.
52 2 EU Becoming a Human Rights Law Organization: Starting from Nowhere with a. . .

Indeed, the 1976 Act cannot be challenged before the European Court of Justice for the very
reason that it is not a “normal” act of the Community, but it is a treaty within the
Community legal order. The Maastricht Treaty, too, is not an act of the Community, but
a treaty by which a revision of the EEC Treaty was brought about. The United Kingdom,
together with all the other parties to the Maastricht Treaty, is responsible ratione materiae
under Article 1 of the Convention and, in particular, under Article 3 of Protocol No. 1, for
the consequences of that Treaty.98

The Court made it clear that it has full jurisdiction to control the Conventional
legality of the EU Treaties,99 as they remained unreviewable for their validity by
the Luxembourg Court. This, as d’Aspremont generally argues, established the
principle that the EU Member States may not remain shielded by their international
responsibility if they effectively control the actions of a non-Convention interna-
tional organization like the EU.100 That said, in Mathews, the Strasbourg Court
embarked upon the EU Treaties to control the human rights protection, with the aim
of playing a role which the Luxembourg Court could not have played. By finding
the United Kingdom and all other parties to the Maastricht Treaty responsible
ratione materiae for the Maastricht Treaty’s violations of the Convention,101 the
Strasbourg Court recalled that it continued to hold EU Member States liable for
their transferred competences to EU primary law.102 By construing this new
relationship, the Strasbourg Court established momentum in enforcing the Conven-
tion rights jurisdictionally upon the EU Treaties but still preserved the place and
function of the Luxembourg Court. This was made—in principle—rather contrary
to the logic crafted by the ICJ’s Advisory Opinion, which for purposes of demon-
strating the nature of international obligations had ruled that: ‘It is well established
that in its treaty relations a State cannot be bound without its consent [. . .].’103 The
same rationale was applied by the Strasbourg Court in Soering v. United Kingdom,
where the Court had argued that ‘the Convention does not govern the actions of

98
ECtHR, Matthews v. United Kingdom, Application no. 24833/94, Decision of 18 February 1999
(Judgement, Grand Chamber).
99
The Strasbourg Court—in Mathews—basically applied the Inter-American Court of Human
Rights Velasquez Rodrıguez v. Honduras concept, where it had ruled that: ‘An illegal act which
violates human rights and which is initially not directly imputable to a State [. . .] can lead to
international responsibility of the State, not because of the act itself, but because of the lack of due
diligence to prevent the violation or to respond to it as required by the Convention.’ (Inter-
American Court of Human Rights, Velasquez Rodrıguez v. Honduras, Judgment of July 29, 1988,
Ser. C No. 4, para. 172) [emphasis added]; It was not essentially the issue whether the attribution of
responsibility should have been primarily placed upon the EU, but rather whether the EU Member
States had taken the necessary steps to remove such violation from the EU Treaties for which they
were liable before the Convention.
100
d’Aspremont (2007), p. 101.
101
See also: Lock (2012), p. 163; On the way the Strasbourg Court used to hold Member States
responsible for EC violations, see also: Klabbers (2009), p. 164; Kuiyper and Paasivirta
(2004), p. 131.
102
O’Meara (2011), p. 1816.
103
International Court of Justice, (Reservations to the Convention on the Prevention and Punish-
ment of the Crime of Genocide), Advisory Opinion, I.C.J. Reports 1951, 19 (28 May 1951), p. 21.
2.3 Strasbourg Looking Towards Luxembourg: What About a Refined Legal. . . 53

States not Parties to it, nor does it purport to be a means of requiring the
Contracting States to impose Convention standards on other States.’104 It would
have therefore been against this principle if the EC/EU would have been bound
directly to ECHR obligations. However, both Mathews and Bosphorus (the latter
examined infra) tend to revise this doctrinal foundation and develop the concept of
functional succession. On the other hand, the Commentary to the Draft Articles on
the Law of Treaties for international organizations explains why organizations like
the EU accept innovations such as those brought by Mathews and Bosphorus, by
asserting that ‘it can happen that an organization will be bound by legal rules
contained in a treaty without being a party to the treaty, either because the rules
have a customary character in relation to the organization, or because the organi-
zation has committed itself by way of a unilateral declaration.’105 Clearly, the EU
was obliged under this logic to allow for Strasbourg’s reign in this regard.
The other important case, Bosphorus Airways, as explained infra,106 kept the
Matthews principle on EU primary law in place, but went further with its position
towards EU secondary law, by ruling that:
The Court finds that the protection of fundamental rights by Community law can be
considered to be, and to have been at the relevant time, “equivalent” to that of the
Convention system. Consequently, the presumption arises that Ireland did not depart
from the requirements of the Convention when it implemented legal obligations flowing
from its membership of the European Community.107

It clearly seems that the Strasbourg Court crafted the ‘equivalent protection’
doctrine,108 further polishing its preliminary nature as set in M. & Co. It established

104
ECtHR (plen.), Soering v. United Kingdom, judgment of 7 July 1989, Series A, No. 161, para.
86 [emphasis added].
105
Draft Articles On The Law Of Treaties Between States And International Organizations Or
Between International Organizations With Commentaries 1982. Text adopted by the International
Law Commission at its thirty-fourth session, in 1982, and submitted to the General Assembly as a
part of the Commission’s report covering the work of that session (at para. 63). The report, which
also contains commentaries on the draft articles, appears in Yearbook of the International Law
Commission, 1982, vol. II, Part Two, p. 20.
106
See Chap. 10.
107
ECtHR, Bosphorus Hava Yollari Turizm ve Ticaret Anonim Şirketi v. Ireland, Application
No. 45036/95 (Judgment of 30 June 2005); Cf.: Gragl (2013), p. 71, who argues that the product
devised by the Court in Bosphorus tries to balance the principles of states’ sovereignty and
autonomy to conclude international agreements, on the one hand, with the human rights obliga-
tions flowing from the Convention system, on the other hand.
108
E.g.: Lock (2010), p. 779; See also: Quirico (2010), p. 42; Lavranos (2008), p. 604; Phelps
(2006), p. 271/2; Craig (2013), p. 1136; Cf.: de Rivery and Chassaing (2013), p. 4, who argues that
the doctrine of equivalent protection is established and will remain applicable only for as long as
EU does not accede to ECHR; Cf.: Gragl basically argues that under the present system an
applicant must make an in-depth investigation to find out whether in his/her case the system of
protection under the EU law is equivalent to that of the ECHR—the Bosphorus presumption—
which is basically a very complex task to be addressed with certain unknown factors (see: Gragl
2013, p. 140). This being the case, it is logically argued that the present Bosphorus formula makes
it not possible for applicants to make use of the ECHR system.
54 2 EU Becoming a Human Rights Law Organization: Starting from Nowhere with a. . .

a rather fine-tuned deference to the Luxembourg Court’s jurisdiction on human


rights, as long as there is no ‘manifest deficiency’,109 the latter being a situation
wherein fundamental rights are blatantly violated. Evidently, the presumption of
equivalent protection is assumed as long as there was no state discretion exercised
by a Member State.110 On the contrary, in a very recent case in 2012, Michaud
v. France, the Strasbourg Court pronounced that it does not apply the doctrine of
equivalent protection to an EU-originated case if it has not gone through the filter of
the Luxembourg Court,111 clearly recognizing the latter’s core role in the design of
the application of the doctrine concerned. The doctrine concerned is also not
applicable in cases where EU Member States exercise state discretion, as decided
in M.S.S. v. Belgium and Greece.112
With Bosphorus Airways, therefore, the Strasbourg Court laid down the method
for a mediated dialogue with the Luxembourg Court,113 which undoubtedly
amounts to a form of immunizing the latter and in turn eroding the real and effective
human rights protection for which the Convention is meant to exist. In addition, the
Court had ruled that the Convention is a ‘constitutional instrument of European
public order,’114 thereby probably attempting to legitimize its passive and defensive
approach towards the Luxembourg Court, although formally holding that the
Convention stands above the EU legal order in the European human rights law
hierarchy. The equivalent protection doctrine left outside Strasbourg’s jurisdiction
all potential claimants which were subject to an EU secondary law action or

109
See e.g.: Phelps (2006), p. 272; Lock (2012), p. 164; See also: Eckes (2013), p. 261; Gragl
(2013), p. 72; Hoffmeister (2006), pp. 447 et seq; See also: Besselink (2013), p. 309; See also:
O’Meara (2011), p. 1817; Cf.: Ryngaert (2011), p. 1014 (‘[. . .]under the Bosphorus standard,
Member States appear to be required to screen the IO’s human rights performance continuously.’)
[emphasis added].
110
Accord.: ECtHR, Connolly v 15 Member States of the European Union, Application No. 73274/
01, (Section V), December 9, 2008; See also: Lock (2010), p. 779; Besselink (2013), p. 308; See
e.g.: Tulkens (2013), p. 4, who refers to Strasbourg’s Michaud v. France case to demonstrate how
Strasbourg may choose to depart from the doctrine of equivalent protection if the act of the
member state has involved state discretion in implementing EU law.
111
ECtHR, Michaud v. France, Application No. 12323/11, 6 December 2012.
112
ECtHR, M.S.S. v. Belgium and Greece, Application No. 30696/09, 21 January 2011.
113
Peers (2006), p. 452 (‘The compromise position permits a conditional, rather than an unlimited,
review of member states’ actions within the scope of Community law’); See also: Gragl (2013),
p. 73, who argues that through the Bosphorus approach, the Strasbourg Court has kept itself
outside a potential ‘power game’ with the EU Court, therefore recognizing and deferring to the
flexibility required for international cooperation and the right of the ECHR contracting parties to
enter into international legal obligations on basis of their own choice. This, according to Gragl,
‘led to the notion that Strasbourg was especially open-minded and tolerant towards international
law and Union law.’ (also referring to Haratsch).
114
Preliminarily ruled in: ECtHR, Loizidou v. Turkey (preliminary objections), judgment of
23 March 1995, Series A no. 310, pp. 27–28, para. 75; Cf.: Sohn (1995), who argues basically
that UN-originated instruments like the International Covenant on Civil and Political Rights lack a
constitutional status within the context of hierarchy of law, which I argue the ECHR enjoys in
Europe.
2.3 Strasbourg Looking Towards Luxembourg: What About a Refined Legal. . . 55

omission, substituting the Strasbourg Court with the Luxembourg Court, the latter
being merely a domestic court for purposes of interpretation of EU law, but indeed
avoiding potential normative conflicts between the two orders. One must also say
that this presumption may well put the EU in a fundamental rights deadlock and
‘undermine the EU’s commitment to genuine accountability in its fundamental
rights activities.’115 Neuman rightly stressed that ‘[m]aking a human right more
“effective” does not necessarily mean giving the right a broader meaning.’116
Bosphorus seems to evidently support a view which gives importance to ‘how
broadly’ is a right guaranteed rather than its effectiveness. The author of this book
would obviously stand by Neuman’s argument when it comes to the fact that
effectiveness of rights is far more important than their scope in form, the latter—
not the former—being something which Bosphorus tries to ‘sell’ as human-right
protection. From a comparative perspective, it is worth mentioning that the Inter-
American Court of Human Rights would have disagreed with Bosphorus, taking
into account that it had ruled in Velasquez Rodrıguez v. Honduras that ‘[t]he
obligation to ensure the free and full exercise of human rights is not fulfilled by
the existence of a legal system designed to make it possible to comply with this
obligation—it also requires the government to conduct itself so as to effectively
ensure the free and full exercise of human rights.’117 Bosphorus clearly places
emphasis on the existence of the system rather than its effectiveness in providing
protection,118 if the issue is considered from the margins of Velasquez Rodrıguez
v. Honduras case. Accession would therefore seem the most reliable response to
this potential legitimacy threat, as Judge Ress similarly asserted in his dissenting
opinion on Bosphorus.119
From a more strategic perspective on Bosphorus, Merrill argues that the Stras-
bourg Court usually follows a non-conflicting approach with other treaty regimes
when it comes to the interpretation of rights, trying always to use them as sources of
assistance for legitimizing its ‘resourcefulness’.120 The author of this book agrees
with Merrill that the approach of the Strasbourg Court—which is also followed in

115
O’Meara (2011), p. 1817.
116
Neuman (2008), p. 115.
117
Inter-American Court of Human Rights, Velasquez Rodrıguez v. Honduras, Judgment of July
29, 1988, Ser. C No. 4, para. 167 [emphasis added].
118
Contra: Tulkens (2013), p. 3, who argues that ‘[. . .] any such finding of equivalence could not
be final and would be susceptible to review in the light of any relevant change in fundamental
rights protection.’. Tulkens, however, seems to refer to a notional picture rather than to the strict
reading of the Bosphorus approach.
119
On Judge Ress, see: White (2010), p. 440; Cf.: Ladenburger (2012), p. 45 (Available at: www.
fide2012.eu/index.php?doc_id¼88), which shows the EU Member States debates disagreeing on
making Bosphorus standard textually part of the DAA; See also: Besselink (2013), p. 311/2, who
argues that the big reason why Bosphorus should be abandoned after accession is because
accession is being undertaken to reach exactly the opposite of this doctrine: namely, to allow
parties reach the Strasbourg Court and attack before it EU-originated violations.
120
Merrill (1993), p. 218.
56 2 EU Becoming a Human Rights Law Organization: Starting from Nowhere with a. . .

Bosphorus—helps it build a friendly connection with other treaty regimes, however


this will not necessarily translate into Strasbourg playing its natural role more
vigorously. Thus, although Merrill’s argument may appear to appreciate Bosphorus
for the way it addresses the EU regime of law, this may not be appreciated if one
looks it from the substantive human rights perspective. All said, the Strasbourg
Court’s approach towards the Luxembourg regime of law has been mainly one of
friendly dialoguer rather than inflexible competitor.121 With this stance, the Con-
vention system became a practical instrument for serving the peculiarities of the EU
system of law,122 adding that ‘both European courts displayed comity towards each
other by allowing each court to “reign over their own kingdoms” without having to
fear any interference from each other – except in exceptional cases.’123 However,
these landmark decisions of the Strasbourg Court proved rather restrictive, and
restrained the effectiveness of the Convention rights, giving priority to the formal-
istic presumption of equivalent protection rather than substantive human rights
protection.124 Trading off the Convention’s effectiveness with the preservation of
EU law peculiarities seems smart neither in face of international law nor in the face
of the unequal footing between the state parties and the EC/EU before the Conven-
tion. Certainly, these loopholes will make it even more necessary for the EU to
accede to the ECHR—such accession being ‘crucial [. . .for. . .] the full effective-
ness of the European human rights system [. . .]’125—an issue that will be covered in
the next chapter.126

121
See e.g.: De Londras and Kingston (2010), p. 401; Cf. Chalmers et al. (2010), p. 260, who
observe: ‘One has the feeling almost of a “nonaggression pact” between the two European courts,
whereby the Court of Justice will slavishly follow the case law of the European Court of Human
Rights, whereas the latter will intervene only in cases of the most grotesque dysfunction.’.
122
See also, e.g.: Olsen (2009–2010), p. 70/1.
123
Lavranos (2008), p. 606.
124
In this regard, Gragl argues that CFTD, M & Co, Cantoni, Mathews and Bosphorus would have
had a different conclusion if EU was a contracting party to the Convention (see: Gragl 2013,
p. 140). Gragl’s point is valid as regards the jurisdictional position of the Strasbourg Court upon
these cases, although this may not be said so indisputably with regard to the substantive conclu-
sions on these cases.
125
AIRE Centre et al. (2013), p. 1.
126
On the need for accession rather than the scenario wherein EU becomes obliged under ECHR
based on the functional succession of its Member States, see: Odermatt (2014), p. 12; Cf. Heringa
and Verhey (2011), p. 31, who argues that: ‘The optimal approach would certainly have been an
accession by the EU to the Council of Europe and the ECHR.’
2.4 EU Charter on Fundamental Rights and Its Normative Relationship with the ECHR 57

2.4 EU Charter on Fundamental Rights and Its Normative


Relationship with the ECHR

European Union—as explained infra—has its own internal bill of rights,127 the
Charter of Fundamental Rights and Freedoms. It is the first written document
containing an explicit set of rights and freedoms in the European Union, which
logically followed a case-law tradition of fundamental rights built by the Luxem-
bourg Court for many decades. The Charter not only represents a document of
rights but also a segment of law that is utilized today by every citizen who is
subject—directly or indirectly—to EU authority. The Charter, therefore, swims in
the waters of EU law, the latter being supported by other sources of human rights
law, such as constitutional traditions of Member States, the ECHR, and general
principles of law. To the extent that the Charter is now the primary source of law for
human rights, it cohabitates with the ECHR both internally and externally. Such
symbiotic interdependence requires analysis.
The Charter was initially adopted by a Convention128 representing EU and
Member States’ multi-level stakeholders in October 2000. It was subsequently
approved by the Biarritz European Council. In justifying the adoption of the
Charter, the European Council has noted that:
Protection of fundamental rights is a founding principle of the Union and an indispensable
prerequisite for her legitimacy. The obligation of the Union to respect fundamental rights
has been confirmed and defined by the jurisprudence of the European Court of Justice.
There appears to be a need, at the present stage of the Union’s development, to establish a
Charter of fundamental rights in order to make their overriding importance and relevance
more visible to the Union’s citizens.129

The Charter was undoubtedly marked as a fundamental tool of legitimacy and/or


relevance for the citizens’ standard of human rights. With its approval, however, the
European Council neither expounded upon its legal authority nor offered grounds
for this assumption. Therefore, the Charter at the outset was adopted to serve
mainly as a moral document that mirrors a bill of rights, without enjoying legal
authority to bind the EU and its institutions. The question of lack of legal authority
for the Charter did not prevent the Luxembourg Court to refer to it,130 and to utilize
it as an informal source of law producing compulsory effect towards potential
claimants. Soon after that, there were several attempts to discuss and resolve the
issue of legal status of the Charter. Both the Nice European Council and the

127
Gragl rightly notes that the Charter refers to several layers of fundamental rights law: EU,
member states’ constitutional laws and international law (ECHR). To this extent, Gragl proposes
that Art. 52 and Art. 53 ChFR take this situation into consideration, and lay down rules to settle
disputes that may arise in the management of this interaction. See: Gragl (2013), p. 58.
128
Callewaert (2014), p. 34.
129
European Council Decision on the Drawing Up of a Charter of Fundamental Right of the
European Union, Annex II, para. 1.
130
E.g. Court of Justice of EU, Unibet v. Justitiekanslern, Case C-432/05 [2007] ECR I-2271.
58 2 EU Becoming a Human Rights Law Organization: Starting from Nowhere with a. . .

Convention on the Future of Europe (which drafted the Constitutional Treaty)


failed to produce an effective decision on the legal status of the Charter. Years
after the failure of the Constitutional Treaty, the issue of the legal standing of the
Charter was finally addressed effectively with the successful approval of the Lisbon
Treaty.131 The latter not only laid down a legal nature for the Charter, but also
incorporated it in the hierarchical domain of EU primary law. With the primary law
status, the Charter vitally empowered its nature of effect and the overall impact on
the EU legal order. With the Charter today having primary law status, one may
argue that the EU has an established human rights law from a constitutional
perspective. The Charter, on the other hand, enshrines not only civil and political
rights like the ECHR, but also social and economic rights which further expand the
reach and mode of protection of the EU’s bill of rights. Balancing these rights at the
same time as adhering to the Convention system and the multilevel interface of
human rights law in Europe remains a rather complex task for the Luxembourg
Court. Technically speaking, the Charter is separated into six individual human-
right categories’ chapters and one chapter on general provisions; such chapters
sorting rights into the following classes: dignity, freedoms, equality, solidarity,
citizens’ rights, justice, and the last chapter setting the general provisions.
Callewaert, for example, argues that the Charter’s provisions are largely an imita-
tion of the parallel provisions of the Convention on human rights, but the Charter’s
drafters have tried to make them simpler and sometimes more current to mirror the
needs of modern situations. Callewaert attractively supports the latter argument, for
instance, with the example of the word ‘communications’ appearing in the Charter
which in fact mirrors the word ‘correspondence’ from the Convention,132 clearly
offering more room to the latest technological developments as to the scale and
nature of communication. In certain cases, the Charter clearly offers broader rights
than those written in the text of the Convention.133
Interestingly enough, the Luxembourg Court—in N.S. and Others—has ruled
that the Charter merely confirms the rights and freedoms ‘recognised in the Union
and makes those rights more visible, but does not create new rights or princi-
ples.’134 Whether the standard on visibility is meant as reinforcing certain rights or
merely making them formally more visible is not noticeable from an outsider’s
point of view. However, one does not understand the motivation behind
Luxembourg’s desire to diminish the function of the Charter. On the other hand,
with regard to the scale of application, Art. 51(2) of the Charter reads:

131
See also: Arold Lorenz et al. (2013), pp. 244 et seq, who cites a wealthy scholarship that
basically argues that with the adoption of the Lisbon Treaty, the Charter and the Convention do not
seem to turn into two mutually exclusive instruments. Instead, she argues, the EU Court pays a
recognized respect for the Strasbourg Court, whereas, with the horizontal clauses which the
Charter possesses, the Charter’s relationship with the Convention will not be one of competition
but rather one of mutual harmony.
132
Callewaert (2014), p. 36; See also: Gragl (2013), p. 58/9.
133
Callewaert (2014), p. 36.
134
Court of Justice of EU, N.S. and Others, C-411/10 and C-493/10, 21 December 2011, para. 119.
2.4 EU Charter on Fundamental Rights and Its Normative Relationship with the ECHR 59

The Charter does not extend the field of application of Union law beyond the powers of the
Union or establish any new power or task for the Union, or modify powers and tasks as
defined in the Treaties.

That observed, the Charter applies only to EU affairs. The substantive test to
apply the Charter—which Luxembourg Court devised in Åkerberg Fransson135—
needs to demonstrate that there is an organic link between the action or omission
and the EU’s involvement in a certain issue for the Charter to apply. The Court
adopted a far more restrictive approach in Siragusa (deferring rather well to the
approaches taken in Iida136 and Annibaldi137) by establishing that:
In order to determine whether national legislation involves the implementation of EU law
for the purposes of Article 51 of the Charter, some of the points to be determined are
whether that legislation is intended to implement a provision of EU law; the nature of that
legislation and whether it pursues objectives other than those covered by EU law, even if it
is capable of indirectly affecting EU law; and also whether there are specific rules of EU
law on the matter or capable of affecting it.138

The Court seems to have developed four criteria—as opposed to a very general
and flexible criterion developed under Fransson—to determine the scope of Mem-
ber States’ obligations under the Charter when they implement EU law, as follows:
first, whether the national legislation regulating the case before the national court is
intended to implement an EU-law provision. In case the national legislation
concerned is not intended to convey an EU-law obligation, it would not fulfill
this criterion. Therefore the Charter’s right does not extend its application on that
national legislation (e.g. Maurin).139 The idea of making a difference between the
national legislation intended to implement EU law and that not intended to imple-
ment EU law is to demonstrate the difference between the EU as a scope of
governance and EU-law specific obligations which member states’ need to address
in the most tangible context; second, if yes, the nature of that legislation and
whether its objectives include non-EU law elements. This criterion seeks to ensure
that the national legislation concerned is within the EU-law objectives proclaimed
by specific EU-law acts. A national measure intending to reach a state-relevant
objective (as opposed to an EU-law objective), as long as there is no such objective
under EU law, would not pass this criterion. One example is when the scope of an
EU act and national act have roughly the same scope, but their objectives remain
different. If the national legislation does not intend to meet the same objectives, it

135
Court of Justice of EU, Åkerberg Fransson, C-617/10, 26 February 2013, paras 19 et seq; Gragl
rightly notes that a similar broad test of the ‘scope’ of Member States fundamental rights
obligations when implementing EU law is the ERT case (C-260/89). See on this: Gragl (2013),
p. 55.
136
Court of Justice of EU, Iida, Case C-40/11 [2012] ECR, para. 79.
137
Court of Justice of EU, Annibaldi, Case C-309/96 [1997] ECR I-7493, paras 21–23.
138
Court of Justice of EU, Cruciano Siragusa v. Regione, Case C-206/13, Judgement of the Court
of 6 March 2014, para. 25.
139
Court of Justice of EU, Maurin, Case C-144/95 [1996] ECR I-2909, paras 11/12.
60 2 EU Becoming a Human Rights Law Organization: Starting from Nowhere with a. . .

would then fall outside the scope of the Charter.140 To digress, with regard to this, in
Iida, the Court conducted this test by noting that ‘[w]hile Paragraph 5 of the
FreizügG/EU, which provides for the issue of a “residence card of a family member
of a Union citizen”, is indeed intended to implement European Union law, it is none
the less the case that the situation of the claimant in the main proceedings is not
governed by European Union law, since he does not satisfy the conditions for the
grant of that card in accordance with Article 10 of Directive 2004/38. Moreover, in
the absence of an application by him for the status of long-term resident in
accordance with Directive 2003/109, his situation shows no connection with
European Union law.’141 Here, the Court applied a negative test, by examining
whether the applicant’s position is affected by the EU legislation and if the latter
intended to regulate his/her position. As the EU legislation had not prescribed the
objective to regulate the situations of those who do not apply for the residence card,
the Court was right to conclude that it is not the objective of EU law to regulate the
status of those who do not apply but rather of those who apply and are given the
residence card. This negative test seems to make the application of this criterion far
more restrictive compared to a normal test. On the other hand, with regard to this
criterion, one also needs ‘consider the objective of protecting fundamental rights in
EU law, which is to ensure that those rights are not infringed in areas of EU activity,
whether through action at EU level or through the implementation of EU law by the
Member States’142; third, whether the national legislation concerned would be able
to indirectly affect EU law. This criterion basically means that even if the national
legislation concerned falls outside the EU-law objectives, the mere fact that it can
affect indirectly the EU law may serve sufficiently to have this criterion passed.
This may be the case when the national legislation X implements simultaneously an
EU law obligation and many other national law objectives; and, fourth, whether
there are specific EU law rules on the subject of the case at hand or capable of
affecting it, meaning if there are rules at the EU law level—explicit or indirect—
that regulate the matter at hand. Fulfilling these four criteria may not only be
complex but sometimes is not uncontroversial. The four criteria do not seem
cumulative.
This means in practice that the Charter may not impose on Member States new
duties except those which they themselves have transferred to the EU.143 Such
limitation serves the Member States’ need to be assured that the Charter does not
impose new duties on them and will not become a source of authority to transfer

140
Court of Justice of EU, Cruciano Siragusa v. Regione, Case C-206/13, Judgement of the Court
of 6 March 2014, note 138, para. 28.
141
Court of Justice of EU, Iida, Case C-40/11 [2012] ECR, op cit. note 136, para. 80.
142
Court of Justice of EU, Cruciano Siragusa v. Regione, Case C-206/13, Judgement of the Court
of 6 March 2014, note 138, para. 31.
143
Contra.: Gragl argues that by extending the basis of the application of fundamental rights
limitations on Member States when they apply EU law, the Luxembourg Court actively expanded
the areas of EU law over national law, therefore simultaneously ‘increasing the EU competences at
the expense of those of the Member States’. See: Gragl (2013), p. 55/6.
2.4 EU Charter on Fundamental Rights and Its Normative Relationship with the ECHR 61

more competences to the EU. With regard to the complex problem of identification
of the extent of Member States’ action within EU law, Timmermans rightly argues
that the scope of application of the Charter on EU Member States’ acts and
omissions when they implement EU law is rather vague and follows a mixture of
accents (as seen also above in Siragusa).144 Agreeing with Timmermans, one
should extend this argument in light of the fact that the lack of a clear set of
standards may amplify the problem of applying the ECHR when implementing EU
law from the perspective of EU Member States as well. Such a problem may mix
the nature of the relationship between the internal ECHR obligations of Member
States’ courts and EU obligations under the ECHR when implementing EU law.
This limited guidance for implementing the ECHR in a non-defined set of situations
will certainly increase legal uncertainty and make human rights protection flawed
from a citizen’s perspective. Therefore, it is suggested that clearer procedural
safeguards be developed to clarify the scope of application and degree of obligation
to apply the Charter by EU Member States from the context of interaction between
the national legal orders and EU law post-accession. Roughly the same view
(although in a different context) with the latter has been maintained by the Lux-
embourg Court in Opinion 2/13, the latter requiring certain additional safeguards in
the DAA—of the type explained above—to ensure that EU Member States’ obli-
gations under the ECHR do not hinder the effectiveness of the Charter when the
latter is applied by national courts. The Luxembourg Court—in this connection—
unfortunately did not understand that such safeguards are of an EU-law domain and
should be set by the Luxembourg Court or EU law, whereas the DAA should in
principle seem as irrelevant towards that need.
The Charter fulfills its role as a bill of rights in the EU by directly interacting
with other instruments such as the ECHR. In setting a basic standard of protection,
the Charter in Art. 52(3) reads:
Insofar as this Charter contains rights which correspond to rights guaranteed by the
Convention for the Protection of Human Rights and Fundamental Freedoms, the meaning
and scope of those rights shall be the same as those laid down by the said Convention. This
provision shall not prevent Union law providing more extensive protection.145

Art. 52 (3) sets a very important rule as to the standard of protection: it makes the
Convention the basic or minimum standard of human rights whereupon the Charter

144
Timmermans (2014), p. 338/9.
145
Gragl argues that the last word of Art. 52 (3) ChFR, establishing that EU law may provide more
extensive protection, marks the mechanism to preserve the EU law autonomy. Art. 52 ChFR,
according to Gragl, is a ‘dynamic norm of reference’ to the Strasbourg’s case-law. See: Gragl
(2013), p. 59. I disagree on this argument with Gragl, as the above provision seems to oblige
systemically the Luxembourg Court to the Convention’s standards rather than request it to refer to
it for ‘cross-referencing’ purposes. This line of subordination to the Strasbourg Court’s case-law
may lead also to interference to EU law autonomy, as the latter would take precedence over the
Luxembourg’s fundamental rights exclusive jurisdiction.
62 2 EU Becoming a Human Rights Law Organization: Starting from Nowhere with a. . .

rights shall be interpreted and built upon.146 In devising this model of protection,
the Charter tries to reach two deep-seated intentions: first, to recognize the function
of the Convention as an instrument of European public order, constituting a
constitutional tradition of all Member States as well; and, second, shielding all
potential criticisms on the level of protection which the EU guarantees: something
that may not go below the recognized Convention’s baseline. It seems clear that the
Charter not only tries to instill a standard of deferring to the Convention system, but
also to keep all potential reactions from Member States’ courts sealed under this
stamp. This has also been further supported with direct references from the Lux-
embourg Court to the caselaw of the Strasbourg Court, such as in Secretary of State
for the Home Department v Akrich.147 Recognizing the Convention as the mini-
mum standard, nevertheless, makes the Charter a rather ambitious and advanced
bill of rights, giving the possibility to the EU to go beyond the standard established
by the Strasbourg regime of law. The EU therefore is not barred from offering
higher human rights protection; instead AG Cruz Villalon argues that ‘[a] contrario
that prevents the European Union adopting measures affording less extensive
protection.’148
Relating to the standard of protection reflected in the Convention system, the
Luxembourg Court has articulated in Otis the mechanics of this principle, although
using rather vague language. It had ruled that: ‘Article 47 of the Charter secures in
EU law the protection afforded by Article 6(1) of the ECHR. It is necessary,
therefore, to refer only to Article 47.’149 Otis exemplifies a landmark case to
demonstrate how the Luxembourg Court utilized and built upon Art. 52(3) of the
Charter, thereby deconstructing the character of the interaction between the Charter
and the Convention from the minimum-standard perspective. Otis illustrates that
the Luxembourg Court merely looks at whether a certain right enumerated in the
Charter is also similarly reflected in the Convention. This is nothing more than a
textual reading of the Convention’s rights—although it was the Luxembourg Court
in Dereci which had itself accepted that the case-law of Strasbourg Court is also
part of this commitment150—which may never show the real nature of the extent of

146
Drawing on this obligation, Gragl (referring to Costa and Skouris) names this interaction
‘parallel interpretation’. See: Gragl (2013), p. 58.
147
Court of Justice of EU, Secretary of State for the Home Department v Akrich, Case C-109/01
[2003] ECR I-9607.
148
Opinion of AG Cruz Villalon in Case C-120/10, European Air Transport SA v. Collège
d’Environnement de la Région de Bruxelles-Capitale, delivered on 17 February 2011, para. 79.
149
Court of Justice of EU, Otis, Case C-199/11 [2012], para. 47; See also: Court of Justice of EU,
Chalkor v Commission, Case C-386/10 P [2011] ECR I-13085, para. 51.
150
Court of Justice of EU, Dereci, C-256/11, 15 November 2011, para. 70; Accord: This is also
supported by the Official Explanations Relating to the Charter of Fundamental Rights, which,
regarding this specific provision, assert: ‘The meaning and the scope of the guaranteed rights are
determined not only by the text of those instruments, but also by the case-law of the European
Court of Human Rights and by the Court of Justice of the European Union.’ (Explanations Relating
to the Charter of Fundamental Rights, OJ C 303, 14.12.2007, pp. 17–35. Explanation on Article
52—Scope and interpretation of rights and principles).
2.4 EU Charter on Fundamental Rights and Its Normative Relationship with the ECHR 63

that right which the Convention system provides through the substantive standards
developed by the Strasbourg Court. There are several other examples demonstrating
the same attitude of the Luxembourg Court as in Otis.151 This shows that the
Luxembourg Court often tries to use a very formal deference to the Convention
system, compared to a substantive one which the Charter logically directs. This
being the practice, one may legitimately argue that Luxembourg’s approach to the
standard of protection in relation to the Convention is hardly within the required
format and does not meet the intended outcome which the Charter makes
obligatory.
This is what De Rivery and Chassaing label as the ‘Charter-centrism’152—as
opposed to the Convention-centrism—which the Luxembourg Court uses as an
autonomous policy, although outside the ambit of the normative obligation of Art.
52(3) of the Charter. This autonomous approach shows the Luxembourg’s concept
on human rights as internally developed—something that goes back to
Internationale Handelsgesellschaft—although it does not stand in accordance
with the treaty obligation to defer substantively to the Convention system, at least
with regard to the minimum standard. In following this formal deference
approach—without substantive consideration of the level of protection which the
Convention system offers—the Luxembourg Court departs significantly from the
rationale on which the Convention system relies, as prescribed in Tyrer. The
Strasbourg Court had ruled in Tyrer that: ‘[t]he Court must also recall that the
Convention is a living instrument which, as the Commission rightly stressed, must
be interpreted in the light of present-day conditions.’153 This means that the
Convention rights need be developed consistently and progressively, with their
scope and nature changing over time and becoming enhanced. The argument goes,
therefore, that it would be impossible to follow the standards of the Convention
system if the Convention were not respected and adhered to as a living instrument.
Luxembourg’s approach of disregarding the substantive nature of rights as built by
Strasbourg’s caselaw makes it impossible for the Charter to follow the Conven-
tion’s minimum standard if the latter’s ‘living’ character is not taken into account.
This said, this kind of ‘autonomous interpretation’ of human rights rooted in the EU
legal order—originating from Internationale Handelsgesellschaft—would misalign
the Charter with the Convention, and the divergence between Otis and Tyrer makes
this apparent. This divergence has also been expressed by the Strasbourg Court in
Micallef v. Malta, where it ruled that:
Article 47 of the Charter of Fundamental Rights of the European Union guarantees the right
to a fair trial. Unlike Article 6 of the Convention, the provision of the Charter does not

151
Callewaert (2014), p. 40.
152
de Rivery and Chassaing (2013), p. 3; Cf.: Gragl in this regard points to the fact that the
Luxembourg Court has regularly used the standards developed under the ECHR system as
‘interpretive tool’ to enlighten the meaning of certain fundamental rights in the EU law. (Gragl
2013, p. 54).
153
ECtHR, Tyrer v UK, No 5856/72, 25 April 1978, para. 31 [emphasis added].
64 2 EU Becoming a Human Rights Law Organization: Starting from Nowhere with a. . .

confine this right to disputes relating to “civil rights and obligations” or to “any criminal
charge” and does not refer to the “determination” of such. In Bernard Denilauler v. SNC
Couchet Frères [. . .] the European Court of Justice (“the ECJ”) held that provisional
measures given ex parte without hearing the defendant could not be recognised according
to its case-law. This implies that such safeguards should apply also outside the context of
final decisions.154

In this line, Gráinne de Búrca rightly argues that:


There are still concerns, despite the ‘judicial diplomacy’ which has developed between the
CJEU and the European Court of Human Rights, that a disparity between the approaches of
the two courts – to the detriment of human rights protection – may grow if the CJEU
increasingly distances itself from the jurisprudence of the Strasbourg Court and places
emphasis on an autonomous EU approach to the interpretation of the Charter.155

The autonomous approach in interpreting rights—as the practice of the Luxem-


bourg Court demonstrates, and as asserted by Callewaert156—may seriously harm
the ambition of the Charter itself, and may put the Charter beyond the redline of the
Convention’s standards. The need for autonomy of the EU legal order does not
justify the contravention of the Charter’s clear obligation to adhere to the standards
of the Convention. Therefore, Luxembourg’s approach in considering the relation-
ship between the Charter and the Convention seems to have a detrimental effect on
human rights protection which an EU bill of rights is logically expected to pursue.
To make a brief discussion, the European Committee on Social Rights in Interna-
tional Federation for Human Rights (FIDH) v. France case has ruled that:
‘according to the Vienna Declaration of 1993, all human rights are “universal,
indivisible and interdependent and interrelated”. The Committee is therefore mind-
ful of the complex interaction between [. . .] rights.’157 Luxembourg’s approach to
the Charter’s autonomous interpretation of rights—detached from the Convention’s
system of substantive standards—seems to contravene the above baseline; as
Luxembourg’s approach does not recognize the Charter’s rights as universal, but
rather an internal concept of the EU detached from their common understanding
which the Convention system portrays as the minimum standard. The Luxembourg
Court wishes to see the Charter rights having independence from the Convention
rights, and their lively nature, and detached from interaction with peers in the
Convention system. Departing from this standard makes the Charter lack a univer-
sal approach, which the Charter itself appears to endorse rather firmly in the text.

154
ECtHR, Micallef v. Malta, Application no. 17056/06, Grand Chamber, Judgment, Strasbourg,
15 October 2009, para. 32 [emphasis added].
155
de Búrca (2013), p. 172.
156
Callewaert (2014), p. 11; Cf.: To this extent, Gragl rightly points that the EU Court never
considered itself bound to the decisions of the Strasbourg Court from the perspective of interna-
tional law. Some Strasbourg standards which the Luxembourg Court uses and refers to in its
decisions—according to Gragl—are merely considered as binding from the perspective of EU-law
internally. See: Gragl (2013), p. 54.
157
European Committee on Social Rights, International Federation for Human Rights (FIDH)
v. France, Collective Complaint No. 14/2003. Decision on the merits, 8 September 2004, para. 28.
2.5 Chapter’s Summary of Conclusions 65

Callewaert makes a distinct argument in this regard. He argues that the Luxem-
bourg Court’s function with regard to the fundamental rights law protection is not
that of a simple national court. Callewaert states that the scale of Luxembourg
human rights standards saturate all EU Member States’ legal orders.158 This being
the problem, Luxembourg’s divergent approach to the consideration of rights
compared to the Strasbourg Court is rather important, as it implicates the position
of EU Member States while they are simultaneously obliged under Convention law
(which the Luxembourg Court forgets to tackle when it deals with Art. 53 of the
Charter in Opinion 2/13). For this reason, Callewaert rightly notes that one should
not make the mistake of comparing Luxembourg’s divergent approach to that of
national courts with regard to deference to Strasbourg’s standards.159 For this
reason, it is argued that an incoherent arrangement between the Charter and the
Convention was not the aim of the Masters of the Lisbon Treaty. Instead, they
wished to see both documents cohabitating in order to establish one integrated layer
of human rights protection, with Luxembourg and Strasbourg courts living concur-
rently within one layer of minimum protection. Callewaert, for these purposes,
argues that the Lisbon Treaty designers ‘wanted the first to be the bedrock of the
second, and this to be clear.’160 It was—for a neutral outside reader—made clear in
the Lisbon Treaty, but not taken seriously in the practice of the Luxembourg Court.
For this reason, EU accession to the ECHR seems to be the method that would make
this disparity controllable. Callewaert hence argues that ‘what is at stake in EU
accession is to some extent the European conception of human rights, which is
measured by the ability of all Europeans to adhere to the same catalogue of
unequivocal minimum fundamental rights.’161 The latter seems not to have been
the practice of interaction between the Charter and the Convention. EU accession to
the ECHR will likely close this debate in favour of a more regulated relationship
among the two.

2.5 Chapter’s Summary of Conclusions

Rightly argued, the EC/EU had a difficult and stony road to becoming a human
rights law organization. Not only internal problems and pressures, but also inter-
national persuasion to build a credible external image, put the EU Court under
constant pressure to devote more room to human rights law. This resulted not only
in expansion of its jurisdictional human rights development but also development
of substantive human rights principles. That said, the EU Court had an important
objective to reach: on the one hand, to manage its Member States aversion towards

158
Callewaert (2014), p. 11.
159
Callewaert (2014), p. 11.
160
Callewaert (2014), p. 11.
161
Callewaert (2014), p. 12.
66 2 EU Becoming a Human Rights Law Organization: Starting from Nowhere with a. . .

its primacy principle, for which it devised human rights as a means to appease the
national courts. Second, it used this stress on rights as a tool to soften its relations
with international tribunals, and to keep the Strasbourg Court swiftly communi-
cated in its relationship with the EU legal order.
It is therefore concluded that the path followed by the EC/EU to become a
human rights law organization was originally neither intended nor required by the
Masters of the Treaties and the treaties themselves. The development of EC/EU
human rights law was managed through an entirely separate process led by the EU
Court, which was paradoxically translated into a source of energy for the subse-
quent treaty-making process. The EU Court, one may legitimately argue, proved to
be an effective manager throughout the process of using human rights law simul-
taneously for building its own reputation and credibility while halting the EU
Member States’ dissatisfaction with its doctrine of primacy and direct effect. The
role played by the EU Court deserves appreciation, as it is a core factor that proved
to draw the supranational doctrine into a landscape of human rights credibility. The
latter route has been described by Besselink with these words: ‘[f]rom fundamental
rights rejection to full scrutiny’.162 On the other hand, the EU Court swiftly
deferred—although in a rather formalistic language—to the basic international
human rights law instruments, thereby suggesting that it favors (at least formally)
friendly relations with international tribunals with regard to human rights protec-
tion. This further demonstrated that the EU Court trusted itself to be entitled to
ownership of human rights review of acts of EU institutions which furthermore
converted into more external visibility for the EU Court itself. Of this approach was
also the EU Court’s relationship with the Strasbourg Court, the latter being a core
component of the European constitutionalist debate and framework. Not only had
the gradual and careful relations whom the EU Court proved building with Stras-
bourg but also the latter’s interest in that path proved the friendly relationship
between the two.
The Strasbourg Court followed a pragmatic approach towards Luxembourg: as it
had no jurisdictional authority to address the EU, it chose to apply equivalent
protection to rationalize its inert behaviour while also building a conflict-free
relationship with the EU Court. This soon translated into a mode where accession
of the EU to the ECHR would be seen as the only way to institutionalize this
relationship, as criticism over the Strasbourg’s passive behaviour grew steadily.
This said, it is concluded that both courts have tried to manage a calm relationship
between themselves, however they never managed to make the accession process
undesirable and unnecessary. Even the Opinion 2/13 of Luxembourg Court should
not be seen from such an outlook. Therefore, the history of cooperation and
competition between the two courts merely proves that accession is the most
efficient and legally certain answer to the intricate and divisive environment in
which both courts operated in the last decades. That being the most problematic
aspect of this discussion, EU accession to the ECHR has proven to be the answer to

162
Besselink (2013), p. 303.
References 67

closing the jurisdictional—but also, hopefully, substantive human rights law—gaps


that currently exist between the two organizations and regimes of law; although the
‘gaps’ concerned have been kept outside of the two courts’ official debate and
almost faultlessly managed by them.

References

AIRE Centre, AI, ICJ (2013) European Union Accession to the European Convention on Human
Rights: briefing note on the accession agreement. Brussels, 6 September 2013
Arold Lorenz N-L et al (2013) The European human rights culture – a paradox of human rights
protection in Europe? Martinus Nijhoff Publishers, Leiden
Balfour ADJ (2007) Eliminating conflicting interpretations of the European Convention on Human
Rights by the European Court of Justice and the European Court of Human Rights: the PDIQ
system as a preventative solution. Intercult Hum Rights Law Rev 2:183–247
Besselink L, Reestman J-H (2008) The relative autonomy of the EU human rights standard. Eur
Constitut Law Rev 4(2):199–204
Besselink LFM (2013) Should the European Union ratify the European Convention on Human
Rights? Some remarks on the relations between the European Court of Human Rights and the
European Court of Justice. In: Føllesdal A et al (eds) Constituting Europe: the European Court
of Human Rights in a national, European and global context. Cambridge University Press,
Cambridge
Breuer M (2013) No Donum Danaorum! A reply to Daniel Thym’s “A Trojan Horse?”, VerfBlog,
2013/9/16, http://www.verfassungsblog.de/en/no-donum-danaorum-a-reply-to-daniel-thymsa-
trojan-horse/
Callewaert J (2014) The accession of the European Union to the European Convention on Human
Rights. Council of Europe, Strasbourg
Chalmers D, Davies G, Monti G (2010) European Union Law. Text and materials, 2nd edn.
Cambridge University Press, Cambridge
Coppel J, O’neill A (1992) The European Court of Justice: taking rights seriously? Common
Market Law Rev 29(4):669–692
Craig P (2013) EU accession to the ECHR: competence, procedure and substance. Fordham Int
Law J 36(5):1114–1150
Craig P, de Búrca G (2007) EU law: text, cases and materials, 4th edn. Oxford University Press,
Oxford
d’Aspremont J (2007) Abuse of the legal personality of international organisations and the
responsibility of member states. Int Organ Law Rev 4(1):91–119
de Búrca G (2011) The road not taken: the European Union as a global human rights actor. Am J
Int Law 105(4):649–693
de Búrca G (2013) After the EU Charter of Fundamental Rights: the Court of Justice as a human
rights adjudicator? Maastricht J Eur Comp Law 20(2):168–184
De Hert P (2005) Crafting European Integration with a Constitution containing an abundance of
substantive human rights. Are freedom, democracy and pluralism served? In: Van Thiel S, De
Gucht K, Lewis R (eds) Understanding the new European Constitutional Treaty. VUB Press,
University Press, Brussels
De Londras F, Kingston S (2010) Rights, security, and conflicting international obligations:
exploring inter-jurisdictional judicial dialogues in Europe. Am J Comp Law 58:359–414
de Rivery EM, Chassaing E (2013) EU accession to the ECHR: bringing about an aggiornamento
for EU case law? Overview of ECHR and EU case law. e-Competitions 1–16, N 61327
de Vries SA (2013) Editorial: EU and ECHR: conflict or harmony? Utrecht Law Rev 9(1):78–79
68 2 EU Becoming a Human Rights Law Organization: Starting from Nowhere with a. . .

Defeis E (2007–2008) Dual system of human rights: The European Union. ILSA J Int Comp Law
14:1–8
Defeis EF (2000–2001) Human Rights and the European Union: Who decides? Possible conflicts
between the European Court of Justice and the European Court of Human Rights. Dickinson J
Int Law 19(2):301–331
Eckes C (2013) EU accession to the ECHR: between autonomy and adaptation. Mod Law Rev 76
(2):254–285
Franck TM (1988) Legitimacy in the international system. Am J Int Law 82(4):705–759
Franklin C (2010–2011) The legal status of the EU Charter of Fundamental Rights after the Treaty
of Lisbon. Tilburg Law Rev 15:137–162
Gragl P (2011) Accession revisited: will fundamental rights protection trump the European
Union’s Legal Autonomy? In: Benedek W, Benoı̂t-Rohmer F, Karl W, Nowak M (eds)
European Yearbook on Human Rights, 2011. Neuer Wissenschaftlicher Verlag – NWV, Wien
Gragl P (2013) The accession of the European Union to the European Convention on Human
Rights. Hart Publishing, Oxford
Groussot X et al (2011) EU accession to the European Convention on Human Rights: a legal
assessment of the draft accession agreement of 14th October 2011. Fondation Robert
Schuman/European Issues N 218/7 November 2011
Heringa AW, Verhey L (2011) The EU Charter: text and structure. Maastricht J Eur Comp Law 8
(1):11–32
Hoffmeister F (2006) International Decisions: Bosphorus Hava Yollari Turizm Ve Ticaret Anonim
Sirket v. Ireland. App. No. 45036/98. Am J Int Law 100:442–449
Jacobs FG (2007) The sovereignty of law: the European way. Cambridge University Press,
Cambridge
Jacque JP (2011) The accession of the European Union to the European Convention on Human
Rights and fundamental freedoms. Common Market Law Rev 48(4):995–1023
Klabbers J (2009) Treaty conflict and the European Union. Cambridge University Press,
Cambridge
Krüger HC (2002–2003) Reflections concerning accession of the European Communities to the
European Convention on Human Rights. Pa State Int Law Rev 21(1):89–99
Kuiyper PJ, Paasivirta E (2004) Further exploring international responsibility: the European
Community and the ILC’s project on responsibility of international organisations. Int Organ
Law Rev 1:111–138
Ladenburger C (2012) FIDE 2012—Session on ‘Protection of Fundamental Rights post-Lisbon’
Report
Landau EC (2008) A new regime of human rights in the EU? Eur J Law Reform 10(4):557–575
Lavranos N (2008) Regulating competing jurisdictions among international courts and tribunals.
Za€oRV 68:575–621
Lawson R (2005) Human rights: the best is yet to come. Eur Constitut Law Rev 1(1):27–37
Lenaerts K (2003) Interlocking legal orders in the European Union and comparative law. Int Comp
Law Q 52(4):873–906
Lenaerts K, de Smijter E (2001) The Charter and the role of the European Courts. Maastricht J Eur
Comp Law 8(1):90–101
Lock T (2010) EU accession to the ECHR: implications for judicial review in Strasbourg. Eur Law
Rev 35(6):777–798
Lock T (2012) End of an epic? The draft agreement on the EU’s Accession to the ECHR. Yearb
Eur Law 31(1):162–197
Martin FF (2004) Our constitution as federal treaty: a new theory of United States constitutional
construction based on an originalist understanding for addressing a new world. Hastings
Constitut Law Q 31:268–354
Martinico G (2012) Is the European Convention going to be ‘Supreme’? A comparative-
constitutional overview of ECHR and EU law before national courts. Eur J Int Law 23
(2):401–424
References 69

Merrill JG (1993) The development of international law by the European Court of Human Rights.
Manchester University Press, Manchester
Miiller CA (2007) Fundamental rights in multi-level legal systems: recent developments in
European human rights practice. Interdiscip J Hum Rights Law 2(1):33–60
Morano-Foadi S, Andreadakis S (2011) Reflections on the architecture of the EU after the Treaty
of Lisbon: the European judicial approach to fundamental rights. Eur Law J 17(5):595–610
Neuman GL (2008) Import, export, and regional consent in the Inter-American Court of Human
Rights. Eur J Int Law 19(1):101–123
O’Meara N (2011) “A More Secure Europe of Rights?” The European Court of Human Rights, the
Court of Justice of the European Union and EU accession to the ECHR. German Law J 12
(10):1813–1832
Odermatt J (2014) The EU’s accession to the European Convention on Human Rights: an
international law perspective. Leuven Center for Global Governance Studies. Working Paper
No. 136
Olsen JB (2009–2010) Protecting fundamental rights and the evolving roles of the Court of Justice
of the European Union and The European Court of Human Rights; Europeanisation in action at
the supranational level. Edinburgh Student Law Rev 1(3):54–73
Peers S (2006) Bosphorus – European Court of Human Rights. Eur Constitut Law Rev 2
(3):443–455
Phelps J (2006) Comment: reflections on bosphorus and human rights in Europe. Tulane Law Rev
81:251–279
Quirico O (2010) Substantive and procedural issues raised by the accession of the EU to the
ECHR. Italian Yearb Int Law 20:31–53
Ryngaert C (2011) The European Court of Human Rights’ approach to the responsibility of
member states in connection with acts of international organizations. Int Comp Law Q 60
(4):997–1016
Scheuner U (1975) Fundamental rights in European Community Law and in National Constitu-
tional Law. Common Market Law Rev 12:171–191
Schimmelfennig F (2006) Competition and community: constitutional courts, rhetorical action,
and the institutionalization of human rights in the European Union. J Eur Public Policy 13
(8):1247–1264
Sera JM (1996) The case for accession by the European Union to the European Convention for the
protection of human rights. Boston Univ Int Law J 14:151–186
Shelton D (2003) The boundaries of human rights jurisdiction in Europe. Duke J Comp Int Law
13:95–153
Škugor EM (2012) European Union and the European Convention on Human Rights – the
accession of a giant. Pravnik 46(1):105–118
Sohn L (1995) How American international lawyers prepared for the San Francisco Bill of Rights.
Am J Int Law 89(1):540–553
Timmermans C (2001) The constitutionalization of the European Union. Yearb Eur Law 21
(1):1–11
Timmermans C (2014) The accession of the EU to the ECHR and the Charter of Fundamental
Rights: enlarging the field of protection of human rights in Europe. In: Kosta V et al (eds) The
EU accession to the ECHR. Hart Publishing, Oxford
Toth AG (1997) The European Union and human rights: the way forward. Common Market Law
Rev 34(3):491–529
Tulkens F (2013) EU accession to the European Convention on Human Rights. Speech at: National
School of Judiciary and Public Prosecution (KSSIP) – Krakow (Poland), Friday 1 March 2013
Von Bogdandy A (2000) The European Union as a human rights organization? Human rights and
the core of the European Union. Common Market Law Rev 37(6):1307–1338
Von Bogdandy A et al (2012) Reverse Solange – protecting the essence of fundamental rights
against EU Member States. Common Market Law Rev 49(2):489–519
70 2 EU Becoming a Human Rights Law Organization: Starting from Nowhere with a. . .

Weiler J (1981) The community system: the dual character of supranationalism. Yearb Eur Law 1
(1):267–306
Weiler J (1986) Eurocracy and distrust: some questions concerning the role of the European Court
of Justice in the protection of fundamental human rights within the legal order of the European
communities. Wash Law Rev 61:1103–1142
Weiler JHH, Lockhart N (1995a) “Taking rights seriously” seriously: the European Court and its
fundamental rights jurisprudence – part I. Common Market Law Rev 32(1):51–94
Weiler JHH, Lockhart NJS (1995b) “Taking rights seriously” seriously: the European Court and its
fundamental rights jurisprudence – part II. Common Market Law Rev 32(2):579–627
Wetzel JR (2003) Improving fundamental rights protection in the European Union: resolving the
conflict and confusion between the Luxembourg and Strasbourg Courts. Fordham Law Rev
71:2823–2862
White S (2010) The EU’s accession to the Convention on Human Rights: a new era of closer
cooperation between the Council of Europe and the EU? New J Eur Crim Law 1(4):433–446
Williams A (2004) EU human rights policies: a study in Irony. Oxford University Press, Oxford
Chapter 3
EU Law Autonomy: Where Does
the Viewpoint for ‘Competition’
of Luxembourg Start From?

3.1 Introduction

This chapter discusses and examines some core issues relating to EU law autonomy,
with a special focus on its external autonomy benchmarks which will become
especially significant in light of the new relationship between the Luxembourg
and Strasbourg courts post-accession. The chapter provides an examination of the
research question: ‘Which are the landmark legal principles that prescribe EU law
external autonomy, especially with regard to an external tribunal such as the
Strasbourg Court?’ By addressing this question, the chapter construes the
Luxembourg’s external borders of jurisdiction and influence, portraying the
image that it likely will try to see even with regard to the Strasbourg regime of
law post-accession.
This chapter starts first by examining the seminal Costa ENEL case, showing
how EU law autonomy from its Member States grew, and the main benchmarks that
it achieved. The chapter then examines how the Luxembourg Court built upon this
internal autonomy, extending its influence to the external borders of the EU. Such
innovative but also authoritative approach of the Luxembourg Court is
deconstructed with reference to the seminal cases of Opinion 1/91, Mox Plant
and finally the landmark Kadi case. These three seminal cases deconstruct both
the Luxembourg Court’s standards of protecting its jurisdiction and EU law from
external interferences, and procedural and material safeguards that it uses to ensure
that EU law is not interpreted outside its perimeter. The chapter carefully introduces
and examines the Luxembourg Court’s intended purpose for its external autonomy
from a tribunal like the Strasbourg Court. The chapter then discusses the sometimes
very aggressive approaches which Luxembourg uses to retain its full exclusivity
and primacy of EU law, but also the latter’s dominance over any class of global law.
The chapter also portrays the means via which Luxembourg attempts to guard its
external borders when certain segments of EU law become interpreted by external
tribunals. A clear jurisdictional examination of the position of international

© Springer International Publishing Switzerland 2015 71


F. Korenica, The EU Accession to the ECHR, DOI 10.1007/978-3-319-21759-8_3
72 3 EU Law Autonomy: Where Does the Viewpoint for ‘Competition’ of. . .

agreements in the EU legal order is also provided, while aiming to demarcate the
borders which the Luxembourg Court will establish in relation to the Strasbourg
Court even when the EU accedes to the ECHR. Luxembourg’s aggressive insis-
tence as demonstrated in Kadi to maintain the prevalence of EU law even upon UN
law is also verified, something that will be argued as space for Luxembourg to act
even upon Convention system if there is an extraordinary scenario. The chapter
concludes that the Luxembourg Court carefully considers its relationship with
external tribunals like the Strasbourg Court, and draws rather strong lines to ensure
that EU law autonomy is exclusively interpreted within its purview.

3.2 ‘EU Law Autonomy’: What Does It Actually Mean?

The concept of legal autonomy appears, however, not infrequently at the center of
legal scholarship relating to the EU and its relationship with international tribunals.
Thus far, only a few sporadic attempts have been made to legally define this obscure
concept. One may argue that legal autonomy may refer to the condition of being
independent in taking decisions, or, the right of being self-governed independent of
an external body. According to Sellers, autonomy in its very ordinary sense implies
the right of states, families, associations or of individuals to construct the rules
which govern their actions.1
However, the focus of the present inquiry does not aim to describe any of the
abovementioned categories. Conversely, it will explicate the concept of legal
autonomy of an enterprise sitting between a federal state and an international
organization,2 namely the autonomy of the European Union legal order.3 Such
autonomy lies at the heart of the relationship between the EU and ECHR after
accession, and this has carefully been considered in their historical path. In order to
have a clearer picture of the autonomy of the EU legal order one must take a closer
look at its foundations. For instance, Schilling attempts to display a typology of
foundations which form the autonomy of the EU legal order. The first being
‘original autonomy’ which has no source originaire to justify its existence; the
second being ‘derivative autonomy’ which is a substrate of an already existing legal
order and, finally, the third type, dubbed ‘interpretive autonomy’ which is construed
by institutions of a particular legal order.4 Although there are many concepts that

1
Sellers (2007), p. 1.
2
Cf: Gragl in this regard argues that EU is neither a pure international organization nor a
federation. EU being a rather special entity, Gragl argues, it needs be treated through a special
consideration by international law. This tension, in Gragl’s words, will likely produce tensions
with regard to the EU law autonomy when contrasted with the process of accession to the ECHR.
See: Gragl (2013), p. 25.
3
See an excellent contribution on this issue, at: Barents (2004).
4
See on this: Schilling (1996), pp. 389 et seq. See the academic debate on EU legal order
autonomy: Weiler and Haltern (1996).
3.3 Autonomy in the ‘European Way’: Tracing Its Origins and Discussing Its Rationale 73

need be explained in this context, we will start with the notional scrutiny of the core
moments that have prescribed EU law autonomy, both internally—but more impor-
tantly—externally. As a final comparison note, since EU law protects so rigidly its
autonomy, the Convention system—Gragl argues—and its Protocol 14 have no
provision to assure the protection of EU law autonomy.5 This basically shows that
‘autonomy’ in this accession debate is predominantly an EU problem.

3.3 Autonomy in the ‘European Way’: Tracing Its Origins


and Discussing Its Rationale

3.3.1 Costa ENEL: ‘EU Law Autonomy’ Where the Whole


Story Began

At the outset, one must recognize that the EU law autonomy concept finds no
formal underpinning in the founding treaties. Even the Treaty of Lisbon brought
into being as a result of a treaty revision process, just as its predecessors, remains
silent on this issue and contains no express reference for the autonomy of the EU
legal order. Thus, it was the Court of Justice which first elaborated on the autonomy
of EU legal order from its inception in Costa ENEL case.6
The case at hand concerned a claim of an Italian national who opposed the
nationalization of the electricity industry by a state owned company. Owing to this,
Mr. Costa refused to pay electricity bills which amounted to a small monetary sum.
However, the company decided to sue Mr. Costa for failing to pay his dues. The
issue reached the top of the Italian judiciary, which uncertain of the obligations
arising as a result of Italy’s membership to the then-EEC, was unable to give an
appropriate answer. Therefore, as the court of last instance, the Court of Milano was
obliged to refer the question for a preliminary ruling to the Court of Justice.
Giudice conciliatore of Milano in its request for a preliminary ruling asked,
amongst others, whether the national court should give precedence to EU law
provisions or whether national law provisions should take precedence. The reason-
ing employed in the decision, one may argue, came as a consequence of a previous
ruling in which the Court of Justice first ruled that the EU comprised a separate
legal order of international law.7 Accordingly, in Costa ENEL the Court of Justice,
when asked to assess the compatibility of the national legislation with the then-
Treaty of Rome, ruled that the law stemming from the treaty is an independent
source of law, and cannot as such be set aside by national law.8 With this said, one

5
Gragl (2013), p. 23.
6
Court of Justice of EU (ECJ), Costa v ENEL, Case 6/64 [1964] ECR 585; See also: Defeis (2000–
2001), p. 309.
7
Court of Justice of EU, Van Gend en Loos v. the Netherlands, case 26/62 Judgment (CJEC).
8
Court of Justice of EU (ECJ), Costa v ENEL, Case 6/64 [1964] ECR 585, note 6.
74 3 EU Law Autonomy: Where Does the Viewpoint for ‘Competition’ of. . .

may logically criticize the Court of Justice’s reasoning for its activist tendencies.
However, the rationale behind this, one may argue, may have rested on the need to
ensure that the common market project is maintained, and that norms emanating
from national constitutions would not call into question the foundational treaties.
By comparison, if one prefers using the formula employed in Van Gend en Loos,
what the Court of Justice ruled is not beyond the intentions of the founding
members, namely to submit to the provisions of the EEC Treaty in the event of a
conflict with national ones. Ultimately, the ruling in Costa ENEL bears consider-
able importance for it established internal autonomy of EU law vis- a-vis the law of
EU member states. This clearly became the landmark decision that opened way for
external autonomy of EU law later on, something having relevance even in today’s
debate regarding EU accession to the ECHR.

3.3.2 Opinion 1/91: CJEU Says No for Two Courts Under


the Same Roof

Although Costa ENEL remains thus far a reliable weight on the internal autonomy
of EU legal order, the concept as such is rather blurred; thus, clarification on the part
of the Court of Justice was deemed indispensable. In 1991, the debate on the
concept of autonomy of EU law re-emerged. This time, it was an international
agreement to be concluded between the then-Community and European Economic
Area (EEA) that came to the forefront of the legal debate on external autonomy of
EU law. The agreement at hand had foreseen, inter alia, the establishment of an
integrated tribunal comprising of judges from both signatories, with compulsory
jurisdiction to interpret its provisions. When asked for an opinion on the compat-
ibility of the draft agreement with the Community treaty system, the Court of
Justice contended that the proposed tribunal and the jurisdiction allocated to it
under the agreement were likely to affect the autonomy of the community legal
order.9 The reasoning of the Court of Justice this time was rather tenacious. In
concreto, paragraph 21 of the opinion places significance on the importance of the
EEC Treaty, which, although concluded as an international agreement, in the view
of the Court of Justice, was perceived as the constitutional charter of the European
Community.10 Surprisingly enough, however, the Court of Justice spoke for the first
time on the character of the EEC Treaty as having peculiarities of a constitutional
document. Nonetheless, this part of the opinion may be criticized for failing to
further elaborate what this may entail for the European legal order. In the following
paragraphs, the Court of Justice seems to have driven the point home by focusing on
the very practical implications of the agreement. Further, it suggested that, the mere

9
Court of Justice of EU, Opinion 1/91, ECR 191 I-06079, para. 35; See also: van den Berghe
(2010), pp. 148 et seq; Lock (2010), p. 781.
10
Court of Justice of EU, Opinion 1/91, ECR 191 I-06079, note 9, para. 21.
3.3 Autonomy in the ‘European Way’: Tracing Its Origins and Discussing Its Rationale 75

existence of similarities in wording and content of the provisions of the agreement,


on the one side, and on the other, the Community law provisions cannot ensure
homogeneity in the application of those provisions.11 This part of the reasoning
may be convincing in that the Court of Justice robustly insisted on the reconciliation
of its case-law with the proposed agreement. This is most notable with regard to its
milestone constructions including the primacy of provisions of EU law over
conflicting laws of the EU member states, and the direct effect of provisions of
EU law conferring rights on citizens that may be invoked before national courts.
Generally, Opinion 1/91 displays reluctance of the Court of Justice to share
discharge of judicial functions with other international courts or tribunals,12 the
Strasbourg Court being one such potential example. Apparently, it rejects the
existence of any jurisdictional competition with EU courts, especially when
imposed by an external agreement which, in turn, would seriously undermine the
unique judicial architecture imposed by the treaties. This, as the argument goes,
may seem convincing in terms of allowing the Court of Justice to duly perform its
tasks without competing claims over jurisdictional matters. Opinion 1/91 signifies a
marked shift in the line of reasoning employed in Costa ENEL, which primarily
focused on the autonomy of the EU legal order vis- a-vis legal orders of its member
states. This time, the emphasis was placed on the autonomy of the EU legal order
vis-
a-vis the international legal order, or, put simply, the external autonomy of the
EU legal order. This case will have a remarkable place in the debate of EU
accession to the ECHR, as it implies the review that Strasbourg Court may
undertake on parts of the ECHR which are already incorporated into the substance
of EU law.

3.3.3 MOX Plant: When Jurisdictions Collide CJEU Should


Decide

Notwithstanding the soundness of the Court’s reasoning in Opinion 1/91, the battle
for autonomy had to be fought for a longer time.13 In the meantime, yet another
controversial case appeared before the Court of Justice, with the issue of the EU
legal order’s autonomy at the centre of discourse, which needed be immunized in
the face of international law jurisdictions.14 The facts of the present case concern a
conflict which arose between two EU Member States, namely the Republic of

11
Court of Justice of EU, Opinion 1/91, ECR 191 I-06079, note 9, para. 22.
12
See also: Eckes (2013), p. 259; See also: Gragl (2013), p. 83.
13
See also two other important and authoritative cases on the autonomy of EU legal order which
due to lack of space will be omitted from consideration: Court of Justice of EU, Opinion 1/00
ECAA Agreement [2002] ECR I-349, and Court of Justice of EU, Opinion 1/09 (Agreement on the
European and Community Patents Court).
14
See e.g. on Intertanko, at: Mendez (2010), p. 1751.
76 3 EU Law Autonomy: Where Does the Viewpoint for ‘Competition’ of. . .

Ireland and the United Kingdom over a nuclear plant at Sellafield site located on the
coast of the Irish Sea. Allegedly, the United Kingdom failed to conduct an impact
assessment on the possible effects that MOX Plant would have on the environment.
Dissatisfied with the negligence of the United Kingdom’s authorities to address this
issue and concerned about a potential nuclear incident, the Republic of Ireland
instituted proceedings before the International Tribunal for the Law of the Sea
(ITLOS),15 established by the United Nations Convention on the Law of the Sea
(UNCLOS).16 Even though both states were parties to the United Nations Conven-
tion on the Law of the Sea, some aspects covered by the Convention fell within the
scope of EU competences. The latter had joined its Member States to the Conven-
tion in the form of a mixed agreement.17 The issue was aggravated when the
European Commission brought proceedings under what is now Article 258 TFEU
against the Republic of Ireland for allegedly violating former Articles 10 and
292 TEC, respectively.18 The former requested EU member states to take all
appropriate measures at their disposal to ensure the fulfillment of obligations
arising out of treaties; whereas, the latter, urged EU member states not to submit
a dispute concerning the interpretation or application of the treaties to any tribunal
other than that provided for thereunder.
In his opinion to the case, Advocate General Maduro, emphasized the impor-
tance of ‘duty of loyalty’ or duty of cooperation under Article 10 TEC, and the
jurisdictional monopoly of EU courts, implying that when provisions of an agree-
ment fall under the scope of an area covered solely by EU law, the EU Court may
rule upon the interpretation and application of such rules (the ECHR may also be
such an example).19 More concretely, related to the question of whether the
jurisdiction of the Court is exclusive with regard to the provisions of UNCLOS,
AG Maduro acknowledged the non-exclusive nature of EU competencies with
respect to environmental policy, an area where both the EU and its member states
enjoy concurring competences. However, after recalling the ERTA/AETR20 for-
mula, AG Maduro opined that once the EU has exercised such competences and
adopted common rules, Member States are prevented from exercising their powers
affecting those rules.21 Furthermore, in the following paragraphs, AG Maduro
condemned endeavors by the Republic of Ireland for inviting the ITLOS for

15
Court of Justice of EU, Commission v. Ireland (Mox Plant case), Case C-459/03 [2006] ECR
I-4635.
16
United Nations Convention on the Law of the Sea was signed at Montego Bay in Jamaica, on
10 December 1982; See also: Lavranos (2006), pp. 224/5 et seq.
17
See: Council Decision 98/392/EC. OJ 1998 L 179, p. 1.
18
Court of Justice of EU, Mox Plant Case C-459/03 [2006] ECR I-4635, note 15.
19
Opinion of Advocate General Poiares Maduro in Case C-459/03 Commission v Ireland (Mox
Plant), delivered on 18 January 2006.
20
Court of Justice of EU, Commission of the European Communities v Council of the European
Communities, Case 22/70, [1971] ECR 263.
21
Opinion of AG Maduro in Case C-459/03 Commission v Ireland (Mox Plant), delivered on
18 January 2006, note 19, para. 23.
3.3 Autonomy in the ‘European Way’: Tracing Its Origins and Discussing Its Rationale 77

interpretation of the United Kingdom’s obligations under EU law, which according


to Ireland’s submissions coincide with those assumed under UNCLOS. This, as one
may observe, displayed reluctance on the part of Ireland to cooperate with EU
institutions on issues concerning EU law. On this point, Neframi suggests that the
duty of loyal cooperation entails the ‘substantive duty in order to guarantee the
effective implementation of the mixed agreement in its entirety, fulfilled through
the procedural obligation of cooperation to refer to the Court of Justice’.22 Simi-
larly, the judgment of the Court of Justice relied largely on submissions of Advo-
cate General Maduro pointing out, amongst others, that the initiation of proceedings
before a tribunal other than that provided for in the treaties runs the risk of seriously
jeopardizing the autonomy of the EU legal order, especially ‘the jurisdictional order
laid down in the Treaties’,23 by divesting EU courts of their jurisdictional
exclusivity.
In turn, the ruling commenced with the Court of Justice opining that agreements
concluded by the EU form an integral part of the EU legal order and thus ‘shall be
binding on the institutions of the Community and on member States’.24 Arguably,
one may contend that, an agreement concluded based on shared competence, split
between the EU on the one hand, and its Member States on the other, equally forms
part of the EU legal order. Consequently, the jurisdiction of the Court of Justice
extends to the whole provisions of the agreement. Subsequently, the following
paragraphs of the opinion centered on whether there existed EU competence over
provisions of the agreement; and, if so, whether the jurisdiction of the Court of
Justice is exclusive such as to halt Ireland from bringing the case before a tribunal
other than that foreseen under the treaties. The Court of Justice responded to both
questions in the affirmative. As to the first, the Court of Justice recognized that the
EU enjoys external competence in matters of environmental protection which is
shared with its Member States; moreover, a majority of provisions relied on by
Ireland were largely covered by secondary EU rules. Whereas, as to the second, the
Court opined that merely an international agreement, in the case at hand the United
Nations Convention on the Law of the Sea, cannot affect the exclusive jurisdiction
of the Court which extends to the interpretation and application of EU law.25 The
judgment and the reasoning employed thereof is, once again, striking for its failure
to take account of the international setting of the dispute. Biehler argues that the
Court of Justice wrongly ruled in favor of its exclusivity, by unscrupulously
neglecting rules of international law ‘governing conflicting assertion of jurisdic-
tions’.26 Moreover, one may rightly point out that Ireland correctly decided to
initiate proceedings before a specialized tribunal on the law of the sea provided

22
Neframi (2010), p. 335.
23
Court of Justice of EU, Mox Plant Case C-459/03 [2006] ECR I-4635, note 15, para. 154; See
also: Gragl (2013), p. 83.
24
Court of Justice of EU, Mox Plant Case C-459/03 [2006] ECR I-4635, note 15, para. 81.
25
Court of Justice of EU, Mox Plant Case C-459/03 [2006] ECR I-4635, note 15, para. 132.
26
Biehler (2008), p. 328.
78 3 EU Law Autonomy: Where Does the Viewpoint for ‘Competition’ of. . .

for in UNCLOS. Such an argument may be considered logical, having in mind that
even the Vienna Convention on the Law of Treaties provides for the application of
lex specialis and lex posterior rules in the event of conflict of norms. In due course,
Klabbers rightly criticizes the approach followed by the Court of Justice, which in
his words ‘does aspire to build a fence around EU law, therewith running the risk of
placing the EU outside international law’.27 Certainly, these standards should be
considered as playing a core role with regard to the relationship between the
Luxembourg and Strasbourg courts in a post-accession phase.

3.3.4 Kadi: CJEU Policing EU External Borders

Soon after MOX Plant, the concept of autonomy appeared in orbit again in the Kadi
case.28Kadi arose in light of a well-orchestrated campaign on the part of the
international community against terrorism. This time, an EU Regulation was
adopted which copy-pasted certain segments of the UN Security Council Resolu-
tion which listed certain organizations and natural persons as suspected of financing
terrorist activities.29 The UN Security Council Resolution made calls to the inter-
national community for more effective ways of combating international terrorism,
through the so-called ‘smart sanctions’.30 In the case at hand, the applicant was a
natural person who appealed the measure for its compatibility with fundamental
human rights standards as guaranteed by EU law. Yassin Abdullah Kadi, a success-
ful Saudi Arabian businessman, found himself listed in the Regulation concerned.
The applicant invoked three possible grounds to substantiate his claims, namely the
right to property, the right to be heard and the right to an effective legal remedy.
Appearing before the Court of First Instance (now the General Court), the applicant
found no commensurate redress. The General Court openly displayed deference to
the Regulation profoundly rooted in the UN system. Nonetheless, in certain para-
graphs, surprisingly enough, the General Court held that it is empowered to check
the measure for its compliance with jus cogens,31 but it observed that that was not
the case. Quite the contrary, the Court of Justice, in a radical decision, overturned
the General Court’s ruling and upheld the claims advanced by the applicant.32 It

27
Klabbers (2009), p. 148.
28
Court of Justice of EU, Kadi and Al Barakaat International Foundation v Council and Com-
mission, C-402/05 P and C-415/05 P [2008] ECR I-6351; See also: Miiller (2007), p. 36; Cf.: De
Londras and Kingston (2010), p. 406, who compares Kadi to US Medellin v Texas seminal case.
29
Council Regulation (EC) No 881/2002.
30
De Londras and Kingston (2010), p. 366; For sanctions at the EU level see: Koutrakos (2001),
pp. 49–90; Generally on the UN sanctions regime see: Farrall (2007).
31
Court of Justice of EU (CFI), Yassin Abdullah Kadi v Council of the European Union &
Commission of the European Communities, Case T-315/01, 21 September 2005, para. 226; For
an account of jus cogens, see this excellent contribution: Orakhelashvilli (2009).
32
De Londras and Kingston (2010), p. 391/2.
3.3 Autonomy in the ‘European Way’: Tracing Its Origins and Discussing Its Rationale 79

rendered, thus far, in the words of Tridimas and Gutierrez-Fons, probably the most
important judgment on fundamental rights and on the relationship between EU law
and international law.33 Generally, the judgment followed the previous stream of
case-law on the autonomy of EU legal order; yet, this time the language employed
was harsher considering that UN law was at stake. Ironically, the Union is attached
to its treaty duty under Art. 3(5) TEU which asserts that it shall ‘contribute . . . to the
strict observance and the development of international law, including respect for
the principles of the United Nations Charter.’
More specifically, the judgment began by recognizing that the EU should honor
its obligations assumed under international law.34 In the words of AG Maduro, EU
law and international law do not ‘pass by each other like ships in the night’, on the
contrary, the EU always strives to play a constructive role on the international plane.
In addition, after recalling a pronunciation first made in Les Verts, it stressed the
robust commitment towards the rule of law, which implies that neither the EU
institutions nor the EU member states can avoid the review of legality of their acts
for compliance with the EC Treaty.35 Notwithstanding these soft international-law
tones used in the beginning, the language used by the Court of Justice in the
following paragraphs erupted into a declaration that ‘an international agreement
cannot affect the allocation of powers fixed by the Treaties or, consequently, the
autonomy of the Community legal system’.36 This assertion is very hardly recon-
cilable with the predominant attitude amongst scholars who argue that the UN
Charter is the constitution of international community; thus, it is binding on all
subject of international law.37 For instance, Fassbender argues that the UN Charter
provides a ‘visible document’, comprising both fundamental rights and responsibil-
ities for the members of the international community to which they are committed.38
Generally, the reasoning utilized in Kadi and the outcome thereof have been
harshly criticized as sharply dualist, in certain parts nuanced with parochial and
chauvinist tones,39 especially for rejecting the internationalist context of the case.
De Búrca suggests that the Court of Justice should have adopted a soft-
constitutionalist approach according to which norms emanating from different

33
Tridimas and Gutierrez-Fons (2008), p. 661.
34
Court of Justice of EU, Kadi and Al Barakaat International Foundation v Council and Com-
mission, C-402/05 P and C-415/05 P [2008] ECR I-6351, note 28, para. 291. See also to this effect:
Opinion of AG Maduro in Kadi and Al Barakaat International Foundation v Council and Com-
mission, C-402/05 P and C-415/05 P, delivered on 16 January 2008, para. 22 (emphasis added).
35
Court of Justice of EU, Kadi and Al Barakaat International Foundation v Council and Com-
mission, C-402/05 P and C-415/05 P [2008] ECR I-6351, note 28, para. 281.
36
Court of Justice of EU, Kadi and Al Barakaat International Foundation v Council and Com-
mission, C-402/05 P and C-415/05 P [2008] ECR I-6351, note 28, para 282; See also: Lock
(2010), p. 782.
37
On how the EU Court views EU’s responsibility under dualistic lens, see e.g.: Kokott and
Sobotta (2012), p. 1017/18.
38
Fassbender (1998), p. 617.
39
de Búrca (2010), p. 4.
80 3 EU Law Autonomy: Where Does the Viewpoint for ‘Competition’ of. . .

legal systems should strive to converge by mediating this relationship through


dialoguing.40 One such example of dialogue may be the wish that the EU accession
to ECHR strives to set. Coming back to the previous issue, when the UN Security
Council is exercising its authority under the UN Charter, the Court of Justice should
not place hurdles along the way by compromising its objectives simply because
fundamental rights concerns come into play. If the contrary appears, one may
legitimately claim that such development would seriously call into question the
effectiveness of the international security system and the rationale for the existence
of the United National as well. Certainly, not too many human rights lawyers would
defer to the previous statement since it merely disregards the human rights position
of those subject to the Security Council decisions. I agree with the latter point too.
However, an important point of departure should be made here. Aware that the UN
Charter does not provide for membership for entities other than states, the EU as a
regional organization lacks this recognition and cannot be formally bound by the
UN Charter. Nevertheless, one may rightly inquire whether it is possible to find an
alternative way of binding the EU through its Member States. Interestingly, this
point was raised by the General Court in Kadi inferring the identical logic from a
previous ruling in International Fruit Company,41 where it held that by conferring
the powers to the EU, member states tacitly intended to bind the EU by virtue of the
obligations assumed under the UN Charter.42 According to Uerpmann, this could be
achievable through the so-called ‘functional succession’ ( funktionsnachfolge),
implying that certain functions previously exercised by the EU member states
have been taken over by the EU as their functional legal successor.43 This invention
was held to have its roots in the case of the EC succeeding its member states in the
framework of the General Agreement on Trade and Tariffs (GATT). Even though
prima facie this reasoning may seem persuasive, nevertheless, it may be subjected
to critique. In this connection, one may arguably call to mind the widely accepted
principle of res inter alios acta, which holds that arrangements entered into by two
or more parties cannot bind a third party. Moreover, this principle is embedded in
Article 35 of the Vienna Convention on the Law of Treaties.44

40
de Búrca (2010), p. 49.
41
Court of Justice of EU, International Fruit Company NV and others v Produktschap voor
Groenten en Fruit, Joined cases 21-24/72, [1972] ECR 1219.
42
Court of Justice of EU (CFI), Yassin Abdullah Kadi v Council of the European Union &
Commission of the European Communities, Case T-315/01, 21 September 2005, note 31, para.
200; Accession of the EU to the ECHR would therefore further enhance its autonomous nature of
law compared to the current scenario wherein EU is bound by ECHR on basis of the doctrine of
functional succession of its Member States. Cf.: Odermatt (2014), p. 12 (‘The succession doctrine
imagines the EU as a sum of the legal obligations of its Member States, rather than as a legally
separate and autonomous legal order.’).
43
Uerpmann (2003); On functional succession with regard to EU obligations under ECHR, see
also: Odermatt (2014), p. 11.
44
Vienna Convention on the Law of Treaties. Signed at Vienna, on 23 May 1969 (UN Doc.
No. 18232). Entered into force on 27 January 1980. Available at: https://treaties.un.org/doc/
Publication/UNTS/Volume%201155/volume-1155-I-18232-English.pdf.
3.3 Autonomy in the ‘European Way’: Tracing Its Origins and Discussing Its Rationale 81

Fig. 3.1 A figurative view


of EU’s internal and
external autonomy

By way of conclusion, rhetoric concerning the external autonomy of the EU


legal order recently seems to appear more often in the Court of Justice’s reasoning
(Fig. 3.1), with Kadi being the sharpest example that needs to be considered in light
of the relationship between Luxembourg and Strasbourg after accession.45 Weiler
has suggested that Kadi may be illustrated as a Happy Ending ‘only to those for
whom outcomes are more important than process and reasoning’; further asserting
that ‘Kadi looks very much like the European cousin of Medellin.’46 It is not a view
which is supported by the author of this book, however, as human rights objectives
should form the central part of the reasoning that may be promoted by any
legitimate court. On the other hand, the EU Court’s pronunciations such as ‘an
international agreement cannot affect the allocation of powers fixed by the treaties’,
then, ‘EEC Treaty constitutes the constitutional charter of the Community’ are
examples of constitutional semantics currently employed by the Court of Justice,
which may play a role in light of the DAA of the EU to the ECHR (with CJEU
Opinion 2/13 being a very recent, although old-fashioned, example on this topic).
This may, as de Búrca argues, seriously threaten the EU’s role in contributing to

45
Two international-law allergic cases that may be taken as an example are also: Court of Justice
of EU, France v. Commission, Case C-327/91 [1994] ECR I-3641 & CJEU, and, Court of Justice
of EU, Germany v. Council, Case C-122/95 [1998] ECR I-973; One needs to consider carefully
how the EU Court ‘devised’ its competence to annul two international obligations which EU had
already taken in face of third parties.
46
Weiler (2008); On Medellin case, see: the US Supreme Court, Medellin v. Texas, 552 U.S. 491
(2008).
82 3 EU Law Autonomy: Where Does the Viewpoint for ‘Competition’ of. . .

‘effective multilaterism’ under international law.47 However, one may argue that
this is the result of constitutionalization of the EU legal order in the eyes of the
Court of Justice, which Halberstam & Stein name as a sort of ‘local constitutional
resistance’ thereby criticizing the approach of Kadi for lacking respect for interna-
tional law.48 Thus, as one may observe, the EU legal order is gaining ever more
characteristics of a distinctive and self-referential49 order with its easily cognizable
features, something that the accession process may or may wish to change.

3.4 Chapter’s Summary of Conclusions

This chapter presented the fundamentals of EU law’s internal and external auton-
omy, defining not only their conceptual meaning but also their technical operation
inside and outside the EU. The construction of this doctrine has been explained
from both the academic and case-law milieu, therefore demonstrating its legal
nature, application, and relevance from a wider viewpoint. The chapter in general
proved that the autonomy of EU law is a rather harshly controlled and watchfully
devised mechanism to defend the EU law’s territory of command, therefore the
Luxembourg Court and EU law at-large will likely keep this predisposition towards
the Convention system post-accession as well. To digress with this conclusion,
Gragl notes that, in view of many scholars, the Luxembourg Court will not become
subordinated to the Strasbourg Court post-accession, as the latter would not have
any jurisdiction to annul the decisions of the EU Court.50 Although this may be true,
one should disagree with Gragl by adding that the aforementioned criterion may not
serve as the only one to weigh up whether Strasbourg will gain primacy over
Luxembourg. One should also refer to the question of whether, practically and
pragmatically, the Luxembourg Court—as a judicature of an international organi-
zation which is seriously committed to the rule of law in international operations—

47
de Búrca (2009), pp. 4 et seq.
48
Halberstam and Stein (2009), pp. 13 et seq.
49
Although Kadi may be described as a strong self-referential concept developed by the EU Court,
one should also remind that the EU Court itself noted that it is possible to apply a solange II
approach to the UN law. It had ruled specifically in this regard that: ‘It has in addition been
maintained that, having regard to the deference required of the Community institutions vis-a-vis
the institutions of the United Nations, the Court must forgo the exercise of any review of the
lawfulness of the contested regulation in the light of fundamental rights, even if such review were
possible, given that, under the system of sanctions set up by the United Nations, having particular
regard to the re-examination procedure which has recently been significantly improved by various
resolutions of the Security Council, fundamental rights are adequately protected.’ (Court of
Justice of EU, Kadi and Al Barakaat International Foundation v Council and Commission,
C-402/05 P and C-415/05 P [2008] ECR I-6351, note 28, p. 318 [emphasis added]). By noting
this possibility, still, the Court did nothing else than showing a note of courtesy which, practically,
was turned down in the dictum of this decision.
50
Gragl (2013), p. 24.
3.4 Chapter’s Summary of Conclusions 83

would reject to respect and implement a decision of the Strasbourg Court. Saying
the truth, one may seriously doubt that Luxembourg Court would ever pursue that
path. This being the case, disagreeing with Gragl, to make this assessment one
should draw attention to the practical output of the relationship between the two
courts post-accession rather than the formal statements that describe such relation-
ship. This basically means that the EU-law autonomy principles and framework
remain very important benchmarks to be evaluated against this accession process.
Let us know turn to the conclusions on individual cases examined in this chapter.
Costa ENEL—as analyzed above—exemplifies the first step in instituting Com-
munity law as a self-regulating order of international law, thereby establishing a
novel instrument to keep the Member States’ under the control of Community law.
Not only did Costa ENEL build a hierarchical relationship between the Community
law and national legal orders, but also opened the way for the gradual transforma-
tion of the Community regime into a self-governing body of law independent of that
of Member States. Costa ENEL hence established the basis of the internal auton-
omy of EU law, a sort of autonomy which makes EU law take precedence over the
laws of Member States. The internal autonomy expressed in Costa ENEL construed
the first step that made it feasible for the Luxembourg Court to build upon this
foundation further in the future.
Opinion 1/91, on the other hand, went much further in devising a fully-fledged
concept of EU law autonomy, this time from an external perspective. External
autonomy—in this regard—reflects the autonomous approach of the Luxembourg
Court towards external/international courts or tribunals and international law
at-large, showing a hostile attitude towards jurisdiction of international courts or
tribunals which may try to encroach on the territory of EU law. The same logic was
repeated in Mox Plant; the latter demonstrated a far more sophisticated mechanism
to detect and proscribe potential interferences from external international courts or
tribunals in the interpretation and application of EU law. This logic of autonomy
and dualism was finally fortified with the Kadi case, which illustrated the Luxem-
bourg Court’s tendency to reign even with regard to Security Council Resolu-
tions—which many consider jus cogens norms—if they contravene the EU
Treaties’ fundamental principles like human rights. Such rights are considered as
the pinnacle of ius cogens norms by the Court. The external autonomy of EU law
may therefore seriously adjust the position of the EU as a High Contracting Party to
the Convention system, to a degree which may make drastically reduce the impact
of accession compared to that of a state contracting party. To recognize this level of
autonomy to the EU Court—however—there may be the need to transform the level
and method of interface which the Convention system typically aims to build
towards a state party.
To summarize, by providing a more inclusive statement on this issue, Gragl
argues that the notion of autonomy of EU law (‘external autonomy’) comprises of
four characteristics: first, that an international court may not be vested with
jurisdiction to delineate the division of competences between the EU and its
Member States, second, the acts of an international court may not produce any
internal effect on the EU institutions, including the Luxembourg Court, third, an
84 3 EU Law Autonomy: Where Does the Viewpoint for ‘Competition’ of. . .

international agreement should not produce the effect of changing the functions of
an EU institution, and, fourth, an international agreement may not produce the
implicit effect of amending the EU Treaties.51 The chapter therefore concludes that
the principles governing EU law’s internal and external autonomy will be very
important guidelines to analyze and calculate the cooperation and competition
between the Luxembourg and Strasbourg Courts post-accession. These guiding
principles must be contrasted against the core mechanisms of the DAA in order
to detect the potential gaps wherein the interaction between the two courts may be
characterized by aversion and friction. From an EU law perspective, external
autonomy, as established by the three landmark cases above, may see serious
difficulties from a post-accession point of view; however, certain safeguards have
been put into place to keep these impending effects controllable. The autonomy of
EU law therefore ought to be the starting point for examination of the DAA’s
loopholes. The same fundament has been used by the Luxembourg Court in
Opinion 2/13.

References

Barents R (2004) The autonomy of community law. Kluwer Law International, The Hague
Biehler G (2008) Procedures in international law. Springer, Berlin
de Búrca G (2009) The European Court of Justice and the international legal order after Kadi. Jean
Monnet Working Paper 1/09. Available at: http://www.jeanmonnetprogram.org/papers/09/
090101.pdf
de Búrca G (2010) The European Court of Justice and the international legal order after Kadi.
Harvard Int Law Rev 51(1):1–49
De Londras F, Kingston S (2010) Rights, security, and conflicting international obligations:
exploring inter-jurisdictional judicial dialogues in Europe. Am J Comp Law 58:359–414
Defeis EF (2000–2001) Human Rights and the European Union: who decides? Possible conflicts
between the European Court of Justice and the European Court of Human Rights. Dickinson J
Int Law 19(2):301–331
Eckes C (2013) EU accession to the ECHR: between autonomy and adaptation. Mod Law Rev 76
(2):254–285
Farrall JM (2007) United Nations sanctions and the rule of law. Cambridge University Press,
Cambridge
Fassbender B (1998) The United Nations Charter as constitution of the international community.
Columbia J Transnat Law 36:529–619
Gragl P (2013) The accession of the European Union to the European Convention on Human
Rights. Hart Publishing, Oxford
Halberstam D, Stein E (2009) The United Nations, the European Union, and the King of Sweden:
economic sanctions and individual rights in a plural world order. Common Market Law Rev 46
(1):13–72
Klabbers J (2009) Treaty conflict and the European Union. Cambridge University Press,
Cambridge

51
Gragl (2013), p. 84.
References 85

Kokott J, Sobotta C (2012) The Kadi case – constitutional core values and international law –
finding the balance? Eur J Int Law 23(4):1015–1024
Koutrakos P (2001) Trade, foreign policy and defence in EU constitutional law: the legal
regulation of sanctions, exports of dual-use goods and armaments. Hart Publishing, Oxford
Lavranos N (2006) The MOX plant and IJzeren Rijn disputes: which court is the supreme arbiter?
Leiden J Int Law 19(1):223–246
Lock T (2010) EU accession to the ECHR: implications for judicial review in Strasbourg. Eur Law
Rev 35(6):777–798
Mendez M (2010) The enforcement of EU agreements: bolstering the effectiveness of treaty law?
Common Market Law Rev 47(6):1719–1756
Miiller CA (2007) Fundamental rights in multi-level legal systems: recent developments in
European human rights practice. Interdiscip J Hum Rights Law 2(1):33–60
Neframi E (2010) The duty of loyalty: rethinking its scope through its application in the field of EU
external relations. Common Market Law Rev 47(2):323–359
Odermatt J (2014) The EU’s accession to the European Convention on Human Rights: an
international law perspective. Leuven Center for Global Governance Studies. Working Paper
No. 136
Orakhelashvilli A (2009) Peremptory norms in international law. Oxford University Press,
New York
Schilling T (1996) The autonomy of the community legal order: an analysis of possible legal
foundations. Harvard Int Law J 37:389–435
Sellers M (2007) Autonomy in the law. Springer, Dordrecht
Tridimas T, Gutierrez-Fons JA (2008) EU law, international law and economic sanctions against
terrorism: the judiciary in distress? Fordham Int Law J 32(2):660–730
Uerpmann R (2003) International Law as an Element of European Constitutional Law: Interna-
tional Supplementary Constitutions. Jean Monnet Working Paper 9/03. Available at <http://
www.jeanmonnetprogram.org/archive/papers/03/030901-02.pdf>
van den Berghe F (2010) The EU and issues of human rights protection: same solutions to more
acute problems? Eur Law J 16(2):112–157
Weiler J (2008) Editorial: Kadi – Europe’s Medellin? EJIL Talk, Blog of the European Journal of
International Law. Available at: http://www.ejiltalk.org/letters-to-the-editor-respond-to-ejil-
editorials-vol-195/
Weiler JHH, Haltern U (1996) The autonomy of the community legal order through the looking
glass. Harvard Int Law J 37:411–448
Part II
The Draft Accession Agreement of the EU
Accession to the ECHR: An Examination
of the Central Mechanisms in Light of EU
Law Pecularities
Chapter 4
A New Start for the Accession of the EU
to the ECHR

4.1 Introduction

This chapter examines the main steps that provoked the accession process and the
debates that surrounded the expression for the need for accession, including the
core results that the latter will bear on the EU’s external picture both with regard to
the Convention system and international law at-large. Internal changes to the EU
position vis-a-vis Member States resulting from the accession process will be also
examined vigilantly. This chapter therefore addresses the research question: ‘What
are the main changes that the DAA will bring to the EU external image, and what
are the core issues settled in the Draft Accession Agreement to this end?’ The
chapter begins by analyzing how the need for EU accession to the ECHR was
originally articulated, and the rationale for it to be a necessity for the organizational
design of the Union. It then turns to the question of the main changes that the treaty
system will experience with the EU placed in a state-like contracting position
within the ECHR, and how will this affect the external treaty-making landscape
of the EU. The chapter carefully analyzes how EU accession to the ECHR will be of
a distinct nature, as the ECHR has been conventionally a state-based instrument of
international law. Unique arguments are also presented in relation to the changes
that international law will experience with the EU situated as a state-like party to a
historically state-owned sovereign domain of human rights, namely the Convention
system. This is contrasted with the Vienna Convention on the Law of Treaties’
principles, showing how conventional views on treaty-making on human rights may
be changed with the EU becoming a contracting party to the ECHR. All this is
analyzed in the context of a multilayered human rights law in Europe debate, with
the EU cultivating its federal attributes with accession to the ECHR. The chapter
then embarks on the issue of treaty competence of the EU to accede to a human-
rights instrument, previously a states-reserved domain of law, and the main impli-
cations that this process may bring to the EU external image, both vis-a-vis
international law but also Member States legal orders. The chapter subsequently

© Springer International Publishing Switzerland 2015 89


F. Korenica, The EU Accession to the ECHR, DOI 10.1007/978-3-319-21759-8_4
90 4 A New Start for the Accession of the EU to the ECHR

offers a Convention-based outlook on the structural changes that EU accession


therein will generate, upholding the argument that the Convention will not only
become a hybrid, but rather complex treaty system. Afterwards, the chapter
embarks on an examination of the post-accession relationship between the EU,
ECHR and EU Member States, arguing that the position of the EU will be
strengthened post-accession, with the judicial dialogue being shifted from Member
States-Strasbourg to a Luxembourg–Strasbourg orientation.
The chapter then shifts its focus to the normative deconstruction of the model of
accession that the EU will pursue, and the main opportunities that the selected
model will likely present. In addition, the chapter then assesses the DAA, and
cautiously but briefly analyzes each article of this agreement, providing specific
analysis to some institutional parts which will not be further discussed or analyzed
in the remaining parts of the book. The chapter concludes that the DAA will have a
activist impact on the EU’s external image, but will also bring major changes in the
structure and nature of the ECHR system of law. Such changes will better establish
the Luxembourg–Strasbourg relationship as a multilayered human rights hub in
Europe.

4.2 Background on EU Accession to ECHR from a Treaty


and Human Rights Law Perspective

‘For most of its first 30 years, the Convention was largely ignored by just about
everybody, including victims of human rights abuses, lawyers, jurists, politicians
and social scientists.’1 In the present day, the Convention has become so attractive
and necessary even for the political legitimacy of the European Union as well.2 In
the words of Schuman: ‘Europe will not be made all at once or according to a single
plan.’3 The author of this book takes the stance that it is now time for Europe to go
further in building its human rights identity, with EU accession to the ECHR being
a core development in that regard. EU accession to the ECHR has been discussed
and considered for a long time, as these two legal orders have cohabitated and

1
Greer and Williams (2009), p. 464.
2
Accord.: Krüger (2002–2003), p. 94, who argues that accession will improve EU’s political
image as the latter remains the only uncontrolled human rights entity in Europe; See also:
Odermatt (2014a), p. 15.
3
Robert Schuman, Declaration of 9 May 1950.
4.2 Background on EU Accession to ECHR from a Treaty and Human Rights Law. . . 91

competed in one way or another.4 ‘It is not simply a matter of asserting that, in a
general sense, these orders influence and look to each other; it is necessary to
analyse and identify their interaction in a more detailed way so as to test any
perceived tendencies and direction in their jostling together.’5 EU accession to the
ECHR has aimed to reconcile or at least institutionalize the appearance of recon-
ciliation between the two orders.6 One of the most serious historical attempts of
promoting this was the 1979 Commission’s Memorandum, wherein the latter
advocated in favour of EU accession to the ECHR7 in order for the EU to become
an area of human rights not only for purposes of its external image but also for the
actual strengthening of human rights protection.8,9 Although the procedure for EU
accession to the ECHR was judicially blocked by Opinion 2/94,10 this issue did not
cease to be an important policy objective in the Union and the European human
rights debate, as ‘fundamental rights are also becoming a more and more important

4
E.g.: Lavranos (2006b), p. 242; Although there have been huge criticisms on the EU’s capacity to
provide for human rights protection against the fundamental economic freedoms, reference should
be made to Kadi II judgment of the Luxembourg Court, wherein it established that even UN law
when EU implements it without discretion is reviewable from the perspective of ius cogens EU
treaties’ norms—in order to offer fundamental human rights protection—with the prime intention
of preserving its own autonomous authority. This must therefore be read as a strategy via which the
Luxembourg Court ‘drafted fundamental rights to defend and justify that autonomy with a
normative layer absent in the early, more formal reasoning supporting that autonomy: EU law
must be autonomous if it is to be a bulwark for fundamental rights.’ (Cuyvers 2011, pp. 485 et seq).
5
Harding (2000), p. 135.
6
Although the EU Court had determined itself to take into account the ECHR. See e.g.: Court of
Justice of EU, Case C-94/00 Roquette Freres [2002] ECR I-9011 and Court of Justice of EU, Case
C-276/01 Steffensen [2003] ECR I-3735, para. 72; On the academic appeal in favor of accession,
see: van den Berghe (2010), p. 112.
7
E.g.: de Búrca (2011), p. 693, who argues that human rights law forms the EU’s international
identity, and a distinguishing feature of it in general.
8
Memorandum on the accession of the European Communities to the Convention for the Protec-
tion of Human Rights and Fundamental Freedoms.COM (79) 210 final, 2 May 1979. Bulletin of
the European Communities, Supplement 2/79. [EU Commission—COM Document], pp. 11 et seq;
One needs to point that a form of cooperation and understanding between the two regimes has
existed far before accession, but not at an institutionalized level. On the latter, see generally:
Morijn (2006), p. 25; See also: Lock (2010), p. 777; Quirico (2010), p. 31; Contra to the argument
for the need on accession of EU to ECHR, see: Balfour (2007), p. 187, who proposes that the
accession process must rest instead concentrating on the reconciliation of the two courts.
9
Besides EU institutions, accession was also supported by scholars and even some of the judges of
Strasbourg Court. See: Balfour (2007), p. 212.
10
E.g. Peers (2006), p. 443; See also: Gragl (2013), pp. 76 et seq, who argues that Opinion 2/94 does
not consider the issue of EU accession to ECHR from the perspective of EU law autonomy with
special reference especially to Opinion 1/91. Gragl argues that Opinion 2/94 merely concludes that as
the treaties stood at that time accession was not authorized, whereas, if accession takes place, it would
produce significant constitutional changes to the EU legal order. This, according to Gragl, was an act
that drew on the Member States’ governments concerns. To this extent, Gragl points out that the
concerns of the Member States were more about the fact that EU will penetrate into a complex and
external institutional system rather than the protection of its autonomy. Such concern, Gragl argues,
would still be relevant although the treaties do authorize accession in explicit terms.
92 4 A New Start for the Accession of the EU to the ECHR

part of EU “political messianism”’.11 The matter of the EU accession to the


ECHR12 took a more visible dynamic in 2001,13 when a working group of the
Council of Europe—Working Group on the Legal and Technical Issues of a
Possible EU Accession to the European Convention on Human Rights (GT-DH-
EU) acting under the auspices of the Steering Committee for Human Rights
(CDDH)—was asked to study the key issues, implications and the means to surpass
them with a view to the accession concerned. The GT-DH-EU made a thorough and
sound preliminary inquiry, and came up with the key issues that needed be
considered in light of the accession of the EU to the ECHR. Since 2001, the issues
raised by GT-DH-EU14 remain those that became the key components of the
today’s Draft Accession Agreement (henceforth: DAA).
A primary issue that needs be pointed out is that the ECHR was an instrument
opened to states only,15 as opposed to non-state actors. That being the case, the
issue of EU accession to the ECHR becomes more complex. The founding idea of
construing the ECHR had been rationalized merely by the need to provide a
regional-international instrument that embedded an external standard-driven mech-
anism of human rights control towards the Council of Europe member states. The
ECHR, therefore, was designed to offer space to states only, not only in terms of
accession but also in terms of the intended outcome with regard to human rights
protection. With the multi-layered European human rights space being developed,
the ECHR nevertheless became a target of an international organization, the
EU. Such an EU aim may be broadly described as intended to increase its demo-
cratic legitimacy in internal and external affairs.16 Considering the complex setting
of accession which the ECHR had provided for state entities only, there was a
definite need to change the design of the ECHR as regards the parties that could
accede therein in order to offer space for the EU’s accession.

11
Muir (2014), p. 222; See also on the political importance for the EU accession to ECHR, at:
Krüger (2002–2003), p. 94.
12
Sera (1996), p. 153 (‘[. . .]accession is the best way to solve the EU’s above-stated problems
regarding human rights protection [. . .]’); Krüger (2002–2003), p. 92 (‘The best means of achieving
the necessary coherence between the European Convention on Human Rights and Community law
is for the European Communities or the European Union to accede to the former.’).
13
Krüger (2002–2003), p. 97; A Memorandum of Understanding between EU and Council of
Europe in 2007 attached special significance to the accession of EU to ECHR, mentioning it as an
explicit objective of both organizations. See: Memorandum of Understanding Between the
Council of Europe and the European Union, 10 May 2007. Available at: http://cor.europa.eu/en/
about/interinstitutional/Documents/5fe3aa86-d3c2-4ac2-a39c-c2ea21618ffe.pdf.
14
See e.g., also: Krüger (2002–2003), p. 98, on a condensed list of points that became later parts of
the draft accession agreement.
15
Lock (2010), p. 777; See also: Landau (2008), p. 560; Odermatt (2014a), p. 7.
16
E.g.: Sera (1996), p. 180/1; On how the EU needed its democratic claim to become internation-
ally visible, see e.g.: von Bogdandy (2012), pp. 318 et seq; See also: Joris and Vandenberghe
(2008–2009), p. 19 (‘Subjecting the EU to the same degree of external scrutiny by the ECtHR
would also strengthen the credibility of the EU human rights policy.’).
4.2 Background on EU Accession to ECHR from a Treaty and Human Rights Law. . . 93

Before analyzing the models which the working group of the Council of Europe
considered with regard to the mode of accession, it is important to offer an initial
analysis of the context of EU accession to the ECHR in light of international law.
As a starting point, the Vienna Convention on the Law of Treaties, the key
international law instrument that regulates the issue of treaty-making and enforce-
ment, stipulates that only states could be parties to international treaties. The
VCLT17 puts this in the context of states as the only sovereign agents under
international law which could be held internationally liable for their obligations
under international law. It is arguable that under the VCLT, the reason that only
states are able to become party to international treaties is rooted in the fact that only
states could be held liable under international law for their duties undertaken
thereof. This said, a logical explanation of the VCLT for only offering space to
states to become parties to international treaties goes arm-in-arm with the acknowl-
edgment that only internationally sovereign agents could hold responsibility under
an international treaty regime. This certainly becomes a founding concept for
arguing the EU’s place in international law as a ius cosmopoliticum.18
It is therefore important to mention that the EU accession to the ECHR does not
in fact become designed in accordance with VCLT principles, but it goes beyond
the more conventional view that VCLT prescribes on this issue. Although the EU
could have some state-like features,19 its accession to instruments like the ECHR
puts more and more emphasis on its supranational character that embeds federal-
state attributes.20 This said, the EU accession to the ECHR is a rather distinctive
case wherein a regional-international organization becomes a state-like party to an

17
Shelton (2003), p. 125 (ECtHR ‘established that the terms of the Vienna Convention concerning
interpretation are applicable to the ECHR because they enunciate “generally accepted principles of
international law.”’).
18
On the latter, Cf.: von Bogdandy (2012), p. 324; Cf: Wildhaber (2002), p. 161, on the context and
understanding of the role of states in international constitutionalism; See de Búrca’s criticism on
how Kadi changes the role of ECJ into one which tries to threaten the ‘effective multilateralism’ in
international law, at: de Búrca (2009), pp. 4 et seq; One may certainly argue that should this type of
mood be further promoted by CJEU in face of the Strasbourg Court post-accession, it may
seriously threaten the effective cooperation which the Accession Agreement intends to establish
between the two orders.
19
E.g.: Daniel Hannan had argued that in accordance with the 1933 Montevideo Convention on the
Rights and Duties of States, the EU fulfills the criteria of a ‘state’ (Hannan 2010).
20
See e.g. generally: Landau (2008), p. 573; Gragl (2011), p. 69; Quinn (2001), p. 853; Jones
(2012), pp. 5–6; Cf.: Alemanno (2010), pp. 468 et seq, where the author examines how Opinion
1/08 defines EU’s competence to act internationally on basis of its Member States’ international
obligations (Opinion 1/08 making bold the shared competences); Contra to the argument men-
tioned in the main text, see: Kumm (2005), p. 304; Cf.: Sera (1996), p. 184, who argues that
Germany wanted a federal Europe attached to ECHR as a federal attribution element.
94 4 A New Start for the Accession of the EU to the ECHR

international treaty which was designed originally for states only.21 The same
finding has been noted by in the very recent Opinion 2/13 (para. 193). Considered
with a view towards the conformity of this accession with the VCLT, one should
first say that the EU is not explicitly a signatory of the VCLT in the contractual law
sense, adding that it could not be so unless it becomes a member of the United
Nations (as VCLT is an instrument under the auspices of UN law). Although there
is no explicit obligation of the EU towards the VCLT, the adaptation of ECHR to
address the EU peculiarities will very much reflect on its conformity with the
VCLT’s principles.
Before embarking on the more analytical context of the accession features, it is
arguable that the EU accession to the ECHR would provide for a more dynamic
system of international treaty-making, and it could be seen as an exemplary
stepping stone that goes beyond the conventional view of a state-based international
treaty system22 as designed by the VCLT.23 On the one hand, such an accession will
first acknowledge the increasingly visible trend of supranational integration of the
EU,24 which also needs be accommodated under international law mechanisms.
Alternatively, such an accession would draw attention to the multi-layered system
of human rights protection—but also integration—that Europe is building: one that
rests on numerous sources of legitimacy,25 both domestic, regional, and interna-
tional. The latter is a core example of the ever-growing legal pluralism in Europe
(and specifically in the EU)26 in light of the latest tendencies of law in a globalized

21
Although EU has also e.g. become a party to the World Trade Organization. Cf.: Analysis
(1997), p. 237, which argues that based on Opinion 2/94 EU was meant as not being a state and
therefore legally incapable of acceding to ECHR. Cf.: Odermatt (2014a), p. 7, who brings the
example of the resistance that UN Member States echoed when EU tried to get an equal-footing
position with the Member States in the UN General Assembly; some UN Member States resisted
due to the allergy that they have towards increasing stateness attitude of the EU, indicating their
disagreement with the practice where sovereignty becomes a non-exclusive state concept.
22
E.g.: Roberts and Sivakumaran (2012), p. 120; Cf. on the people-centered doctrine of interna-
tional law: Grossman and Bradlow (1993), pp. 23–25; See also: Buxbaum (2005), p. 314.
23
See the insistence of an EU Member State’s constitutional court to keep the state-oriented VCLT
principles on EU Treaties as the only source for the latter’s constitutional validation, at: Polish
Constitutional Court, Polish Membership of the European Union (Accession Treaty), Judgment
K18/04 of 11 May 2005, para. 6 et seq; This certainly shows the allergy that national constitutional
courts of EU Member States will have towards the increasing capacity of EU as a supranational
organization changing the nature of treaty law as well, with constant increasing attributes of a
sovereign agent of international law.
24
See by example, at: Licková (2008), p. 467.
25
Accord: Olsen (2009–2010), p. 56, who suggests that EU accession to ECHR will increase the
bona fide function and legitimacy of the EU Court towards guarding the legality of EU law; Cf.:
Neuman (2008), p. 115 (‘The “humanization” of international law has not proceeded so far as to
make international human rights tribunals self-legitimating on the basis of their direct relationship
with individual human rights.’).
26
See: Barents (2009), p. 422; Von Bogdandy et al. (2012), p. 492 (‘[..]in a Union based on
constitutional pluralism [. . .]’).
4.3 Can EU Be a Master of Treaty in ECHR? 95

world.27 In view of this, the accession concerned would therefore construe a more
modern construction of the treaty regime acknowledging the fact that not only
states could be seen as sovereign agents that could be liable under international law,
but so could international organizations with very deep-seated supranational fea-
tures like the EU as well.28 This could fill in the bigger picture of a real ‘pluralist
European constitutional area’29 with proliferated courts30 and also signify the end
of the solange approach of the Strasbourg Court on the EU.31 This new picture will
arguably not ‘threaten[. . .] but rather confirm and strengthen the concept of a
European multilevel constitution.’32

4.3 Can EU Be a Master of Treaty in ECHR?

Before accession, the EU has mostly penetrated to its chosen international law
fragments via its Member States’ relationships with international law, such as in
Kadi II33 wherein the Luxembourg Court reviewed UN law as an obligation
stemming from its Member States adjacent to EU human rights law.34 Furthermore,
the member states’ constitutional courts have insisted that they are the sole Masters
of Treaties of the EU,35 implicitly excluding the possibility that the EU itself
become a Master of Treaty there where its Member States are simultaneously

27
See generally: Tamanaha (2008), pp. 389 et seq; See also the overlapping jurisdictions between
EU, Council of Europe and Member States courts, at: Sweet (2009), p. 637.
28
Accord.: the trend of transformation of state-centered international law towards a more human-
right-based international law, at: Petersmann (2002), pp. 32–67; See also: Odermatt (2014b), p. 4
(‘EU accession has great legal significance in this regard since it is the first time that an
international organization will formally submit itself to a system of external human rights
review.’).
29
Barents (2009), p. 445; See also: Joris and Vandenberghe (2008–2009), p. 31, referring to the
term Pan-European legal area.
30
On the global trend of proliferation and use of international tribunals like this, see: Lavranos
(2006a), p. 457.
31
The solange formula has been hugely criticized for the lack of substantive human rights
protection as in Bosphorus, at: Peers (2006), p. 455; On the vertical relationship between courts
as a matter of the solange formula, see: Lavranos (2008), p. 612; Cf.: Besson (2009), p. 258; On the
nature of solange between Luxembourg Court and German Constitutional Court, see: Thym
(2013b), p. 398; See also on solange with Strasbourg Court: Papangelopoulou (2006), p. 134;
and Kokott and Sobotta (2012), p. 1018.
32
Pernice and Kanitz (2004), p. 20.
33
Cf.: Mendez (2010), p. 1719, on the criticisms on the EU’s consideration of international law.
See also: Licková (2008), p. 464.
34
Cuyvers (2011), p. 509; See also: van Rossem (2009), p. 219, who legitimately questions how to
tackle international treaties which EU should respect because of its Member States’ international
obligations but which clash with the EU law; See also: Mendez (2010), p. 1720.
35
Bieber (2009), p. 397.
96 4 A New Start for the Accession of the EU to the ECHR

masters therein (not two levels of masters for one treaty). Practically, this rule
pointed to the fact that there may not be concurrently the EU and its Member States
as masters of a treaty, internally or externally. This legal scenario will start
changing tremendously, as the EU will act and rely on its own capacity in the
face of international law obligations (the Convention being a core example).36 It is
worth noting that EU accession to the ECHR would entitle the EU to become a High
Contracting Party,37 wherein it becomes one of the founding institutional parties
that ensure the authorship over and take over the duty to implement the ECHR. A
key issue here is whether the EU could become a High Contracting Party to the
ECHR, or whether it has the legal capacity to join such a treaty (jus tractandi)? It is
important to note that becoming a High Contracting Party to the ECHR sets out a
very important feature for the EU’s external independence, as it quite well portrays
the EU’s international attributes and forms part of its increasingly stateness atti-
tude.38 Both substantively and formally, becoming a High Contracting Party to the
ECHR causes a certain level of burden sharing between the EU and its member
states, but it also causes a certain level of external-sovereignty sharing between
them.39 The former, for sure, would enable the EU to be held accountable under the
ECHR system for its own human rights law and behavior. The latter, most notably,
will transfer another layer of EU member states’ sovereignty over the contracting
authority on the ECHR to the EU. In this regard, the EU will account for a more
independent entity under international law, with special regard to the ECHR,
wherein a part of the EU member states’ almost exclusive international authority
over human rights will be transferred to the EU.40 One must note here that the
Luxembourg Court has already taken significant, informal steps with Schmidberger
and Omega to grasp human-rights law domain from what might originally be

36
See a request on this, at: Douglas-Scott (2006), p. 252.
37
Something that will lead to the growing notion of ‘European international law’. On the latter’s
growth, see: Smith (2007), p. 438; See also generally on these effects: Sera (1996), p. 180.
38
Cf: Dyevre (2013), p. 139; Somewhat a contrary argument on the insistence that the EU is an
international organization may be found at: Dijk (2007), pp. 2–4; Accord: Schuman Declaration
(Paris, 9 May 1950), which shows some of the above-mentioned tendencies even in written.
39
See e.g.: von Bogdandy (2012), p. 322, who shows how the EU itself has a dual concept of
democratic legitimacy. After accession to ECHR, such dual democratic legitimacy becomes even
more complex in terms of human rights protection and the legitimacy deriving therefrom; See a
similar argument at: Licková (2008), p. 489; See also the overall context of how Member States are
becoming limited in space in international law due to these legal situations, at: Neframi (2010),
pp. 323–359.
40
E.g.: The German Federal Constitutional Court insisted that the EU powers may not go so far as
to hamper the constitutional identity of Germany as an EU Member State. If applied against the
issue of human rights domain, one may argue that the German Court would argue through
Gauweiler that accession of the EU to the ECHR may not go beyond the powers set on EU by
the Treaties, as that would simultaneously violate Art. 4(2) TEU. With this in mind, one may not
look at the human rights external identity of EU in face of ECHR as going so far as to violate the
portrait of Member States’ constitutional identity. See: BverfG, 2 BvE 2/08 Gauweiler v Treaty of
Lisbon, Judgment of 30 June 2009, paras. 240 and 339.
4.3 Can EU Be a Master of Treaty in ECHR? 97

considered as belonging to Member States,41 something that is supposed to be


further institutionalized and formalized with the DAA practical impact in this
picture.
This development, however, should be also seen in light of the growing body of
modern international law which tends to regulate the law of treaties between states
and international organizations or international organizations themselves.42 Such a
core development is the, still legally unforced, Vienna Convention on the Law of
Treaties between States and International Organizations or between International
Organizations (henceforth: VCLTSIOIO).43 VCLTSIOIO,44 which is still not rat-
ified by most of the EU Member States, foresees the possibility for international
organizations to become part of treaties and international treaty mechanisms, one
such example being the DAA. Art. 6 VCLTSIOIO states that:
The capacity of an international organization to conclude treaties is governed by the rules of
that organization.

The Official Commentary to VCLTSIOIO asserts that the latter provision


‘reflects the fact that every organization has its own distinctive legal image which
is recognizable, in particular, in the individualized capacity of that organization to
conclude international treaties.’45 The abovementioned clause would thus turn us to
the EU Treaties to examine whether the EU possesses jus tractandi to join ECHR
and correspondingly conclude the DAA. Of course, two fundamental EU Treaties’
provisions explicitly provide for this basis: first, Art. 218 TFEU, which establishes
the authority of the EU to enter into international agreements in the sphere of its
exclusive and shared competences, and, Art. 6 (2) TEU, which establishes the duty
of the EU to accede to the ECHR46 (the principle delegatus non potest delegareis is
not therefore violated). Therefore, it should be argued that the EU has a full legal

41
Sabel and Gerstenberg (2010), p. 511; Cf.: Gragl (2013), p. 56, who argues that both Omega and
Schmidberger prove that the Luxembourg Court has a wealthy experience of giving precedence to
fundamental rights over EU-law economic freedoms, the latter being crucial for the maintenance
of the EU common market.
42
A case which should make the Luxembourg Court accept the primacy of Strasbourg Court as
regards fundamental rights, may be found e.g.: Lavranos (2006b), p. 243.
43
Vienna Convention on the Law of Treaties between States and International Organizations or
between International Organizations. Done at Vienna on 21 March 1986. Not yet in force.
44
Referring to Akehurst, Gragl notes that this instrument—although still not legally effective—
may be considered as forming a declaratory part of customary international law. See: Gragl
(2013), p. 89.
45
Draft Articles On The Law Of Treaties Between States And International Organizations Or
Between International Organizations With Commentaries 1982. Text adopted by the International
Law Commission at its 34th session, in 1982, and submitted to the General Assembly as a part of
the Commission’s report covering the work of that session (at para. 63). The report, which also
contains commentaries on the draft articles, appears in Yearbook of the International Law
Commission, 1982, vol. II, Part Two, p. 24.
46
E.g.: van den Berghe (2010), p. 149.
98 4 A New Start for the Accession of the EU to the ECHR

basis to accede to the ECHR, by correspondingly concluding the DAA, within the
sphere of its competences for which the human rights law can bind the EU.47

4.4 ECHR Becomes a Hybrid and Complex Treaty


System?

In admitting the EU as a party, the ECHR evolves/will evolve as a hybrid and


complex treaty regime, ‘not to mention conceiving of a horror juris.’48 In view of
this, the ECHR thereby accommodates a rather supranational governing system and
allows its judicial review mechanism to rule over it. This transforms it into a multi-
level system of responsibility attribution, wherein either states alone, EU and states
mutually, or the EU alone is held liable for a violation of the ECHR. Being
‘multiple’-level49 would most simply mean facing issues of ‘overlapping jurisdic-
tions, fractured authority, horizontal rather than nested ordering and territorial
fluidity.’50 This said, the accession not only provides for a new party to the
ECHR, but it also introduces an innovative dual level of membership where the
structure of liability for the ECHR violations turns into a multi-level business. With
this new design, it will appear that the ECHR would have to design a system that not
only finds whether there is a violation of a duty but also the level/s to which that
violation should be attributed, clearly instituting a more complex system of naming
the liable entity. This so-called progress could then lead to satisfying AG Jacobs’
appeal in the Wachauf case, where he argued:
[. . .]it appears to me self-evident that when acting in pursuance of powers granted under
Community law, Member States must be subject to the same constraints, in any event in
relation to the principle of respect for human rights, as the Community legislator.51

This standard may certainly become applicable only if the Bosphorus presump-
tion is revoked,52 something which is expected to result from EU accession to the

47
As set in Demirel, where the Court ruled: ‘although it is the duty of the Court to ensure
observance of fundamental rights in the field of Community law, it has no power to examine the
compatibility with the European Convention on Human Rights of national legislation lying outside
the scope of Community law.’ Court of Justice of EU, Demirel V Stadt Schwäbisch Gmünd, Case
12/86, Judgment Of The Court, 30 September 1987, para. 28.
48
Barents (2009), p. 441.
49
Cf.: Dimopoulos (2011), pp. 64 et seq, to compare this with the inter and intra-EU investment
agreements, many of which may be of bilateral nature.
50
Harding (2000), p. 145; See also: Quirico (2010), p. 33.
51
AG Jacobs Opinion para 22, in Court of Justice of EU, Case 5/88, H. Wachauf v. Germany, of
13 July 1989; See also: Court of Justice of EU, Baustahlgewebe GmbH v Commission of the
European Communities, Case C-185/95, [1998] ECR 8417, where the court relies explicitly on
Strasbourg’s standards.
52
E.g.: Albi (2009), p. 63/4; See also: Eckes (2013), p. 280.
4.5 Accession Enables the EU to Enjoy the Benefit of a primus inter pares Position 99

ECHR.53 It is this path that is leading to a new concept of treaty-making from an EU


perspective. Forrest Martin rightly notes that, in this regard, ‘[t]he effective erosion
of national frontiers, the increasing interdependence of states, and the multiplica-
tion of multilateral approaches to global problems demand a construction of the
Constitution that can respond to these new challenges – this New World.’54 EU
accession to the ECHR will obviously lead to the formal demise of the EU external
borders in face of international human rights law, clearly making space for the EU
within the Forrest Martin’s ‘New World’ concept of global law.

4.5 Accession Enables the EU to Enjoy the Benefit of a


primus inter pares Position

As a common argument, one may assert that ‘the Union remains internally and
externally a complex and compound structure in which numerous actors participate
and struggle for visibility.’55 It is important to note that the DAA provides for a
rather distinguished position of the EU in this regard, going beyond the position of a
simple High Contracting Party. One may recognize the EU as enjoying a primus
inter pares56 position in the ECHR system, because it is rather privileged in
comparison with other contracting parties (although Opinion 2/13—considering
its para. 193—does not pragmatically value this opportunity).57 Further, the EU
becomes institutionally in the lead of its Member States when it comes to the issue

53
See generally the concept on this: LB and JHR (2008), p. 203; Craig (2013), p. 1140/1; See also:
Conforti (2010), p. 84; Cf.: On possible scenarios with the doctrine of equivalent protection post
accession, see: Quirico (2010), p. 42; See also on the ‘equivalent protection’: Gragl (2013),
pp. 70 et seq.
54
Martin (2004), p. 270; See a similar argument on pan-European human rights landscape based
on the accession context, at: White (2010), p. 435.
55
Eckes (2012), p. 232.
56
Eckes (2013), p. 265.
57
One additional example brought into another perspective, but relevant for this analysis, is
Gragl’s point. He argues that post-accession national judges will have to consider better the
accession ramifications and issues, which will bring major impact to their work and interaction
with EU law. One such major impact is the fact that national courts will need to tighten their policy
on posing preliminary reference questions to the Luxembourg Court, as, the failure to do so, may
well put that Member State in the terrain of violations of ECHR for which the EU would not need
to maintain any burden. See: Gragl (2013), p. 279.
100 4 A New Start for the Accession of the EU to the ECHR

of liability and ‘who comes first’ question.58 It is important to note that as opposed
to all other contracting parties, the EU changes the ECHR system to adapt it to its
legal peculiarities, such as the exclusive share of competences defined in the EU
Treaties or the exclusive jurisdiction of its court. Such a need for revising the ECHR
with the DAA to adapt to the EU needs makes the latter rather honored in face of all
other contracting parties.59 Although the need to save the EU Treaties’ special
definition of powers as defined by Protocol 8 can be regarded as a very ambitious
EU promotion within the ECHR system, nevertheless one may not put in doubt the
fact that the DAA still allows for rather effective protection of human rights through
the Strasbourg regime of law for applications originating in EU law.
Next, one may observe a more substantive point whereupon the EU becomes
primus inter pares in face of its Member States before the Strasbourg Court. To
start, as the core and fundamental principle of EU Treaties is the EU law primacy/
supremacy over the Member States’ legal orders,60 one needs to question how can
this principle be reflected in the new and rearranged position between the EU and its
Member States within the ECHR system’s landscape. Up until now, the EU was
represented and kept indirectly liable for its violations through its Member States
direct obligations towards the ECHR system. Being based on the Bosphorus
formula,61 the EU Member States, although having transferred responsibilities to
the EU, were not freed from their duty to implement the Convention even though
the violations over the latter were done by the organization to which those Member
States had transferred some of their competencies. Practically, with the old prac-
tice—where the EU is not a contracting party of the ECHR—EU Member States
represented and kept EU law chained to their legal personality before the Stras-
bourg regime of law. The EU law supremacy principle was therefore largely hidden
due to the fact that EU Member States had to respond to the Strasbourg regime of
law for any alleged violation of the EU, in spite of the latter arguing its law is
supreme towards its Member States’ laws. Once the EU accedes to the ECHR, the
real effect of the primus inter pares proposition will come into effect. Since EU law

58
Komárek (2013), p. 421, showing how there is an ongoing empowerment of the Luxembourg
Court’s position against the Member States’ courts, something flowing from the rather rigid
application of the principles of primary and direct effect of EU law; Cf.: A leading remark for
the border of the scale of primus that the EU may capture in this regard is Art. 4(2) TEU, which
states that: ‘The Union shall respect the equality of Member States before the Treaties as well as
their national identities, inherent in their fundamental structures, political and constitutional,
inclusive of regional and local self-government. It shall respect their essential State functions,
including ensuring the territorial integrity of the State, maintaining law and order and safeguarding
national security. In particular, national security remains the sole responsibility of each Member
State.’.
59
Eckes (2013), p. 265.
60
See: Article 351 TFEU, wherein the Member States are requested to remain loyal to the
recognition of EU law supremacy; See generally, also: Court of Justice of EU, Intertanko,
C-308/06, [2008] ECR I-4057; Accord: Reestman (2005), p. 104.
61
Cf.: Lavranos (2008), p. 612 (‘[. . .] the Strasbourg Court ‘applied the Solange method in its
Bosphorus judgment for regulating its horizontal jurisdictional relationship vis-a-vis the ECJ.’).
4.5 Accession Enables the EU to Enjoy the Benefit of a primus inter pares Position 101

will be supreme over its Member States’ legal orders, many of the violations of EU
Member States will originate from their duty to respect obligations under EU law,
respectively the principle of supremacy. That being the case, the EU would need to
represent its supreme law62 before the Strasbourg Court with its Member States
standing as mere agents of implementation of such law.63 This direct attribution of
EU supreme law, wherein many violating acts and omissions of the Member States
originate, brings the EU to the position to stand as the primus party that can remedy
violations and ensure the implementation of ECHR rights. The Member States
would be bound to that new setting as a duty to sincere cooperation under the EU
Treaties,64 therefore having little room to escape.65 When this primus position
before the Strasbourg regime of law is combined with the principle of supremacy
of EU law over its Member States law,66 one can argue that EU accession to the
ECHR further fortifies not only the external personality of the EU in international
law but also takes over to correspond to the Strasbourg regime of law on its own

62
Although one needs to note that the concept of supremacy of EU law is being criticized for its
narrowness. It is now being suggested that the EU-law picture is rather a legal pluralism system, at:
Gerards (2011), p. 80/1.
63
E.g.: Alter (1998), pp. 135 et seq, questioning whether the Member States may regain the
authority over the Luxembourg Court. Compare this to the end result of the EU accession to
ECHR, and the international law role that the Luxembourg Court will attain in face of the EU
Member States’ courts.
64
See e.g. on the duty of loyal cooperation in this context: Eckes (2013), p. 272.
65
Having estimated this, the Czech Constitutional Court argued that it ‘must emphasize that
Article 216 [article on the capacity of EU to enter into international treaties], because of its
vagueness, is on the borderline of compatibility with the requirements for normative expression of
a legal text that arise from the principles of a democratic, law-based state. The Constitutional Court
[. . .asserted. . .] that this transfer must be delimited, recognizable, and sufficiently definite. It is
precisely the ‘definiteness’ of a transfer of powers to an international organization that is quite
problematic in Article 216 TFEU’. See: Czech Constitutional Court, US 19/08 Treaty of Lisbon,
Judgment of 26 November 2008, para. 186; This estimated vagueness noted by the Czech
Constitutional Court may have led to a sort of growth of EU’s international personality in a way
which may have not been permitted initially by the Masters of its Treaties. The DAA may be one
such example.
66
Cf: Liisberg (2002), pp. 1172 et seq, to see why some argue—as opposed to the argument of the
author—that more human rights protection within and for the EU could result in threatening the
EU law supremacy; To note, the German jurisprudence, on the other hand, insists that ‘supremacy
of EU law is therefore limited by the [national] constitutional authorization.’ Payandeh (2011),
p. 12; Cf.: Neframi (2010), pp. 323–359, on how Member States of EU are becoming limited in
their external relations as a result of the growing EU position in international law.
102 4 A New Start for the Accession of the EU to the ECHR

right.67 This would evidence the ongoing ‘struggle for external visibility’ that exists
between the EU and its Member States.68 This practice will certainly lead to an
upcoming situation wherein instead of the European human rights law being
developed by the Strasbourg Court and national courts as was the practice up to
now,69 a new practice would develop wherein the main institutions that can
organize this dignified dialogue and development become the Luxembourg and
Strasbourg courts.70 Unfortunately, the Luxembourg Court’s recent Opinion 2/13
fails to take account of this ‘big fact’—either erroneously or intentionally—that
goes to the benefit of EU. Whether or not this will lead to a better fortified ‘judicial
colonialism’71 remains to be seen.72
DAA may therefore logically amount to a change in the relationship between the
Luxembourg Court and national courts in face of the ECHR system, something that
may go beyond the borders of the Luxembourg Court’s arguments established in

67
Balfour (2005), p. 22; Contra: Bieber (2009), p. 392/3, wherein it is argued that it is rather hard
for the Luxemburg Court to use the principle of supremacy of EU law as a means to impair the
sovereign constitutional character of the Member States’ legal orders. This argument does not,
however, as such contradict this book’s point, as it merely connotes to a situation without the
external picture in place such as the Convention system; Cf. Weiß (2011), p. 65, arguing that with
the Lisbon Treaty and the new position of the Convention in the EU law, even member states’ legal
orders will see their positions subsumed more to the Convention; See also Art. 47 TEU, as one of
the core innovations and developments of the Lisbon Treaty in this regard.
68
Eckes (2012), p. 231.
69
A good example on this would be the assertion of the Czech Constitutional Court’s ‘argument’
on the renowned Lisbon case, where it argued that transformations which advance the EU legal
position in face of the Member States should be read in this language: ‘[The] transfer of certain
state competences that arises from the free will of the sovereign and will continue to be exercised
with the sovereign’s participation in a manner that is agreed upon in advance and is reviewable, is
not ex definitionem a conceptual weakening of the sovereignty of a state, but, on the contrary, it can
lead to its strengthening within the joint actions of an integrated whole.’ Lisbon Decision of the
Czech Constitutional Court, quoted from: Břı́za (2009), p. 149; Such argument seems rather
formal and intended to calm the troubled waters from the EU’s constitutional expansion, rather
than a legally substantive point.
70
Eckes (2013), p. 284; Heifer and Slaughter (1997), pp. 273 et seq; Although this may not arrive
at a point wherein national courts will be seriously dismissed from playing a crucial role; Cf.: Alter
(1998), p. 121, who argues that there is no doubt that the Luxembourg Court is an influential
international court, something that is thought to be beyond a normal international tribunal; Cf.
Olsen (2009–2010), p. 56, who argues that a jurisdictional conflict between the two courts over
their primacy should not be underestimated.
71
Cartabia (2009), p. 17.
72
On the increased primacy that Strasbourg Court is supposed to gain over Luxembourg Court in
the human rights law domain, see e.g.: Schimmelfennig (2006), p. 1248; See also a note on
pre-accession communication between the two courts: Martı́n and De Nanclares (2013), p. 2/3
(‘Until now this has been based not so much on specific legal texts but rather a sort of legal
dialogue’ [between the two courts]); Cf.: On the contrary, one must also raise the point that
accession may perhaps produce the ‘sandwich effect’, as Luxembourg Court will become con-
trolled from now on two sides: one, Member States constitutional courts through their solange
formula, and two, Strasbourg Court through its external control mechanism. See on the latter:
Martı́n and De Nanclares (2013), p. 16.
4.6 The Accession Model and Possible Implications 103

Opinion 1/09.73 Such a change may not be of the same nature as was in the case of
the Unified Patent Litigation System, however it may roughly produce the outcome
that national courts become weakened towards the Luxembourg Court,74 as the
latter takes the prerogative of becoming primus before the ECHR system. This
would, on the other hand, further promote the argument that the Strasbourg Court
‘has effectively become the constitutional court for greater Europe.’75 To make the
final point, Eeckhout argues that it is always complicated to understand in detail and
portray the cohabitation between the EU and its Member States in the international
scene.76 EU accession to the ECHR will allow even less leeway for such details.

4.6 The Accession Model and Possible Implications

In general, becoming a party to the ECHR was a relatively easy step for Council of
Europe member states: each of them had to sign and ratify the ECHR as such,
automatically becoming a party thereof. This issue is a bit more complicated in the
case of the EU. As the EU is not a state party, the ECHR as such would need certain
adaptations in terms of treaty law to make space for the EU and its peculiarities. As
an example, this new rearrangement will make ‘lawyers no longer to approach the
object of their analysis in normative terms of unity of law and its inherent require-
ment of an exclusive relationship between law and territory as well as its implicit
state-centred thinking.’77 This said, accommodating the EU78 within the ECHR
may be accomplished by addressing two issues: first, amending the ECHR and
allowing for it to offer eligibility to the EU and its peculiarities, and second,
acceding the EU to the ECHR via an international law procedure. CDDH had
considered the main legal repercussions and possible proposals since 2002,79

73
Court of Justice of EU, Opinion 1/09 (Creation of a unified patent litigation system—European
and Community Patents Court—Compatibility of the draft agreement with the Treaties), 8 March
2011, para. 60 et seq.
74
Cf.: BverfGE, 2 BvE 2/08 Gauweiler v Treaty of Lisbon, Judgment of 30 June 2009, where the
German Federal Constitutional Court recalled quite strongly that constitutional sovereignty
remains with the Member States and not with the EU. It seemed quite allergic to accept that EU
legal order is becoming constitutionally sovereign. If this approach will be followed, the primus
position of CJEU towards national tribunals proposed here may seriously become an illusion.
75
Greer and Williams (2009), p. 465.
76
Eeckhout (2004), p. 191/2.
77
Barents (2009), p. 441.
78
Cf.: Licková (2008), p. 463/4, who argues that the EU ‘escapes the traditional categories of
constitutional and international law.’ One need to compare this to the accession procedure and its
effects on the EU law.
79
See: Study Of Technical And Legal Issues Of A Possible EC/EU Accession To The European
Convention On Human Rights, Report adopted by the Steering Committee for Human Rights
(CDDH) at its 53rd meeting (25–28 June 2002). DG-II(2002)006. [CDDH(2002)010 Addendum 2].
104 4 A New Start for the Accession of the EU to the ECHR

many of which have been heavily revised in the meantime to produce the final Draft
Accession Agreement (referred to as the DAA).80
The GT-DH-EU had well thought-out the possibilities to accommodate the EU
within the ECHR. It had found that two possibilities exist to address the
abovementioned issue: first, that an amending protocol be attached to the ECHR,
adding then to this a further accession procedure for the EU, or, second, that an
accession treaty be adopted between the EU and ECHR High Contracting Parties.81
Both options envisaged by GT-DH-EU seemed reasonable, although the latter was
preferred more than the former.
The initial option would have necessitated two separate procedures that would
accommodate the EU within the ECHR. First, an amending protocol would need to
be signed and ratified by all ECHR High Contracting Parties, which would make the
necessary adaptations to the treaty system to adapt it to the EU peculiarities and
offer space for EU accession therein. The first procedure would consume a rather
long period of time. And second, an accession procedure whereby the EU would
consent to becoming a party to the ECHR, wherein EU member states would also
need to vote via their parliaments for such a decision. This certainly was a two-stage
procedure that would have consumed much more time and efforts that the second
option. Gragl takes the example of the ratification process for Protocol 14 ECHR, in
which case certain political obstacles by contracting parties had seriously
undermined the efficacy of the process. The latter, according to Gragl, demonstrates
that if this procedure would have been followed for the DAA as well it would have
practically elongated the accession procedure.82
The second option instead was simpler and more efficient. It followed the
common logic and normative procedure via which states accede to the EU. The
option for an accession treaty whereby the EU would accede to the ECHR seemed
therefore more reasonable, effective, and less time-consuming. The accession treaty
would consequently be a one-stage procedure, whereby both the ECHR High
Contracting Parties and the EU would approve a treaty which simultaneously
amends the ECHR and accedes EU to it (‘two in one’ package). The accession
treaty would therefore need to be signed and ratified by both the ECHR High
Contracting Parties, the EU relevant institutions and EU member states (all of
which are High Contracting Parties to the ECHR; they act only once by consenting
their both decisions). With the second option, goes the argument, the procedure for
EU accession to the ECHR would seem simpler, more efficient, and also faster.
According to Gragl, in contrast to an amending protocol which would ‘cease to

80
See also: Lock (2012), p. 162.
81
Doc. CDDH(2002)010 Addendum 2, see note 79; See also: Gragl (2013), p. 92.
82
Gragl (2013), p. 92.
4.6 The Accession Model and Possible Implications 105

exist’ after ratified, the accession treaty will continue to have effect and stand on its
own right as a legal instrument post-accession in addition to the Convention and the
revisions which it will produce to the latter.83 Relating to the second option,84 it is
worth noting that via Protocol 14, the ECHR had in principle already offered space
for EU accession.85 To digress, if Protocol 14 would have not been adopted, the
Luxembourg Court would have used the standard of Opinion 2/91 to rule that it
would not be compatible with EU Treaties if the EU concludes international
agreements (such as the ECHR before Protocol 14 entered into force) which are
limited to states only.86 The latter standard may not apply anymore, as Protocol
14 explicitly offers room to the EU (accession) as an international organization.
Opting for the second choice, the GT-DH-EU also proposed numerous issues
that needed be considered in the accession treaty. It is important to note that the
accession treaty as such would constitute merely a legal relationship that lays down
the route for EU accession to the ECHR. However, it would not entail anything with
regard to EU accession to the Council of Europe as a whole, as accession to the
Council of Europe would require the ratification of its Statute.87 This said, although
the ECHR is an instrument of the Council of Europe, EU accession to the ECHR is
limited into the latter only, establishing no further relationship between the EU and
the Council of Europe.88 In this context, it is arguable that EU accession to the
ECHR has a rather limited effect in terms of a broader integration within the
Council of Europe framework, thereby confirming the argument that such accession
has simply a focused human rights law target.
Digging further into the issue of the accession treaty, one should note that the
final outcome of the latter should reflect the EU peculiarities and features. Gragl
rightly argues in this regard that the tension on EU accession to ECHR is not so
much about substantive fundamental rights protection but rather about the manage-
ment of the relationship between the two legal orders.89 Although it may seem
simple, adapting the ECHR with regard to EU peculiarities would involve changing
the mechanisms and/or normative principles wherein doubt would exist over the

83
Gragl (2013), p. 93.
84
This second option, according to Gragl, would better enable the agreement to be negotiated
more flexibly to allow a more genuine accommodation of EU-law peculiarities in it. See: Gragl
(2013), p. 111.
85
Landau (2008), p. 572; See also: Joris and Vandenberghe (2008–2009), p. 36; van Dooren
(2009), p. 50; Odermatt (2014a), p. 6.
86
Court of Justice of EU, Opinion 2/91 [1993] ECR I-1061, paras. 4, 5 and 37.
87
Statute of the Council of Europe, London, 5.5.1949. Available at: http://conventions.coe.int/
Treaty/en/Treaties/Html/001.htm.
88
Although it is generally argued that the EU accession to the ECHR will itself improve the
communication and cooperation between the EU and the Council of Europe as a whole. See e.g.:
van Dooren (2009), p. 51/2.
89
Gragl (2013), p. 80.
106 4 A New Start for the Accession of the EU to the ECHR

possibility that the ECHR encroach upon the autonomy of EU law.90 Therefore,
preserving the autonomy of the EU legal order91 would be the key intention of the
adaptations that the accession treaty ought to introduce to the ECHR—such inten-
tion having been cautiously addressed by negotiators from both sides. Although this
seems rather difficult, the accession treaty would need to ensure that the ECHR
system conform with the key EU features, which certainly suggests the necessity of
amending the ECHR which could entail changes within the basic meanings of the
latter. In ensuring that the ECHR will not infringe upon the EU legal order’s
autonomy—which negotiators addressed carefully—the accession agreement
strengthens the concept of an autonomous EU legal order rather than a legally
certain human rights control over the latter.92 Demonstrating the negotiating cli-
mate, Odermatt rightly notes that ‘EU Member States were sitting on both sides of
the table in a sense. This meant that the EU was able to attain concessions that it
would not have otherwise attained had it been negotiating with another interna-
tional organization with a more global membership.’93
Within the context of accession negotiations, the recommendation of the GT-
DH-EU as regards the modality of accession became later the official standard of
the negotiations between the EU and the Council of Europe. As a starting point, on
26 May 2010, the Council of Europe Committee of Ministers authorized the
Steering Committee for Human Rights (CDDH) to start the negotiations with the
European Union as regards the accession of the latter to the ECHR.94 On its part, the
EU Justice Ministers, on 04 June 2010, authorized the European Commission to
negotiate with the Council of Europe for the accession procedure.95 An informal
group representing both parties construed the Draft Accession Treaty for EU

90
On the obligations of the EU Member States to preserve the autonomy of the EU Court when
they sign treaties outside the EU, see: Peers (2013), p. 69, compare this to the preservation of EU
Court autonomy by the DAA mechanisms and safeguards.
91
On the EU legal order, see: Barents (2009), p. 426; See also: Lavranos (2006a), p. 464; On the
question whether there is an EU legal order, or instead, an EU international legal order, see: Smith
(2007), p. 444.
92
Cf.: Balfour (2007), p. 186, who sees accession as being made to eliminate the problems of
human rights and not capturing the huge adaptation which this process will produce on ECHR.
93
Odermatt (2014a), p. 7.
94
Decision No. CM/882/2605201, Adopted by the Ministers’ Deputies at their 1085th meeting
(26 May 2010). See also: Steering Committee For Human Rights (CDDH), Ad hoc terms of
reference concerning accession of the EU to the Convention given to the CDDH by the Ministers’
Deputies during their 1085th meeting (26 May 2010). Strasbourg, 3 June 2010 CDDH(2010)008.
Available at: http://www.coe.int/t/dghl/standardsetting/hrpolicy/accession/Working_documents/
ToR_CDDH-UE_en.pdf; See also: Martı́n and De Nanclares (2013), p. 4; Craig (2013), p. 1117;
Odermatt (2014a), p. 6.
95
Press Release, ‘European Commission and Council of Europe kick off joint talks on EU’s
accession to the Convention on Human Rights’. European Commission, Strasbourg, 7 July 2010.
IP/10/906; See also: White (2010), p. 437/8; Callewaert (2014), p. 48.
4.6 The Accession Model and Possible Implications 107

accession to the ECHR and submitted it to the CDDH for adoption—such accession
negotiations, according to De Vries, were in that time ‘in full swing’.96 Following
that, the Committee of Ministers directed the CDDH, on 13 June 2012, to seek from
the EU to consent the DAA and to proceed with the accession voting.97 Gathered in
a mutual format of ‘47+1’ (Council of Europe Members + EU Commission), the
CDDH-EU Negotiating Group—after five negotiating meetings—reached a final
agreement on 5 April 2013,98 known officially as the ‘Draft revised agreement on
the accession of the European Union to the Convention for the Protection of Human
Rights and Fundamental Freedoms’. Such a ‘draft agreement might mean the end to
a discussion of almost epic proportions.’99 The 2013 Draft Accession Agreement
agreed by both parties100 will therefore lead to the EU becoming the 48th
contracting party to the Convention.101 All this process was appreciated as being
transparent.102 Evidently, the Luxembourg Court was asked to give an opinion103
on the DAA in accordance with Art. 218 (11) TFEU,104 by considering the legality
and conformance of the DAA with the Treaties (see infra Chap. 11 for a thorough
examination of this opinion).105 On the latter procedure, the Luxembourg Court
has ruled in Opinion 1/94 that the opinion may be requested ‘at any time before
the [. . .EU’s. . .] consent to be bound by the agreement is finally expressed.

96
de Vries (2013), p. 79.
97
Report to the Committee of Ministers on the elaboration of legal instruments for the accession of
the European Union to the European Convention on Human Rights. Steering Committee for
Human Rights (CDDH), (CM/Del/Dec(2011)1126/4.1, CM(2011)149). Committee of Ministers
of Council of Europe, 1145th meeting—13 June 2012. Available at: http://www.coe.int/t/dghl/
standardsetting/hrpolicy/Accession/Accession_documents/2012_CM_Decision_accession_en.
pdf.
98
Callewaert (2014), p. 49.
99
Lock (2012), p. 162 [emphasis added].
100
Cf.: Council of Bars and Law Societies of Europe (CCBE) Statement on the European Union
accession to the European Convention on Human Rights, July 2013, p. 1, which welcomes the final
draft agreement.
101
Callewaert (2014), p. 7.
102
O’Meara (2011), p. 1817/8; Contra.:Jones (2012), p. 8, who complains that the negotiating
mandate for the Commission has been obsessively kept closed for the public.
103
Regarding the Court’s outcome of opinion, it is worth reading a core component of Opinion
1/75, at: Court of Justice of EU, Opinion 1/75, [1975] ECR 1355, p. 1360/1 (‘[. . .]a possible
decision of the Court to the effect that such an agreement is, either by reason of its content or of the
procedure adopted for its conclusion, incompatible with the provisions of the Treaty could not fail
to provoke, not only in a Community context but also in that of international relations, serious
difficulties and might give rise to adverse consequences for all interested parties, including third
countries.’).
104
Eckes (2013), p. 264; Raba (2013), p. 571.
105
On when an opinion may be asked from the Court, see also: Court of Justice of EU, Opinion
2/94 para. 16; See also: Court of Justice of EU, Opinion 1/78 International Agreement on Natural
Rubber [1979] ECR 2871, para 35; See also: Analysis (1997), p. 237.
108 4 A New Start for the Accession of the EU to the ECHR

Unless and until that consent is given, the agreement remains an envisaged agree-
ment’.106 DAA therefore remains currently with the status of an ‘envisaged agree-
ment’ from the EU law perspective. If all this process ends successfully, then the
new European human rights law—with both European courts ‘living next to each
other’107—will produce the taste of ‘Europeanization in Action’.108

4.7 Outline of the Draft Accession Agreement of the EU


to the ECHR: What Substantial Issues Does It
Address?

The Draft Accession Agreement contains 12 articles, 5 annexes to the agreement,


one of which is the Explanatory Report.109 Most of the articles of DAA directly
revise the Convention, or at least provide more explicit—now legally binding—
meanings for some of the core Convention provisions. They also establish new
mechanisms which are introduced with new provisions that will become part of the
Convention once DAA comes into force. Four of the five annexes tackle the issue of
declarations of the EU on the use of the co-respondent mechanism, the procedure of
third party intervention when the addressee is a non-EU member state, and the
supervision of friendly settlements by the Committee of Ministers of the Council of
Europe. Annex 5—being the most important one—is named as the Explanatory
Report to the DAA of the EU to the ECHR. It is therefore very important to examine
here the nature of the Explanatory Report. It is basically argued that the DAA
package was construed in a rather general language in order to offer space for
protection to the EU law autonomy.110
To note, the Explanatory Report provides additional in-depth clarifications of
DAA, showing ways to interpret it from both the perspective of the Strasbourg
Court and Luxembourg Court, and better specifies the framework whereupon the
EU and the Council of Europe conclude the DAA. Most of the explanations in the
Explanatory Report not only show the intention for which most of the provisions of
DAA were established—something that will later be used for the teleological
interpretations of both Courts – but also gives direction on how to systemically
interpret the practical adjudication of cases which involve EU-law genesis before

106
Court of Justice of EU, Opinion 1/94,[1994] ECR I-5267, para 12 [emphasis added].
107
Lavranos (2008), p. 606.
108
On this terminology, see: Olsen (2009–2010), pp. 58 et seq.
109
Cf.: Lock (2012), p. 164, who argues that the rules in the DAA were purposely drafted very
generally and with an open-ended language.
110
On the latter, see: Raba (2013), p. 563.
4.7 Outline of the Draft Accession Agreement of the EU to the ECHR: What. . . 109

the Strasbourg Court.111 On the other hand, the Explanatory Report submits argu-
ments as to why some of the mechanisms are meant as judicially effective and help
better the administration of justice. These clarified concepts push forward a more
thorough argumentative position which also the Luxembourg Court needs to take to
reposition itself before the Strasbourg Court. As the Explanatory Report to DAA
explains in a more detailed version the meaning and the means of using many of the
newly introduced mechanisms of the DAA, one should question whether the
Explanatory Report is legally binding, whether it forms part of DAA itself, and
whether it can be interpreted as supplementing the DAA. To offer an answer to this,
the Explanatory Report itself establishes that:
The present explanatory report is part of a package of instruments prepared by the
negotiating group which all form part of the context underlying the accession of the EU
to the Convention. Explanatory reports have been used by the European Court of Human
Rights as a means of interpretation.112

In addition, this was also supported by the document through which the final
report to the CDDH was officially submitted to the relevant institutions, which
mentions the fact that113:
The participants agreed to recommend to the Committee of Ministers that, when taking note
of the “package” of instruments, it also stress the importance of all the instruments
elaborated, including the explanatory report, which all form part of the context underlying
the accession of the EU to the Convention.114

Therefore, it is argued that the Explanatory Report is meant to be the foundation


upon which DAA relies, having a legal nature in terms of the interpretation of DAA
from all contracting parties. Such legal nature of the Explanatory Report does not
reflect its direct applicability or effect individuals rely upon, but rather provides an
institutional framework which is obligatory for the contracting parties. Such obli-
gation can go as far as DAA itself possibly will, as the contracting parties of the
latter have ‘agreed’ and ‘ratified’ DAA simply on the foundation of its annexes, the
most important of which is the Explanatory Report. Hence, even though the
Explanatory Report does not form part of DAA within the text, it forms part of it
in the legal and factual sense. To note here, DAA along with its five annexes is
meant to serve as a package of the agreement as such, making the overall legal box
of EU accession to the ECHR legally broader than the strict text of the DAA per se.
In its Preamble, DAA makes reference to Art. 59 (2) ECHR which prescribes
that the EU may accede to the ECHR—which Protocol 14 of the ECHR had already

111
E.g. for such possible conflicts that are supposed to exist between the EU law and ECHR, one
can take the example of the compliance of Dublin II Regulation with ECHR, the latter certainly
providing higher standards compared to that EU act. Lenart (2012), pp. 9 et seq.
112
Explanatory Report to the DAA, para. 15.
113
In the final and Fifth Negotiation Meeting Between the CDDH Ad Hoc Negotiation Group and
the European Commission for the EU accession to ECHR, on Strasbourg 3–5 April 2013, CoE
Doc. 47+1(2013)008rev2.
114
CoE Doc. 47+1(2013)008rev2, para. 9.
110 4 A New Start for the Accession of the EU to the ECHR

provided for115—showing the legal basis whereupon the EU accession to the ECHR
is based. In the Preamble, DAA makes the legal assumption that it can enter into
force only if both the current ECHR Contracting Parties and the European Union
ratify it in accordance with their own constitutional procedures. It is postulated that
the EU, in light of the Preamble, stands as an independent party with its own legal
personality (something Opinion 2/13 seeks to amend), although its Member States
(which are already parties to the ECHR) also need to assent to the EU’s accession
through ratifying the DAA. Hence, logically, EU Member States which are simul-
taneously ECHR Contracting Parties with the act of ratification of the DAA give
their assent to EU accession to the ECHR as well. In addition, these member states
transfer to the EU portions of their own singular liability towards ECHR even for
EU acts or omissions which they have held alone to date (Bosphorus or M.S.S.
practices116). ‘Should a Member State court accept an interpretation of the ECHR
by the ECJ, which in its view would bring its jurisdiction into violation of an
international obligation of the highest order, a risk which the ECJ does not
have?’117 Such a question would not be a concern any longer after accession.
Although the Preamble does nor require further analysis, it is important to note
that it mentions that the accession of the EU to the ECHR is being undertaken to
ensure two broad-scope goals: first, to further establish the fact that the EU is
founded on the respect for human rights, as a value stemming from the EU Treaties,
and second, to enhance coherence of human rights protection in Europe,118 mean-
ing to set up an institutional mechanism of submission and interaction between the
Luxembourg and Strasbourg Courts.119 Such coherence, Callewaert argues, will be
of both formal and substantive effect.120 The latter may finally materialize White’s
proposal for ‘mutual trust’ between the two courts.121 However, it seems that
neither of these two intentions can be regarded as the sole outcome of EU accession
to the EU, as the core human-right counter argument would be that the accession
per se is being undertaken to shift the position of the EU from an isolated landscape

115
Joris and Vandenberghe (2008–2009), p. 36; See also: Ribble (2010–2011), p. 225; Martı́n and
De Nanclares (2013), p. 5.
116
On M.S.S., see e.g.: Nanopoulos (2013), pp. 277/8 et seq.
117
Editorial (2013), p. 473.
118
Quirico (2010), p. 33; See also: Lenaerts and de Smijter (2001), p. 100; Odermatt (2014a),
pp. 12–13; White (2010), p. 435; Tulkens (2013), p. 16.
119
Which have had times of dialogue but also of competition, at: Peers (2006), p. 443; On the
enhancement of human rights protection after accession, see also: Groussot et al. (2011), p. 17; See
also another relevant interesting claim on this topic, at: Miiller (2007), p. 34 (‘. . .as long as an
“International Court of Human Rights” is not in place and as long as the EU is not a member of the
ECHR, great uncertainty and a lack of judicial remedy could remain.’).
120
Callewaert (2014), pp. 9–10.
121
White (2010), p. 433.
4.7 Outline of the Draft Accession Agreement of the EU to the ECHR: What. . . 111

of human rights protection to an externally controlled human rights organization.122


It should be noted that, coming back to the second point, the ‘coherence of human
rights’ concept does not necessarily (strictu sensu) mean full coherence but rather a
coherence on the minimum standards,123 as, even post-accession, the Charter will
continue to permit the EU to offer broader human rights protection compared to the
Convention system. A further point, crafted rather covertly, of the Preamble
mentions the fact that the EU accession to the ECHR will make it possible for
private parties to access the Strasbourg Court as a means to file applications against
the EU for its acts or omissions which may have allegedly violated the Convention.
In truth, the factual and legal importance of the EU accession to the ECHR may
have this sole advantage as human rights protection in the EU may not be entirely
possible if the EU is isolated from the control of international law human rights
courts.124 Ryngaert argues rightly that ‘accession should signify a shift from the
application of a standard of equivalent rights protection to a standard of identical
rights protection’,125 the latter representing the wish of all human rights lawyers.
Nevertheless, from another perspective, as there exist multilayered human rights
regimes in Europe, ‘[i]t is inevitable that these different layers of governance will at
times rub against each other, producing inter-layer “irritation”.’126 Dealing with
these irritations would seem as the more problematic issue which the DAA would
need to tackle effectively and with caution. To this extent, one needs to add that the
DAA does not provide for an explicit mechanism to resolve the potential treaty
conflicts between ECHR and EU Treaties (e.g. the conflict between EU law
autonomy and the Convention’s human rights protection), although the latter
strongly insists on its autonomy. One may use the example of the UN Charter,
which in Art. 103 reads:
In the event of a conflict between the obligations of the Members of the United Nations
under the present Charter and their obligations under any other international agreement,
their obligations under the present Charter shall prevail.

122
On this, see e.g.: Garcı́a (2002)), p. 501; On the submission of the Luxembourg to the
Strasbourg Court, see e.g.: Odermatt (2014a), p. 5; See also a general note on this, at: Tulkens
(2013), p. 9.
123
See a similar argument, at: Callewaert (2014), p. 19.
124
E.g. and Cf.: Muir (2014), p. 219, arguing that accession will increase human right protection in
the EU; See also: Groussot et al. (2011), p. 1.
125
Ryngaert (2011), p. 1016.
126
O’Cinneide (2009), p. 1.
112 4 A New Start for the Accession of the EU to the ECHR

UN Charter seemed to provide an intentional primary position to itself in the


hierarchy of international law.127 DAA, by contrast, seems to neglect this issue,
clearly disregarding the question of any potential treaty conflict between ECHR and
EU Treaties, but also aiming to build a consensual relationship with the EU
Treaties—one which would not ignite aversion to the EU legal order.128 Lacking
an explicit mechanism of treaty conflict within the DAA—however—leaves this
question to the VCLT principles on treaty law.
A final point to be noted in the Preamble is the DAA’s assertion that accession
should end up with an outcome which both respects and does not affect the specific
nature of the EU legal order. As Thym puts it—in a balanced way—‘the objective is
evident: to support oversight of EU law by the [. . .Strasbourg Court. . .], while
preventing that Strasbourg intervenes instead of Luxembourg.’129 Such mentioning
refers mainly to the EU law autonomy, as the very specific nature of the EU legal
order, but also reflects the need to protect EU law as a state-like independent source
of law. This intention shown in the Preamble is in fact the only worry which the EU
would like to have addressed most directly and thoroughly with the DAA, some-
thing which stand in line with Opinion 2/94 on EU competence to accede to the
ECHR130 taken in conjunction with Opinion 1/91 and Opinion 1/00. This being an
institutional intention of the EU rather than a need for better ensuring human rights
protection in the EU, the DAA generally aims to reflect a rather protective approach
to ensuring that the EU law characteristics are all sheltered from any interference of
the Strasbourg Court or the ECHR regime of law in general. Preserving the
characteristics of EU law—such as its autonomy—may also be interpreted as
limiting the jurisdiction of the Strasbourg Court over the EU as a High Contracting
Party. One may rightly use the Hilaire v. Trinidad and Tobago case of the Inter-
American Court of Human Rights, where the Court—by refusing a reservation of
Trinidad Tobago on limiting its jurisdiction—ruled that: ‘[. . .] the purported
restriction relates to the State’s “recognition” of the Court’s compulsory jurisdiction
so that its judgments do not “infringe, create or abolish any existing rights or duties
of any private citizen” (sic). Again, though the precise meaning of this condition is
unclear, without a doubt it cannot be utilized with the purpose of suppressing the

127
On a general examination of this issue, see also: Pauwelyn (2003), pp. 99 et seq; In addition, for
instance, the Luxembourg Court has ruled in The Queen ex parte Centro-Com that the Community
law may not prevent the EU Member States from implementing UN law. See: Court of Justice of
EU, The Queen ex parte Centro-Com v. HM Treasury, Case C-124/95 [1997] ECR I-81, para. 61.
128
One may recall that the Court of First Instance had accepted in Yusuf the primacy of interna-
tional law in face of EU law, asking EU to disapply EU law if it contradicts to international law.
However, that approach was later changed by the European Court of Justice acting as a second/
appeal instance in that case. See: Court of First Instance, Yusuf v. Council and Commission, Case
T-306/01 [2005] ECR II-3353, paras. 234–240.
129
Thym (2013a), p. 3.
130
See: Muir (2014), p. 243, arguing that the EU Court maintained that the EU has no competence
to accede, which must be understood as that competence remains to the Member States, the latter
having the competence to decide otherwise; See also: Analysis (1997), pp. 235 et seq; Quirico
(2010), p. 31/2.
4.7 Outline of the Draft Accession Agreement of the EU to the ECHR: What. . . 113

jurisdiction of the Court to hear and decide an application related to an alleged


violation of the State’s conventional obligations [. . .]’.131 The standard introduced
in Hilaire v. Trinidad and Tobago may well serve to inform the boundaries of
accession in the face of the Convention’s efficiency. The question that needs be laid
down from the very beginning therefore is whether the protection of EU law
peculiarities could be achieved without breaching the proclaimed intention to
enforce a Strasbourg-like system of human rights upon the EU on equal footing
with other High Contracting parties.132 This is something this book will tackle in
the following chapters in depth—adding that Luxembourg and Strasbourg may be
seen as singing the/a ‘je t’aime, moi non plus?’133 song.
Article 1 DAA tackles the scope of EU accession to the ECHR and its pro-
tocols.134 The scope of accession is a rather important issue, as it sets the limits of
control that the Strasbourg system has over EU law.135 As it is well known, there
have been 14 protocol annexed to the ECHR to date, many of which are no longer in
force. Therefore, with the EU acceding to the ECHR, and becoming a segment of
the Convention institutional framework, the EU would not automatically become
part of the ECHR protocols as accession to those protocols may thereafter involve a
separate process.136Article 1 establishes that under the current DAA the EU accedes
to the ECHR, and to Protocol of the ECHR (which regulates the right to protection
of property, right to free elections and the right to education) and Protocol 6 of the
ECHR, which abolishes the death penalty. These are the only protocols which all
EU Member States have joined.137 Accession only to these two protocols,
according to Gragl, means that the EU is following the ‘lowest common denomi-
nator’ of human rights in terms of the entire pool of protocols.138 Four other

131
Inter-American Court of Human Rights, Hilaire v. Trinidad and Tobago, Preliminary Objec-
tions, judgment of 1 September 2001, Series C, No. 80, para. 86.
132
Von Bogdandy (2012), p. 517; Lenaerts and de Smijter (2001), p. 100; Craig (2013), p. 1142;
On the request for an equal footing position of EU with other Convention Member States, see:
European Parliament resolution of 19 May 2010 on the institutional aspects of the accession of the
European Union to the European Convention for the Protection of Human Rights and Fundamental
Freedoms (2009/2241(INI)), p. 1; Compare the latter to the Luxembourg’s Opinion 2/13, which
basically requests the contrary; Cf.: Tulkens (2013), p. 10 (‘Absolute equality between the EU and
the other Contracting States is therefore neither possible nor even desirable.’); Cf.: de Rivery and
Chassaing (2013), p. 4 (In a post-accession perspective, ‘[t]he CJEU ought therefore to no longer
be able to hide behind the Bosphorus veil, because in accordance with ECHR provisions, it will
just become one among the many supreme courts of the Contracting Parties.’).
133
de Rivery and Chassaing (2013), p. 1.
134
The Council had proposed two decades ago that the then EC should have acceded to all
protocols in which all Convention contracting parties had also acceded. See: Analysis (1997),
p. 238/9.
135
Jacque (2011), p. 1002.
136
The European Parliament had proposed that the EU join only those protocols which include the
Charter of Fundamental Rights’ rights. See: Quirico (2010), p. 36.
137
Jacque (2011), p. 1004; See also: Craig (2013), p. 1118.
138
Gragl (2013), p. 94.
114 4 A New Start for the Accession of the EU to the ECHR

protocols remain in force139 wherein the EU might join, and the procedure to join
those protocols is tackled by those protocols themselves (separately and outside the
DAA). That means specific and separate accession procedures must be pursued by
the EU if it wishes to join those protocols.140 It must be said here that, indispensable
from the fact that the EU narrows the scope of joining the Convention system by not
acceding to those remaining protocols, nevertheless, this narrow scope of accession
does not per se prejudice the EU’s position as a fully-fledged High Contracting
Party to the Convention’s institutional and legal systems. It was argued during the
negotiations process that it would not be appropriate to allow the EU to accede to all
protocols, as that might turn into giving to the EU additional competences by
Member States outside those already transferred by the Treaties (an argument
that may be used for the DAA as well; see supra, subchapter on EU becoming
primus inter pares).141 The selected option of acceding only to these protocols (and
not to all others) hence has been made to ‘reassur[e] those Member States who fear
that protocols they have not ratified would become applicable in their national legal
order by means of EU law.’142 This, according to Gragl, basically means that EU
will uphold the principle of neutrality towards the Member States’ positioning in
face of the remaining protocols.143 A legitimate question would therefore be
whether the Luxembourg Court will continue to consider those protocols where
the EU has not acceded as general principles of EU law, and whether the Strasbourg
Court would continue to apply the Bosphorus formula with regard to Member
States actions (originating in EU law) violating a non-acceded protocol.144 This
remains to be seen in practice, but part of it may be answered with reference to
Fransson case where the Luxembourg Court refused to rely on a Convention

139
On the existing protocols and their material nature, see: Shelton (2003), pp. 100 et seq.
140
Raba (2013), p. 563; See also: Gragl (2013), p. 94.
141
E.g.: Groussot et al. (2011), p. 4.
142
Jacque (2011), p. 1003; The latter, however, argues that such resistance of Member States
would seem impracticable as all Member States are already submitted to the Charter of Funda-
mental Rights, which is effectively far broader than all Convention protocols; See also on this
resistance: Odermatt (2014a), p. 30.
143
Gragl (2013), p. 94.
144
See also: Jacque (2011), p. 1004/5; See also: Gragl (2013), p. 95.
4.7 Outline of the Draft Accession Agreement of the EU to the ECHR: What. . . 115

Protocol to which not all EU Member States had acceded.145 One must also mention
Gragl’s point that the remaining protocols’ rights are already guaranteed with the
EU’s Charter,146 therefore the EU fundamental rights pool will not suffer anything
from that deficit in the internal context. Gragl also rightly questions the application
of Art. 52 (3) of the Charter when it comes to interpreting the Charter’s rights which
correspond to the EU-unacceded ECHR protocols. He basically argues that the
Luxembourg Court may either choose to demonstrate judicial activism and consider
Art. 52 (3) of the Charter as covering the rights of the EU-unacceded ECHR
protocols in the same line with the acceded protocols, or merely consider their
rights as general principles of law in light of Art. 6 (3) TEU.147 Both points of Gragl
seem valid.
Besides tackling the scope of accession, Article 1 DAA also establishes that the
DAA legally forms part of the Convention, therefore having the same legal rank as
the Convention itself. Article 1 also outlines the attribution of jurisdiction between
the EU and its Member States in the face of the ECHR, something which is tackled
infra in Chap. 6. Besides making this division of responsibility at least in theory,
Article 1 also amends the ECHR to apply concepts like ‘state’ or ‘state party’ to the
EU as a non-state party to the ECHR and having the same legal status as a High
Contracting Party with all other contracting parties. The interpretation of such
terms are made to ensure that the EU finds a functional place as a High Contracting
Party to the Convention, allowing the EU to stand as at least an equal contracting
party. Article 1 establishes a jurisdictional model which allows individuals to be
subject to the Convention on the basis of their EU member State territory.148 As
such, Article 1 (on EU’s espace juridique) reads that: ‘the expression “everyone
within their jurisdiction” appearing in Article 1 of the Convention refers to persons
within the territory of a High Contracting Party, it shall be understood, with regard
to the European Union, as referring to persons within the territories of the member
States of the European Union to which the Treaty on European Union and the

145
In examining the claimant’s reliance upon an ECHR Protocol to which not all EU Member
States had acceded, AG Cruz Villalon had argued that ‘the fact is that not all the Member States
have ratified that provision, while others have adopted reservations or interpretative declarations in
relation to it. The effect of that situation is that the requirement to interpret the Charter in the light
of the ECHR and the case-law of the European Court of Human Rights (Article 52(3) of the
Charter) becomes, so to speak, asymmetrical, leading to significant problems when it is applied to
this case.’ See: Opinion of AG Cruz Villalon delivered on 12 June 2012, on Case C-617/10,
Åklagaren v Hans Åkerberg Fransson, para. 70. Although AG Cruz Villal on and the Luxembourg
Court in Fransson did not explicitly suggest how to tackle these protocols, their basic argument
was that as long as EU has not acceded to the ECHR and those protocols, the EU Court may not
tackle that issue as, if it does, that would mean as it is regulating the relationship between the
Convention and its Member States. The Court and AG Cruz Villal on seem to suggest that such
issue should remain outside the answer of the EU Court, which practically means that such
protocols have no value in EU legal order.
146
Gragl (2013), p. 94.
147
Gragl (2013), p. 95.
148
See also: Quirico (2010), p. 40.
116 4 A New Start for the Accession of the EU to the ECHR

Treaty on the Functioning of the European Union apply.’ This jurisdictional model
of the Convention’s application to territories of the EU Member States makes a
direct link between the EU Member States’ territories and the EU Treaties’ territo-
rial application, adding that Article 355 TFEU149 defines the territorial range of the
EU Treaties on its Member States. Such territorial definition of the EU ensures that
for the Convention’s purposes, EU law autonomy is better protected, thereby
excluding the possibility that territorial jurisdiction be determined through the
bottom-level law which is applied upon the individual whose rights are at stake
before the Strasbourg Court. The latter bottom-up approach would have allowed the
Strasbourg Court to intervene in EU law autonomy, therefore it was assumingly
dismissed as inadvisable. Although one may legitimately propose the argument that
even the current model of shaping the jurisdiction of EU under the Convention may
leave room for the Strasbourg Court to enter into EU Treaty law with the intention
of verifying whether a certain claimant has been subject to TEU or TFEU applica-
tion, thereby asserting whether the case falls under EU jurisdiction.
Article 2 regulates the issue of reservations to the ECHR and its protocols. That
being the case, Article 2 merely establishes that the EU may make reservations in
line with Art. 57 of the ECHR, which is a normal space provided for every High
Contracting Party. The EU may not be an exception, although it is normally not
customary to make reservations to the Convention.150 Allowing for the EU to make
reservations, however, is conditioned with Art. 10 of the DAA which does not in
itself make any reservation to the Convention. If, as seen infra, the DAA makes no
reservation—such as prohibiting interference to EU law autonomy, something
Protocol 8 to the EU Treaties provides for—then the argument would remain that
Art. 2 & Art. 10 of the DAA do not place a limit on Strasbourg’s usual jurisdictional
territory and therefore allow room for such potential interferences in EU law
autonomy (even though the EU, for its internal use, has the legally-effective
Protocol no. 8).151
Article 3 tackles the issue of division of responsibility between the EU and its
Member States for Convention violations which may have originated under EU law
but have been implemented by EU Member States, or vice versa in very exceptional

149
On the latter part of the argument, see: Odermatt (2014a), p. 33.
150
Analysis (1997), p. 238.
151
I use a synonymous argument here inspired by the general note of: Ribble (2010–2011), p. 226;
Cf.: Gragl (2013), p. 130, who, by referring to an ICJ judgment, comes to the conclusion that a
reservation, e.g. to exclude the EU primary law from the Strasbourg’s jurisdiction would go
contrary to VCLT’s Art. 19b (in view also of the terms of Art. 2 (2) of the DAA).
4.7 Outline of the Draft Accession Agreement of the EU to the ECHR: What. . . 117

cases.152 By building the co-respondent mechanism,153 examined infra, Article 3


aims to reach two basic goals: first, to allow for the Convention system to function
also upon the multilevel EU-Member States types of responsibility for human
rights, wherein the EU or Member States may be separately or jointly responsible
for certain human rights violations, and, second, and the most important for the EU
law itself, to preserve EU law autonomy internally but more importantly, the
external autonomy of the EU towards the Strasbourg regime of law.154 Article 3
therefore not only establishes the co-respondent mechanism as a means for the
Strasbourg Court to simultaneously attribute the responsibility for violations of the
Convention to both the EU and its Member States, but also prevents the exercise of
any jurisdictional competence for which the Luxembourg Court possesses exclu-
sive authority. Article 3 also introduces the prior involvement mechanism, which is
a means for the Luxembourg Court to judge a case that involves an alleged
violation—the latter assumed to originate from EU law—if it reaches the Stras-
bourg Court without having had the possibility to be judged directly or indirectly by
the Luxembourg Court first. Both of these mechanisms under the DAA seem to
privilege the EU in comparison to other High Contracting Parties to the Conven-
tion. This being the case, the author of the book offers Strasbourg’s Pfunders case,
in which the Court ruled that:
[t]he obligations undertaken by the High Contracting Parties in the European Convention
are essentially of an objective character, being designed rather to protect the fundamental
rights of individual human beings from infringements by any of the High Contracting
Parties than to create subjective and reciprocal rights for the High Contracting Parties
themselves.155

The same argument finds support in the Inter-American Court of Human Rights
Advisory Opinion OC-2/82, where the Court ruled that:
In concluding [. . .] human rights treaties, the States can be deemed to submit themselves to
a legal order within which they, for the common good, assume various obligations, not in
relation to other States, but towards all individuals within their jurisdiction.156

152
It should be noted that although the EU has rather clearly established competences, there are
usual tendencies to recognize the limitations to those competences, which can sometime arrive in a
position wherein the EU takes over to guarantee rights which may be of the nature that go beyond
what can the EU possess in terms of its competences. See on this e.g.: L€o€of (2006), p. 426.
153
It is noted that both the co-respondent mechanism and the prior involvement procedure have
been influenced by the joint statement of the two presidents of European courts. See on this:
O’Meara (2011), p. 1814.
154
On the external autonomy, Cf.: Besson (2009), p. 240; On the question of whether there is an
ongoing extention of international tribunals’ jurisdiction on matters that were previously state-
owned, see: Lavranos (2008).
155
European Commission on Human Rights, Austria v. Italy (the ‘Pfunders’ Case), Application
No. 788/60, Decision as to Admissibility (11th January 1961), p. 19.
156
Inter-American Court of Human Rights, (‘The Effect of Reservations on the Entry into Force of
the American Convention on Human Rights (Arts. 74 and 75)’), Advisory Opinion OC -2/82 of
24 September 1982, Inter-American Court of Human Rights (Ser. A, no. 2) (1982), para. 29.
118 4 A New Start for the Accession of the EU to the ECHR

It seems clear that with Pfunders, the Strasbourg Court has prohibited the
recognition of the possibility that High Contracting Parties use their Convention
obligations as means to establish subjective and reciprocal rights for themselves
(the same line of reasoning is followed in the Inter-American Court of Human
Rights Advisory Opinion OC-2/82). Both the co-respondent mechanism and the
prior involvement established by the DAA seem to offer the EU certain subjective
rights in the face of other contracting parties of the Convention system—obviously
diminishing the objective character which the Convention enjoys on all original
High Contracting Parties. This said, it is hereby argued that Pfunders proscribes
options for the EU to gain subjectively-privileged rights in the Convention system
such as the co-respondent and the prior involvement mechanism—as the Conven-
tion is merely an objective system of human rights protection and may not be used
for any other purpose. This author therefore considers that these two mechanisms
seem to deconstruct a very important standard against which the Convention should
be measured—hence proving rather problematic to argue that the DAA provides an
accession process with objective human-rights intentions for the EU. In sum,
Article 3 of the Draft Accession Agreement stands as a largely important segment
of the DAA and probably the most worrying and criticized mechanism which this
book will tackle. It is argued also that this article seems rather imprecise and of a
broad nature.157
Article 4 addresses the inter-state mechanism of the Convention, and revises the
current legal understanding of Art. 33 ECHR so that the inter-party mechanism
becomes applicable to the EU, thereby resolving the legal conflicts that have existed
between the EU Treaties (Art. 344 TFEU) and the Convention in this regard. For
this purpose, Article 4 amends the title of Art. 33 from ‘inter-state cases’ to ‘inter-
party cases’, thus accommodating the EU as a possible party (both claimant and
respondent party). Further, Article 5 establishes that proceedings before the Lux-
embourg Court ‘shall be understood as constituting neither procedures of interna-
tional investigation or settlement within the meaning of Article 35, paragraph 2.b,
of the Convention, nor means of dispute settlement within the meaning of Article
55 of the Convention.’ Article 4 therefore ensures that the EU is not blocked from
the possibility of challenging a High Contracting Party—by using Art. 33 ECHR—
before the Strasbourg Court. This also allows other High Contracting Parties to
challenge an act or omission of the EU before the Court inter se. Article 4 therefore
will assure an equal position of the EU with other High Contracting Parties as
regarding the applicability of Art. 33 ECHR, thereby improving the Convention’s
character as an equal-treatment instrument not only for the EU in relation to other
High Contracting Parties but also vice versa.158 Finally, it is important to note that
through the alteration that Article 4 entails, the Convention becomes a possible
arena for the inter-High Contracting Party submissions wherein an international

157
On the latter, see: Raba (2013), p. 563.
158
On the need for equal-footing position of EU and Member States, see also: Conforti
(2010), p. 83.
4.7 Outline of the Draft Accession Agreement of the EU to the ECHR: What. . . 119

organization is allowed to initiate proceedings for human rights violations against


state parties and vice versa, clearly introducing a new standard in international
human rights law and revising the conventional model of pure states-centered
human rights law.
In order to ensure fully-fledged integration of the EU into the ECHR system, the
DAA not only tackles the legal issues relating to the submission and adjudication of
cases before the Strasbourg Court, but also institutionally integrates the EU in more
or less the standard footing in the Convention’s institutional structures.159 This will
produce dual representation, wherein both the EU and its Member States participate
in their own right.160 As a point of digression, this position is rather different from
other international organizations, such as the World Trade Organization, wherein
the EU does not have simultaneous voting rights alongside its Member States.161
Returning to the issue of dual representation in the Convention’s institutional
structure,162 it is important to note that the Member States might be forced under
their duty of sincere cooperation163 to have their positions coordinated while
participating together with the EU in the Convention statutory bodies, of course
only within the ambit of an EU-related obligation.164 A similar position was
decided in a case regarding the arrangement between the EU and its Member States
in the Food and Agricultural Organization (FAO).165 As to the entry point of

159
Joris and Vandenberghe (2008–2009), p. 41 (Participation in these Convention/Council of
Europe structures ‘[. . .] could result in a better division of tasks and activities, and would allow the
EU to speak for itself in Council of Europe bodies on all issues which affect its interests and which
fall within its competence.’); See also: Raba (2013), p. 559.
160
Wetzel (2003), p. 2849; Cf.: Odermatt (2014a), p. 20 (In this case one is ‘faced with the issue of
how to reconcile the fact that one the one hand, the EU is a separate legal entity that should prima
facie have the right to vote, and the fact that the EU is constituted by other Member States who will
retain their right to vote.’).
161
Art. 9 (1) of the Marrakesh Agreement Establishing the World Trade Organization reads:
‘Where the European Communities exercise their right to vote, they shall have a number of votes
equal to the number of their member States which are Members of the WTO.’; Accord: A similar
example is the Statute of the International Renewable Energy Agency, which states: ‘In the case of
any regional intergovernmental economic integration organisation, the organisation and its Mem-
ber States shall decide on their respective responsibilities for the performance of their obligations
under this Statute. The organisation and its Member States shall not be entitled to exercise rights,
including voting rights, under the Statute concurrently.’ Council Decision of 24 June 2010 on the
conclusion of the Statute of the International Renewable Energy Agency (IRENA) by the
European Union (2010/385/EU), Art. 6 (C) [emphasis added].
162
Cf.: Odermatt (2014a), p. 21, who argues that one reason for allowing the dual representation
with DAA is because it would not be possible and feasible to define the division of competences
between the EU and Member States regarding the Convention.
163
On the duty of sincere cooperation in international practice from the perspective of Opinion
1/94, see: Editorial Comments (1995), pp. 385 et seq; See also in this regard an interesting case, at:
Court of Justice of EU, The Queen, ex parte Centro-Com Srl v HM Treasury and Bank of England,
Case C–124/95 [1997] ECR I-81, para. 27.
164
Eckes (2013), p. 273.
165
Court of Justice of EU, Commission v Council, Case C-25/94 [1996] ECR I-1469.
120 4 A New Start for the Accession of the EU to the ECHR

participation of the EU in the Council of Europe-Convention-related statutory


bodies, Article 6 (in conjunction with Art. 20 ECHR) makes room for the EU to
have its own judge in the Strasbourg Court similar to other state High Contracting
Parties,166 and makes mention of the appointment procedure that must be followed.
The EU judge would be inter alia considered as the person with knowledge of EU
peculiar features167—to which the DAA refers—when the EU is the accused party
before the Strasbourg Court. Such an EU judge, however, will not represent the EU
within Strasbourg Court proceedings, as the latter’s judges act on their independent
personal capacity. The EU judge will therefore act on an equal footing with all other
contracting parties’ judges.168 Moreover, Article 6 sets forth the method on the
basis of which the European Parliament—as the supposed state-like legislative
body of the EU—would participate in the Parliamentary Assembly of the Council
of Europe when the latter exercises its function of electing Strasbourg Court judges.
To institute this procedure, Article 6 establishes that: ‘The delegation of the
European Parliament shall have the same number of representatives as the delega-
tion of the State which is entitled to the highest number of representatives under
Article 26 of the Statute of the Council of Europe.’169 Although not fully equal with
the other contracting parties, the EU will achieve equality and a nucleus of
representation in the Parliamentary Assembly of Council of Europe when it elects
the Strasbourg Court judges. The mere existence of an EU presence in the Parlia-
mentary Assembly in these cases will suffice for the EU to exercise influence in the
Council of Europe institutional structures when it comes to election of judges for
the Convention system. One must also add here that the EU’s position as a High
Contracting Party to the ECHR may not be seen elsewhere with this advantaged/
honored status (although—ironically—this is not mentioned in Opinion 2/13); an
example is the EU’s enhanced observer participation in UN General Assembly,
wherein one may see that the ‘EU [. . .is. . .] not [. . .] regarded as an equal partic-
ipant on par with UN member states’.170 The EU’s equal position with states in the

166
Lock (2010), p. 777; See also: Groussot et al. (2011), p. 3; Conforti (2010), p. 86; Ribble
(2010–2011), p. 224 (‘[. . .]having an EU judge participate in ECHR cases concerning the EU may
help alleviate stress between different interpretations of the Convention as it relates to EU law.’);
Cf.: Odermatt (2014a), p. 18 (‘The concept of having a judge in respect of the EU is somewhat
novel, and does not appear in other Conventions where the EU is a party.’).
167
On the latter, more broadly, see also: Odermatt (2014a), p. 18.
168
See also: Odermatt (2014a), p. 19.
169
See also e.g.: Jacque (2011), p. 1009; O’Meara (2011), p. 1827; Martı́n and De Nanclares
(2013), p. 7/8.
170
Odermatt (2014a), p. 8.
4.7 Outline of the Draft Accession Agreement of the EU to the ECHR: What. . . 121

Convention system must therefore be taken as an opportunity for its fully-fledged


participation in international organizations.
Article 7 establishes the procedure for EU participation in the meetings of the
Committee of Ministers of the Council of Europe as part of the Convention’s
system,171 since the Committee concerned is the responsible institution for moni-
toring the execution of the Strasbourg Court’s judgments.172 The EU’s right to vote
there will be parallel to that of its Member States173 in cases involving EU issues
under the Convention. This said, originally, the Committee of Ministers is a
Council of Europe body—as defined by the Statute of the Council of Europe—
which also has specific authorizations under the Convention. Therefore, Article 7
merely tackles EU participation in the Committee of Ministers when it acts upon its
duty to fulfil the Convention functions, but nothing more beyond this. Through
Article 7 the EU becomes a member of the Committee of Ministers in line with Art.
54 ECHR functions of the Committee, and has a right to vote in cases when the
Committee decides on issues of Art. 26 (2) ECHR, Art. 39 (4) ECHR, Art. 46 (2–5)
ECHR, Art. 47 and 54 (1) ECHR. The participation of the EU in the Committee of
Ministers,174 however, is defined as one which may not prejudice the competences
that the Committee enjoys from the Council of Europe Statute, as that may establish
‘a risk of internal imbalance’175 in favor of the EU. On the other hand, Article 7
makes it clear that the participation of the EU in the Committee of Ministers when
the latter decides on issues relating to the supervision of the execution of judgments
and friendly settlements as to Art. 39 ECHR and Art. 46 ECHR may not prejudice

171
Hart (2010), p. 551; Martı́n and De Nanclares (2013), p. 7/8; Jones (2012), p. 4; On the EU
institutional participation in other international organizations and treaty bodies, see e.g.:
Hoffmeister (2007), pp. 41 et seq; See also: Groussot et al. (2011), p. 8; and, Conforti (2010), p. 86.
172
As a core statutory body of the Council of Europe, see e.g.: Joris and Vandenberghe (2008–
2009), p. 5; See also: Raba (2013), p. 569.
173
Some international organizations prohibit parallel voting. One example is FAO, which in its
Rule II (5) reads: ‘Before any meeting of the Commission [. . .] in which a Member Organization is
entitled to participate, the Member Organization or its Member States shall indicate in writing
which, as between the Member Organization and its Member States, has competence in respect of
any specific question to be considered in the meeting and which, as between the Member
Organization and its Member States, shall exercise the right to vote in respect of each particular
agenda item.’ (FAO Rules of Procedure, available at: http://www.fao.org/docrep/006/y4800e/
y4800e0m.htm); Prohibiting parallel voting may have probably been a better model of excluding
coordinated voting between the EU and its Member States in the Committee of Ministers of the
Council of Europe.
174
Lock (2010), p. 777; See also: Quirico (2010), p. 52; Jacque (2011), p. 1010, arguing that EU
should make a declaration to the DAA undertaking not to coordinate decisions with its Member
States within the Committee of Ministers if the issue involves a non-EU Member State. It also
makes the argument that with EU part of the Committee, EU and Member States may practically
dominate the Committee’s work; Cf.: Tulkens (2013), p. 15, who argues that, nevertheless, the
EU’s position within the Committee of Ministers would at least be important for the ‘symbolic
power games between the EU and its Member States on the one hand and the non Member States,
on the other.’
175
Martı́n and De Nanclares (2013), p. 8.
122 4 A New Start for the Accession of the EU to the ECHR

the effective and functional exercise of the Committee’s supervisory functions.


Then again, Article 7 makes it clear that practices wherein the EU and its Member
States take ‘coordinated’ decisions176—on issues where the Committee should
supervise the implementation of a judgment relating to the EU and its Member
States to ensure the matter is concluded—should be precluded, as that may diminish
the effective exercise of the Committee’s overseeing functions. This may be seen as
‘the thorniest’177 issue, as having a block of 29 parties (including the EU and its
Member States) potentially ‘acting’ under a single direction may well ruin the very
concept on the original function of the Committee.178 Therefore, it should be
argued that Article 7 tends to ensure that the EU right to vote in the Committee
of Ministers should not influence or condition the votes of the EU Member
States179—even though the issue of enforcing judgments might be of EU law
substance. Such a practice would seriously weaken the possibility of the Committee
of Ministers to stand as a supervisory body for judgments to which EU primary and
secondary law is subject. No bond or prior agreement between the EU and its
Member States may be allowed within the Committee of Ministers voting patterns,
Article 7 seems to establish—although this may conflict with the Luxembourg’s
insistence to see the EU and its Member States having a unitary representation in
the international scene.180 The prohibition analyzed above, one may argue, may be
easily considered by the Luxembourg Court as breaching the principle of loyal/
sincere cooperation under the EU Treaties, nevertheless.181

176
One example of ‘coordinated position’ between the EU and its Member States as regards the
participation in the decision-making body of an international treaty may be found at: ‘Internal
Agreement between Representatives of the Governments of the Member States, meeting within the
Council, on the Financing and Administration of Community Aid under the Financial Protocol to
the Partnership Agreement between the African, Caribbean and Pacific States and the European
Community and its Member States signed in Cotonou on 23 June 2000 and the allocation of
financial assistance for the Overseas Countries and Territories to which Part Four of the EC Treaty
applies (2000/770/EC)’. OJ L 317, 15.12.2000, pp. 355–357; See in this regard a decision of
Luxembourg Court as regards voting rights between EU and its Member States in FAO, at: Court
of Justice of EU, Commission v Council (FAO), Case C-25/94 [1996] ECR I-1469.
177
Martı́n and De Nanclares (2013), p. 8; Cf.: Odermatt (2014a), pp. 8 and 20, arguing that it is
normal for non-EU Member States to consider EU’s dual representation at international organi-
zations as a risky practice, as the EU may encourage block voting with its Member States.
178
Cf.: Kuiper (1995), p. 243/44, who suggests that it is desirable to adopt a code of conduct to
regulate the coordination between the EU and its Member States when they participate in WTO
mechanisms—as to comply with the duty of sincere cooperation. Although the desirability of such
mechanisms may seem legitimate for technical reasons, it would however seriously undermine the
external independence of EU Member States when they participate in international organizations
along the EU.
179
Cf.: See also in this regard the EU’s participation in Korean Peninsular Energy Development
Organisation Executive Board, at: Council Common Position 2001/869/CFSP [2001] OJ L 325/1,
article 1.
180
Court of Justice of EU, Opinion 2/91, [1993] ECR I-1061, para. 36.
181
Court of Justice of EU, Opinion 2/91, [1993] ECR I-1061, paras. 36–38.
4.7 Outline of the Draft Accession Agreement of the EU to the ECHR: What. . . 123

Article 8 regulates the issue of the EU’s financial contribution to the Council of
Europe to maintain the functioning of the Convention’s system. Article 8 therefore
provides for a contribution that the EU should award to the Council of Europe
‘equal to 34% of the highest amount contributed in the previous year by any State to
the Ordinary Budget of the Council of Europe.’182 The EU’s contribution to the
Convention system, however, should have logically been even higher as there are
additional costs which the Strasbourg Court would have to undertake when it deals
with longer procedures and EU-centered delays for which the DAA leaves room.
Relations of the EU and the DAA with the other agreements to which the
Convention system usually refers are regulated by Article 9. Article 9, for this
purpose, establishes that the EU will respect—as opposed to get internationally
bound to—Art. 1 to 6 of the European Agreement relating to Persons Participating
in Proceedings of the European Court of Human Rights,183 Art. 1 to 19 of the
General Agreement on Privileges and Immunities of the Council of Europe,184 and
Art. 1 to 6 of the Protocol 6 to the General Agreement on Privileges and Immunities
of the Council of Europe.185 Further, Article 9 reads that ‘[f]or the purpose of the
application of the agreements and protocols referred to [. . .above. . .], the
Contracting Parties to each of them shall treat the European Union as if it were a
Contracting Party to that agreement or protocol.’186 It seems that Article 9 tends to
‘accede’ the EU to the abovementioned instruments without the EU having gone
through their ratification procedures (or act as if the EU has acceded). In other
words, the EU will be unilaterally undertaking an obligation before international
law to respect those agreements187 as if they were separately concluded. In addi-
tion, Article 9 tends to make the fact and effect that the Convention and the Council
of Europe ‘should’ take for granted that the EU will respect those instruments as if it
were in fact a party to them. This situation is similar to the argument illustrated in
the ICJ’s North Sea Continental Shelf case,188 where the ICJ accepted the fact that
an international law entity may unilaterally accept an obligation—without having

182
Art. 8 (1) DAA; A general note on the flat-rate participation on expenditures rule that EU
should adhere to, see: Jacque (2011), p. 1010; See also: Martı́n and De Nanclares (2013), p. 13
(‘The contribution of approximately 9.34m euros to the Council of Europe’s shaky finances could
go a long way to overcoming the reluctance which is still shown in the negotiations by some States
which are parties to the ECHR but not members of the EU.’); Odermatt (2014a), p. 21.
183
European Agreement relating to Persons Participating in Proceedings of the European Court of
Human Rights of 5 March 1996 (ETS No. 161).
184
General Agreement on Privileges and Immunities of the Council of Europe of 2 September
1949 (ETS No. 2).
185
Protocol 6 to the General Agreement on Privileges and Immunities of the Council of Europe of
5 March 1996 (ETS No. 162).
186
Art. 9 (2) DAT.
187
On the latter argument, see also: Odermatt (2014a), p. 23.
188
International Court of Justice, North Sea Continental Shelf (Federal Republic Of Germany/
Denmark; Federal Republic Of Germany/Netherlands), Judgment, ICJ Reports 1969, 20 February
1969, para. 27.
124 4 A New Start for the Accession of the EU to the ECHR

entered into any corresponding contractual duty. This model of unilateral reception
seems to be read objectively as an obligation under international law, although it is
not concluded in a normal format with another entity. Although Article 9 seems to
go ultra vires from the real substance and object of the DAA, it provides for the idea
that the EU should become a fully-fledged party to the Convention and Council of
Europe-Convention-centered instruments.
Article 10 regulates the signature, ratification and entry into force of the DAA,189
a moment in which the EU will become a High Contracting Party to the Conven-
tion, Protocol to the Convention and Protocol 6 to the Convention. Certainly, the
means via which the EU will ratify the DAA are not established here, as this is an
exclusive competence issue regulated by Art. 218 TFEU. In order for the DAA to
become ratified and enter into force, it will need to be consented to by each of the
current High Contracting Parties and the EU as a new190 High Contracting Party to
the Convention, in accordance with their domestic constitutional procedures and in
line with the prescriptions specified in accordance with the VCLT (Art. 1 (b)). This
is expected to take considerable time to become concluded from all parties.191
Article 11 regulates the issue of reservations to the DAA by any of the
contracting parties,192 however it mentions that there are no reservations submitted
to the DAA by any of the parties.193 Being based in Art. 2.1.d VCLT—the specific
provision which regulates reservations—the reservation to a treaty ‘purports to
exclude or to modify the legal effect of certain provisions of the treaty in their
application to that’ party. Clearly, as the EU does not make any reservations to the
DAA, and according to the understanding of Art. 2.1.d VCLT, the EU does not seek
to specifically protect its autonomy in light of the Strasbourg Court’s potential
behaviour in the future. Otherwise, there seems no legal positivist reason for the EU
to have refrained from inserting such a reservation explicitly in the DAA. Is it
possible that the EU did not intentionally make reservations as that would have
required the consent of all Convention contracting parties? This author makes the
argument that this may not be the case, as VCLT permits the EU alone to make the
reservation concerned, as expressed concretely in Art. 20.1 VCLT: ‘A reservation
expressly authorized by a treaty does not require any subsequent acceptance by the

189
Accord.: Art. 11 (2), Art. 14 & Art. 15 VCLTSIOIO.
190
E.g.: Quirico (2010), p. 32.
191
On the latter, see e.g.: Martı́n and De Nanclares (2013), p. 7.
192
Cf.: Art. 19 VCLTSIOIO.
193
On this, see also: Martı́n and De Nanclares (2013), p. 7.
4.7 Outline of the Draft Accession Agreement of the EU to the ECHR: What. . . 125

other contracting States unless the treaty so provides.’194 On the other hand, one
would legitimately question whether the lack of reservations from the side of the
EU affects the individual reservations of the EU Member States with regard to the
Convention. The European Convention Secretariat, making an observation on this
issue, had argued that ‘[t]he Member States’ individual reservations made in respect
of the ECHR and additional protocols [. . .] would in any event remain unaffected
by accession since they concern the respective national law, whereas accession by
the Union would have legal effect only insofar as Union law is concerned.’195
Unfortunately, this argument has not been understood by Opinion 2/13 (para. 227)
either. Therefore, with regard to the latter argument, everything related to the
DAA’s reservations would not in practice need to affect the individual position of
EU Member States in relation to the Convention.
Article 12, on the other hand, as the very final provision of the DAA, specifies the
parties which should be notified by the Secretary General of the Council of Europe
in respect to consent instruments of ratification or approval and other important
issues relating to the DAA. The latter stands in conformity with Art. 11.2
VCLTSIOIO which prescribes that ‘[t]he consent of an international organization
to be bound by a treaty may be expressed by signature, exchange of instruments
constituting a treaty, act of formal confirmation, acceptance, approval or accession,
or by any other means if so agreed.’196 After the DAA enters into force, then, every
‘new Council of Europe state will therefore be bound [to it], not only by the
amended Convention and Protocols, but also to the Accession Agreement, which
is to have an “explicit link” with the Convention.’197
Before concluding this chapter, one may consider an additional brief discussion
point. From the overall accession stipulations’ picture—Lixinski198 opines that the
DAA packages accession as a court-oriented product and does not reflect on a larger
integration in the sense of legal pluralism in Europe. The author of this book
disagrees with Lixinski for two fundamental reasons. First, the DAA does not
pack the accession elements merely within the context of the two courts’ relation-
ship. It covers numerous additional issues relating to the overall mechanisms that

194
Accord, also: Inter-American Court of Human Rights, (on the effect of reservations on the entry
into force of the American Convention on Human Rights) Advisory Opinion OC -2/82 of
24 September 1982 (Arts. 74 and 75), Series A, No. 2, para. 34 (‘A treaty which attaches such
great importance to the protection of the individual that it makes the right of individual petition
mandatory as of the moment of ratification, can hardly be deemed to have intended to delay the
treaty’s entry into force until at least one other State is prepared to accept the reserving State as a
party.’); Cf.: International Law Commission, (‘The unity or diversity of the juridical regime for
reservations’), Preliminary Conclusions of 1997, para. 1 (‘the object and purpose of the treaty is
the most important of the criteria for determining the admissibility of reservations’).
195
The European Convention, The Secretariat, Brussels, 22 October 2002, CONV 354/02 WG II
16 Report, Chairman of Working group II “Incorporation of the Charter/accession to the ECHR”,
Final report of Working Group II, p. 15.
196
VCLTSIOIO.
197
Odermatt (2014a), p. 31.
198
Lixinski (2014), p. 232.
126 4 A New Start for the Accession of the EU to the ECHR

will make the two regimes compliant under a certain legal framework, such as the
scope of accession, nature of obligations upon the EU, means of delineating the
jurisdictions of human rights between the two orders, etc. There is nothing very
special with regard to the manner in which other High Contracting Parties are
framed under the same accession process. Second, although Lixinski may be
perhaps right that most of the mechanisms established by the DAA relate to the
methods of cooperation between the two courts, this argument may not go so far as
to become construed in such a technical language. Legal pluralism in Europe, first
of all, is reflective mainly of the two courts’ relationship, as they have played the
core function of building the nature and scope of their regimes’ credibility, sub-
stance, and interconnected policy of cooperation. Therefore, disagreeing with
Lixinski, it is argued that the DAA instead contributes to the institutionalization
of legal pluralism in Europe by establishing a qualified method of cooperation and
competition between the two regimes of law, most of the latter contextualized under
the jurisdictional correlation connecting the two courts.
Finally—as a general point on the nature of equality of the DAA—Harmsen
rightly opines199 that accession of the EU to the ECHR should not be seen as an
isolated objective, but rather as the chief means by which to rearticulate
Strasbourg’s role in human rights protection in the new constitutional debate in
Europe, while increasing the democratic legitimacy of the pan-European human
rights landscape. If accession leads to privileging the EU as a High Contracting
Party at the expense of human rights protection equality within the EU, then this
should not be part of the debate. The author of this book stands in the same line of
argument by noting that accession as such would have no value if the EU becomes
privileged to the extent of harming the normal objectives of the Convention system
in relation to its contracting parties. However, it is generally argued that the extent
to which the Convention system privileges the EU is not of that scope so as to make
the Convention system not functional in the EU; nonetheless, the mere fact that
concession elements exist hinders the real intended aim of the accession process.

4.8 Chapter’s Summary of Conclusions

The chapter offered an overall examination of accession objectives, its intended


outcomes and its overall effect from the perspectives of EU constitutional law and
international law. It is concluded here that the analysis would have been incomplete
if these two perspectives were not considered dynamically and mutually. The
analysis then deconstructed each DAA article, concentrating especially on those
regulating the institutional aspects of accession—as the latter aspect will not be
examined anywhere else in the book. The chapter concludes that EU accession to
the ECHR will likely cause several transformations in EU law, the Convention

199
Harmsen (2014), p. 216.
4.8 Chapter’s Summary of Conclusions 127

system, and, most importantly, the interaction between the EU legal order and
Convention system from an international law perspective. Such transformations
will likely reflect that a more innovative system of international human rights law is
gaining momentum, with international organizations like the EU acquiring sover-
eign powers in conventionally state-owned domains like international human rights
obligations. The transformation which the accession process will engender will
likely produce a more plural and fragmented European/international law landscape,
with the multilevel regimes of law seeing constant development and functional
integration.
The chapter therefore concludes that with EU accession to the ECHR, the EU
will become one of the Masters of the EU Convention from the perspective of a
treaty-making authority in international law. Such an attribute will certainly
strengthen the EU’s stateness attitude and shape in international law, but not only
will the latter phenomenon prove of high relevance, but will also impact the internal
relationship between the EU and its Member States to the benefit of the former. On
the other hand, with the EU becoming a High Contracting Party to the Convention,
the latter will structurally alter its systemic picture, being offered to the increasing
human rights law pluralism concept. The chapter therefore concludes that EU
accession to the ECHR will not only cause certain changes in the internal and
external shape of EU, but will also have a certain cost for the Convention system
itself, the latter being for the first time opened to an international organization as a
contracting party. The chapter also examined the mode of accession, therefore it is
hereby concluded that the model of accession that the negotiators arranged—the
one with an Accession Treaty that serves simultaneously two functions, one, to
amend the Convention, and two, for the EU to accede to the Convention—was
examined as the most opportune and legally certain method.
The chapter also discussed separately the preamble and each article of the DAA,
considering their reach, nature of effect, and potential implications that their
application may pose. Certain reflection was given to the EU’s institutional partic-
ipation in the Convention system, therefore concluding that the latter has offered a
sufficiently accommodating environment to the EU—something that may go
beyond the practices of other international organizations. On the other hand, as
regarding the scope of accession, it is concluded that accession of the EU to certain
protocols—and not to all of them—may be a safer route to keep Member States
assured that accession will not directly change their position with regard to the
Convention protocols.
Finally, it is concluded that EU accession to the ECHR will provide for a new
pool of developing concepts within the pan-European human rights landscape,
something that may enlarge the spill-over effect of EU integration into the entire
European continent. Such an effect—one should conclude—will certainly be of
significance for the mode of interaction that certain self-contained regimes of law
will experience under the rule of a changing law of ‘nations’.
128 4 A New Start for the Accession of the EU to the ECHR

References

Albi A (2009) Ironies in human rights protection in the EU: pre-accession conditionality and post-
accession conundrums. Eur Law J 15(1):46–69
Alemanno A (2010) Opinion 1/08, Community competence to conclude with certain members of
the World Trade Organization Agreements modifying the schedules of specific commitments
of the community and its Member States under the General Agreement on Trade in Services.
Am J Int Law 104(3):467–474
Alter KJ (1998) Who are the “Masters of the Treaty”? European Governments and the European
Court of Justice. Int Organ 52(1):121–147
Analysis (1997) The European Community cannot accede to the European Convention on Human
Rights. Eur Law Rev 1:235–249
Balfour ADJ (2005) Application of the European Convention on Human Rights by the European
Court of Justice. Harvard Law School Student Scholarship Series. Paper 4:1–57
Balfour ADJ (2007) Eliminating conflicting interpretations of the European Convention on Human
Rights by the European Court of Justice and the European Court of human rights: the Pdiq
System as a preventative solution. Intercult Hum Rights Law Rev 2:183–247
Barents R (2009) The precedence of EU law from the perspective of constitutional pluralism. Eur
Const Law Rev 5(3):421–446
Besson S (2009) European legal pluralism after Kadi. Eur Const Law Rev 5(2):237–264
Bieber R (2009) An association of sovereign states. Eur Const Law Rev 5(3):391–406
Břı́za P (2009) The Czech Republic: the constitutional court on the Lisbon Treaty Decision of
26 November 2008. Eur Const Law Rev 5(1):143–164
Buxbaum RM (2005) A legal history of international reparations. Berkeley J Int Law 23:314–346,
Available at: http://scholarship.law.berkeley.edu/bjil/vol23/iss2/5
Callewaert J (2014) The accession of the European Union to the European Convention on Human
Rights. Council of Europe, Strasbourg
Cartabia M (2009) Europe and rights: taking dialogue seriously. Eur Const Law Rev 5(1):5–31
Conforti B (2010) Comments on the accession of the European Union to the ECHR. Italian Yearb
Int Law 20:83–86
Craig P (2013) EU accession to the ECHR: competence, procedure and substance. Fordham Int
Law J 36(5):1114–1150
Cuyvers A (2011) The Kadi II Judgment of the General Court: the ECJ’s predicament and the
consequences for Member States. Eur Const Law Rev 7(3):481–510
de Búrca G (2009) The European Court of Justice and the International Legal Order after Kadi.
Jean Monnet Working Paper 1/09. Available at http://www.jeanmonnetprogram.org/papers/09/
090101.pdf
de Búrca G (2011) The road not taken: the European Union as a global human rights actor. Am J
Int Law 105(4):649–693
de Rivery EM Chassaing E (2013), EU accession to the ECHR: bringing about an aggiornamento
for EU case law? Overview of ECHR and EU case law. e-Competitions | No 61327:1–16
de Vries SA (2013) Editorial: EU and ECHR: conflict or harmony? Utrecht Law Rev 9(1):78–79
Dijk V (2007) The accession of the European Union/European Community to the European
Convention on Human Rights, pp 2–4. Available at: http://www.statewatch.org/news/2007/
sep/van-dijk-eu-echr.pdf
Dimopoulos A (2011) The validity and applicability of international investment agreements
between EU Member States under EU and International Law. Common Market Law Rev 48
(1):63–93
Dooren V (2009) The European Union and human rights: past, present future. Merkourios
70:47–52
Douglas-Scott S (2006) Bosphorus Hava Yollari Turizm Ve Ticaret Anonim Sirketi v. Ireland,
(2006) Casenote. Common Market Law Rev 43(1):243–254
References 129

Dyevre A (2013) European Integration and National Courts: defending sovereignty under institu-
tional constraints? Eur Const Law Rev 9(1):139–168
Eckes C (2012) Protecting supremacy from external influences: a precondition for a European
Constitutional legal order? Eur Law J 18(2):230–250
Eckes C (2013) EU accession to the ECHR: between autonomy and adaptation. Modern Law Rev
76(2):254–285
Editorial (2013) Human rights: Member State, EU and ECHR levels of protection; P.S. Catalonia;
why does it take so long for my article to be published? In this issue: human rights: Member
State, EU and ECHR levels of protection. Eur J Int Law 24:471–475
Editorial Comments (1995) The aftermath of Opinion 1/94 or how to ensure the Unity of
Representation for Joint Competences. Common Market Law Rev 32:385–90
Eeckhout P (2004) External relations of the European Union. Oxford University Press, Oxford
Garcı́a RA (2002) The general provisions of the charter of fundamental rights of the European
Union. Eur Law J 8(4):492–514
Gerards J (2011) Pluralism, deference and the margin of appreciation doctrine. Eur Law J 17
(1):80–120
Gragl P (2011) Accession revisited: will fundamental rights protection trump the European
Union’s legal autonomy? In: Benedek W, Benoı̂t-Rohmer F, Karl W, Nowak M (eds)
European yearbook on human rights. Neuer Wissenschaftlicher Verlag (NWV), Wien
Gragl P (2013) The accession of the European Union to the European Convention on Human
Rights. Hart Publishing, Oxford
Greer S, Williams A (2009) Human rights in the Council of Europe and the EU: towards
‘Individual’, ‘Constitutional’ or ‘Institutional’ justice? Eur Law J 15(4):462–481
Grossman C, Bradlow DD (1993) Are we being propelled towards a people-centered transnational
legal order? Am Univ Int Law Rev 9(1):1–25
Groussot X et al (2011) EU accession to the European Convention on Human Rights: a Legal
Assessment of the Draft Accession Agreement of 14th October 2011. Fondation Robert
Schuman/European Issues No 218/7 November 2011
Hannan D (2010) The EU demands recognition as a state by the United Nations. The Telegraph
Blog. Available at: http://blogs.telegraph.co.uk/news/danielhannan/100053681/the-eu-wants-
to-be-treated-as-a-state-by-the-united-nations/
Harding C (2000) The identity of European law: mapping out the European legal space. Eur Law J
6(2):128–147
Harmsen R (2014) The (geo-)politics of EU accession to the ECHR: democracy and distrust in the
wider Europe. In: Kosta V et al (eds) The EU accession to the ECHR. Hart Publishing, Oxford
Hart JW (2010) The European human rights system. Law Libr J 102(4):533–559
Heifer LR, Slaughter A-M (1997) Toward a theory of effective supranational adjudication. Yale
Law J 107:273–392
Hoffmeister F (2007) Outsider or frontrunner? Recent developments under international and
European law on the status of the European Union in international organizations and treaty
bodies. Common Market Law Rev 44:41–68
Jacque JP (2011) The accession of the European Union to the European Convention on Human
Rights and Fundamental Freedoms. Common Market Law Rev 48(4):995–1023
Jones C (2012) State watch analysis: the EU’s accession to the European Convention on Human
Rights: a cause for celebration or concern? Available at: http://www.statewatch.org/analyses/
no-187-echr.pdf, pp 5–6
Joris T, Vandenberghe J (2008–2009) The Council of Europe and the European Union: natural
partners or uneasy bedfellows? Columbia J Eur Law 15:1–43
Kokott J, Sobotta C (2012) The Kadi Case – constitutional core values and international law –
finding the balance? Eur J Int Law 23(4):1015–1024
Komárek J (2013) The place of constitutional courts in the EU. Eur Const Law Rev 9(3):420–450
Krüger HC (2002–2003) Reflections concerning accession of the European communities to the
European Convention on Human Rights. Pa State Int Law Rev 21(1):89–99
130 4 A New Start for the Accession of the EU to the ECHR

Kuiper PJ (1995) The conclusion and implementation of the Uruguay Round results by the
European Community. Eur J Int Law 6(1):222–244
Kumm M (2005) The jurisprudence of constitutional conflict: constitutional supremacy in Europe
before and after the Constitutional Treaty. Eur Law J 11(3):262–307
Landau EC (2008) A new regime of human rights in the EU? Eur J Law Reform 10(4):557–575
Lavranos N (2006a) MOX Plant Dispute – Court of Justice of the European Communities. Eur
Const Law Rev 2(3):456–469
Lavranos N (2006b) The MOX Plant and IJzeren Rijn disputes: which court is the supreme arbiter?
Leiden J Int Law 19(1):223–246
Lavranos N (2008) Regulating competing jurisdictions among international courts and tribunals.
Za€oRV 68:575–621
LB and JHR (2008) The relative autonomy of the EU human rights standard. Eur Const Law Rev 4
(2):199–204
Lenaerts K, de Smijter E (2001) The charter and the role of the European Courts. Maastrich J Eur
Comp Law 8(1):90–101
Lenart J (2012) ‘Fortress Europe’: compliance of the Dublin II Regulation with the European
Convention for the protection of human rights and fundamental freedoms. Utrecht J Int Eur
Law 28(75):04–19
Licková M (2008) European exceptionalism in international law. Eur J Int Law 19(3):463–490
Liisberg JB (2002) Does the EU Charter of fundamental rights threaten the supremacy of
Community law? Common Market Law Rev 38(5):1171–1199
Lixinski L (2014) Taming the fragmentation monster through human rights? International consti-
tutionalism, ‘Pluralism Lite’, and the common territory of the two European Legal Orders. In:
Kosta V et al (eds) The EU accession to the ECHR. Hart Publishing, Oxford
Lock T (2010) EU accession to the ECHR: implications for judicial review in Strasbourg. Eur Law
Rev 35(6):777–798
Lock T (2012) End of an epic? The Draft Agreement on the EU’s accession to the ECHR. Yearb
Eur Law 31(1):162–197
o€
L€ of R (2006) Shooting from the hip: proposed minimum rights in criminal proceedings through-
out the EU. Eur Law J 12(3):421–430
Martin FF (2004) Our constitution as federal treaty: a new theory of United States Constitutional
construction based on an originalist understanding for addressing a new world. Hastings
Constit Law Q 31:268–354
Martı́n J, De Nanclares P (2013) The accession of the European Union to the ECHR: more than
just a legal issue. Working Papers on European Law and Regional Integration WP IDEIR no
15. Instituto de Derecho Europeo e Integracion Regional (IDEIR), Universidad Complutense
Mendez M (2010) The enforcement of EU agreements: bolstering the effectiveness of treaty law?
Common Market Law Rev 47(6):1719–1756
Miiller CA (2007) Fundamental Rights in multi-level legal systems: recent developments in
European human rights practice. Interdiscip J Hum Rights Law 2(1):33–60
Morijn J (2006) Balancing fundamental rights and common market freedoms in union law:
Schmidberger and Omega in the light of the European Constitution. Eur Law J 12(1):15–40
Muir E (2014) The fundamental rights implications of EU legislation: some constitutional
challenges. Common Market Law Rev 51(1):219–246
Nanopoulos E (2013) Trust issues and the European Common Asylum System: finding the right
balance. Camb Law J 72(2):276–280
Neframi E (2010) The duty of loyalty: rethinking its scope through its application in the field of EU
external relations. Common Market Law Rev 47(2):323–359
Neuman GL (2008) Import, export, and regional consent in the inter-American Court of human
rights. Eur J Int Law 19(1):101–123
O’Cinneide C (2009) Human rights and within multi-layered systems of constitutional gover-
nance: rights cosmopolitanism and domestic particularism in tension. UCD Working Papers in
Law, Criminology & Socio-Legal Studies. Research Paper No. 12/2009
References 131

O’Meara N (2011) “A More Secure Europe of Rights?” The European Court of Human Rights, the
Court of Justice of the European Union and EU accession to the ECHR. German Law J 12
(10):1813–1832
Odermatt J (2014a) The EU’s accession to the European Convention on Human Rights: an
international law perspective. Leuven Center for Global Governance Studies. Working Paper
No. 136, p 15
Odermatt J (2014b) The EU’s accession to the European Convention on Human Rights: an
international law perspective. ExpressO. Available at: http://works.bepress.com/jed_
odermatt/1
Olsen JB (2009–2010) Protecting Fundamental Rights and the evolving roles of the Court of
Justice of the European Union and the European Court of human rights; Europeanisation in
action at the supranational level. Edinburgh Stud Law Rev 1(3):54–73
Papangelopoulou E (2006) Introductory note to European Court of Human Rights (ECHR) Grand
Chamber: Bosphorus Hava Yollari Turizm Ve Ticaret Anonim Sirketi v. Ireland. Int Legal
Mater 45(1):133–135
Pauwelyn J (2003) Conflict of norms in public international law. Cambridge University Press,
Cambridge
Payandeh M (2011) Constitutional review of EU law after Honeywell: contextualizing the
relationship between the German Constitutional Court and the EU Court of Justice. Common
Market Law Rev 48(1):9–38
Peers S (2006) Bosphorus – European Court of human rights. Eur Const Law Rev 2(3):443–455
Peers S (2013) Towards a new form of EU law? The use of EU Institutions outside the EU legal
framework. Eur Const Law Rev 9(1):37–72
Pernice I, Kanitz R (2004) Fundamental rights and multilevel constitutionalism in Europe. White
Paper 7/04, Humboldt University, p 20
Petersmann E-U (2002) Constitutionalism and WTO law: from a state-centered approach towards
a human rights approach in international economic law. The political economy of international
trade law, 1st edn. Cambridge University Press, Cambridge, pp 32–67
Quinn G (2001) The European Union and the Council of Europe on the issue of human rights:
twins separated at birth? McGill Law J 46(849–874):853
Quirico O (2010) Substantive and procedural issues raised by the accession of the EU to the
ECHR. Italian Yearb Int Law 20:31–53
Raba K (2013) The accession of the European Union to the European Convention for the
protection of human rights and fundamental freedoms—overview of the accession agreement.
ERA Forum 14:557–572
Reestman JH (2005) Primacy of union law. Eur Const Law Rev 1(1):104–107
Ribble MR (2010–2011) I Don’t Trust Your Judgment: the European Convention on Human
Rights meets the European Union on new grounds? Pa State Int Law Rev 29(1):211–232
Roberts A, Sivakumaran S (2012) Lawmaking by nonstate actors: engaging armed groups in the
creation of international humanitarian law. Yale J Int Law 37(1):108–152
Ryngaert C (2011) ‘The European Court of Human Rights’ approach to the responsibility of
Member States in connection with acts of international organizations. Int Comp Law Q 60
(4):997–1016
Sabel CF, Gerstenberg O (2010) Constitutionalising an overlapping consensus: the ECJ and the
emergence of a coordinate constitutional order. Eur Law J 16(5):511–550
Schimmelfennig F (2006) Competition and community: constitutional courts, rhetorical action,
and the institutionalization of human rights in the European Union. J Eur Public Policy 13
(8):1247–1264
Sera JM (1996) The case for accession by the European Union to the European Convention for the
protection of human rights. Boston Univ Int Law J 14:151–186
Shelton D (2003) The boundaries of human rights jurisdiction in Europe. Duke J Comp Int Law
13:95–153
132 4 A New Start for the Accession of the EU to the ECHR

Smith M (2007) The European Union and International Order: European and Global Dimensions.
Eur Foreign Aff Rev 12(4):437–456
Sweet AS (2009) Constitutionalism, legal pluralism, and international regimes. Indiana J Global
Legal Stud 16(2):621–645
Tamanaha BZ (2008) Understanding legal pluralism: past to present, local to global. Sydney Law
Rev 30:374–411
Thym D (2013a) A Trojan Horse? Challenges to the primacy of EU law in the Draft Agreement on
accession to the ECHR. VerfBlog, 2013/9/11, http://www.verfassungsblog.de/en/atrojan-
horse-challenges-to-the-primacy-of-eu-law-in-the-draft-agreement-on-accession-to-the-echr/,
p3
Thym D (2013b) Separation versus Fusion – or: how to accommodate national autonomy and the
charter? Diverging visions of the German Constitutional Court and the European Court of
Justice. Eur Const Law Rev 9(3):391–419
Tulkens F (2013) EU accession to the European Convention on Human Rights. Speech at: National
School of Judiciary and Public Prosecution (KSSIP) – Krakow (Poland), Friday 1 March 2013
van den Berghe F (2010) The EU and issues of human rights protection: same solutions to more
acute problems? Eur Law J 16(2):112–157
van Rossem JW (2009) Interaction between EU law and international law in the light of Intertanko
and Kadi: the dilemma of norms binding the Member States but not the Community. Nether-
lands Yearb Int Law 40:183–227
von Bogdandy A (2012) The European lesson for international democracy: the significance of
Articles 9 to 12 EU Treaty for International Organizations. Eur J Int Law 23(2):315–334
Von Bogdandy A et al (2012) Reverse solange – protecting the essence of fundamental rights
against Eu Member States. Common Market Law Rev 49(2):489–519
Weiß W (2011) Human Rights in the EU: rethinking the role of the European Convention on
Human Rights after Lisbon. Eur Const Law Rev 7(1):64–95
Wetzel JR (2003) Improving fundamental rights protection in the European Union: resolving the
conflict and confusion between the Luxembourg and Strasbourg Courts. Fordham Law Rev
71:2823–2862
White S (2010) The EU’s accession to the convention on human rights: a new era of closer
cooperation between the Council of Europe and the EU? New J Eur Crim Law 1(4):433–46
Wildhaber L (2002) A constitutional future for the European Court of human rights. Hum Rights
Law J 23:161–165
Chapter 5
Status of ECHR and DAA in EU Legal Order

5.1 Introduction

The relationship between the Luxembourg and Strasbourg courts, EU law and the
Convention system at-large are fundamentally conditioned by the legal status of the
Convention and Accession Agreement in EU law. This chapter offers a novel
examination of the status of the Convention and the DAA in EU law post-accession,
analyzing not only the overall picture of cohabitation and/or competition but also
the internal effect that accession will likely pose to the EU legal order (in relation to
the Convention and the DAA). This chapter responds to the research question:
‘What is the legal status of the ECHR and DAA in the EU legal order post-
accession?’ The chapter starts by examining the position and mode of penetration
of international agreements in EU legal order, giving special emphasis to the effect
that such agreements produce within the EU legal order. Haegeman, Commission
v. Germany, Demirel and Bananas are examined to deconstruct Luxembourg’s
stance on international agreements concluded by the EU, either alone or together
with Member States. A careful deconstruction of Luxembourg’s jurisdiction to
observe those agreements is also provided, showing how similarities may be
drawn with regard to both the Convention and Accession Agreement. Then, the
chapter embarks on an examination of the status of the Convention and the DAA,
contrasting them with current Luxembourg benchmarks on international agree-
ments. Arguing that neither the Convention nor the DAA form part of the usual
international agreements which the EU has concluded, the chapter goes on to
provide a novel analysis of the status of Convention and the DAA in the EU legal
order arguing that this should be taken as a very unique case. The chapter then
provides a comprehensive analysis on the specific status of the Convention and the
DAA post-accession, examining both their rank and implied position within a
multilayered system of human rights, while considering potential implications.
To examine carefully both the status of the Convention and the DAA post-
accession, this chapter provides a systemic interpretation of Art 6 (1) TEU, Art

© Springer International Publishing Switzerland 2015 133


F. Korenica, The EU Accession to the ECHR, DOI 10.1007/978-3-319-21759-8_5
134 5 Status of ECHR and DAA in EU Legal Order

6 (2) TEU, Art 6 (3) TEU, Art 218 (8) TFEU and Protocol 8, combined with the
case-law cited beforehand. The chapter tries not only to interpret the positive
meaning of the Treaty-based prescriptions on this issue but also the negative
meaning of those provisions, and the practical implicit situations which will
imply a hidden understanding for those two acts’ status in EU legal orders once
the Accession Agreement is concluded. Five distinct arguments are provided in this
chapter to support these claims. Although this chapter goes rather far in
deconstructing the systemic picture that portrays the status of the Convention and
the DAA in the EU legal order, it simultaneously draws attention to the fact that a
distinguished explanation should be innovated to construe this relationship more
thoroughly and clearly. There is no dispute over the fact that the Convention and the
DAA must be set as being distinguished even in view of the EU Treaties and post-
accession implied practice. The chapter then concludes that the Convention and the
DAA will enjoy a rather distinguished position within EU law, something that may
serve as ground for a new understanding of the EU Treaties’ setting, and may play a
core role on the new relationship between the EU and Convention regimes of law.

5.2 Examining the Status of ECHR and DAA in EU Legal


Order

5.2.1 The Status of International Agreements in EU Legal


Order

Before delving into the more advanced question of the effects of accession in the
EU legal order, one should consider the question of the Convention’s status within
the hierarchy of the EU legal order. This is an essential question in the relationship
between the two orders, but also between the two courts. The issue therefore poses
the question of whether the Convention penetrates the EU legal order, its hierar-
chical status, nature of effect and applicability, and the more general legal outcomes
of such a result. In general, this question leads to the mode of regulation of the
relationship between international law and EU municipal law, bearing in mind
special features of the Convention.
A basic argument is that the issue of the relationship between national law
(in this case, the EU law) and international law should be governed by the
international law itself.1 On the other hand, the prevalent practice is that it is
usually constitutional law or national law at large that should and that usually
regulates the relationship concerned.2 One must however acknowledge that the

1
Lauterpacht, as an example, stands in this line, arguing that international law, not constitutional
law, should govern the relationship concerned. See: Lauterpacht in Maniruzzaman (2001), p. 311
et seq.
2
For the second part of the sentence, see for more details a very good book, at: Cassese (1985). For
the first part of the argument, please read: Shaw (2008), p. 129; See also: Gragl (2013), p. 97.
5.2 Examining the Status of ECHR and DAA in EU Legal Order 135

presence of international law in the municipal law structure is growing exponen-


tially in the last decades,3 with EU law generally following the same track. The
issue of the relationship between international law and national law mainly circles
around the question of whether international treaties penetrate the national legal
order, the position that they assume in the hierarchy of national law, and the effect
of such treaties on domestic law.4 In practice, the issue of the relationship between
international law and national law, in our case between international law and EU
law, is located in one of the theoretical models, namely monism and dualism.
The question of whether the EU has authority to enter into certain treaty regimes
has been debated since the Maastricht Treaty.5 With the EU today having a Treaty-
prescribed international legal personality within the competences conferred to it by
the Member States in the Treaties, as defined in Art 47 TEU, it is important to
initially review the status of international treaties in the EU legal order. In this
context, international agreements6 concluded by the EU, as decided in Haegeman,7
form part of EU law.8 This speaks for a rather monist relationship model between
international agreements and the EU legal order,9 as concluded agreements do
automatically penetrate in the EU legal order.10 Although there is no doubt that
international agreements concluded by the EU directly penetrate and form part of
the EU legal order,11 one should question the hierarchical status that such

3
See a very good contribution on this topic, at: Wilson (1964).
4
See a general study on this, at: La Pergola (1993).
5
See why this issue should not anymore be so relevant, at: Eeckhout (2004), p. 155.
6
The EU currently is part of around 90 international agreements. See on this: Wessel
(2008), p. 152.
7
Court of Justice of EU, Haegeman, Case 181/73 [1974] ECR 449, judgment of 30 April 1974,
para 5; See also: Court of Justice of EU, Hauptzollamt Mainz ν Kupferberg, Case 104/81 [1982]
ECR 3641, paragraph 13 (‘In ensuring respect for commitments arising from an agreement
concluded by the Community Institutions the Member States fulfil an obligation not only in
relation to the non-member country concerned but also and above all in relation to the Community
which has assumed responsibility for the due performance of the agreement’).
8
De Londras and Kingston (2010), p. 367; Including customary international law, with a rule
deriving from VCLT as the case. See on the latter: Court of Justice of EU, Racke v Hauptzollamt
Mainz, Case C-162/96, judgment of 16 June 1998, paragraph 46-8; Court of Justice of EU, R & V
Haegeman v Belgian State, Case 181/73 [1974] ECR 449, para. 5; Court of Justice of EU,
Andersson and Wåkerås-Andersson, Case C-321/97 [1999] ECR I-3551; See also: Mendez
(2010), pp. 1721–1723; Similarly, see: van Rossem (2009), p. 187; Some insist that the Universal
Declaration of Human Rights has become a central part of customary international law in this
regard. See e.g.: Rodley (1989).
9
E.g.: Quirico (2010), p. 34.
10
Klabbers therefore rightly argues that ‘Haegeman [is seen] as a manifesto of monism, pointing
out that to claim that treaties form an integral part of the legal system is typically an exercise in
monism.’ Klabbers (2002), p. 277.
11
AG Gulmann, in this context, had argued that: ‘That was particularly emphasized in the Court’s
judgment in Case 104/81 Kitpferberg, in which the Court stated that the effects in the Community
of the provisions of an international agreement cannot be determined without account being taken
of the origin in international law of its provisions and that it follows from the principles of
international law that the contracting parties may determine what effects the provisions of the
136 5 Status of ECHR and DAA in EU Legal Order

agreements hold within the EU legal order. In light of Commission v. Germany,12


the Luxembourg Court has determined that agreements concluded by the EU are
located between primary and secondary law of the EU.13 With this decided, the
secondary law of the EU must be compatible with the agreements concluded by the
EU,14 however not ruling out mandatory consistency between the concluded inter-
national agreements and the primary law of the EU. It is also important to mention
that the Luxembourg Court has clearly set that the level of implementation of
international agreements concluded by the EU should not vary according to
whether the implementation of such agreements is within EU or Member States
competence,15 or vary due to Member States different national rules on interna-
tional agreements position and interpretation.16 The uniform application of such
international agreements comes therefore under the authority of the Court, on the
basis of the law that it observes. It should be highlighted that in light of Demirel,
such agreements are self-executing in the EU legal order if the wording and purpose
of such agreements are clear enough to be understood as not requiring a subsequent
legal measure for their implementation,17 an issue which must be properly evalu-
ated rather than presumed.18 It is important to note that the EU Court uses a very
simplistic standard on titling international agreements—as it had held in Opinion

agreement are to have in their internal legal order.’ (Opinion Of Advocate General Gulmann in
C-280/93, Germany v Council. Opinion delivered on 8 June 1994, para. 127). With a similar view
on the Convention, it is needs be discussed whether the latter must enjoy a special status in the EU
legal order, something that is predetermined by its origin in the terms of Kitpferberg case; Cf.:
Gragl in this regard argues that with the Convention becoming part of EU law, the Luxembourg
Court will interpret it within its jurisdiction to observe the law. Such fact will ensure that the
Convention becomes first interpreted and applied in individual cases by the national court of the
contracting party, the EU, and thereafter if a case of alleged violation would arise, by the
Strasbourg Court. This fact, according to Gragl, perfectly maintains the principle of subsidiarity
of the Convention system, as it allows the Convention to become part of the law that will be
considered from within against potential cases before the Strasbourg Court will have a chance to
assess the case from its external perspective. See: Gragl (2013), p. 265.
12
Court of Justice of EU, Commission of the European Communities v. Federal Republic of
Germany, Judgment, Case C-61/94, para 52.
13
See also a logical conclusion on this, at: Court of Justice of EU, Commission of the European
Communities v Federal Republic of Germany (International Dairy Arrangement), Case C-61/94
[1996] ECR I-3989, para. 52; See also: Court of Justice of EU, Agrover Srl v Agenzia Dogane
Circoscrizione Doganale di Genova, Case C-173/06 [2007] ECR I-8783, para. 17.
14
See also: Opinion of AG Tesauro in case Commission v. Germany, C-61/94, delivered on
7 May 1996.
15
Court of Justice of EU, Case 104/81, Hauptzollamt Mainz ν Kupferberg [1982] ECR 3641,
judgment. Para. 14; Cf. Kapteyn (1974), p. 74.
16
Opinion of Mr. Advocate General Darmon, Demirel v. Stadt Schwäbisch Gmünd. C-12/86,
delivered on 19 May 1987.
17
Court of Justice of EU, Demirel v. Stadt Schw€ abisch Gm€ und, Case 12/86, Judgment Of The
Court, 30 September 1987.
18
Opinion of Mr. Advocate General Darmon, Demirel v. Stadt Schwäbisch Gmünd. C-12/86,
delivered on 19 May 1987.
5.2 Examining the Status of ECHR and DAA in EU Legal Order 137

1/75—by ruling that it accepts an act as an international agreement if it is simply an


‘undertaking entered into by entities subject to international law which has binding
force, whatever its formal designation’.19
It clearly follows from Art 216 (2) of TFEU that agreements concluded by the
EU are binding upon its institutions and Member States, adding that the EU is
obliged to implement them bona fidei.20 With this being noted,21 the failure of a
Member State to implement an agreement concluded by the EU would not only be a
violation of the agreement concerned but also of the Art 216 (2) TFEU. The same
would apply to the situation wherein an EU Member State infringes upon the
Convention, an act which would simultaneously breach EU law if the act or
measure which the Member State performed originated from their obligations
under the Union. Therefore, international agreements concluded by the EU (includ-
ing the Convention)22 form part of the law observed and uniformly interpreted by
the Luxembourg Court. In cases of international agreements that are concluded
separately and individually by the EU and by Member States, such as the ECHR, it
would follow from the interpretation of Art 216 (2) that Member States’ breach of
such agreements would violate both the relevant agreement and the TFEU. On the
other hand, a logical consideration of Costa/ENEL would also extend the principle
of supremacy of EU-law-penetrated international agreements upon the legal orders
of EU Member States. In view of this, Gragl notes that the internal effect of
international agreements is made to assure the ‘external projection’ of EU law
primacy,23 therefore better guaranteeing the uniformity of EU legal order as a
whole.
A key question that arises in the context of penetration of international agree-
ments in the EU legal order is whether the Luxembourg Court has jurisdiction to

19
Court of Justice of EU, Opinion 1/75 (OECD Local Cost Standard) [1975] ECR 1355, p. 1359/
60.
20
Court of Justice of EU, Portugal v. Council, Case C-149/96, Judgment of the Court,
23 November 1999. para 35; See also: Court of Justice of EU, Anklagemyndigheden v Peter
Michael Poulsen and Diva Navigation Corp, Case C-286/90 [1992] ECR I-6019, para. 9; Gragl
(2013), p. 98/9.
21
To note, the EU Court has ruled in Demirel that an international agreement in this sense needs to
be ‘sufficiently precise and unconditional’ on its stipulations in order to be of direct effect. (Court
of Justice of EU, Case C-12/86 Demirel [1987] ECR 3719, para. 23; emphasis added); On the
latter, see also: Court of Justice of EU, Sevince, Case C-192/89 [1990] ECR 3461, para. 14 et seq;
Court of Justice of EU, Simutenkov, Case C-265/03 [2005] ECR I-2579; On the direct effect of
international agreements of EU in Member States’ legal orders, see e.g.: Lenaerts and Corhaut
(2006); On some exceptions on direct effect of international agreements in EU legal order, see
also: Court of Justice of EU, Foster and others v British Gas plc, Case C-188/89 [1990] ECR
I-3313, para 18.
22
See e.g.: Lock (2012), p. 190; See also: Gragl (2013), p. 99; Contra, a previous and old case of
the EU Court exhibiting ECHR as a supporting tool of interpretation rather than an incorporated
part of EU law (in a pre-accession scenario), see: Court of Justice of EU, Mayr-Melnhof
Kartongesellschaft mbH v Commission of the European Communities, Case T-347/94 [1998]
ECR II-1751, para. 311/2.
23
Gragl (2013), p. 98.
138 5 Status of ECHR and DAA in EU Legal Order

observe the law of international agreements in each case. This question arose in
Demirel, wherein a Member State had claimed that the jurisdiction of the Court
under Art 19 (1) TEU to ensure the uniform interpretation and observance of the
law does not extend to mixed international agreements, like the ECHR. The Court
ruled in Demirel24 that it has jurisdiction to observe the law of international
agreements that form part of EU law: both with regard to international agreements
that fall in the exclusive competence of the EU and mixed international agree-
ments.25 Member States’ duty to implement those agreements (such as the Con-
vention) does not merely form part of the principle of primacy of EU law wherein
the international agreement originates, but also of the duty of loyal cooperation
deriving from the Treaties.26 In Bananas,27 however, the Court has rightly stressed
that it is within the nature and objectives of the concluded international agreement
the basis upon which one could request the review of legality of a Community act in
terms of its compliance with that international agreement, adding that ‘the juris-
diction of the Court cannot be limited by the grounds on which the validity of those
measures may be contested.’28 That being the result, the Convention becomes
embedded as an act of EU law—embedded in EU law primacy as well—therefore
becoming superior also in relation to national constitutions, something which the
Convention might have not been able to impose/institute before accession.29 On the
other hand, ‘the judgments of the ECtHR take effect within the binding force and
the primacy and direct effect of EU law upon the domestic legal orders.’30
The Luxembourg Court therefore has Art 19 (1) TEU wide-ranging jurisdiction
to ‘examine whether their validity may be affected by reason of the fact that they are
contrary to a rule of international law.’31 There are two conditions to examine as to
whether a secondary legislation act of the EU could be invalidated by a rule of a
concluded international agreement: first, that the Community should be bound by

24
Court of Justice of EU, Demirel v. Stadt Schw€ abisch Gm€und, Case 12/86, Judgment, CJEU,
30 September 1987, paras. 8–9.
25
In this context, similarly, the Luxembourg Court has ruled in Court of Justice of EU, Commis-
sion v. France (Case C-239/03, judgment, para. 25), that: ‘In accordance with case-law, mixed
agreements concluded by the Community, its Member States and non-member countries have the
same status in the Community legal order as purely Community agreements in so far as the
provisions fall within the scope of Community competence.’ [emphasis added].
26
Neframi (2010), p. 331.
27
Court of Justice of EU, Germany v. Council, Case C-280/93, Judgment of the Court, 5 October
1994, para 111.
28
Court of Justice of EU, International Fruit Company v Produktschap voor Groenten en Fruit,
Joined Cases 21 to 24/72 [1972] ECR 1219, para 5.
29
See e.g.: Quirico (2010), p. 33; Lock (2012), p. 191; See also: De Schutter (2007), p. 13.
30
Weiß (2011), p. 91; On direct effect of international agreements, see also: Court of Justice of
EU, Texaco A/S v Middelfaert Havn and others, Joined Cases C-114/95 and C-115/95 [1997] ECR
I-4263.
31
Court of Justice of EU, International Fruit Company, Joined Cases 21 to 24/72 [1972] ECR
1219, note 28, para 6.
5.2 Examining the Status of ECHR and DAA in EU Legal Order 139

that agreement, and second, the provision in question of the concerned agreement
should be capable of conferring rights on the citizens of the EU which they could
invoke before national courts.32 It should be noted that Art 275 TFEU exempts such
jurisdiction from the area of foreign and security policy, except cases when Council
decisions limit rights of legal and natural persons deriving from the TEU.
In this context, it is important to locate both the Convention and the Accession
Agreement of the EU to the ECHR within the legal hierarchy of the EU legal order.
Upon identifying the hierarchical position of the Convention and the DAA within
the EU legal order,33 one would be able to question the impact and effects of the
Convention on the overall relationships between primary and secondary law and
also between EU law and national law of the Member States.
There appear to be two important characteristics that need be considered before
embarking on an examination of the more analytical dimension of the Convention’s
status in the EU legal order. First, the Convention enjoys a rather distinctive
historical and functional standing in the law of the EU34 and case-law of the
Luxembourg Court,35 and also within the EU integration process at-large; this
examination should be considered significant, as it could quite well shape the
difference between the Convention and all other international agreements wherein
the EU is a party. Second, the Convention has a rather special legal dimension
prescribed to it by the Treaties, a fact that needs to be considered as having
important meaning in the context of interpretation of its legal status in the EU

32
Court of Justice of EU, International Fruit Company, Joined Cases 21 to 24/72 [1972] ECR
1219, note 28, paras. 7 & 8.
33
Cf.: Gragl (2013), p. 105–108, who argues that the fact that the Convention will be duplicated—
once by becoming incorporated into EU law and once through its previous incorporation into
Member States’ law—may produce interesting and conflicting situations. Gragl rightly notes that
the EU-law incorporated Convention will comprise part of the principle of primacy of EU law,
therefore take precedence over national laws of member states. On the other hand, the incorporated
Convention at the level of national law will be subsumed to the EU-law principle of supremacy
(which includes the Convention as well). In this situation, the Luxembourg Court’s decisions on
the Convention will take primacy over potential Strasbourg Court decisions on national law of
Member States; Another scenario, according to Gragl, would be if Member States’ obligations
under the Convention and EU law were to clash. In such a scenario, the Member States would first
be obliged to the Luxembourg Court and only after that to the Strasbourg Court. In such scenario,
Gragl proposes that the national court should pose a preliminary reference question and leave it to
the EU Court to decide how to tackle those conflicting situations. See: Gragl (2013), pp. 108–110.
Gragl’s latter scenario supposes that the conflicting obligation stems on the one hand from EU
secondary law, and on the other, from the Convention. Gragl therefore finds the answer on the
preliminary reference procedure which may invalidate the secondary EU law act and therefore
possibly defer to the Convention, therefore resolving the potential conflict. Gragl’s answer,
however, does not work if the Member States’ conflicting obligations were to stem from the EU
primary law and the Convention, in which case the EU Court would have no choice but to violate
the Convention and defer to its Treaties, therefore pulling the Member States with it in that terrain
of violation (which member states would need to commit due to the principle of primacy).
34
E.g.: Weiß (2011), p. 65.
35
E.g.: van Rossem (2009), p. 208.
140 5 Status of ECHR and DAA in EU Legal Order

legal order. It should nonetheless be noted that the responsibility of the EU to


respect the Convention stems not merely from its treaty obligations but also from its
international law obligations.36
As importantly noted, however, Jacobs argued that the EU Court has been
growing in its friendly approach towards the Convention even though EU was not
a party to it.37 From a post-accession outlook, therefore, there are two issues that
need be considered separately: first, the Convention’s status in the EU legal order,38
and second, the DAA’s status in the EU legal order.39 Both issues are intertwined,
however there might be diverging answers as to their status. While one refers to the
status of the Convention and the DAA in the legal order, the point is made to answer
four possible questions: first, the hierarchical position of the Convention and the
Accession Agreement in the hierarchy of legal acts in EU law; second, the rela-
tionship between the Convention and the Accession Agreement to other levels of
law in EU legal order, such as the relationship between the Convention and EU
secondary law; third, the indication and effect of the Convention and the Accession
Agreement’s status in the Member States legal orders; and, fourth, the Convention
and Accession Agreement’s effect on both persons having an interest in EU law and
those having an interest in Member States’ laws. Before going into a thorough
examination, it is important to note that it may be accepted only exceptionally—but
also from a very general point of view—that the Convention is part of the EU
primary law.40 As a caveat, in addition, with its Opinion 2/94, the Luxembourg
Court had warned that the accession of the EU to the ECHR ‘would be of
constitutional significance’,41 pointing out that the process will bring primary law

36
Odermatt (2014), p. 12 & 36; Cf.: On the obligation of every entity of international law to
respect international agreements, see: International Court of Justice, Cameroon v Nigeria, Pre-
liminary Objections Judgment, ICJ Reports, 1998, para. 38 (‘The Court observes that the principle
of good faith is a well-established principle of international law.’); On the latter, see also: The 1970
Declaration on Principles of International Law Concerning Friendly Relations and Co-operation
among States. UN General Assembly resolution 2625 (XXV)—October 1970.
37
Jacobs (2008), p. 31.
38
In this regard, see: See also: Court of Justice of EU, Åklagaren v Hans Åkerberg Fransson, Case
C-617/10 [2013] para. 44: ‘[. . .]it is to be remembered that whilst, as Article 6(3) TEU confirms,
fundamental rights recognised by the ECHR constitute general principles of the European Union’s
law and whilst Article 52(3) of the Charter requires rights contained in the Charter which
correspond to rights guaranteed by the ECHR to be given the same meaning and scope as those
laid down by the ECHR, the latter does not constitute, as long as the European Union has not
acceded to it, a legal instrument which has been formally incorporated into European Union law.’
[emphasis added].
39
Cf.: Gragl notes that neither the DAA nor the Convention make any implicit or explicit mention
of the status that they will have in the EU legal order post-accession. See: Gragl (2013), p. 97.
40
Morijn (2006), p. 21.
41
Court of Justice of EU, Opinion 2/94, para. 34/5.
5.2 Examining the Status of ECHR and DAA in EU Legal Order 141

implications and/or modifications. The need to make an in-depth analysis of this


issue is therefore highly recommended, as seen infra.

5.2.2 Analyzing the Hierarchical Position of the Convention


and Accession Agreement in EU Legal Order

There appear to be four fundamental Treaty bases regarding the analysis of the
position of the Convention and the DAA in the EU legal order, namely, Art
6 (1) TEU, Art 6 (2) TEU, Art 6 (3) TEU,42 Art 218 (8) TFEU and Protocol
8. Some of these treaty bases may even conflict with each other.43 In this
subchapter—as a short digression—one needs to start with the examination of a
suggestion coming from the Permanent Court of International Justice which sup-
ports the method applied here; the Permanent Court basically arguing that ‘it is
obvious that [. . .a. . .] Treaty must be read as a whole, and that its meaning is not to
be determined merely upon particular phrases which, if detached from the context,
may be interpreted in more than one sense.’44 We will follow this perspective
exactly. Therefore, a systemic analysis of these four Treaty foundations regarding
the position of the Convention and the DAA in the EU legal order is mandatory,
building upon many court cases that have shed light on such issues. Although it
would seem that the position of the Convention and the DAA are not well explained
theoretically, a complete and complex practical deconstruction of such Treaty
articles would help to construe a model of explanation relating to the legal status
of these two legal acts. With this burden in mind, it is necessary not only to engage
in the positive interpretation of such articles, what such articles say, but to engage
also with the negative interpretation of such articles, connoting to what the articles
do not say. This is made with the intention of saving the analysis from a ‘chicken-
and-egg’ dilemma.45
The most noteworthy article is Art 6 (2) TEU which lays down the duty of the
EU to accede to the Convention, by setting that: ‘The Union shall accede to the
European Convention for the Protection of Human Rights and Fundamental Free-
doms. Such accession shall not affect the Union’s competences as defined in the
Treaties.’ Art 6 (2) TEU is therefore the Treaty-based duty for the EU to accede to

42
Cf.: Morano-Foadi and Andreadakis (2011), p. 600, who argues that there may be no hierarchy
between the Charter, ECHR and general principles of law deriving from Member States’ consti-
tutional traditions, something that suggests a primary law function for all of them; See a rather
similar consideration at: Muir (2014), p. 219; These three articles that were devised with the
Lisbon Treaty are seen as enhancing the overall human rights protection in the Union, at:
Carruthers (2009), p. 804.
43
E.g.: Hofmann and Mihaescu (2013), p. 73.
44
Permanent International Court of Justice, Advisory Opinion on the Competence of the ILO to
Regulate Agricultural Labour, P.C.I.J. (1922), Series B, No. 2/3, p. 23.
45
van Rossem (2009), p. 209.
142 5 Status of ECHR and DAA in EU Legal Order

the Convention,46 supposing that the latter agrees to such accession. One must
argue that with the prescription of Art 6 (2) TEU, the EU is granted Treaty-based
competence (which it lacked earlier),47 permission,48 and duty to accede to the
Convention,49 clearly making constitutional space for such accession and fulfilling
the request of Opinion 2/94. This is all made under the assumption that the EU
should be considered as possessing the pouvoir constituent to enter the domain of
human rights law under the Convention system,50 if not, then this provision pro-
vides it for the EU. This ‘amounts to an important step forward in the progressive
process of jurisdictional development’51 for Union and Europe as a whole. Digging
further into Art 6 (2) TEU would lead to four basic observations which derive from
the legal construction of it: first, that there is a Treaty space for the EU’s accession
to the Convention; second, that there is simultaneously an obligation for the EU to
accede to the Convention52; third, that both the space and the duty concerned are
located in the primary law of the EU; and, fourth, the accession concerned is linked
to/bound to the approval of the Masters of the Treaty that it have constitutional
connotation.
Next, the second most noteworthy provision is Art 6 (3) TEU which supplies a
Treaty-based expression for the general rank of the Convention in the EU legal
order, by setting that: ‘Fundamental rights, as guaranteed by the European Con-
vention for the Protection of Human Rights and Fundamental Freedoms and as they
result from the constitutional traditions common to the Member States, shall
constitute general principles of the Union’s law.’53 It is explicitly stated here that
the Convention provides a general principle of the Union law,54 being construed as
such in Treaty-based language.55 ‘This basic provision, therefore, asserts the

46
Muir (2014), p. 219; See also: European Union commits to joining European Human Rights
Court. International Law Update. Sep 2007, Vol. 13: 182–183, p. 182; Eckes (2013), p. 2.
47
Shelton (2003), p. 115; See also: Martı́n and De Nanclares (2013), p. 3.
48
See e.g.: Hart (2010), p. 550.
49
The same was said also in the Constitution for Europe, at: Lawson (2005), p. 27; See also: Craig
(2013), p. 1117; de Rivery and Chassaing (2013), p. 3/4.
50
van Rossem (2009), p. 210.
51
Martı́n and De Nanclares (2013), p. 3.
52
E.g.: Lock (2010), p. 777; See also: Groussot et al. (2011), p. 1.
53
See also, generally: Muir (2014), p. 243.
54
Regarding this relationship, the EU Court has ruled in Cinéthèque that: ‘[A]lthough it is the duty
of the Court to ensure observance of fundamental rights in the field of Community law, it has no
power to examine the compatibility with the European Convention on Human Rights of national
legislation lying outside the scope of Community law.’ Court of Justice of EU, Cinéthèque v
Fédération nationale des cinémas français [1985] ECR 2605, Judgment of 11 July 1985 in Joined
Cases 60 and 61/84, p. 2618.
55
Gragl argues in this regard that Art. 6 (3) TEU—in view of the Luxembourg Court’s stance—
binds EU internally but not externally in the context of the ECHR as an international law
instrument (see: Gragl 2013, p. 54). Gragl’s point makes sense if one tries to understand the
facade which is often used by the Luxembourg Court to package its use of rights rhetoric, however,
this may not change the fact that ECHR system is an instrument of international law, which,
5.2 Examining the Status of ECHR and DAA in EU Legal Order 143

European/Member State democratic governance heritage and, more specifically,


the content of the ECHR as a non-derogable underpinning for the whole range of
EU/EC activity.’56 Having a non-derogable status signifies a rather privileged
treaty-like position, one could argue.57 On the other hand, as a hypothetical starting
point, the claim that the EU has no competence to place the ECHR within the rank
of primary law should be considered as unfounded. The Member States have
already conferred certain primary-law authorities to the EU, which, also on the
basis of Member States constitutional traditions, should be confronted with
primary-law rights mainly deriving from the ECHR.58
The third important provision is Art 6 (1) TEU. It asserts that the Charter of
Fundamental Rights has the same legal value as the Treaties,59 explicitly attributing
a Treaty-rank of legal position to the Charter.60 This said, the Charter61 automat-
ically forms part of the primary law of the EU.62 Being part of the primary law, Art
52 (3) ChFR specifically reads: ‘In so far as this Charter contains rights which
correspond to rights guaranteed by the Convention for the Protection of Human
Rights and Fundamental Freedoms, the meaning and scope of those rights shall be
the same as those laid down by the said Convention. This provision shall not
prevent Union law providing more extensive protection.’63 The Convention is

although provided on a domestic constitutional obligation by TEU, still penetrates into EU law
from its international law position and remains a standard of international law rather than domestic
EU law.
56
Harding (2000), p. 141.
57
Besson (2009), p. 254, who argues that in Kadi II, the Court argued that there was no contest that
international law was confirmed as having primacy over the EU legal order, of course excluding
this possibility when the compliance of international law with ius cogens is in question.
58
Cf. with Court of Justice of EU, Case T-112/98, Mannesmannr€ ohren-Werke v Commission
[2001] ECR II-729, para 75, wherein the Court in 1998 argued that an applicant cannot invoke
ECHR directly before it.
59
See e.g.: Franklin (2010–2011), p. 142.
60
Lenaerts (2012), p. 376, who argues in favour of making the Charter and human rights in the EU
a ‘federalized device’; See also: Lenaerts and Gutiérrez-Fons (2010), p. 1658, on how EU human
rights law can be read as affecting the powers assigned by the Treaties in the ‘limiting’ sense;
Editorial (2013), p. 472 et seq.
61
‘It could be easily predicted that the approval of the Charter of Fundamental Rights would
produce a centralising effect, gradually drawing the protection of human rights to the European
level and at the same time sterilising the protection guaranteed by the national Constitutions and
breaking the limits of jurisdiction in which the action of the Community institutions should be
carried out.’ Cartabia (2009), p. 17; Compare this with supra Chap. 4 on how the EU will become
primus inter pares once EU accedes to ECHR.
62
Franklin (2010–2011), p. 161 (‘[. . .] the entry into full legal force of the Charter would at the
very least serve to render the relationship not only between the Charter and the Convention but also
between the two Courts as less legally ambiguous and less contingent upon the mere goodwill of
the ECJ to accept in practice.’).
63
See also: O’Meara (2011), p. 1819; Martı́n and De Nanclares (2013), p. 4; Cf.: Referring to
another agreement, ECJ had ruled that: ‘The fact that the provisions of the agreement and the
corresponding Community provisions are identically worded does not mean that they must
144 5 Status of ECHR and DAA in EU Legal Order

automatically viewed as a constitutional minimum standard.64 ‘This provision pays


tribute to the fact that the Charter was drafted in the image of the Convention, as is
recognized in the Fifth Recital of its Preamble and Declaration No. 1 to the Lisbon
Treaty.’65 This is a substantial incorporation of the Convention to the Charter,
wherein this author does not accept the Convention being made part of primary law
but rather asserts that Convention rights (meaning its substance) are being incor-
porated into the primary law of the EU.66 Another author insists that it ‘flows from
the intention of those who drafted the Charter [. . .] that the whole meaning and
scope of the ECHR will be incorporated in the Charter.’67 One should also point out
that at the time of the drafting of the Charter, there was a proposal to keep the word

necessarily be interpreted identically.’ (Court of Justice of EU, Opinion 1/91 (1991) ECR I-6079,
para. 14); Cf.: de Rivery and Chassaing (2013), p. 3, who argue that the Charter is put by Art.
6 TEU in the same level of hierarchy with the general principles of law and ECHR in the pool of
EU legal order.
64
Heringa and Verhey (2011), p. 16; See also: Krüger (2002–2003), p. 91; Ribble (2010–2011),
p. 224; Lenaerts and de Smijter (2001), p. 97; Craig (2013), p. 1148; Cf.: Garcı́a (2002), p. 497/8,
although there exists the possibility still for divergent interpretations between the Luxembourg and
Strasbourg courts.; See also: Landau (2008), p. 565; Cf.: Wetzel (2003), p. 2839 (‘[. . .] the ECJ
appeased those concerned that fundamental rights protection at the Community level would fail to
meet the minimum standards in the ECHR without making itself subservient to the Strasbourg
Court.’); See also: Von Bogdandy et al. (2012), p. 517; See also: Gragl (2013), p. 61.
65
Weiß (2011), p. 69; See also: Heringa and Verhey (2011), p. 17 (‘It seems plausible to assume on
the basis of the final sentence of Article 52 section 3, (“This provision shall not prevent Union law
providing more extensive protection”), that the case law of the Court will be considered as defining
the minimum level of protection, from which the ECJ may certainly deviate, but only in order to
provide a more extensive protection.’); On the latter argument, see also: Balfour (2005), p. 34;
Lenaerts and de Smijter (2001), p. 99.
66
Weiß (2011), p. 70; Cf.: Gragl makes the point, referring to some authors, that the EU is not
bound by the Convention as long as it has not acceded to it (Gragl 2013, p. 54). This argument may
be to a certain degree valid if one looks the Convention from its international law perspective and
the fact that it may not put obligations on entities which have not become contracting parties to
it. However, from the domestic constitutional perspective, one must disagree with the Gragl’s
point, by arguing that the primary law stipulations established in Art. 6 (3) TEU and Art. 52 &
53 ChFR may not be understood in another way but as indisputably binding the EU to the
Convention’s rights. It seems pointless to make the argument that—as some, including generally
Gragl, try to make—that the Convention as such is different from the rights enshrined by it. One
may in no way pretend that the rights enshrined by the Convention form part of a different body of
law from the Convention, as this would then make the point of effective human rights protection
totally moot.
67
Heringa and Verhey (2011), p. 16; Cf.: Gragl notes that although this may be read as incorpo-
ration of the Convention to the Charter it may not be read as subordination of the Luxembourg
Court to the Strasbourg Court: See: Gragl (2013), p. 62. I disagree with Gragl in this regard, not for
the overall conclusion he arrives at but rather for the fact that, according to him, such ‘incorpo-
ration’ should be read as having no relevance in terms of the relationship between the two courts.
Such argument seems moot for the mere fact that the relationship between the two acts should
definitely produce some effect also with regard to the relationship between the two courts which
hold exclusive jurisdiction on those acts.
5.2 Examining the Status of ECHR and DAA in EU Legal Order 145

‘similar’ instead of the word ‘corresponding’,68 but the latter was intentionally
selected. It is therefore argued that it was the intention of the Masters of the Treaties
to attach that explicit effect to the Convention in relation to the Charter. In addition,
one may add that this may go as far as recognizing Strasbourg’s case law as forming
part of that meaning (the Luxembourg Court has already shown willingness in
Schecke GbR v. Land Hessen to recognize this69). The expressional linkage made
here between the Charter and the Convention provides a framework within which
one should include analysis of the status of the Convention and the DAA, some-
thing which will help analyze the legal significance of the Convention in the EU
legal order. Let us examine each of these points.

5.2.3 Five Specific Arguments: Shaping More Concretely


This Undeveloped Relationship

First, the legal definition of Art 6 (2) TEU that the ‘Union shall accede’ to the
Convention provides a lot of room for examination. Essentially, such legal pre-
scription provides a special obligation for the Union to accede to the Convention.
The term “shall” could not be seen as merely an option for the Union.70 It (‘shall
accede’) is made to ensure that the EU becomes obliged to accede to the Conven-
tion, and there is no second option or discretion left for another alternative.
Expressio unius est exclusio alterius—the explicit mention of one thing excludes
all others, is postulated. The question therefore remains: Should it have been
reliable not to mention that there is a “duty” for the Union to accede—a duty
which also is meant to entail a Treaty-based source for the accession to the ECHR?
Should the Masters of the Treaty have liked to make it simply an alternative rather
than an obligation, they could have construed the provision in a way similar to
Protocol 14 of the Convention, which reads that the EU ‘may accede’ to the
Convention. Similarly, the masters of the treaty could have stated that the EU is
‘authorized’ to accede to the ECHR. The first version of this proposed provision in
the Constitutional Treaty for the EU had envisaged the wording the ‘EU may
accede to the Convention’.71 Such wording was then intentionally removed and
replaced with the word ‘shall’. The lack of the intention to word the article as
optional, avoiding use of the word ‘may’, is in itself of significant quality to explain

68
Craig (2013), p. 1149.
69
Court of Justice of EU, Schecke GbR v. Land Hessen, Cases C-92-93/09.[2010] E.C.R. I-11117,
para. 51.
70
If the Union were to make this as non-obligatory, it would use the verb ‘may’ instead. See a
comparison for the same legal comparison of this verb, at: Mendez (2010), p. 1729; Cf.: Licková
(2008), p. 490, who argues that—as exemplified from similar cases like Bosphorus—such type of
relationship is meant as furthering the European integration concept in the context of the Luxem-
bourg Court’s pattern as followed to date.
71
See on the latter: Callewaert (2014), p. 48.
146 5 Status of ECHR and DAA in EU Legal Order

the intention of the later wording as ‘shall’. This being said, Art 6 (2) TEU is written
to mean that there is a Treaty-based obligation to accede to the Convention.72
Therefore, the “duty” to accede to the Convention is certainly located within the
primary law of the EU.
The above paragraph, nevertheless, should be further extended to identify the
real positive and negative meaning of Art 6 (2) TEU. The Treaty-based obligation
for the EU to accede to the Convention should not be examined merely from its
static dimension, but also from the dimension of its intended outcome. Clearly, by
‘obliging’ the EU to accede to the Convention, the Masters of the Treaty have also
assented and requested that the EU receive and undertake the obligations stemming
from accession to the Convention. It is clearly argued that by obliging the EU to
accede to the Convention, the Masters of the Treaty have obliged the EU to
undertake the obligations deriving from the Convention not merely as an interna-
tional agreement. This said, Art 6 (2) TEU should be read to mean that the intended
outcome of the accession concerned is that the EU undertake the obligations and
receive the Convention as a source of law deriving from a Treaty basis.73 The ‘duty
to accede’ is written to mean that whatever the output of the accession process, the
Treaties recognize and accept such output as being the ‘intention’ and ‘function’ of
the Treaties themselves. If this had not been the real meaning of the ‘duty to
accede’, the Treaties could have merely ‘motivated’ or ‘invited’ the Union institu-
tions to accede to the Convention, but not obliged them. More importantly, the
nature of the ‘duty to accede’ and the intended ‘output’ from the implementation of
this duty is that the Convention rights become implanted in the Treaties’ system.
There seems to be a clear linkage between the ‘duty to accede’ and the intended
‘output’ through which the Convention rights become constitutionally established
in Union law. Otherwise, the question remains what would be the result of
breaching Art. 6 (2) TEU, namely failing to fulfill the ‘duty to accede’ and inability
to produce the intended ‘output’? Of course, the result would be that the Union and
its Member States have violated the Treaties.74 Likewise, the argument stands that
if it is a violation of the Treaties should the Union fail to accede to the ECHR75 and
implant the Convention’s rights to the EU constitutional system as the intended
‘output’, then there is no dispute that the duty articulated in Art. 6 (2) TEU is meant
as providing a Treaty stipulation and position to the Convention and accession
thereto, probably not longer than this, however.

72
Accord.: Franklin (2010–2011), p. 159, who argues that the accession is a mandatory positive
duty for the EU.
73
Contra: Weiß (2011), p. 71/2, where the author tries to portray the Convention merely as an
international agreement standing between the primary and secondary law.
74
Cf.: ‘In the light of article 6(3) TEU in particular, it would be contrary to EU law to disregard the
Convention.’ Eckes (2013), p. 278/9.
75
It is argued that should the EU choose not to accede, that would give entitlement to anyone to
use the action for failure to act on basis of the Treaties’ obligation on the Union before the
Luxembourg Court. See on this argument: Jacque (2011), p. 995.
5.2 Examining the Status of ECHR and DAA in EU Legal Order 147

In relation to the previous argument, one should also analyze the systemic
picture of the Treaties regarding the Convention. Many argue that the Convention
should be merely seen as an international agreement to which the EU is obliged to
become a party. Nevertheless, the systemic picture of the Treaties suggests the
opposite. At the outset, it is important to observe the wording of the Treaties
regarding the Convention and all other international agreements. A basic observa-
tion in this regard is that the Treaties uniquely mention and bring up the accession
of the EU to the Convention, something that does not appear in any other interna-
tional agreement in the Treaties’ text. Nowhere else in the Treaties could one find
an obligation for the EU to accede to an international agreement, as that is simply an
issue under the auspice of Art 216 TFEU. A systematic observation therefore
reveals that the Treaties provide special significance to the Convention, one
which the Masters of the Treaties would not have intended unless they meant to
provide this privileged status to the Convention. This noted, the Convention’s status
could not be considered that of a usual international agreement, as there would have
been no need to attribute to it a special significance with a special Treaty provision.
Art. 6 (2) TEU therefore stands as a lex specialis to the general rules on treaty
conclusion in the Union, clearly attributing the Convention and the accession to it a
privileged position.
Of further importance in examining the Convention’s status76 is the second
sentence of Art 6 (2) TEU, which sets forth that the accession of the EU to the
Convention ‘shall not affect the Union’s competences as defined in the Treaties.’ In
addition to that, Art. 2 of Protocol 8 sets forth the following safeguard: ‘The
agreement referred to in Article 1 shall ensure that accession of the Union shall
not affect the competences of the Union or the powers of its institutions.’77 This
prohibitive rule plays a huge role with regard to the testing of the negative meaning
of Art 6 (2) TEU. If one would follow the argument that the Treaties provide no
room for the Convention and the DAA to be located within the primary law of the
EU, then, the logical question would be why would Art 6 (2) TEU and Art 2 of
Protocol 8 state that the accession ‘shall not affect’ the competences as defined in
the Treaties, meaning both the competences of the Union or the authority of its
institutions that are set forth by the Treaties. More practically, how can accession
and the Convention affect the Treaties’ materie if they are inferior to the Treaties?
Or, even more practically, would it be logical to set forth a prohibitive rule for

76
In many national legal orders of EU Member States, the Convention retains a constitutional rank
in the hierarchy of law. See e.g.: Martinico (2012), p. 404; See also on the latter: Callewaert
(2014), p. 21; See also: Sweet (2009), p. 630/1, arguing that ECHR system is meant to be of a
general constitutional nature.
77
See e.g.: Groussot et al. (2011), p. 16, who argues that it would suffice to be considered that the
EU law autonomy is not touched by the DAA if it establishes no new competences for the Union;
See also: Gragl (2013), p. 8, who rightly notes that it would not be logical to expect that the EU and
its Court will give up from their competences and authority which they have so hardly taken from
the sovereign Member States. In view of this, the limitations drawn with Protocol 8 seem to
accommodate this rationale.
148 5 Status of ECHR and DAA in EU Legal Order

something which may have not have a chance to be in the position of causing the
prohibited act? The fundamental argument, therefore, is that if the Convention were
placed by the Treaties as having a subordinate position (to primary law) there would
be no need to put a prohibitive rule in the Treaties ruling out that the Treaty
competences of the Union be changed by the act of accession. Should Art 6 TEU
be read as giving a legal status lower than the primary law to the Convention and the
Accession Agreement, there would be no need to prohibit an interference from a
lower act such as the Convention or the DAA, as both would not have a legal role in
changing any of the competences of the Union.
Therefore, the argument is that so long as the Treaties prohibit the DAA from
affecting the competences of the Union, such a provision is made to be negatively
read as saying that the Accession Agreement shall have the status of an act that
could affect the Treaties. Such an act should be seen as having the status of Treaties
themselves, otherwise there would be no need to put such a prohibitive rule in place.
Art. 2 of Protocol 8 therefore should not be read as prohibiting the fact that the
Convention becomes—implicitly or explicitly—part of primary law, but rather
should be read as a negotiating principle78 which guides Union institutions on the
extent of their negotiating powers over the DAA in a way which may not explicitly
impair or shorten the competences of the Union or its institutions. It is therefore
argued here that the DAA does not make any substantive or explicit change of the
competences, and the Convention’s status in the EU legal order may not be
considered as serving that aim.
If one were to present the counter argument that the Treaties could not be
amended but by the will of the Masters of the Treaties, then it is important to
note that such provisions that led to the abovementioned arguments are part of the
Treaties. They have been adopted under the ordinary revision procedure of the
Treaties, and have had support of the Masters of the Treaty as long as they were
adopted by each Member State in accordance with their national constitutional
procedures. As such, the arguments deriving from the negative interpretation of Art
6 TEU could not be seen as amending the Treaties, as those arguments pursue
strictly the very essence of the Treaties’ provisions.
Third, as argued above, the Charter has a Treaty-level position in the hierarchy
of EU law. In Art. 52 (3), the Charter establishes the principle that its rights which
correspond to the Convention rights shall have as a minimum ‘the meaning and
scope’ of Convention rights. Certainly, this stands as a minimum standard which
the Charter rights should fulfill consistently. One could legitimately question
whether it would be possible for an inferior act, such as the Convention, to set
out minimum standards that the superior act should pursue (the Charter)? Of course
not. Establishing the Convention as the threshold which the Charter should meet,
places the Convention in at least the same or more superior position in relation to
the Charter. This argument further asserts that the Convention may not stand as the
threshold to be superseded in terms of its minimum standard if it is not at least

78
On the argued complexity of these negotiations, see e.g.: Olsen (2009–2010), p. 73.
5.2 Examining the Status of ECHR and DAA in EU Legal Order 149

accepted or recognized as having the same rank with the Charter. Otherwise, one
should have upheld the argument that the Charter, as the superior act, is being
infringed upon by a lower act, the Convention, which would originate from the
Charter. Therefore, the only logical meaning of Art. 52 (3) of the Charter is that the
minimum standard set by the Convention is one which rests at least at the primary
law level, and as such, it has the capacity to legally oblige the interpretation of the
Charter in conformity with Convention rights. The contrary would lead us to the
argument that the minimum standard principle set forth in Art. 52 (3) of the Charter
infringes upon the Charter itself, as the minimum threshold—being inferior to the
Charter—may not impose on any act above itself, and the Charter may violate its
own provisions.
Fourth, Art 6 (3) TEU states that: ‘Fundamental rights, as guaranteed by the
European Convention for the Protection of Human Rights and Fundamental Free-
doms and as they result from the constitutional traditions common to the Member
States, shall constitute general principles of the Union’s law.’79 It clearly follows
from this provision that the Convention’s rights are regarded by the Treaty as also
having the status of general principles of EU law, adding that general principles of
law are considered by some to enjoy a primary law status.80 Nevertheless, from a
more neutral point of analysis, general principles of EU law serve as unwritten
foundations of EU law. Most general principles are either explicitly devised by the
Court of Justice81 or are interpreted in light of the mechanisms and practical
objectives of the Treaties, although they have had no Treaty basis. The Court
draws such general principles either from more general principles of law or the
general principles common to the laws of Member States of the Union. One should
argue that the general principles of EU law are not necessarily derived from the
Member States, adding that some of the general principles are simply not present at
all at any of the Member States constitutional traditions, for instance the right
against self-incrimination.82 Ackerhust argues that the Luxembourg Court does not
apply those general principles which are most widely spread within the Community
legal order, but those having the most advanced and progressive function.83 The
identification and the construction of the general principles of EU law follow both

79
See also: Sweet (2009), p. 637.
80
Quirico (2010), p. 34; See also: Gragl (2013), p. 102; Accord: De Londras and Kingston (2010),
p. 370 (‘[. . .] general principles of Community law rank at the same level as the EC and EU
Treaties themselves [. . .]’); For the utilization of general principles of EU law to invalidate
secondary law acts, see e.g.: Court of Justice of EU, Omega Spielhallen, Case C-36/02, [2004]
ECR 1-9609; Court of Justice of EU, Schmidberger v. Austria, Case C-112/00, [2003]ECR 1-5659;
Cf.: Balfour (2005), p. 12 (Advocate-General Cosmas argued in Van Der Wal that: ‘The Court of
Justice and the Court of First Instance do not apply the ECHR, but rather the general principles of
Community law.’).
81
Utilizing its authority under Art 19 (1) TEU.
82
Court of Justice of EU, Orkem SA v. Commission, Case 374/87, [1989] ECR 3283.
83
Akehurst (1981), p. 33–35; E.g. only fundamental general principles of law can override the acts
of EU, at: Ibid, Akehurst (1981), p. 40–47.
150 5 Status of ECHR and DAA in EU Legal Order

the very objectives and needs of the Community system. General principles as such,
although unwritten sources of law, are meant as the founding principles on which
the Treaties are codified.
General principles in the EU legal order are used for various reasons, and
therefore they are meant to have different legal ranks. Most commonly, the general
principles are utilized as tools to interpret the Treaties and/or legal acts of the EU
institutions, as tools to fill loopholes in the EU legal order, as legal benchmarks for
assessing the validity of various EU secondary legal acts, and as founding charac-
teristics which the Union’s special nature requites for its existence.84
From a more theoretical perspective, many general principles of law are seen as
sources of rationale for a certain norm’s existence. Gerald Fitzmaurice argued that
general principles of law draw attention to, and elucidate, a written rule.85 In this
vein, general principles of law are meant to be seen as the foundation upon which
the associated written rules should be interpreted, signalling that general principles
of law provide for the basis of a Treaty regime, rather than being an inferior source
of law. In the words of Fitzmaurice, the EU legal order is based in numerous general
principles of law which are neither written nor codified, the vast majority of which
have been devised by the Luxembourg Court. Otherwise, general principles in the
EU legal order are considered ‘[h]orizontally, as a source of “primary law”’86 at the
position of constitutional foundations. Such general principles of law include the
principle of supremacy, the principle of direct effect,87 the right to defence
(as recognized by the court in Transocean Marine Paints Association
v. Commission),88 the principle of legal certainty (T€ opfer & Co. GmbH
v. Commission),89 the principle of non-retroactivity (as validated by R v. Kirk),90
the right to effective judicial protection as seen in Johnston v. Chief Constable of
Royal Ulster Constabulary,91 the principle of proportionality as observed in the
decision of the court in Macchiorati Dalmas e figli v. High Authority,92 etc. Most of
these general principles of law, as Tridimas considers, are argued to have a
constitutional class of legal authority.93
The Convention certainly is indirectly attributed the standing of a general
principle of EU law given the prescription of Art 6 (3) TEU. The argument is that

84
Akehurst (1981), pp. 29 and 30.
85
Fitzmaurice (1957), p. 7.
86
Lenaerts and Gutiérrez-Fons (2010), p. 1629.
87
Court of Justice of EU, Case 6/64 Costa v ENEL [1964] ECR 585.
88
Court of Justice of EU, Transocean Marine Paints Association v. Commission, Case 17-74, 1974
E.C.R. 1063.
89
Court of Justice of EU, T€opfer & Co. GmbH v. Commission, C-112/77, 1978 E.C.R. 1019.
90
Court of Justice of EU, R v. Kirk, C-63-83, 1984 E.C.R. 2689.
91
Court of Justice of EU, Case 222/84, Johnston v. Chief Constable of Royal Ulster Constabulary
[1986].
92
Court of Justice of EU, Case 1/59, Macchiorati Dalmas e figli v. High Authority [1959].
93
Tridimas (2006), p. 4.
5.2 Examining the Status of ECHR and DAA in EU Legal Order 151

although the Convention is somehow brought under this umbrella, one could not
argue that general principles of law are explicitly subordinated to the Treaties and
are not considered part of the Treaties. Such an argument would seem unconvincing
for two core reasons: first, although the Luxembourg Court has never given an
explicit rank of legal power to the general principles—which is to a certain extent
rightly maintained—the Court for instance explicitly argued in Opinion 1/91 that
the principle of direct effect and primacy of EU law comprise ‘the essential
characteristics of the Community legal order.’94 Such meaning reveals the fact
that the Court places such general principles of the EU legal order at the top of the
EU law hierarchy, as such principles are named as being essential to the very
organizational nature of the EU: an indication that is clearly of constitutional rank
and significance. Second, general principles of EU law serve as the very character-
istics of the Treaties, and are essentially made to serve the Treaties rational/
foundational existence and the key objectives of the organization itself.
It is of course admitted that a general principle of law in the EU legal order may
not in principle prevail over a Treaty rule.95 However, it is also argued that as long
as the general principle does not contradict a Treaty provision it is used as a
foundational guiding rule, resting at the same level as the Treaties. Remark should
in this regard be made to the Luxembourg’s Audiolux SA case, where it ruled that
‘[t]he general principles of Community law have constitutional status’.96 This being
an indication that certain general principles may sit at the level of EU primary law,
one may go even further and bring up Kadi where the Luxembourg Court, amongst
others, referring to some primary-law provisions that set certain common market
stipulations, established that they may not ‘be understood to authorize any deroga-
tion from the principles of liberty, democracy and respect for human rights and
fundamental freedoms enshrined in Article 6(1) EU as a foundation of the Union.’97
Furthermore, it ruled that:
Article 307 EC may in no circumstances permit any challenge to the principles that form
part of the very foundations of the Community legal order, one of which is the protection of
fundamental rights, including the review by the Community judicature of the lawfulness of
Community measures as regards their consistency with those fundamental rights.98

This basically means that Kadi allows some general principles of law, such as
those mentioned above, to take precedence even upon the Treaties. Former Art.
307 EC was therefore set aside by the Court on basis of some general principles of

94
Court of Justice of EU, Opinion 1/91 on Draft Agreement Relating to the Establishment of the
European Economic Area, para 21; See also: Reestman (2005), p. 104.
95
Akehurst (1981), p. 30.
96
Court of Justice of EU, Audiolux SA and Others v Groupe Bruxelles, Case C-101/08, Judgement
of the Court of 15 October 2009, para. 63.
97
Court of Justice of EU, Kadi and Al Barakaat International Foundation v Council and Com-
mission, C-402/05 P and C-415/05 P [2008] ECR I-6351, para. 303.
98
Court of Justice of EU, Kadi and Al Barakaat International Foundation v Council and Com-
mission, C-402/05 P and C-415/05 P [2008] ECR I-6351, note 97, para. 304.
152 5 Status of ECHR and DAA in EU Legal Order

EU law. This may therefore be a huge step forward to establish the position of the
general principles of law in the EU legal order. This being noted, the fact that the
Convention is implicitly prescribed with the status of a general principle of law in
the EU legal order—an indication that derives from the explicit mentioning in the
Treaties that the Convention’s rights form general principles of law in the EU legal
order—does not prevent it from being considered an implied part of the Treaties. As
fundamental rights are explicitly given Treaty standing, it logically follows that the
interpretation of those fundamental rights in light of the Convention as a general
principle of EU law and as a tool for identifying the rationale of such rights, the
Convention is not precluded from enjoying Treaty status by Art 6 (3) TEU. Instead,
the Convention is given the status of a foundational principle whereupon the
Treaties’ interpretation of rights should be based.
To dig a bit further into whether the issue of the Convention being a general
principle of EU law99 prevents it from becoming part of primary law, one should
examine the Declaration relating to Art 55 (2) of the Treaties. The Declaration, with
the intention of reaffirming the legal nature of the unwritten principle of primacy in
EU law, states that ‘primacy of EC law is a cornerstone principle of Community
law’ and that ‘[. . .] fact that the principle of primacy will not be included in the
future treaty shall not in any way change the existence of the principle [. . .].’ One
should reiterate that such Declarations are annexed to the final act of the intergov-
ernmental conference that adopted the Treaty of Lisbon. It is obvious therefore that,
as stated in the Declaration concerned, although the principle of primacy of EU law
has not been explicitly mentioned in the Treaties, it is implied that such general
principle constitutes an essential unwritten constitutional characteristic of the
Union which forms part of the Treaty system. It is therefore argued that the fact
that a general principle is not mentioned in the Treaty does not preclude it from
having constitutional status, similar to the case of the general principle of pri-
macy.100 The same logic should also be extended on the Convention, meaning that,
although the Treaties make no explicit mention of the Convention’s constitutional
status, such status is provided within the context of the Treaty foundational nature,
wherein the Convention played and plays a core role accepted by the CJEU as well.
Fifth, it is important to examine once more the fact whether the Convention
could be considered a mere international agreement. Initially, the DAA is a mixed
agreement, as it involves competences shared between the EU and its Member
States.101 However, the conclusion and legal effectuation of the DAA is

99
It is not the intention to argue here that the Convention as such is formally recognized as a
general principle of EU law, but rather that the rights which the Convention enshrines form general
principles of EU law. Such fact, one should say, leads to the argument that the Convention is—
substantively speaking—a general principle of EU law as well. It seems not wise and even relevant
to make a distinction between the rights guaranteed by the Convention and the Convention itself:
such distinction is practically moot. Therefore, substantively, the Convention is intentionally
regarded as a general principle of EU law in this chapter.
100
On primacy, and its legal nature: Reestman (2005), p. 104 et seq.
101
The general rule to sign mixed agreements in external relations seems the usual path followed
in practice. See: Leal-Arcas (2001), p. 483.
5.2 Examining the Status of ECHR and DAA in EU Legal Order 153

distinguished with a special procedure under Art 218 (8) TFEU, which distin-
guishes it from all other international agreements under Chapter V of TFEU.102
More specifically, Art 218 (8) TFEU states that: ‘The Council shall also act
unanimously for the agreement on accession of the Union to the European Con-
vention for the Protection of Human Rights and Fundamental Freedoms; the
decision concluding this agreement shall enter into force after it has been approved
by the Member States in accordance with their respective constitutional require-
ments.’103 All this being a very special procedure, quite similar—and fulfilling the
substantive, if not formal, elements of an ordinary treaty revision—to an implied
treaty revision, or treaty supplement in the form of a lex specialis.104
It follows from the general observation of the above provision that in order for
the DAA to be concluded and effectuated, it needs first be consented by the
European Parliament,105 and then concluded unanimously by the Council upon
the proposal of the Negotiator.106 Such Accession Agreement will be legally
effective only when it has been ratified by all EU Member States in accordance
with their constitutional provisions on ratification of international treaties.107
Although the DAA is a ‘mixed agreement’,108 as a shared function of both the
EU and Member States in the same international convention, it needs this require-
ment to fulfil the need to obtain the Member States will for such Accession
Agreement and seeks unanimous vote for concluding it in the Council. In this
light, as the Member States will have to ratify the decision of the Council conclud-
ing the DAA, in line with their constitutional ratification provisions, one cannot
argue that it would be outside the will of the Masters of the Treaties that the DAA
not mean what it actually does. The Convention’s arguably constitutional status in
the EU legal order does not therefore contravene ‘Kompetenz-Kompetenz’,109 as it
is the Member States playing their role unanimously twice to define this issue: once
in the Council with unanimous voting, and once through their constitutional

102
On ordinary trade agreements, e.g., see: Woolcock (2005), p. 245 et seq.
103
Emphasis added; See also: Groussot et al. (2011), p. 17; It was not initially required unanimity
in the Council, but this was later put so with the insistence of some Member States. This fact
further supports the argument presented in the main text. See on this: Jacque (2011), p. 996.
104
See the ordinary revision procedure, at: Peers (2012), pp. 19 et seq.
105
Art 218 (6) [a: iii]; On how the EP protects its right on international agreements, see a landmark
case, at: Court of Justice of EU, Parliament v Council, Case C-189/97 [1999] ECR I-4741.
106
Art 218 (6) TFEU.
107
This is also done with the aim of ensuring that the external relations of the Union are meant
within the essential characteristics of the Union. See e.g.: Leal-Arcas (2001), p. 510; On the
ratification steps, see also: Lock (2010), p. 778.
108
Cf.: Koutrakos (2006), p. 152 (‘Mixity enables the Community institutions and the Member
States to avoid debates about the legal bases of envisaged agreements and arguments about precise
delineation of competences.’).
109
On ‘Kompetenz-Kompetenz’ in cases involving international agreements in the EU, see: Kumm
(2005), p. 264.
154 5 Status of ECHR and DAA in EU Legal Order

ratification procedures. Callewaert supports a similar view by arguing that ‘[t]his


procedure is undoubtedly extremely unwieldy. It is largely accounted for by the
exceptional significance of accession for the European Union, described by the
CJEU in its Opinion 2/94 as “constitutional”.’110
It is not the intention to argue here that the DAA will revise the Treaties (even
though it likewise fulfills the expected needs of Art. 48 TEU),111 although the
procedure under Art 218 (8) TFEU fulfils the goal of having the full Member States
permission to have the agreement concluded. The conclusion of such an agreement
entails the stance that Member States agree to the fact that such an agreement be
treated in line with the overall legal picture of Art 6 (2) TEU. This being said, the
argument could not be that the Member States have not individually ‘ratified’ the
DAA in their capacity of the Masters of Treaties (as the collective author of the
Treaties). Instead, the argument is that the DAA, and its conclusion entails a
procedure wherein Member States are directly involved, therefore it has the assent
of the Member States to enjoy the status provided for in Art 6 (2) TEU, and that the
consideration of such an agreement as part of the primary law, as long as it does not
contradict with the Treaties, is not excluded.112 Going far with such an argument,
one may invoke two cases, as examined infra,113 in which the DAA is thought to
shift the meaning of EU Treaties to new legal understandings.114 One, the case may
be the prior involvement mechanism, which is argued—as explained infra (chapter
on prior involvement mechanism)—to produce the effect of making the under-
standing of the Treaties different from what it is now. Second, the inter-party
mechanism within DAA will definitely produce the effect of altering the meaning
of the Treaties on international settlement of issues pertaining to EU law. Conse-
quently, the fact that the Treaties become fine-tuned to new concepts and practical
understandings after the DAA becomes effective may not be rejected with an
absolute right.
Therefore, as a whole, one may argue that the Convention and the DAA are
reserved a special position within the EU Treaties,115 a position which may not be
as simple as that of an international agreement.116 This special status is not merely

110
Callewaert (2014), p. 52.
111
See also: Peers (2012), p. 19 et seq.
112
Cf.: ‘Considering that the Court of Justice in Kadi has elevated human rights (together with
other core principles of EU law) to the status of being the ‘very foundations’ of constitutional law
that rank above ‘ordinary’ EU primary law, a breach of the ECHR would logically make the EU
Treaties unlawful under EU law.’ Eckes (2013), p. 282.
113
In the chapter infra on inter-party complaints and the chapter infra on prior involvement of EU
Court.
114
In addition, the entire effect of the DAA seems also very similar in substantive terms with Art.
49 TEU.
115
To note, there are some EU member states which place ECHR in the rank of their own
constitutions. See e.g.: Miiller (2007), p. 35.
116
E.g. in Ireland v. United Kingdom, application no. 5310/71, para 239, the Strasbourg Court had
ruled: ‘Unlike international treaties of the classic kind, the Convention comprises more than mere
5.2 Examining the Status of ECHR and DAA in EU Legal Order 155

supposed but rather logically argued if positive and negative meanings of the EU
Treaties are combined, and a systemic and thorough interpretation of all treaties’
relevant provisions are examined.117 That said, the balanced argument would be
that the Convention and the DAA—although not explicitly given a Treaty-rank of
legal position—have a constitutional significance and position in the EU Treaties.
Even the EU Court stated in Opinion 2/94 that accession of the EU to the ECHR
would be of ‘constitutional significance’,118 most simply demonstrating that both
the ECHR and the DAA would have to be accommodated in this domain of law.
Their meaning, therefore, poses an understanding that they are located
unreceptively—if not actively—at the EU primary law,119 certainly their intention

reciprocal engagements between contracting States. It creates, over and above a network of
mutual, bilateral undertakings, objective obligations which, in the words of the Preamble, benefit
from a “collective enforcement”.’ This does of course support our statement above; Contra.: Gragl
(2013), p. 102, who argues that the Convention will have the status of a normal international
agreement sitting between primary law and secondary law post-accession. Such position,
according to Gragl, will force the Luxembourg Court to review the legality of the secondary law
not only in light of the Treaties but also the Convention.
117
It would suffice for purposes of respecting Protocol 8 to the Treaties merely ‘not to touch’ upon
the competences of the Treaties, however there is room made available to allow an understanding
that the DAA ‘may’ change positively the understanding of Treaties of course without making any
change to the competences. In addition, e.g.: Groussot et al. (2011), p. 16, puts the argument that
the intention was in fact to uphold the autonomy which is translated as a prohibition on touching
upon the competences, for a point that makes room for our argument to be still valid.
118
Landau (2008), p. 560; See also: Shelton (2003), p. 115; Cf.: Gragl in this regard argues that
Opinion 2/94 merely mirrors the concerns of the Member States on potential accession, and it may
not be concluded whether the Court’s thought was as such against accession due to the potential
interference to the EU law autonomy. See: Gragl (2013), p. 80.
119
Contra: With regard to the status of the Convention in EU law hierarchy, Gragl raises the
question of whether the Convention may gain a primary law position due to the fact that the
Treaties make it obligatory on basis of Art. 6 (2) TEU EU’s accession to the Convention. Gragl
responds to this question rather simply, by maintaining that even if EU would like to do that it lack
the kompetenz-kompetenz to amend the treaties, as it is the Member States which are the Masters of
the Treaties. Gragl goes further by arguing that EU Member States play no significant role in the
accession process, and there is no reason why they should play any such role. He argues that the
DAA will be an EU-exclusive international agreement on basis of Art. 218 TFEU, as the DAA will
be concluded merely between the EU and Council of Europe contracting parties (but not EU
Member States). Gragl in this regard notes that Member States do have only one competence: that
of requiring an opinion on the DAA from the Luxembourg Court. In this regard, finally, Gragly
rightly points that Art. 218 TFEU may not be drawn as a basis for primary law amendments. See:
Gragl (2013), pp. 100–101. I disagree with Gragl on all points except the last one. First of all, it
would not be true to maintain that EU Member States have no function in the conclusion of the
DAA. As shown above, the DAA may only become effective if the decision concluding it will be
ratified by EU Member States on basis of their constitutional rules. It is the TFEU providing for
this derogation to the general rule on conclusion of international agreements. The Masters of the
Treaties do have a constitutive role in the procedure for concluding the DAA, and, teleologically,
one wonders why should the TFEU have provided for this special rule. Second, I disagree with
Gragl who connects the meaning of Art. 6 (2) TEU with the status of the DAA. If one considers the
status of the Convention, it does not necessarily have to be linked to Art. 218 TFEU, as the latter
merely sets the procedure for the conclusion of international agreements but does not govern their
156 5 Status of ECHR and DAA in EU Legal Order

and practical effect not being one which changes the Treaties but rather affirma-
tively become accommodated therein. Such a position given to the Convention and
the DAA is one which may not result in breaching the explicit meanings of the
Treaties, something that needs be acknowledged expressly. With the Convention
and the DAA having this unreceptive constitutional position at the EU primary law
landscape, their effects may be so as to influence the overall functioning of EU law,
however not likely at shifting the latter’s form or functioning through explicit and
ambitious effects or practices. Whether this result would nevertheless impair the

status and their effect. Art. 6(2) TEU is a lex specialis to the general treaty rules on international
agreements, and has its relevance and effect independent from the effect that DAA will receive on
basis of Art. 218 TFEU once it gets concluded. Limiting the scope of the special rule (Art. 6.2
TEU) with the general rule (Art. 218 TFEU) seems a moot point. The latter is the rule whereas the
former is the exception. Third, it is not true to regard the DAA as an EU-exclusive agreement.
There are two basic reasons for that: (a) the sort of obligations in the DAA are not those relating
merely to the EU competence, but also to the shared competences. The co-respondent mechanism
is one good example to prove this claim. The CJEU’s Opinion 2/13, while making the argument
that the Member States’ position may change in face of the Convention if their reservations
become subject to a mutual responsibility decision of the Strasbourg Court (para. 227/8), merely
confirms this argument. (b) There would have been no need for the member states’ to participate in
the effectuation procedure for the DAA if the latter would have not been of a mixed nature. The
mere fact that member states are a constituent player in the latter procedure seems to suggest that
teleologically the TFEU regards them as maintaining certain competence in relation to that
agreement’s substance. Gragl’s final point needs a certain answer as well. No one has ever claimed
that the DAA will amend the treaties in the procedural sense, something I agree with Gragl.
However, the position which Art. 6 (2) TEU may give to the Convention stands as an independent
point from the effect that the conclusion of DAA will draw on EU law. The DAA may even never
be concluded, however, that is absolutely irrelevant to the nature of effect which Art. 6(2) TEU
recognizes to the Convention. To go further, I first use Kaczorowska (2013), p. 109, who states that
EU primary law, amongst others, comprises of ‘[a]cts adopted by the Council, or the Council and
the EP for the adoption of which approval by the Member States in accordance with their
respective constitutional requirements is necessary’. The decision concluding the DAA and the
accession procedure sits in the same procedural row as well. Kaczorowska therefore seems to
support the fact that the DAA and the Convention sit within EU primary law. Second, I bring here
again Opinion 2/94, wherein the Court established that EU accession to ECHR will be of a
constitutional significance. It was the Luxembourg Court itself which had proclaimed this fact
even before the Lisbon Treaty’s adoption. With the Lisbon Treaty there are even more arguments
in support of this. Third, the fact that CJEU in Opinion 2/13 seeks a certain coordination
mechanism between the Charter’s Art. 53 and ECHR—which seems to say ‘coordination’ between
the two rather than ‘subordination’ of the Convention to the Charter—shows a horizontal level of
arrangement between the two acts. Should it have said that such coordination needs not be made
because ECHR as an international agreement would be bound to the Charter and therefore would
not be able to challenge the Charter’s effectiveness, then the argument would have been that the
Court is considering the incorporated ECHR as an inferior instrument compared to the Charter
(primary law). However, with this assertion, the Court’s concern seems to suggest that ECHR will
sit with EU primary law, otherwise the CJEU’s Melloni standard alone would have safeguarded the
subordination of the Convention to the Charter and there would be no need for further mecha-
nisms. Since the EU Court seems not convinced that Melloni’s safeguard may deliver this output, it
is therefore suggesting that the Convention sits at the same level with the EU Treaties. These being
some counter-arguments, this book maintains another direction of argument compared to Gragl,
and, it is argued that such arguments presented in the body of this chapter appear to be valid still.
5.3 Chapter’s Summary of Conclusions 157

autonomy of EU law—adding that Strasbourg decisions would be binding on the


Union according to Art. 216 (2) TFEU—remains to be seen in practice.120 Of
course, this conclusion may not be taken to mean that by accepting the Convention
as part of primary law the Luxembourg Court subsumes to the Strasbourg Court.121
Instead, the conclusion here merely clarifies the argument on the position of the
Convention, keeping still the note that the Luxembourg Court will still insist that
the ECHR be interpreted within its own jurisdiction and on its own territorial
authority and right.122

5.3 Chapter’s Summary of Conclusions

The chapter examined the Convention and the DAA’s status in the EU legal order
post-accession, giving direction on the hierarchical status that both of these instru-
ments will enjoy and their effect mainly from an EU constitutional law perspective.
The chapter therefore concludes that the Convention and the DAA will not enjoy a
normal ‘international agreement’ status in EU law post-accession, but a far more
honored position, one that lays within the foundation of EU primary law. Such
positioning, the chapter concludes, makes both the Convention and DAA with the
capacity to adjust affirmatively—for the benefit of the EU’s internal needs to
comply with the Convention system—some treaty mechanisms already in place.
The latter conclusion is supported by several arguments presented infra, and will
further advance the EU’s interaction with international law instruments like the
Convention.

120
See e.g. European Parliament, Resolution of 19 May 2010 sub. 1, which requests that the
accession effects not hinder the autonomy of EU law and its Court’s jurisdiction; See also:
Franklin (2010–2011), p. 159 (‘As a consequence of this, the ECJ would no longer be the ultimate
arbiter of the lawfulness of EU action alleged to be in breach of human rights, thus signifying a
potentially major incursion into its jurisdiction.’).
121
Cf.: In Gragl’s view, with the Convention taking a ‘mezzanine’ status in EU law, there may
appear clash of principles between Art. 53 ECHR—which portrays the Convention as the
minimum standard as it grants discretion to Member States to apply a higher standard of
protection—and Art. 53 ChFR which establishes that the standard of protection provided by the
Charter may not restrict the level of protection afforded, amongst others, by the Convention
system. The latter, according to Gragl, allows Member States to grant higher protection that the
level provided by the Charter. In view of this, Gragl rightly raises the question that the space to
provide provided to Member States to provide higher protection may certainly take precedence
over the EU law primacy principle as well. Such scenario would practically allow Member States
to apply the higher standard by superseding the EU-law primacy limitations, therefore harming the
uniformity of EU law and its effectiveness. Gragl rightly argues that the more extensive protection
doctrine naturally contradicts with the EU-law primacy doctrine. Gragl also questions why should
then the EU accede to the ECHR if it was ‘punished’ to comply with the principle of primacy of
EU law to the detriment of more extensive protection that the status quo. See: Gragl (2013),
pp. 103–105. This is also an argument raised by the EU Court in its Opinion 2/13.
122
A similar, but more general argument, may be also found at: Sera (1996), p. 177.
158 5 Status of ECHR and DAA in EU Legal Order

The chapter first presented an analysis of the main concepts and caselaw
arguments that set forth the status of international agreements in EU law, demon-
strating Luxembourg’s approach to international law at-large. Then, the chapter
embarked on an exhaustive analysis of the EU constitutional provisions with regard
to the Convention and the DAA, questioning the latter’s status from a constitutional
point of view. The analysis included six landmark arguments on the fact that the
Convention and DAA enjoy a primary law rank. Such ranking was envisaged to
serve in the context of good administration of the relationship between the Treaties
and the two instruments—from which the Treaties will benefit in the positive
sense—as opposed to an external interference that may harmfully affect the allo-
cation of powers and interpretation of EU law. Those being the core arguments
presented by the chapter, it is concluded that the Convention and DAA will enjoy a
privileged position within EU law, something which no other international agree-
ment may have reached before.
The chapter therefore concludes that although it may sound odd to consider the
Convention and DAA enjoying a privileged position in EU law, that fact alone does
not compromise or harm the internal balances of powers within the EU institutional
structure. It is not the aim to conclude that the Convention and DAA will produce
internal competition to the EU Treaties; rather, they will merely serve to interface
with regard to a more coordinated approach to bring the EU legal order into line
with the Convention system’s normal operation. On the other hand, providing a
privileged position to the Convention and DAA does not per se compromise the
autonomy of EU law, as the latter does not intend to prohibit EU law from allowing
the penetration and cohabitation of unharmful international law to its order. This
being the case—with the privileged position of the Convention and the DAA—one
may well support the conclusion that EU accession to the ECHR will likely make
the EU legal order even more open to international law, and will build a practice of
accommodating international agreements with the purpose of functionalizing a
natural relationship which Luxembourg and Strasbourg ought to have in a multi-
polar European law landscape. Whether this logic will be received in that format
from the Luxembourg Court remains to be seen from the future practice.

References

Akehurst M (1981) The application of the general principles of law by the Court of Justice of the
European communities. Br Yearb Int Law 52(1):29–51
Balfour ADJ (2005) Application of the European Convention on Human Rights by the European
Court of Justice. Harvard Law School Student Scholarship Series. Paper 4, pp 1–57
Besson S (2009) European legal pluralism after Kadi. Eur Constitut Law Rev 5(2):237–264
Callewaert J (2014) The accession of the European Union to the European Convention on Human
Rights. Council of Europe, Strasbourg
Carruthers S (2009) The Treaty of Lisbon and the reformed jurisdictional powers of the European
Court of Justice in the field of justice and home affairs. Eur Hum Rights Law Rev 6:784–804
Cartabia M (2009) Europe and Rights: taking dialogue seriously. Eur Constitut Law Rev 5(1):5–31
References 159

Cassese A (1985) Modern constitutions and international law. Martinus Nijhoff, Hague
Craig P (2013) EU accession to the ECHR: competence, procedure and substance. Fordham Int
Law J 36(5):1114–1150
De Londras F, Kingston S (2010) Rights, security, and conflicting international obligations:
exploring inter-jurisdictional judicial dialogues in Europe. Am J Comp Law 58:359–414
de Rivery EM, Chassaing E (2013) EU accession to the ECHR: bringing about an aggiornamento
for EU case law? Overview of ECHR and EU case law. e-Competitions 1–16, N 61327
De Schutter O (2007) Accession of the European Union to the European Convention on Human
Rights, p 13. Available at: http://www.statewatch.org/news/2007/sep/decchutte-contributin-
eu-echr.pdf
Eckes C (2013) EU accession to the ECHR: between autonomy and adaptation. Mod Law Rev 76
(2):254–285
Editorial (2013) Human rights: member state, EU and ECHR levels of protection; P.S. Catalonia;
Why does it take so long for my article to be published?; In this Issue: Human rights: member
state, EU and ECHR levels of protection. Eur J Int Law 24:471–475
Eeckhout P (2004) External relations of the European Union: legal and constitutional foundations.
Oxford University Press, Oxford
Fitzmaurice G (1957) The general principles of international law considered from the standpoint
of the rule of law. Collected Courses of the Hague Academy of International Law, vol 92. Brill|
Nijhoff, Boston
Franklin C (2010–2011) The legal status of the EU Charter of fundamental rights after the Treaty
of Lisbon. Tilburg Law Rev 15: 137–162
Garcı́a RA (2002) The general provisions of the Charter of fundamental rights of the European
Union. Eur Law J 8(4):492–514
Gragl P (2013) The accession of the European Union to the European Convention on Human
Rights. Hart Publishing, Oxford
Groussot X et al (2011) EU accession to the European Convention on Human Rights: a legal
assessment of the draft accession agreement of 14th October 2011. Fondation Robert
Schuman/European Issues N 218/7 November 2011
Harding C (2000) The identity of European law: mapping out the European legal space. Eur Law J
6(2):128–147
Hart JW (2010) The European human rights system. Law Libr J 102(4):553–559
Heringa AW, Verhey L (2011) The EU Charter: text and structure. Maastricht J Eur Comp Law 8
(1):11–32
Hofmann HCH, Mihaescu BC (2013) The relation between the Charter’s fundamental rights and
the unwritten general principles of EU law: good administration as the test case. Eur Constitut
Law Rev 9(1):73–101
Jacobs FG (2008) Direct effect and interpretation of international agreements in the recent case
law of the European Court of Justice. In: Dashwood A, Maresceau M (eds) Law and practice of
EU external relations: salient features of a changing landscape. Cambridge University Press,
Cambridge
Jacque JP (2011) The accession of the European Union to the European Convention on Human
Rights and fundamental freedoms. Common Market Law Rev 48(4):995–1023
Kaczorowska A (2013) European Union law, 3rd edn. Routledge, London
Kapteyn PJG (1974) The “Domestic” law effect of rules of international law within the European
community system of law and the question of the self-executing character of GATT rules. Int
Lawyer 8(1):74–82
Klabbers J (2002) International law in community law: the law and politics of direct effect. Yearb
Eur Law 21:263–298
Koutrakos P (2006) EU international relations law. Hart Publishing, Oxford
Krüger HC (2002–2003) Reflections concerning accession of the European Communities to the
European Convention on Human Rights. Pa State Int Law Rev 21(1):89–99
160 5 Status of ECHR and DAA in EU Legal Order

Kumm M (2005) The jurisprudence of constitutional conflict: constitutional supremacy in Europe


before and after the constitutional treaty. Eur Law J 11(3):262–307
La Pergola A (1993) The relationship between international and domestic law: traditional prob-
lems and new trends. In: ‘European Commission for Democracy Through Law (Venice
Commission),’ ‘The Relationship Between International and Domestic Law,’ proceedings of
the UniDem Seminar organized in Warsaw on 19–21 May 1993 in cooperation with the
University of Wroclaw and the Poznan Human Rights Centre. Science and Technique of
Democracy No. 5. CDL-STD(1993)005. Strasbourg, 15 September 1993
Landau EC (2008) A new regime of human rights in the EU? Eur J Law Reform 10(4):557–575
Lawson R (2005) Human rights: the best is yet to come. Eur Constitut Law Rev 1(1):27–37
Leal-Arcas R (2001) The European Community and mixed agreements. Eur Foreign Affairs Rev 6
(4):483–513
Lenaerts K (2012) Exploring the limits of the EU Charter of fundamental rights. Eur Constitut Law
Rev 8(3):375–403
Lenaerts K, Corhaut T (2006) Of birds and hedges: the role of primacy in invoking norms of EU
law. Eur Law Rev 31(3):287–315
Lenaerts K, de Smijter E (2001) The Charter and the role of the European Courts. Maastricht J Eur
Comp Law 8(1):90–101
Lenaerts K, Gutiérrez-Fons JA (2010) The constitutional allocation of powers and general
principles of EU law. Common Market Law Rev 47(6):1629–1669
Licková M (2008) European exceptionalism in international law. Eur J Int Law 19(3):463–490
Lock T (2010) EU accession to the ECHR: implications for judicial review in Strasbourg. Eur Law
Rev 35(6):777–798
Lock T (2012) End of an epic? The draft agreement on the EU’s accession to the ECHR. Yearb Eur
Law 31(1):162–197
Maniruzzaman MAF (2001) State contracts in contemporary international law: Monist versus
Dualist controversies. Eur J Int Law 12(2):302–329
Martı́n J, De Nanclares P (2013) The accession of the European Union to the ECHR: more than
just a legal issue. Working Papers on European law and Regional Integration WP IDEIR n 15.
Instituto de Derecho Europeo e Integracion Regional (IDEIR), Universidad Complutense
Martinico G (2012) Is the European Convention going to be ‘supreme’? A comparative-
constitutional overview of ECHR and EU law before national courts. Eur J Int Law 23
(2):401–424
Mendez M (2010) The enforcement of EU agreements: bolstering the effectiveness of treaty law?
Common Market Law Rev 47(6):1719–1756
Miiller CA (2007) Fundamental rights in multi-level legal systems: recent developments in
European human rights practice. Interdiscip J Hum Rights Law 2(1):33–60
Morano-Foadi S, Andreadakis S (2011) Reflections on the architecture of the EU after the Treaty
of Lisbon: the European judicial approach to fundamental rights. Eur Law J 17(5):595–610
Morijn J (2006) Balancing fundamental rights and common market freedoms in union law:
Schmidberger and Omega in the light of the European constitution. Eur Law J 12(1):15–40
Muir E (2014) The fundamental rights implications of EU legislation: some constitutional
challenges. Common Market Law Rev 51(1):219–246
Neframi E (2010) The duty of loyalty: rethinking its scope through its application in the field of EU
external relations. Common Market Law Rev 47(2):323–359
O’Meara N (2011) “A More Secure Europe of Rights?” The European Court of Human Rights, the
Court of Justice of the European Union and EU accession to the ECHR. German Law J 12
(10):1813–1832
Odermatt J (2014) The EU’s accession to the European Convention on Human Rights: an
international law perspective. Leuven Center for Global Governance Studies. Working Paper
No. 136
References 161

Olsen JB (2009–2010) Protecting fundamental rights and the evolving roles of the court of justice
of the European Union and the European Court of Human Rights; Europeanisation in action at
the supranational level. Edinburgh Student Law Rev 1(3):54–73
Peers S (2012) The future of EU treaty amendments. Yearb Eur Law 31(1):17–111
Quirico O (2010) Substantive and procedural issues raised by the accession of the EU to the
ECHR. Ital Yearb Int Law 20:31–53
Reestman JH (2005) Primacy of union law. Eur Constitut Law Rev 1(1):104–107
Ribble MR (2010–2011) I don’t trust your judgment: the European Convention on Human Rights
meets the European Union on New Grounds? Pa State Int Law Rev 29(1):211–232
Rodley NS (1989) Human rights and humanitarian intervention: the case law of the world court.
Int Comp Law Q 38(2):321–333
Sera JM (1996) The case for accession by the European Union to the European Convention for the
protection of human rights. Boston Univ Int Law J 14:151–186
Shaw M (2008) International law, 6th edn. Cambridge University Press, Cambridge
Shelton D (2003) The boundaries of human rights jurisdiction in Europe. Duke J Comp Int Law
13:95–153
Sweet AS (2009) Constitutionalism, legal pluralism, and international regimes. Indiana J Global
Leg Stud 16(2):621–645
Tridimas T (2006) The general principles of EU law, 2nd edn. Oxford University Press, New York
van Rossem JW (2009) Interaction between EU law and international law in the light of Intertanko
and Kadi: the dilemma of norms binding the member states but not the community. Neth Yearb
Int Law 40:183–227
Von Bogdandy A et al (2012) Reverse Solange – protecting the essence of fundamental rights
against EU member states. Common Market Law Rev 49(2):489–519
Weiß W (2011) Human rights in the EU: rethinking the role of the European Convention on
Human Rights after Lisbon. Eur Constitut Law Rev 7(1):64–95
Wessel RA (2008) The EU as a party to international agreements: shared competences, mixed
responsibilities. In: Dashwood A, Maresceau M (eds) Law and practice of EU external
relations: salient features of a changing landscape. Cambridge University Press, New York
Wetzel JR (2003) Improving fundamental rights protection in the European Union: resolving the
conflict and confusion between the Luxembourg and Strasbourg courts. Fordham Law Rev
71:2823–2862
Wilson RR (1964) International law in new national constitutions. Am J Int Law 58(2):432–436
Woolcock S (2005) European Union trade policy: domestic institutions and systemic factors. In:
Kelly D, Grant W (eds) The politics of international trade in the twenty-first century: actors,
issues and regional dynamics. International political economy series. Palgrave, Basingstoke
Chapter 6
Attribution of Liability Under
the Co-respondent Mechanism

6.1 Introduction

This chapter offers an analysis of the individual complaint mechanism in


EU-related cases, with a specific view to the recently innovated co-respondent
mechanism. The latter being a core but also the most essential device to ensure
the autonomy of EU law, and to functionalize the multifaceted nature of EU–
Member States relation before the Convention system. The chapter responds to
the research sub-question: ‘How may the distribution of burden for ECHR viola-
tions be shared between the EU and Member States under the co-respondent
mechanism, and where do loopholes remain?’ Essentially, this chapter engages
with the model of liability sharing between the EU and Member States which the
DAA installs within the context of EU Treaties’ prescriptions and foundational
limitations.
The chapter starts by analyzing the EU Treaties’ specific guarantees made to
ensure that the division of liability between the EU and Member States before the
Strasbourg Court respect EU law autonomy at-large, and purposely protect EU law
from being interpreted outside of the EU law landscape. This is not merely
examined in the context of jurisdictional exclusivity of the Luxembourg Court,
but also in the context of Strasbourg’s likely international outlook on the EU as a
Convention contracting party. The chapter then compares the nature of the
co-respondent mechanism with the third party intervention mechanism, outlining
the fundamental mechanical disparities and demonstrating that the former is far
more complex and possesses an outlying singular intention. The chapter also
comprehensively tackles the functional mechanics of the co-respondent mecha-
nism, paying special attention to implications that it may pose once the admissibil-
ity test is performed before the Strasbourg Court. The co-respondent mechanism is

© Springer International Publishing Switzerland 2015 163


F. Korenica, The EU Accession to the ECHR, DOI 10.1007/978-3-319-21759-8_6
164 6 Attribution of Liability Under the Co-respondent Mechanism

then tested against DARIO standards, and evaluated through the latter’s lens. On
the latter, the chapter carefully draws the core differences that the International Law
Commission would design through DARIO in terms of distribution of liability of
international organizations before another international organization (EU vis-a-vis
the Convention system being an example), keeping the note that the co-respondent
mechanism appears to be a mechanism intended to preserve EU law autonomy
rather than oriented towards better attributing liability to the EU. Certain answers as
to whether the latter practice may hinder the goal of the Convention’s human rights
protection is provided through a careful analysis of these issues.
The chapter also offers in depth scrutiny of the means via which friendly
settlements will be processed within the co-respondent process, noting the main
difficulties to reconcile the current understanding of friendly settlement within the
role established by the DAA. The same is also assessed in terms of the unilateral
declarations between the EU and Member States once they appear jointly before the
Strasbourg Court. The chapter then carefully and substantively examines situations
and normative directions in which the EU, its Member States or jointly appear
together as co-respondents before the Strasbourg Court, and the main loopholes that
the DAA tolerates in this course. This is not only examined in terms of how parties
may join as co-respondents, but also the nature and effect that the Strasbourg Court
may deliver in its decisions with regard to co-respondents. The chapter then
embarks in analysing the central ambiguities identified in the co-respondent mech-
anism, placing special emphasis on situations which may erode EU law autonomy
as a result of Strasbourg Court rulings/decisions.
In its final part, the chapter also examines the issue of appeals of co-respondents
to the Grand Chamber of the Strasbourg Court, examining how the convenience of
this mechanism may be risked from the potential disjointing between the respon-
dent and co-respondent in the appeal procedure. Finally, the chapter—after provid-
ing three practical tests on the co-respondent mechanism—makes a grounded
comparison between it and two other international instruments that provide joint
liability distribution models, namely UNCLOS and UNCILDCSO. Strengths and
weaknesses are part of the latter comparison.
The chapter concludes that the co-respondent mechanism is an internationally
appreciated innovation of the DAA, however it will lead to numerous uncertainties
that may provoke EU law autonomy erosion and/or impair the functionality of the
Convention system and EU responsibility towards the latter.
6.2 A Background on the Complex Task of Attributing Liability in a Post. . . 165

6.2 A Background on the Complex Task of Attributing


Liability in a Post-accession Scenario

The Convention’s primary role is to protect individual rights,1 ensuring that indi-
viduals2 are given specific protection that guarantees their personal rights enjoy-
ment and/or possession.3 It must be said that the core function of the Convention,
therefore, is to put in practice a system for ensuring the protection of individual
rights rather than serving the inter-state conflict resolution regarding human rights
issues.4 This foundational attribute of the Convention has been seen as the most
significant dimension of its international relevance and distinctiveness compared to
the vast majority of other international human rights treaties. But beyond this, a
‘major issue today seems to be one of the coherence of human rights protection in
Europe’5 which intensifies the debate as ‘[. . .] a legal or jurisdictional conflict
between these two ‘supreme’ courts [in Europe] would erode confidence in the
rule of law with the result that the authority of both systems would suffer. Even
worse, the end result of such a conflict would be the erosion of fundamental rights
protection for people.’6
To note, the Convention not only provides for the legal recognition of individual
rights, but also ensures a mechanism to judge individual rights-related cases. This
characteristic makes the Convention a genuine avenue wherein individuals may
seek external human rights shelter, although up to now the Strasbourg system has
called inadmissible ratione personae applications against EU acts.7 One must
mention here that, as opposed to the inter-state conflicts that could arise in light

1
Orakhelashvili (2003), p. 529; See also: Odermatt (2014), p. 9.
2
See e.g.: van Dooren (2009), p. 50; See also: Greer and Williams (2009), p. 463, where the
question remains: ‘is the trans-national protection of human rights in Europe a matter of ‘individ-
ual’, ‘constitutional’ or ‘institutional’ justice?’. This being the doubt, some do argue that the
Convention is not nevertheless very appropriate for very ‘individual’ human rights protection; See
also: Sweet (2009), p. 624/5.
3
From a historical perspective, the individual complaint mechanism foreseen by the Convention
was a rather significant development in terms of human rights protection at the time when it was
initiated, adding that in that period there was no view beyond the inter-state mechanism in terms of
international human rights treaties. See: Kerson (1961), p. 177; See also: Sera (1996), p. 154/5.
4
For instance, with respect to Article 2 ECHR the Court held that High Contracting Parties are
bound to take appropriate steps to ensure the right to life of those within their jurisdiction. See:
ECtHR, LCB v UK, Application no. 14/1997/798/1001, Judgment, Strasbourg, 9 June 1998, para.
36; With regard to the Convention as a core international example of international human rights
law protection, see: Shelton (2003), p. 123.
5
Higgins (2003), p. 1; See more on the need for coherence of human rights, with accession being a
core moment: Joris and Vandenberghe (2008–2009), p. 12.
6
Olsen (2009–2010), p. 56; See also: Harding (2000), p. 142, who speaks on the influence of
ECHR system on the legal products of the Luxembourg Court.
7
ECtHR, Confédération Française Démocratique du Travail (CFDT) v European Communities
[1979]; Accession will allow individuals to request individual responsibility of the EU before the
Strasbourg Court. See on this e.g.: Phelps (2006), p. 278/9.
166 6 Attribution of Liability Under the Co-respondent Mechanism

of the Convention’s framework, the individual protection of human rights by the


Strasbourg Court is called into question when it comes to a two-level membership—
which might either be of an origin of dual/competing federalism or cooperative
federalism in the context of the EU–Member States’ relationship8—that will appear
in the aftermath of EU accession to the Convention.9 This said, it is important to
note that once the EU accedes to the Convention, an individual might face the
problem of who to call into responsibility for his/her own right violation: the EU or
Member State.10 On the other hand, either the EU or the Member State11 could be
outside of the liable entity with regard to a violation that should have been
attributed to it, or vice versa. These kinds of situations have led the Drafting
Group to devise the mechanism referred to as the Co-respondent Mechanism12 in
the DAA, which is supposed to ensure that one party may join the proceedings in a
certain case being adjudicated before the Strasbourg Court if it considers that such
access in the procedure is necessary to better channel the liability and ensure that
the Court targets in the right direction. To note, the Strasbourg Court has had varied
and blurred experiences of shared responsibility with multiple state parties,13
something that may start to be better developed post accession. Relating to the
latter practice, Sarvarian rightly argues that the Strasbourg Court’s experience of
attribution of liability with regard to the UN, NATO and EU is rather defective and
notionally vague. To this background, Sarvarian argues, the EU desired to see a
special mechanism such as the co-respondent mechanism in order to better ensure
the certainty of rules on attribution of responsibility from the Convention system.14
This is why this chapter deserves considerable attention with regard to the question
of effectiveness of EU accession to the ECHR.

8
de Búrca (2005), p. 95.
9
Literature on the accession of the EU to the ECHR and the possible implications for individual
judicial review in Strasbourg has grown in recent years: See e.g.: Lock (2010), p. 777; Jacque
(2011), p. 995–1023; Lock (2011), p. 1025–1054, etc;
10
A normal feature of the increasing legal pluralism being crafted with the latest developments in
Europe. See: van Rossem (2009), p. 206.
11
Cf.: Ryngaert (2011), p. 1005, who argues that in Gasparini case the Strasbourg Court never-
theless ruled that it may call Member States responsible if the international organization to which
they are members has a ‘structural lacuna’ of human rights protection. Based on the Gasparini
approach, one may argue that the Strasbourg Court has not fully omitted from finding Member
States responsible for their membership in IOs, if the latter have ‘structural lacuna’ somewhere in
their framework (including secondary law of EU, one may say). This needs be compared to the
DARIO’s standards applied upon the EU scenario.
12
Some dispute its need and argue that it will prove very complex for usage: Conforti
(2010), p. 85.
13
den Heijer (2013), p. 438.
14
Sarvarian (2014), p. 102.
6.2 A Background on the Complex Task of Attributing Liability in a Post. . . 167

Upon Whom to Place the Blame would seem the most consistent phrase that
would indicate the rationale on which the co-respondent mechanism was built.
There is a theoretical and practical ongoing dilemma to the question ‘which is the
prior or supreme order’,15 that of the EU or the Member States.16 ‘Either the
national legal order is subordinate to the (higher) legal order of the EU or the EU
legal order (in the national territory) is subordinate to the (higher) national legal
order.’17 Since there is no definite answer to this, and there are conflicting stances
on this, the only way to hold the EU and Member States responsible for EU primary
and secondary law without troubling these waters would be to apply a ‘muddling
through’ mechanism that establishes a mutual share of burden. That being an
important underlying rationale, one should stress that although the question of
upon whom placing the blame might seem technical in nature, its legal relevance
gives the impression of an extremely important topic with regard to the overall
integrity, autonomy, and fairness in the relationship between the Convention, EU
and Member States, as well as the relationship between the EU legal order and
Member States legal orders. Upon whom to place the blame, therefore, might
become the point of tension that would amuse many of the core principles already
in existence in the current EU legal order, or even threaten the stability of the
unharmed relationship until today between the laws of the EU and Member States.
This chapter provides a detailed examination of the relevance and function of the
co-respondent mechanism within the entirety of the Convention, EU and Member
States legal affairs, and examines in-depth the legal implications that could derive
therefrom. A comprehensive examination of the supposed effects of such mecha-
nism will reveal the more practical effects that the DAA will produce in the daily
life of interaction in the above triangle. International law principles, being revisited
with the co-respondent mechanism, will see many innovations whose success will
need time to be tested.

15
Cf.: Lenaerts and de Smijter (2001), p. 96 (‘Not only does such divergence entail a dilemma for
the national judiciary bound as it is by two distinct forms of legal loyalty, it also increases the risk
of liability of the EU Member States for violations of the ECHR committed by the institutions of
the Union.’).
16
It is interesting to mention the fact that the EU Treaties oblige the EU and its Member States to
build a level of consistency in their external action policy. This level of consistency—seen from an
international liability viewpoint—somehow favors models of mutual responsibility for EU and its
Member States when it comes to violations consumed by the EU and its Member States in their
external relations. On the former note, see e.g.: Wessel (2000), p. 1135–1171.
17
Barents (2009), p. 429.
168 6 Attribution of Liability Under the Co-respondent Mechanism

6.3 EU Treaty-Based Provisions: Which Were the Initial


‘Peculiar’-Related Instructions for DAA?

The Lisbon Treaty, which [. . .] ‘as we shall see, [is] in part comparable to a
Constitution for the European Community’,18 introduced numerous standards to
ensure the protection of EU legal order autonomy,19 and the guidelines relating to
the latter. In doing so, the Treaty provided for Protocol 8 which sets some basic
limitations—condicio sine qua non—with regard to the accession of the EU to the
Convention. That specific limitation being inclusive, Protocol 8 (1) amongst others
reads:
The agreement relating to the accession of the Union to the European Convention on the
Protection of Human Rights and Fundamental Freedoms (hereinafter referred to as the
‘European Convention’) provided for in Article 6(2) of the Treaty on European Union
shall make provision for preserving the specific characteristics of the Union and Union
law, in particular with regard to:
(a) the specific arrangements for the Union’s possible participation in the control bodies of
the European Convention;
(b) the mechanisms necessary to ensure that proceedings by non-Member States and
individual applications are correctly addressed to Member States and/or the Union as
appropriate.20

Protocol 8 (1) clearly sets the limits to the negotiating authorizations for the EU,
meaning that this is a limitation that obliges the EU internally with regard to its
negotiating mandate with the Council of Europe. Such limitation,21 however, could
not bind the Council of Europe or the Strasbourg Court at a later stage (post-
accession phase). Two are the principles deriving from Protocol 8 (ceteris paribus):
first, that the DAA should explicitly provide for the preservation of the Union

18
Jacobs (2007), p. 13.
19
On the autonomy of the EU legal order: See: Court of Justice of EU, Opinion 1/91 [1991] ECR
I-6079, para. 35; Court of Justice of EU, Opinion 1/00 [2002] I-3493, paras. 11–13; Court of
Justice of EU, Case C-450/03, Commission of the European Communities v Ireland (MOX Plant),
[2006] ECR I-04635, paras. 123–124; Court of Justice of EU, Joined Cases C-402/05 P and C-415/
05 P, Yassin Abdullah Kadi and Al Barakaat International Foundation v Council of the European
Union and Commission of the European Communities, [2008] ECR I-06351, para. 282; Court of
Justice of EU, Opinion 1/09[2011] ECR I-1137, para. 67; See generally on this topic, at:
Barents (2004).
20
Gragl points out that some of the Protocol 8 prohibitions are written to address the Member
States’ concern that post accession the EU Court would possibly engage with a form of interpre-
tation of the Convention that would take more competences from the Member States. See: Gragl
(2013), p. 96.
21
To this end, in addition, Protocol 8(2) to the EU Treaties reads: ‘The agreement referred to in
Article 1 shall ensure that accession of the Union shall not affect the competences of the Union or
the powers of its institutions. It shall ensure that nothing therein affects the situation of Member
States in relation to the European Convention, in particular in relation to the Protocols thereto,
measures taken by Member States derogating from the European Convention in accordance with
Article 15 thereof and reservations to the European Convention made by Member States in
accordance with Article 57 thereof.’
6.3 EU Treaty-Based Provisions: Which Were the Initial ‘Peculiar’. . . 169

characteristics,22 in a way to keep the autonomy of the EU court system sealed,23


and second, that the Accession Treaty specifically provide for the preservation of
the Union law individuality.24 The latter is the point that needs further discussion.25
In light of such limitation, Protocol 8 requires that the DAA in particular preserve
the Union law individuality with regard to the ‘the mechanisms necessary to ensure
that proceedings by non-Member States and individual applications are correctly
addressed to Member States and/or the Union as appropriate.’26 With regard to the
protection of EU law peculiarities—namely the EU law autonomy—it is important
to bring Strasbourg’s landmark case T.I. v. United Kingdom, where it established
that: ‘Where States establish international organisations, or mutatis mutandis inter-
national agreements, to pursue co-operation in certain fields of activities, there may
be implications for the protection of fundamental rights. It would be incompatible
with the purpose and object of the Convention if Contracting States were thereby

22
See e.g.: Ribble (2010–2011), p. 225; Jacque (2011), p. 1011; White (2010), p. 442; See also:
Gragl (2013), p. 91/2, who argues that the accession agreement should not only preserve the
characteristics of the EU legal order but also entail as little as possible changes to the Convention
system, in order for the latter to uphold its capacity to serve effective human rights protection; On
such core characteristics of EU law, see the Luxembourg’s landmark opinion on ‘Draft Agreement
on a European Common Aviation Area, Re(Opinion 1/00), op. cit. note 19.
23
See in this regard the indispensability of this characteristic of the EU law, at: Court of Justice of
EU, Opinion 1/92 [1992] E.C.R. I-2825, para. 24, and, Court of Justice of EU, Commission v
Ireland (C-459/03) [2006]; Ribble (2010–2011), p. 225 (‘This provision poses an interesting
question: what will affect the power of EU institutions? Without an additional explanation, it is
unclear whether an adverse ruling by the ECHR would inhibit an EU institution’s power.’); See
also: Odermatt (2014), p. 16; See also: Gragl (2013), p. 265, who rightly argues that ‘the
negotiators emphasised that the special legal characteristics of the Union must be taken into
account, but not at the price of impairing the Convention’s human rights protection system.’ One
needs to argue that, however, negotiators of the DAA were not that well courageous to maintain
this promise until the end, as the DAA demonstrates that there are certain loopholes which do
affect negatively the protection system under the Convention post-accession.
24
However, one must also mention that in light of international law the EU may not use Protocol
8 afterwards (post-accession) to denounce its obligations undertaken with the DAA, although
DAA may perhaps seem later on as interfering to the limitations of Protocol 8. See in this regard:
Art. 46.2 VCLTSIOIO, which reads: ‘An international organization may not invoke the fact that its
consent to be bound by a treaty has been expressed in violation of the rules of the organization
regarding competence to conclude treaties as invalidating its consent unless that violation was
manifest and concerned a rule of fundamental importance.’
25
This is written inter alia to mean that the Convention system may not interfere to the division of
responsibilities between the EU and its Member States. The EU Court—in this regard—has ruled
in Ruling 1/78 that it is not necessary to set out the division of powers between the EU and its
Member States in an international agreement. However, it would be ‘sufficient to state to the other
contracting parties that the matter gives rise to a division of powers within the Community, it being
understood that the exact nature of that division is a domestic question in which third parties have
no need to intervene.’ (Court of Justice of EU, Ruling 1/78 (on the Draft Convention of the
International Atomic Energy Agency on the Physical Protection of Nuclear Materials, Facilities
and Transports) [1978] ECR 2151, para. 35). The latter seems well embedded in the Protocol
8, however it is not something faultlessly guaranteed by the DAA.
26
See also: den Heijer (2013), p. 434.
170 6 Attribution of Liability Under the Co-respondent Mechanism

absolved from their responsibility under the Convention in relation to the field of
activity covered by such attribution.’27 It is perfectly obvious—and arguably fair—
that T.I. v. United Kingdom—if applied against the EU scenario—would not allow
the EU law autonomy issue to become something that either distorts or hinders the
nature and mission of the Convention system; the author holds the argument that
this standard needs therefore be carefully considered against the DAA mechanisms
devised to uphold the EU law autonomy post-accession. Imposing EU-law-auton-
omy related benchmarks on the Convention and DAA—if such benchmarks would
result in hindering the expected output of the Convention system—would arguably
be contrary to the standard required by T.I. v. United Kingdom.
In light of the above provision of Protocol 8, it is argued that the specification is
made to connote to a mechanism that addresses the violations of the Convention to
the correct respondents, ensuring that the EU and/or Member State(s) are given
correct direction as to their individual responsibility arising from an individual case
before the Strasbourg Court while the EU law peculiarities are not affected.28 This
should be crafted in a way that would accommodate the applicant’s legal situation
in the sense that the applicant should ‘know’ how to attempt it.29 This would not be
at all problematic if it is the EU which is both legislating and directly implementing
its own law towards an individual party (the cases of direct implementation of EU
law). This may particularly be the case in fields of law ‘such as trademarks,
antitrust, and state aid, where it is the EU which makes decisions that affect the
individual directly.’30 In the latter case, the applicant would attack the EU as a
normal High Contracting Party before the Strasbourg Court, whereas the EU would
be the sole respondent with no right for anyone to initiate the co-respondent
mechanism. However, in cases when the EU legislates but does not directly
implement its law—the latter is implemented by its Member States indirectly—
then that scenario becomes problematic in terms of whom to call into responsibil-
ity.31 It is obvious from the latter scenario that the nature of the EU law allows for

27
ECtHR (European Court of Human Rights) (3d sect.), T.I. v. United Kingdom (Appl. No. 43844/
98), decision (inadmissibility) of 7 March 2000, page. 15.
28
Such peculiarities mainly refer to standards deriving from Costa v ENEL [1964], Opinion 1/91
[1991] and Commission v Ireland (C-459/03) [2006], op. cit. note 19; See also on the ‘correct
addressee’ debate during the negotiation process: Groussot et al. (2011), p. 7; Lock (2012),
pp. 165 et seq; Eckes (2013), p. 267.
29
E.g.: Lock (2010), p. 780; See also: Gragl (2013), p. 147, who argues that the idea of making it
simpler for the applicant to know how to proceed a certain violation to the Strasbourg Court
through the entity that implemented it—rather than the one that legislated it—is more rational and
the best possible choice. In addition, Gragl argues that this model saves better the autonomy of EU
law as it does not engage the Strasbourg Court with the adjudication of the levels of responsibility
within EU law that it would need to tackle if the co-respondent model would have not been
selected.
30
Streinz (2013), p. 1.
31
See on this, e.g.: Streinz (2013), p. 1.
6.3 EU Treaty-Based Provisions: Which Were the Initial ‘Peculiar’. . . 171

an insufficient distinction between the EU law and Member States law,32 especially
in cases when there is a legal act adopted by the EU but implemented by the
Member States (illustrated in Fig. 6.1).33 ‘Should a violation be found [in that
scenario], the Member State concerned is also unable to take responsibility for
removing the violation.’34 To note is the fact that as the majority of EU law is
implemented by Member States, with the EU only rarely acting directly in the
implementation of law in individual cases, the problem that will most regularly
appear will be whom to call into responsibility35: the EU as the legislator of the
supposed human-right violating legal provisions or the Member States as the
implementers of such EU legal provisions, or both. As Bjoerklund puts it, the
core problem in this regard is: ‘Who – if anyone – is responsible for what?’36
This option rests mainly with the applicant, who must also meet the admissibility
conditions and name the ‘violator’ of his/her right deriving from the Convention,
the assumed violator being ‘the party which has acted vis-a-vis the applicant in the
concrete case.’37 The ‘targeted’ addressee, the respondent, might be or not the real
violator of the human right for the purpose of which an applicant lodged the
application before the Strasbourg Court, an issue which brings the question of
whether the Member States should be held responsible even if they did not have
state discretion in implementing an EU legal act which infringed one’s rights.38 If

32
One must add here that there is an ongoing tension between the EU Court and Member States’
courts, the former trying sometimes to ‘grab’ portions of competences from Member States. A core
example is case Secretary of State for the Home Department v. Akrich (Court of Justice of EU,
Secretary of State for the Home Department v. Akrich, Case C-109/01, [2003] E.C.R. I-9607, para.
61), which shows a tendency of the ECJ to get into the domain of Member States’ law territory.
This makes the overall picture of ‘who to call into responsibility’ (the EU or Member States)
becomes even more problematic.
33
E.g.: den Heijer (2013), p. 434; Raba (2013), p. 564; Gaja (2013), p. 1; See also: Lock (2012),
p. 175; Ribble (2010–2011), p. 221 (In terms of the complexity of this picture of legal borders,
‘[. . .]member states of the EU are in a precarious situation where they are bound to abide by both
EU law and the Convention. It is possible, for example, that a state could implement an EU
regulation and simultaneously violate the Convention.’); On the latter, see also: Wetzel (2003),
p. 2825; This may lead to Member States needing to choose between violating the primacy of EU
law or the Convention. See: Lenaerts and de Smijter (2001), p. 92.
34
Raba (2013), p. 564.
35
See e.g.: Art. 4 (1) ILC Articles on State Responsibility in this regard reads: ‘The conduct of any
State organ shall be considered an act of that State under international law, whether the organ
exercises legislative, executive, judicial or any other functions [. . .].’; The same approach has been
applied until now with regard to EU Member States’ responsibility for EU acts which allow no
state discretion. See also: ECtHR, Bosphorus Hava Yollari Turizm ve Ticaret Anonim Şirketi
v. Ireland, Application No. 45036/95 (Judgment of 30 June 2005), para. 153.
36
Bjoerklund (2001), p. 374.
37
Lock (2010), p. 784.
38
This was the case in e.g. Bosphorus Hava Yollari Turizm ve Ticaret Anonim Şirketi v. Ireland,
Application No. 45036/95, op. cit. note 35; However, the Strasbourg court was reluctant to review
the measure knowing that a Security Council Resolution was at stake. Moreover, the Court
(ECtHR) argued that EU offers human rights protection equivalent to that of the ECHR (see
172 6 Attribution of Liability Under the Co-respondent Mechanism

Fig. 6.1 The Multi-Level Commission of a Violation of ECHR in an EU-Member State(s) scenario

one would allow this confusion to appear, the Strasbourg Court would have to
engage in the justification of EU–Member States borders of competences with the
aim of targeting the correct respondent, an action that clearly would have gone
against EU law autonomy and the requirements deriving from Protocol 8 (1)b. It is
no innovation that international tribunals, such as Strasbourg, normally play the
function of ‘lawmaking [. . .by filling. . .] gaps in existing legal rules, or, in the
terminology of rational choice, complet[ing] incomplete contracts.’39 The Stras-
bourg Court’s role in the face of DAA may become that of completing the loopholes
of the latter, adding that there are in fact ‘too many’ loopholes in place. Therefore,
basically, DAA would need to make the Convention more complex and suitable for
this scenario, something that would go beyond the traditional notion of jurisdiction.
The latter is linked merely to the effective control, such as in Ilascu and others
v. Moldova and Russia, where the Court had ruled: ‘The exercise of jurisdiction is a
necessary condition for a Contracting State to be able to be held responsible for acts
or omissions imputable to it which give rise to an allegation of the infringement of

para. 155); For an analysis of the doctrine of equivalent protection, see De Hert and Korenica
(2012), p. 874–895. (In Bosphorus the EU Regulation was adopted relying on a UNSC Resolution;
Republic of Ireland implemented the Regulation with no margin of discretion as to the alteration of
the measure); Cf.: Conforti (2010), p. 85, who argues that the new mechanism must establish that
there should be no responsibility for the Member State if it exercised no discretion.
39
Alter and Helfer (2010), p. 566.
6.3 EU Treaty-Based Provisions: Which Were the Initial ‘Peculiar’. . . 173

rights and freedoms set forth in the Convention.’40 It seems clear that with the EU
as a High Contracting Party to the Convention, the Ilascu model on defining the
jurisdiction of the High Contracting Parties—as precondition to attribute Conven-
tion responsibility to them—will not seem viable anymore, as a more advanced
mechanism needs to be devised: in order to attribute liability to the EU as the
legislator of a certain violation, even though Member States undertake the effective
measures to violate the Convention. With the DAA, certainly, the concept on Art.
1 ECHR will need to provide a far more advanced model of attribution of liability
and definition of jurisdiction41 in cases when both the EU and its Member States
appear together as respondents.
To ensure that the above situations are resolved by respecting EU law autonomy,
and preventing the Strasbourg Court from engaging in the adjudication of who is
responsible for a supposed violation filed before it,42 the DAA has envisaged the
co-respondent mechanism,43 which is the apparatus that allows the EU or the
Member States to join the proceedings before the Strasbourg Court where an
individual case involves their legal ‘land’, and accordingly the addressees are
given the chance to assume their responsibility and/or competency in that area
more fairly and in solido.44 Having that aim in mind, the co-respondent mechanism
seems to be a very significant instrument that should ensure the integrity and
credibility of addressing the violations of the Convention in a multi-level system

40
ECtHR (GC), Ilascu and others v. Moldova and Russia, Appl. No. 48787/99, judgment of 8 July
2004, para. 311.
41
The traditional model of definition of jurisdiction was based on the notion of territory, with a
single level of entity being responsible in face of the Convention system. Accord.: The Court had
argued that it ‘must be considered to reflect this ordinary and essentially territorial notion of
jurisdiction, other bases of jurisdiction being exceptional and requiring special justification in the
particular circumstances of each case.’ (ECtHR, Bankovic v Belgium and others, Application
no. 52207/99, Grand Chamber, Decision as to the Admissibility, para. 61); For a critique to the
Strasbourg Court’s Bankovic approach—due to its vague nature on determining the notion of
jurisdiction—see e.g.: Martin (2003), pp. 106–109.
42
See a similar example, at: Lavranos (2006), pp. 464 et seq; Cf.: Lock (2010), p. 780, where the
author argues that the co-respondent mechanism should have been suitable for usage for one which
has no legal education.
43
Cf.: European Parliament’s Committee on Constitutional Affairs (in May 2010) had proposed a
different model for the ‘correct addressee’, by proposing this provision: ‘[A]ny application by a
natural or legal person concerning an act or failure to act by an institution or body of the Union
should be directed solely against the latter and that similarly any application concerning a measure
by means of which a Member State implements the law of the Union should be directed solely
against the Member State, without prejudice to the principle that, where there might be any doubt
about the way in which responsibility is shared, an application may be brought simultaneously
against the Union and the Member State.’ European Parliament Committee on Constitutional
Affairs Resolution (May 2010) (P7_TA(2010)0184), para. 9.
44
E.g.: Quirico (2010), p. 50.
174 6 Attribution of Liability Under the Co-respondent Mechanism

of authority and responsibility, wherein EU law autonomy seeks careful protection


from an external judicial body such as the Strasbourg Court.45 This would seem to
fit with Stone Sweet’s ‘new constitutionalism’ concept of the two European courts
becoming the constitutional courts of ‘Europe’.46
To make a short digression, the author would like to discuss previous examples
regarding Strasbourg’s examples involving shared/mixed responsibilities of Con-
vention High Contracting Parties. One such landmark case is the Ilse Hess v. United
Kingdom—which involved a prisoner’s claim that his Convention rights were
violated in prison and caused equally and simultaneously by four states (referred
to as powers). The claimant attacked the United Kingdom before the Strasbourg
Court—UK being one of those states part of the joint command of the prison—for
having violated his Convention rights. In dealing with the multi-parties’ responsi-
bility, the Court in this case concluded ‘that the responsibility for the prison at
Spandau, and for the continued imprisonment of Rudolf Hess, is exercised on a
Four Power basis and that the United Kingdom acts only as a partner in the joint
responsibility which it shares with the three other Powers.’47 The Court further
ruled that ‘the joint authority cannot be divided into four separate jurisdictions and
that therefore the United Kingdom’s participation in the exercise of the joint
authority and consequently in the administration and supervision of Spandau Prison
is not a matter “within the jurisdiction” of the United Kingdom, within the meaning
of Art. 1 of the Convention.’48 It seems clear that under the logic of Ilse Hess
v. United Kingdom, the Court had refused to accept to provide neither joint
responsibility for all parties—as only one of them was attacked before the Stras-
bourg Court—nor separation of joint responsibility into individual responsibility
for each party.49 Clearly, the DAA and the co-respondent mechanism establish a
rather more complex and new model of locating and attributing responsibility,
something that goes far beyond the scope of the practicability and rationale in
Ilse Hess. Still, the DAA does not provide any tool for separating the individual
responsibility of a party when there are many parties located as being jointly
responsible for a violation. However, it builds upon the idea that although it is
not possible for the Court to separate/divide joint responsibility into individual
ones, it may not refuse to call parties’ actions or omissions under their jurisdiction
and find them jointly liable—leaving the separation of their joint responsibility into

45
On the general trends of fragmentation of international law, wherein each treaty regime tries to
build its own autonomous system of rules of interface with others, and the overall treaty-conflict
resolution mechanisms, see two very interesting sources: Simma (1985), p. 111–136; and, ILC
(2006). U.N. doc. A/CN.4/L.702 of 18 July 2006.
46
Tabarelli (2013), p. 342, citing Stone Sweet.
47
ECtHR (European Commission of Human Rights), Ilse Hess v. United Kingdom, Appl.
No. 6231/73, Decision of 28 May 1975 on the admissibility of the application, p. 73/4.
48
ECtHR (European Commission of Human Rights), Ilse Hess v. United Kingdom, op. cit. note
47, p. 73/4.
49
See in this regard a wealthy contribution on the EU and its Member States under the ILC Articles
on Responsibility of International Organizations, at: Kuijper and Paasivirta (2004), p. 111–138.
6.3 EU Treaty-Based Provisions: Which Were the Initial ‘Peculiar’. . . 175

individual ones for their own internal methods of law—thereby assuring that the
claimants remain protected before the Convention although they have been violated
by (a multilevel) parties’ action or omission (clearly improving the efficacy of the
Convention compared to the Ilse Hess standard).
Coming back to the main issue over again, the Explanatory Report to the
Accession Treaty provides some central explanations on the rationale of the
co-respondent mechanism. It argues that the co-respondent mechanism is not a
procedural privilege for the EU or its Member States, ‘but a way to avoid gaps in
participation, accountability and enforceability in the Convention system. This
corresponds to the very purpose of EU accession and serves the proper administra-
tion of justice.’50 It is of course obvious that the mechanism will provide for the
addressees to become identified and properly called into responsibility with a more
sophisticated device, wherein the liable addressee is asked to improve the violation
of the human right deriving from the individual case. Although the Explanatory
Report sees the co-respondent mechanism as the means to ensure the proper
administration of justice, its voluntary nature will likely risk such an attribute of it.
It is important to mention here that EU liability (and jurisdiction) for violations
of the Convention is demarcated in the context of its outer border with Art.
1 (3) DAA, which reads:
Accession to the Convention and the protocols thereto shall impose on the European Union
obligations with regard only to acts, measures or omissions of its institutions, bodies,
offices or agencies, or of persons acting on their behalf. Nothing in the Convention or the
protocols thereto shall require the European Union to perform an act or adopt a measure for
which it has no competence under European Union law.

DAA therefore makes it clear that although the possibility might exist for the EU
Member States to become liable for Convention breaches even if the legal basis for
the violation stems from an EU law provision,51 which leaves no state discretion,
the counter scenario might not come under consideration. The jurisdiction of EU in
the context of Art. 1 ECHR is therefore rather well clarified not only in terms of the
subjects (persons) but also of scope of responsibility that may be attributed to it on
that basis.52 It becomes clear from Art. 1 (3) DAA that the EU may only hold
responsibility for the acts or omissions of its institutions, meaning EU secondary

50
CDDH-UE (2011)16fin, para. 33 et seq.
51
See very generally, also: Rieder (2005), p. 102; Cf.: Lenaerts (2012), p. 377, on the responsi-
bility on the application of the Charter of Fundamental Rights.
52
See also: Lock (2012), p. 174/5; and it includes all policy fields of EU, including CFSP. See on
this, e.g.: Balfour (2007), p. 214; Compare to: Art. 6(1) of Annex 9 of UNCLOS (‘Parties which
have competence under article 5 of this Annex [the part on the declaration of competence of each
contracting party] shall have responsibility for failure to comply with obligations or for any other
violation of this Convention.’); Clearly, the DAA establishes another model of clarifying the
jurisdiction on basis of the legal acts which EU produces on basis of its own law, rather than on
basis of a declaration of competence which the Convention system does not require neither for the
EU nor for its Member States. The DAT seems more reliable in terms of claimants’ ability to know
when EU is responsible for an alleged violation and when not.
176 6 Attribution of Liability Under the Co-respondent Mechanism

law,53 excluding liability for the Treaties and/or the Member States acts or omis-
sions when they exercise state discretion54 (Accord: Bosphorus, Cf.: M.S.S.55).56
CFSP measures are also attributed to the EU on this basis57 (including individual
decisions under Art. 25(2c) TEU; even those having an extraterritorial nature such
as those ruled in Bankovic and others v. Belgiumor Coard v. United States58). To
digress, the jurisdiction on CFSP acts and omissions was ruled as incompliant with
the EU Treaties in Opinion 2/13 (see supra the subchapter on Opinion 2/13).
Coming back to the previous issue, in terms of Art. 1 (3) DAA, as a point of further
examination, this means in practice that—in cases of positive obligations—the EU
may not be called to legislate in areas in which it has no competence. Strasbourg’s
approach in that regard would be one which inquires into EU law to determine the
competent authority for that omission—the EU or its Member States59: the latter

53
den Heijer (2013), p. 433.
54
On the latter, see e.g. the EU Court’s reasoning in the Bosphorus case, at: Hoffmeister
(2006), p. 446.
55
On the scope of Member States’ discretion on ‘implementing measures’ of the EU act, see:
Nanopoulos (2013), p. 277/8 et seq.
56
Of course, this does not influence in any form the Member States’ human rights obligations on
their exclusive competences. Cf.: Court of Justice of EU, Cinetheque S.A. and Others
v. Federation Nationale des Cinemas Francais, Cases 60–61/84, 1985 E.C.R. 2627, para.
26, where the court ruled: ‘Although it is true that it is the duty of this Court to ensure observance
of fundamental rights in the field of Community law, it has no power to examine the compatibility
with the European Convention of national legislation which concerns, as in this case, an area
which falls within the jurisdiction of the national legislator.’; See also: Court of Justice of EU,
Demirel v. Stadt Schwdbisch Gmtnd, Case 12/86, [1987] E.C.R. 3719; Court of Justice of EU,
Kremzow, Case C-299/95, [1997] ECR I-2629, para 15.
57
See e.g.: Raba (2013), p. 568; Cf.: Jacque (2011), p. 1005, arguing that the inclusion of CFSP
actions under the accession will generate asymmetry between Luxembourg and Strasbourg courts,
as the former has no treaty competence to review such measures whereas the latter is being given
the chance to serve as a court of first instance on CFSP materie.
58
ECtHR (GC), Bankovic and others v. Belgium and 16 Other States (Appl. No. 52207/99),
decision (inadmissibility) of 12 December 2001, para. 67 (‘the Court has accepted only in
exceptional cases that acts of the Contracting States performed, or producing effects, outside
their territories can constitute an exercise of jurisdiction by them within the meaning of Article 1 of
the Convention.’); On Bankovic, see also: Gondek (2005), p. 356 et seq; Accord: Inter-American
Commission of Human Rights, Coard v. United States, Report No 109/9 of 29 September 1999
(DC 215-6), para. 37 (‘While the extraterritorial application of the American Declaration has not
been placed at issue by the parties, the Commission finds it pertinent to note that, under certain
circumstances, the exercise of its jurisdiction over acts with an extraterritorial locus will not only
be consistent with but required by the norms which pertain.’).
59
Gragl proposes that in order to overcome this potential interference to EU law autonomy, the
Strasbourg Court should call them inadmissible and leave it to the Luxembourg Court to decide on
those issues. See: Gragl (2013), p. 165. One should fully disagree here with Gragl, as it makes not
sense to exclude EU’s accession to ECHR for the bulk of positive obligations the definition of
which may need to put the Strasbourg Court to find out the competent and liable entity. It makes no
sense to have the EU responsible under the Convention system only for actions and not for
omissions. Their value is absolutely equal in terms of the persons enjoyment of rights under the
Convention.
6.3 EU Treaty-Based Provisions: Which Were the Initial ‘Peculiar’. . . 177

being a case of interference into EU law autonomy. On the other hand, Member
States’ implementing measures include situations wherein EU Member States are
obliged to respect EU human rights law, which, while ‘acting as part of the
decentralized EU administration, provide the necessary “hands and feet” for EU
law to have effects within national territories.’60 Member States’ EU human rights
obligations in this context are not obligatory merely when they implement EU
law,61 but also when they derogate from it or when falling outside the justified remit
of it.62 This said, the EU is protected against becoming liable before the Strasbourg
regime of human rights for an issue which goes beyond its competence stemming
from the Treaties and secondary law. This boundary is rather well demarcated,
however the question remains whether there is any portion of liability for the EU
Treaties if the EU Court, as an EU institution, takes part in interpreting it and
making Treaty-level case law. Certainly, as one will observe supra, practically, the
EU will become still responsible even for primary law interpretations and case-law
of its own Court, which still makes Art. 1 (3) DAA a bit surpassed by the overall
organization and functioning of EU law and Strasbourg law interaction under the
future practice.63
This model of division of labor in the face of the Convention is rather complex
and may produce several legal ramifications in practice when it comes to identify-
ing the correct addressee, a procedure which may not only generate a potential clash
between the EU and Member States, but also them and the Convention system.64
One should also remind that should the applicant not name the ‘correct addressee’
the application may even risk to be called inadmissible.65 The legitimate question
will remain how can one ‘guarantee the ECHR’s uniform application across the EU
and the rest of Europe in light of the ECJ’s intense interest in maintaining its
supremacy, as well as an autonomous Community legal order?’66 Altogether, the

60
Hancox (2013), p. 1418; See a similar concept at: Lock (2012), p. 193; Von Bogdandy
et al. (2012), p. 497.
61
E.g.: Muir (2014), p. 228; See also two cases of this kind: Court of Justice of EU, Wachauf
v. Forstwirtschaft, Case 5/88, [1989] E.C.R. 2609; Court of Justice of EU, Tihelorassi
v. Pliroforissis, Case C-260/89, [1991] E.C.R. 1-2925; To note, implementing measures of
Member States are interpreted rather broadly by the EU Court, in order to ensure that fundamental
rights are observed as broadly as possible from the perspective of MS actions. See an exemplary
case to taste the latter argument, at: Court of Justice of EU, K€uc€ukdeveci, Case C-555/07, [2010]
ECR I-00365, paras. 23–26.
62
See e.g.: Garcı́a (2002), p. 495; Gragl (2013), p. 55; On the prohibited derogation from EU law,
see also: Franklin (2010–2011), p. 153.
63
See infra the chapter on exhaustion of EU law direct remedies.
64
Gragl rightly argues that these blurred borders between the EU and Member States’ responsi-
bility may become even more problematic if Member States were recognized a certain margin of
discretion when they implemented the alleged violation. See: Gragl (2013), p. 143.
65
Gragl (2013), p. 142.
66
Wetzel (2003), p. 2825.
178 6 Attribution of Liability Under the Co-respondent Mechanism

Treaty-based instructions on the co-respondent mechanism67 need to address these


loopholes and leave little space for such clashes and challenges to EU law auton-
omy. The co-respondent mechanism, however, might prove financially burdensome
for applicants who have no legal education, as its use requires the support of an
excellent lawyer.68
As a final point of analysis and discussion, Jean d’Aspremont argues that the
DAA’s rules on the attribution of responsibility will become internal rules within
the Convention system and international law at large. Compared to the previous
position—in which case the rules on external responsibility of the EU and its
Member States were considered possibly lex specialis in international law—the
DAA’s rules do make those rules merely internal, therefore no longer capable of
providing lex specialis to the law of international responsibility. In this regard,
d’Aspremont argues that the DAA’s rules on attribution of responsibility will no
longer be of high relevance to the law of international responsibility post-accession
as before.69 The author of this book partially agrees with d’Aspremont with regard
to the nature of these rules having an internal effect with regard to the EU-ECHR
systemic relationship in the attribution of responsibility. However, the situation was
not so different before accession for two basic reasons. First, even before accession,
the rules on attribution of responsibility of the Strasbourg Court on the EU were not
applicable to all other international organizations. The difference recognized by the
Strasbourg Court between Bosphorus and Michaud v. France provided validity to
this argument. Second, even though the EU will be a contracting party to the ECHR,
the rules on attribution of responsibility applied in this context may well serve the
needs of a lex specialis order under international law—with two legal regimes being
rather well founded on the grounds of law of international organizations. This said,
the practice and mechanisms established by the DAA with regard to the attribution
of responsibility to the EU and its Member States may well be used by the
Strasbourg Court on other international organizations as well from the perspective
of judicial innovations which Strasbourg may push forward. That said, it should not
be taken for granted that d’Aspremont distinction between the model of attribution
of responsibility before accession and under DAA will be so significant as to make

67
Gragl in this regard argues that should the Strasbourg Court be given jurisdiction to rule on the
apportioning of responsibility between the EU and its Member States that would violate Art. 1 of
Protocol 8 which requests that the accession process preserve the characteristics of EU law. See:
Gragl (2013), p. 144; Cf.: Gragl in this regard also mentions the suggestion of the EP’s Committee
on Constitutional Affairs which had basically proposed that when there are doubts as to whether a
certain violation is caused by the EU or the Member State(s), then the applicant direct the
application concurrently to both of them. Gragl criticizes this model, with a valid argument, as
that would have complicated even more the procedure before the Strasbourg Court, especially if
one takes into account the fact that the rule on exhaustion should have been applied in both levels
in that scenario. Gragl (2013), p. 148.
68
Lock (2010), p. 783; See also: Gragl (2013), p. 141–143; Accord: Raba (2013), p. 564, who
argues that it is rather difficult for the applicant to define which level of legal order (EU v. Member
States) caused the violation; On the latter, see also: Craig (2013), p. 1122.
69
d’Aspremont (2014), p. 84.
6.4 The Differences Between Third-Party Interventions and the Co-respondent. . . 179

it impossible for the DAA to serve as lex specialis in regard to the law on
international responsibility of international organizations.

6.4 The Differences Between Third-Party Interventions


and the Co-respondent Mechanism: Where Does
the DAA Go Blurry?

Initially, Rober Badinter, the French well-known jurist, had proposed that third
party intervention would be a safer method for the Union compared to a method
such as the co-respondent mechanism. Badinter argued that—should his argument
not become applicable in practice—the Union would become the most convicted
European entity for human rights breaches in Strasbourg as it is a Union of more
than 20 states,70 clearly supporting a model wherein the Union would not become a
party in a proceeding before Strasbourg Court where the original challenged
respondent is a Member State which implemented an EU law duty.71 Those who
drafted the DAA did not of course agree with Badinter, and as a result crafted the
co-respondent mechanism. It is important to note that there is a clear difference
both in nature and function between the third-party interventions provided for by
Art 36 ECHR72 and the co-respondent mechanism provided for by Art 3 of the
DAA. Similar to the third party intervention before the Strasbourg Court is the
procedure of third party intervention to the International Court of Justice, as
established by Art. 62 and Art. 63 of the Statute of International Court of Justice73;
using the latter as an additional example on basis of which one can draw clearly its
difference with the co-respondent mechanism. There are three key differences
between the third party intervention and the co-respondent mechanism, which
need appropriate attention.

70
Lock (2010), p. 785.
71
On Germany’s proposal for merely a redesigned third party mechanism, see: Groussot
et al. (2011), p. 7.
72
The procedure and nature of third party intervener is explained officially in the following
manner: ‘The President of the Court may authorize any person other than the applicant, or another
State Party to the Convention other than that against which the application has been lodged, to
intervene in the proceedings. This is called third-party intervention. The person or State in question
is entitled to file pleadings and take part in public hearings.’, at: The European Court of Human
Rights (2014), p. 8; See also: Leach (2011), p. 16 et seq; Art. 36 (1 & 2) ECHR, regarding third
party intervention procedure, reads: ‘1. In all cases before a Chamber or the Grand Chamber, a
High Contracting Party one of whose nationals is an applicant shall have the right to submit written
comments and to take part in hearings. 2. The President of the Court may, in the interest of the
proper administration of justice, invite any High Contracting Party which is not a party to the
proceedings or any person concerned who is not the applicant to submit written comments or take
part in hearings.’
73
See e.g.: Chinkin (1986), pp. 495 et seq; See also: Statute of International Court of Justice,
available at: http://www.icj-cij.org/documents/index.php?p1¼4&p2¼2&p3¼0&#CHAPTER_III.
180 6 Attribution of Liability Under the Co-respondent Mechanism

First, the third party intervention procedure gives the possibility to any High
Contracting Party to submit written comments and to participate in the hearing
sessions,74 and there would be no ‘obligation’ to intervene.75 The third party,
therefore, does not become a party to the case,76 consequently the decision of the
Court does not oblige it. Third party intervention is used by a High Contracting
Party merely as an option to offer its view solely based on its concern and/or desire.
By contrast, in a situation wherein a High Contracting Party becomes a
co-respondent, the co-respondent takes the legal status as a party to the case,
enjoying the legal guarantees to argue and submit legal reasons as if it were the
original party, adding that the decisions of the Court also oblige the co-respondent
as they do with the original respondent.77 The co-respondent ‘enjoys full procedural
rights.’78 Being a co-respondent, therefore, entitles the latter to the rights and duties
which the judicial course of action produces for the original respondent, clearly
making the co-respondent parallelly situated to the procedural standing of the
original respondent both in the technical steps and its legal outcomes.
Second, the right to intervene as a third party in accordance with Art 36 ECHR
belongs merely to a High Contracting Party whose national is an applicant, meaning
that only states with a direct interest, tested through the citizenship of the appli-
cant,79 could join the proceedings before the Court. By contrast, EU Member States
could take the status of co-respondent in accordance with Art 3 (3) of the DAA
independent from the fact whether the case being judged is filed by a citizen holding
the co-respondent’s nationality. The ground for joining the proceedings is therefore
very different, and is meant to permeate the co-respondent to all substantive objects
for which the procedure is being undertaken.
In addition, the third-party intervention is invented as a tool only to allow a High
Contracting Party to submit arguments in relation to a case involving its citizen as
an applicant before the Court. The co-respondent mechanism, by contrast, is
foreseen as a mechanism to call into responsibility the correct entity for violating
human rights of the applicant. The interest and duty of the co-respondent here is not
merely the defense of certain arguments but rather the undertaking of liability to
remedy the violation. This said, the co-respondent mechanism simply serves the
purpose of identifying and entitling the legal responsibility to the right addressee,80

74
See e.g. the third party intervention in case: ECtHR, KAOS-GL v Turkey Application No. 4982/
07; See also generally: Leach (2011), pp. 16 et seq.
75
Lock (2010), p. 785, as one finds it in the case of co-respondent.
76
E.g.: Groussot et al. (2011), p. 10; See also: Lock (2012), p. 167; See also: O’Meara
(2011), p. 1821.
77
CDDH-UE (2011)16fin, paragraph 39; See also: Raba (2013), p. 565; Martı́n and De Nanclares
(2013), p. 10/11; See also: Gragl (2013), p. 151.
78
Raba (2013), p. 565; Odermatt (2014), p. 24.
79
E.g.: Lock (2010), p. 785.
80
See e.g.: Gragl (2011), p. 71.
6.5 The Core Function of the Co-respondent Mechanism: What Is It Made for? 181

for which to happen, the right addressee should be acting either as a respondent or
as a co-respondent.
Fourth, the third-party intervention could also be utilized by persons not directly
related and with no direct interest in the case being adjudicated before the Court,
upon the invitation of the President of the Court. Such access as a third-party
intervener is also granted to the Council of Europe Commissioner for Human
Rights,81 as an example. This could not be the case with the co-respondent
mechanism, as the status of co-respondent is strictly linked to the EU and its
Member States, in line with the permissibility criteria listed in Art 3 of the DAA.
In addition to the above, one should mention here that a logical interpretation of
Art 3 of the DAA does not aim to maintain that if a party has joined once as a third-
party intervener it could not ask to change to the status of a co-respondent. This
said, having been a third-party intervener does not prohibit (and does not exhaust
the right of) the same party from requesting to become a co-respondent in the same
case, conditional to meeting the admissibility criteria from Art 3 of the DAA.
It is important to note that due to the fact that the co-respondent mechanism is
made to serve and could merely be utilized in EU-related issues, this does not
prevent the EU from joining cases related to non-Member States before the Court as
a third-party intervener, or even joining as third party intervener to cases wherein it
does not fulfill the conditions to become a co-respondent. The same is true for
non-Member States regarding EU cases before the Court. This being said, if a case
appears before the Court wherein a non-Member State submits an issue partially
linked to the EU, such as issues pertaining to e.g. the Schengen Agreement,
European Economic Area or the Dublin II Regulation, the EU would have the
opportunity on basis of Art 36 (1) ECHR to join as a third-party intervener.82 Third-
party intervention procedure, nevertheless, allows room for the EU to join as a third
party intervener in cases originating from individual applications from non-EU
Member States, and this becomes especially important for cases which touch upon
questions interrelated to EU law.

6.5 The Core Function of the Co-respondent Mechanism:


What Is It Made for?

It is important to note that the co-respondent mechanism is made to act in response


to the requirement of Protocol 8 to ensure that the addressee(s) is/are correctly
determined, and that the special legal nature of the EU law be protected against

81
Art 36, para. 2 and 3 ECHR.
82
CDDH-UE (2011)16fin, para. 40.
182 6 Attribution of Liability Under the Co-respondent Mechanism

external interference, if the latter is a probable scenario.83 To meet this aim, Art
3 (1) b of the DAA amends the Convention’s Art 36, wherein a paragraph on the
co-respondent mechanism will be added, reading as follows:
The European Union or a member State of the European Union may become a
co-respondent to proceedings by decision of the Court in the circumstances set out in the
Agreement on the Accession of the European Union to the Convention for the Protection of
Human Rights and Fundamental Freedoms. A co-respondent is a party to the case. The
admissibility of an application shall be assessed without regard to the participation of a
correspondent in the proceedings.84

According to the above provision, the co-respondent is a function that could be


utilized by either the EU or a Member State to proceedings before the Strasbourg
Court. Of course, there should be a respondent, against which the applicant has
submitted the application, in order for a co-respondent to become eligible to join the
proceedings.85 The aim of co-respondent, therefore, is to join the respondent in two
respects: first, by submitting its own arguments and reasons for a certain issue being
proceeded in that case therefore fully defending its legal position before the Court
and the claimant(s), and second, by taking over the burden for the violation which
the application was filed for, and sharing it either with the respondent or holding it
alone. This will also be an argument that will obviously ‘reduce the scope for
Member States to argue that their hands are tied due to their obligations under EU
law’.86 On the other hand, the Luxembourg Court has already dealt with the issue of
international liability in the context of it and its Member States’ relationship, in
Parliament v Council, wherein it ruled that: ‘The Convention was concluded [. . .]
by the Community and its Member States of the one part and the ACP States of the
other part. [. . .] In those circumstances, in the absence of derogations expressly laid
down in the Convention, the Community and its Member States as partners of the
ACP States are jointly liable to those latter States for the fulfilment of every
obligation arising from the commitments undertaken [. . .].’87 Joint liability thus

83
See: Lock (2010), p. 782/3; See also: Eckes (2013), p. 267; Gragl notes that this mechanism was
first introduced in 2002, in which case it was argued that the EU should be given the chance to
defend itself before the Strasbourg Court if the alleged violation derives from EU law. See: Gragl
(2013), p. 148/9.
84
CDDH-UE (2011)16fin, Art 3 (1)b; See also: Gaja (2013), p. 3 (‘The main purpose of the EU
becoming a co-respondent is to allow the EU to enjoy all the rights of a party to the proceedings in
order to defend what it considers to be the proper interpretation of the relevant provisions of EU
law and of the ECHR.’).
85
Gragl in this regard notes that—referring to Lock—in the past there have been three options
presented in the table as regards the manner in which the co-respondent mechanism may be
triggered: first, by the request of the original respondent, second, by the request and decision of the
Strasbourg Court, and, third, by the request of the co-respondent itself. See: Gragl (2013), p. 153.
The current DAA foresees two possibilities: first, that the co-respondent request to join and the
Court decide on it, and, second, that the Court invite the co-respondent to join.
86
O’Meara (2011), p. 1820; See also: Gragl (2013), p. 149.
87
Court of Justice of EU, Parliament v Council, Case C-316/91 [1994] ECR I-625, para.
29 [emphasis added]; See also on joint responsibility: Eeckhout (2004), p. 222/3.
6.5 The Core Function of the Co-respondent Mechanism: What Is It Made for? 183

is not new for the EU Court (see e.g. AG Tesauro’s argument in Hermès),88 and it is
something that it follows with full desirability.89 To digress, before embarking on
this issue more in depth, it seems odd to read the recent argument of the Luxem-
bourg Court in Opinion 2/13 (para. 226 et seq) which basically shows allergy even
on the model of joint liability—which, by the way, fits best with the requirement to
safeguard the EU law autonomy. Such argument of the Court in Opinion 2/13
contravenes its own earlier case-law demonstrated above.
One must point here that the function of the co-respondent is not merely to
access a procedure with the final intention of sharing the burden of its decision with
the respondent. Although this might be the final effect or outcome of the
co-respondent apparatus, the genuine aim of the latter is to protect its position
before the Court in a case which encapsulates segments and/or constitutional
competences of its own. The co-respondent, a fortiori, would not be in principle
willing to be ‘judged’ for a human right violation. However, with its presence in the
proceeding, the co-respondent becomes the competent authority to speak for such
violation, wherein, practically, the Strasbourg Court would not be able to review its
internal competences and/or touch upon its internal division of authority. By
protecting its own law under the co-respondent mechanism, the co-respondent
ensures its autonomy of law by not allowing someone else to ‘articulate’ on behalf
of its legal order.90 It is important to argue that the co-respondent mechanism under
DAA is built on the presumption that the relationship between EU law and national
legal orders is one which has no tensions—similar to the relationship between the
federal law and federal entities’ law—which is not that true in practice. Such
tensions may in fact impair the entire concept in which the co-respondent mecha-
nism is rooted, as loyal cooperation and full submission of national legal orders to
the EU Treaties might not always be appropriate. Otherwise, if this presumption

88
As a comparative example, AG Tesauro affirmed in Hermès that ‘[i]n these circumstances, it
should be recognised that the Member States and the Community constitute, vis-a-vis contracting
nonmember States, a single contracting party or at least contracting parties bearing equal
responsibility in the event of failure to implement the agreement’. [emphasis added]. (See:
Court of Justice of EU, Hermès International v FHT Marketing Choice BV, Case C-53/96
[1998] ECR I-3603, para 14). It seems clear that basically AG Tesauro proposed a two-level
incremental model of responsibility in Hermès, the first level being joint responsibility, which, if
not accepted by the parties, then equal responsibility of the parties (Community and Member
States) will be the principle applying in the second level—something which may not be departed
any further. This option seems to have been totally excluded in the DAA.
89
Contra.: Opinion of the Advocate-General Mischo (Berne Convention for the Protection of
Literary and Artistic Works), Commission v Ireland, Case C-13/00 [2001] ECR I-2943, para.
30 (‘the very fact that the Community and its Member States had recourse to the formula of a
mixed agreement announces to non-member countries that that agreement does not fall wholly
within the competence of the Community and that, consequently, the Community is, a priori, only
assuming responsibility for those parts falling within its competence.’) [emphasis added]. It seems
clear that AG Mischo promotes a shared responsibility model for the EU and its Member States,
certainly such model would offer space for interference to EU law autonomy from
Strasbourg’s side.
90
On the latter, see a general opinion at: Lock (2010), p. 784.
184 6 Attribution of Liability Under the Co-respondent Mechanism

would not be valid, then the co-respondent mechanism might seriously threaten the
effectiveness of the Convention system to identify the violator and attribute proper
responsibility to it.
It is very clear from the above provision that the co-respondent is a party to the
case being adjudicated by the Court. As a party, it shares the burden to both protect
its position and to assume the responsibility. This speaks for the fact that the
co-respondent becomes a full party to the case, and lacks nothing compared to
the respondent but the ‘co-’ before its name. It must purposely be acknowledged
that the co-respondent does not serve to help or to support the respondent. It merely
accesses the procedures to protect its own law, its authority, and to share the burden
of the Convention responsibility. This said, the positions of the respondent and the
co-respondent are not said to follow the same direction and to represent the same
interest (although they are in pari delicto): they are merely the target of the
application concerning a violation of the Convention’s right. Beyond this point,
there is nothing else that would naturally join the respondent and the co-respondent.
Gragl tries to draw a difference between the ‘co-respondents’ and ‘multi-respon-
dents’ scenarios: the former, according to Gragl, applies when two respondents
have a joint interest, whereas the latter applied when they have been given auton-
omy to address their conduct in the case.91 One may not take Gragl’s account for
being safely argued: co-respondents may not necessarily need to follow the same
interest and direction. They are part of the same procedure simply due to the fact
that one of them legislated a certain violation whereas the other implemented it. The
same background may capture ‘multi-respondents’ in such scenario. Therefore, the
only difference between the two seems to be the fact that co-respondents are not the
original addressees, whereas all multi-respondents are always the original
addressees of a certain case.
The above-mentioned provision makes it clear that admissibility of the case is
assessed without making reference to the co-respondent in the proceeding, adding
that ‘the draft agreement does not address the question of the latest possibility for
the co-respondent to join proceedings.’92 This said, the application’s admissibility
is considered with regard to the criteria under Art 35, and the presence or not of the
co-respondent in the admissibility procedure is absolutely irrelevant and not linked
with the admissibility of the case. It should be argued here that the co-respondent
does not have any direct link with the application’s intention: the co-respondent
mechanism is merely a space for others to join the court proceedings against them.
The applicant, therefore, has absolutely no burden to argue/reason whether and why
a co-respondent should be held responsible (actori incumbit probation does not
apply to the applicant with regard to the co-respondent). The application’s admis-
sibility is solely linked with the respondent, and the reasoning that the latter should
be held responsible for the violation of the human right concerned.

91
Gragl (2013), p. 150.
92
Lock (2012), p. 168.
6.5 The Core Function of the Co-respondent Mechanism: What Is It Made for? 185

It is important to question the period within which a co-respondent would be able


to join the proceedings. The Explanatory Report makes it clear that the
co-respondent mechanism will not change the current practice in which the Stras-
bourg Court undertakes a preliminary assessment of a case filed before it to check
its admissibility. Should the Court find in a preliminary check that the case is
inadmissible or is manifestly ill-founded, it will not communicate this to a High
Contracting Party, meaning the respondent. In this regard, the applications are
communicated to the respondent(s) only if they are found admissible and not
manifestly ill-founded by the Court. The Explanatory Report makes it clear that
the co-respondent mechanism applies from the moment the application is commu-
nicated to the respondent and beyond.93 This said, it is argued that a party could not
join as a co-respondent to the proceedings if the case is still being judged for its
admissibility under Art 35 ECHR.
One should dig a bit deeper in the impact of the situation wherein a
co-respondent is able to join the proceedings only once the admissibility test has
passed successfully. Given the fact that the co-respondent mechanism has been
devised to avoid the Court to engage in the review of EU internal competences
deriving from the Treaties,94 it appears that in numerous cases this issue could arise
even at the admissibility phase. The problem starts with Art. 34 ECHR, which, in
order for a case to be admissible, requests that claimants ‘be the victim of a violation
by one of the High Contracting Parties of the rights set forth in the Convention or
the Protocols thereto.’95 At the admissibility phase, therefore, the Strasbourg Court
would need to assess whether there is a connection between the interference to the
claimant and the act or omission of the challenged respondent. In finding this link,
the Court—in an EU–Member State related violation—would need to assess
whether according to the internal law of the respondent, the latter could have
factually caused that violation. If the Court finds that there was no opportunity
for the respondent to cause the violation—based on an assessment of the internal
law of the respondent—the application is dismissed as inadmissible, therefore there
would be no way for the co-respondent (the real respondent) to join the proceed-
ings. For such admissibility to be assessed by the Strasbourg Court, review of EU
law may sometimes be the only way to find whether the violation was in fact caused
by the attacked respondent (High Contracting Party). Of course, in a contrary
scenario, wherein the co-respondent would be allowed to join from the moment
the case was filed—therefore giving the chance to co-respondent to enter the
admissibility phase—such internal review would not have been undertaken. Such
an example may be Kokkelvisserij U.A. v Netherlands, wherein the claimant had
exhausted Netherlands-based legal remedies, but not those of the EU. The Court
considered this application inadmissible. In Kokkelvisserij U.A. v Netherlands the
claimant challenged the EU’s treaty provisions on the impossibility to reply to

93
CDDH-UE (2011)16fin, para. 45 et seq.
94
Generally established in Art. 4 and 5 TEU and Art. 2 to 4 TFEU.
95
Emphasis added.
186 6 Attribution of Liability Under the Co-respondent Mechanism

Advocate General opinion within the context of the preliminary reference proce-
dure before the Luxembourg Court. In such a scenario, after accession, the Stras-
bourg Court would find that in Kokkelvisserij U.A. v Netherlands there is no linkage
between the actions or omissions of Netherlands with reference to the alleged
violation stemming from the Treaties, wherein the Netherlands did not take an
action nor originated any legal basis for that alleged violation. In that context, the
Strasbourg Court would find the application of Kokkelvisserij U.A. as inadmissible
on the basis of Art. 34 ECHR. Such inadmissibility would allow the Strasbourg
Court to enter into the space of EU Treaties to identify—as explained infra96—
whether Netherlands did in fact have any power to amend the EU Treaties or if it
has any competence to offer to the claimants before Luxembourg Court standing to
reply to the Advocate General’s opinion. That said, it is argued here that as long as
the procedure on co-respondent admissibility starts from the moment the claimant’s
admissibility is passed, this allows for space wherein the Strasbourg Court would
inefficiently dismiss an application as inadmissible by entering into the review of
EU internal law to identify whether the conditions of Art. 34 ECHR on admissi-
bility are met. Therefore, it is suggested that for better ensuring the efficiency of
adjudication and better preserving the autonomy of EU law, the co-respondent
mechanism should have been allowed to be utilized before or at least during the
early stage of the admissibility check which the Strasbourg Court performs on an
EU-related case.97 If that scenario would be practical, then applications would not
have been dismissed on the basis of the missing linkage between the act/omission
which caused the alleged violation, whereas the EU law autonomy would have been
fare better protected. Interestingly enough—unfortunately as well—the Luxem-
bourg Court did was incapable to identify this big loophole in the DAA, and did not
pronounce on it in its recent Opinion 2/13.
To make a short digression on the rationale for the design of the abovementioned
mechanism and the implication discussed above, ICJ’s Judge Gaja argues:
Thus, when the EU is only a co-respondent, an application could be declared admissible
even if the local remedies have not been exhausted within the EU legal system. This looks
odd, but may find a justification in the fact that the addition of the EU as co-respondent
depends on the EU’s consent, which may be taken as a waiver of any objection to the
admissibility of the application concerning the EU which might otherwise have been
raised.98

96
See also the part of this chapter where three cases are tested with the co-respondent mechanism
(Chap. 10).
97
The Strasbourg Court—when it comes to cases involving both the EU and its Member States—
could easily follow the approach of the Inter-American Court of Human Rights, which in the
Velasquez Rodrıguez v. Honduras ‘maintain[ed] that the issue of exhaustion of domestic remedies
must be decided jointly with the merits of this case, rather than in the preliminary phase.’ (Inter-
American Court of Human Rights, Velasquez Rodriguez Case, Preliminary Objections, Judgment
of June 26, 1989, Inter-Am.Ct.H.R. (Ser. C) No. 1 (1994), para. 83). If this approach would have
been followed, there would seem far less spaces for interfering to the EU law autonomy.
98
Gaja (2013), p. 3 [emphasis added].
6.6 A Comparison Note Between DAA’s Co-respondent Mechanism and DARIO 187

Inherently, Judge Gaja tries to argue that this rule has been purposely devised in
this form to disregard/ignore EU’s potential disagreement on cases that have been
called admissible without passing through the EU court system. The author agrees
with Gaja that—if not intentionally—this device appears to be a perfect plan to keep
the EU outside the admissibility stage, thereby isolating its potential concerns
regarding why such cases become admissible without the EU courts having had the
chance to rule on the issue. Pragmatically, one may also put forth the argument that
the EU is still not totally barred from the opportunity to rule on that issue, as it has the
prior involvement mechanism that it may utilize. The latter, still, does not have the
power of the objections on admissibility which the EU could have raised at the
admissibility stage, as the latter could have resulted in pressure for the Strasbourg
Court to stop cases originating in EU law at the admissibility stage if they have not
been exhausted before the Luxembourg Court. So, Gaja’s argument stands rather
strongly in this pool of facts, and this author agrees with its central implication;
although this author insists that this model still seems to be an opportunity to interfere
to EU law autonomy. Hence, it is not the best option for EU law autonomy, although
it may perhaps be a better option for the prospect of human rights protection.
That being understood, as a result, the co-respondent mechanism could be utilized
by a party from the moment the application is notified by the Court to the respondent.
Arguably, the question would also be whether the respondent and co-respondent must
act in full agreement during all stages of the procedure. Certainly, the respondent and
co-respondent are not meant to have an agreed position before the Court, except when it
comes to the Court’s decision stage—after admissibility is passed—in which case the
respondent and co-respondent(s) should agree through a friendly settlement on the
division of burden (examined infra).

6.6 A Comparison Note Between DAA’s Co-respondent


Mechanism and DARIO

It is also very important to contrast the design and function of the co-respondent
mechanism with some of the core provisions of the Draft articles on the responsi-
bility of international organizations (DARIO).99 DARIO has been cited by the
Strasbourg Court as relevant law and practice of the field.100 Although DARIO is

99
Draft articles on the responsibility of international organizations (DARIO). ‘Adopted by the
International Law Commission at its sixty-third session, in 2011, and submitted to the General
Assembly as a part of the Commission’s report covering the work of that session (A/66/10, para.
87).’ It is argued that DARIO, if not legally binding, may still be considered as customary
international law for purposes of Strasbourg Court’s interpretations as well. See a similar argu-
ment, at: Larsen (2008), p. 517/8.
100
Larsen (2008), p. 512/3; See also the ICJ referring to the articles on state responsibility, at:
International Court of Justice, Gabcı́kovo-Nagymaros Project (Hungary/Slovakia) case, Judgment,
ICJ Reports, 1997, p. 7.
188 6 Attribution of Liability Under the Co-respondent Mechanism

still not in force101—and its provisions may be surpassed if a lex specialis102 like
DAA regulate it differently—its principles play a core function103 in delineating the
borders of international responsibility of international organizations like the EU. To
note, Art. 1(2&2) DARIO establishes that it applies both ‘to the international
responsibility of an international organization for an internationally wrongful act’
and ‘to the international responsibility of a State for an internationally wrongful act
in connection with the conduct of an international organization.’ Three provisions
of DARIO would provide a wealth of knowledge if contrasted to the co-respondent
mechanism established by DAA. Those three provisions are Art. 7 DARIO, Art.
15 DARIO and Art. 59 DARIO, all of them regulating the relationship between an
international organization and entities like its Member States in relation to their
international responsibility for wrongful acts. Art. 7 DARIO therefore reads:
The conduct of an organ of a State or an organ or agent of an international organization that
is placed at the disposal of another international organization shall be considered under
international law an act of the latter organization if the organization exercises effective
control over that conduct.104

This would basically mean, according to ICJ’s Judge Gaja, in a DARIO applied
upon the EU scenario, ‘[. . .] State organs act as quasi organs of the EU when they
implement an obligation under EU law which leaves them no discretion. Respon-
sibility for their conduct would have to be attributed only to the EU [. . .].’105 In
addition, Art. 15 DARIO reads:
An international organization which directs and controls a State or another international
organization in the commission of an internationally wrongful act by the State or the latter
organization is internationally responsible for that act [. . .]

101
See: Odermatt (2014), p. 5, who argues that during the drafting of DARIO, International Law
Commission had refused to contrue specific rules on the EU—the latter being quite a special case
even in terms of DARIO regulative landscape.
102
Art. 64 DARIO.
103
See the Strasbourg Court’s view on the need to engage with international law when it comes to
the attribution of liability, while pointing at the Convention’s special status as a human rights
treaty, at: ECtHR, Bankovic v Belgium and others, Application no. 52207/99, Grand Chamber,
Decision as to the Admissibility, para 57. (‘The Court must take into account any relevant rules of
international law when examining questions concerning its jurisdiction and, consequently, deter-
mine State responsibility in conformity with the governing principles of international law,
although it must remain mindful of the Convention’s special character as a human rights treaty.’
Emphasis added). The international law referred to in the above provision may certainly cover
DARIO as well.
104
See e.g., the ICJ’s view on the effective control test, at: International Court of Justice,
Nicaragua v United States of America, Merits, Judgment, ICJ Reports, 1986, pp. 14–65 (para
115 et seq).
105
Gaja (2013), p. 2.
6.6 A Comparison Note Between DAA’s Co-respondent Mechanism and DARIO 189

Within the context of Art. 7 DARIO, in conjunction with Art. 15 DARIO, EU


Member States should be considered as being at the disposal of the EU to imple-
ment EU law—being both directed106 and controlled by EU law—in a situation
when Member States implement EU law without exercising state discretion.107
Two arguments support this claim: first, because Member States implement EU
law—in a situation when they exercise no state discretion—merely as its agents and
on its behalf, secondly, because in a situation when Member States have no state
discretion in implementing EU law, they are effectively directed and controlled by
the EU,108 in the sense that should the EU law provide for a different stipulation,
Member States have no other choice but implementing it. It is also assumed that
Art. 7 & 15 DARIO apply not only to situations when Member States have no
discretion—as they are under the disposal of the EU for their conducts
implementing the EU law—but also when they have no authority to change that
law, which directs us to EU secondary law (EU may not have the authority to
change or hold authorship for the primary law, only the secondary law is under its
origin of authority). Art. 7 & 15 DARIO compared to the co-respondent mechanism
design a different model of attribution of responsibility for the EU and its Member
States when the latter implements EU secondary law without enjoying state discre-
tion. DARIO seems to support a model of subjective responsibility, like the ICJ’s
Corfu Channel case,109 as opposed to the model of objective responsibility which
DAA seems to establish. The co-respondent mechanism in the latter scenario would
certainly make both the EU and its Member State which conducted the EU
secondary law implementation as responsible for the violation of the Conven-
tion,110 whereas Art. 7 & 15 DARIO would not attribute any responsibility to the

106
Meaning the ‘effective control’ of EU over Member States’ conduct. See a similar notion
explained in cases of share of burden for military actions under UN umbrella, at: Larsen
(2008), p. 514.
107
Two exemplary cases may be of this nature: Court of Justice of EU, Wachauf, Case C-5/88,
[1989] ECR 2609, para 19, and, Court of Justice of EU, Bostock, Case C-2/92, [1994] ECR I-955,
para 16.
108
Larsen (2008), p. 515 (‘The organization must exercise effective control over the conduct of an
organ of a state that is placed at the organization’s disposal in order for the conduct to be
attributable to the organization.’); In Loizidou, the Strasbourg Court used ‘effective overall
control’ to exhibit this notion. Larsen (2008), p. 522; Cf. to Al-Jedda case, wherein the Strasbourg
Court applied the effective control test in a different way, by calling UK responsible for its actions
instead of UN.
109
International Court of Justice, United Kingdom v Albania (Corfu Channel), ICJ Reports,
1949, p. 15.
110
A proposal for following DARIO in the issue of division of labour between EU and Member
states when it comes to joint responsibility has been given since Bosphorus. See: Hoffmeister
(2006), p. 447; Cf.: Gaja (2013), p. 3, who argues that the DAA builds upon the idea that Member
States—when implementing an EU law obligation—are kept responsible for the mere fact of
implementing the EU law.
190 6 Attribution of Liability Under the Co-respondent Mechanism

Member State for conducts made on behalf and under the effective control of the
EU.111 DARIO therefore seems to be more appropriate in dividing/isolating the
responsibility for a wrongful act, compared to the co-respondent mechanism which
applies a joint/mutual responsibility although the Member State conduct has been
ordered and originates from the international organization (in this case the EU).
E.g. In a scenario like Bosphorus, Ireland—also in conjunction with Art.
43 DARIO—would be entitled to invoke the international responsibility of the
EU. The same would apply to Kadi case. Of course, DARIO’s model of setting forth
the responsibility may have not been accepted by DAA for practical reasons, as it
would have allowed the Strasbourg Court to enter into EU law to inquire about the
violating party, which would amount to an interference in EU law autonomy.
A different scenario would be if EU primary law were at stake, a situation in
which the real authors are the Member States and not the EU. Art. 59 (1) DARIO
therefore reads: ‘A State which directs and controls an international organization in
the commission of an internationally wrongful act by the latter is internationally
responsible for that act [. . .]’. Art. 59 (1) DARIO clearly applies to a situation like
Mathews or Kokkelvisserij, wherein the violation stemmed from EU primary law,
which is in fact a violation of EU Member States by adopting the primary law,
rather than the EU itself. Therefore, if there appears a situation in which the
violation/wrongful act stems from EU primary law, DARIO would have called
into responsibility merely the EU Member States and not the EU, as it is the
Member States which have enforced and have the authority to revise that law. It
would be the Member States which would invoke their responsibility for a scenario
like in Kokkelvisserij. Compared to the co-respondent mechanism, DARIO seems
to better allocate the responsibility of the EU Member States on the EU Treaties,
rather than holding both the EU and Member States jointly responsible for such a
wrongful act—something the co-respondent mechanism does. DARIO does not
properly address, however, cases when EU Member States cause violations to the
Convention by exercising state discretion in implementing EU law.112
As a final point, two relevant cases may help show the before-accession practice
of the Strasbourg Court with regard to the attribution of responsibility on violations
involving international organizations and states. First, the Strasbourg Court in
Behrami & Saramati had ruled:

111
A case of attributing responsibility in this regard is the following provision of the Inter-
American Court of Human Rights case: ‘An illegal act which violates human rights and which
is initially not directly imputable to a State [. . .] can lead to international responsibility of the State,
not because of the act itself, but because of the lack of due diligence to prevent the violation or to
respond to it as required by the Convention.’ (Inter-American Court of Human Rights, Vel asquez-
Rodrı́guez v. Honduras, judgment of 29 July 1988, at para. 172) [emphasis added].
112
See e.g. the explanation at: Franklin (2010–2011), p. 153 (‘Member States will clearly be
understood as acting within the scope of application of EU law where they act in order to
implement, enforce or interpret EU secondary legislation such as regulations or directives at
national level. The Court has indeed come to view this simply as alogical extension of its
jurisdiction to assess the legality of EU action, since Member States in such cases are seen to be
acting as agents of the EU.’).
6.6 A Comparison Note Between DAA’s Co-respondent Mechanism and DARIO 191

Accordingly, the Court notes that UNMIK was a subsidiary organ of the UN created under
Chapter VII of the Charter so that the impugned inaction was, in principle, “attributable” to
the UN in the same sense.113

Quite differently—apparently not in a very different attribution scenario—the


Strasbourg Court in Nada v. Switzerland ruled:
Relying on the Court’s decision in Behrami and Saramati (cited above), the intervening
French Government, in particular, argued that the measures taken by the member States of
the United Nations to implement Security Council resolutions under Chapter VII of the
Charter were attributable to the United Nations and were thus incompatible ratione perso-
nae with the Convention. The Court cannot endorse that argument. It would point out that it
found in Behrami and Saramati that the impugned acts and omissions of the Kosovo Force
(KFOR), whose powers had been validly delegated to it by the Security Council under
Chapter VII of the Charter, and those of the United Nations Interim Administration Mission
in Kosovo (UNMIK), a subsidiary organ of the United Nations set up under the same
Chapter, were directly attributable to the United Nations, an organisation of universal
jurisdiction fulfilling its imperative collective-security objective (ibid., § 151). In the
present case, by contrast, the relevant Security Council resolutions, especially Resolutions
1267 (1999), 1333 (2000), 1373 (2001) and 1390 (2002), required States to act in their own
names and to implement them at national level.114

From the two cases, one may understand that the Strasbourg Court had stood
quite well in the understandings of DARIO in Behrami & Saramati, something that
if applied upon the EU, would hold EU responsible for all violations in which the
EU is the normative author while its Member States have had no state discretion.
The contrary may be observed regarding DARIO in Strasbourg’s Nada
v. Switzerland, where the Court makes a difference on basis of whether individual
member states have been mentioned individually as authorized to apply certain
norms—with or without state discretion—to their own territories. Although rather
contrary to DARIO, Nada v. Switzerland stands more in the lines of DAA and its
co-respondent mechanism, as it operates under the meaning that if EU Member
States—whether or not possessing discretion—have been individually called to
apply EU law, they still have a sort of responsibility for that violation; something
which the DAA packages in the format of joint responsibility for the EU and ECHR.
DAA therefore is more or less packaged under the calculated meaning of Nada
v. Switzerland, quite the contrary being the example of Behrami & Saramati.
Now, let us now jump to the friendly settlement mechanism under DAA and
its use.

113
ECtHR, Behrami &Saramati, Grand Chamber, Decision As To The Admissibility, Application
no. 71412/01 Agim BEHRAMI and Bekir BEHRAMI v. France and Application no. 78166/01 by
Ruzhdi SARAMATI v. France, Germany and Norway, para. 143.
114
ECtHR, Nada v. Switzerland, Grand Chamber, Application no. 10593/08, Judgment, Stras-
bourg, 12 September 2012, para. 120 [emphasis added].
192 6 Attribution of Liability Under the Co-respondent Mechanism

6.7 Friendly Settlements and Unilateral Declarations: Any


Interference in the EU Law Autonomy?

It is important to clarify that—in order for the co-respondent and respondent to have
a single position before the Strasbourg Court—the Explanatory Report states that
both ‘the respondent and the co-respondent will need to agree to a friendly
settlement under Article 39 of the Convention.’115 This seems somehow obligatory
that both the respondent and co-respondent jointly agree to a friendly settlement
before the Court decides on the merit of the case. It would not seem feasible for the
DAA to ‘oblige’ the respondent and co-respondent to reach a friendly settlement, as
the respondent and co-respondent are not necessarily in the proceeding to support
each other, but rather to share the responsibility and to ‘protect’ their own domestic
competences in the face of the other. Furthermore, one may not take for granted that
the respondent and co-respondent will agree to the manner of sharing the burden.
Should the DAA ‘aim’ to oblige the respondent and the co-respondent to become
united for the purpose of not allowing the Court to review their division of
responsibility, this would be logical merely to the extent of fulfilling the obligation
of Protocol 8 (1) b of not allowing that the Strasbourg Court review the compe-
tences prescribed by the Treaties.
The fact that the DAA requests a compulsory friendly agreement between the
respondent and co-respondent seems doubtful also in light of Art 39 of the Con-
vention, to which the Explanatory Report refers explicitly. Art 39 (1) of the
Convention sets forth that: ‘At any stage of the proceedings, the Court may place
itself at the disposal of the parties concerned with a view to securing a friendly
settlement of the matter on the basis of respect for human rights as defined in the
Convention and the Protocols thereto.’ Rule 62 (1) of the Rules of the Court
consolidates this further, by establishing that: ‘Once an application has been
declared admissible, the Registrar, acting on the instructions of the Chamber or
its President, shall enter into contact with the parties with a view to securing a
friendly settlement of the matter in accordance with Article 39 § 1 of the Conven-
tion. The Chamber shall take any steps that appear appropriate to facilitate such a
settlement.’ It is obvious that Art 39 of ECHR and Rule 62 (1) are made to serve the
friendly settlement between the applicant and the respondent, which have
contrasted positions in the proceedings.116 The use of friendly settlement serves
merely as a means to surpass the Court’s proceedings and judgment on the merits of
the case.117 It is not made, therefore, as a mechanism to find a unique position
between the applicants, or the respondent and co-respondent which together stand

115
CDDH-UE (2011)16fin, para. 52.
116
See e.g.: Hart (2010), p. 540, showing that friendly settlement is a mechanism that was
originally and now used for the consensus between the applicant and respondent, something
differing from its use in the co-respondent mechanism.
117
See e.g.: Friendly Settlement and Strike Out (Articles 37–38). Article 3 of the European
Convention on Human Rights: A Practitioner’s Handbook. p. 193, available at: http://www.
6.7 Friendly Settlements and Unilateral Declarations: Any Interference in the. . . 193

as the side of the addressee. Moreover, as long as Art 39 of the Convention


establishes the friendly settlement mechanism as a voluntary one, it would not be
logical to consider that there is a duty for both the respondent and co-respondent, as
written in the Explanatory Report, to reach a friendly settlement on their position
before the Court and the case being judged thereto. The accommodation of the
mechanism of friendly settlement for the purposes of DAA seems both legally
tenuous and technically inappropriate to reach the aim for which it is established.
On the other hand, the friendly settlement between the respondent and
co-respondent would not allow the Court to review the division of responsibility
between them, which would entail that the Court is not touching upon the division
of competences deriving from the Treaties118; therefore preserve the autonomy of
the EU law, which does not seem guaranteed at all. This is due to the fact that under
Art 39 (1) ECHR the friendly settlement procedure is enlightened and headed by the
Strasbourg Court, which puts itself at the disposal of both parties. Rule 62 (3) of the
Rules of Court further states that the friendly settlement is reviewed for its com-
pliance with human rights of the Convention by the Court.119 This being said, one
could not argue that the friendly settlement procedure prevents the Court from
engaging virtually in the review of the division of responsibilities between the
respondent and co-respondent, as the procedure as such should be made possible by
the Court’s disposal and engagement, although it is for the parties to agree on it
without the Court’s directions. The friendly settlement, nevertheless, supposes that
there is still a role for the Court to aid the parties in reaching a friendly settlement on
the ‘basis of respect for human rights as defined in the Convention and the Protocols
thereto.’120 The role of the Court, even so, is to supervise whether the settlement
being reached, in this case, between the respondent and co-respondent is standing in
line with the human rights envisaged by the Convention ( jura novit curia), an
involvement which might place the Court in the position of directing the parties on
the ‘eligible’ settlement. Although this procedure must happen confidentially under
Art 39 (2) ECHR, it may certainly provide room for the Court to interfere in the
autonomy of EU law. In addition, one may wonder whether such ‘confidentiality’ of
the friendly settlement procedure between the EU and one or more of its Member
States to locate their share of the burden under EU law amounts to a violation of
Article 15(1) TFEU. This provision requires that openness in judicial processes be
extensively guaranteed by the Union institutions, which also includes the Commis-
sion which will represent the Union in this procedure.121 Friendly settlement

omct.org/files/2006/11/3633/handbook1_eng_08_part8.pdf; See also: Weber (2007), pp. 223 et


seq.
118
As prohibited by Protocol 8 (1)b.
119
‘[T]he Court will examine the terms with a view to establishing whether respect for human
rights as defined in the Convention and the protocols is upheld in the declaration.’ Friendly
Settlement and Strike Out, op. cit. note 117, p. 194.
120
Art 39 (1) ECHR.
121
On the openness concept of the EU Treaties, see: Alemanno and Stefan (2014), p. 104.
194 6 Attribution of Liability Under the Co-respondent Mechanism

mention in the DAA seems inappropriate with its intended architecture and use, and
does not solve the problem of protecting EU law autonomy as there is no assurance
that respondent and co-respondent(s) will settle the share of burden, which might
compromise effective human rights protection from the Strasbourg Court’s side. It
is paradoxical that the EU Court does not refer to this ambiguity in its Opinion 2/13.
The same is true regarding the unilateral declarations, which, according to the
DAA must be reached by the respondent and the co-respondent. The DAA stipu-
lates that: ‘Both the respondent and the co-respondent will need to agree to make a
unilateral declaration of a violation for which they are both responsible.’122 Here
again, the unilateral declaration would be used in contrast with its function under
Rule 62A of the Rules of Court, as therein unilateral declarations are meant to be
utilized only if the respondent and the claimant have failed to reach a friendly
settlement.123 This is, again, a procedural obligation that aims to put both the
respondent and co-respondent on the same path regarding their responsibility.
With the unilateral declaration, the DAA intends to hold both the respondent and
co-respondent in one seat before Strasbourg Court, therefore preventing the Court
to give a single ruling but with separate conclusions for the respondent and
co-respondent: a practice which would allow the Court to apportion the individual
responsibility of respondent and co-respondent, something that will certainly touch
upon the review of competences124 of the EU and Member States as deriving from
the Treaties. The Luxembourg Court’s Opinion 2/13 does not take note of this huge
safeguard which the DAA installs, unfortunately.
It must be argued that the DAA makes it an obligation for the respondent and
co-respondent to present a single unilateral declaration for the violation which the
application is filed for before the Court. Under a teleological interpretation, such
unilateral declaration should not be merely a single document, but also present a
single joint position of both the respondent and co-respondent. This said, the
respondent and co-respondent are obliged to have one single stance, and they are
not independent in presenting their own position. The respondent and
co-respondent are thus required to fix their own share of external responsibility
regarding a violation through a friendly settlement procedure, without the engage-
ment of the Court, or any judicial body of the EU, and present their unique position
to the Court with a single voice, the latter needing to be in line with the principle of
loyal cooperation and primacy of the Treaties. It is not clear, however, what would
happen if the respondent and co-respondent are not capable of having a unified
voice in this regard, and they would need the interpretation of the CJEU, something

122
CDDH-UE (2011)16fin, para. 53.
123
See also: European Court of Human Rights, Unilateral declarations: policy and practice, 2012,
available at: http://www.echr.coe.int/Documents/Unilateral_declarations_ENG.pdf; See e.g.:
ECtHR, Bekerman v. Liechtenstein (no. 15994/10), 29 November 2011; ECtHR, Liptay
v. Hungary, Application no. 12144/09, 22 May 2012.
124
Something that might result in the Strasbourg Court dealing with the ‘separation thesis’
between the EU Court and national constitutional courts’ competences. On the ‘separation thesis’,
see: Thym (2013b), p. 404.
6.8 EU and Member States as Co-respondents 195

which the procedure before the Strasbourg Court would absolutely proscribe. To
digress, the Luxembourg Court’s Opinion 2/13 (in para 234) questions the reliabil-
ity of this mechanism—calling it not sufficiently safe—therefore requesting that the
division of responsibility between the respondent and co-respondent be regulated
by EU law as well (although principles relating to Art. 4 (3) TEU may seem
sufficient EU law safeguards), and provide jurisdiction—should a dispute arise
between the respondent and co-respondent—to the Luxembourg Court to review
the legality of the division of such responsibility. Luxembourg Court’s point is
nothing more than a request to sustain its jurisdiction and keep the issue legal in
nature in cases where political consensus between the respondent and
co-respondent may appear in practice.
On the whole, coming back to the central issue, the unilateral declaration tool
requires the respondent and co-respondent to undertake a self-judgment and com-
pulsorily agree on their division of responsibility towards the violation being
adjudicated before the Court. This process involves an informal arrangement
between the respondent and co-respondent, wherein each of them are obliged to
find a consensual standing before the Strasbourg Court. The obligation to reach a
unilateral declaration could hinder the individual responsibility of Member States
and the EU towards the Convention, and might lead to the Court losing its capacity
to address individual violators of human rights, not necessarily related EU legal
acts. On the other hand, the inability of the respondent and co-respondent(s) to
reach both a friendly settlement and a unilateral declaration might violate the duty
to loyal cooperation under Art. 4 (3) TEU, although for the sake of the procedure
before the Strasbourg Court there is no instrument to remedy a potential situation
wherein they refuse a unilateral position with their burden shared through an
internal agreement.

6.8 EU and Member States as Co-respondents

6.8.1 EU as Co-respondent: Examining Its Normative


Architecture and Potential Implicative Legal Outcomes

To start, Tomuschat had rightly observed that:


An entity discarding any notion of liability for its conduct could not be taken seriously in
international dealings. As strange as it may seem, the capacity to incur international
responsibility is an essential element of the recognition of international organisations in
general and of the European Union in particular as entities enjoying personality under
international law.125

125
Tomuschat (2002), p. 183.
196 6 Attribution of Liability Under the Co-respondent Mechanism

The possibility for the EU to join as co-respondent has been devised to serve
inter alia exactly the above intention. With the EU joining through an institutional
procedure as co-respondent in a certain case before the Strasbourg Court, it not only
fortifies its international recognition but also ensures its status as an entity with
external legal personality under international law. The EU’s participation as
co-respondent is not merely a formal issue, but—in Tomuschat’s words—a sub-
stantive point that makes EU participating convincingly in the domain of interna-
tional responsibility. Therefore, to start, one of the cases whereupon the
co-respondent mechanism could be triggered under the DAA is when an application
is directed against one or more EU Member States (the latter being the primary
respondents), wherein the EU could join as a co-respondent. The alleged violation
may be assumed in two scenarios in this situation: ‘either the underlying provision
of EU law was faulty, which automatically renders its implementation incompatible
with the ECHR, or the legislation was compliant but was implemented in a way
which was not in accordance with the ECHR.’126 Regarding the procedure for the
EU to join as a co-respondent, the DAA reads:
Where an application is directed against one or more member States of the European Union,
the European Union may become a co-respondent to the proceedings in respect of an
alleged violation notified by the Court if it appears that such allegation calls into question
the compatibility with the Convention rights at issue of a provision of European Union law,
notably where that violation could have been avoided only by disregarding an obligation
under European Union law.127

It is obvious from the above paragraph that the EU could join as a co-respondent
if an application has been notified to one or more Member States in respect of an
alleged violation of the Convention by EU law.128 The Explanatory Report, and the
logical interpretation of the above paragraph, leave open the option that the EU join
as a co-respondent in cases both where the alleged violation is supposed to derive

126
Groussot et al. (2011), p. 11.
127
CDDH-UE (2011)16fin, Art 3 (2).
128
See: den Heijer (2013), p. 434; One should note that, although the Member States’ actions or
omissions might have resulted from the application of an EU law obligation, many of such actions
may have already been checked abstractly by the Luxembourg Court’s preliminary reference
procedure. See e.g.: Carrubba and Murrah (2005), p. 401; The European Parliament had proposed
a less mature model of making this established, by proposing that: ‘[. . .]any application by a
natural or legal person concerning an act or failure to act by an institution or body of the Union
should be directed solely against the latter and that similarly any application concerning a measure
by means of which a Member State implements the law of the Union should be directed solely
against the Member State, without prejudice to the principle that, where there might be any doubt
about the way in which responsibility is shared, an application may be brought simultaneously
against the Union and the Member State.’ European Parliament resolution of 19 May 2010 on the
institutional aspects of the accession of the European Union to the European Convention for the
Protection of Human Rights and Fundamental Freedoms (2009/2241(INI)), para. 9 [emphasis
added]. The latter proposal innovates the notion of ‘doubt’ on the share of burden, which is not a
practical instrument as a party may not be checked admissibility before Strasbourg on its targeted
respondents on basis of its ‘doubt’. The current proposal on the co-respondent therefore seems
safer and more legally certain.
6.8 EU and Member States as Co-respondents 197

from the incompatibility of a provision of both the EU primary law and secondary
law with the Convention. This being said, it is argued that the applicant could have
filed an application for an alleged violation deriving from a provision of EU
primary law or secondary law, or both, upon which condition the EU could join
as a co-respondent.
The first condition in this regard, therefore, is that the alleged violation of the
Convention should derive from a provision of EU law, which the applicant should
have specified in writing before the Court in order for the application to become
admissible. It follows logically from Art 3 (2) of the DAA that in order for the EU to
fulfill the criterion of joining as co-respondent, there must be a clear linkage
between the application’s allegation for a violation of the Convention and one or
more provisions of EU law. Should the applicant not have made such allegation,
explicitly or implicitly, there would be no space for the EU to join as a
co-respondent. This condition speaks to the fact that the provision of the EU law
that allegedly violates the Convention must be acknowledged by the application
filed before the Court. Otherwise, there would not be space for the co-respondent
mechanism to be utilized by the EU if nowhere in the application the Court finds
that the allegation leads to an infringement deriving from an EU law provision.
It is also important to note that the wording of the above article makes it clear
that there must be an allegation by the applicant that the violation derived from an
obligation of the State to the EU law, in order for the EU to be able to gain standing
as a co-respondent. It is therefore logically argued that the construction of the above
article leaves space for contention, as it states that there must be an alleged violation
deriving from the EU law in order for the EU to fulfill the condition to join as a
co-respondent. One should also point here to the fact that should the interpretation
of the above article not mean that, then, Art 3 (2) of the DAA would make it a duty
of the Court to check whether the application’s allegation for a violation of the
Convention by one or more Member States derives from an EU law provision.
Should this be intended by the above article of the DAA, then the Court would have
to engage in the interpretation of the EU law, and, notably, to penetrate into its
autonomy,129 which would be in opposition to Protocol 8 of the Treaties, but not
necessarily in opposition to the DAA.
It flows from the construction of the article concerned that the connection
between the alleged violation and the provision of the EU law must not necessarily
be explicit; it would suffice to be implicit. It is argued that the alleged violation is
not required to originate merely from the said provision of the EU law; it suffices
that such violation partially derives from the EU law. One would question whether
it would suffice for the applicant simply to claim that the alleged violation origi-
nates in full or in part from the EU law for the EU to be able to join as a
co-respondent under Art 3 (2) of the DAA. The latter leaves the impression that it
is sufficient for the claimant (or the respondent in its submissions to the Court) to

129
Adding here that such autonomous consideration of its own jurisdiction is kept strongly also by
the Strasbourg Court. See e.g.: Sweet (2009), p. 640.
198 6 Attribution of Liability Under the Co-respondent Mechanism

have simply claimed that the alleged violation, inter alia, originates from EU law,
for the EU to gain standing as a co-respondent.
A core question at this point is whether the meaning of Art 3 (2) of the DAA
seeks a connection between the alleged violation and the provision of the EU law
for the EU to join as a co-respondent for the whole proceeding. It stems from the
logical interpretation of the above provision that the standing of the EU as a
co-respondent could be maintained only for the part of the proceeding wherein
the violation is considered from the EU law perspective. Given that an application
could target more than just the EU law as the origin of the alleged violation,
meaning the Member States’ laws being part of the targeted pool of alleged
violations, would be it be reasonable to question whether the EU’s standing as
co-respondent would be necessary in those parts of proceedings and Court’s
decisions wherein the EU law is not questioned? Given the fact that allowing the
Court to enter the internal borders of competences between the EU and Member
States would constitute a breach of the autonomy of EU law,130 it would be logical
to argue that once the EU joins as co-respondent, it will have the standing of a full
party with the Member State(s) as respondents, and that the Court will lead a joint
proceeding against both parties, not specifically mentioning which party is respon-
sibility for which part of the violation. If the Court would act otherwise, meaning if
it would engage with directing the respondents and the EU as co-respondent on
which part they are obliged to stand responsible for the violation, it would constitute
an act whereupon the Court would engage with the review of the Treaties, thereby
surpassing the autonomy of EU law.131
Art 3 (2) of the DAA explains that the EU could join as co-respondent if an
alleged violation originates in an EU law provision, ‘notably where that violation
could have been avoided only by disregarding an obligation under European Union
law.’132 As also explained in the Explanatory Report, the test upon the EU’s
standing as co-respondent should meet the condition that the alleged violation
would have been avoided if the Member States would have disregarded an obliga-
tion under the EU law (which Gragl, referring to Lock, calls a situation of ‘norma-
tive conflicts’ between the obligation of MSs under the Convention and under the
EU law).133 This test, nevertheless, is complicated with the word ‘notably’ which
appears at the beginning of the last sentence of Art 3 (2). There are two key
questions: first, whether it is mandatory that the standing of the EU as
co-respondent be tested against the condition that the respondent Member State
(s) violation originate directly from an obligation of EU law; and second, whether

130
Cf.: One may apply an argument from the following article with analogy. Alter and Helfer
(2010), p. 586, who argues that this kind of role, which we think Strasbourg Court may attain, at
the end of the day may be one which is made to fill in the treaty gaps—many of which do really
need Strasbourg to go beyond it a logically allowed function.
131
For a general scenario on this, see also: Lock (2010), p. 784.
132
CDDH-UE (2011)16fin, Art 3 (2) [emphasis added].
133
Gragl (2013), p. 158. To this extent, Gragl insists that this situation is one in which the Member
State(s) had no discretion in implementing EU law. Gragl (2013), p. 158.
6.8 EU and Member States as Co-respondents 199

this implies that the co-respondent mechanism offers room for the EU to join
proceedings strictly in cases wherein Member States134 exercised no state discre-
tion but merely implemented an EU law provision.135 State discretion for this
purpose,136 as the Luxembourg Court’s formula establishes, is evaluated in light
of the fact ‘that as long as a member state enjoys a discretionary power the exercise
of which must comply with other provisions of EU law, that member state is
“implementing EU law”.’137
With regard to the first question, the word ‘notably’ appearing in Art 3 (2) is
written to connote something that is particularly the case. It is not meant as
something that could be maintained as a full condition. It rather stands as a general
rule, which implies that the ‘majority’ of the violations should be particularly from
those wherein the respondent Member State(s) would have not violated the Con-
vention should they have disregarded the obligation of the EU law.138 The word
‘notably’, therefore, muddles the waters of interpretation of Art 3 (2), as it clearly
leaves the option that the EU also gain standing as a co-respondent if a Member
State(s) violation of the Convention could not have been avoided by disregarding an
obligation from the EU law. If this would be the most logical interpretation—as the
need to stand within the box of the meaning of ‘notably’ could not be
circumvented—then it flows from such an interpretation that the gateway for the
EU to join as co-respondent usually occurs when the alleged violation could have
been avoided by the Member State(s) by disregarding an obligation from the EU
law, but also includes other cases where that violation could have not been only
avoided by disregarding an obligation under the EU law. This opens the way to join
as co-respondent merely in cases wherein an EU law provision is concerned, not
necessarily being linked with the condition on the strict interpretation of whether
the alleged violation could have been avoided by the respondent by disregarding the
EU law obligations. Paragraph 42 of the Explanatory Report, building upon this,
does not suggest a rigid interpretation of the rule concerned, arguing that it also
stands in the line of the arguments shown above.

134
See: Lenaerts (2012), p. 378, on the scope of EU Member States implementing actions that may
be reviewed against human rights law; See also on this issue: Gragl (2013), p. 158.
135
Gragl takes the examples of Bosphorus and Operation Atalanta to demonstrate examples of
Member States implementing an EU law obligation without state discretion. See: Gragl
(2013), p. 163.
136
Cf.: Eckes (2013), p. 268, who argues that ‘notably’ refers to the cases wherein Member States
did also use discretion when implementing an obligation of EU law.
137
Lenaerts (2012), p. 380.
138
Contra: Gragl insists that, as opposed to my view on the criterion of ‘majority’, this criterion
basically refers to the fact that it would suffice for the Court that ‘the element of a normative
conflict is simply one of several such elements which are not included in this non-exhaustive
enumeration.’ See: Gragl (2013), p. 161; In Gragl’s view, therefore, this standard is fulfilled simply
when the Court is convinced that merely one of the normative conflicts—as opposed to the
majority of normative conflicts, which is my view—is found to derive from the alleged violation
and its relationship with EU law.
200 6 Attribution of Liability Under the Co-respondent Mechanism

On the other hand, if one would argue that the EU could join as co-respondent
only if there were an alleged violation which could have been avoided by the
respondent Member State(s) if they only disregarded an obligation under the EU
law, then it would appear in practice that the EU involvement as co-respondent
could risk the mechanism as such, and the overall autonomy of the EU law. To
demonstrate this with an example, there cases could appear in which a respondent
Member State would have avoided the violation not only by disregarding the
obligation from the EU law, but also from refraining from its sovereign action in
the form of an omission (e.g. M.S.S. & Mathews).139 In such case,140 although the
alleged violation could have not been avoided only by disregarding an obligation of
the Member State under EU law, the Court would have necessarily engaged in the
review of the Member States responsibility under EU law, whereas that would also
imply a possible review upon the interpretation of the EU law. Such action, should
the co-respondent mechanism be interpreted that strictly, would have permitted the
Court to penetrate into the autonomy of the EU law,141 and accordingly interfere

139
Gragl rightly makes the argument, as shown above as well, that the application of the
co-respondent admissibility criteria are especially problematic in the case of omissions, as the
Strasbourg Court would in principle need to embark on EU law to find out who is the competent
authority that should have taken the relevant decision—and which it omitted to take. See: Gragl
(2013), p. 164.
140
In this regard, the Strasbourg Court in MSS had noted ‘that Article 3 § 2 of the Dublin
Regulation provides that, by derogation from the general rule set forth in Article 3 § 1, each
member State may examine an application for asylum lodged with it by a third-country national,
even if such examination is not its responsibility under the criteria laid down in the Regulation.
This is the so-called “sovereignty” clause. In such a case, the State concerned becomes the member
State responsible for the purposes of the Regulation and takes on the obligations associated with
that responsibility.’ (ECtHR, M.S.S. v. Belgium and Greece, Application no. 30696/09, Judgment,
Strasbourg, 21 January 2011, para. 339). MSS is taken as an example to demonstrate that the
commission of a violation could have been done simultaneously by disregarding an EU law
provision (which gives right to the use of the co-respondent mechanism) and by omitting from
making use of its sovereign actions (which would not give the right to anyone to join as
co-respondent). This basically means that to make the assessment on EU’s eligibility to stand as
co-respondent, the Strasbourg Court may only rely on the part of the argument that the alleged
violation has been caused by the respondent Member State’s use of ‘sovereignty clause’, some-
thing that goes outside the scope of EU law, therefore call the EU’s request to join as
co-respondent as not admissible. On the other hand, in this scenario, the EU could have been
admitted to join as co-respondent, but the respondent Member State(s) may claim that the alleged
violation did not take place within the scope of EU law obligations. In deciding whether to choose
between the first or second option, as shown here, the Strasbourg Court would basically be allowed
to interpret EU law, as the Member States’ obligations under it and possible derogations from such
obligations would need to be defined as a prior issue before deciding on the admissibility of the EU
as co-respondent; On the MSS’s sovereignty clause, see also: Gragl (2013), p. 120/1.
141
Cf.: Lang (1986), p. 174, who promotes the idea that there is no room for third parties (like
Strasbourg Court) to enter into the borders of EU law for purposes of demarcating the competences
and consequently the responsibility between the EU and its Member States—something that is
promoted as being of an internal nature.
6.8 EU and Member States as Co-respondents 201

with the Luxembourg Court’s exclusive jurisdiction on that matter.142 Such a


scenario would lead to the EU—as an autonomous legal order143—losing its
integrity within the context of accession to the ECHR, as prescribed by Protocol
8 to the Treaties.
Lock makes a very substantial argument regarding the word ‘obligation’
appearing in Art 3 (2) of the Accession Treaty. He argues:
By referring to ‘an obligation’ the draft does not make it sufficiently clear whether this
obligation is an obligation contained in the allegedly incompatible provision of EU law or
whether it can be any obligation under EU law.

Should Lock’s argument be considered a bit more in-depth, implications arise


from the interpretation of the above provision of the Accession Treaty. It should
first be noted that the standing of the EU as co-respondent would make sense only in
cases where the respondent Member State(s) have had no discretion in
implementing an obligation arising from the EU law. However, Lock’s argument
makes sense, as Art 3 (2) of the DAA leaves open the option that not necessarily the
alleged EU law provision wherefrom the violation originates should derive from the
obligation of the Member State(s) in the context of the last sentence of the above
article. This said, it is argued that there seems no need for a linkage between the
provision of EU law where the applicant alleges the violation originates and the
respondent Member State(s) obligation to implement such provisions under EU
law; more concisely, the argument is that the obligation of the Member State under
EU law could not be the alleged violating provision of EU law pointed by the
application. If this were the case, then it would stem from such provision that the
EU could join as co-respondent also in cases where Member State(s) have
implemented an EU law obligation not necessarily without state discretion, but
merely on the basis of a more general obligation such as the principle of supremacy
or primacy.144 It needs be pointed out that, should the argument of Lock appear
correct, namely that the alleged violating provision of EU law not necessarily be the
same with the obligation of the Member State(s) as respondent to implement it
under EU law, then Art 3 (2) would leave room for the Strasbourg Court to enter
into the borders of EU law, as the Court must need to determine whether a linkage
exists between the alleged violating provision of EU law and the Member State

142
See on this: Court of Justice of EU, Case C-13/00 Commission v. Ireland (Mox Plant),
Judgment of the Court, 19 March 2002, para. 154; Many see the Member States’ courts as the
agents of the Luxembourg Court, but the original argument maintains that it was Luxembourg
Court which was foreseen to stand as a Member States’ agent, something which cannot anymore
remain a reality. See: Alter (1998), pp. 123 et seq; See also: Gragl (2013), p. 158.
143
Contra on whether the Luxembourg Court is autonomous towards the Member States: Alter
(1998), pp. 126 et seq, who argues that the Luxembourg Court at least in its internal autonomy
context may be considered an agent of Member States in controlling the EU institutions.
144
Accord: On the nature of primacy of EU law and its effect on Member States actions, see:
Herman Reestman (2005), pp. 103 et seq.
202 6 Attribution of Liability Under the Co-respondent Mechanism

(s) obligation to implement it: a path that would involve the Court in the interpre-
tation of EU law.145 Such a scenario would impair the EU law autonomy, as the
Court would need to enter into the interpretation of the EU law when testing the
admissibility of the EU as co-respondent under Art 3 (2) of the DAA. Quite
interesting, this huge gap in the DAA has not been captured by the Luxembourg
Court’s eye in Opinion 2/13.
There is one last issue that needs careful consideration. Art 3 (2) of the DAA
leaves the option that the EU join as co-respondent both in cases where respondent
Member State(s) is implementing an obligation from EU secondary law and
primary law. There is no dispute on the rationale for the EU to join the proceedings
as co-respondent if its secondary law is being checked against compatibility with
the Convention. There is however doubt on the standing of EU as co-respondent if
there is the primary law of the EU being checked against compatibility with the
Convention, with Member State(s) being held as respondents. It is a commonly
accepted principle that the Treaties, namely the primary law of the EU, are a matter
of authorship of the High Contracting Parties. The Masters of the Treaties have
been accepted as the only authorized entities to enact and amend the Treaties of the
EU.146 The EU as such has no obligation, but also no authorization, in its own
capacity as an international sui generis organization, to take responsibility over the
Treaties. Art 3 (2) of the DAA makes it evident that the EU could join as
co-respondent in both cases: where respondents are called responsible under EU
primary law and EU secondary law. The argument is that it would be not logical,
but also acceptable, that the EU join as co-respondent in a case which is submitting
the compatibility of a provision of the Treaties with the Convention, as this rests
completely within the realm of responsibility of Member States as Masters of the
Treaties, as also confirmed logically and substantively with Mathews. If the respon-
dent and co-respondent in this case would be held responsible under the Convention
for a provision of EU primary law, it would be logically submitted that the EU
become part of the responsibility which it does not possess and could not maintain.
This tendency, therefore, speaks for room being made to the EU to penetrate
through Art 3 (2) of the DAA into a higher level of authorization regarding the
EU primary law, wherein no space has been given to the EU as such to maintain any
responsibility or competence regarding the Treaties.
It should be argued here that Art 3 (2) is also uncertain with regard to the other
Member States—which are not respondents to the case—participation as
co-respondents in a case involving EU primary law. One may suppose that there
emerges a situation like this: the claimant files an application to the Strasbourg
Court wherein a provision of EU Treaties is alleged to be incompatible with the

145
In this regard, Gragl demonstrates that in the previous drafts of the DAA this provision had
required a ‘substantive link with European Union legal acts or measures [. . .]’. Gragl demonstrates
that the drafters at that time had considered it sufficient to leave this wide-scope criterion. See:
Gragl (2013), p. 157/8.
146
Alter (1998), pp. 121 et seq.
6.8 EU and Member States as Co-respondents 203

Convention, and only one Member State has been addressed as respondent, as the
applicant has exhausted the remedies only in that Member State. If this comes up to
be the case, Art 3 (2) of the DAA allows the EU to join as co-respondent, but it does
not allow other Member States to join to the proceedings as co-respondents. This
would result in a confusing situation wherein other Member States (which are not
respondents) could not join in a proceeding that judges a provision of the Treaties
compatibility with the Convention, taking into account the fact that only all
Member States (as a college) could and should be held responsible for the violation
concerned. This describes the fact that the article concerned leaves no space to other
Member State(s) to join as co-respondents and to take the responsibility over a
violation originating in the primary law, whereas the EU as the non-responsible
party is allowed to join the proceeding in a case involving primary law. Again, the
Luxembourg Court stands silent on this issue, although this is a huge gap of the
DAA. This issue will be analyzed below a bit more comprehensively.
The idea that Art 3 (2) of the DAA allows for the EU to join as co-respondent
when there is an issue of EU primary law checked against the Convention is
rationalized with the fact that, if it would have been provided otherwise, the
Strasbourg Court would have needed to evaluate the EU law and to interpret the
division between the primary and secondary law with the intention of finding where
the alleged violating provision originates from. With the current wording of Art
3 (2), it is acknowledged that the reasons of not specifying whether the EU would
join in the case being considered belongs to the primary law of EU remain in favor
of the autonomy of EU law, which sometimes needs blurred provisions. The fact
that the Strasbourg Court would have needed to interpret the lines between primary
and secondary law when testing the request of EU to join as co-respondent would
have been realistic, as there is no guidance provided for separating an alleged
violating provision of EU law in terms of its origination; the case could have
appeared that the violating provision of secondary law derives from a more general
primary law provision, and therefore the Court’s engagement into the review of the
division between primary law and secondary law could not have been omitted in
such a scenario.

6.8.2 Member State(s) as Co-respondent(s): Examining


the Normative Architecture and Potential Implicative
Legal Outcomes

The DAA enables Member State(s) too to join as co-respondent(s) if an application


is filed before Strasbourg Court on an alleged violation of EU primary law.147 The
rationale for including Member States stands in line with d’Aspremont requirement
that if member states control the decision-making within an international

147
Groussot et al. (2011), p. 12; See also: Streinz (2013), p. 2.
204 6 Attribution of Liability Under the Co-respondent Mechanism

organization like the EU,’ it is posited here that the legal personality of that
organisation can no longer constitute a shield behind which member states can
evade a responsibility that they would have incurred if they had themselves
committed the contested action [. . .].’148 The author agrees with d’Aspremont,
adding here that the EU has rather well shielded its Member States from a more
rigorous review from the Strasbourg Court in the perspective of d’Aspremont’s
argument. That being the risk that needed to be removed, the DAA prescribes that
Member States constitute the second category whereupon the co-respondent mech-
anism may be utilized. Art 3 (3) of the DAA expressly reads:
Where an application is directed against the European Union, the European Union member
States may become co-respondents to the proceedings in respect of an alleged violation
notified by the Court if it appears that such allegation calls into question the compatibility
with the Convention rights at issue of a provision of the Treaty on European Union, the
Treaty on the Functioning of the European Union or any other provision having the same
legal value pursuant to those instruments, notably where that violation could have been
avoided only by disregarding an obligation under those instruments.149

It is clearly understood from the above provision that Member State(s) could join
as co-respondents in a case filed against the EU wherein the alleged provision
violating the Convention right is placed at the primary law of the EU (Member
States being the guardian ad litem for the EU primary law150).151 The rationale
after this solution is to ensure that the Member States, as the only Masters of the
Treaties,152 are attached to a proceeding which engages the compatibility of the EU
primary law with the Convention. It is clear enough that the EU has no competence
to remove the violating primary law provision. If there is a violation found with the
primary law, it would be solely Member States which would have the authority to
remove it, as they stand as the only authorized party to amend the Treaties and/or
ensure the authorship upon them. Therefore, the single idea that Member States are
given the chance to join as co-respondents in a proceeding against EU in this
specific case seems not only rational but also the only way to bind and carry the
responsibility for violations at the level of EU primary law to the parties in charge.
One should also hint to the fact here that the co-respondent mechanism in this case

148
d’Aspremont (2007), p. 101.
149
CDDH-UE (2011)16fin, Art 3 (3); Gragl points that the previous version of this provision had
also mentioned the fact that it is only the Member States which have the constitutional capacity to
amend the Treaties. In this new version of this provision that note, according to Gragl, has been
removed. See: Gragl (2013), p. 166.
150
E.g.: Raba (2013), p. 564; See also: Gragl (2013), p. 166.
151
It is suggested that the EU Court, in its functional position, is always pressured by the member
states in the sense that a ruling of it could be overcome by a Treaty revision. If this would be an
argument, the Strasbourg Court’s position in this context would be to strengthen the role of the EU
Court as untouchable from member states when it comes to fundamental rights enshrined by the
Convention. See the first argument at: Carrubba et al. (2008), p. 438 et seq.
152
This being a normal international organizations’ feature, wherein member states do still remain
the masters of international treaties. See: Alter and Helfer (2010), p. 568/9.
6.8 EU and Member States as Co-respondents 205

allows that there be no hurdle as to whether there is a procedural path to ensure that
Member States are formally held responsible jointly (in solido) with the EU is an
issue that is of internal nature, namely involving the EU and its Member States. The
proposal to exclude the EU from the responsibility over its primary law—wherein
the EU has no competence to remove violations as only Member States may revise
the Treaties—would seem doubtful for the fact that the Strasbourg Court would
then need to find EU responsible for violations stemming from EU secondary law
and the Member States for violations stemming from the EU primary law. Such
delineation of borders between the primary and secondary law would pull the
Strasbourg Court in the review of EU law, and construe the margins between the
primary and secondary law, which would certainly amount to interference in EU
law autonomy.153
The core issue that needs be analyzed here is the question of EU standing as the
respondent for an alleged violation located at the EU primary law, and the Member
States simply having the possibility to join as co-respondents. Understandably, EU
as such has no competence to adopt, amend or hold responsibility for primary law:
that is reserved only for Member States. Lock gives a few arguments to that
question: first, because the DAA provides for the intention of the EU to join as an
equal High Contracting Party to the Convention and this normally requests that EU
be given the chance somehow to act as the responsible party regarding any segment
of its law; second, although the EU has no authority over the adoption or amend-
ment of its primary law, it is nevertheless involved in the amendment process in
accordance with Art 48 TEU, which certainly points to the symbolic character of
such involvement but also its symbolic responsibility for its primary law.154 With
these two reasons, it would make sense to have the EU as a respondent for merely
symbolic reasons.
The question, nonetheless, is a bit more complex, namely would the EU be able
to stand responsible for its primary law. Certainly, there is no means by which the
EU could be made responsible regarding its primary law, and this makes the
indication that it seems doubtful how come that primary law issues be addressed
by the applicant solely to the EU. It is understandable that the Member States could
join as co-respondents in such a case, but that is a matter of their will. Otherwise, it
is only the EU which has been targeted by the applicant for an alleged violation of

153
Lock (2010), p. 783; Contra.: Gragl keeps insisting that the current provision does not leave
room to the Strasbourg Court to interpret EU law in a binding manner. Moreover, Gragl makes the
argument that the Court will only look at where the legal origin of the alleged violation is in this
regard, but does not mention anything in regard to the worry which I have expressed above. See:
Gragl (2013), p. 166; One should disagree again with Gragl as the Strasbourg Court may not make
such an assessment that simply, as it would need to examine whether—for the purposes of the
criteria specified in Art. 3 (3) DAA—a certain legal basis clearly rests with EU primary law or not.
Making this analysis is not possible with an external outlook, as the EU Treaties do have very
complex structures which need be interpreted to reveal whether a certain legal basis stands merely
at the level of the Treaties and not at the level of other secondary sources of EU law.
154
Lock (2012), p. 172/2.
206 6 Attribution of Liability Under the Co-respondent Mechanism

EU primary law, against which the admissibility test by the Court will be
performed. It would make sense nevertheless if an application alleges that the
violation stems from both EU primary and secondary law, which would rationalize
the EU’s standing as respondent in that case. However, there appears to be no
rationale for the EU standing alone as the addressee of an application attacking the
compatibility only of the primary law with the Convention.
In order to test the logic of Art 3 (3) of the DAA, it is indispensable to question
the technical rationale of the situation wherein there appears an application against
the EU for an alleged violation originating merely from the EU primary law. For
such an application to be admissible in Strasbourg, it must have first been exhausted
at the Luxembourg Court in line with Art 35 (1) ECHR. One should question
therefore if it is at all possible for persons, be them legal of natural, to exhaust an
issue of EU primary law before the Luxembourg Court. At the level of EU court
system, there are three legal remedies that could be utilized by individuals to bring a
case before the Strasbourg Court.155 Such legal remedies are: first, actions for
annulment, as regulated by Art 263 TFEU, wherein an individual might request
the annulment of a legislative act, acts of the Council, the Commission, the
European Parliament and European Council, and legal acts of all other institutional
offices/segments of the European Union, wherein the Luxembourg Court is asked to
verify and review the legality of such acts, with the case being made that the
individual applicant has direct and individual interest; second, action for failure
to act, as defined by Art 265 TFEU, wherein an individual lodges a complaint
before the Court that a specific institution of the EU, or a related office, has failed to
act in response to its legal obligation from EU law towards the applicant156; and
third, direct actions for damage, as defined in Art 268 TFEU and Art 340 TFEU,
wherein individuals might ask the Court to act in disputes relating to contractual
and non-contractual damage caused by the EU or a staff working under its
authority.

155
On the debate on locus standi for non-privileged applicants, e.g. see the following CJEU cases:
Case 25/62, Plaumann & Co v Commission [1963] ECR 95; Case C-309/89, Codorniu SA v
Council [1994] ECR I-1853, para. 20; Case T-177/01 Jégo-Quéré v Commission [2002] ECR
II-2365, para. 51; Case C-209/94P, Buralux and Others v Council [1996] ECR I-615, para. 25;
Case 11/82, A.E. Piraiki-Patraiki v Commission [1985] ECR 207; Joined Cases 106-7/63, Alfred
Toepfer and Getreide-Import Gesellschaft v Commission [1965] ECR 405; Case C-50/00, P Uni on
de Pequeňos Agricultores v Council [2002] ECR I-6677—Opinion by Advocate General Jacobs,
para. 60; See also academic literature on this matter: Arnull (1995), p. 7–49; Tridimas and Poli
(2008); Balthasar (2010), pp. 542–550. (ToL introduced a new category of acts [regulatory acts]
that could be challenged using action for annulment mechanism).
156
E.g. cases on failure to act before Luxembourg Court: Cases 10 and 18/68, Societa ‘Eridania’
Zuccherifici Nazionali v Commission [1969] ECR 459; Case 247/87, Star Fruit Company v
Commission [1989] ECR 291; Case T-277/94, Associazione Italiana Tecnico Economica del
Cemento (AITEC) v Commission of the European Communities [1996] ECR II-351; Joined
cases 166 and 220/86, Irish Cement Limited v Commission of the European Communities
[1988] ECR 6473; and, Case T-387/94, Asia Motor France SA v Commission [1996] ECR II-961.
6.8 EU and Member States as Co-respondents 207

As these are the only three legal remedies that could be utilized by individuals
(both legal and natural persons) to file a case before the Luxembourg Court, one
observes that there is nowhere a chance under this remedy system for an applicant
to bring an issue before the Luxembourg Court asking that it review the legality of
the Treaties.157 It is clear enough that the legal remedies and the competence given
to the Luxembourg Court do not provide for the Luxembourg Court’s jurisdiction to
review the legality of the Treaties. The Court’s only competence with regard to the
Treaties is to interpret them. This being argued, Art 3 (3) of the Accession Treaty
seems doubtful when it foresees the possibility that the EU be a respondent in a case
exclusively tackling its primary law before the Strasbourg Court, as the alleged
provision of the primary law should have first been exhausted at the level of the
Luxembourg Court to become entitled to enter the door of the Strasbourg Court.
This being said, the fact appears that no applicant could bring an application
before the Luxembourg Court alleging that a provision of EU primary law is in
contradiction to the Convention, as there is no jurisdiction that would allow the
Luxembourg Court to enter into such a scenario. So, here one has the problem of
qualifying whether it is practical that there be an application exhausted at the EU
level concerning an alleged violation deriving from EU primary law. Of course,
there could not appear a single case exhausted in meritum before the Luxembourg
Court which had alleged that the Court review the legality of the Treaties, or a
provision located thereto. If this would be the case, Art 3 (3) of the Accession
Treaty would not be practicable to be put in place with EU as a respondent
concerning an alleged violation located at the EU primary law. If this stands as it
does here, Art 3 (3) seems useless and not applicable in practice.
However, it is logical also to consider this issue with a motion of flexibility, the
latter being the criteria to be applied in cases wherein the EU is a respondent. It is
possible for the Strasbourg Court to consider the rule on exhaustion with some
degree of flexibility when it comes to cases when EU is the respondent, and the
applicant raises the question of EU primary law compatibility with the Convention.
The Strasbourg Court could consider a case as exhausted at the level of Luxem-
bourg Court if it simply raised the question of a provision located at the secondary
law, which, in the view of Luxembourg Court, was interpreted in line with the
Treaties. This said, the Strasbourg Court would have to consider any case which
implicitly raises an issue of EU primary law, not necessarily exhausted with the
very same allegation at the level of Luxembourg Court, as sufficient for it to fulfill
the rule on exhaustion. Only in this scenario could Art 3 (3) of the DAA be logically
interpreted and deconstructed, and the EU standing as respondent before Strasbourg
in a case alleging the violation from a provision of primary law rationalized. In this
context, the application filed at Strasbourg against the Luxembourg Court would

157
Gragl also questions whether the entire exclusion of primary law from the Strasbourg’s
jurisdiction may be an option. He maintains that ‘since the European Union is unable to redress
human rights violations rooted in primary law, this specific layer of EU law [primary law] should
arguably not be covered by the ECtHR’s jurisdiction’. See: Gragl (2013), p. 127.
208 6 Attribution of Liability Under the Co-respondent Mechanism

simply need to have a linkage with the allegation that the violation is supposed to
rest with the primary law, and not necessarily exhausted before Luxembourg Court
with the same allegation and object.
With the supposed situation that the EU could be brought before Strasbourg
Court as the respondent in a case concerning an allegation of violation deriving
from primary law, Art 3 (3) of the DAA would allow Member States to join as
co-respondent(s). This being said, Art 3 (3) provides for the voluntary clause that
Member States could join as co-respondent in case the EU is the respondent for a
violation alleged to be deriving from primary law. It is also clearly argued that, as
this specific mechanism has been made to serve merely applications before the
Strasbourg Court that concern an EU primary law provision, it is solely the Member
States which would be able and authorized to remove the violation from the EU
law. Hence, the question follows, what if Member State(s) do not join as
co-respondents? Could the EU alone take over the responsibility to remove the
violation originating in its primary law? Probably this situation shows the ‘grey’
story of the efficacy that the co-respondent mechanism could produce in practice
(which, paradoxically, has not been observed in any way by the Opinion 2/13). To
give the answer, it is important to say that the co-respondent mechanism is not put
in place to resolve all potential challenges with regard to the sharing of responsi-
bility; there are many lacunas wherein such challenges could appear. The fact that
Art 3 (3) makes it clear that Member State(s) may join as co-respondents, it surely
gives the option for them not to join at all as co-respondents. In such a scenario, EU
would stand as the only addressee of the case, and logically, Art 3 (3) of the DAA
would give no direction on how to proceed with a case if no Member State joined as
co-respondent; whereas the violation originating from primary law would have to
be attributed to the EU. The only possibility to surpass this hurdle would be the
mandatory invitation of the Strasbourg Court which would bind all parties to join as
co-respondent (which is foreseen in the final DAA draft), adding that Opinion 2/13
(para. 222) considers that ‘invitation’ as not problematic at all from the perspective
of the EU Treaties.
Were the EU held alone responsible by the Court for the violating provision of
its primary law, the argument would be that the mechanism for the Court’s
allocation of responsibility to the responsible High Contracting Party would fall
short. On the other hand, the Court could use the option to call all 27 Member States
responsible in a proceeding wherein the EU alone is held as respondent—something
which it may utilize by posing an invitation to all EU Member States to join as
co-respondents. Should such scenario appear in practice (meaning the scenario with
the Court’s invitation for parties to join as co-respondents, which is now foreseen
with the latest version of the DAA), the Court would have to get into the Treaties
and observe the fact that for them to be changed and for Convention’s violation
addressee(s) to become liable, the Member States would have to be called into
responsibility. That fact alone would be seen as interfering in EU law autonomy, as
the Strasbourg Court would need to interpret the Treaties and find out who are the
addressees for a violating provision originating therefrom. It is therefore suggested,
but also logically submitted, that the Strasbourg Court would need to have no
6.8 EU and Member States as Co-respondents 209

authority to call Member State(s) responsible for a violating provision of EU


primary law, should Member State(s) have refused to join in proceedings as
co-respondents. In such scenario, therefore, Strasbourg Court would have to call
the EU alone responsible although the EU is not the right and liable addressee. It
would then be for the EU and the Member States to resolute this issue internally,
most likely via political consensus, in order for the Member States to be convinced
to change the violating provision of the Treaties. The EU Treaties’ provision on
sincere cooperation (such as e.g. Opinion 1/94)158 can help to solve this issue on
internal legal basis. To digress briefly, it is very ironic to question why the
Luxembourg Court did find this scenario as compliant with the EU Treaties—
noting that it did not observe any problem when it examined this part of the DAA—
in its Opinion 2/13. Such loopholes examined here would need to have been raised
by the Luxembourg Court and examined properly, as thy seriously harm the
Protocol 8 safeguards.
Another scenario would be if there appears that only one or some Member
States, as opposed to all Member States, join as co-respondents to a case against
EU as the main respondent where an issue of primary law is being judged against
the Convention (in a case where no invitation of the Strasbourg Court has been
delivered to all MSs). It is apparent that a violating provision in EU law could only
be removed if all Member States agree to such decision, connoting the fact that the
co-respondent mechanism in Art 3 (3) of the DAA would only make sense if it holds
all Member States jointly responsible for a violation originating from EU primary
law. The question would therefore be whether the Court would hold responsible all
Member States if there is simply one or some Member States which joined as
co-respondents. Logically, again, the Court would not need to enter into the
Treaties and interpret the provisions on how and who can amend them, as this
would lead to interference in EU law autonomy. Therefore, the Court would not
need to hold all Member States liable for the violating provision in EU law, but
merely the respondent and co-respondents. In that case, again, the solution would
be that the EU as respondent and/or some Member States as co-respondent resolve
this issue politically with all other Member States, in order to ensure that the
violating provision in EU law be removed. Luxembourg Court—in Opinion 2/13
(para. 234)—resisted to accept political consensuses in this regard as reliable means
to address the Convention violations (although it pronounced in a very general
language), therefore it requested that this issue be determined fully by EU law and
that sufficient jurisdiction be provided to itself to review potential problems in this
regard.
The Explanatory Report asks that the procedure for Member States to join as
co-respondent(s) to a case in line with Art 3 (3) of the Accession Treaty be
construed in a way wherein Member State(s) make a reasoned request to join as
co-respondent(s).159 This refers to the fact that Member States should state the

158
Court of Justice of EU, Opinion 1/94 [1994] ECR I-5267, paras. 107–109.
159
CDDH-UE (2011)16fin, Explanatory Report to the DAA, para. 46.
210 6 Attribution of Liability Under the Co-respondent Mechanism

reasons that they consider the criteria of the above provision to be fulfilled. In doing
so, Member States would have to mention the fact that they are the Masters of the
Treaties, and according to the Treaties they are the only parties which could be
responsible for and remove the violation originating in the Treaties themselves.
Such reasoning of Member State(s) would need to be considered for its validity
from the Court, in order to test their eligibility to become co-respondent(s). Such a
review from the side of the Court could merely be undertaken if the Court goes into
the Treaties, and checks the validity of the Member State(s) reasoning; this fact
alone would engage the Court in the review of EU Treaties, and would assign itself
to interpret it for the co-respondent admissibility test to be performed. This would
be understood as permitting the Court to interfere to the Luxembourg Court’s
autonomy on the issue. Luxembourg Court’s recent Opinion 2/13 supports exactly
the same line of argument in its paras. 224/5, although with a less academic
language.
There is another layer wherein the autonomy of EU law might be jeopardized.
Art. 3 (3) of the DAA provides an open clause when it comes to the determining
which provisions are considered to be part of primary law. It sets that Member
States could join as co-respondent(s) if it appears that the alleged violation of the
‘Convention rights at issue of a provision of the Treaty on European Union, the
Treaty on the Functioning of the European Union or any other provision having the
same legal value pursuant to those instruments, [. . .].’ Art 3 (3) is clear enough
when it comes to the two founding treaties of the EU, namely the Treaty on the
European Union and the Treaty on the Functioning of the European Union. How-
ever, it leaves open the path on the qualification of the wording ‘any other provision
having the same legal values pursuant to those instruments’. The latter wording
would normally mean all other provisions that have the status of primary law, such
as the Charter on Fundamental Rights,160 possibly any general principle of EU law,
etc. However, to apply that provision in practice, the Strasbourg Court would need
to interpret that opened wording in a specific case, and in order to do so, it would
need to assess the TEU and TFEU to find out which are those other sources which
could be considered of having the same legal value. In pursuing this task, the Court
would certainly encroach on EU-law territory, therefore interfering in the autonomy
of the latter. Such a clause in Art 3 (3) is certainly poorly drafted, and it leaves way
for the Court to encroach on the EU law autonomy even if not intended, but merely
to find out what it means according to the Treaties themselves. The argument is that
the DAA should have prevented every need for the Strasbourg Court to inquire into
the Treaties and interpret them for its admissibility requirements under Art
35 ECHR. Opinion 2/13 makes no observation on this important fact that may
seriously interfere to the EU law autonomy.
In order to test the credibility of the co-respondent mechanism in this case, one
should also consider the situation wherein an application with the EU as a respon-
dent is filed before the Strasbourg Court, whereas the EU submits in the

160
On the legal status of the Charter, see: Lenaerts (2012), p. 377 et seq.
6.8 EU and Member States as Co-respondents 211

admissibility phase that the alleged violating provision does not stem from primary
law, rather it stems from the secondary law. Should that argument be accurate, then
there would be no room for Member States to join as co-respondents. But the
situation might prove to be one in which one or more Member States would apply to
join as co-respondents. In that situation, the Court would have the jurisdiction to
examine and rule whether the Member States are entitled to join as co-respondents
according to Art 3 (3) of the Accession Treaty, whereupon the examination of
whether the alleged violating provision is located at the primary law needs be
examined by the Court itself. This would certainly engage the Court in a review of
the Treaties, and therefore interfere with EU law autonomy. Nevertheless, one
should consider that there might appear a situation wherein one or more Member
States could have diverging positions with the EU as respondent, and that would
certainly need be solved between the EU and its Member States internally in order
to prevent the Court from entering into EU law territory. One such mechanism to
address that potential conflict between the EU and its Member States does not exist
within the Treaties or within the EU judicial system, but needs merely political
consensus. It would be judicious to expect that the Strasbourg Court would not
allow itself to counter the EU’s stance that the alleged violation does not stem from
the EU law, as this would be the only way for the Strasbourg Court to respect the
internal autonomy of the EU. It is unfortunate that the Opinion 2/13 does not
suggest anything with regard to this important issue.
One final issue that is of high significance is the problem with the domestic
exhaustion of legal remedies in the context of Art 3 (3) of the DAA. As argued in
the scenario under Art 3 (2) of the DAA, if there appears a case against Member
State(s) as respondents, the EU could join as co-respondent having preliminarily
done a sort of internal review by the Luxembourg Court on that application. This
was done to ensure that, when there is a case against Member State(s) with an
alleged violation standing somewhere in the EU law, the case not be judged by the
Strasbourg Court without the Luxembourg Court making a review on it. The issue
seems quite the contrast to the procedure envisaged by Art 3 (3) of the DAA. As
Member States are the Masters of the Treaties, when there appears a case which
calls into question the compatibility of EU primary law with the Convention, the
liability for such a violation should be directed towards the Member States only. As
there is the possibility that applicants file an application against the EU for an
alleged violation located in a provision of the EU primary law, Member States
could join as co-respondents. However, the co-respondents are not required to pass
the admissibility test under Art 35 ECHR, and they might join the respondent
simply on their own wish. The application filed against EU primary law needs
not have been exhausted at the level of any Member State legal order.161 Therefore,
one could observe here a situation wherein an application against the EU primary
law, the authority and responsibility over which could be attributed to the Member

161
However, to note, the EU law may not be invalidated before Member States’ national courts:
Court of Justice of EU, Foto-Frost v Hauptzollamt Lübeck-Ost, Case 314/85 [1987] ECR 4199.
212 6 Attribution of Liability Under the Co-respondent Mechanism

States only, be judged at Strasbourg without it first having been exhausted at the
domestic level of the entities that hold authority over it. This said, there could
appear a situation where Member States are held liable for the EU primary law
without their courts having had first the opportunity to judge and possibly avoid
such violation located within the EU primary law, by asking their institutions to
start the process of amending the Treaties with the purpose of removing the
concerned violation.
This issue certainly hinders the ability of the Strasbourg Court to adhere to the
principle of subsidiarity, by allowing first the Member States as the responsible and
only authorized parties to remove the alleged violation from primary law through
exhausting the case domestically. With this in mind, Art 3 (3) of the DAA allows
room for this paradoxical situation, wherein the rule of exhaustion and the principle
of subsidiarity are not applied upon the responsible parties which are liable for EU
primary law, whereas the rule of exhaustion is merely applied to the respondent,
that is the EU, which in fact holds no responsibility and could not remove the
alleged violation from the primary law.162 One must argue therefore that this lacuna
seriously harms the concept of subsidiarity in the Strasbourg human rights regime
when it comes to EU primary law with the EU as the respondent, something that
could enter into the limits of prohibition of the Patent Opinion—as the Member
States are neglected the opportunity to judge the alleged violation of primary law at
the domestic level first (Cf: Treaty of Lisbon case before German Cons Court).163
The argument should be extended in the direction that it is obvious that national
courts only rarely engage in the review of the legality of the Treaties, however, that
possibility still exists, and the rule on exhaustion has been applied almost without
exception. This was certainly not the case with Art 3 (2) of the DAA, as the EU as
co-respondent was given the chance to provide for an internal review before its acts
or the Treaties were checked against the standards of the Convention. Art 3 (3) there-
fore seems to provide room wherein Member States are unprivileged and excluded
from the principle of subsidiarity in the context of the standing of the EU as sole
respondent regarding its primary law.164

162
See e.g.: van den Berghe (2010), p. 122.
163
BverfGE, 2 BvE 2/08 Gauweiler v Treaty of Lisbon, judgment of 30 June 2009, available at
http://www.bundesverfassungsgericht.de/entscheidungen/es20090630_2bve000208en.html;
Thym (2009), p. 1795–1822; Editorial Comments (2009), p. 1023–1033.
164
Gragl, referring to O’Meara, notes that it would be especially interesting to see Member States
joining as co-respondents in a case involving an obligation relating to the primacy of EU law
question (Gragl 2013, p. 167).
6.8 EU and Member States as Co-respondents 213

6.8.3 EU and Member State(s) as Joint Respondents: Still


a Possible Scenario

The Explanatory Report foresees the possibility that there be applications against
both the EU and Member State(s) as respondents in the same case.165 In such a
scenario, the rule of exhaustion would have to be applied in both respondents’
judicial systems, meaning the Luxembourg Court and the Member State(s) national
courts. Although this might be theoretically possible, in practice it would be rather
unusual to see an application that has exhausted substantively the same case in both
the EU and member State(s) legal orders, as there is a rather well defined division of
labor between the Luxembourg Court and Member State(s) courts when it comes to
individual legal remedies that could be utilized on the basis of EU law. However,
should there appear an application that is filed against both the EU and Member
States jointly as respondents, one should wonder whether there is a possibility for
any of the respondents to become a co-respondent.
The Explanatory Report makes it clear that in such a situation, after the case has
passed the admissibility test under Art 35 ECHR, and after having been notified to
the respondents, one of the respondents could request to turn its status into that of
the co-respondent. In such a case, the change of status would have to meet the
requirements of Art 3 of the DAA for parties willing to become co-respondents. The
choice therefore rests merely with the respondents. The Explanatory Report makes
it clear that the respondent that could use this right is the one which is not
responsible for having caused the violation by acting or omitting to act, but merely
for the legal provision wherefrom such action or omission derived. This needs be
tested against a supposed practical example. If there is an application alleging that
one or more Member States have performed an action or have omitted to act under
an obligation from the EU law, then it would be merely the EU which could use the
right to turn its status from respondent to the co-respondent. Should there be an
alleged violation deriving from the primary law of the EU, whereas the alleged
violating action or inaction of EU derived therefrom, then it would be in principle
the Member States which would have the right to turn their status from respondent
(s) to co-respondent(s), as they are the parties responsible for legislating the
primary law. However, the latter case could be doubtful in one dimension: that
there could be a situation where the action in case of an obligation flowing from
primary law could have been undertaken jointly by the EU and the Member States;
in such a case, it is submitted that none of the respondents would have been able to
use the right of turning their status from respondent into co-respondent.
One would question why should there be the option that if there are more than one
respondents, whose admissibility has been checked under Art 35 ECHR, that they be
permitted to change their status into co-respondents. The rationale would be that
nevertheless there is a moral difference between the status of the respondent and the

165
Explanatory Report to the DAA, Art. 49.
214 6 Attribution of Liability Under the Co-respondent Mechanism

co-respondent, independent from the fact that there is no legal difference between
them once they are notified of the application by the Court. The DAA recognizes that
moral difference. Such moral difference submits the picture that the co-respondent
would not be as responsible for the application as the respondent, otherwise there
would be no reason for the Accession Treaty to provide for the option that one or
more respondents be turned into co-respondents. To bring a brief discussion, ICJ’s
Judge Gaja gives a ground-breaking argument by asserting that ‘[a]nother conse-
quence of there being a co-respondent is that it becomes possible to trigger the
procedure for “prior involvement” of the Court of Justice.’166 Practically, this
means that if EU would like to utilize its prior involvement mechanism, it would
need and wish to turn its status to that of ‘co-respondent’ from the previous ‘respon-
dent’ status. One may add to this that as the e.g. EU’s legal remedies have been
exhausted—since the applicant had attacked the EU directly as respondent before the
Strasbourg Court—why would EU need to utilize prior involvement afterwards? This
is essential, as exhaustion of legal remedies never guarantees that the EU Court has
essentially reviewed the very EU legal norm which is the core norm attacked before
the Strasbourg Court. This may provide up-to-the-minute room for the Luxembourg
Court to review it through the prior involvement mechanism. Therefore, the oppor-
tunity for the EU in this scenario to alter its status to co-respondent may well be
within the intention of Luxembourg Court of reviewing what it had not done before,
and trying to protect its autonomy from a later-phase perspective.
The chance that one of the respondents be turned into co-respondent, on the
other hand, puts into question the entire application of the applicant before Stras-
bourg. It is well known that the applicant directs the arguments and reasoning
towards all respondents consecutively: the fact being made that one of the respon-
dents is turned into co-respondent, it would remain not possible for the applicant to
direct the arguments towards the co-respondent that was turned so from the position
of respondent. The application normally is against an action or omission to act that
should be linked with the respondent, and this practice does not allow it to occur.
This remains an ambiguity that may seriously harm the position of the claimant as
regards his/her right to safe fair trial before the Strasbourg Court.

6.9 Ambiguities in the Co-respondent Mechanism: Why Is


It So Blurred?

The co-respondent mechanism seems to fulfill some of the needs of the EU system
of law, to preserve the autonomy of the EU law and the exclusive jurisdiction of the
Luxembourg Court. Gragl in this regard argues that this mechanism does not seem
to interfere to the EU law autonomy167—but he does not examine whether this

166
Gaja (2013), p. 6.
167
Gragl (2013), p. 156.
6.9 Ambiguities in the Co-respondent Mechanism: Why Is It So Blurred? 215

mechanism interferes to the credibility on human rights protection by the Conven-


tion system, which is a more important issue than the previous one. This being the
issue, there appear many problems with regard to the good administration of justice
that this mechanism is supposed to convey. Such implications seem to rationalize
the view that the co-respondent mechanism was deliberately construed as vague,
with many ambiguities in its design, aiming to allow for a flavor where neither the
Luxembourg Court168 nor the Strasbourg Court would see fundamental rights
protection being traded off for EU legal autonomy. In order to give a general
picture of the ambiguities observed in the co-respondent mechanism, it is necessary
to use presumptions which could happen in practice, and to test them against the
design and application of the co-respondent mechanism. Let us start with the most
challenging ambiguities. To note, Opinion 2/13 merely confirms some—out of
many—ambiguities which this book has observed on this mechanism.

6.9.1 Ambiguity 1: The Discretionary Nature


of the Co-respondent Mechanism

One of the core ambiguities of the co-respondent mechanism, but also of the entire
DAA, is that the status of co-respondent is one that could be gained in no other way
but by the will of the party willing to join as co-respondent (including the scenario
when the Court invites the co-respondent(s) to join on basis of Art. 3 (5) DAA). The
Court may not force a party to become a co-respondent.169 The Accession Treaty
sets this nature of the co-respondent mechanism by deciding that the ‘European
Union or a member State of the European Union may become a co-respondent to
proceedings by decision of the Court in the circumstances set out’170 in the DAA.
This clearly shows that the co-respondent mechanism is a discretionary tool that
might be used only by those willing to, although the Court itself may also invite a
party to join—such invitation being non-mandatory in nature.171

168
In this regard, it is worth noting that the Luxembourg Court had ruled in Opinion 2/00 that its
opinion on an international agreement ‘is not intended to solve difficulties associated with
implementation of an envisaged agreement which falls within shared Community and Member
State competence.’ Court of Justice of EU, Opinion 2/00 [2001] ECR I-9713, paras. 17/8. This
assumed, the Luxembourg Court would not be intended to solve any of the difficulties that would
seem in the co-respondent mechanism or prior involvement, and other issues, through its opinion
on the DAA. These ambiguities will remain to be solved through the regular case-law.
169
E.g.: den Heijer (2013), p. 434; Gragl in this regard questions why co-respondents should not be
forcibly requested to join. In this regard, Gragl advocates for inserting a provision in the DAA that
would make it an obligation to co-respondents to join the procedure whenever their law is at stake.
See: Gragl (2013), p. 156. This position of Gragl has been roughly in the same way argued by the
Luxembourg Court in its Opinion 2/13. See supra the subchapter on Opinion 2/13 which explicates
this in details.
170
Art 3 DAA [emphasis added]; See also: Raba (2013), p. 565.
171
Explanatory Report to the DAA, para. 53.
216 6 Attribution of Liability Under the Co-respondent Mechanism

The core function of the co-respondent mechanism is to address the liability for
an alleged violation to the right addressee, which is the party that is both liable and
that should be capable of removing the violation. This stands in line with the
requirement of Protocol 8 of the Treaties, which requests that the DAA ensure
that ‘individual applications are correctly addressed to Member States and/or the
Union as appropriate.’ It is obvious that the discretionary nature of the
co-respondent mechanism does not allow for such intention to be reached.
The key question in this regard is whether the discretionary nature of the
co-respondent mechanism may force liable parties to claim their liability over an
alleged violation. The argument here should be seen from two sides. Initially, it
must be acknowledged that the discretionary nature of the co-respondent mecha-
nism was made to assure that the autonomy of EU law is not jeopardized, as the
Strasbourg Court in contrast, would have had to enter into the territory of EU law in
order to find the right addressee for an alleged violation (which may still be the case
in cases of Court invitations on basis of Art. 3 (5) DAA). But the question remains
what if the right addressee does not join as co-respondent, having in mind the fact
that the co-respondent mechanism does not possess an apparatus to address the
alleged violations to the correct addressee(s). Gragl in this regard also notes that the
DAA remains silent on the method that may lead to the correct addressee.172 In that
situation, the ambiguity of the co-respondent mechanism is evident. This said, the
discretionary nature of the co-respondent mechanism does not assure that the
violations are to be attributed to the correct addressees,173 and there is no duty
for the liable addressees to join as co-respondents.174 This still may not be in
compliance with Protocol 8 to the Treaties, as the latter request that the attribution
of responsibility be is obliged to be related to the right addressee. To mention, the
very recent Opinion 2/13 of the Luxembourg Court stands silent on this very
important issue.
This scenario brings the answer back to the rationale that the discretionary nature
of the co-respondent mechanism was built upon the foundational concept that the
Strasbourg Court should not deal with the process of sentencing the liable addressee
for an alleged violation when it comes to issues wherein both EU and Member
States are engaged. The Strasbourg Court is, in such context, merely to rule whether
there was a violation or not, and to leave the rest of the issue to the EU and its
Member States. This model of policy setting, therefore, makes impossible for the
Strasbourg Court to locate not only the violation, but also the responsible violator,

172
Gragl (2013), p. 142.
173
One should take the example of the European Arrest Warrant to consider a good case of close
judicial cooperation between the EU and its Member States as regards arrest warrants. This might
serve as a promising example to optimistically trust that the EU and its Member States can deal
themselves in finding a balanced and fair division of burden when they are found as co-respondents
liable for an ECHR right violation. See generally: Alegre and Leaf (2004), p. 201 et seq.
174
Cf.: Kokott and Sobotta (2012), pp. 1016 et seq, who argue—by contrast to this possibility—
that the EU Court itself applies a rather high standard on human rights, at the level to which UN
law would be refused application if it contradicts EU Treaties human rights law.
6.9 Ambiguities in the Co-respondent Mechanism: Why Is It So Blurred? 217

when there is an EU–Member States case before it, clearly dismissing its role as the
locater of the liable party that needs to remove the violation. This obviously limits
the function of the Strasbourg Court, and dismisses the concept of fair administra-
tion of justice, although the contrast appears formally proclaimed in the DAA.
The discretionary nature of the co-respondent mechanism builds upon the
presumption that the EU, and its internal affairs with the Member States, are
projected as one, single picture in the view of the Strasbourg Court. In turn, this
speaks to the detail that the discretionary nature of the co-respondent mechanism is
built upon the rationale that there should be a political consensus between the EU
and its Member States that each stand, due to their Treaties’ obligation, liable for
their part towards the Strasbourg Court through an internal agreement where no
Strasbourg access is given thereto. Such internal agreement between the EU and
Member States to appear as one voice before Strasbourg, wherein Strasbourg has no
authority upon deciding which the liable subject is, seems to be of a political nature.
This said, the division of responsibility between the respondent and co-respondent,
including those would-be co-respondents that have not joined, should be agreed in
no other way but through a political consensus between them, allowing no external
access of the Strasbourg Court therein. In such a state of affairs, the political
consensus to take over the liability of an alleged violation towards the Convention
law between the EU and its Member States automatically excludes the legal nature
of the Convention system, and allows that the correct liable addressee be left to
political consensus.
The question that arises in the context of the discretionary nature of the
co-respondent mechanism is therefore how would the Strasbourg Court behave
were that the respondent, in a situation where no co-respondent has joined, refuses
to accept that it is the responsible entity to be worn with the alleged violation.175 It
clearly follows from the nature of the co-respondent mechanism that it is for the EU
and its Member States to resolve that problem internally, whereas the Strasbourg
Court would have no authority to enter into that political dilemma upon which the
Union is built. Leaving this issue to the EU internal affairs domain, nevertheless,
puts forward the argument that the Convention system does not provide for a fully
fledged system of human rights protection, and a lot of direction on choosing the
methods to divide the responsibility remains within the domestic affairs domain of
the EU and its Treaties’ law. This said, in a scenario wherein a correct addressee
does not join as co-respondent, whereas the respondent attacked with the applica-
tion of the claimant does not accept its responsibility over that violation, the way
forward is something that should not be ruled by the Strasbourg Court and is merely
an issue of the domestic affairs of the Union. In such a scenario, it would suffice for
the Strasbourg Court to find the case admissible and to rule on the alleged violation:

175
As an additional example, Gragl argues that in a scenario where EU does not joint as
co-respondent and the violation stems from primary law, the Member State(s) as the original
respondents may not counter the EU’s decision not to join, therefore the Strasbourg Court would
apply on them the Mathews standard on Member States as the sole respondents for primary law
violations. See: Gragl (2013), p. 157.
218 6 Attribution of Liability Under the Co-respondent Mechanism

the rest of the story remains for EU domestic law to resolve. Opinion 2/13 should
have carefully observed this lacuna, and rule on the treaty mechanisms that need be
devised to surpass this hurdle. Unfortunately, and ironically, it did not tackle this
significant issue.
Finally, as a point of further discussion, Delgado Casteleiro basically argues that
Member States’ duty to intervene—although the mechanism is of a voluntary
nature from the perspective of the Convention system as an instrument of interna-
tional law—will be an issue governed by domestic law of the EU. He takes the
example of the Etang de Berre case where the EU Court ordered a kind of unitary
representation in external affairs for the Community and its Member States.
Delgado Casteleiro goes on by noting that the obligation of unitary representation
before the Convention system is an issue of domestic law, rather than international
law. Such duty, he argues, if violated, may end up with the Commission initiating
an infringement action against a Member State before the Luxembourg Court.
Therefore, refusing the invitation to join as co-respondent would violate the Con-
vention which, he argues, would go against the interests of the EU and duty to
sincere cooperation, therefore violating internal EU law as well.176 In essence, the
author of this book agrees with Delgado Casteleiro that—from a fundamental
perspective developed by the case-law of the EU Court—the refusal of an EU
Member State to join as co-respondent in a certain case may qualify as a violation of
the principle of sincere cooperation under the EU Treaties. However, seen from the
complex picture of scenarios in which the co-respondent mechanism may be
triggered, this does not seem as simple as portrayed by Delgado Casteleiro, and
that for two basic reasons: first, it is not always clear when the criteria to join as
co-respondent for a Member State are satisfied. Interpreting those conditions may
often seek for the Strasbourg Court to enter into the borders of EU law, which EU
may easily argue as violating the autonomy of EU law (this is also supported by
Opinion 2/13, para. 224). And, second, EU Member States refusal to accept joining
as co-respondent may be often an issue deriving from the reasons not pertaining to
EU law but rather to the applicant’s wrong identification of the potential respon-
dents. This said, one should not so rigidly consider the refusal to join as
co-respondents for the EU Member States as constituting a violation of the principle
of sincere cooperation.

6.9.2 Ambiguity 2: Strasbourg Court’s Plausibility

The co-respondent mechanism sets that a party ‘may become co-respondents to the
proceedings in respect of an alleged violation notified by the Court if it appears that
such allegation calls into question the compatibility with the Convention rights at

176
Casteleiro (2014), p. 118/9.
6.9 Ambiguities in the Co-respondent Mechanism: Why Is It So Blurred? 219

issue of a provision’177 of the EU law. It is clear enough that the engagement of the
co-respondent could be materialized only with the decision of the Court.178 Thym
asserts that—in this regard—DAA allows the Strasbourg Court ‘to handle the
trigger for the corespondent mechanism generously.’179 In order for the Strasbourg
Court to issue a decision on a party to join as co-respondent, it must appear180 to the
Court that such allegation points to a violation originating in the EU law. This
process, wherein the Court is convinced that a party fulfills the admissibility criteria
to join as co-respondent, requires that it must be visible to the Court that the party
concerned has a legal interest in the law wherein the violation is supposed to be
located. In order for it to appear before the Court as fulfilling the criteria to join as
co-respondent, the Court would normally need to engage into a fundamental review
of the application of the party to join as co-respondent.181 Arguendo, this leads to
the question of whether the Court needs to enter into the EU law arena to evaluate
whether the party seeking to join appears to fulfill the admissibility criteria laid
down in the DAA. One such requirement includes that the case in question involves
an alleged violation originating in the EU law. Such scenario would put the Court
into utilizing the plausible clause as a means to engage with the EU law interpre-
tation for its own purposes.182 The same view has been upheld by the Luxembourg
Court in its Opinion 2/13 (para. 221 et seq; see infra subchapter on Opinion 2/13).
To this end—while discussing the voluntary nature of this mechanism—Thym
brings the amusing question that ‘[i]f the Commission’s ex ante-screening fails to
identify relevant cases involving the implementation of Union law among the
21,189 complaints brought against 28 EU Member States in 2012, the EU loses

177
Art. 3 (2 & 3) DAA; On the voluntary nature of co-respondent mechanism, see also: O’Meara
(2011), p. 1821.
178
Contra: DAA ‘leaves open the question as to whether it is for the EU or for the Court to decide
when and whether the requirements to trigger the co-respondent mechanism have been met.’
(AIRE centre, AI & ICJ (2013) ‘European Union Accession to the European Convention on
Human Rights: Briefing Note on the Accession Agreement’. (Brussels, 6 September 2013, p. 3).
179
Thym (2013a), p. 2.
180
Contra: Gragl insists that the wording ‘appear’ basically means that the Strasbourg Court is
now allowed to examine in-depth the EU law for the purpose of this provision. Gragl insists that
the wording ‘appear’ is basically similar to the Strasbourg Court needing to ‘view’ whether the link
required in the above mentioned provision exists. See: Gragl (2013), p. 159. In contrast to Gragl, I
argue that one may not that simply calculate the range of jurisdiction which comes under the term
‘appear’, since, as the provision stands right now, the Strasbourg Court may use the jurisdiction of
‘appearing’ to the extent of being convinced that the linkage between the obligation and alleged
violation is clear enough. Such jurisdiction will certainly provide the Court with indefinite capacity
to enter into EU law borders in order to find out that linkage. It would not be valid to uphold the
point that ‘appearing’ would merely allow the Court to find this link from an external position, as
EU law is so complex that such ‘view’ to be taken without clear in-depth analysis is not possible.
181
Contra: Gragl insists that this wording does not give to the Strasbourg Court any ‘persuasive’
jurisdiction to embark on the EU law. I disagree for the abovementioned reasons with Gragl on this
issue. See: Gragl (2013), p. 160.
182
Cf.: Licková (2008), p. 468, arguing that: ‘[b]ut whether the division of powers within the EU
will be taken into consideration internationally depends entirely on the agreement of third parties.’
220 6 Attribution of Liability Under the Co-respondent Mechanism

out on the option to be a co-respondent.’183 This speaks to the fact that not only does
the EU needs put in place a mechanism to ensure that it reviews all cases which
involve EU law substances, but also that the EU’s potential error to identify one
certain case may result in the Convention system losing its capacity to apportion
responsibility to the correct parties.

6.9.3 Ambiguity 3: Share of Burden Between the EU


and Member States When They Appear
as (Co)respondents Jointly

In case the Strasbourg Court finds that there has been a violation committed in a
case involving both the EU and Member States, it needs to issue a decision
confirming that the Convention has been violated by its High Contracting Parties,
at least one of them being co-respondents. Such a decision of the Court would
merely need to rule that there has been a violation committed, leaving the rest of the
story to the EU and its Member States. This scenario, where the EU and Member
States are part of the case as respondent(s) and co-respondent(s) ad quod damnum,
binds both the respondent(s) and co-respondent(s) jointly for the committed viola-
tion.184 Tomuschat would argue that—as regards joint responsibility—‘there are
instances where the strict application of this doctrine would be manifestly unrea-
sonable’.185 To jump to the core issue, the question then remains how the share of
burden would be determined between the respondent and co-respondent in such a
case, wherein both the EU and its Member State(s) are bound by a single decision of
the Court. The latter, as the Court has noted, for a different reason, in Assanidze
v. Georgia, is a result of the fact that: ‘Unlike the American Convention on Human
Rights of 22 November 1969 (Article 28), the European Convention does not

183
Thym (2013a), p. 2.
184
E.g.: Tulkens (2013), p. 11; See also: Gragl (2013), p. 168; Contra.: An example of separate
responsibility is the Community’s declaration of competence annexed to the Convention on the
Transboundary Effects of Industrial Accidents, which, amongst others, reads: ‘[a]s regards the
application of the Convention, the Community and its Member States are responsible, within their
respective spheres of competence’. Emphasis added. See: Annex II to Council Dec 98/685 [1998]
OJ L 326/1; The latter clearly seems to establish separate responsibility between the EU and its
Member States, which goes contrary to the model of responsibility established by the DAA’s
co-respondent mechanism.
185
Tomuschat (1983), p. 130/1; Cf.: Ryngaert (2011), p. 1015 (‘[. . .] allowing a Member State to
be held responsible for an IO’s own conduct during its lifetime may tend to negate the separate
personality of the [. . .International Organization. . .].’); Cf.: Gragl (2013), p. 168, who notes that
there may be the possibility for each co-respondent to make arguments before the Strasbourg Court
on issues relating to their individual liability for the alleged violation.
6.9 Ambiguities in the Co-respondent Mechanism: Why Is It So Blurred? 221

contain a ‘federal clause’ limiting the obligations of the federal State for events
occurring on the territory of the states forming part of the federation.’186 For this
purpose, we bring the example of Art. 28 of the American Convention on Human
Rights, which reads187:
1. Where a State Party is constituted as a federal state, the national government of
such State Party shall implement all the provisions of the Convention over
whose subject matter it exercises legislative and judicial jurisdiction.
2. With respect to the provisions over whose subject matter the constituent units of
the federal state have jurisdiction, the national government shall immediately
take suitable measures, in accordance with its constitution and its laws, to the
end that the competent authorities of the constituent units may adopt appropriate
provisions for the fulfillment of this Convention.
3. Whenever two or more States Parties agree to form a federation or other type of
association, they shall take care that the resulting federal or other compact
contains the provisions necessary for continuing and rendering effective the
standards of this Convention in the new state that is organized.188
The American Convention on Human Rights is a contrary example compared to
the co-respondent mechanism—the latter seeks a joint position of the respondent
and co-respondent(s) before the Court, whereas the Court is obliged to find them
jointly liable in spite of the fact that one of them may have neither legislated nor
implemented the violation of the Convention. The federal clause of ACHR makes
the division/attribution of burden direct and thin, without involving any form of
liability for an unliable party. However, it allows the Convention to penetrate into
the domestic constitutional structures of the contracting party to identify which
could have been the ‘real violator’ (which level of government) on the basis of the
constitutional competence. Art. 28.1 and Art. 28.2 ACHR makes it clear that the
attribution of responsibility goes hand in hand with the level and scope of compe-
tence that the level of entity has in a federal state—quite similar to the EU’s vertical
levels of power (although far less complex compared to the EU). In a scenario like
the ACHR, the EU would have not held any type of responsibility for its Member
States and vice versa. This would have made the attribution of responsibility—but
also the effectiveness of the Convention system—far more appropriate and suitable
for the human rights protection mission. The Strasbourg Court on the other hand
would have located the real violator with its decision, and the claimant would have
had it far easier to file the application against the liable party before the Strasbourg
Court. The author—however—upholds the argument that the ACHR model would

186
ECtHR (GC), Assanidze v. Georgia (Appl. No. 71503/01), judgment of 8 April 2004, para. 141.
187
Adopted at the Inter-American Specialized Conference on Human Rights, San José, Costa
Rica, 22 November 1969.
188
Emphasis added.
222 6 Attribution of Liability Under the Co-respondent Mechanism

have seriously harmed the external autonomy of the EU,189 adding that it may have
brought several ramifications in the relationship between the EU and its Member
States in cases when there are shared competences. It is obvious therefore that—as
opposed to the EU—parties to ACHR seem absolutely not concerned about their
federation’s external autonomy. As this is not the case with DAA, let us turn to the
in-depth deconstruction of the ambiguity introduced in the title of this section.
To note, coming back to the main issue, it is implied from the general rationale
of the DAA that the division of responsibility between the EU and its Member
States in a case where they are joined respondent(s) and co-respondent(s) could not
be ruled by the Court190: the only way is that they agree themselves on the share of
burden.191 Such internal agreement between the respondent(s) and co-respondent(s)
on the share of burden would not seem to fulfill the criteria for a legally certain
protection of human rights by the Strasbourg Court, adding that such agreement
could not in principle fall into the review jurisdiction of the Court of Strasbourg.192

189
Cf.: Opinion of AG Tesauro delivered on 13 November 1997, in case CJEU, Hermès Interna-
tional v FHT Marketing, Case C-53/96 [1998] ECR I-3603, p. 3621 (‘The [. . .EU. . .] legal system
is characterised by the simultaneous application of provisions of various origins, international,
[. . .EU. . .] and national; but it nevertheless seeks to function and to represent itself to the outside
world as a unified system. That is, one might say, the inherent nature of the system which, while
guaranteeing the maintenance of the realities of States and of individual interests of all kinds, also
seeks to achieve a unified modus operandi. Its steadfast adherence to that aim, which the Court
itself has described as an obligation of solidarity, is certainly lent considerable weight by the
judicial review mechanism which is defined in the Treaty and relies on the simultaneous support of
the [. . .EU. . .] court and the national courts.’). This balance sought to be established by the above
statement may seriously become at risk if the model of distributing liability between the EU and
Member States would have followed the ACHR approach.
190
Another scenario is provided for in Art. 3 (7) DAT which establishes that there is no exception
to the rule on joint responsibility ‘[. . .] unless the Court, on the basis of the reasons given by the
respondent and the co-respondent, and having sought the views of the applicant, decides that only
one of them be held responsible.’ Attributing individual responsibility under this scenario,
therefore, may be possible only with the assent of the respondent and co-respondent and their
reasoned claim. That means that, practically, both of them need to prove before the Court that they
have reasons—including a plan of reasons to attribute the burden to only one of them—which
would convince the Court to take that individualized decision. This scenario, however, seems to
leave room to the Strasbourg Court to enter into the EU law domain, since the Court would need to
review the reasons which the respondent and co-respondent(s) have provided and rule upon their
justification as means to accept or deny the individual responsibility of one of them; The latter
argument is supported in the same form with the recent Opinion 2/13 (para. 234).
191
Lock (2010), p. 787; Gragl (2013), p. 168/9; Gragl mentions the fact that there may be
situations in which Member State(s) will be pushed internally for certain political reasons not to
accept the blame on the violation judged by the Strasbourg Court, especially in cases when there
are monetary compensations (Gragl 2013, p. 168).
192
For instance, the UN Convention on the Law of the Sea provides a different model of separating
the responsibility between state parties while not interfering to their internal legal orders. Art. 6
(2) (Annex IX)—for that purpose—reads: ‘Any State Party may request an international organi-
zation or its member States which are States Parties for information as to who has responsibility in
respect of any specific matter. The organization and the member States concerned shall provide
this information.’ [emphasis added]; Should DAA have adopted this model of separating
6.9 Ambiguities in the Co-respondent Mechanism: Why Is It So Blurred? 223

Given that the share of burden between the EU and Member State(s) in such
scenario would seem to be an issue of the EU internal law,193 one should also
consider the view that there might be disagreements between the respondent(s) and
co-respondent(s) on the means and way of sharing the burden and then removing
the violation.194 This leads to consensus facit legem—concensus makes the law—
which is rather implicative for the human rights legal certainty on these cases. If
there would be disagreement on sharing this burden—which can also lead to
breaching the principle of sincere cooperation—then the Luxembourg Court may
be given the jurisdiction to attribute the burden to each of them accordingly.195 In
that case, Gragl argues that there would be the need to pass an EU-treaty amend-
ment: validly, one should adhere to Gragl’s point because, I argue, otherwise,
Protocol 8 would be violated.196 Although this proposal may seem fair, one may

responsibility without entering the territory of EU law, it may have been a solution that would have
both respected the EU law autonomy and provided for an individualized responsibility between
EU and its Member States. This model—however—may have only worked if EU and its Member
States would have offered a fully bona fide respect for announcing their individual responsibility:
otherwise, this model would have not worked effectively.
193
On the role of Member States on the constitutional Competence-Competence of the EU
Treaties, see: Dyevre (2013), p. 149.
194
For this reason, Gragl argues that there is the need to establish an internal EU mechanism to
attribute responsibility between the EU and its Member State(s) when they find themselves subject
to a decision on joint-liability of the Strasbourg Court. See: Gragl (2013), p. 168/9. I argue that this
mechanism may be the Luxembourg Court, which needs a certain jurisdiction to be added to it to
be able to rule on such issues. However, legalizing this issue too much may push the Luxembourg
Court on the borders of sovereign rights of Member States to decide themselves through political
means on the fulfillment of obligations which they undertake under international law; Gragl also
argues that attaching this role to the Luxembourg Court may be time-consuming, which would
then undermine the credibility of the procedure. However, Gragl notes that if Luxembourg Court
was given this authority then it would play an important role to better delineating the borders of
responsibility between the EU and its Member States in the context of EU law obligations. In such
circumstances, Gragl notes that for legislative acts, it is the EU legislature (institutions) which will
be held liable. However, Gragl rightly points that this would be a bit more difficult when it comes
to attributing the burden for directives, in which case the Luxembourg Court may either choose EU
or its Member States: depending on the margin of appreciation which the latter had enjoyed in
implementing a certain legal obligation. See: Gragl (2013), p. 169/70.
195
Lock (2010), p. 787; Contra: Gragl, in addition to the Luxembourg Court, proposes that such a
mechanism may be in the form of a committee composed of EU and member states’ representa-
tives. See: Gragl (2013), p. 169; On the model of internal apportion of the monetary burden in
cases of joint responsibility, Cf.: Gragl (2013), p. 169, who requests a mechanism to play this role.
As regards the share of burden, Gragl proposes that if the Member State(s) acted without state
discretion, it is the EU which should cover the costs of compensation. If the Member State(s) acted
with a certain state discretion, and they did not request the preliminary reference of the EU Court
under Art. 267 (3) TFEU, then it is the member states which in principle should incur such costs
(Gragl 2013, p. 169).
196
Gragl (2013), p. 169. With regard for a treaty amendment, I argue that such type of jurisdic-
tion—if added through an internal EU regulation—would prejudice the competences of the EU
Court, therefore violate the prohibition which Protocol 8 which requires that accession not affect
the treaty-competences of the EU institutions.
224 6 Attribution of Liability Under the Co-respondent Mechanism

not see it as very practical as the Luxembourg Court may not jurisdictionally
provide for that ‘service’ in a safe and sound way as the respondent and
co-respondent should have already provided through a friendly settlement and the
unilateral declaration their agreement on this share of burden, otherwise the
Strasbourg’s decision would have been blocked. Involving the Luxembourg Court
in this intermediate stage (before Strasbourg gives its decision) seems rather
unrealistic and not feasible at least for the time the latter takes to clear the issue.
Moreover, the unilateral declaration and the friendly settlement declaration may not
fulfill the criteria of acts that may become attacked before the Luxembourg Court.
Rules on sincere cooperation—on the other hand—seem rather general and vague
to offer safe departure from such situations.197 Opinion 2/13 (para. 264) supports
the latter line of argument, although with a very general language (see supra the
subchapter on Opinion 2/13). That said, the blurred nature of this mechanism—as
noted above—seems to contradict the natural mission which the Convention system
has been authorized to follow. An example from the Inter-American Court of
Human Rights, in Hilaire v. Trinidad and Tobago, may well serve this argument.
The latter Court has ruled amongst others that: ‘It would be unacceptable to
subordinate the said mechanism to restrictions that would render the system for
the protection of human rights established in the Convention and, as a result, the
Court’s jurisdictional role, inoperative . . .’198 It is not the intention to argue here
that the problems explained above make the Convention system inoperative, how-
ever, they seem to erode its effectiveness which would quite well fall within the
borders of Hilaire v. Trinidad and Tobago.
From another perspective of discussion, it is important to bring the presumption
of Jacque, who questions whether the current model of recognition of the margin of
appreciation which the Strasbourg Court recognizes to certain states may make the
issue further blurred if applied against the joint responsibility system which the
co-respondent mechanism installs.199 The author agrees with Jacque that recogniz-
ing the margin of appreciation200 in a system of joint responsibility seems rather
odd and legally uncertain. Two fundamental argumentative scenarios support this.

197
On the latter argument, see e.g.: Chatháin (1999), p. 466, referring to the uncertainty over the
use of this rule on the participation of EU and its Member States to the WTO mechanisms; See
also: Heliskoski (1996), p. 116, referring to the Commission’s proposal to adopt a code that would
regulate the means of participation of EU and its Member States in the WTO mechanisms; The
same seems requested by the Luxembourg Court in Opinion 2/13 as regards the mode of sharing
the burden between EU and its Member States for ECHR violations.
198
Inter-American Court of Human Rights, Hilaire v. Trinidad and Tobago, Preliminary Objec-
tions, judgment of 1 September 2001, Series C, No. 80, para. 82.
199
Jacque (2011), p. 1016.
200
To note, the margin of appreciation is an instrument of the Court which is not applied uniformly
on all parties: the Court chooses to decide on the degree to which margin of appreciation is
recognized to a certain state on a certain segment of the Convention on basis of its own
appreciation of national legal and factual factors. On the divergent degrees of the margin of
appreciation recognized to states within the context of the Court’s approaches, see e.g.: Helfer
(1993), pp. 136 et seq.
6.9 Ambiguities in the Co-respondent Mechanism: Why Is It So Blurred? 225

(A) Let us suppose that on a certain right—e.g. the right to property—the Stras-
bourg Court recognizes a wide margin of appreciation to the EU but a rather narrow
margin of appreciation to Poland.201 If Poland and the EU were co-respondent(s) in
a case before the Strasbourg Court, the Strasbourg Court would need to find both of
them jointly liable: however, it may pursue three options as regards its position on
the margin of appreciation towards them: (a) abandon its recognition of the margin
of appreciation to both of them, (b) equalize the level of margin of appreciation to
both of the parties, (c) or, locate the responsibility on the basis of their level of
margin of appreciation, a case which needs to pull the Strasbourg Court into the EU
law. Any of these three options would seem to harm the legal certainty of the
Convention system in applying the margin of appreciation tool in a system of joint
responsibility. (B) Let us suppose that the Strasbourg Court recognizes a certain
level of margin of appreciation to the United Kingdom and does not recognize any
margin of appreciation to Romania on the right to e.g. vote. If Member States stand
as one side of the respondents, the decision of the Strasbourg Court would in
principle examine the Convention rights against one or more of the Member States
which were respondents therein (attacked by the application of the claimant). That
decision takes into account only the attacked respondent situation with regard to the
use of the margin of appreciation. However, effectively, the decision will bind all
EU Member States as it is only them enjoying the power to amend the EU Treaties.
The co-respondent mechanism in this regard would produce a situation wherein the
use of margin of appreciation is projected by the Strasbourg Court only against the
respondent that is part of the case—e.g. United Kingdom—but it will effectively
produce the same outcome also for e.g. Poland (which was not a co-respondent to

201
One may counter-argue this by maintaining that the margin of appreciation does not vary
according to the identity of the state or contracting party, as the Strasbourg Court applies a one-fits-
all model of margin of appreciation on individual rights. I tend to disagree with this from a
particularistic perspective, since the Strasbourg Court often devises the extent of the margin of
appreciation on basis of the context in which the human rights operate in a certain constitutional
system. One example may be the right to property: in eastern (former communist) countries, the
Court recognizes a rather broad margin of appreciation which it does not recognize on western
European countries (due to the pressing needs which the Eastern European countries have in this
regard, that makes their context difference from those of the Western Europe). One needs to
recognize that the ‘margin allows a country a degree of defence at the European Court where
judges are obliged to take into account the cultural, historic and philosophical differences between
Strasbourg and the nation in question since what is right for Spain may not be right for the UK.’
(See: ‘Human Rights Act: How it works.’ BBC, 29 September 2000. Available at: http://news.bbc.
co.uk/2/hi/uk_news/946390.stm); To support this, Lavender further demonstrates that the use of
margin of appreciation depends from the context of the case and legal system where it applies. See:
Lavender (1997), p. 382. Another example may be the difference between the EU under the
Bosphorus margin of appreciation and state contracting parties which have no systemic presump-
tion like Bosphorus. Another example may the right limitations to the right to vote and prohibition
of discrimination, which the Strasbourg Court has recognized to Bosnia until the Sejdic case, a sort
of margin of appreciation which the Court did not recognize to any other contracting party
(it removed it even from Bosnia when it recently adopted Sejdic case, and then Zornic case).
Therefore, the note made above remains valid in the form in which it was given.
226 6 Attribution of Liability Under the Co-respondent Mechanism

the case), although the Court would not allow any margin of appreciation to Poland.
The co-respondent mechanism certainly makes the use of the margin of apprecia-
tion rather vague and essentially based on an unequal format—which also derives
from the nature of the EU law and its relationship with the Member States’ legal
orders.
To finalize this discussion—Sarvarian argues that the idea of EU and Member
States dividing their internal competences externally for the purpose of attribution
of liability—as for example in the Convention system—will be even further
complicated with the co-respondent mechanism of the DAA.202 In principle, the
author of this book disagrees with that argument. It is true that the Convention
system may put its contracting parties to a certain position where it needs to tackle
their division of competences of joint conduct from the perspective of their external
competences. However, the co-respondent mechanism basically tries to prevent
such situations from appearing in practice. The mutual responsibility model which
the co-respondent mechanism involves produces exactly the contrary to the argu-
ment of Sarvarian. The join liability model in fact tries to keep the issue of division
of external competences of EU and its Member States as an issue of internal law, of
no interest for the Convention system and its court’s jurisdiction; for this reason,
Gragl argues that there is the need for the EU to adopt internal rules.203 The correct
argument in this regard is that the co-respondent mechanism makes the division of
external competence irrelevant for the Strasbourg Court’s function to attribute
liability when both the EU and its Member States appear as joint (co-)respondents.
This being the case, in contrast to Sarvarian, the author of this book argues that the
co-respondent mechanism intentionally tries to keep the issue of division of com-
petences between the EU and its Member States as an issue of EU internal law, in
order to preserve the autonomy of the latter and to release the Court from additional
burden on attributing liability between the normative violator and the implementer
of the violation. Whether this mechanism makes it more efficient for the Strasbourg
Court to protect human rights is not necessarily the case.

6.9.4 Ambiguity 4: Lack of the Right Addressee—No


Answer—Political Consensus

The last issue that remains unsolved is that of the scenario wherein one, or the
correct, addressee has not joined as co-respondent, and the Court would need to
attach the violation to the respondent. It is submitted that in such scenario the entire
efficiency of the co-respondent mechanism may become at risk while the efficiency
of the Convention system would lose its practicability. Furthermore, there would
seem no EU internal Treaty mechanism to solve that problem, while EU would

202
Sarvarian (2014), p. 102.
203
Gragl (2013), p. 168/9.
6.10 Referral to the Grand Chamber: Is There Space for Divorce Between (Co). . . 227

become internationally liable for a violation which practically is not hers. The
author would like to bring the argument of the UN Committee on Economic, Social
and Cultural Rights which—in relation to the issue of implementation of interna-
tional obligations of treaty contracting parties—has argued: ‘States have a joint and
individual responsibility, in accordance with the Charter of the United Nations, to
co-operate in’ implementing their treaty obligations.204 It is in the nature of the
Convention system—as a European-wide instrument of international law—to
become effective only in light of bona fide cooperation between its contracting
parties. Under this standard—one should argue—neither the EU nor its Member
States may refuse to join as co-respondents, if the Convention’s system of attribu-
tion of liability is at risk. Their responsibility for not having joined is not only
individual—but also collective, as the Convention—as ruled in Soering v. United
Kingdom—is a system of ‘special character as a treaty for the collective enforce-
ment of human rights’.205 The standard ruled by the ICJ in this regard—in a similar
case relating to the reservations on a certain convention against genocide—asserts
that ‘the contracting States do not have any [. . .] interests of their own, but merely a
common interest[. . .]’ to reach the objectives for which the Convention exists.206
Refusing to join as a co-respondent—applying this logic to the EU Member
States—would simply ruin the common interest for which the DAA and the
Convention exist. That said, if a co-respondent would refuse to become a party to
the case in which it has certain liability for the alleged violation—it would of course
make both the EU and the Member States as a whole fail to respect their duty to
cooperate, thus consuming both individual and collective responsibility that exists
under international law—in addition to their EU Treaties. Ironically, this issue
remains fully unobserved by the Luxembourg Court’s Opinion 2/13, although its
relevance for the EU Treaties effectiveness is crucial.

6.10 Referral to the Grand Chamber: Is There Space


for Divorce Between (Co)-respondents?

It is important to make note to the fact that the Convention foresees the possibility
to have certain individual cases appealed to the Grand Chamber, which is the last
instance of the Strasbourg Court. For this reason, Art. 43 (1) ECHR establishes that:
‘Within a period of three months from the date of the judgment of the Chamber, any

204
Committee on Economic, Social and Cultural Rights, [The Right to Adequate Food (Art. 11 of
the International Covenant on Economic, Social and Cultural Rights)], General Comment
No. 12 (1999), E/C.12/1999/5, para. 38 [emphasis added].
205
ECtHR, Soering v. United Kingdom, ECtHR, Series A No. 161 (1989), para. 34; See also,
EctHR, Ireland v. United Kingdom judgment of 18 January 1978, Series A No. 25, p. 90 (para.
239).
206
International Court of Justice, Reservations to the Convention on the Prevention and Punish-
ment of the Crime of Genocide, ICJ Reports, 1951, p. 23.
228 6 Attribution of Liability Under the Co-respondent Mechanism

party to the case may, in exceptional cases, request that the case be referred to the
Grand Chamber.’ If the Grand Chamber accepts the request, it issues a judgment
with the effect of a final legal ruling within the Strasbourg Court. The real question
therefore is how will the respondent and co-respondent(s) appear before the Grand
Chamber if their case is referred there. The Explanatory Report to DAA makes it
clear that the possibility exists for EU-related cases to reach the Grand Chamber as
well. It therefore gives this legal explanation:
Any Party may request the referral of a case to the Grand Chamber under Article 43 of the
Convention; the respondent or co-respondent could therefore make such a request without
the agreement of the other. Internal EU rules may, however, set out the conditions for such a
request. Should a request be accepted, the Grand Chamber would re-examine the case as a
whole, in respect of all alleged violations considered by the Chamber and with regard to all
Parties.207

It is obvious that the possibility for either the respondent or the co-respondent(s)
to file an appeal to the Grand Chamber exists, however, they are divorced in this
procedure. The above quoted explanation clearly establishes that the respondent
and the co-respondent(s) need no agreement to move forward with the appeal,
although there may be set forth EU internal rules to concretize this. It is not logical
for one party to retain the right to appeal alone a jointly received and attributed
judgment of the Chamber, as that would make one party enjoy the right to appeal
something which may interfere to the right not to appeal of the other party with
which the first judgment is received jointly. The central issue here is that although
the respondent and co-respondent(s) have been mutually and jointly held responsi-
ble for a certain violation by a Chamber, they may make their own ways to appeal
the decision to the Grand Chamber. The question, nonetheless, is whether this
‘divorce’ between the respondent and co-respondent(s) may result in space for
the EU law autonomy to be interfered by the Grand Chamber. The answer is
positive, of course. The Grand Chamber is expected to do no less review of the
alleged violation and its cause than the Chamber, with such review having the
possibility to revisit the entire case adjudicated in the first-instance Chamber.
However, in doing so, the Grand Chamber will have the opportunity to adjudicate
the case as a whole, meaning that nothing less than what has been presented during
the first-instance adjudication may come out revisited.
The first argument in this regard is that the Explanatory Report wrongly con-
siders that there is no need for the respondent and co-respondent to appear as one
before the Grand Chamber. In fact, allowing them to have their own positions of
arguments and facts, gives the Grand Chamber a chance to revisit segments of EU
law in order to find whether the respondent’s or co-respondent’s claims are correct.
Without them appearing as one party, there exists no means for the Grand Chamber
to be prohibited from dealing with their division of responsibility or their respon-
sibility according to their laws, which might lead the Grand Chamber to EU internal
rules. One must argue that the mere fact that only the respondent—but not the

207
Explanatory Report to the DAA, para. 63.
6.10 Referral to the Grand Chamber: Is There Space for Divorce Between (Co). . . 229

co-respondent(s)—or vice versa, might appeal the case to the Grand Chamber,
namely without any ties/duty to do that jointly, may put in danger the entire
co-respondent mechanism and its logic. Suppose that X is a respondent and Y a
co-respondent: both of them have been found liable for the Convention’s breach.
Only Y files an appeal to the Grand Chamber, whereas X does not. This mere fact
would mean that Y will use arguments which might put it irresponsible or less
responsible for the appealed breach, a space which the Grand Chamber would need
to visit in order to understand whether ‘there is less or far less’ responsibility for Y
in that case. Such visit of the Grand Chamber might need to capture the EU law
obligations on basis of which X and Y have been found responsible by the decision
of the Chamber, clearly allowing space for interference to the EU law autonomy.
Besides, Y may become released from a responsibility which in fact belonged to X,
whereas the latter would be released at all from that responsibility. This would bring
then the entire application at the risk of failing to be addressed properly by the
Strasbourg Court.
The second argument is that Y did not take the consent of X for two core reasons:
either X was satisfied with the decision of the Chamber, or X has escaped from
some obligations thus in turn putting more obligations to Y. Faced with this
situation, Y would therefore need to remedy its ‘extra’ liability, something for
which it needs to present arguments to the Grand Chamber which put the latter in a
position to revisit the internal linkage between X and Y on their joint liability.
Third, Y might consider—against or without the consent of X—that the entire
responsibility attributed to them has been wrongly or manifestly illegally applied.
In either scenario, Y would need to defend X as well before the Strasbourg Court, as
the only way to prove their joint lack of responsibility for the wrongful act which
they have been called responsible in the first-instance Chamber. In doing so, Y
might use arguments which originate in X and which only X could have submitted
legitimately to the Grand Chamber. In that scenario, the Grand Chamber may need
again to revisit some of the causes of responsibility in the law of X in order to
decide whether the claim of Y for defending their joint position is right or not. That
could again produce a practice that interferes to the EU law autonomy.
It is therefore argued that the possibility for the respondent and co-respondent to
get divorced in the stage of Grand Chamber mislays the rationality of the
co-respondent mechanism if it is only applied in its original form in the proceedings
of first instance. This also unchains the possibility for the Grand Chamber to get
involved in the ‘interpretation’ of EU law as a means to decide whether the divorced
and now separate respondent and co-respondent(s) relationship as to the responsi-
bility for the breach of Convention decided by a Chamber. Of course, the EU Court
might interpret such divorce as against the principle of sincere cooperation
established by the Treaties. Still, the mere fact that the respondent and
co-respondent(s) are given the chance to defend their own positions and not their
position before the Grand Chamber, allows for each of them to present separate and
also diverging arguments which altogether provide space for the Grand Chamber to
visit their internal law as a means to rule on the appeal. Paradoxically, the
230 6 Attribution of Liability Under the Co-respondent Mechanism

Luxembourg Court did not observe this huge loophole when it examined the
compatibility of the DAA with the EU Treaties in its recent Opinion 2/13.

6.11 A Comparison on the Model of Sharing the Liability


Between DAA, UNCLOS and UNCILDCSO: Which
Are the Strengths and Weaknesses of DAA?

Besides the Accession Agreement of the EU to ECHR, there exist two other
international agreements which have established systems of mutual responsibility
for state parties and international organizations which have together caused a
certain violation of the respective treaty. These two international instruments are:
United Nations Convention on International Liability for Damage Caused by Space
Objects (henceforth: UNCILDCSO) and United Nations Convention on the Law of
the Sea (henceforth: UNCLOS). Since the accession of the EU to ECHR will be
accommodated amongst others with the innovated co-respondent mechanism, the
question remains how these two comparable instruments convert the relationship
between the liable parties if there is no exact means of holding parties indepen-
dently accountable.
UNCLOS provides a rather broad-scope model of sharing the liability between
two or more parties. In this context, Art. 139 UNCLOS reads:
2. Without prejudice to the rules of international law[. . .], damage caused by the failure of a
State Party or international organization to carry out its responsibilities under this Part
shall entail liability; States Parties or international organizations acting together shall
bear joint and several liability. A State Party shall not however be liable for damage
caused by any failure to comply with this Part by a person whom it has sponsored under
article 153, paragraph 2(b), if the State Party has taken all necessary and appropriate
measures to secure effective compliance under article 153, paragraph 4, and Annex III,
article 4, paragraph 4.
3. States Parties that are members of international organizations shall take appropriate
measures to ensure the implementation of this article with respect to such
organizations.208

As seen from Art. 139 UNCLOS, one can observe the notion of ‘without
prejudice to international law’ which seems to suggest that UNCLOS’ model of
liability distribution may not exclude high-rank instruments of international law
which tackle the issue of liability sharing between states and/or international
organizations. This seems to suggest that UNCLOS does merely build an auxiliary
rule on the distribution of liability between states and/or international organiza-
tions, adding that it is explicitly accepted that if such international law would exist
and seem applicable it would be superior to Art. 139 UNCLOS. This is clearly not
the case in the DAA of EU to ECHR, as the latter does not establish but also clearly

208
Emphasis added.
6.11 A Comparison on the Model of Sharing the Liability. . . 231

excludes any norm of international law to serve as a legal basis for sharing the
liability between the EU and its Member States as regards the violations on ECHR.
As opposed to UNCLOS, DAA seems to be closed for any norm of international
law to serve as legal basis for the distribution of liability in case the Strasbourg
Court finds both the EU and Member State(s) jointly responsible for the violation.
Such closure to international law makes DAA more autonomous and protects better
the EU legal order against potential interferences to its external autonomy that
might be based in other rules of international law.
UNCLOS as a result establishes a different model of attributing liability in case
there appears a violation done by more than one state or international organization.
As opposed to DAA which allows no room but for simply a joint responsibility
between the EU and its Member State(s), UNCLOS prescribes that if there are more
than one state or international organization responsible for a certain failure to
respect UNCLOS they will have joint and several liability.209 Under UNCLOS,
that model of liability has a cumulative nature, meaning that the states and/or
international organizations are liable once jointly and then separately for the
same violation. Clearly, UNCLOS applies a double responsibility ensuring that if
parties refrain from holding the joint responsibility then they will still remain
individually responsible for their part of liability towards UNCLOS. As opposed
to the latter, DAA makes no room for several responsibility and foresees that only
the co-respondents may decide themselves if they wish to exclude one of them from
such responsibility. In addition, it is only the co-respondents themselves who can
decide how to share their responsibility, as the DAA does not allow for individual
and separated responsibility if the co-respondent mechanism is activated and the
parties have not utilized Art. 3 (7) DAA.
Another point under Art. 139 UNCLOS is the issue of state liability if the state
has sponsored someone to act as its agent in exploring and exploiting the Area. It
seems rather clear that if the state party has taken all necessary and appropriate
measures to ensure the application of UNCLOS, then a potential failure to meet
those obligations by the state agent would not bring any form of liability to the
sponsoring state. This is rather dissimilar to DAA once EU accedes to ECHR. If
Member States are taken as agents of EU law when implementing the latter through
their national measures210 (e.g. in the case of Directives), the question would be
how would the co-respondent mechanism distribute the liability between the EU
and Member States in such instances. Since the EU would join as a co-respondent in
a case wherein the implementing measure of the EU law undertaken by the Member
State is challenged before the Strasbourg Court, then the latter would find both the

209
See also: Advisory Opinion of the Seabed Disputes Chamber of the International Tribunal for
the Law Of the Sea on Responsibilities and Obligations of States sponsoring persons and entities
with respect to activities in the Area, para. 201 et seq. Available at: http://www.itlos.org/fileadmin/
itlos/documents/cases/case_no_17/adv_op_010211.pdf.
210
E.g. On member states as agents of EU for implementing EU law, see: Franklin (2010–2011),
pp. 153 et seq; An exemplary case may be: Court of Justice of EU, Promusicae, Case C-275/06,
[2008] ECR I-271, para 68.
232 6 Attribution of Liability Under the Co-respondent Mechanism

EU and Member State(s) in principle responsible even for a violation of a Member


State’s implementing measure in which case the EU would have done the most to
ensure through its general act full compliance with the Convention rights. There-
fore, DAA seems more distributive in terms of liability compared to UNCLOS,
which tends to make precluded from liability a state party if it has done everything
possible to comply with UNCLOS.
Finally, UNCLOS requires the state parties which are members or international
organization to ensure that they distribute the provision of liability also in respect to
such organizations. In this vein, UNCLOS tries to ensure that liability is distributed
to other international organizations by state parties if they transfer to them certain
competences which might involve the application of obligations from UNCLOS.
DAA by contrast does not allow any form of transfer of human rights liability of the
EU in face of the Convention to any other international organization to which the
EU could accede, requiring that the EU alone be held responsible for its own law
indispensable where the EU could accede in the future (the Bosphorus principle).
As regards UNCILDCSO, one can observe rather similar liability distribution
model with UNCLOS. That being the point of examination, Art. 4 of UNCILDCSO
reads:
1. In the event of damage being caused elsewhere than on the surface of the earth to a space
object of one launching State or to persons or property on board such a space object by a
space object of another launching State, and of damage thereby being caused to a third
State or to its natural or juridical persons, the first two States shall be jointly and
severally liable to the third State, to the extent indicated by the following:
(a) If the damage has been caused to the third State on the surface of the earth or to aircraft
in flight, their liability to the third State shall be absolute; (b) If the damage has been
caused to a space object of the third State or to persons or property on board that space
object elsewhere than on the surface of the earth, their liability to the third State shall be
based on the fault of either of the first two States or on the fault of persons for whom
either is responsible.
2. In all cases of joint and several liability referred to in paragraph 1 of this article, the
burden of compensation for the damage shall be apportioned between the first two
States in accordance with the extent to which they were at fault; if the extent of the fault
of each of these States cannot be established, the burden of compensation shall be
apportioned equally between them. Such apportionment shall be without prejudice to
the right of the third State to seek the entire compensation due under this Convention
from any or all of the launching States which are jointly and severally liable.

As seen in Art. 4 (1), UNCILDCSO foresees the joint and several responsibility
for the violating parties. The same arguments from UNCLOS as regards the joint
and several liabilities apply here as well, of course keeping the note that this model
of liability distribution seems different from what DAA establishes in relation to the
EU and Member States violations of ECHR rights. However, such joint and several
liability of UNCILDCSO may be of two forms: first, as an absolute joint and several
liability towards the third state whom a UNCLOS right was violated, and, second,
an individual liability based on the fault respectively the state party which caused
the violation. DAA applies the first model of UNCILDCSO liability, that of
absolute liability if a Convention right is violated. However, DAA does not foresee
and allow for any form of liability based on that who undertook the fault, as such
6.11 A Comparison on the Model of Sharing the Liability. . . 233

individual liability if applied upon DAA would leave room to the Strasbourg Court
to enter the internal borders that define the relationship EU and Member States’ law,
which would amount to a violation of the EU law autonomy. Besides, in the DAA
case, if an implementing measure of the Member State211 deriving from EU
secondary law (e.g. regulations) wherein the Member State has no discretion
from applying it or not becomes attacked before the Strasbourg Court, the latter
may not deal with what UNCILDCSO allows: the identification of the party that has
the fault for the violation. Were DAA to allow for such possibility, then, in cases
wherein Member States undertake implementing measures wherein they have no
state discretion—the duty to undertake such implementing measures stemming
from an EU law provision—then the Strasbourg Court would have had to find
merely the EU responsible for such violation, and not the Member States as well. As
opposed to UNCILDCSO, therefore, DAA applies a model of identifying the liable
party by calling liable both the party which effected the violation (the Member
States), namely the party which produced the result of the violation, and the party
which had legally made that measure obligatory (the EU), namely the legal origin
of the violation.
On the other hand, once two or more parties have been ruled as having joint and
several liability under UNCILDCSO, then the burden of compensation for the
violation caused needs to be apportioned proportionally between the violating
parties. If such proportional apportionment cannot be established, meaning by an
agreement between the liable parties, then UNCILDCSO establishes that such
apportionment of burden will be established equally between the parties. This
shows that UNCILDCSO offers space for the liable parties, if they are jointly and
severally liable, to agree on sharing the burden of liability proportionally. If that is
not agreed between the parties, then UNCILDCSO recognizes parties as equally
responsible for the purposes of burden sharing. The first option seems similar with
the friendly settlement provisions of DAA which requires that the EU and Member
State(s) internally agree for the share of burden as co-respondents, and then submit
their agreement to the Strasbourg Court. In such a case, the Strasbourg Court seems
to have been given no space to intervene between the co-respondents when reaching
the friendly settlement agreement on the share of burden of liability, and DAA
allows no space for the Strasbourg Court to share the burden between the EU and
Member State(s) equally if they have made no agreement for the friendly settlement
(as opposed to UNCILDCSO). The lack of possibility of the Strasbourg Court to
apply the UNCILDCSO principle of equal-burden-sharing principle if parties have
not been agreed on the proportional share, makes DAA lack a very important legal
certainty instrument which stands as highly important for the effectiveness of
human rights protection by the Strasbourg Court.
Finally, UNCILDCSO establishes that whether the violating parties choose to
use the proportional burden sharing through an agreement or they become other-
wise equal in the share of burden, that internal issue of the mechanism for sharing

211
E.g. Hancox (2013), p. 1426, on how far the notion of implementing measure may go.
234 6 Attribution of Liability Under the Co-respondent Mechanism

the burden between the parties may not prejudice the right of the victim party from
seeking its compensation. This standard seems rather in the same lines with DAA
and the Convention, as the latter allow for no room when it comes to the liable
parties. Although the EU and Member State(s) might have huge problems in
relation to apportioning their burden for the joint liability, that may not prejudice
or put into risk the efficiency of the victim to have its rights restored and compen-
sated. Therefore, this UNCILDCSO principle likely seems well installed also in the
Convention and DAA system, adding that there is almost no doubt that DAA
establishes that even the decision of the Strasbourg Court may not be prejudiced
by the fact if there is or not an internal agreement on friendly settlement between
the liable parties of the co-respondent mechanism. In the same line as with
UNCILDCSO, DAA and the ECHR built the adjudication procedure in a way in
which the victim party may not become a ‘second’ victim of the Court’s impossi-
bility to decide on the violation on the basis of the fact that the EU and Member
State(s) might disagree on their burden sharing resulting from the use of the
co-respondent mechanism.

6.12 Chapter’s Summary of Conclusions

The chapter examined the design, legal effect and reach of the co-respondent
mechanism in the context of attribution of Convention responsibility for alleged
violations involving simultaneously the EU and its Member States. The chapter
concludes that the distribution of liability for the EU and its Member States under
the co-respondent mechanism follows a complex and sophisticated methodology,
with certain remaining loopholes which may permit the Strasbourg Court to impede
on EU law autonomy.
The chapter first examined the treaty-based instructions and limitations on the
scope of accession and the potential effect of the co-respondent mechanism. The
chapter concludes that—as Protocol 8 requests that the mechanism address the
liability to the correct addressees—the co-respondent mechanism carries out pre-
cisely the contrary of Protocol 8. The latter conclusion refers to the fact that the
co-respondent mechanism produces a sort of joint liability model, wherein a certain
party may become liable before the Convention system even though it had no direct
responsibility for legislating the alleged violation; this latter scenario is exactly the
contrary of what Protocol 8 requests, namely to address liability only to the correct
addresses and not to any other party. Considering the nature of effect of Protocol
8, the chapter concludes that the EU Treaties’ limitations on the accession process
may be of effect merely to the negotiating capacity of the EU, but they do not
directly shape or influence the interpretation of the DAA per se.
The chapter then carefully analyzes the normative construction of the
co-respondent mechanism, considering criteria which need be satisfied for it to be
applied to the EU, its Member States, or both of them. The chapter concludes that
6.12 Chapter’s Summary of Conclusions 235

the normative construction of the co-respondent mechanism has too many loop-
holes in its structure, therefore allowing too many gaps that may permit the entrance
of the Strasbourg Court into the EU law autonomy—a point in which this book
disagrees with Gragl.212 The normative construction of the co-respondent mecha-
nism, however, operates in a highly advanced normative prescription, therefore its
use will need complex and advanced methodology from all parties to make its use
certain and efficient. To this degree, the chapter concludes that it is especially tricky
to construe the internal rules of the EU law that would prescribe the use of it from
the EU and its Member States, as such rules may not interfere with the efficiency
and legal mechanics of the DAA.
The chapter therefore concludes that the co-respondent mechanism proves to be
a highly specialized and distinguished method of attributing liability, something
that goes far beyond the principles of DARIO. This being the issue, the chapter
stands in the same line of argument that the co-respondent mechanism has a lex
specialis nature in face of DARIO. That being its principal setting, it is concluded
that the co-respondent mechanism is a highly innovative and complex mechanism
in international law, its intended effect being a distinguished devise to show how
self-contained regimes in international law may regulate their competing jurisdic-
tions without harming their primary organizational values and goals. Then, the
chapter also presented an analysis to the issue of appeals that may be filed before the
Grand Chamber of the Strasbourg Court. It is here concluded that—in those cases—
the co-respondent mechanism seems with little relevance, and, therefore, there is
high potential for interference to the EU law autonomy once cases become filed to
the Grand Chamber. Finally, the chapter—having shown how the co-respondent
works technically and identifying its loopholes—embarked on comparing the issue
of attribution of liability between DAA, UNCLOS and UNCILDCSO. The chapter
concludes that DAA follows a rather different model of attributing responsibility
between EU and Member States, something that is apparently dedicated to the
protection of EU law autonomy rather than the efficiency of the task on attributing
of responsibility.
In this regard, the chapter concludes that the co-respondent mechanism will be a
highly valuable mechanism that will make the interaction between the Strasbourg
and Luxembourg regimes of law more feasible than previously. This interaction
will be less combative and less harmful, therefore proving that the attribution of
liability upon the EU and its Member States becomes embedded in a comfortable
legal environment.

212
Gragl (2013), p. 167. Gragl basically argues that the co-respondent mechanism takes into
account the EU specificities, and relies on the good will of the Strasbourg Court on EU law
autonomy: something Gragl says has been demonstrated also with the previous case-law on EU of
the Strasbourg Court (Gragl 2013, p. 167). One additional argument (in addition to those
mentioned above in the body of the text) which supports this book’s line of reasoning—and also
disagrees with Gragl—is Luxembourg Court’s Opinion 2/13, which tries, and sometimes also
finds, some of these loopholes which the Court says may seriously undermine EU law autonomy
(see especially para. 222 et seq of Opinion 2/13).
236 6 Attribution of Liability Under the Co-respondent Mechanism

References

Alegre S, Leaf M (2004) Mutual recognition in European judicial cooperation: A step too far too
soon? Case study—the European arrest warrant. Eur Law J 10(2):200–217
Alemanno A, Stefan O (2014) Openness at the Court of Justice of the European Union: Toppling a
taboo. Common Market Law Rev 51(1):97–140
Alter KJ (1998) Who are the “Masters of the Treaty”? European Governments and the European
Court of Justice. Int Organ 52(1):121–147
Alter KJ, Helfer LR (2010) Nature or nurture? Judicial lawmaking in the European Court of Justice
and the Andean Tribunal of Justice. Int Organ 64(4):563–592
Arnull A (1995) Private applicants and the action for Annulment under Art. 173 of the EC Treaty.
Common Market Law Rev 32(1):7–49
Balfour ADJ (2007) Eliminating conflicting interpretations of the European Convention on Human
Rights by the European Court of Justice and the European Court of Human Rights: the Pdiq
system as a preventative solution. Intercult Hum Rights Law Rev 2:183–247
Balthasar S (2010) Locus Standi rules for challenges to regulatory acts by private applicants: the
new article 263(4) TFEU. Eur Law Rev 35(4):542–550
Barents R (2004) The autonomy of community law. Kluwer Law International, The Hague
Barents R (2009) The precedence of EU law from the perspective of constitutional pluralism. Eur
Const Law Rev 5(3):421–446
Bjoerklund M (2001) Responsibility in the EC for Mixed Agreements—should Non-member
Parties Care? Nordic J Int Law 70(3):373–402
Carrubba CJ, Murrah L (2005) Legal integration and use of the preliminary ruling process in the
European Union. Int Organ 59(2):399–418
Carrubba CJ et al (2008) Judicial behavior under political constraints: evidence from the European
Court of Justice. Am Polit Sci Rev 102(4):435–452
Casteleiro AD (2014) United we stand: the EU and its Member States in the Strasbourg Court. In:
Kosta V et al (eds) The EU accession to the ECHR. Hart Publishing, Oxford
Chatháin CN (1999) The European Community and the Member States in the dispute settlement
understanding of the WTO: united or divided? Eur Law J 5(4):461–478
Chinkin CM (1986) Third-party intervention before the International Court of Justice. Am J Int
Law 80:495–531
Conforti B (2010) Comments on the accession of the European Union to the ECHR. Italian Yearb
Int Law 20:83–86
Craig P (2013) EU accession to the ECHR: competence, procedure and substance. Fordham Int
Law J 36(5):1114–1150
d’Aspremont J (2007) Abuse of the legal personality of international organisations and the
responsibility of Member States. Int Organ Law Rev 4(1):91–119
d’Aspremont J (2014) A European law of international responsibility? The articles on the
responsibility of international organizations and the EU. In: Kosta V et al (eds) The EU
accession to the ECHR. Hart Publishing, Oxford
de Búrca G (2005) Limiting EU powers. Eur Const Law Rev 1(1):92–98
De Hert P, Korenica F (2012) The Doctrine of Equivalent Protection: its life and legitimacy before
and after the European Union’s accession to the European Convention on Human Rights.
German Law J 13(7):874–895
den Heijer M (2013) Shared responsibility before the European Court of Human Rights. Nether-
lands Int Law Rev 60(3):411–440
Dyevre A (2013) European integration and national courts: defending sovereignty under institu-
tional constraints? Eur Const Law Rev 9(1):139–168
Eckes C (2013) EU accession to the ECHR: between autonomy and adaptation. Mod Law Rev 76
(2):254–285
Editorial Comments (2009) Karlsruhe has spoken: “Yes” to the Lisbon Treaty, but. . .. Common
Market Law Rev 46(4):1023–1033
References 237

Eeckhout P (2004) External relations of the European Union: legal and constitutional foundations.
Oxford University Press, Oxford
Franklin C (2010–2011) The legal status of the EU charter of fundamental rights after the Treaty of
Lisbon. Tilburg Law Rev 15:137–162
Gaja G (2013) The “Co-Respondent Mechanism” according to the Draft Agreement for the
accession of the EU to the ECHR. ESIL Reflect 2(1):1–6
Garcı́a RA (2002) The general provisions of the charter of fundamental rights of the European
Union. Eur Law J 8(4):492–514
Gondek M (2005) Extraterritorial application of the European Convention on Human Rights:
territorial focus in the age of globalization? Netherlands Int Law Rev 52(3):349–387
Gragl P (2011) Accession revisited: will fundamental rights protection trump the European
Union’s legal autonomy? In: Benedek W, Benoı̂t-Rohmer F, Karl W, Nowak M (eds)
European yearbook on human rights, 2011. Neuer Wissenschaftlicher Verlag (NWV), Wien
Gragl P (2013) The accession of the European Union to the European Convention on Human
Rights. Hart Publishing, Oxford
Greer S, Williams A (2009) Human Rights in the Council of Europe and the EU: towards
‘Individual’, ‘Constitutional’ or ‘Institutional’ Justice? Eur Law J 15(4):462–481
Groussot X et al (2011) EU accession to the European Convention on Human Rights: a legal
assessment of the Draft Accession Agreement of 14th October 2011. Fondation Robert
Schuman/European Issues No 218, 7 November 2011
Hancox E (2013) The meaning of “implementing” EU law under Article 51(1) of the Charter:
Åkerberg Fransson. Common Market Law Rev 50(5):1411–1431
Harding C (2000) The identity of European law: mapping out the European legal space. Eur Law J
6(2):128–147
Hart JW (2010) The European human rights system. Law Libr J 102(4):553–559
Helfer LR (1993) Consensus, coherence and the European Convention on Human Rights. Cornell
Int Law J 26:133–165
Heliskoski J (1996) The “Duty of Cooperation” between the European Community and its
Member States within the World Trade Organization. Finnish Yearb Int Law 7:59–133
Higgins R (2003) The ICJ, the ECJ, and the integrity of international law. Int Comp Law Q 52
(1):1–20
Hoffmeister F (2006) International decisions: Bosphorus Hava Yollari Turizm Ve Ticaret Anonim
Sirket v. Ireland. App. No. 45036/98. Am J Int Law 100:442–449
ILC (2006) Fragmentation of international law: difficulties arising from the diversification and
expansion of international law. Report of the Study Group of the International Law Commis-
sion, Fifty-eighth session, Geneva, 1 May–9 June and 3 July–11 August 2006
Jacobs FG (2007) The sovereignty of law: the European way. Cambridge University Press,
Cambridge
Jacque JP (2011) The accession of the European Union to the European Convention on Human
Rights and fundamental freedoms. Common Market Law Rev 48(4):995–1023
Joris T, Vandenberghe J (2008–2009) The Council of Europe and the European Union: natural
partners or uneasy bedfellows? Columbia J Eur Law 15:1–43
Kerson DLA (1961) The European Convention for the protection of human rights and fundamental
freedoms. Calif Law Rev 49(1):172–186
Kokott J, Sobotta C (2012) The Kadi case – constitutional core values and international law –
finding the balance? Eur J Int Law 23(4):1015–1024
Kuijper PJ, Paasivirta E (2004) Further exploring international responsibility: the European
Community and the ILC’s project on responsibility of international organizations. Int Organ
Law Rev 1(1):111–138
Lang JT (1986) The ozone layer convention: a new solution to the question of community
participation in “Mixed” International Agreements. Common Market Law Rev 23(1):157–176
Larsen KM (2008) Attribution of conduct in peace operations: the ‘Ultimate Authority and
Control’ Test’. Eur J Int Law 19(3):509–531
238 6 Attribution of Liability Under the Co-respondent Mechanism

Lavender N (1997) The problem of the margin of appreciation. Eur Hum Rights Law Rev 4
Lavranos N (2006) MOX Plant dispute – court of justice of the European Communities. Eur Const
Law Rev 2(3):456–469
Leach P (2011) Taking a case to the European Court of Human Rights. Oxford University Press,
New York
Lenaerts K (2012) Exploring the limits of the EU charter of fundamental rights. Eur Const Law
Rev 8(3):375–403
Lenaerts K, de Smijter E (2001) The charter and the role of the European Courts. Maastrich J Eur
Comp Law 8(1):90–101
Licková M (2008) European exceptionalism in international law. Eur J Int Law 19(3):463–490
Lock T (2010) EU accession to the ECHR: implications for judicial review in Strasbourg. Eur Law
Rev 35(6):777–798
Lock T (2011) Walking on a tightrope: the Draft ECHR Accession Agreement and the autonomy
of the EU legal order. Common Market Law Rev 48:1025–1054
Lock T (2012) End of an epic? The Draft Agreement on the EU’s accession to the ECHR. Yearb
Eur Law 31(1):162–197
Martin FF (2003) The international human rights aspects of the forum non conveniens doctrine.
Univ Miami Inter-Am law Rev 35:101–123
Martı́n J, De Nanclares P (2013) The accession of the European Union to the ECHR: more than
just a legal issue. Working Papers on European Law and Regional Integration WP IDEIR no
15. Instituto de Derecho Europeo e Integracion Regional (IDEIR), Universidad Complutense
Muir E (2014) The fundamental rights implications of EU legislation: some constitutional
challenges. Common Market Law Rev 51(1):219–246
Nanopoulos E (2013) Trust issues and the European common asylum system: finding the right
balance. Camb Law J 72(2):276–280
O’Meara N (2011) “A More Secure Europe of Rights?” The European Court of Human Rights, the
Court of Justice of the European Union and EU accession to the ECHR. German Law J 12
(10):1813–1832
Odermatt J (2014) The EU’s accession to the European Convention on Human Rights: an
international law perspective. Leuven Center for Global Governance Studies. Working Paper
No. 136
Olsen JB (2009–2010) Protecting fundamental rights and the evolving roles of the Court of Justice
of the European Union and the European Court of Human Rights; Europeanisation in action at
the supranational level. Edinburgh Stud Law Rev 1(3):54–73
Orakhelashvili A (2003) Restrictive interpretation of Human Rights Treaties in the recent juris-
prudence of the European Court of Human Rights. Eur J Int Law 14(3):529–568
Phelps J (2006) Comment: reflections on Bosphorus and human rights in Europe. Tulane Law Rev
81:251–279
Quirico O (2010) Substantive and procedural issues raised by the accession of the EU to the
ECHR. Italian Yearb Int Law 20:31–53
Raba K (2013) The accession of the European Union to the European Convention for the
Protection of Human Rights and Fundamental Freedoms—overview of the Accession Agree-
ment. ERA Forum 14:557–572
Reestman JH (2005) Primacy of union law. Eur Const Law Rev 1(1):104–107
Ribble MR (2010–2011) I don’t trust your judgment: the European Convention on Human Rights
meets the European Union on new grounds? Pa State Int Law Rev 29(1):211–232
Rieder C (2005) Protecting human rights within the European Union: who is better qualified to do
the job-the European Court of Justice or the European Court of Human Rights? Tulane Eur
Civil Law Forum 20:73–107
Ryngaert C (2011) ‘The European Court of Human Rights’ approach to the responsibility of
Member States in connection with acts of international organizations. Int Comp Law Q 60
(4):997–1016
Sarvarian A (2014) The EU accession to the ECHR and the law of international responsibility. In:
Kosta V et al (eds) The EU accession to the ECHR. Hart Publishing, Oxford
References 239

Sera JM (1996) The case for accession by the European Union to the European Convention for the
protection of human rights. Boston Univ Int Law J 14:151–186
Shelton D (2003) The boundaries of human rights jurisdiction in Europe. Duke J Comp Int Law
13:95–153
Simma B (1985) Self-contained regimes in international law. Netherlands Yearb Int Law
16:111–136
Streinz T (2013) Avoiding an Odyssey – the EU’s accession to the ECHR. VerfBlog, 2013/9/30.
Available at: http://www.verfassungsblog.de/en/avoiding-an-odyssey-the-eus-accession-to-
theechr/
Sweet AS (2009) Constitutionalism, legal pluralism, and international regimes. Indiana J Global
Legal Stud 16(2):621–645
Tabarelli M (2013) The influence of the EU and the ECHR on “Parliamentary Sovereignty
Regimes”: assessing the impact of European integration on the British and Swedish Judiciaries.
Eur Law J 19(3):340–363
The European Court of Human Rights (2014) The ECHR in 50 Questions, p 8. Available at: http://
www.echr.coe.int/Documents/50Questions_ENG.pdf
Thym D (2009) In the name of sovereign statehood: a critical introduction to the Lisbon judgment
of the German Constitutional Court. Common Market Law Rev 46(6):1795–1822
Thym D (2013a) A Trojan horse? Challenges to the primacy of EU law in the Draft Agreement on
Accession to the ECHR. VerfBlog, 2013/9/11. http://www.verfassungsblog.de/en/atrojan-
horse-challenges-to-the-primacy-of-eu-law-in-the-draft-agreement-on-accession-to-the-echr/
Thym D (2013b) Separation versus fusion – or: how to accommodate national autonomy and the
charter? Diverging visions of the German Constitutional Court and the European Court of
Justice. Eur Const Law Rev 9(3):391–419
Tomuschat C (1983) Liability for mixed agreements. In: O’Keeffe D, Schermers HG (eds) Mixed
agreements. Kluwer, Deventer
Tomuschat C (2002) The international responsibility of the European Union. In: Cannizzaro E
(ed) The European Union as an actor in international relations. Kluwer Law International, The
Hague
Tridimas T, Poli S (2008) Locus Standi of individuals under Article 230 (4): the return of
Euridice? In: Arnull A, Eeckhout P, Tridimas T (eds) Continuity and change in EU law –
essays in honour of Sir Francis Jacobs. Oxford University Press, Oxford
Tulkens F (2013) EU accession to the European Convention on Human Rights. Speech at: National
School of Judiciary and Public Prosecution (KSSIP) – Krakow (Poland), Friday 1 March 2013
van den Berghe F (2010) The EU and issues of human rights protection: same solutions to more
acute problems? Eur Law J 16(2):112–157
van Dooren MI (2009) The European Union and human rights: past, present future. Merkourios
70:47–52
van Rossem JW (2009) Interaction between EU law and international law in the light of Intertanko
and Kadi: the dilemma of norms binding the Member States but not the community. Nether-
lands Yearb Int Law 40:183–227
Von Bogdandy A et al (2012) Reverse Solange – protecting the essence of fundamental rights
against EU Member States. Common Market Law Rev 49(2):489–519
Weber GS (2007) Who killed the friendly settlement? The decline of negotiated resolutions at the
European Court of Human Rights. Pepperdine Disp Resolut Law J 7(2). http://digitalcommons.
pepperdine.edu/drlj/vol7/iss2/4
Wessel RA (2000) The inside looking out: consistency and delimitation in EU external relations.
Common Market Law Rev 37(5):1135–1171
Wetzel JR (2003) Improving fundamental rights protection in the European Union: resolving the
conflict and confusion between the Luxembourg and Strasbourg Courts. Fordham Law Rev
71:2823–2862
White S (2010) The EU’s accession to the Convention on Human Rights: a new era of closer
cooperation between the Council of Europe and the EU? New J Eur Crim Law 1(4):433–446
Chapter 7
Inter-Party Mechanism and the EU: Possible
Implications from the Strasbourg’s
Jurisdiction?

7.1 Introduction

This chapter provides an analysis of the inter-party complaint procedure in an


EU-ECHR post-accession context, tackling not only the reconciliation of EU and
Convention regimes to enable the utilization of this mechanism but also means
which show evidence of how EU law autonomy may be protected therein. The
chapter responds to the research sub-question: ‘Whether and, if yes, how can ECHR
and EU Law reconcile to allow for EU/Member States to use inter-party mechanism
of ECHR?’ The chapter examines carefully the EU law limitations on the use of
external tribunals for purposes of interpreting segments of EU law, under the note
that ECHR has already become part of EU law. To this purpose, the chapter looks at
the extent to which the EU Treaties’ limitations on external tribunals’ jurisdiction
should be interpreted, and shows ways which may reconcile it with ECHR pro-
visions on inter-party jurisdiction of the Strasbourg Court. Mox Plant and Ireland
v. UK are used and systematically examined against the inter-party mechanism and
mechanics provided for by the Accession Agreement, with the intention of testing
the applicability, integrity and functionality of its application with EU as one of the
parties. Careful examinations of how the EU Treaties may become accommodated
to the DAA’s inter-party complaint procedures are provided, whereas several
consensual arguments have been laid down to make certain that the inter-party
procedure may be invoked in the EU context as well however several safeguards
need be introduced to ensure the EU law autonomy in this regard. The issue of
abstract control of EU law under the inter-party complaint mechanism is also
considered in the light of EU law autonomy sensitivity. Briefly, the chapter also
covers the possibility for CFSP measures to become attacked before the Strasbourg
Court, the latter being a first-instance court for such branch of EU law. The chapter
concludes by arguing that the inter-party mechanism under the DAA and ECHR
may allow space for reconciliation with the EU law, with EU being a party that may

© Springer International Publishing Switzerland 2015 241


F. Korenica, The EU Accession to the ECHR, DOI 10.1007/978-3-319-21759-8_7
242 7 Inter-Party Mechanism and the EU: Possible Implications from the. . .

file and become attacked under this mechanism before the Strasbourg Court
(certainly EU law autonomy questions implicitly raised there as well).

7.2 A Background on the Inter-Party Complaint


Mechanism with a View to EU-ECHR Peculiar Context

International human right treaties usually provide for individual protection of


human rights, wherein individuals are given access to certain judicial institutions
requesting the restoration of a right deriving from an international human right
instrument.1 Individual complaint procedures before international human rights
tribunals comprise the common legal practice within most international human
rights instruments. However, besides individuals, many international human rights
instruments provide for the possibility that a certain human right complaint origi-
nate and be adjudicated between states inter se, namely the inter-state complaint
procedure.2 Inter-state complaint procedures are not the common, and probably the
most usual form, to require the protection of human rights.3 However, the mere
existence of such a mechanism within many international human right treaties
points to the need for analyzing its possible impact, especially in the context of
very specific international organizations law like the EU.
The inter-state complaint4 mechanism is normally anticipated to raise political
ramifications for the relations between the applicant and respondent state.5 The

1
This type of action is the most common of its kind in conventional international law. See:
Robertson (1959), p. 397.
2
This type of dispute settlement has existed even before the WWII. However, its use had followed
mainly informal patterns and did not entail human rights disputes as such, but stuff like borders
between states or commerce between them. See an account of the historical route of the interna-
tional dispute settlement and its development, at: Caldwell (1920); See also: Shelton (1989), p. 15,
on the basic idea in international law that states are entitled to protect their citizens against the
human rights violations of other states.
3
Licková (2008), p. 479; To note, ECHR gives full authority to each state party to utilize and to
become subject to this mechanism, which does not appear with e.g. in the International Covenant
on Civil and Political Rights, wherein many state parties have not given jurisdiction to become
subjects to the inter-state mechanism. See: Heffernan (1997), p. 88; In the ICCPR case, even
though the parties might have submitted to the Commission’s inter-state jurisdiction, they are not
obliged to implement its recommendations, adding that there is no binding effect of such
recommendations. See e.g.: Comment (1978), p. 899; Cf: Pasqualucci (1994/1995), pp. 313 et seq.
4
An EU inter-state complaint, e.g.: Court of Justice of EU, France v UK, Case 141/78 [1979]
ECR 2923.
5
The UN Charter, e.g., establishes that the disputes between the UN members should normally be
referred to the International Court of Justice for resolution. See: Article 36(3) of the UN Charter;
See also: Schreuer (2008), p. 965.
7.2 A Background on the Inter-Party Complaint Mechanism with a View. . . 243

political implications6 supposed to follow the use of the inter-state mechanism are
seen as the key barrier for the normal and frequent use of such procedure in
international judicial proceedings. It also ‘highlights difficult problems of
co-ordination between competing judicial procedures which culminate in a signif-
icant risk of incompatible judgments.’7 Due to this reason, the very exceptional use
of this mechanism seems to support the argument that its suitability for human
rights protection does not reach the basic expectations. However, given the reality
that the mere use of this mechanism could produce several inter-state, but also
broader, legal implications directs us to the fact that the need to explain the legal
consequences of such mechanism are highly important even if one overlooks the
legal dimension, wherein jurisdiction versus treaty interpretation often becomes a
huge stone to be passed.8
The configuration of the inter-state mechanism provides for the possibility that a
state appear as the applicant before an international tribunal against another state,
both being parties to the same international human right treaty, requesting that the
respondent state be held responsible for the violation of human rights in its territory/
jurisdiction or its dependent territory. Certainly, such act of the applicant state
could be seen as interfering to the domestic affairs of the respondent state, therefore
there are numerous reasons for the latter to trust that the applicant state is provoking
a political battle through international judicial means. Such circumstances however
need be rationalized with the fact that the observance of international human rights
treaties for purposes of implementing them at the domestic level could not be seen
as interference in the domestic affairs of the respondent state. Rather, their aim is
merely to support the protection of human rights through means that go beyond
individual complaint procedures.
The European Convention on Human Rights provides also for the inter-state
complaint procedure, wherein a High Contracting Party submits an application
against another High Contracting Party regarding an alleged violation of the
Convention rights.9 This mechanism makes it possible for the Convention parties
to observe collectively the compliance of any party with the Convention system.10

6
To this point, Gragl argues that it is the political implications which have basically prevented the
use of the inter-state mechanism under the ECHR, therefore today there are only a few inter-state
cases that have appeared in practice. See: Gragl (2013), p. 139.
7
Shany (2004), p. 816.
8
Pauwelyn (2004), p. 913.
9
See: Costa (2003), p. 455; See also: Grabenwarter (2009) (the entire section on inter-state
procedure in the ECHR system); Lock (2010), p. 795; See also relevant caselaw of this type,
such as: ECtHR, Ireland v. United Kingdom, Appl. No. 5310/71, Judgment of the Court of
18 January 1978; ECtHR, Denmark v. Turkey, Appl. No. 34382/97, Judgment (Friendly Settle-
ment) of the Court of 5 April 2000; ECtHR, Cyprus v. Turkey (IV), Appl. No. 25781/94, Judgment
of the Court of 10 May 2001.
10
Gragl (2013), p. 174.
244 7 Inter-Party Mechanism and the EU: Possible Implications from the. . .

The parties named as respondents are obliged and should submit themselves to the
Strasbourg Court’s jurisdiction,11 ‘[. . .] which is essentially an international court –
although one with a remarkable jurisdiction – and a court which does not seek to
unify the law, but rather to set a minimum European standard.’12 The DAA,
however, will replace the word “inter-state” with the word “inter-party”, aiming
to enable the EU to appear as a High Contracting Party to participate as a fully-
fledged party in the previously named inter-state procedure.13 The drafters of this
provision (Art. 4 DAA), according to Gragl, considered it too privileging to grant
the EU immunity from the inter-party jurisdiction. Therefore, they decided to
enable such jurisdiction of the Court upon the EU as well.14 This would allow
that three possible scenarios appear within the context of EU accession to ECHR.
First, that one EU Member State file an application against another EU Member
State, second, that an EU Member State file an application against the EU, and
third, that the EU file an application against an/some EU Member State(s). Each of
these scenarios touches the very existence of the EU Treaties exclusivity to regulate
EU-Member States relations.15 However, as Gragl rightly points out, this mecha-
nism may also be used in the future by EU against non-EU contracting parties to the
Convention, as part of its external human rights policy.16
With regard to the inter-state mechanism, Gragl argues that there exist two
possibilities for utilizing it: first, with the aim of protecting the interests and
human rights status of its citizens, a contracting party files an application against
another contracting party with an individual case at its object, and, second, a
contracting party—in the format of an ‘actio popularis’—files an application before
the Court against another contracting party for having violated the European public
order.17 The Convention provides specifically that:
Any High Contracting Party may refer to the Court any alleged breach of the provisions of
the Convention and the Protocols thereto by another High Contracting Party.18

11
Compare this jurisdiction with that of the The International Convention on the Elimination of
All Forms of Racial Discrimination, wherein the jurisdiction of the Court does not depend from the
addresses. See e.g. Comment (1978), p. 895.
12
Jacobs (2007), p. 11.
13
This provision, according to Gragl, by changing the name of the heading from ‘inter-state’ to
‘inter-party’, accommodated the supranational features of the EU as a non-state entity into the
ECHR system. See: Gragl (2013), p. 184.
14
Gragl (2013), p. 184.
15
It is important to note, however, that there was no application under this mechanism between EU
Member States since the founding of the EU structures. See: Gragl (2013), p. 175.
16
Gragl (2013), p. 175.
17
Gragl (2013), p. 174/5.
18
Art. 33, ECHR.
7.2 A Background on the Inter-Party Complaint Mechanism with a View. . . 245

Art 33 of the Convention clearly allows that both the EU and each individual EU
Member State be recognized as equal High Contracting Parties regarding the inter-
state complaint procedure. It is important to bring the ICJ’s perspective on
interpreting Art. 33 ECHR. ICJ, in an Advisory Opinion, ruled that ‘the first duty
of a tribunal which is called upon to interpret and apply the provisions of a treaty, is
to endeavour to give effect to them in their natural and ordinary meaning in the
context in which they occur. If the relevant words in their natural and ordinary
meaning make sense in their context, that is an end of the matter.’19 If this ICJ
standard was to be applied upon Art. 33 ECHR, the Convention system should
make room for EU Member States to utilize the inter-party procedure uncondition-
ally. This, however, could touch upon the borders of the EU law, as the Strasbourg
Court could be given authority to interpret the EU-Member States relations in a
possible scenario wherein Member State(s) refer questions of EU law (direct or
indirect) for resolution at the Strasbourg Court, which certainly interferes to the EU
law autonomy. In principle, it was long argued that the inter-state procedure should
be explicitly excluded from the DAA, and from the entire EU standing according to
the Convention. Such exclusion, it was argued, would prevent the Strasbourg Court
from interfering to the EU law autonomy.20 The Luxembourg Court—in its Opinion
2/13 (para. 213)—does not request that the inter-party procedure be entirely
excluded, but that the DAA specifically mention that the Strasbourg Court will
have no jurisdiction ratione materie to rule on such cases (supra the subchapter on
Opinion 2/13).
With the DAA clearly providing for the possibility of inter-party applications,21
it is necessary to analyze the overall but also two specific dimensions of such
mechanism. Such two specific issues that need be considered are: first, the juris-
dictional exclusivity of the Luxembourg Court over the EU law and the jurisdic-
tional exclusivity of the Strasbourg Court over the Convention, and their potential
clash within the inter-party mechanism; and second, the delimitation of their
authority within the context of the inter-party mechanism, while analyzing means
and spaces through which these two orders may become reconciled on this issue.

19
International Court of Justice, Advisory Opinion on the Competence of the General Assembly
for the Admission of a State to the United Nations, I.C.J. Reports 1950, p. 8 [emphasis added].
20
Lock (2010), p. 797; Disagreeing with Lock, see: Analysis (1997), p. 238/9.
21
To make a short digression from the subject, the EU Court has always liked to match the
understandings of concepts in Member States’ legal orders and the EU legal order. For instance, it
had ruled that the notion of commercial policy has ‘the same content whether it is applied in the
context of the international action of a State or to that of the Community’. See: Court of Justice of
EU, Commission v Council, Case 45/86 [1987] ECR 1493, para 16. If the EU Court would apply
this view with the same degree of deference to the understanding of the inter-party mechanism,
there would seem no dilemma on recepting the latter as fully tolerable also under the EU law.
246 7 Inter-Party Mechanism and the EU: Possible Implications from the. . .

7.3 The Possible Impairment of the Luxembourg Court’s


Exclusive Jurisdiction by the DAA: A Perspective
on the Inter-Party Mechanism

It stems from the general nature of the EU law autonomy that the Luxembourg
Court is granted with the exclusive jurisdiction of Art 19 (1) TEU to interpret and
observe the EU law. This jurisdiction is complemented with Art 344 TFEU, which
clearly sets out the prohibition that disputes between EU Member States should be
solely resolved within the EU courts’ system.22 The autonomy of EU law, therefore,
makes it clearly visible that no external interference is allowed when it comes to the
disputes arising among EU and Member States upon EU law matters, for which the
Luxembourg Court holds an exclusive jurisdiction.23
The prohibition of Art 344 TFEU24 has been explicitly ruled upon also in Mox
Plant, where the Luxembourg Court25 ruled that no international agreement, to
which the Union is a party, be it mixed agreement or not, could affect the allocation
of powers set by the Treaties, whereas the Luxembourg Court is the sole organ
responsible to observe the law.26 It was reiterated in this case the duty of the
Member States not to submit any dispute arising from the EU law to any other
method of dispute settlement other that the mechanisms deriving from the Treaties
themselves.27 Opinion 2/13 (para. 213) supports this claim further, although with a
less rigid accent. This instruction stands clearly within the words of Protocol 8 to
the Treaties, wherein not only the prohibition of affecting the allocation of respon-
sibilities provided for by the Treaties but also specifically the prohibition of
affecting Art 344 of the Treaties was laid down. On the other hand, Art.
33 ECHR provides for the possibility that the Convention parties refer to the
Strasbourg Court an inter-party case, wherein one High Contracting Party chal-
lenges the acts or omissions of the another High Contracting Party, something
which could lead to EU Member States challenging EU law at Strasbourg.28
However, in principle, there seems no overlap of authorizations as long as Art

22
See, e.g.: on the case of arbitrations, how far can this relationship go, at: Von Papp (2013),
p. 1051 and 1054; See also: Odermatt (2014), p. 13.
23
E.g. Intertanko case, see at: Mendez (2010), p. 1751; See also: Groussot et al. (2011), p. 5.
24
One of the interesting proposals to accommodate the possibility for the EU-originated parties to
get access to international arbitration tribunals—without violating Art. 344 TFEU—is to consider
those tribunals as being Member States’ arbitration courts. See this, at: Von Papp (2013), p. 1081.
25
It is normally proposed that the EU Court increase its deference to its co-European court, the
Strasbourg Court, in order to envisage a more structured human right protection that currently does
not seem to be the case in Europe. See e.g. Gerards (2011), p. 81 & 85 et seq.
26
Court of Justice of EU, Case C-459/03 Commission v Ireland (Mox Plant) [2006] ECR I-4635,
para 123; See also: Court of Justice of EU, Opinion 1/91 [1991] ECR I-6079, para. 35; Court of
Justice of EU, Opinion 1/00 [2002] ECR I-3493, paras. 11 and 12; See also: Martinico
(2012), p. 424.
27
This includes any international law tribunal. See generally: Lavranos (2006), p. 224 et seq.
28
E.g.: Lock (2010), p. 795; See also: Gragl (2013), p. 176.
7.4 Prohibition of Protocol 8(3) and Art 344 TFEU: Is There an. . . 247

33 ECHR and Art 344 TFEU are concerned. This stems from the fact that Art
33 ECHR refers merely to the jurisdiction of the Court to resolve cases contesting
the applicability of ECHR, whereas Art 344 TFEU refers to the interpretation and
application of the Treaties. However, once Treaties or legislation deriving thereof
becomes the source of violation of ECHR (with ECHR and DAA themselves being
incorporated in EU law29), the two articles could become conflicting and
overlapping.30 The latter possible scenario therefore should be extensively exam-
ined below, although Opinion 2/13 (para. 203 et seq) comes to this conclusion as
well with an extremely brief and non-convincing reasoning.

7.4 Prohibition of Protocol 8(3) and Art 344 TFEU: Is


There an ‘Intended’ Understanding?

It is clearly defined in Article 3 of Protocol 8 to the Treaties that the DAA should
not affect Art 344 of TFEU,31 making it exclusively prohibited that such an article
be arranged differently. Art 344 TFEU reads:
Member States undertake not to submit a dispute concerning the interpretation or applica-
tion of the Treaties to any method of settlement other than those provided for therein.

Referring to Lock, Gragl argues that the core objective of Art. 344 TFEU was to
predominantly block EU Member States from having recourse to the International
Court of Justice, as the EU Member States had accepted its inter-state jurisdiction
either on basis of Art. 36 (1) or 36 (2) of the ICL Statute.32 In this regard, Gragl
notes that Art. 344 TFEU is a lex specialis to the general principle of sincere
cooperation under the EU Treaties.33 From a more accession-related perspective,
Art 344 TFEU may be inclusively commented by six noteworthy assessments.
First, that EU Member States undertake not to submit a dispute over the
interpretation of application of the Treaties to another mechanism not foreseen by

29
On basis of the Haegeman doctrine, as ruled in: Court of Justice of EU, R & V Haegeman v
Belgian State, Case 181/73 [1974] ECR 449, paras. 4–6.
30
As opposed to the scope of this chapter, Gragl also considers it relevant to examine (and offers a
separate chapter on) the EU’s use of inter-party complaint procedure before the Strasbourg Court
as an external human rights policy against potential third countries. This book clearly—in the same
line with Gragl—considers that there is no dispute on the fact that the DAA and Art. 344 TFEU do
not make any prohibition—therefore permit—the utilization of the inter-party mechanism by the
EU and its Member States against non-EU ECHR contracting parties. The use of the inter-party
jurisdiction of the Strasbourg Court against the non-EU ECHR contracting parties as an external
human rights policy by the EU is examined profoundly by Gragl. See: Gragl (2013), p. 197–204.
31
See e.g. Raba (2013), p. 559, arguing that the DAA may not interfere to Art. 344 TFEU; See
also: Gragl (2013), p. 177; Cf. the latter to CJEU Opinion 2/13 para. 203 et seq; See also: Odermatt
(2014), p. 16.
32
Gragl (2013), p. 21.
33
Gragl (2013), p. 22.
248 7 Inter-Party Mechanism and the EU: Possible Implications from the. . .

the Treaties, such as the Convention system. From a textual interpretation, Art
344 refers to the word ‘dispute’34 as a device rather than an objective of this
stipulation. One could argue that disputes arise merely in cases when there are at
least two conflicting positions, in this case either between two Member States or
one Member State and the EU. ‘A dispute may exist even if one party does not
oppose the other party’s position but fails to provide a remedy.’35 For the dispute to
appear, the conflicting parties must have already taken a stance over the interpre-
tation or application of the Treaties. Supposing, one might ask, that there is no
dispute over the interpretation of application of the Treaties, would there be any
prohibition resulting from Art 344 TFEU? Such an example would exist when two
parties, say two Member States of the EU wish to have an EU legal measure
checked against the Convention standards, not contesting at all its applicability.
Answering to this question, seeks to consider the teleological rationale of Art 344.
In fact, the latter is provided for ensuring that the Luxembourg Court remains the
only last-resort domestic body to deal exclusively with the EU law.36 Therefore, it
seems that the concept ‘dispute’ within Art 344 TFEU refers to the prohibition of a
result which would engage an external mechanism to judge an issue relating to the
EU law, rather than to the existence of the dispute as such.
Second, Art 344 TFEU attaches the word ‘dispute’ to the context of interpreta-
tion and application of the Treaties. The nature of prohibition located within the
context of this provision seems to invalidate any action from interpretation to the
application of Treaties submitted outside the realm of the Luxembourg Court,
which in fact encompasses the entire range of legal actions that could appear
therein. This is very relevant to the inter-party procedure as well.
Third, this provision does not bind EU to the duty to refrain from submitting a
dispute arising from the Treaties to an external mechanism. Although it might seem
funny to have EU contesting its own legal act in an external mechanism, such as the
Strasbourg Court, it might be the case that the EU submit an action to an external
mechanism such as Strasbourg against one of its Member States or Third States
(association partners),37 on an issue indirectly or implicitly relating to EU law. Such
a scenario, that could fall within the CFSP policy field, would very well be founded
and not prohibited by Art 344. Its result, however, would directly interfere to the
autonomy of the EU law. The question of whether EU is self-censored from
submitting a legal action which would indirectly interfere to its autonomy is
debatable, and rather a nonrepresentational exercise of this scenario. The latter
scenario—interestingly enough—is not captured by the eyes of the Luxembourg
Court in Opinion 2/13.
Fourth, the question arising out of this provision is also whether the word
‘Treaties’ encapsulates merely the Treaties as such, or the entire spectrum of

34
On this notion, see e.g.: Schreuer (2008), p. 965.
35
Schreuer (2008), p. 978.
36
See also: Court of Justice of EU, Opinion 1/91 (1991) ECR I-6079.
37
Quirico (2010), p. 38.
7.4 Prohibition of Protocol 8(3) and Art 344 TFEU: Is There an. . . 249

legislation deriving thereof. Textually, the provision suggests that it prohibits


Member States merely to submit the dispute over the Treaties to an external
mechanism, leaving outside the legislation originating from the Treaties. However,
seen objectively, the term Treaties in this context should be read as covering the
entire spectrum of legislation that derives therefrom,38 as the interpretation and
application of the Treaties is merely the cap which the provision refers to as the
point that encapsulates everything therein. One may also argue that international
agreements are also included in that wording, including the DAA and the Conven-
tion itself.
Fifth, that the EU court system, namely the Luxembourg Court, retains its
exclusive jurisdiction to interpret and decide on the application of the Treaties
and legal acts deriving thereof. In that scenario, however, Art 344 TFEU merely
assures that this relationship is binding on EU Member States, but not on the EU as
such. The prohibition to submit disputes to other means of dispute settlement is
attached to the Member States only, not to the EU. This said, Art 344 TFEU is the
point that regulates and assures the relationship between the internal and external
autonomy of EU law, ruling out that Member States interfere to the Luxembourg
Court’s jurisdictional exclusivity from an international tribunal position.
Sixth, the point is what if the dispute is not directly linked to the Treaties,
e.g. such as those rights which are written both in the Treaties and in the ECHR?
Or, alternatively, if the Convention is considered to be part of EU primary law
(or simply EU law), or at least in between the primary and secondary law, would
Art. 344 TFEU mean as even if there is a question relating to the Convention be
prohibited to get outside of the Luxembourg Court? Of course, as long as the
question relates to the Convention being part of the EU law obligation, Art.
344 TFEU is meant as covering all possible spectrums of questions that might
arise in that context, therefore prohibiting their submission to the Strasbourg Court.
To dig deeper in this regard, one needs to analyze the systemic picture of the EU
Treaties on this aspect. Hence, analyzing Art 344 TFEU, interpreted in conjunc-
tion with Art 3 of Protocol 8, produces two cumulative explanations:
First, that the DAA should preserve the nature and substance of Art 344 TFEU,
namely the prohibition that EU Member States submit a request of interpreting or
applying the Treaties to an external method of dispute settlement. This would mean
that, practically, the DAA should extend the prohibition for EU Member States to
submit such a request to the European Court of Human Rights, which in Art

38
E.g.: see Bernhard (2011), para 1; Gragl also proposes that the wording ‘treaties’ of the above
provision covers also secondary law and international agreements of EU legal order. Gragl argues
that this approach has also been confirmed by the Luxembourg Court in Opinion 1/91, although he
does not reason carefully such statement. See: Gragl (2013), p. 176; One may agree with Gragl,
however, if reference is made to Mox Plant, which rather well takes that position (see Mox Plant
infra).
250 7 Inter-Party Mechanism and the EU: Possible Implications from the. . .

344 TFEU terms is considered an external method of settlement.39 The two pro-
visions could also be broadly commented as meaning that the DAA should preserve
the entire nature of the existing authority of EU law not only in terms of its external
but also internal autonomy. Such prohibition, however, would not apply if Member
States challenge each other’s non-EU law. Member States, in this respect, may not
surpass their EU Court in exchange to the Strasbourg Court’s jurisdiction in any
matter relating to EU law.
Second, read especially in conjunction with Art 1 of Protocol 8 (point 2), the
DAA should ensure that submissions by non-Member States are appropriately
addressed to Member States and/or Union as appropriate. This speaks for a sort
of co-respondent mechanism that needs be applied also in the context of inter-party
cases, in order to ensure that the respondents’ responsibility is addressed correctly
by the addressee and that the Court does not embark into the division of that
responsibility in cases where both the EU and one of its Member States are mutually
responsible. This said, the division of responsibility between the EU and its
Member States remains merely an internal issue, with the Luxembourg Court
preserving its exclusive jurisdiction to rule on Treaty interpretation and application
even in inter-party cases.
Let us however, consider the fact whether there is any segment of law that may
become reviewed at Strasbourg Court, in which case Art. 344 TFEU would not apply.

7.5 ‘Escaped’ Exclusivity in Inter-Party Complaints: Art


344 TFEU in Light of Art 19(1) TEU and with
Reference to Art 275 TFEU: What Would This Entail
in Terms of the DAA?

It is important to note that the inter-party mechanism would have to respect the
legal autonomy of the EU law, and the Luxembourg Court’s exclusive jurisdiction.
One such issue however needs further examination when it comes to the Common
and Foreign Security Policy of the EU. To examine this issue systemically, let us
undertake a systemic interpretation of the Treaties’ provisions with regard to the
common foreign and security policy.
Art 275 TFEU clearly sets that:
The Court of Justice of the European Union shall not have jurisdiction with respect to the
provisions relating to the common foreign and security policy nor with respect to acts
adopted on the basis of those provisions.

39
Gragl notes that EU Member States may use Art. 258/259 TFEU (infringement proceedings),
Art. 263 TFEU (action for annulment), and Art. 265 TFEU (action for failure to act) as tools to
solve their potential EU law conflicts inter se before the Luxembourg Court—as opposed to Art.
33 ECHR. See: Gragl (2013), p. 177.
7.5 ‘Escaped’ Exclusivity in Inter-Party Complaints: Art 344 TFEU. . . 251

In light of Art 275 TFEU, it is clearly understood that the Luxembourg Court has
no authority to rule on the provisions relating to the CFSP,40 with the same
prohibition extending to the entire spectrum of acts and policies deriving thereof
(under Art. 24 TEU).41 Read in conjunction with Art 19(1) TEU, Art 275 TFEU
means that the context of jurisdictional exclusivity of the Luxembourg Court does
not cover anything relating to the CFSP policy area (excluding restrictive measures
under Art. 275(2) TFEU).42 In that context, the argument is that the Luxembourg
Court has no jurisdiction at all on this policy field, whereas the prohibition to rule
thereon appears not to touch upon its jurisdictional exclusivity. Therefore, the lack
of authority to rule on the CFSP arena of the Luxembourg Court is not respectively
included in the EU law autonomy principle.
From an outdoor point of view, one should question whether Art 275 TFEU
plays a core role in delimitating the external borders of Art 344 TFEU. Arguendo,
the answer is affirmative, as Art 275 TFEU clearly defines the scope of application
of Art 344 TFEU. In this respect, having argued that Art 275 TFEU does clearly set
the authority on CFSP arena outside the framework of the autonomy of EU law
provided for by Art 19(1) TEU, it is submitted that Art 344 TFEU is narrowed by
Art 275 TFEU. This said, the Luxembourg Court has no jurisdiction to rule on the
CFSP arena. Therefore, Art 344 TFEU should be read as not covering the EU
Court’s lack of jurisdiction to rule on the CFSP arena. In this regard, the prohibition
that Member States submit a dispute on the interpretation and application of EU law
to an external method of settlement does not extend on the CFSP ground.
This said, it is clear that Art 344 TFEU does not prohibit Member States from
submitting a dispute regarding the interpretation and application of CFSP to an
external method of dispute settlement. In this light, Art 344 TFEU does not prohibit
Member States from submitting an issue relating to the interpretation or application
of CFSP to another method of dispute settlement, be it the European Court of
Human Rights or another international tribunal. This is so far very clear. However,
the point that needs be further clarified is whether the lack of jurisdiction of the
Luxembourg Court to rule on the interpretation and application of CFSP includes
merely the policy context or also the normative context?
Through reading carefully Art 275 TFEU, one observes that it undoubtedly rules
out any jurisdiction of the Luxembourg Court both to the provisions relating to the
CFSP and the acts deriving from such provisions. In this regard, the argument is that
Art 275 TFEU rules out any jurisdiction for the Luxembourg Court first of all on

40
See also: Quirico (2010), p. 39; See also: Lock (2012), p. 188; De Londras and Kingston
(2010), p. 365.
41
Regarding the scope of acts that may fall under this constitutional domain, Art. 24 (1) TEU
reads: ‘The Union’s competence in matters of common foreign and security policy shall cover all
areas of foreign policy and all questions relating to the Union’s security, including the progressive
framing of a common defence policy that might lead to a common defence.’.
42
This therefore means that the EU Court has jurisdiction to rule on restrictive measures against
persons adopted by Council on the basis of Chapter 2 of Title V of the Treaty on European Union.
Such jurisdiction of the EU Court thus comes under the scope of prohibitions of Art. 344 TFEU.
252 7 Inter-Party Mechanism and the EU: Possible Implications from the. . .

interpreting and applying the Treaties’ provisions on CFSP, including then the
entire spectrum of secondary legislation deriving from such treaty provisions. In
essence, Art 275 TFEU does evidently exclude the Luxembourg Court from any
form of engagement with the interpretation, observation and application of the
normative acts, be them primary, secondary or in-between sources of law, therefore
thinning the scope of the Court’s jurisdictional authority both in terms of internal
and external autonomy.
Read altogether as a package, through a systemic analysis of Art 19(1) TEU, Art
344 TFEU and Art 275 TFEU, it is argued that Member States are not prohibited
from submitting to an external method of dispute settlement issues relating to the
entire spectrum of EU law provisions and acts based on those provisions in the
CFSP field. In that scenario, an EU Member State is not prohibited by Art
344 TFEU to submit directly to the Strasbourg Court an issue of law or policy
within CFSP for its compliance with the Convention. Such a scenario would mean
that a provision of the Treaties on CFSP could be submitted for review with the
Convention to the Strasbourg Court, while not interfering to the EU law autonomy.
It remains debatable, however, whether this lacuna also covers acts adopted on
ordinary procedure based on a prior CFSP common joint action under Art.
75 TFEU, such as the issue of freezing of assets through a Regulation under Art.
215 TFEU, and acts adopted under the flexibility clause provision of Art. 352 TFEU
wherein a CFSP measure is linked to a certain EU objective and then transformed
into ordinary legislation through Art. 298 TFEU, as was the case in Kadi for
example. It is implied that, nevertheless, the lacuna does not cover any of the
above mentioned acts,43 as they originate from an ordinary procedure and may not
be seen merely as CFSP measures and/or acts. The Kadi case would confirm this
unequivocally. The category of acts whereby CFSP policy becomes applied is
defined by Art. 25 TEU, which reads that:
The Union shall conduct the common foreign and security policy by:
1. defining the general guidelines,
2. adopting decisions defining:
a. actions to be undertaken by the Union;
b. positions to be taken by the Union;
c. arrangements for the implementation of the decisions referred to in points (i) and (ii);
and by
3. strengthening systematic cooperation between Member States in the conduct of policy.

Cremona rightly argues that only decisions under Art. 25 (2) TEU would be of an
individual nature and produce a binding effect, therefore having the legal capacity
to produce direct concern to addressants.44 Only the category of acts under Art. 25
(2c) TEU may therefore harm one’s rights under the Convention. These types of

43
See e.g. generally: Eckes (2013), p. 282.
44
Cremona (2003), p. 1356.
7.6 Inter-Party Cases in Light of Mox Plant and Art 344 TFEU: Which. . . 253

acts will of course have the capacity to access the Strasbourg Court as a first-
instance court post-accession.
In that regard, such a scenario could make the Strasbourg Court capable of
reviewing the EU law on CFSP field compatibility with the Convention, touching
upon the internal affairs of EU law. This was interestingly requested in a positive
way by a European Parliament resolution which called this a ‘compensation’ for the
lack of human rights control within the Union for this policy field.45 On the other
hand, given that there is no jurisdiction to observe the law of CFSP within the EU’s
judicial system, the Strasbourg Court could become the first-instance but also the
only layer of judicial control for such a branch of EU law. With this function, the
Strasbourg Court would decisively become a ‘domestic’ court, as opposed to a
subsidiary one on this segment on law, clearly putting its authority into a delicate
context of EU law autonomy and its internal maintenance. Given the current gap in
the non-reviewability of the CFSP law and acts, the Strasbourg Court’s mechanism
for inter-party cases would seem a good point for the human rights protection in that
branch of law, although its utilization would be politically sensitive for many
reasons. In doing so, the EU might bring up the argument that the treaty provision
on sincere cooperation is being violated, either within the context of CFSP or
outside it, as an issue of treaty has been pulled out for interpretation. Overall,
including CFSP under the review of the Convention system will of course improve
the human rights protection in the Union for this field of law.46

7.6 Inter-Party Cases in Light of Mox Plant and Art


344 TFEU: Which Standards Derive Thereof?

It is highly vital to see how the Luxembourg Court visualizes its own exclusive
jurisdiction over EU law in light of Art 344 TFEU.47 To analyze this legal context,
the Mox Plant case serves as the most relevant case on both testing the Luxembourg
Court’s view and demarcating the borders of Member States’ prohibited conduct on
basis of Art 344 TFEU. The Mox Plant case, as well known and as explained before,
involved two EU Member States into an external dispute settlement procedure that
produced EU law implications in the context of the Luxembourg Court’s jurisdic-
tion on the issue.

45
European Parliament Resolution of 19 May 2010 on the institutional aspects of the accession of
the European Union to the European Convention for the Protection of Human Rights and
Fundamental Freedoms (2009/2241(INI)), para. 1.
46
E.g.: Eckes (2013), p. 283.
47
Lavranos (2006), p. 234.
254 7 Inter-Party Mechanism and the EU: Possible Implications from the. . .

As relates to the facts of the case, Ireland had submitted a dispute against United
Kingdom to the Tribunal of the United Nations Convention on the Law of the Sea48
and Arbitral Tribunal of Convention for the Protection of the Marine Environment
of the North-East Atlantic,49 which are both international law instruments wherein
both states and the EU were parties.50 This case provoked an issue involving
overlapping jurisdictions between three treaty regimes. Having filed the dispute
for settlement to the Arbitral Tribunal, an institution outside the realm of EU law,
the Commission filed an action against Ireland to the Luxembourg Court for failure
to comply with Art 344 TFEU.51 Interestingly enough, the Luxembourg Court ruled
numerous principles regarding the way in which Art 344 TFEU must be read, with
such principles serving the internal and external autonomy of the EU law.
In this part of analysis, I will examine the core principles derived from Mox
Plant case with a view to the inter-party mechanism within ECHR. The core
argument on the relationship concerned is given by the Court in Demirel, Commis-
sion v. Ireland and Mox Plant, stating that:
[. . .] in ensuring respect for commitments arising from an agreement concluded by the
Community institutions, the Member States fulfil, within the Community system, an
obligation in relation to the Community, which has assumed responsibility for the due
performance of the agreement.52

In this light, it is clearly stated by the Court that the application of an interna-
tional commitment deriving from an international instrument wherein the EU and
one or more Member States are a party is an issue of EU law, and the Member
States’ duty for their implementation is towards the Treaties rather than interna-
tional law. In designing this type of relationship, the Court has clearly shifted
international obligations wherein EU and Member States are jointly parties to
obligations of internal EU law.53 By devising this shift, the Court has intended to
hold any dispute and/or request for interpretation of such international obligations
at the EU law level, seriously preserving its exclusive jurisdiction also on interna-
tional commitments wherein EU and Member States are mutually answerable.
Wiler rightly argues that ‘[m]ixed agreements, especially when they do not specify
the demarcation line between Community and Member States, defuse at a stroke the
explosive issues of the scope of Community competences (and treaty making

48
Marsden (2009), p. 314; See also: Lavranos (2008), pp. 578–580; On the substantive claims of
Ireland relating to UNCLOS, see: Volbeda (2006–2007), pp. 218 et seq; Lavranos (2008), p. 581.
49
See also: Shany (2004), p. 816.
50
See generally also: Churchill and Scott (2004), p. 645 et seq; See also: Cardwell and French
(2007), p. 121.
51
See generally: Lavranos (2006), p. 224/5; See also: Cardwell and French (2007), p. 122.
52
Court of Justice of EU, Case 12/86, Demirel [1987] ECR 3719, para. 11; Court of Justice of EU,
Case C-459/03, Commission v. Ireland (Mox Plant) [2006] ECR I-4635, note 26, para 85.
53
See also: Court of Justice of EU, Hauptzollamt Mainz v C.A. Kupferberg & Cie KG a.A., Case
104/81 [1982] ECR 3641, para. 11–13.
7.6 Inter-Party Cases in Light of Mox Plant and Art 344 TFEU: Which. . . 255

power). . .’54 It is the author’s view that exactly this type of vague demarcation line
has lead to the EU Court substantiating its external jurisdiction quite broadly and
authoritatively. Therefore, there are three questions which must be considered in
light of MoxPlant: Luxembourg’s exclusive jurisdiction under Art 19(1) TEU,55
duties under Art 344 TFEU and duties of EU and Member States under Art 4
(3) TEU on loyal cooperation.
To clear out some doubts on the nature of the obligation towards the Convention
for EU Member States, in light of Art 344 TFEU, I will examine the specific
contexts under which Mox Plant becomes very relevant to shed light on the inter-
party mechanism post-accession of EU to the Convention period. In relation to the
outer borders of Art. 344 TFEU, which will serve as the borderline also in the inter-
party mechanism under the Convention, four tests seem to derive from Mox Plant,
all such tests needing to pass in order for an issue to get outside the borders of
prohibition of Art. 344 TFEU.

7.6.1 The First Test: Mix Agreement or Not: Defining


the Attribution of Competence?

A first doubt would be whether the mere fact that the dispute is about a mixed
international agreement,56 such as the Convention, excludes Member States from
Art 344 prohibition. The Luxembourg Court, examining the exclusive or shared
nature of international agreements, had ruled that:
[. . .] the question as to whether a provision of a mixed agreement comes within the
competence of the Community is one which relates to the attribution and, thus, the very
existence of that competence, and not to its exclusive or shared nature.57

The first test that the Luxembourg Court applies in this case is to check whether
it has jurisdiction to observe it on domestic grounds. To do that test, two things have
been analyzed in Mox Plant.
The first issue that needs be examined is determining whether the international
agreement is mixed or not.58 The Court had ruled that mixed agreements have the

54
Weiler (1999), p. 177.
55
Where EU Court tends to be rigid, e.g.: Shany (2004), p. 827.
56
Facilitating mixed nature of international agreements has been a rather loaded function of the
Luxembourg Court. See a rich example, at: Court of Justice of EU, Opinion 2/00 [2001] ECR
I-9713; See also a general article on the nature and scope of Luxembourg’s jurisdiction, at:
Heliskoski (2000).
57
Court of Justice of EU, Case C-459/03 Commission v. Ireland (Mox Plant) [2006] ECR I-4635,
note 26, para 93.
58
Regarding the exclusive competence of the EU, Art. 3(2) TFEU reads: ‘The Union shall also
have exclusive competence for the conclusion of an international agreement when its conclusion is
provided for in a legislative act of the Union or is necessary to enable the Union to exercise its
internal competence, or insofar as its conclusion may affect common rules or alter their scope.’.
256 7 Inter-Party Mechanism and the EU: Possible Implications from the. . .

same position in EU legal order as pure EU agreements.59 The question however is


not whether both mixed and pure community agreements have the same status of
hierarchy in EU legal order. The question is rather whether the Luxembourg Court
has jurisdiction to observe mixed international agreements60 also in the context of
Art 344 TFEU. In Mox Plant, the Court made it clear that the test on this issue
should not be laid down on the question whether the international agreement is
mixed or not, but rather whether the concerned mixed agreement rests in the field of
EU competence.61 This being the case, the ECHR would certainly not be consid-
ered as standing outside the Luxembourg Court’s observing jurisdiction for the
mere fact that it is a mixed agreement.62 Another assessment on this first check
needs be added. The following issue would thus clarify the first test on Art
344 TFEU scope and design.
The second issue that comes under this test, as determined in Mox Plant, is
whether the mixed agreements rest within the competences of EU. To drive this
test, the Court asks whether there exists any competence on the EU for that part of
the mixed agreement—meaning the substance of the case being adjudicated—that
needs be tested against Art 344 TFEU. I call this the test on the attribution of
competence. In undertaking the ‘competence attribution’ test,63 the Court asked
‘whether the provisions of that agreement relied on by [. . .a Member States] before
the [. . .international tribunal/external means of dispute settlement] in connection
with the dispute [. . .in question] come within the scope of Community compe-
tence.’64 Of course there is no need to make a detailed test on whether such a

59
Court of Justice of EU, Case C-13/00 Commission v. Ireland, para. 14; See also: Court of Justice
of EU, Meryem Demirel v Stadt Schw€ abisch Gm€ und, Case 12/86 [1987] ECR 3719, para. 12.
60
E.g.: Marsden (2009), p. 312, on the mixed nature of this agreement. See also: Koutrakos (2010),
p. 485, on the jurisdiction of the Court on mixed agreements. Cf.: Court of Justice of EU, Hermès
International v FHT Marketing Choice BV, Case C-53/96 [1998] ECR I-3603, para. 29, where the
EU Court ruled that it may not nevertheless interpret the exclusive competences of Member States
that fall within a mixed agreement.
61
See e.g.: Koutrakos (2010), p. 485.
62
E.g.: Cardwell and French (2007), p. 123; See e.g.: Court of Justice of EU, Hauptzollamt Mainz
v CA Kupferberg & Cie KG aA, Case C-104/81 [1982] ECR 3641, para. 14, where the EU Court
insists that it has the exclusive competence to define the legal nature of effect of an international
agreement; Contra: Gragl insists that the Convention, from the perspective of EU law post-
accession, will not be a mixed agreement. See: Gragl (2013), p. 177. I fully disagree with Gragl
for two basic reasons: first, the respect for the Convention remains in the competence of both the
EU and Member States when they implement EU law, therefore acting as agents of EU, and,
secondly, it is both the EU and Member States which stand as contracting parties to the Convention
on their own right, but often for the same subject of actions or omissions (wherein both the EU and
its Member States participate in taking measures through a joint responsibility and shared
competence in many policy fields, such as CFSP, judicial and home affairs, and many others.
63
See a rather thorough analysis of the nature and level of competence with regard to the exclusive
jurisdiction of EU Court, at: Romano (2007), pp. 171 et seq; See also: Lavranos (2007), pp. 83 et
seq.
64
Court of Justice of EU, Case C-459/03 Commission v. Ireland (Mox Plant) [2006] ECR I-4635,
note 26, para 86.
7.6 Inter-Party Cases in Light of Mox Plant and Art 344 TFEU: Which. . . 257

competence exists or not.65 This seems the most significant part of the test on the
attribution of competence, on basis of which the Court rules whether it has
jurisdiction to observe that mixed agreement and if it falls within the scope of Art
344 TFEU.
It is obvious that the ‘competence attribution’ test is a bit complex, as it enquires
to verify whether there exist competences on the EU to deal with the part of the
mixed agreements submitted for review to another external method of dispute
settlement. Most of such competences, however, are construed very broadly
which one would argue makes the test highly discretionary and not measurable
and objective.66 That said, the EU list of elected competences on external issues
often seem rather formless and always left open for the Luxembourg’s reign. A
remarkable example is Annex B to the Cartagena Protocol on Biosafety, which had
interestingly established that: ‘[t]he exercise of Community competence is, by its
nature, subject to continuous development’.67 The latter is not only too vague but
also open-ended, or, as Bjoerklund puts it, this ‘appears to leave quite a bit in the
gray zone’.68 The competence attribution test—on the other hand—indeed seeks to
determine whether and on basis of which competence the EU has entered into that
mixed agreement, and where are the borders of the Court’s jurisdiction to observe
that mixed agreement in light of Art 19 (1) TEU. When this becomes clarified, it
would then be easy to conclude whether the ‘competence attribution’ test puts the
issues raised before the external methods of dispute settlement as being within the
scope of prohibition of Art 344 TFEU.
To examine whether there exists any external competence for the EU on an issue
deriving from a mixed agreement, the Court in Mox Plant ruled that there is in
principle no need to look for secondary legislation in order to define the external
competence of the EU to conclude that mixed agreement. It would suffice that the
competence derive merely from the Treaties, and that there be no secondary
legislation giving concrete effect to that competence, for the EU to enter into
international agreements (going beyond ERTA formula).69 The EU ‘can enter into

65
E.g.: Koutrakos (2006), p. 152; See also: Court of Justice of EU, Merck Genéricos – Produtos
Farmacêuticos v Merck, Case C-431/05 [2007] ECR I-7001, para. 31 et seq.
66
E.g.: Koutrakos (2010), p. 485.
67
Annex B to Council Decision 2002/628/EC [2002] OJ L 201/48; See similarly the Statute of the
International Renewable Energy Agency, which in this regard states: ‘In their instruments of
ratification or accession, the organisations referred to above shall declare the extent of their
competence with respect to the matters governed by this Statute.’ Council Decision of 24 June
2010 on the conclusion of the Statute of the International Renewable Energy Agency (IRENA) by
the European Union (2010/385/EU), Art. 6 (C).
68
Bjoerklund (2001), p. 380.
69
Court of Justice of EU, Case C-459/03, Commission v Ireland (MOX Plant) [2006] ECR I-4635,
note 26, para. 94 et seq.; Cf.: Koutrakos (2010), p. 482, arguing that the Court’s case law
demonstrates that implied external powers may exist even when they are not EU exclusive; On
a general analysis on the implied competence, see also: Schütze (2004), p. 229 et seq (The Court in
ERTA ‘speaks the language of teleological interpretation under the aegis of the effect utile logic: in
the light of the general scheme of the Treaty, the Community’s power to adopt “any other
258 7 Inter-Party Mechanism and the EU: Possible Implications from the. . .

agreements [. . .] even if the specific matters covered by those agreements are not
yet, or are only very partially, the subject of rules at Community level, which, by
reason of that fact, are not likely to be affected.’70 The act of accession into such
mixed agreement produces as such the effect of EU electing to exercise its external
competence71 on something the Member States are also agreeing to.
This being said, this test is passed only if the provisions of the mixed agreement
which have been submitted to an external dispute settlement method are not found
in EU primary or secondary law and there was no external competence elected from
the EU on basis of domestic competences to exercise the competence residing in the
contested provisions of the mixed agreement.72 Should there appear that the EU has
elected to exercise an external competence, based either on primary or secondary
legislation, when concluding a mixed agreement, and that EU-attributed compe-
tence is being adjudicated by the external method of dispute settlement, certainly
the test is not passed and the external procedure of dispute settlement is considered
to violate Art 344 TFEU. In specific terms, ‘[a]s mixed agreements are by their
nature externally focused, they are a major feature of those aspects of external
relations over which the EC has competence.’73 The mere fact that the provisions of
the mixed agreement submitted before the external tribunal ‘come within the scope
of [. . .EU] competence which the Community exercised by acceding to [. . .that
mixed agreement], with the result that those provisions form an integral part of the
Community legal order [.]’,74 violate the EU Court’s exclusive jurisdiction to
observe EU law comprising also of that mixed agreement. The question whether
such competence is exclusive for the EU is less important for Mox Plant, adding
that the mere fact that a competence becomes under the control of an external
method of settlement suffices for it to violate Art 344 TFEU. This being said, the
EU Court has an exclusive competence to rule on Art 344 TFEU, whereas Art

appropriate provision” to give effect to the Community’s transport policy objectives must be
interpreted, in the perception of the Court, as including the legal power to enter international
agreements.’ Ibid., p. 231); See also on this: Tridimas and Eeckhout (1995).
70
Court of Justice of EU, Case C-459/03, Commission v Ireland (MOX Plant) [2006] ECR I-4635,
note 26, para. 95.
71
By contrast, in Opinion 2/91 the Court has ruled that: ‘In any event, although, under the ILO
Constitution, the Community cannot itself conclude Convention No 170, its external competence
may, if necessary, be exercised through the medium of the Member States acting jointly in the
Community’s interest.’ See: Court of Justice of EU, Opinion 2/91 (Convention No. 170 ILO on
safety . . .) [1993] ECR I-1061, para 5; This shows that the EU Court would not nevertheless bar
EU Member States from exercising its competences in cases when the international treaty—
because of its organizational nature—does not make this possible for the EU itself. In this scenario,
it should be assumed that although Member States are practically exercising that competence it is
the EU which has elected and to which that competence belongs.
72
See e.g.: Churchill and Scott (2004), p. 664, on the same issue on the Mox Plant case.
73
Cardwell and French (2007), p. 123.
74
Court of Justice of EU, Case C-459/03, Commission v Ireland (MOX Plant) [2006] ECR I-4635,
note 26, para. 126.
7.6 Inter-Party Cases in Light of Mox Plant and Art 344 TFEU: Which. . . 259

344 TFEU applies both in cases when the competence is exclusive for the EU and
when it is shared.

7.6.2 The Second Test: Parallel (Mirrored) Provisions

The second test includes assessing and identifying whether parallel provisions exist
both in EU law and the mix agreement, and if yes, do parallel provisions give the
right to the Member State to submit a complaint from such provision(s) to an
external method of dispute settlement with the argument that the provisions the
Member State is being based is that of the mixed agreement and not of the EU law.
In Mox Plant, Ireland argued that although it used as supplementary tools the texts
of an EU Directive,75 it could not be taken as a basis to conclude that Ireland had
asked the Arbitral Tribunal to interpret EU law. The Court answered by ruling that
Ireland had used many of the EU Directive’s norms as means to support its
interpretation before the Arbitral Tribunal, therefore pulling the interpretation of
an EU Directive to an external means of dispute settlement.76
What in fact Ireland tried to argue was that basing its arguments and claims in an
instrument of international law (mixed agreement), adding that such similar norms
also exist somewhere in EU law,77 would not prove Ireland’s standing as bringing
the EU law to an external mechanism of dispute settlement but rather would prove
Ireland supporting its claims on basis of the agreement to which Ireland is a
contracting party. The Court’s view on this point is rather broad and implies the
duty of every Member State not to bring issues of international agreement wherein
they are parties if such claims are also based somewhere in the interpretations of EU
law either in terms of the mirrored substance or in terms of the fact that such
agreements form the law of the Union as well.78 With this in mind, the test is
positive if a Member State uses means of interpretation before an external dispute
settlement mechanism which are similar to those written somewhere in the EU
law.79 The mere act of states in this context, according to the Court, produces the
effect of pulling the EU law to an external mechanism of dispute settlement,
therefore violating its exclusive jurisdiction. Therefore, Member States cannot

75
See e.g.: Marsden (2009), p. 316 et seq.
76
Court of Justice of EU, Case C-459/03, Commission v Ireland (MOX Plant) [2006] ECR I-4635,
note 26, para. 112 et seq.; See also: Churchill and Scott (2004), pp. 658 et seq; Lavranos (2008),
p. 582 (‘The ECJ concluded that the matters covered by the provisions of UNCLOS relied upon by
Ireland before the arbitral tribunal are “very largely” regulated by Community law.’).
77
Cf.: Churchill and Scott (2004), p. 664 et seq, on how this was interpreted in Mox Plant.
78
On the Mox Plant case, see e.g.: Churchill and Scott (2004), p. 674; See also: Cardwell and
French (2007), p. 122/3.
79
On the ‘mirrored provisions’ concept on the Mox Plant, see: Cardwell and French (2007),
p. 122; Cf.: Court of Justice of EU, Opinion 1/91 [1991] E.C.R. I-6079, para. 39 et seq, for
similarities with this argument.
260 7 Inter-Party Mechanism and the EU: Possible Implications from the. . .

claim that they were based in the interpretation of international agreements, if such
interpretations are also found somewhere in the EU law, when they submit claims to
external methods of dispute settlement. For instance, in Opinion 1/03, the Luxem-
bourg Court had argued that ‘[. . .] account must be taken not only of the area
covered by the Community rules and by the provisions of the agreement envisaged,
insofar as the latter are known, but also of the nature and content of those rules and
those provisions, to ensure that the agreement is not capable of undermining the
uniform and consistent application of the Community rules and the proper func-
tioning of the system which they establish.’80 This test on the Lugano Convention
seemed to make the EU Court going as far as taking it for granted that—outside the
competence debate—it has the power to review an agreement if it simply may
undermine the effectiveness of the EU law system.

7.6.3 The Third Test: It Is for the Luxembourg Court


to Delineate Its External Jurisdiction Borders
in Each Case

Let one suppose that a case based merely on parts of an international mixed
agreement that are an exclusive competence of the applicant Member State81 is
submitted before an external tribunal for dispute settlement against another Mem-
ber State. The question is would the mere fact that the application of the Member
State does involve an issue which is absolutely of Member State exclusive compe-
tence suffice for it not to be in violation of the Luxembourg Court’s exclusive
jurisdiction. The Court has given a rather ironical and self-ruling answer to this
issue, setting out that:
Without it being necessary to rule on the question as to whether [. . .] the dispute comes
within the scope of [. . .EU] law, suffice it to hold that [. . .] a significant part of the dispute
in this case between [. . . the two Member States] relates to the interpretation or application
of [. . .EU] law. It is for the Court, should the need arise, to identify the elements of the
dispute which relate to provisions of the international agreement in question which fall
outside its jurisdiction.82

80
Court of Justice, Opinion 1/03 (Lugano Convention), [2006] ECR I-1145, para. 133.
81
For instance, the Luxembourg Court has ruled in Parliament v Council and Commission that if
the EU does not have exclusive competence in a certain policy field then ‘the Member States are
not precluded from exercising their competence in that regard collectively in the Council or
outside it.’ Court of Justice of EU, Parliament v Council and Commission, Joined Cases C-181/
91 and C-248/91 [1993] ECR I-3685, para. 16; The latter standard, however, does not render any
importance when it comes to the external borders of jurisdiction with international tribunals, as,
there, the EU Court would still insist to have itself on the front.
82
Court of Justice of EU, Case C-459/03, Commission v Ireland (MOX Plant) [2006] ECR I-4635,
note 26, para 135.
7.6 Inter-Party Cases in Light of Mox Plant and Art 344 TFEU: Which. . . 261

The Court clearly ruled that although in a certain dispute the applicant Member
State claims that the dispute submitted to the external method of settlement is of a
nature that does absolutely not cover any element of EU law, explicitly or implic-
itly, still such an argument is to be done by the Court itself and could not be
accepted that it be taken for granted by the Member State’s claim. This being said,
the Court has made it clear that it wants each case, whatsoever the level of the
dispute or its base on EU law, to come first to it for qualification on whether it
touches its jurisdiction.83 Only after such a test has been performed by the Court,
supposing that the latter decided that the case passes the test positively, could the
Member State submit the part of the dispute not relating at all to EU law to an
external method of dispute settlement. The mere fact that there exists a substantive
parallel rule in EU law, in whichever form it is construed, that does not allow the
Member State to depart from the prohibition of Art 344 TFEU.
This requirement put by Mox Plant does seek that whatever the nature of the
disputes between the EU Member States in a case involving a mixed agreement, it is
necessary to bring the case first to the Luxembourg Court to check whether it
involves an issue of EU law, and if not partially, which parts of the dispute do not
touch upon the EU law and could be allowed to be submitted to another means of
dispute settlement. This being the case, this standard established by Mox Plant
serves as a type of the duty for prior involvement of the Luxembourg Court on any
case involving dispute between two or more EU Member States on a mixed
international agreement. Such requirement is also substantiated within the context
of the need to observe the application of Art 344 TFEU.

7.6.4 The Fourth Test: Use of Assurances Not Allowed

The fourth issue is whether the guarantee that a Member State could put such as
assuring the Luxembourg Court that it has not called on the external mechanism of
dispute settlement to interpret the EU law seems not sufficient to fulfill the
requirements of Art 344 TFEU. Mox Plant made it clear that although a Member
States gives written assurances that the external tribunal will not, or has been asked
not to, engage with the examination of EU law, that would not suffice and is not
relevant in the context of protecting the autonomy of EU legal order. In Mox Plant,
the Court also ruled that:
That risk exists even though, as Ireland avers, it has given a formal assurance that it has not
called on, and will not call on, the Arbitral Tribunal to examine or appraise, pursuant to

83
Accord.: Court of Justice of EU, Parfums Christian Dior SA and Tuk Consultancy BV and Assco
Ger€uste GmbH, Rob van Dijk, and Wilhelm Layher GmbH & Co. KG, Joined Cases 300/98 and
392/98 [2000] ECR I-11307, paras. 35 [emphasis added] (‘. . . where a provision [. . .of an
international agreement. . .] can apply both to situations falling within the scope of national law
and to situations falling within that of Community law, as is the case in the field of trademarks, the
Court has jurisdiction to interpret it in order to forestall future differences of interpretation.’).
262 7 Inter-Party Mechanism and the EU: Possible Implications from the. . .

Article 293 of the Convention or any other provision, whether the United Kingdom has
breached any rule of Community law.84

Such assurances are something which the Court considers absolutely not a
pledge to protect Art 344 TFEU integrity, and the overall EU legal order autonomy.
The issue of assurance that a Member State gives to the EU Court that, e.g. the
Strasbourg Court will not engage with the review of EU law, does not seem to fulfill
the requirements of the EU Court for a fully-fledged applicable Art 344 TFEU and
its jurisdiction to observe the application of the latter. Such assurance could have
made sense in light of the Doctrine of Equivalent Protection developed e.g. in
Bosphorus.85
The assurance that a Member State could make to the Luxembourg Court,
regarding the Strasbourg’s inter-party procedure, is that the latter will not interpret
EU law as it has already put in place the Bosphorus Doctrine of Equivalent
Protection on basis of which it does not consider itself authorized to enter the
borders of jurisdictional exclusivity of the EU Court due to the fact that the latter
offers equivalent protection with the Convention system. Certainly, although the
Bosphorus standard serves as a very good example of how the Strasbourg Court
would behave, it still how not suffice to fulfill the standard of Mox Plant that the EU
Court have full jurisdiction to deal with any issue involving an EU competence, be
it shared or exclusive. This said, such assurances from any Member State would not
pass the test of being allowed to be pulled outside the Luxembourg Court’s
observance of issues between member states inter se, although the three previous
tests could have been passed successfully.

7.7 Beyond the Conventional Concept on Competing


Jurisdictions of the Two European Courts:
‘Reconciling’ Art 55 ECHR with Art 344 TFEU

Although the Luxembourg Court seems allergic to accepting the fact that there is in
place an inter-party mechanism like this one in the DAA—something which it ruled
in Opinion 2/13 (para. 213) should leave no jurisdiction ratione materie to the
Strasbourg Court—it is still argued that possibilities for reconciling these two
orders do exist. To this extent, Gragl notes that the Steering Committee for
Human Rights (CDDH) has proposed that there was no need to exclude the EU
from the inter-party mechanism, as that was basically an internal issue for the EU
and its Member States which could be regulated through an internal agreement inter
se.86 To start, it is worth noting that both treaty regimes, the EU and the Convention,

84
Court of Justice of EU, Case C-459/03, Commission v Ireland (MOX Plant) [2006] ECR I-4635,
note 26, para. 155.
85
On the Doctrine, see e.g.: Miiller (2007), p. 43.
86
Gragl (2013), p. 183.
7.7 Beyond the Conventional Concept on Competing Jurisdictions of the Two. . . 263

as commonly recognized, have specific provisions assigning the exclusive jurisdic-


tion for dispute settlements arising out of them. In principle, such provisions have
pure conflicting stances in the context of EU as a High Contracting Party of the
Convention. As explained above, the two conflicting provisions as regards inter-
party cases once EU accedes to the Convention will be Art 55 ECHR and Art
344 TFEU, both of them providing exclusive jurisdictions to their respective
courts87 regarding dispute settlement between member states of such treaty
regimes. In order to examine this context of conflict between the two European
courts, it is important to shed light first on the Convention’s jurisdictional clauses
regarding dispute settlements deriving from its parties.
The Convention in Art 55 explicitly reads:
The High Contracting Parties agree that, except by special agreement, they will not avail
themselves of treaties, conventions or declarations in force between them for the purpose of
submitting, by way of petition, a dispute arising out of the interpretation or application of
this Convention to a means of settlement other than those provided for in this Convention.

In principle, Art 55 ECHR seems in direct conflict with Art 344 TFEU, as the
latter provides an exclusive jurisdiction to the Luxembourg Court.88 A core issue
that one could witness in this regard is that in principle Art 55 ECHR is made to
ensure that the interpretation of the Convention by two or more parties in a dispute
should not get beyond the doors of the Strasbourg Court. On the other hand, Art
344 TFEU puts the exclusive jurisdiction of the Luxembourg Court only on
disputes between its parties arising from the interpretation or application of the
EU Treaties. This said, in principle, Art 55 ECHR and Art 344 TFEU have not
diverging or conflicting positions. However, as explained above, upon accession of
the EU to the Convention, the latter becomes part of the EU municipal law, as ruled
in Demirel.89 This being the problem,90 upon accession to the Convention, the latter
will become part of the law to be observed by the Luxembourg Court in line with
the exclusive jurisdiction of Art 19 (1) TEU.91

87
Gragl arrives at the same general conclusion, although with a rather different direction of
argument. See: Gragl (2013), p. 178; See also a ‘proposal’ for an institutionalized system of
relationship between the two courts’ jurisdiction, at: Greer and Williams (2009), p. 463 et seq.
88
Gragl rightly argues that the drafters of the DAA did not intend to put a reservation on the issue
of inter-party jurisdiction of the Strasbourg Court on EU-related parties. With that decision,
implicitly understood, they basically agreed to conclude an internal ‘special agreement’ that refers
directly to the Convention to exclude the inter-party jurisdiction of the Strasbourg Court. See:
Gragl (2013), p. 184.
89
Court of Justice of EU, Demirel, Case 12/86 [1987] ECR 3719, para. 11; On Demirel, see also a
broad analysis, at: Weiler (1992), p. 72 et seq.
90
See some general criticisms on the way the Luxembourg Court has extended its jurisdiction on
the interpretation of international agreements which are considered part of EU law, at: Hartley
(1999), p. 31 et seq.
91
See e.g.: Cardwell and French (2007), p. 123, on the Mox Plant example; See also: Gragl
(2013), p. 179.
264 7 Inter-Party Mechanism and the EU: Possible Implications from the. . .

With the Convention as part of the domestic law observed by the Luxembourg
Court, Art 55 ECHR becomes directly in conflict with Art 344 TFEU which
considers the interpretation and application of the Convention an issue of exclusive
jurisdiction provided for by the means under the Treaties. The question that remains
therefore is whether this conflict is manageable and reconcilable? To answer this
question, an in-depth analysis of Art 55 ECHR is necessary in order to deconstruct
the limits of the Convention’s exclusive mechanisms for dispute settlement
between its parties.
To make a short digression, in principle, locking EU to its internal Art.
344 TFEU would violate its international obligations under the Convention to
allow the implementation of Art. 55 ECHR (something which, nonetheless, Opin-
ion 2/13 tries to clarify from the very beginning). The EU Court may post-accession
argue that it had not considered carefully the DAA, therefore—as they do under-
stand it better—there appear new circumstances on basis of which the EU might
wish to implement its Art. 344 TFEU even if the Convention system would insist in
allowing EU Member States in making use of Art. 55 ECHR. One may legitimately
ask whether this ‘new understanding’ on the insistence of the EU to respect its
internal-law obligation may serve as ground to release it under the DAA and
Convention’s Art. 55? Art. 62.3 VCLTSIOIO prescribes: ‘A fundamental change
of circumstances may not be invoked as a ground for terminating or withdrawing
from a treaty if the fundamental change is the result of a breach by the party
invoking it either of an obligation under the treaty or of any other international
obligation owed to any other party to the treaty.’92 Therefore, there is a clear ‘no’.
Coming back to the core point, Art 55 ECHR does clearly state that High
Contracting Parties will not submit to any other dispute settlement mechanism
any issue of interpretation or application of the Convention other than the means of
settlement provided for in the Convention. The only exception to this rule would be
if there is a special agreement between High Contracting Parties to submit this
jurisdiction to another method of dispute settlement. The special agreement men-
tioned in the Convention stands also in line with the stipulation mentioned in Art.
41.1 VCLTSIOIO: ‘Two or more of the parties to a multilateral treaty may conclude
an agreement to modify the treaty as between themselves alone if: (a) The possi-
bility of such a modification is provided for by the treaty.’ These two parties are the
EU, on the one side, and its Member States, on the other side. The notion of the
special agreement mentioned in Art 55 ECHR needs in-depth assessment.
In principle, the special agreement which Art 55 ECHR93 provides the possibil-
ity for is something that could derogate the authority of the Strasbourg Court with
regard to the exclusivity of interpreting the Convention when it comes to inter-party

92
See also: Draft articles on the law of treaties between States and international organizations or
between international organizations with commentaries 1982, p. 59 et seq.
93
Gragl in this regard argues that Art. 55 ECHR seems to be far more flexible on the exclusive
jurisdiction of the Strasbourg Court compared to Art. 344 TFEU. Such nature of this provision
allows for more coherence and harmony between the two legal regimes. See: Gragl (2013), p. 179.
7.7 Beyond the Conventional Concept on Competing Jurisdictions of the Two. . . 265

complaints. The special agreement therefore would assign the competence to


resolute disputes between the Convention’s High Contracting Parties to another
international dispute-settlement mechanism to interpret the Convention. This said,
the special agreement clause allows that other international mechanisms interpret
the Convention, which directly forms room to interfere or even exclude the auton-
omy of the Strasbourg Court on the subject of inter-party disputes jurisdiction. Of
course, all contracting parties to the Convention must assent that, otherwise the
special agreement would not be considered established. The special agreement
clause portrays the Convention system as far more elastic than the EU Treaties’
system, as the latter allows for no means or ways to interfere to its own Court’s
exclusive jurisdiction.

7.7.1 Conditions for the Special Agreement: What About


More Details?

In light of the words of Art 55 ECHR, the special agreement needs to be a sort of
international accord which all high contracting parties consent to. Its nature must
logically be meant as written in form. The question would be, however, which are
the attributes of an agreement that would make it a ‘special agreement’ under Art
55 ECHR? It is argued that along the line, there exists no description of the qualities
that such special agreement should enjoy. However, logically interpreted, I argue
that such special agreement should at least fulfill these four conditions:
a) The first condition is that all High Contracting Parties to the Convention agree/
consent to such special agreement. Frowein maintains also that a special agree-
ment should have been assented by all contracting parties to the Convention.94
All parties would mean not only the EU Member States as part of the Conven-
tion, but also all other Council of Europe non-EU Member States.95 The nature
of the special agreement, in this regard, seems to be one that needs unanimous
approval by all High Contracting Parties to the Convention,96 although Gragl

94
See Frowein cited at: Gragl (2013), p. 180 (footnote 40).
95
Cf: Analysis (1997), p. 239/40, which questions the suitability if such an agreement could also
be made with a smaller group of contracting parties.
96
Contra: Gragl (2013), p. 180, who maintains, referring to Cyprus v. Turkey, that such an
agreement may also be formed if assented merely a by smaller number of contracting parties.
The Strasbourg Court had ruled in Cyprus v. Turkey that: ‘The Commission concludes that it is not
prevented from examining the present application on the ground that there exists a “special
agreement” to this effect between the two High Contracting Parties concerned. The Commission
would add that, generally speaking, the performance of its functions under Article 19 (Art. 19) of
the Convention cannot in any way be impeded by the fact that certain aspects of the situation
underlying an application filed with it are being dealt with, from a different angle, by other
international bodies.’ (ECtHR, Cyprus v. Turkey, Application no. 25781/94, Decision on the
Admissibility of 28 June 1996, part. III). Reading this finding of the Court carefully, one may
266 7 Inter-Party Mechanism and the EU: Possible Implications from the. . .

and Lock insist that there may be derogation to this rule on basis of the overall
view of Cyprus v. Turkey.97
b) The second condition is the ‘special’ nature of the Agreement, which should
normally be reflected in two directions. First, the agreement should delegate this
dispute-settlement authority to another instrument of international law, which is
normally a mechanism capable of resolving disputes between the parties to the
Convention. And, second, that the agreement must tackle specifically the Con-
vention in its substance, something Gragl agrees as well from a general per-
spective.98 That said, the agreement concerned would not normally be accepted
as existing if it does not specifically refer to the Convention’s interpretation
outside the jurisdiction of the Strasbourg Court. In that way, the agreement could
not be considered as existing between the contracting parties of the Convention
if its object is not specifically the Convention’s Art 55 derogation.
c) The third ordinary condition is that the special agreement must have been
reached in line with the Vienna Convention on the Law of Treaties rules and
principles, wherein contracting parties have fulfilled their constitutional will to
become part of the agreement on basis of their domestic legal order’s ratification
procedures.
d) The fourth condition is that the special agreement could allow only Convention
contracting parties to submit disputes over the Convention’s interpretation to the

not gain the impression that the Court fully agrees that the ‘special agreement’ be allowed also
between a smaller number of high contracting parties, and that for two reasons: first, in the above-
mentioned conclusion, the Court does not say it explicitly that such ‘special agreement’ may be
made by a smaller number of parties in the form of an agreement that excludes the applicability of
Art. 55 ECHR; second, the Court basically only refers to the applicant’s argument, and tries to
negatively reply by basically saying that even though there may be such an agreement—which it
does not say whether there is one and on basis of which conditions that may be formed—it does not
take that as a basis for derogating from its jurisdiction, therefore calling the case admissibly.
Should the Court have considered this ‘special agreement’ as eligible, it would have of course
called the case inadmissible. However, it did not for the mere reason that it considered such
‘special agreement’ as not qualified to stand as derogation for the Court’s inter-state jurisdiction.
97
Lock (2009), p. 392/3; Lock’s point, which is also recalled by Gragl, that the travaux
preparatoires confirm the argument that the ‘special agreement’ may be made also by a smaller
number of parties does not seem fully convincing due to these reasons: first, the travaux
preparatoires merely demonstrate that there was a proposal from the Swedish Delegation for
the ‘special agreement’ to mean that it may be made between a smaller number of parties. There is
no confirmation in the travaux preparatoires that such proposal was voted as based in the meaning
which the Swedish Delegation had proposed to give to it. The mere fact that there was no debate on
that proposal, and no other delegation proposing something else, is insufficient to take it for
granted that all parties agreed with the proposed meaning of the Swedish Delegation; second, Lock
himself (Ibid, p. 392/3) notes that the proposal of the Swedish Delegation was slightly changed
when it got voted by other delegations, which basically means that no one took it for granted that
the original meaning of the Swedish-version proposal for current Art. 55 ECHR was the final one.
98
Gragl argues generally in the same line, by maintaining that the inter-state jurisdiction for the
Strasbourg Court may allow for derogations. However, as those derogation need to be narrowly
construed, the Convention should logically be mentioned and referred to in that ‘special agree-
ment’. See: Gragl (2013), p. 180–182.
7.7 Beyond the Conventional Concept on Competing Jurisdictions of the Two. . . 267

mechanism of dispute-settlement foreseen thereupon, and that the dispute-


settlement mechanism be able to adjudicate upon all Convention contracting
parties’ inter-state complaints.
In this context, it is generally argued that the qualification of an agreement as a
special agreement in light of Art 55 ECHR is rather complex,99 with certain
conditions needing to be fulfilled. To note, the Luxembourg Court in Opinion
2/13 does not tackle this perspective at all.

7.7.2 Questioning Whether EU Treaties Could Serve


as a Special Agreement in Light of Art 55 ECHR?

Given that Art 55 ECHR requests that the exclusive jurisdiction of the Strasbourg
Court be delegated to another international mechanism only with a special agree-
ment, one could ask whether the EU Treaties, namely Art 19 (1) TEU and Art
344 TFEU,100 could be considered such an agreement for the purposes of EU
relations towards the Convention.101 To these meet the criteria, one needs to

99
Gragl comes immediately to the conclusion that, due to the fact that the Convention is not
mentioned specifically in this special agreement, there seem little or no chances of reconciling the
exclusive jurisdictions on inter-party issues of the Luxembourg and Strasbourg Court. See: Gragl
(2013), p. 182.
100
Gragl follows a shorter path to making this observation. He, at first and from a very general
perspective, notes that it would not be possible to regard these two bases as ‘special agreement’ in
light of Art. 55 ECHR for two basic reasons: first, because these two bases do not mention ECHR,
and, second, because not all parties to the ECHR have assented that special agreement. However,
he goes further to argue that, referring to Lock, such two bases have been drafted too generally to
be able to be regarded as referring to a ‘special agreement’ in light of Art. 55 ECHR. See Gragl’s
chapter on inter-party jurisdiction under the DAA: Gragl (2013), pp. 174 et seq.
101
Gragl proposes that in this kind of normative conflict one may apply the ‘disconnection clause’,
which basically means that if the contracting parties to an international agreement are EU
members, they will apply EU law inter se to the exclusion of the law of that international
agreement. Referring to Klabbers, Gragl proposes that the law of that international agreement
may be decided to be applied only in the external relations of the EU and its members. See: Gragl
(2013), p. 189. Although the idea on the ‘disconnection clause’ may sound well in terms of the
inter-party jurisdiction, it may however lead to the wrong direction, as the Luxembourg Court may
wish to see the same clause applied also to the individual complaint jurisdiction with regard to
human rights protection under the Strasbourg system. That would render the entire context of EU
accession to ECHR moot and practically without any taste of relevance; In addition, Gragl (Ibid,
p. 189) goes further by arguing that Art. 5 DAA serves the idea of the ‘disconnection clause’. I
fully disagree with this. Art. 5 DAA does merely qualify the nature of the Strasbourg Court
proceedings as subsidiary to the EU law judicature: it does in no way try to say that the
Convention’s inter-party mechanism is being replaced with the EU law inter-state mechanism.
Such intention may be observed nowhere in that provision. This basically means that Gragl is not
right with the argument that Art. 5 DAA provides for a ‘disconnection clause’ that excludes the
Convention law.
268 7 Inter-Party Mechanism and the EU: Possible Implications from the. . .

examine the EU Treaties against the four criteria mentioned above, adding that the
EU Treaties were ratified after the Convention, and therefore the idea as such could
seem generally sound.
To test whether the EU Treaties, which were ratified as international agreements
by the EU Member States being at the same time as Convention parties, one needs
to receive some help from the Vienna Convention on the Law of Treaties. For this
specific relationship between the EU Treaties (Art 19(1) TEU & Art 344 TFEU)
with Art 55 ECHR, one needs to refer to Art 31.2 VCLT, which reads:
The context for the purpose of the interpretation of a treaty shall comprise, in addition to the
text, including its preamble and annexes: [. . .]any instrument which was made by one or
more parties in connection with the conclusion of the treaty and accepted by the other
parties as an instrument related to the treaty.102

With the Convention being the primary treaty in this regard, the question is
whether the EU Treaties could be considered an instrument which was made by at
least some parties in connection to the conclusion of the Convention and accepted
by other parties as such. There seems some room for considering the EU Treaties as
interpreting tools which could make the ‘special agreement’ of Art 55 ECHR, as the
EU Treaties were ratified only by some of the Convention states. The question
would remain here whether the non-EU Convention parties considered the EU
Treaties as having that effect. It would be hard to trust that non-EU Convention
parties have ever been aware that they are granting this ‘special’ agreement to the
EU, adding that there has never been such decision even implicitly put forward by
any Convention party as regards inter-state complaint mechanism. This being the
case, it logically follows that the EU Treaties could not be considered an instrument
which has been made to add upon the Convention’s Art 55, with non-EU parties
never having agreed neither formally nor indirectly to such a development as by Art
31.2 VCLT.
Of significant importance in this regard, and a more suitable basis to consider
whether the EU Treaties could be qualified as a special agreement, is also Art 31.3
VCLT, which reads:
There shall be taken into account together with the context: (a) any subsequent agreement
between the parties regarding the interpretation of the treaty or the application of its
provisions; (b) any subsequent practice in the application of the treaty which establishes
the agreement of the parties regarding its interpretation;[. . .]103

In light of this provision, the EU Treaties could be qualified either as a subse-


quent agreement or any subsequent practice in terms of the Convention’s Art
55 application. Certainly, the EU Treaties could in no way be qualified as subse-
quent agreement for the mere fact that they were not attached or submitted to the
Convention. This perspective of understanding is also supported by the Official
Commentary to VCLT, which states that in this regard ‘[t]he principle on which this

102
VCLT Art 31, 2(b).
103
VCLT Art 31, 3 (a, b).
7.7 Beyond the Conventional Concept on Competing Jurisdictions of the Two. . . 269

provision is based is that a unilateral document cannot be regarded as forming part


of the “context” within the meaning [. . .the cited VCLT article. . .] unless not only
was it made in connexion with the conclusion of the treaty but its relation to the
treaty was accepted in the same manner by the other parties.’104 Clearly, EU
Treaties may not be deemed of such a nature. However, the question remains
whether there was any subsequent practice which led into the understanding that
there is a ‘special agreement’ regarding Art 55 ECHR. To utilize this point of
VCLT, one could refer to the Bosphorus case wherein the Strasbourg Court,
although not within an inter-state complaint procedure, recognized the fact that
there exists a practice wherein the exclusive jurisdiction of the EU Court should one
way or another be maintained. With that reading in mind, a practice of recognizing
the ‘special’ nature of the Luxembourg Court’s exclusive jurisdiction was
established by the Convention Court itself. The Bosphorus appraisal seems to
support the fact that the practice as by VCLT Art 31.3 emerged as being one of
the Convention’s case-law principles in relation to the EU judicial system.
From the four points of VCLT which were tested against Art 55 ECHR, only the
‘practice’ remains a point which could possibly rationalize the EU Treaties as being
the ‘special agreement’ under Art 55 ECHR. To clean further the ground of the
argument, one needs to look a bit deeper to Art 31.1 VCLT. Art 31.1 VCLT sets that
the meaning of a treaty should be established on basis of its ‘object and purpose’.105
The object and purpose of a treaty’s meaning should be sought somewhere in its text.
To this purpose, the ‘object and purposes’ of the EU Treaties, or the practice installed
with Bosphorus, should have searched to constitute a special agreement in light of Art
55 ECHR. It is clearly argued that neither the EU Treaties nor the practice established
with Bosphorus have had an explicit object and purpose of constituting a special
agreement in light of Art 55 ECHR. The implicit meaning would however be that
both the EU Treaties and the Bosphorus recognition have led to the understanding
that there is already a common practice which respects the Luxembourg Court’s
exclusive jurisdiction on matters related to the EU law. However, this practice was
never made direct towards the Convention as part of the domestic EU law. What has
been the general observance was the practice of recognizing a sort of exclusivity of
EU Court to deal with dispute settlements between the EU member states within the
EU-originated law, not necessarily covering the Convention as an international treaty
domesticated by the Treaties. On the other hand, it is clearly understood within this
context that there was never one single moment when Convention High Contracting
Parties explicitly agreed to form a special agreement to delegate the dispute-
settlement mechanism from Art 55 ECHR to the EU Court: the fact is that the
delegation of that authority could have happened in the light of delegating this entire

104
Draft Articles On The Law Of Treaties With Commentaries 1966. Text adopted by the
International Law Commission at its eighteenth session, in 1966, and submitted to the General
Assembly as a part of the Commission’s report covering the work of that session (at para. 38). The
report, which also contains commentaries on the draft articles, appears in Yearbook of the
International Law Commission, 1966, vol. II, p. 221.
105
See on this in general: Crnic-Grotic (1997); See also: Villiger (2009), p. 428.
270 7 Inter-Party Mechanism and the EU: Possible Implications from the. . .

mechanism, for all Convention parties, to the EU Court. For sure, there was never a
practice which allowed the understanding that the transfer of Art 55 ECHR authority
to EU Treaties would cover all Convention parties, as the EU Treaties provide no
jurisdiction and no logical reasoning for its Court to have jurisdiction on inter-state
cases outside the realm of EU membership. This argued, it is hard to conclude that the
EU Treaties or the practice recognized by Bosphorus have led to any identification of
a ‘special agreement’ under Art 55 ECHR intended to mean that the Convention
parties agree to Strasbourg Court losing its exclusive jurisdiction on dispute settle-
ment between themselves.
Above all, checked against the more general criteria mentioned at the very
beginning, the EU Treaties seem not to have had the intention and objective of
constituting a special agreement in light of Art 55 ECHR. In addition, the EU
Treaties could not be taken as a basis to conclude that all Convention High
Contracting Parties have wished to delegate that authority to the EU Court as a
whole. This said, it is not normal to consider EU Treaties as forming a ‘special
agreement’ in light of Art 55 ECHR wherein there was a transfer of authority for
inter-state cases from the Convention to the EU Court. The core rationale behind
this argument is that the EU Court has no jurisdiction and there is no logical practice
that could put it on a position to accept a transfer of authority to rule on inter-state
cases for non-EU Convention contracting parties.106 As such, EU Treaties seem not
to be able to serve as a ‘special agreement’ between Convention parties in terms of
Art 55 ECHR. This being the case, there seems that a jurisdictional conflict between
the Luxembourg Court and Strasbourg Court is inherent in the type of allergic
approaches that these two follow and the treaty regimes they are based upon.

7.7.3 Accession Agreement as a ‘Special Agreement’ Under


Art 55 ECHR? A Second Try. . .

The second option to consider in light of Art 55 ECHR is questioning whether the
EU Accession Agreement to the Convention could be considered a ‘special agree-
ment’ in light of the confer of authority to Luxembourg Court to deal with inter-
state disputes. For this purposes, Art 5 of the DAA reads:
Proceedings before the Court of Justice of the European Union shall be understood as
constituting neither procedures of international investigation or settlement within the
meaning of Article 35, paragraph 2.b, of the Convention, nor means of dispute settlement
within the meaning of Article 55 of the Convention.107

106
Cf.: Marsden (2009), p. 312, explaining how this was refused in Mox Plant.
107
Art 5 DAA; Cf.: Gragl basically argues that this provision as such does not bar the function and
operation of Art. 344 TFEU. See: Gragl (2013), p. 186. Gragl may be right in the sense that this
provision merely serves the Convention system, but it does not in principle define the meaning of
dispute settlement under EU law and Art. 344 TFEU scope.
7.7 Beyond the Conventional Concept on Competing Jurisdictions of the Two. . . 271

There seem to be two separate issues that the Art 5 of the DAA seems to
establish. First, the relationship between the EU Treaties and Art 35.2.b of the
Convention, and second, the relationship between the EU Treaties and Art 55 of the
Convention. Art 33.2.b and Art 55 of the Convention have been devised to protect
slightly the margins of the Convention system, therefore their adaptation to the
Treaties was a compulsory need.
To start with the first clause of Art 5 of the DAA, it states that cases filed before
the EU Court ‘shall be understood as constituting neither procedures of interna-
tional investigation or settlement within the meaning of Article 35, paragraph 2.b
[. . .]’. Art 35.2.b ECHR, on the other hand, reads:
The Court shall not deal with any application submitted under Article 34 that [. . .](b) is
substantially the same as a matter that [. . .] has already been submitted to another procedure
of international investigation or settlement and contains no relevant new information.108

Art 35.2.b of the Convention clearly refers to Art 34 which covers individual
application. At the first site, Art 35.2.b does not deal with inter-state complaints.
However, if one considers this issue a bit deeper the argument would be that even in
inter-state cases the state submitting the application must prove that the legal
remedies of the respondent state have been exhausted.109 That said, the first clause
of Art 5 of the DAA does clearly preclude applications from being prohibited to be
filed at Strasbourg if they have already gone through Luxembourg.110 In this regard,
that clause reconciles the Strasbourg’s jurisdictional exclusivity with that of
Luxembourg.
On the other hand, the second more important clause of Art 5 in the DAA clearly
states that proceedings at the EU Court would not be considered as a dispute
settlement means in light of Art 55 ECHR. In that context, even though a case
might have been filed at Luxembourg’s inter-state complaint procedure (infringe-
ment proceedings), such case would not be considered as interfering to the
Strasbourg’s exclusive jurisdiction under Art 55 ECHR. However, the question
remains whether the inter-state case that has gone through the EU Court complaint
mechanism could thereafter be submitted before Strasbourg Court under Art 55 of
the Convention. This clearly seems to be the route that could be followed in such a

108
Art 35.2.b ECHR; Gragl, in this regard, notes that with this provision the Strasbourg Court is
assured the function of the subsidiary court; one that does not compete with the Luxembourg Court
on horizontal level. See: Gragl (2013), p. 188.
109
Cf.: Gragl notes that the rule on exhaustion does not apply on every inter-party case; it applies
only when a contracting party submits an inter-party application against another contracting party
with the aim of protecting its own individual citizens. If a contracting party files an inter-party
application of the actio popularis form, it needs not to exhaust legal remedies. See: Gragl (2013),
p. 187/8. This said, it is basically maintained that, according to this line of reasoning, Art. 5 DAA
should be read as having an effect only in relation to inter-party cases that contracting parties
submit to protect the interests of their individual citizens; it does not apply to actio popularis type
of inter-party applications.
110
See the overall meaning of Art. 5 DAT as a means to start this analysis: Groussot
et al. (2011), p. 10.
272 7 Inter-Party Mechanism and the EU: Possible Implications from the. . .

scenario, however, in that case, the mood of Luxembourg would be either to


consider it a violation of Art 344 TFEU or to omit from its exclusive jurisdiction
on domesticated international instruments like the Convention which it considers
the law it observes.111

7.7.4 The Effect of Art 5 (Second Clause) of the DAA


to the Relationship Between Art 344 TFEU and Art
55 ECHR: Why Is the ‘Special Agreement’ Special?

It is essentially argued that Art 5 of the DAA could constitute the ‘special agree-
ment’ under Art 55 ECHR. Three arguments support this conclusion and one does
not. First, the DAA refers explicitly to the EU Treaties dispute settlement jurisdic-
tion, meaning the EU Court jurisdiction, and provides for the reading that the
dispute-settlement mechanism under the EU Treaties would not conflict with the
one under Art 55 ECHR. Second, the DAA will have to be ratified by all High
Contracting Parties to the Convention, meaning both EU and non-EU Convention
contacting parties. Such unanimous consent on the agreement fulfils the condition
that the special agreement be agreed by each party to the Convention on this point
(as also determined by Art 31.3 VCLT). Third, the DAA does not however resolve
the dispute-settlement question for non-EU Convention parties, which clearly
would put doubts on whether this is a real ‘special agreement’ in light of Art
55 ECHR.
The more important issue is analyzing why is the ‘special agreement’ special?
This certainly needs another layer of analysis to portray the entire picture of
cohabitation between Art 344 TFEU and Art 55 ECHR after EU accedes to the
Convention. To undertake that analysis, I put five reasons on the special nature of
this ‘special agreement’, starting with the more general arguments and ending with
those less significant.
First, Art 5 of the DAA merely precludes the foundational conflict between Art
344 TFEU and Art 55 ECHR. It does this on a one-side formula. However, it does
not rule out such potential conflict. By categorizing the proceedings between
Member States inter se before the Luxembourg Court not as procedures of inter-
national dispute settlement, the DAA only makes room for the inter-party jurisdic-
tion of the Strasbourg Court to cohabitate with that of the EU Court on principle
terms (meaning that EU Member States could use Art. 344 TFEU/Luxembourg

111
Contra: Gragl, through a brief pathway, comes to the conclusion that Art. 5 DAA reconciles the
EU law autonomy (including Art. 344 TFEU) with the Convention’s inter-party mechanism. See:
Gragl (2013), p. 188. It is argued here, however, that Gragl’s point does not address profoundly the
implications which the Luxembourg Court would usually pushes for—as examined below, and as
also demonstrated with the Luxembourg’s Opinion 2/13—something that may mislead the overall
expectation from this process.
7.7 Beyond the Conventional Concept on Competing Jurisdictions of the Two. . . 273

Court’s jurisdiction as the first instance for their dispute settlement relating to the
Convention, and then the Strasbourg Court as the second instance for merely
Convention-related disputes). However, Art 5 of the DAA does not confer the
dispute-settlement of inter-state complaints under the Convention to the EU
Court neither for the entire Convention parties nor for the EU Convention parties.
Qualifying the proceedings before Luxembourg Court as not international dispute
settlement procedure, the DAA in fact intends to make the entire EU-related
judicial procedures as procedures of domestic law, which is a normal prerequisite
for a case before it becomes eligible to be filed at Strasbourg. As such, the inter-
state proceedings at the EU level seem to be qualified as procedures of internal law,
whereas the intention is to make the view of the Luxembourg Court as a court of
domestic law not of international law. That being the case, Art 5 of the DAA is
supposed to be more dedicated to making a basis for Art 55 of the Convention to
cohabit abnormally with Art 344 TFEU rather than to offer a fully-fledged solution
for the possible conflicts.
Second, it is obviously argued that the innovation served with Art 5 of the DAA
does not prohibit any party from making use of the inter-state complaint procedure
under the Convention. It suffices to say that Art 5 of the DAA neither prohibits
inter-party cases from being adjudicated at Strasbourg nor undertakes the transfer
of the jurisdiction to the Luxembourg Court on inter-party cases flowing from the
Convention. This argued, parties to the Convention, be them EU Member States,
non-EU Member States or the EU itself will still have the possibility to make use of
Art 55 inter-party mechanism. In no way can one find an argument to say that Art
5 of the DAA should be read as preventing the Strasbourg Court to rule on inter-
party issues. If the drafters of the Draft Accession Agreement would have intended
that, they could have made that explicit in the Agreement, which they did not.
Moreover, the article concerned merely makes it possible for the Convention bodies
to retain their competence even though the same case being judged has been ruled
by another court of international law, the Luxembourg Court. This being the case,
Art 55 ECHR will still be an effective mechanism for all Convention parties, if their
domestic legal obligations allow for such a path to be followed.
Third, Art 5 of the DAA does not solve the issue whether the EU must first solve
an inter-state issue prior to Strasbourg Court taking that jurisdiction. It does not also
prohibit EU Member States from submitting inter-party requests to the Convention
in the same time as to the Luxembourg Court (or the Commission as the first
instance). With this reading, Art 5 of the DAA allows that the EU-Member States
relationship regarding issues of Convention be an issue of the domestic EU law, and
an obligation which Member States could have towards the EU Treaties but nothing
more than this. This being the case, one may perhaps question whether EU Member
States are violated their rights of petition under the Convention if Art 344 TFEU is
the domestic obstacle for such action. Certainly, within the context of a joint
reading of Art 5 of the DAA and Art 55 of the Convention, prohibiting EU
Convention Member States from such rights would amount to a violation for the
core principle of human right protection under an international law instrument such
274 7 Inter-Party Mechanism and the EU: Possible Implications from the. . .

as the Convention. Such prohibition would certainly amount to a violation of Art


27 of the Vienna Convention on the Law of Treaties, which sets that:
A party may not invoke the provisions of its internal law as justification for its failure to
perform a treaty. [. . .].112

Although VCLT is not binding in principle upon the EU, it still comprises a
restatement of customary international law which the EU may be bound to
according to its Treaties’ deference to customary international law and the EU
Court specific deference to parts of the customary international law that have the
nature of ius cogens (adding here that according to Lauterpacht ‘[t]he concept of jus
cogens operates as a concept superior to both customary international law and
treaty.’113). The EU Court confirmed judge Lauterpacht’s argument unequivocally
in Kadi.114 The question in this regard therefore is whether Art 344 TFEU could be
seen as the domestic source which the Luxembourg Court might invoke to justify a
decision asking from an EU Member State not to submit an inter-state dispute
settlement to the Strasbourg Court. In commenting the latter question, attention
should be made to the Commentary on DARIO which states ‘that internal law,
which depends on the unilateral will of the State, may never justify what constitutes
the breach by that State of an obligation under international law.’115
A further question previous to this one is whether Art 344 TFEU is considered
part of ‘internal law’ in terms of Art 27 VCLT. Certainly, there exist two reasons for

112
Art 27 VCLT. This rule being so old as also found in: Permanent Court of International Justice,
Greco-Bulgarian Communities Advisory Opinion PCIJ (1930), no. 17 (“it is a generally accepted
principle of international law that in the relations between Powers who are contracting Parties to a
treaty, the provisions of municipal law cannot prevail over those of the treaty.” Para. 81);
Available at: http://www.worldcourts.com/pcij/eng/decisions/1930.07.31_greco-bulgarian.htm.
113
International Court of Justice, Order of 8 April 1993 on the request for the indication of
provisional measures in the case of the Application of the Convention on the Prevention and
Punishment of the Crime of Genocide (Bosnia and Herzegovina v. Serbia and Montenegro).
Separate opinion of Judge ad hoc Elihu Lauterpacht (I.C.J. Reports 1993, 440), para. 100; One
should note that it is not the intention here to regard the VCLT as having the nature of ius cogens,
however, referring to Kadi, the EU Court left the impression that, as ius cogens form part of
customary international law, it respects the entire spectrum of customary international law
(including VCLT which has some very relevant sections that regulate the latter and thus have
the status of customary international law) as something it defers to rather respectably. VCLT’s
mentioning in Kadi makes this argument specific.
114
Court of Justice of EU (Court of First Instance of the European Communities), Yassin Abdullah
Kadi v. Council of the EU and Commission of the European Communities, Case T-315/01,
judgment of 21 September 2005, para. 227 (‘it must be noted that the Vienna Convention on the
Law of Treaties, which consolidates the customary international law and Article 5 of which
provides that it is to apply “to any treaty which is the constituent instrument of an international
organisation and to any treaty adopted within an international organisation”, provides in Article
53 for a treaty to be void if it conflicts with a peremptory norm of general international law ( jus
cogens), defined as “a norm accepted and recognised by the international community of States as a
whole as a norm from which no derogation is permitted and which can be modified only by a
subsequent norm of general international law having the same character”’).
115
Official Commentary on DARIO, p. 15.
7.7 Beyond the Conventional Concept on Competing Jurisdictions of the Two. . . 275

saying yes to this question. First, the argument is that Art 5 of the DAA clearly
portrays the jurisdiction of the EU Court as something belonging to the domestic
law of a party to the Convention. This is clearly said in Art 5 of the DAA when the
inter-state procedures within the EU law mechanisms are named as not constituting
international dispute settlement procedures. Second, with the EU acceding to the
Convention, for the latter’s context of human rights protection, the EU is considered
a party equal to a state High Contracting Party, and for such purposes, the EU will
have a state alike position in the Convention’s system. There seems no contest on
this issue regarding the substantive perspective of the accession outputs.116 These
two being the reasons, one should argue that EU law is considered an ‘internal law’
for purposes of interpreting Art 27 VCLT. The latter analysis is also supported by
the non-binding Art. 27.2 VCLTSIOIO which prescribes that: ‘An international
organization party to a treaty may not invoke the rules of the organization as
justification for its failure to perform the treaty.’117 ‘Internal organization’s rules’
in this regard are equal with the concept of ‘internal law’ applied above. Therefore,
there seems no dispute on this.
Therefore, the exclusive jurisdiction of the Luxembourg Court to observe the
Convention as part of EU law seems to violate the international obligation of the
EU118 not to invoke domestic law as means for precluding an obligation towards the
Convention as an international law instrument. Such potential violation of Art
27 VCLT regarding the Convention’s inter-party procedure would not only produce
the result that EU violate it on its part but also the Member States as parties to the
Convention. Of course, the argument for an EU Member State would be that it
could not file a petition to the Strasbourg Court under Art 55 ECHR as that would
violate the EU Treaties in which that member state is a party. This being argued, the
expense for saving the exclusive jurisdiction of the EU Court to rule on EU law,
especially its Art 344 TFEU, is to violate the Convention and Art 27 VCLT. That
being the scenario, a special feature of this ‘special agreement’ is the room for an
international law violation to appear invisible from anyone.
Fourth, Art 5 of the DAA allows the EU Court to continue to consider Conven-
tion part of the own law that it observes. However, a more in-depth analysis of
whether Art 344 could be reconciled with Art 55 ECHR is needed. For sure, the EU
Court will insist to have Art 344 further applied upon all EU Member States, as it
did rigidly in Mox Plant and Opinion 1/91 as the most brilliant examples. The
question however is whether this special nature of the ‘special agreement’ that Art
5 of the DAA provides/amounts for another layer of competition between the two

116
One should omit from dealing with the ‘names’ that the DAA revises in the Convention, as that
is done more or less for symbolic reasons, whereas the accession as such produces a state-like
position for the EU in the Convention system. The same view has been ruled by the Luxembourg
Court in Opinion 2/13 (para. 193).
117
Emphasis added.
118
Two cases of the EU Court seem to suggest the later pays rather important respect for
international law. These two cases are: Court of Justice of EU, Poulsen and Diva, Case C-286/
90 [1992] ECR 1-6019, and, Court of Justice of EU, Racke, Case C-162/96 [1998] ECR 1-3655.
276 7 Inter-Party Mechanism and the EU: Possible Implications from the. . .

European courts. The answer is simple: certainly, Art 5 of the DAA basically
legalizes the conflict that both courts could have on inter-party cases. On the one
hand, the EU Court would insist to hold the EU Member States under the control of
Art 344 TFEU, clearly not allowing that they submit a complaint deriving from the
Convention to the Strasbourg Court’s inter-party mechanism. Furthermore, if one
would argue that the EU Court would allow a case to be filed at Strasbourg’s inter-
party procedure only after it has been exhausted at Luxembourg, this still would not
be what Luxembourg normally is expected to do with Mox Plant serving as the best
example.
The matter is that Art 344 TFEU makes it clear that no EU Member State could
file an application for the interpretation or application of EU law to another
mechanism of dispute settlement other that the EU Treaties methods. Art
344 TFEU is clearly made to stop any application, reviewed or not by Luxembourg
Court. A teleological interpretation of Art 344 reveals that it is intended to stop
Member States from going to another court/tribunal for finding a settlement on the
interpretation of EU Treaties. So, the intention of Art 344 is not only to protect the
exclusive jurisdiction of the Luxembourg Court but also the autonomy of the EU
law at-large. Certainly, with the EU acceding to the Convention, Art 344 could not
in principle be thought to exclude this duty on the Convention’s inter-party cases as
well. At least this could be the picture that the rigid and allergic nature of external
tribunals that Mox Plant and Opinion 1/91 provide for this argument.
These being the arguments, a general reading of Art 5 of the DAA, Art
344 TFEU and Art 55 ECHR give the impression that the conflict between the
two courts on inter-party cases has not been fully resolved.119 Only the practice will
show whether these legal loopholes will be ‘dialoged’ between the two courts
constructively.

7.7.5 An Additional, More Hypothetical Explanation


on the Effect of Art 5 of the DAA on Art 344 TFEU
(A Second, More Constructive Scenario)?

Another more constructive explanation on this issue would be to consider the DAA
as such producing adjustment effects on EU Treaties interpretation. As argued in
the Chapter on the status of the Convention on EU law, the DAA of the EU to
Convention has a special constitutional significance. Besides, the Human Rights

119
Cf.: With regard to the potential exclusion of the EU from the ECHR’s inter-party mechanism,
Gragl, referring to Christoffersen, rightly argues that the system of protection under the Conven-
tion—after the EU accedes to the ECHR—must be the same for the EU as for the other high
contracting parties and subject to the same terms of obligations. (See: Gragl 2013, p. 184).
Otherwise, the proclaimed objective of the DAA to offer include the EU into an equal footing
with the other European high contracting parties would make no sense in practice.
7.7 Beyond the Conventional Concept on Competing Jurisdictions of the Two. . . 277

Committee has ruled a very important principle with regard to the possible circum-
ventions to the International Covenant on Civil and Political Rights mechanisms
using states’ domestic legal means. It has ruled that: ‘Domestic laws may need to be
altered properly to reflect the requirements of the Covenant; and mechanisms at the
domestic level will be needed to allow the Covenant rights to be enforceable at the
local level. Reservations often reveal a tendency of States not to want to change a
particular law. And sometimes that tendency is elevated to a general policy.’120
This standard does clearly provide for an international law duty to refrain from
interfering to the mechanisms under a ratified international agreement. The same
may be well fitted in the EU and ECHR inter-party problem, wherein the author
proposes a constructive scenario to mind the existing gap between the two legal
orders (adding that the EU has not placed a reservation to DAA to stop the potential
use of the inter-party procedure from its Member States). This being the case—in a
constructive scenario—the possibility exists for arguing that the DAA in fact
‘adjusts’ the EU Treaties by passive means.121 More important for this subchapter
is to take for granted the fact that Art 5 of the DAA adjusts the conventional
meaning of Art 344 TFEU (the latter may be one reason why Opinion 2/13, in para
201 et seq, seems to refuse the Strasbourg Court’s ratione materie jurisdiction on
inter-party cases).
In principle, Art 5 of the DAA, and its second clause, are mainly aimed at
amending and adjusting the meaning of Art 55 ECHR. However, the possibility also
exists that the DAA produce effects and correspondingly adjust the meaning of Art
344 TFEU, with the latter being the main problem in reconciling the conflict
between the two systems. In this context, one could argue that the second clause
of Art 5 of the DAA produces the effect of adjusting the meaning of Art 344 TFEU,
in order to exclude the Convention’s Strasbourg Court as the external ‘method of
settlement’ in the context of Art 344 TFEU. There exist three reasons for adhering
to this form of interpretation:
First, Opinion 2/94 ECJ had concluded that EU accession to the Convention
would entail substantive changes into the Community Treaty System. Such con-
clusion makes manifest the fact that ECJ had considered that the accession proce-
dure would make at least ‘some’ constitutional and systemic changes to the EU
Treaty system in order to enable the latter to submit to the Convention system. One

120
Human Rights Committee, (Issues Relating to Reservations made upon Ratification or Acces-
sion to the Covenant or the Optional Protocols thereto, or in Relation to Declarations under
Article 41 of the Covenant), General Comment No. 24, 4 November 1994 (CC PR/C/21/Rev.1/
Add. 6), para. 12.
121
Cf: Gragl basically argues that the DAA does not govern the internal EU law issues relating to
the inter-party jurisdiction in the context of Art. 344 TFEU, and that it not something that one may
expect. See: Gragl (2013), p. 185; However, it is the Luxembourg Court’s Opinion 2/13 which
requires the opposite: it invites the DAA to regulate this ‘issue of EU law’, by specifically
requiring that the Strasbourg Court be denied jurisdiction to adjudicate cases under the inter-
party procedure if EU or its Member States wish to use it inter se. See supra the subchapter on
Opinion 2/13.
278 7 Inter-Party Mechanism and the EU: Possible Implications from the. . .

such systemic change is supposed to be the exclusion of the Convention from the
scope of prohibition of Art 344 TFEU. At least, hypothetically, this is one of the
core systemic changes which Opinion 2/94 implicitly referred to. With the acces-
sion duty made explicit in the Lisbon Treaty, one cannot rule out that the DAA
adjust the meaning of Art 344 TFEU to make the EU submitted to the Convention
system for Convention-related cases. This risk has not been captured by the Court’s
eye in Opinion 2/13, the latter making only a very broad opposing point on the inter-
party mechanism.
Second, the travaux préparatoires of the DAA reveal the fact that the drafters
did not have the intention to permit for the Convention inter-state mechanism to
exist in relation to the EU. The situation was later turned over as a result of the
insistence and pressures that this could damage the Convention’s integrity to
protect human rights. This being the case, it is argued that the drafters of the
DAA were conscious that the introduction of the inter-state mechanism would not
be done only for symbolic reasons but also for practical reasons, as at the end of the
day no EU member state could be banned from filing an inter-party complaint to the
Strasbourg Court if such possibility would exist in the DAA.
Third, the EU Treaties could not continue to conflict with the Convention on an
issue as big as the inter-party mechanism. For this reason, a constructive reading of
Art 5 of the DAA which the Luxembourg Court could consent to surpass the Mox
Plant standard, might allow that it be read as adjusting the meaning of Art
344 TFEU in order to ensure that there exists no conflict between the EU and
international commitments undertaken with the Convention.
Therefore, supposedly, although never tested practically, the Luxembourg Court
could accept to read Art 5 of the DAA as adjusting Art 344 TFEU to read that the
Convention inter-party mechanism could be utilized by the EU Member States as
well only after the case has been exhausted domestically at the EU Court.122 This
‘constructive’ reading of Art 5 of the DAA, one must make the notice, would seem
to follow such a consensual approach which the Luxembourg Court has almost
never practiced to date.
Should this practice be allowed, then the logical reading that could be produced
by the Luxembourg Court would be to interpret Art 344 TFEU broadly, and to
request that the Member States first exhaust their EU mechanisms to settle a dispute

122
Cf. Gragl takes the example of Mathews v. UK imagined in an inter-party scenario to
demonstrate that it may be well operated also within the Luxembourg Court’s current jurisdiction.
In that imaginary case, Spain would file an inter-party application against UK before the Luxem-
bourg Court (infringement proceedings) by arguing that UK had violated EU law (the Conven-
tion’s right to vote) by not providing the right to stand for elections to Ms. Mathews from Gibraltar.
Only after this procedure would have been exhausted, then Spain could have applied before the
Strasbourg Court. In that scenario, Gragl proposes that the Strasbourg’s inter-party review would
not touch upon EU law autonomy, because it would merely declare on the Luxembourg’s decision
on this issue from the perspective of the Convention standards. This may be an example to
demonstrate an EU Member State applying against another EU Member State. The same proce-
dure would apply if an EU Member State would intend to file an application before Strasbourg
Court against the EU. See: Gragl (2013), p. 190–197.
7.7 Beyond the Conventional Concept on Competing Jurisdictions of the Two. . . 279

between them concerning the Convention. Once that issue becomes exhausted
before the EU Court as an inter-state procedure, then the parties might send it to
the Strasbourg Court for a subsidiary level of review. In this type of consensus
between Art 344 TFEU and Art 55 ECHR, the Luxembourg Court would construe
its 19 (1) TEU jurisdiction to rule on EU Member States cases inter se as a
jurisdiction of domestic law, and that of Strasbourg Court as a subsidiary jurisdic-
tion like in the individual-complaint cases.
The ultimate point that needs analysis here is whether within this constructive
and questionable scenario there would exist the possibility to ban EU Member
States from submitting an inter-party case to the subsidiary Strasbourg Court
without first exhausting it at the EU level. Certainly, such a rationale could also
be found within the meaning of Art 4(3) TEU, which seeks that loyal cooperation
between EU Member States be demonstrated within the context of loyalty to the
Treaties as well.123 Therefore, a more structured view of this constructive scenario
is that the Luxembourg Court view Art 5 of the DAA as excluding the Convention’s
inter-party mechanism as being looked as a method of settlement within the context
of Art 344 TFEU,124 and holding Art 4 (3) TEU as the means to seek from EU
Member States to bring the issue first before Luxembourg Court before they are
permitted to file at the subsidiary Court of Strasbourg.
As a last point, one should mention the fact that, by deconstructing Mox Plant,
one comes to the view that the Luxembourg Court seems to accept the fact that the
Union can accede to international mixed agreements which have their own dispute
settlement mechanisms (the same approach has also been recalled in para. 181 of
Opinion 2/13). A logical and systemic reading of Mox Plant, makes it clear that
Member States are not barred from acceding, together with the EU, to mixed
international agreements which have their own dispute-settlement mechanisms/
tribunals. On the other hand, an in-depth reading of the overall argument of Mox
Plant reveals that the existence of conflicting dispute-settlement mechanism
between the EU Treaties and a mixed international agreement wherein EU and its
Member States are jointly contracting parties does not amount to a paradoxical

123
See in the context of Mox Plant: Churchill and Scott (2004), p. 663; See also: Court of Justice of
EU, Case C-25/94 Commission v Council (FAO) [1996] ECR I-1469, and, Court of Justice of EU,
Case C-266/03 Commission v Luxembourg [2005] ECR I-4805.
124
Referring to Schott, Gragl points that—in the context of the inter-party debate—the scope of
Art. 344 TFEU should not in principle cover issues that may be subject to the Convention but
outside the scope of EU law (Gragl links this to the Member States’ obligation within the ‘scope of
EU law’ formula, under Charter’s Art. 51). This means that, in Gragl’s view, which seems quite
contrary to the findings of the Mox Plant approach in which Luxembourg Court chose a wide-
discretion formula to perform the competence attribution test examined supra (and quite differ-
ently to the conclusions of this book), issues outside the EU competence are precluded from the
scope of Art. 344 TFEU and therefore may be filed before the Strasbourg Court without needing to
be exhausted or even considered in light of the Luxembourg’s restrictive jurisdictional criteria.
See: Gragl (2013), p. 196/7. Gragl, however, does not offer a feeding debate on how broadly the
EU Court interprets the scope of EU ‘competence’ for purposes of Art. 344 TFEU, something that
would have likely changed Gragl’s direction of this argument.
280 7 Inter-Party Mechanism and the EU: Possible Implications from the. . .

legal situation which goes beyond the prohibition of Art 344 TFEU. However, in
terms of Mox Plant, the mere utilization of such external dispute settlement
mechanism from an EU Member State conflicts with Art 344 TFEU, and put the
respective Member State in violation of the Luxembourg Court’s exclusive juris-
diction. With this in mind, it seems that the EU Court sees the concept of conflict
with the external mechanisms of dispute-settlement only in light of the de facto
rationales but not de iure ones. The plain fact that the Convention, as a mixed
agreement wherein EU and Member States are jointly contracting parties, has its
own dispute-settlement procedure does not legally produce a violation of Art
344 TFEU in light of Mox Plant. The operation of such clause of the Convention
however is supposed to produce such violation on factual basis.125 This said, the
context of reconciliation of Art 5 of the DAA in conjunction with Art 55 ECHR and
with Art 344 TFEU seems to be something that could, in either scenarios, be
devised on de iure or de facto bases, whichever solves better the practical conflicts
that could arise between the two courts. EU may not invoke Art. 344 TFEU as
means to surpass its international obligation to respect Art. 55 ECHR. At the end of
this stony road of analysis, the Permanent Court of International Justice Advisory
Opinion may help this scenario in making Art. 344 TFEU ‘constructive’. The
Permanent Court had ruled in this regard that ‘it is obvious that the Treaty must
be read as a whole, and that its meaning is not to be determined merely upon
particular phrases which, if detached from the context, may be interpreted in more
than one sense.’126 This constructive scenario offered here stands in the lines of this
standard.127
To make a short digression, finally, one needs to question whether the Lisbon
Treaty—with Art. 6 (2) TEU—has intended to make Art. 344 TFEU of an exclusive
position towards EU’s international law obligations under the upcoming accession
to the Convention? One may be served with a landmark decision of the US Supreme
Court (as a comparative case), namely Murray v. The Charming Betsy where the
Court—with regard to the intention of the domestic law towards international
law—had ruled that ‘an act of congress ought never to be construed to violate the
law of nations, if any other possible construction remains. . ..’128 In the words of
Curtis A. Bradley, this is the so called ‘“legislative intent conception” [which] rests
on the assumption that Congress generally does not wish to violate international law
[. . .].’129 Fitzpatric argues that Murray v. The Charming Betsy seeks to provide ‘an
internationally law-abiding character to the law-giver whose unclear enactments

125
See with reference to Mox Plant: Churchill and Scott (2004), p. 663.
126
Permanent Court of International Justice, Advisory Opinion on the Competence of the ILO to
Regulate Agricultural Labour, P.C.I.J. (1922), Series B, No. 2/3, p. 23.
127
Cf.: Gragl (2013), p. 272, who, although not taking the road taken by the above chapter to
examine this issue from an in-depth perspective, comes to the general conclusion that the DAA
should not be read as prohibiting the operation of the inter-party mechanism post-accession.
128
US Supreme Court, Murray v. The Charming Betsy, 2 Cranch 64, 118 (1804), p. 6.
129
Bradley (1998), p. 495.
7.7 Beyond the Conventional Concept on Competing Jurisdictions of the Two. . . 281

are at issue.’130 In Bradley’s words—and within the frames of Murray v. The


Charming Betsy—one may put the argument that the EU Treaties may have not
intended to make Art. 344 TFEU harm EU’s obligations under the Convention
(including the provision on inter-party mechanism).131 Teleologically speaking, the
Masters of the EU Treaties may have not relied on the assumption that any
provision of the Treaties should or will violate international law, respectively the
Convention in this case. Had the Masters of the EU Treaties intended to make Art.
344 TFEU violate the mechanisms of the Convention once EU accedes to the latter,
they would have simply refused to make the accession of EU to the Convention as a
core obligation for the EU. The argument goes, there may be no judge with the EU
Court that would use the argument that the Masters of the EU Treaties ought to
make the EU Treaties violate obligations which EU undertakes under international
law. Therefore, Murray v. The Charming Betsy would very well support the
argument presented in this constructive scenario of reconciling Art. 344 TFEU
with Art. 5 DAA, thereby ensuring that the EU respects and does not put barriers to
the use of the Convention’s core mechanisms. As a final note, it is unfortunate that
the Luxembourg Court’s Opinion 2/13 does not follow such line of argument132;
rather it finds the mere fact of the existence of the inter-party mechanism as
incompliant with the EU Treaties (as noted at the very beginning of this chapter).
The latter is a basic argument which this chapter has shown in the beginning;
however it would not be practically feasible to keep the language so simple and
irresponsible to the nature and existence of international law—which, ironically,
the Luxembourg Court demonstrates in this recent opinion.

130
Quoting Fitzpatric, at: Bradley (1998), p. 495/6.
131
One may disagree with this statement with the argument that Murray standard may not apply in
case there is a horizontal interface between two acts of international law, namely the EU Treaties
and the Convention. I oppose this view and therefore counter-argue that with the EU becoming a
contracting party to the Convention, its law (including the EU Treaties) takes the status of
domestic law in face of the Convention. From that perspective, such post-accession scenario
which is discussed above considers the EU Treaties as domestic law in face of the Convention,
therefore the Murray standard applies here as well since EU Treaties (as a domestic law of one of
the contracting parties) may not have intended to violate the Convention.
132
Gragl supports roughly the same line of argument of this thesis. He rightly points that the
exclusion of the EU from the inter-party mechanism would seriously reduce the scope of EU
accession to ECHR on the normal standards of human rights protection, but would also run counter
the concept of the Convention as a collective system of human rights law protection. Gragl,
referring to Christoffersen, moreover argues that the inter-state mechanism is even more original
(as it was the first mechanism installed in the Convention when it was concluded) than the
individual complaint mechanism under the Convention system, therefore there seem even less
reasons to reject including the EU in the inter-party procedure. See: Gragl (2013), p. 183.
282 7 Inter-Party Mechanism and the EU: Possible Implications from the. . .

7.8 Testing the Inter-Party Procedure to Ireland v. UK:


What Does It Mean in Practice? From Theoretical
to a Practical Scenario

Although the inter-state complaint mechanism of the Convention has been utilized
only a few times throughout the entire history of the Convention, there have been a
few cases which proved highly important to delineating the jurisdiction and the
contexts under which such jurisdiction could be used in complaints between
Convention parties inter se. For the purposes of this thesis, it is important to test
an inter-state case that was submitted to the Strasbourg Court in the past against the
current version of the DAA, and the overall analysis of the interaction between Art
55 ECHR and Art 344 TFEU, not excluding Art 19 (1) TEU. The only case that
serves best this effort is Ireland v. United Kingdom, brought before the Strasbourg
Court in 1971 on basis of then inter-state procedure of the Convention.133 Two
dimensions make this case most appropriate to be tested against the current version
of the Accession Treaty: first, because it involved two EU Member States, which,
upon accession of the EU to the Convention would be bound by the DAA, and
second, because the substance and the object of this inter-state case involves a very
interesting material of law, one that is rather shared between the EU and Member
States wherein the human rights law layer plays an important role upon that
consideration.134

133
See e.g.: Donahue (1980), pp. 379 et seq.
134
Cf.: One may legitimately question whether this scenario may appear in practice as the EU
Court has repeatedly ruled that its fundamental rights jurisdiction is limited to situations where
Member States act within the scope of EU law, and that scope, in view of Siragusa (examined
supra in the subchapter on the ChFR), requires a certain connection between the obligations and
objectives of EU law with the Member States’ implementing measures. I tend to disagree with this
argument. There are three basic reasons for this. First, there is a huge difference between the
Luxembourg Court’s view on the Member States obligations under Art. 344 TFEU with its view on
the Member States’ obligations under Art. 51 ChFR. These two treaty bases have different
intentions, meaning and scopes. One may not apply the concept which Siragusa has developed
with regard to the Member States’ obligations under Art. 51 ChFR to Art. 344 TFEU, as the latter
has a very different nature, intention and scope. Just to make one brief argument: merely the
‘competence attribution’ test which the Luxembourg Court has developed in Mox Plant makes the
application of the Member States’ scope of obligations under Art. 51 ChFR irrelevant. The
Luxembourg Court does not care whether Member States are implementing an EU law objective
when they file an application to an external means of dispute settlement, but whether such act may
push EU law (on basis of the four tests explained supra) beyond the jurisdictional borders of the
EU Court. Second, Ireland v UK is taken as an example of demonstrating how the Luxembourg
Court devises its exclusive jurisdiction with regard to external tribunals and treaty regimes. It has
very little to do with fundamental rights jurisdiction and the Member States’ obligations under
that. The basis whereupon the Member States’ acted when implementing a certain EU law
obligation are not relevant on whether the EU Court will bar a certain EU law issue from being
reviewed in an external means of dispute settlement or not. And, third, Ireland v UK serves as a
perfect case to demonstrate that—when it comes to the competence attribution test which the
Luxembourg Court applies with regard to the application of Art. 344 TFEU in a specific case—the
7.8 Testing the Inter-Party Procedure to Ireland v. UK: What Does It Mean in. . . 283

Ireland v. UK was submitted to the Strasbourg Court by the Republic of Ireland


against the United Kingdom, the latter being the sole respondent of the case. It was
lodged to the Court on 1976. The action of Ireland against the United Kingdom
concerned the following complaint:
to ensure the observance in Northern Ireland of the engagements undertaken by the
respondent Government as a High Contracting Party to the Convention and in particular
of the engagements specifically set out by the applicant Government in the pleadings filed
and the submissions made on their behalf and described in the evidence adduced before the
Commission in the hearings before them.135

According to that time Convention, the Court was to have jurisdiction only upon
the respondent party having previously accepted the jurisdiction of the Court to act
on inter-state procedure, which UK has already accepted.136 Practically, the appli-
cation of Ireland against the United Kingdom requested the Court to find the
Government of the United Kingdom culpable for the use of the five interrogation
techniques which United Kingdom made use of in the territory of the Northern
Ireland,137 namely hooding, subject to noise, wall-standing, deprivation of food and
drink and deprivation of sleep. Such interrogation techniques, known otherwise as
‘disorientation’ techniques, Ireland claimed, amount to torture in violation of Art
3 of the Convention. Those techniques, Ireland claimed, were used in unknown
interrogation centers where certain detention practices have been accomplished.
Ireland submitted other claims attached to the ‘five techniques’, which also pointed
to the assertion that United Kingdom has not respected the minimum rights of the
criminal procedure to those persons detained in Northern Ireland.
The Strasbourg Court, regarding the use of the ‘five techniques’ by the United
Kingdom in the territory of Northern Ireland, ruled that United Kingdom violated
Art 3 of the Convention. The Court ruled that the use of the ‘five techniques’ were
not however of the ‘intensity and cruelty implied by the word torture as so
understood [. . .]’, however such behavior of the United Kingdom amounted to
inhuman and degrading treatment in violation of Art 3 of the Convention.138
I will use Ireland v. United Kingdom case to test the application of the inter-party
mechanism once EU accedes to the Convention. Such analysis will be done on the
presumption that the case happened in a situation where EU is a party to the

EU Court specifically draws attention to the fact that certain pieces of legislation may even not be
legislated or have not been legislated under EU law, but the mere fact that they stand as the
objectives of EU Treaties in the broad sense makes this test relevant. Fundamental rights remain a
core objective of the democratic principles which the EU Treaties’ proclaim; therefore the author
of this book considers the main-text argument as valid.
135
ECtHR, Ireland v. United Kingdom, Appl. No. 5310/71, Judgment of the Court of 18 January
1978, para 1 & 2.
136
ECtHR, Ireland v. United Kingdom, Appl. No. 5310/71, Judgment of the Court of 18 January
1978, note 135, para 2.
137
See e.g.: Donahue (1980), p. 390 et seq.
138
ECtHR, Ireland v. United Kingdom, Appl. No. 5310/71, Judgment of the Court of 18 January
1978, note 135, para 167 & 168.
284 7 Inter-Party Mechanism and the EU: Possible Implications from the. . .

Convention, and that the current Draft Accession Agreement is already legally
effective. Such examination would reveal whether Art 5 of the DAA will prove
helpful and effective to solve a case like Ireland v. United Kingdom in a policy
setting where the EU Court watchdogs the EU law autonomy very rigidly (suppos-
ing that Luxembourg Court Opinion 2/13 is not intended to block the mere
existence of the inter-party mechanism, which it frankly does not). To consider
this case against the legal picture that the EU Accession Agreement to the Con-
vention will establish one must deal with three individual issues and tests that were
devised by Mox Plant, as this thesis argued in the previous pages. First, checking
and identifying the ‘competence attribution’ test, which has been analyzed above;
second, performing the test on the external binderies of the Luxembourg Court’s
jurisdiction, and third, performing the test on the parallel provisions.

7.8.1 Testing the Case

The first issue, respectively test, is to qualify whether the material object of Ireland
v. United Kingdom case is one that could be attributed to the EU. More important,
Ireland v. United Kingdom has a more difficult ‘competence attribution’ test
compared to Mox Plant, as it involves not only issues related to criminal law but
also to the protection of human rights in criminal procedures. This double-hat of the
material object of Ireland v. United Kingdom makes it more complex for the case to
be deconstructed in the context of a hypothetical situation wherein the case appears
before Strasbourg Court supposing that the EU has acceded to the Convention.
Ireland v. United Kingdom involves a claim, for the most part, that the five
interrogation techniques used by United Kingdom in the territory of Northern
Ireland violate Art 3 ECHR, namely the prohibition of torture and inhuman or
degrading treatment or punishment of persons. On the one hand, the object of the
case is an issue relating to criminal procedural law and practice, meaning whether
there are procedural guarantees that the interrogation of persons be undertaken
under certain procedural assurances. On the other hand, that substance of the case
touches on the question of human rights of persons under a criminal interrogation
procedure. This being the case, the competence Ireland v. United Kingdom deals
with is both the competence on the minimal criminal procedure guarantees and the
competence on assuring human rights protection at such procedures.
To undertake the test of ‘competence attribution’ in this case that is being
hypothesized as if it has happened in a situation where EU is a party to the
Convention, one needs to dig into the nature of these two competences and check
whether the EU has any attribution to on that issue, and if yes, whether the
competence is shared or exclusive. The Treaties specify that the EU has the shared
competence on judicial criminal cooperation based on Art 67.3 TFEU and Art
82 TFEU, amongst others. For this specific issue, Art 82.2 TFEU reads:
7.8 Testing the Inter-Party Procedure to Ireland v. UK: What Does It Mean in. . . 285

To the extent necessary to facilitate mutual recognition of judgments and judicial decisions
and police and judicial cooperation in criminal matters having a cross-border dimension,
the European Parliament and the Council may, by means of directives adopted in accor-
dance with the ordinary legislative procedure, establish minimum rules. Such rules shall
take into account the differences between the legal traditions and systems of the Member
States.
They shall concern:
(a) mutual admissibility of evidence between Member States;
(b) the rights of individuals in criminal procedure;
(c) the rights of victims of crime;
(d) any other specific aspects of criminal procedure which the Council has identified in
advance by a decision; for the adoption of such a decision, the Council shall act unani-
mously after obtaining the consent of the European Parliament.
Adoption of the minimum rules referred to in this paragraph shall not prevent Member
States from maintaining or introducing a higher level of protection for individuals.

Art 82.2 TFEU makes it clear that EU is given the competence to facilitate the
judicial cooperation between EU Member States, which, inter alia, includes
adopting EU legislation to deal with criminal procedure admissibility of evidence
and/or the rights of individuals in criminal procedures. Art 82.2 TFEU makes it
clear that as far as for the benefit of EU judicial cooperation there is a need, EU will
set minimum rules which will not prevent Member States for introducing higher
levels of protection for individuals (as also set in bold in Art 83.1 & 83.2 TFEU).
This makes it clear that the EU possesses competences on criminal matters that can
also cover the minimum standards of interrogation in criminal procedure such as the
‘five techniques’ in Ireland v. United Kingdom.
To support even more this argument, EU has a rather developed secondary
legislation covering the fields of criminal procedure that would deal also with the
minimum procedural standards of ‘five techniques’ alike case. The question of
whether the EU has elected to exercise this competence is clearly affirmative, as
there are already a set of EU secondary legislation covering this field. Such
secondary legislation, amongst others, related to the substance of Ireland
v. United Kingdom includes Directive 2012/13/EU on the right to information in
criminal proceedings,139 Directive 2012/29/EU establishing minimum standards on
the rights, support and protection of victims of crime140 and Directive 2010/64/EU
on the right to interpretation and translation in criminal proceedings.141 This being
the case, there seems no objection to conclude that there is an EU, almost special,
competence on judicial cooperation in criminal matters, that also covers the com-
petence of the EU to set minimum standards on criminal procedure rather well
covering the questions on the ‘five techniques’ of Ireland v. United Kingdom.

139
Directive 2012/13/EU of the European Parliament and of the Council of 22 May 2012 on the
right to information in criminal proceedings.
140
Directive 2012/29/EU of the European Parliament and of the Council of 25 October 2012
establishing minimum standards on the rights, support and protection of victims of crime.
141
Directive 2010/64/EU of the European Parliament and of the Council of 20 October 2010 on
the right to interpretation and translation in criminal proceedings.
286 7 Inter-Party Mechanism and the EU: Possible Implications from the. . .

Therefore, there is an attributed competence to the EU regarding the substance of


this case, meaning that such a case would rather well fall within the scope of EU law
that is observed by the Luxembourg Court.
The other question regarding the attribution of competence would be to check
the second side of the material object that Ireland v. United Kingdom involves,
namely the prohibition of torture and inhuman or degrading treatment. The EU has
surely the competence to observe the right to prohibition of torture and inhuman or
degrading treatment in two bases: first, on basis of Art 4 ChFR, which is part of the
Treaty law, and second, on basis of Art 3 of the European Convention on Human
Rights, which forms a general principle of law and a document of constitutional
significance in the EU legal order directly effective and applicable in the Union
(even through upon accession such position will be even further fortified for this
instrument). It is therefore absolutely clear that the EU possesses the competence to
observe the human rights in the application of its law, in this case the EU criminal
law as argued above, the observance of which is a condition for the legality of the
implementation of such law.
This being the case, the ‘competence attribution’ test is not passed in the
hypothetical Ireland v. United Kingdom case, as the latter is considered to fall
within the scope of EU competence. The argument is that if not all, at least some of
the core parts of this case fall within the scope of the EU competence which the
Luxembourg Court is authorized to observe.
However, were Ireland to argue that the ‘five techniques’ of interrogation did not
fall in their entirety with the EU competence, and thus, Ireland could have argued
that it has filed before the Strasbourg Court only those parts of the claim that do not
fall within the EU competence. Certainly, this argument of Ireland would have not
also passed the test on the Luxembourg Court’s exclusivity to delineate its external
jurisdiction borders in each case. Luxembourg Court would argue that if there are
parts of such material of law falling outside its competence,142 it is for itself to
judge that first and to say which of those parts of the case fall outside its compe-
tence. This test would again reveal Ireland v. United Kingdom as not passing the
minimum threshold of the Luxembourg Court criterion. The third test would also
have not been passed in this hypothetical scenario (the test on parallel provisions):
the EU Court would argue that both the ChFR and the Convention form part of EU
law which it is authorized to observe. A potential argument by Ireland that it has
based the claim in the provisions of the Convention, and not of the ChFR, would be
absolutely disregarded on basis of the Court’s reason that both the ChFR and the
Convention are considered domestic law of the EU which the Court is entitled to
rule upon with an exclusive jurisdiction. The fourth test would be irrelevant to be
applied upon Ireland v. United Kingdom.
It is therefore argued that Ireland v. United Kingdom would not pass any of the
EU Court’s tests whereupon it assesses whether the substance of the case filed from
an EU Member State against another EU Member State before an external dispute-

142
See e.g.: Cardwell and French (2007), p. 124.
7.9 Inter-Party Procedure as Room for the Strasbourg Court to Interpret EU Law. . . 287

settlement mechanism, in our case the Strasbourg Court, infringes the exclusive
jurisdiction of the EU Court (Art 19 (1) TFEU), the prohibition to pull out an issue
of EU law to an external method of settlement (Art 344 TFEU) and Art 4 (3) TEU
on loyal cooperation. In light of the first scenario which I explained in chapter
supra, Ireland v. United Kingdom would prove as violating the Luxembourg
Court’s jurisdiction, and that is something the latter would find Ireland responsible
for infringing the Treaties. However, if the second scenario which I explained in
chapter supra would appear, the Luxembourg Court would at least insist that
Ireland v. United Kingdom be first reviewed and adjudicated before Luxembourg,
as a precondition for it to become admissible for a potential submission to the
Strasbourg Court’s jurisdiction later on.
Ireland v. United Kingdom, the argument goes, explains how difficult it is to
escape from the Luxembourg Court’s jurisdiction when it comes to inter-state
cases. Not that there is very little room for escaping from such jurisdiction, but
that the possibility to do so is highly contested from whether the Luxembourg Court
would follow scenario one or scenario two (explained in this chapter supra) in this
type of jurisdictional enterprise. This being the case, however, it seems that EU
accession to ECHR would leave very little room for cases such as Ireland v. United
Kingdom to be submitted directly before Strasbourg Court on basis of Art
55 ECHR. Even though this subchapter proves that with the DAA in force it
would not be likely that EU Member States utilize the inter-party mechanism due
to EU internal law limitations—something this chapter argues is rather well
safeguarded within EU internal law—the Luxembourg Court still insists in Opinion
2/13 that the jurisdiction of the Strasbourg Court ratione materie be removed on
this type of matters (para. 213; see also the subchapter supra on the Opinion 2/13).
One may not understand how come that the Luxembourg Court makes such ironical
general statements which do not have the deepness required, seen from the per-
spective of the relevance and problem which it tackles.

7.9 Inter-Party Procedure as Room for the Strasbourg


Court to Interpret EU Law: Where Does the Risk
for EU Law Autonomy Stand in Exceptional Cases?

It is important to mention that the protection of EU law autonomy within the


accession of EU to ECHR remains the key priority from the side of EU. To ensure
that the EU law autonomy is protected, the DAA has put several instruments in place
to prohibit the review of EU law from the Strasbourg Court. The entire co-respondent
mechanism built by the DAA is dedicated to pursuing this aim. However, not all
things are settled when it comes to the loopholes that might allow for the Strasbourg
Court to engage with the review of EU law. Such aim, yet, seems to be problematic in
terms of the inter-party procedure143 which the DAA does not outlaw.

143
This doubt has been raised a long time ago, e.g.: Analysis (1997), p. 240.
288 7 Inter-Party Mechanism and the EU: Possible Implications from the. . .

As well known, the core argument that has been pushed forward in relation to the
claim that the DAA protects EU law autonomy has had two dimensions: first, the
argument was that the Strasbourg Court does never engage with the interpretation
of High Contracting Parties domestic laws. It takes such laws as mere facts and does
not interpret them. On the other hand, it merely issues a judgment whether there has
been an act or omission, indispensible whether deriving or not from a specific law,
that caused the violation of the Convention. This assurance has been argued as the
core protection mechanism that will prevent the Strasbourg Court into engaging
with the interpretation of EU law when there appears a case that derives from an
obligation of EU law. Second, the co-respondent mechanism applied in individual
cases only, has been devised to prohibit the Strasbourg Court to engage with the
internal division of labor if there appears a case involving both EU and Member
State(s) legislation and action, in order to ban the possibility that the Strasbourg
Court get into the borders of EU law to check the correct violator of the Conven-
tion. This second assurance has been designed to prohibit that any of the exclusive
authorities of the EU Court become under the ‘attack’ of Strasbourg Court.
This supposed, the Strasbourg Court, when dealing with individual cases, but
also with almost the entire inter-party cases dealt to date, normally does not review
or interpret the High Contracting Parties’ laws, and uses them merely as facts.
Whenever there is a case being adjudicated before Strasbourg Court, the Court’s
assessment on the violation or not of the Convention will normally be dealt merely
in the context of acts or omissions of the respondents. Such practice and jurisdiction
of the Strasbourg Court has ensured that there is a clear division between the High
Contracting Parties domestic laws and the Convention as an international
instrument.
One basic problem however appears to exist within the context of inter-party
procedure under Art 55 ECHR. In Ireland v. United Kingdom, the Court also dealt
with the question whether there is jurisdiction to rule on claims of one party against
the laws of another contracting party as being in violation of the Convention.144
While clarifying its jurisdiction, the Court also engaged with the review of its
jurisdiction in inter-state cases. To consider its jurisdiction in this context, the Court
amongst others asked ‘whether a Contracting State is entitled to challenge under the
Convention a law in abstracto.’145 This question proves highly important especially
in light of the context of EU as a contracting party to the ECHR, and submitted to
the jurisdiction of the Strasbourg Court under Art 55 ECHR. There are two remarks
that need be raised regarding the abstract control of norms by Strasbourg Court in
an inter-party procedure: first, that abstract control of norms would mean that the
Court engaged with the review of a High Contracting Party domestic law, and this
makes the object of the case the interpretation of that law against the Convention,

144
ECtHR, Ireland v. United Kingdom, Appl. No. 5310/71, Judgment of the Court of 18 January
1978, note 135, para. 239.
145
ECtHR, Ireland v. United Kingdom, Appl. No. 5310/71, Judgment of the Court of 18 January
1978, note 135, para. 240.
7.9 Inter-Party Procedure as Room for the Strasbourg Court to Interpret EU Law. . . 289

rather than the law as a fact, and second, the abstract control of norms appear as the
loopholes wherein the Strasbourg Court enters into the domestic borders of national
legal orders of the High Contracting Parties, without any need there to appear an
individual case of action or omission in violation of the Convention. These two
dimensions would have huge significance for the EU law autonomy as an indepen-
dent source of law for the purposes of Luxembourg’s jurisdiction.146
In devising this abstract jurisdiction in inter-state procedure, the Court in Ireland
v. United Kingdom made the difference between individual applications (now)
under Art 34 ECHR and inter-state applications under Art 33 ECHR. The Court
drew the difference by arguing that Art 34 ECHR does in fact request that in order
for an individual application to be procedurally admissible, the applicant must
prove that he/she was a victim of a violation of one of the High Contracting Parties
of the rights envisaged by the Convention. On the hand, Art 33 ECHR merely sets
that a High Contracting Party may submit to the Court an alleged ‘breach’ of the
Convention by another High Contracting Party. The Court went further to rule that:
Such a “breach” results from the mere existence of a law which introduces, directs or
authorizes measures incompatible with the rights and freedoms safeguarded; this is con-
firmed unequivocally by the travaux préparatoires.147

Arguendo, the Court used a rather broad term to qualify the notion of ‘breach’
used by the Convention in Art 33, which seems logical if this interpretation is
undertaken in comparison with the bar introduced by Art 34 ECHR. This being said,
the Court argued that there is no need in principle for a High Contracting Party to
prove that there is an individual violation resulting from an action or omission of
the respondent High Contracting Party. The mere existence of an abstract violation,
in the form of a national legal measure/law which directs or authorizes measures in
violation of the Convention rights, suffices for the Court to call into responsibility
the respondent High Contracting Party for the violation of the Convention. The test
on abstract control of domestic laws within the context of Art 55 ECHR jurisdiction
may be passed only if the Court:
[m]ay find a breach of this kind only if the law challenged pursuant to Article 24 (art. 24) is
couched in terms sufficiently clear and precise to make the breach immediately apparent;
otherwise, the decision of the Convention institutions must be arrived at by reference to the
manner in which the respondent State interprets and applies in concreto the impugned text
or texts.148

146
On the latter, see e.g.: de Waele (2011), p. 142.
147
The Court here referred to the travaux préparatoires, Document H (61) 4, pp. 384, 502, 703 and
706 (Ireland v. United Kingdom, op. cit. note 9, para. 240); See the original version of the travaux
préparatoires, at: Travaux préparatoires of ECHR. Official Version Of The Travaux Preparatoires
Of The European Convention On Human Rights ‘Collected edition of the “Travaux préparatoires”
of the European Convention on Human Rights/Council of Europe ¼ Recueil des travaux
préparatoires de la Convention européenne des droits de l’homme/Conseil de l’Europe’. The
Hague: Martinus Nijhoff, 1975–1985.
148
ECtHR, Ireland v. United Kingdom, Appl. No. 5310/71, Judgment of the Court of 18 January
1978, note 135, para. 240.
290 7 Inter-Party Mechanism and the EU: Possible Implications from the. . .

In addition, the Court made it clear in this case that the absence of a law
prohibiting specifically a certain violation of the Convention could not be consid-
ered as such as a violation of any of the rights enshrined in the Convention. This
said, the in abstracto jurisdiction of the Court to control the compatibility of
national laws of High Contracting Parties with the Convention, within the inter-
party procedure, seems to be possible, although under very rigid bars that need be
passed by a case. Being sufficiently clear and immediately apparent may be most of
EU Regulations, which are abstract in nature, but very concrete in the terms and
procedures that they introduce.
To prove that the Strasbourg Court undertakes such abstract control of national
laws, which also enters into the interpretation of the national laws, reference is
made to Ireland v. United Kingdom, where the Court said:
Examination in abstracto of the legislation in force at the relevant time in Northern Ireland
reveals that it never introduced, directed or authorized recourse to torture or to inhuman or
degrading treatment.149

It becomes evident, therefore, that the abstract control of norms under the inter-
party mechanism may become a real challenge to the EU law external autonomy, as
all EU acts that fulfill the criteria of being sufficiently clear and immediately
apparent in the potential violation to the Convention that they may introduce
become eligible to be reviewed under this mechanism. This loophole—which
does not exist within the individual complaint procedure—would not seem highly
problematic even in terms of interpretation of Art. 344 TFEU, as the Luxembourg
Court does in many occasions, either in terms of the action for annulment, request
for an opinion or preliminary review procedure provide rather abstract rulings,
which may become reviewed by the Strasbourg Court for their compliance with the
Convention thereby producing the effect of interference to EU law autonomy. It is
unfortunate to conclude that the EU Court’s Opinion 2/13 does not make any note to
this huge gap that may provide space for interference to the EU law autonomy.

7.10 Chapter’s Summary of Conclusions

The chapter examined the formerly inter-state mechanism—now the inter-party


mechanism—of the Convention system against EU’s internal prohibitions of stand-
ing in it. The chapter concludes that although the intention may have been to
prohibit EU to possibly utilize the inter-party mechanism of the Convention system
(something which Opinion 2/13 still insists upon), a systematic and watchful
analysis reveals that the DAA does not rule out EU from having standing in
being a party to the inter-party complaint mechanism. This conclusion does not
per se diminish the value of the EU’s internal treaty limitations such as Art.

149
ECtHR, Ireland v. United Kingdom, Appl. No. 5310/71, Judgment of the Court of 18 January
1978, note 135, para. 241.
7.10 Chapter’s Summary of Conclusions 291

344 TFEU (although the Luxembourg Court in Opinion 2/13 tries to maintain the
contrary), and also does not make the latter’s handling impracticable. The latter
conclusion is supported with the argument that there is nevertheless a consensual
reading of the interaction between Art. 344 TFEU, the Convention’s inter-party
complaint mechanism provisions and DAA provisions on this issue which leads to
the understanding that the inter-party mechanism is neither meant to violate the
internal treaty limitations of the EU nor to make this central mechanism of the
Convention system impractical for the EU.
The chapter first offered an analysis on the treaty limitations that bar EU law
from getting interpreted by external courts or tribunals, examining their design,
intended effect and modes of testing their application. Having deconstructed the
nature of those limitations, the chapter then examined the ways that can reconcile
the EU Treaties’ limitations on this issue, Convention system and DAA. In making
this analytical exercise, the chapter found that it is possible to have these three
instruments reconciled in a constructive scenario, therefore making it possible for
the inter-party mechanism to function with EU as a party therein. In offering this
constructive scenario, the chapter argued that there need be made certain departures
from the conventional views on EU Treaties’ autonomous mechanisms (which the
Luxembourg Court seems unwilling to undertake if one reads the general lines of
Opinion 2/13, para. 201 et seq). Reading these interacting provisions constructively
was supported with certain provisions of international treaty law principles.
The chapter therefore concludes that the possibility to have EU as both an
addressee and applicant party in the inter-party complaint mechanism is both
possible and required from the Convention system itself. The procedural contexts,
in which this may become utilized, however, needs be carefully and strategically
addressed by both courts to ensure that accession becomes fully fledged and equal
for the EU compared to all other state High Contracting Parties. Taking this
conclusion for granted, the use of the inter-party mechanism with the EU being
an addressee and applicant party becomes a very pioneering feature being seen
uniquely in international law. Applying the inter-party mechanism in a scenario of
multi-level human rights law and framework becomes even more interesting, as the
Convention system will gradually become home to inter-party applications involv-
ing an international organization as well. This development will enrich EU’s place
in the pan-European human rights framework, but will also make the Convention
system a modern space for developing treaty law from multipolar perspectives.
Therefore, the chapter concludes that the accommodating EU in the inter-party
mechanism of the Convention is a doable function and will advance further the
EU’s capacity to operate in multi-level international law instruments in the same
capacity as other state High Contracting Parties. This may be a symbol of EU being
further expanded its stateness attitude and shape.
292 7 Inter-Party Mechanism and the EU: Possible Implications from the. . .

References

Analysis (1997) The European Community cannot accede to the European Convention on Human
Rights. Eur Law Rev 1:235–249
Bernhard W (2011) Art 344 AEUV. In: Calliess C, Ruffert M (eds) EUV/AEUV (Das
Verfassungsrecht der Europäischen Union mit Europäischer Grundrechtecharta Kommentar).
C.H.Beck (Auflage: 4)
Bjoerklund M (2001) Responsibility in the EC for mixed agreements—should non-member parties
care? Nord J Int Law 70(3):373–402
Bradley CA (1998) The Charming Betsy canon and separation of powers: rethinking the interpre-
tive role of international law. Georgia Law J 86:479–537
Caldwell RG (1920) The settlement of inter-state disputes. Am J Int Law 14(1/2):38–69
Cardwell PJ, French D (2007) Who decides? The ECJ’s judgment on jurisdiction in the MOX Plant
dispute. J Environ Law 19(1):121–129
Churchill R, Scott J (2004) The Mox Plant litigation: the first half-life. Int Comp Law Q 53
(3):643–676
Comment (1978) The international human rights treaties: some problems of policy and interpre-
tation. Univ Pa Law Rev 126(4):886–929
Costa J-P (2003) The European Court of Human Rights and its recent case law. Texas Int Law J
38:455–468
Cremona M (2003) The draft constitutional treaty: external relations and external action. Common
Market Law Rev 40(6):1347–1366
Crnic-Grotic V (1997) Object and purpose of treaties in the Vienna Convention on the Law of
Treaties. Asian Yearb Int Law 7:141–174
De Londras F, Kingston S (2010) Rights, security, and conflicting international obligations:
exploring inter-jurisdictional judicial dialogues in Europe. Am J Comp Law 58:359–414
de Waele H (2011) The EU, the Member States and international law. In: Layered global player.
Springer, Berlin, Chapter (141–152)
Donahue DE (1980) Human rights in Northern Ireland: Ireland v. the United Kingdom. Boston
Coll Int Comp Law Rev 3(2):377–432
Eckes C (2013) EU accession to the ECHR: between autonomy and adaptation. Mod Law Rev 76
(2):254–285
Gerards J (2011) Pluralism, deference and the margin of appreciation doctrine. Eur Law J 17
(1):80–120
Grabenwarter C (2009) Europäische Menschenrechtskonvention. Verlag C.H. Beck, Munchen
Gragl P (2013) The accession of the European Union to the European Convention on Human
Rights. Hart Publishing, Oxford
Greer S, Williams A (2009) Human rights in the council of Europe and the EU: towards
‘Individual’, ‘Constitutional’ or ‘Institutional’ justice? Eur Law J 15(4):462–481
Groussot X et al (2011) EU accession to the European Convention on Human Rights: a legal
assessment of the draft accession agreement of 14th October 2011. Fondation Robert
Schuman/European Issues N 218/7 November 2011
Hartley T (1999) Constitutional problems of the European Union. Hart Publishing, Oxford
Heffernan L (1997) A comparative view of individual petition procedures under the European
Convention on Human Rights and the international covenant on civil and political rights. Hum
Rights Q 19(1):78–112
Heliskoski J (2000) The jurisdiction of the European court of justice to give preliminary rulings on
the interpretation of mixed agreements. Nord J Int Law 69:395–412
Jacobs FG (2007) The sovereignty of law: the European way. Cambridge University Press,
Cambridge
Koutrakos P (2006) EU international relations law. Hart Publishing, Oxford
Koutrakos P (2010) I. External relations. Int Comp Law Q 59(2):481–489
References 293

Lavranos N (2006) The MOX Plant and IJzeren Rijn disputes: which court is the supreme arbiter?
Leiden J Int Law 19(1):223–246
Lavranos N (2007) The scope of the exclusive jurisdiction of the court of justice. Eur Law Rev 32
(1):83–94
Lavranos N (2008) Regulating competing jurisdictions among international courts and tribunals.
Za€oRV 68:575–621
Licková M (2008) European exceptionalism in international law. Eur J Int Law 19(3):463–490
Lock T (2009) The ECJ and the ECtHR: the future relationship between the two European courts.
Law Pract Int Courts Tribunals 8:375–398
Lock T (2010) EU accession to the ECHR: implications for judicial review in Strasbourg. Eur Law
Rev 35(6):777–798
Lock T (2012) End of an epic? The draft agreement on the EU’s accession to the ECHR. Yearb Eur
Law 31(1):162–197
Marsden S (2009) MOX Plant and the Espoo Convention: can Member State disputes concerning
mixed environmental agreements be resolved outside EC law? Rev Eur Commun Int Environ
Law 18(3):312–327
Martinico G (2012) Is the European Convention going to be ‘Supreme’? A comparative-
constitutional overview of ECHR and EU law before national courts. Eur J Int Law 23
(2):401–424
Mendez M (2010) The enforcement of EU agreements: bolstering the effectiveness of treaty law?
Common Market Law Rev 47(6):1719–1756
Miiller CA (2007) Fundamental rights in multi-level legal systems: recent developments in
European human rights practice. Interdiscip J Hum Rights Law 2(1):33–60
Odermatt J (2014) The EU’s accession to the European Convention on Human Rights: an
international law perspective. Leuven Center for Global Governance Studies. Working Paper
No. 136
Pasqualucci JM (1994/1995) The Inter-American Human Rights System: establishing precedents
and procedure in human rights law. Univ Miami Inter Am Law Rev 26(2):297–361
Pauwelyn J (2004) Bridging fragmentation and unity: international law as a universe of inter-
connected islands. Mich J Int Law 25:903–927
Quirico O (2010) Substantive and procedural issues raised by the accession of the EU to the
ECHR. Ital Yearb Int Law 20:31–53
Raba K (2013) The accession of the European Union to the European Convention for the
Protection of Human Rights and Fundamental Freedoms—overview of the accession agree-
ment. ERA Forum 14:557–572
Robertson AH (1959) The European Court of Human Rights. Int Comp Law Q 8(2):396–403
Romano C (2007) Commission of the European Communities v. Ireland. Am J Int Law 101
(1):171–179
Schreuer C (2008) What is a legal dispute? Available at: http://www.univie.ac.at/intlaw/
wordpress/pdf/95.pdf
Schütze R (2004) Parallel external powers in the European Community: from “Cubist” perspec-
tives towards “Naturalist” constitutional principles? Yearb Eur Law 23:225–274
Shany Y (2004) The first MOX Plant award: the need to harmonize competing environmental
regimes and dispute settlement procedures. Leiden J Int Law 17(4):815–827
Shelton DL (1989) Private violence, public wrongs, and the responsibility of states. Fordham Int
Law J 13:1–34
Tridimas T, Eeckhout P (1995) The external competence of the community and the case-law of the
court of justice: principle versus pragmatism. Yearb Eur Law 14(1):143–177
Villiger ME (2009) Commentary on the 1969 Vienna Convention on the law of treaties. Brill/
Martinus Nijhoff Publishers, Leiden
Volbeda MB (2006–2007) The MOX Plant case: the question of supplemental jurisdiction for
international environmental claims under UNCLOS. Texas Int Law J 42:211–240
294 7 Inter-Party Mechanism and the EU: Possible Implications from the. . .

Von Papp K (2013) Clash of “Autonomous Legal Orders”: can EU Member State courts bridge the
jurisdictional divide between investment tribunals and the ECJ? A plea for direct referral from
investment tribunals to the ECJ. Common Market Law Rev 50(4):1039–1082
Weiler J (1999) The constitution of Europe. Cambridge University Press, Cambridge
Weiler JHH (1992) Thou shalt not oppress a stranger: on the judicial protection of the human rights
of non-EC nationals—a critique. Eur J Int Law 3(1):65–91
Chapter 8
EU Prior-Involvement Review

8.1 Introduction

This chapter examines the mechanism on prior involvement of Luxembourg Court


as established by the DAA, and the main ambiguities which the autonomy of EU
law requires to keep sealed therein.1 This analysis is made against a complex
background of EU law autonomy, EU peculiar institutional features and the Stras-
bourg Court’s natural jurisdictional function (subsidiary review). The chapter
responds to the research sub-question: ‘What are the legal implications that the
use of the prior-involvement mechanism may reflect, and how can one deconstruct
its mechanics to make it readable to the individual claimant?’ The chapter therefore
provides an in-depth examination of the prior-involvement provisions of the DAA
and circumstances when they may be utilized, paying singular attention to the effect
on norms that may become reviewed and jurisdictional means that may become
utilized under this procedure both at EU law and Strasbourg level.2 This is done by

1
Contra: Referring to Lock, Gragl basically argues that there is in principle no need to establish a
prior involvement mechanism in order to safeguard the autonomy of EU law, as the Strasbourg
Court decisions are merely declaratory and may not question the validity of EU law from the
internal perspective. See: Gragl (2013), p. 234/5. This issue is counter argued below in this
chapter, adding that the baseline argument is that, still, this mechanism may pose challenges to
the EU law autonomy, indispensable of the fact that Strasbourg’s decisions produce no direct
internal effect in EU legal order.
2
Cf.: Considering previous possibilities, Gragl notes that as a first option one has to consider
whether it would be possible to establish an external mechanism in the form of a ‘reference
procedure’ from Strasbourg Court to Luxembourg Court instead of the prior involvement mech-
anism. Referring to Lock, Gragl argues that such external mechanism would directly interfere to
the EU law autonomy and provide an amendment to the EU Treaties. Such an external mechanism
in the form of a ‘reference procedure’—similar to the preliminary reference procedure under the
EU Treaties—via which Strasbourg Court would have referred a case to the Luxembourg Court to
rule on the EU law question, in Gragl’s words, would have moreover added a new function to the
EU Treaties and change the nature of the function which the latter has recognized to the

© Springer International Publishing Switzerland 2015 295


F. Korenica, The EU Accession to the ECHR, DOI 10.1007/978-3-319-21759-8_8
296 8 EU Prior-Involvement Review

cautiously assessing the origins of norms that may become reviewed in the course
of prior involvement, and eligibility means which EU would need to fulfill to
become entitled to request this procedure. Although prior involvement is argued
as a privilege for EU, the chapter also tries to depict legal implications which
Luxembourg Court will face to assess the Convention-violating EU law provisions,
something that is tackled both in terms of scope of review and nature of effect of
Luxembourg’s filters. Besides examining the conditions which EU would need to
fulfill to operate the prior involvement procedure, this chapter also tackles the
current EU Treaties’ spaces which may be utilized to reconcile this procedure
with the EU Treaties’ instructions. The chapter in addition analyzes the integrity
of this procedure in EU current legal remedies’ system, offering a number of
arguments in relation to the fact that this new procedure may introduce certain
implicit Treaty changes in EU law and institutional structure. This is tackled from
the perspective of whether prior involvement introduces a new remedy, to a degree
which also silently amends the EU Treaties. The chapter concludes that prior
involvement procedure may perhaps implicate certain aspects of the EU legal
remedies’ system, however it seems a very good device to defend the EU Court’s
exclusive jurisdiction.

Luxembourg Court. It is therefore, in Gragl’s words, maintained that the DAA’s reference to an
internal EU law mechanism—the prior involvement—as opposed to an external one, seems the
only way to keep EU law autonomy respected. Such external reference mechanism would have
also privileged EU too much, and destroy the concept of EU accession to ECHR on ‘equal
footing’, as the Member States’ senior courts have no such authority under the current Convention
system. Finally, Gragl points that—referring to Opinion 1/91 and 1/09—such reference procedure
would prejudice the role of national courts to refer preliminary reference questions and therefore
deprive them from this natural role which they currently possess under the EU Treaties. See: Gragl
(2013), p. 237–239. A similar argument on the reference procedure under Protocol 16 ECHR has
been said by the Luxembourg Court in its recent Opinion 2/13; Cf. also: According to Gragl,
another option mentioned in the negotiations process was to devise the mechanism of a reference
for an ‘opinion’ via which the Strasbourg Court requires an opinion from the Luxembourg Court
for the question of compatibility of the alleged violation with the EU legal basis and EU law in
general. The unenthusiastic aspect of this proposal, however, would be that the Luxembourg Court
would merely provide an opinion but would have not been able to remedy the violation. That being
the case, Gragl argues that such mechanism would have not been in compliance with the
Strasbourg Court’s subsidiary character (Art. 35.1 ECHR). See: Gragl (2013), p. 240/1.
8.2 A Background on the Subsidiary Nature of the Convention System from the. . . 297

8.2 A Background on the Subsidiary Nature


of the Convention System from the Perspective
of EU as a Party to the Latter

The Convention system, to be noted, is embedded on the principle of subsidiarity,3


which implies that it is primarily for the Convention High Contracting Parties to
ensure the respect for the Convention’s rights whereas the Strasbourg Court is
thought as an auxiliary intervener only if one of the High Contracting Parties has
not been able to ensure the respect for the rights concerned. This being the systemic
code whereupon this system envisages the intervention of the Strasbourg Court, the
Convention applies the rule of exhaustion of domestic legal remedies as a condition
for one case to enter the Strasbourg Court review.4 Although the EU legal order—in
this new picture—‘may be compared in more than one respect with a domestic legal
order, it also counts states as its subjects and regards itself as an order of interna-
tional law, albeit a new one.’5
Since this rule applies mostly to individuals, but also in inter-party cases, it is of
utmost importance to shed light on its structure and practical application in multi-
faceted legal environments like the EU and its Member States post-accession. To
note, the rule on exhaustion of legal remedies causes numerous problems to parties
which have limited access to courts on multiple circumstances, but also makes a lot
of expenses to them in order to follow the entire domestic court structure in
sequence for the case to be eligibly exhausted to enter the Strasbourg gates.
Individuals, therefore, must pass numerous and sometimes very complex legal
procedures to ensure that they are eligible to enter the Strasbourg review procedure.
Only in this way, however, one can portray the Strasbourg system as an external and
subsidiary human rights control mechanism, rather than a regular one. And that
would in the case of EU play a core role in order to better shield EU law autonomy
from the Convention’s potential interferences.6
Likewise, the application of the rule on exhaustion of legal remedies causes
certain problems within the context of Strasbourg Court mechanisms, as this stands

3
On the principle of subsidiary and ECtHR, see e.g.: Krüger (2002–2003), p. 96; See also: Hart
(2010), p. 552.
4
On the rule on exhaustion, see e.g.: Ribble (2010–2011), p. 214; The International Court of
Justice has argued in Interhandel Case that in order to get access to a tribunal of international law,
it should be first the national tribunals which should be given the chance to rule on the issue. See:
International Court of Justice, Interhandel Case (Switz. v. U.S.), [1959] I.C.J. 6, 27 (Mar. 21); One
should also note that in the hearing sessions on Opinion 2/13, the Commission seemed to argue that
one of the safeguards of the EU law autonomy is the rule on exhaustion which would ensure that
the Strasbourg Court could not rule on EU law compliance with the Convention without the
Luxembourg Court having had the opportunity first to do that.
5
Besson (2009), p. 239.
6
On the latter, see e.g.: Jacque (2011), p. 1016/7.
298 8 EU Prior-Involvement Review

as a core admissibility criterion for every submitted case.7 However, a certain degree
of simplicity has been managed for decades by the Strasbourg Court as the Conven-
tion system had only two levels: one, the Convention High Contracting Parties, and
two, the Strasbourg Court. With the EU accession to ECHR the legal-remedy
environment for the application of the rule on exhaustion will become even more
problematic, and genuinely complex to be interpreted by both the Strasbourg and
from time to time Luxembourg Court as well. In the framework of interaction
between Strasbourg and Luxembourg courts, the application of the rule on exhaustion
of legal remedies needs certain clarifications and thorough assessments to ensure that
both Luxembourg retains its autonomy to rule of EU law remedies and Strasbourg
ensures parties are given due access to its system of subsidiary review.
Notwithstanding the usual path that applicants use to exhaust legal remedies
within the EU system8 to get access to the Strasbourg Court, a certain degree of
uncertainty or loophole exists even when it comes to issues that interlink EU-Member
States and EU law, which stand unclear sometimes even within the context of
EU-Member States legal orders relations. With this in mind, the application of the
rule on exhaustion of legal remedies does not only implicate many matters within the
context of Luxembourg Court’s struggle to retain its autonomy but also involves
uncertain issues that might exist within the EU-Member States legal orders shared
competences.9 Therefore, it is necessary to deconstruct thoroughly the application of
this rule within the context of interaction between Luxembourg, Strasbourg and EU
Member States, at which point one needs to explain the manners and interpretations
that need be used in this regard. Two routes exist to exhaust legal remedies in
EU-originated case: one, the direct actions, which claimants exhaust by filing their
applications directly before the Luxembourg Court, and two, indirect action, namely
the preliminary reference procedure, which claimants exhaust by filing an application
to the national court and asking the latter to refer preliminary questions to the
Luxembourg Court.10 As both routes are well examined by the existing literature,
we will merely discuss here why the current system of legal remedies in the EU is not
compliant with the ECHR standards, and then embark on the examination of the prior
involvement review of the Luxembourg Court.

7
E.g.: Lock (2010), p. 788; Hart (2010), p. 551.
8
The EU law being itself a regime/system of law, with fully fledged structural organization. See
generally: Harding (2000), p. 130.
9
One should add that the appeal to the European Ombudsman may not be qualified as a remedy
that can make a party eligible to enter Strasbourg gates. See e.g.: Lock (2010), p. 789.
10
For a general note on this, see e.g.: Craig (2013), p. 1125; See also Callewaert (2014), p. 30.
8.2 A Background on the Subsidiary Nature of the Convention System from the. . . 299

8.2.1 Limited Access to Justice Under EU-Law Direct


Actions: In the Borders of Violation of the Right
to Access the Court Through an Effective Legal
Remedy

As well known, the space for direct actions under EU law remains rather limited for
private parties11 resulting into serious limitations to the right to access the EU Court
through effective legal remedies.12 The narrowed space of ‘the judicial restraint can
be justified by the ECJ’s desire to keep the EU immune to challenges.’13 This
limited space for private parties does critically implicate the exhaustion of domestic
legal remedies that the former needs to fulfill in order to obtain access to the
Strasbourg Court.14
The main criticism on the EU direct actions relates to the fact that there is a clear
gap in the efficiency of judicial protection by the EU Court,15 something the
claimants and AGs have also examined in Uni on de Peque~ nos Agricultores and
Commission v Jégo-Quéré.16 Asked to surpass this gap, the Court has argued that it
may not exceed the borders of allowed authority by the Treaties, even if that gap
would risk the application of Art. 47 of the Charter of Fundamental Rights of the
EU and Art 6 & 13 ECHR.17 On the other hand, if one would push the argument that
although the EU Court could limit the access to court of certain parties, the latter
might still use national courts as indirect means of access to the EU Court, does not
stand practically true. To denounce this, the Court has ruled in Uni on de Peque~
nos
Agricultores that the admissibility of an action for annulment before it is indepen-
dent of the fact whether there is any legal remedy available before the national

11
Phelps (2006), p. 273; Balfour (2005), p. 31; The same can also be said about actions for
damages, see e.g.: Ward (2001), pp. 423–424; See also: Eliantonio and Stratieva (2003), p. 3.
12
Ragolle (2003); See also: Usher (2005); Gormley (2006); Craig and De Búrca (2008), p. 512,
which generally argues that this makes it virtually almost not possible to use direct actions; See
also: Analysis (1997), p. 242; Landau (2008), p. 570; Phelps (2006), p. 277.
13
Eliantonio and Stratieva (2003), p. 11.
14
Cf.: Court of Justice of EU, Les Verts v European Parliament, C-294/83 [1986] ECR 1339, para.
23, where the Court argued that ‘the Treaty established a complete system of legal remedies and
procedures designed to permit the Court of Justice to review the legality of measures adopted by
the institutions. Natural and legal persons are thus protected against the application to them of
general measures which they cannot contest directly before the Court by reason of the special
conditions of admissibility laid down in the second paragraph of [. . .Art. 263(4). . .] of the Treaty.’
This is still not a complete answer, as most regulations are directly applicable and not of general
nature but do not pass the test of individual concern, which in fact makes it not possible for private
parties to attack them before the EU Court.
15
Albors-Llorens (2012), p. 534; See also: van den Berghe (2010), p. 113.
16
Court of Justice of EU, Case C-263/02 P Commission v Jégo-Quéré [2004] ECR I-3425.
17
Court of Justice of EU, Case C-263/02 P Commission v Jégo-Quéré [2004] ECR I-3425, note
16, para 36; See also: Court of Justice of EU, Order of 9 January 2007 in Case T-127/05 Lootus
Teine Osa€ uhing v Council, para 50.
300 8 EU Prior-Involvement Review

court.18 Even if a party would prove that there is no remedy under national law to
attack indirectly an EU act which affects him/her faultily, the EU Court would not
take this as a factor to determine the admissibility of any direct action. In addition,
as is well known, the Court has ruled that private parties are not entitled to any right
to force the national court to address their case through the preliminary reference
procedure before the EU Court.19 Those who might therefore become prohibited
standing for direct actions based on the argument that preliminary reference
procedure provides a remedy, might be totally denied access to justice if the
national court refuses to refer a question to the EU Court.20 This being the case,
if a certain private party is wrongly affected by an EU act which entails no Member
State implementing measures, he/she would need to pass the tough test on direct
and individual concern; the latter being too difficult to be passed when legislative
acts are in question.
AG Jacobs has argued in UPA that the individual concern test seems to go
beyond the redline of a minimum effective judicial protection, as it is illogical to
seek the ‘requirement that an individual applicant seeking to challenge a general
measure must be differentiated from all others affected by it in the same way as an
addressee.’21 He argues that the dose of irrationality becomes even higher when one
argues that with the individual concern test the broader the number of persons
affected by an EU act less probable the right of a person to pass this test.22
Therefore, one affected by a certain measure is not simultaneously individually
concerned. The EU Court has for example ruled in Deutsche Lebensmittelwerke
that there is a clear difference between being affected and being individually
concerned by arguing that:
Although the contested decision, affects the applicants, that is only because of the effects it
produces on their position in the market. In that regard, the decision is of concern to the
applicant just as it was to any other person supplying margarine on the West Berlin market
while the contested operation was in progress, and it is not therefore of individual concern
to them.23

Not that the Plaumann formula is far from accepting someone to pass the test on
basis of ‘being affected’, but even the minimum proposals for changing it remain
still far from what one can expect. AG Jacobs instead proposes that the individual
concern test be passed merely if a person is adversely affected,24 by reason of

18
Court of Justice of EU, Union de Peque~nos Agricultores v Council, Case C-50/00 P [2002] ECR
I-6677, para 46.
19
Court of Justice of EU, C.I.L.F.I.T. v Ministry of Health, Case 283/81, Judgment OF 6.10.1982,
para. 9.
20
Koch (2005), p. 515; See also a similar argument, at: Lock (2010), p. 784.
21
AG Jacobs Opinion in UPA, Case C-50/00 P, delivered on 21 March 2002, para. 59.
22
AG Jacobs Opinion in UPA, Case C-50/00 P, delivered on 21 March 2002, note 21, para 59.
23
Court of Justice of EU, C-97/85, Deutsche Lebensmittelwerke v Commission [1987] ECR 2265,
para. 11; See also: Balfour (2005), p. 38.
24
E.g.: Ward (2001), p. 433.
8.2 A Background on the Subsidiary Nature of the Convention System from the. . . 301

particular circumstances, by that act or measure. Such adverse effect would allow
that persons be given effective legal protection through the EU Court merely if they
were faultily affected by the challenged act or measure, therefore not needing to
breach the law as a condition to get access to the Court. Such a proposal would
make it possible to dismiss the very narrow Plaumann test which had left persons
without substantive remedies to challenge adversely affecting EU acts needing no
implementing measures. This proposal of AG Jacobs has of course been refused by
the Court in UPA, therefore arguing that the current access to direct remedies in the
EU law remains rather limited. A more liberal interpretation of the individual
concern, nevertheless, remains a precondition for the enjoyment of the right to
access the court in the EU.25 The Plauman formula clearly becomes a hurdle which
hugely impairs individual standing.26 To consider the latter, if the Luxembourg
Court would defer to its own arguments e.g. in Jusuf case, for the same substance if
not for the same motives and formal procedures, it would have to call its system of
remedies as infringing private parties’ fundamental rights.27 This would also
contradict with the call of Van Gerven who seeks more judicial protection in
trade off for lack of democratic legitimacy,28 with EU being an example of serious
democratic deficits.
Finally, it should be noted that the EU Court, through the direct actions, controls
the legality of EU acts on basis of facts and law which existed in the time when the
challenged EU act was adopted.29 This approach clearly limits the full judicial
protection of individuals, as the ECHR has installed the standard that rights should
be assessed on present situations,30 the ECHR itself being a living instrument.31
Besides, the Strasbourg Court has loudly prescribed in Beer and Regan v. Germany
that:
To read [. . .] the Convention and its guarantee of access to court as necessarily requiring the
application of national legislation in such matters would, in the Court’s view, thwart the

25
Which may certainly violate Art. 6 and Art. 13 ECHR. See: Balfour (2005), p. 35; Adding here
that e.g. the German jurisprudence does not restrain itself to consider itself constitutionally
authorized to review EU acts if they rebut the principle of solange human rights protection. On
this tension which partially derives also from the lack of reliable access of private parties to the EU
Court, see e.g.: Payandeh (2011), p. 15; See also: Miiller (2007), p. 59 et seq, on Solange’s current
understanding in European human rights law.
26
See also criticisms at: Ragolle (2003), p. 90.
27
On human rights approach of the Court on Jusuf case, see: Miiller (2007), p. 59; Cf.: Heringa and
Verhey (2011), p. 20, which states that ‘. . .there seems to be a related aspect which should be
considered. Article 47 entitles everyone to ‘an effective remedy before a tribunal’. If we can safely
assume that Article 6 of the ECHR is fully incorporated within Article 47, then Article 47 requires
the establishment of court proceedings in which a court can give binding decisions which provide
for adequate relief to the petitioner.’.
28
van Gerven (2005), p. 63.
29
Court of Justice of EU, France v Commission, Joined Cases 15/76 and 16/76 [1979] ECR
321, para 7.
30
On the Convention as a modern-day instrument, see e.g.: Heringa and Verhey (2011), p. 19.
31
Shelton (2003), p. 126.
302 8 EU Prior-Involvement Review

proper functioning of international organisations and run counter to the current trend
towards extending and strengthening international co-operation.32

It is obvious that Beer and Regan v. Germany does not accept limitations in
national legislation for claimants’ access to court as valid under the Convention as
that would run counter to the obligation of the high contracting parties to develop
international cooperation. Therefore, one needs to argue that the limitation set in
France v Commission limits the rights of claimants to have their rights judged
against present factual and legal circumstances, rather than having to adhere to the
principle of assessing the validity of EU acts at the moment when they were
adopted—adding that the Strasbourg Court, based in Beer and Regan
v. Germany, would consider that such limitations need be removed33—putting the
latter in the context of the known fact that becoming a party to the Convention
needs be traded off with certain domestic changes that may affect the Convention’s
efficacy; recalling that the France v Commission practice could lead to a model of
assessing the validity of potential infringing acts based on an artificial method,
which does not evaluate the real interference that the individuals’ rights could
receive.34 As a comparative example, the Inter-American Court of Human Rights
would have been in the same line with Strasbourg’s Beer and Regan, as it had ruled
that effective access to court means not only to ‘organize the governmental appa-
ratus [. . .but also. . .] all the structures through which public power is exercised, so
that they are capable of juridically ensuring the free and full enjoyment of human
rights.’35 That speaks for the needs of the EU to carefully consider reframing its
legal remedies’ system in order to comply with the international standards on access
to effective judicial protection. It is a pity that the Luxembourg Court remains
absolutely silent on this issue in its Opinion 2/13, adding that this topic remains of a
crucial relevance for the efficiency and quality of the accession procedure and its
practical operation.

32
ECtHR (GC), Beer and Regan v. Germany, Application No. 28934/95, judgment of 18 February
1999, para. 62.
33
Cf.: ECtHR, Michaud v. France, Application No. 12323/11, judgment of 6 December 2012,
para. 111, where the Strasbourg Court absurdly and weirdly calls the EU system of legal remedies
limited but comparable to that of the Convention system.
34
On the latter part of the argument, see: Balfour (2005), p. 39 (‘[. . .] the individual is currently
unable to enforce these rights in an effective and practical manner.’).
35
Inter-American Court of Human Rights, (Exceptions to the Exhaustion of Domestic Remedies in
Cases of Indigency or Inability to Obtain Legal Representation Because of a Generalized Fear
Within the Legal Community (Arts. 46(1), 46(2)(a) and 46(2)(b) of the American Convention on
Human Rights), Advisory Opinion OC-11/90 of August 10, 1990, para. 23 [emphasis added].
8.2 A Background on the Subsidiary Nature of the Convention System from the. . . 303

8.2.2 Whether Preliminary Reference Procedure Under Art.


267 TFEU Complies with the ECHR Standard
on Access to Court Through Efficient Legal Remedies?

The Court has ruled in UPA that every Member State must provide, ‘so far as
possible’, applicants with sufficient space to challenge national implementing acts/
measures deriving from EU law,36 as a means to ensure that the national court could
then utilize the preliminary reference procedure.37 This is now also a matter of the
Treaties, as Art. 19 TEU asserts that ‘Member States shall provide remedies
sufficient to ensure legal protection in the fields covered by Union law.’38 AG
Jacobs, in UPA case, has fundamentally examined the efficiency of legal remedies
under Member States’ domestic law as means to have questions of EU law heard
incidentally before the EU Court: all this within the framework of the right to
effective legal remedy and protection. AG Jacobs argues that in the time when the
EU Court favors most issues to come through the national courts’ preliminary
reference procedure, there are examples when there is almost no legal remedy
available before the national courts whatsoever.39 The Court in UPA went even
further by arguing that it is not for the EU Court to examine whether there are

36
Court of Justice of EU, Union de Peque~ nos v Council Case C-50/00 P [2002] ECR I-6677, note
18, para. 42.
37
See: Komárek (2013), p. 436, on the so called ‘parallel references’, a situation in which a certain
measure is not merely against the EU law but also the national constitutional provisions. In that
case, the question would have to be referred simultaneously to the EU Court and the national
constitutional court, each of them having the possibility to give diverging verdicts thus complicat-
ing the issue even more. See e.g. Czech Constitutional Court, Decision on Admissibility, 2008/12/
02 – Pl. ÚS 12/08: Non-Applicability of Contested Provision, available at: http://www.usoud.cz/en/
decisions/?tx_ttnews%5Btt_news%5D¼483&cHash¼6b8830d1a16e4ba14dcbc3b678bb7bb2.
38
Emphasis added. Accord: The Luxembourg Court has tried to argue that the failure to offer
sufficient national legal remedies to access the court violates the principle of loyal cooperation
under the Treaties, as follows: ‘Applying the principle of cooperation laid down in Article [4
(3) TEU], it is the national courts which are entrusted with ensuring the legal protection which
citizens derive from the direct effect of the provisions of Community law.’ See: Court of Justice of
EU, Rewe-Zentralfi nanz and Others v Landwirtschaftskammer f€ ur das Saarland, Case 33/76
[1976] ECR 1989, para. 5.
39
AG Jacobs Opinion in UPA, Case C-50/00 P, delivered on 21 March 2002, note 21, para 37; Cf.:
The EU Court, on the other hand, has ruled that ‘the principle of effective judicial protection of an
individual’s rights under Community law must be interpreted as meaning that it does not require
the national legal order of a Member State to provide for a free-standing action for an examination
of whether national provisions are compatible with Article [56 TFEU], provided that other
effective legal remedies, which are no less favourable than those governing similar domestic
actions, make it possible for such a question of compatibility to be determined as a preliminary
issue, which is a matter for the national court to establish.’ See: CJEU, Unibet v Justitiekanslern,
Case C-432/05 [2007] ECR I-2271, para. 65; I take the view that although this approach of
argument seems rightly stressed by the EU Court, there are too many obstacles in practice
which make this proclamation merely of a theoretical nature.
304 8 EU Prior-Involvement Review

effective legal remedies in Member States as a precondition of admissibility for an


action for annulment, as, the Court argued, ‘such an interpretation cannot have the
effect of setting aside the condition in question, expressly laid down in the Treaty,
without going beyond the jurisdiction conferred by the Treaty on the Community
Courts.’40 On the other hand, the EU Court paradoxically requested the Member
States, beyond what the Treaties establish, to reform their system of legal remedies
in order to accommodate claimants’ right to the EU Court through the incidental
review. The IATA and ELFAA41 case does clearly bound the national court to
proceed with the preliminary reference procedure only if there are ‘well founded’
arguments to trust that an EU act could be invalid; Such huge scope of appraisal—
certainly not a minimum bar of e.g. one argument claiming invalidity of an EU
act—makes it very unfounded to trust that the national court would request an EU
act to be reviewed for its validity by the EU Court merely for simple doubts on its
legality. If the latter stands true, the preliminary reference procedure can only be
considered a high-level procedure which can only be used if extreme doubts are
proved against the validity of an EU act.42
Both AG Jacobs43 and AG Darmon44 have argued that the caselaw of the EU
Court does require that every claimant have access to an effective legal remedy if
he/she is wronged by an act of the EU, which impairs his/her rights or benefits in
order to receive judicial protection for such infringement. Such requirement also
stands in line with the Court’s standard to rest in the same line with the Conven-
tion’s rights, most concretely Art. 6 & 13 ECHR and the Charter of Fundamental
Rights of the EU provisions as well.45 AG Jacobs did contest the full availability of
this right under the Member States legal orders,46 arguing that this situation can
impair the right of claimants to access effectively the EU Court through the national
courts’ preliminary reference procedure. This said, the Court’s argument in UPA
that ‘the Treaty has established a complete system of legal remedies and procedures
designed to ensure judicial review of the legality of acts of the institutions, and has

40
Court of Justice of EU, Uni
on de Peque~
nos Agricultores v Council, Case C-50/00 P [2002] ECR
I-6677, note 18, para 44.
41
Court of Justice of EU, IATA and ELFAA, Case C-344/04 [2006] ECR I-403, para. 30 [Emphasis
added].
42
Cf.: ECtHR, Michaud v. France, Application No. 12323/11, judgment of 6 December 2012,
para. 115, where the Strasbourg Court ruled that if the case has not gone through the review of the
Luxembourg Court (because the national court did not refer a preliminary reference question), the
state’s act—although stemming from an EU law obligation—is not deemed as shielded with the
equivalent protection doctrine.
43
AG Jacobs Opinion in UPA, Case C-50/00 P, delivered on 21 March 2002, note 21, para 38.
44
Opinion of AG Darmon in Borelli, Case C-97/91 [1992] ECR I-6313, para 31.
45
AG Jacobs Opinion in UPA, op. cit. note 115, para. 39; Court of Justice of EU, Commission v
Austria, Case C-424/99 [2001] ECR I-9285, para 45; See also: Court of Justice of EU, Johnston,
Case 222/84 [1986] ECR 1651, para 18; See also: Balfour (2005), p. 41.
46
AG Jacobs Opinion in UPA, Case C-50/00 P, delivered on 21 March 2002, note 21, para. 40.
8.2 A Background on the Subsidiary Nature of the Convention System from the. . . 305

entrusted such review to the Community Courts’ seems almost insecure.47 Such
insecurity does not merely derive from the Treaties system but also from the overall
complications produced in the division of duties between the EU and Member
States in cases when the EU acts need be enacted by national implementing
measures. It is clear that the Court’s argument in UPA goes in contradiction with
the assertions of the same Court in C.I.L.F.I.T.48 and IATA49 that the preliminary
reference procedure is not a right of the parties in procedure before a national
court,50 but rather a supporting discretion of the latter.51 To prove this hesitation,
the Court sometimes has tried to deny itself and to call the preliminary reference
procedure as a remedy, clearly excluding the argument on how it may be of such a
nature.52
The Foto-frost formula makes it clear that national courts have no authority to
call invalid an act of the EU. Such a formula makes it necessary for individuals to
have access to a judicial protection system wherein the EU act adversely affecting
their rights be judged for its legality, which certainly brings the argument that
access for these individuals should be proved to the EU Court. AG Jacobs argues
that effective judicial protection indicates a system wherein individuals are given
access to a legal remedy if they are negatively affected by a legal measure or act.53
This said, Art. 267 TFEU is not a right of the claimants but a competence of the
national court.54 AG Jacobs tests this claim by arguing that national courts can
refuse55 to proceed with a preliminary reference procedure even if that is a request

47
Court of Justice of EU, Uni on de Peque~ nos Agricultores v Council, Case C-50/00 P [2002] ECR
I-6677, note 18, para. 40; See also: Court of Justice of EU, Les Verts v Parliament, Case 294/83
[1986] ECR 1339, para 23.
48
Court of Justice of EU, C.I.L.F.I.T. v Ministry of Health, Case 283/81, Judgment OF 6.10.1982,
para. 9.
49
Court of Justice of EU, IATA and ELFAA, Case C-344/04 [2006] ECR I-403, note 41, para. 28.
50
E.g.: van den Berghe (2010), p. 125; See also: Discussion document of the Court of Justice of the
European Union on certain aspects of the accession of the European Union to the European
Convention for the Protection of Human Rights and Fundamental Freedoms, para 10 (‘. . .it is not
possible for the parties to set this procedure in motion.’); See also: Craig (2013), p. 1125, arguing
that the claimant can in no way be in control of this procedure; Odermatt (2014), p. 28.
51
The EU Court has itself accepted that the national courts enjoys the widest discretion within this
context. See: Court of Justice of EU, Rheinm€ uhlen-D€ usseldorf v Einfuhr- und Vorratstelle f€
ur
Getreide, Case 166/73 [1974] ECR 33, para. 4.
52
Court of Justice of EU, FMC and Others v Intervention Board For Agricultural Produce And
Another, Case C-212/94 [1996] ECR I-389, para 56.
53
Albors-Llorens (2012), p. 514.
54
It is suggested that the better use of state liability against Member States which fail to use the
preliminary reference procedure before the Luxembourg Court when the case involves a substance
of EU law may seem a good strategy to force national courts to better apply Art. 267 TFEU. See on
this: O’Meara (2011), p. 1830/1.
55
See: Carrubba and Murrah (2005), p. 403 et seq, which shows which are those states more likely
to use the preliminary reference procedure and reasons for such behavior; See also: Broberg and
Fenger (2013), p. 501, who suggests that reasons for refusals to proceed with preliminary reference
questions may be structural but also behavioral, making the point that there are clear differences
306 8 EU Prior-Involvement Review

of the claimant,56 or if the latter would insist to have EU law issues contested in
his/her case before the national court57; adding here the fact that such a procedure
takes too much time until it becomes replied by the EU Court, which certainly
interferes with the principles of legal certainty.58 Moreover, arguing that this cannot
be considered a right, AG Jacobs argues that the national court—even if accepting
the claim of the applicant that his/her case involves the validity of an EU act—can
mistakenly accept the contested EU act as valid therefore not proceeding with the
preliminary reference question to the EU Court.59 Furthermore, the national court
has full discretion to evaluate whether the case of the claimant involves issues of
EU law, therefore a mere evaluation by the national court that the case does not
encompass any EU law issue suffices to reject to consider the preliminary reference
procedure as a need for the case to be resolved. Alter proposes a very good example
to support this claim. He argues that: ‘If the lower court, however, did not think that
it would like what that other parent might say, it could follow the “don’t ask and the
ECJ can’t tell” policy and not make a referral.’60 On the other hand, even if the
national court decides to proceed the question for a preliminary reference by the EU
Court, two things remain still uncertain here. First, the national court would still
have full discretion to formulate the question that will be processed for the
preliminary reference before the EU Court,61 such question being rather abstract
and not relating to the very petitum and allegations of the case and claimant. The
latter certainly makes it imperative to argue that the preliminary reference answer
of the EU Court does not resolve the case but rather answers more generally to the
question of the national court with the purpose of resolving the question of the latter
and not the case of the claimant. Second, the national court is not obliged to follow
the arguments of the claimant as to what is in fact the contested EU act and how the
question should be formulated for having the EU Court answering the very issue in
question. That full discretion to formulate the question that will be proceeded to the

between member states’ courts as to the use of this mechanism (therefore confirming our
argument); See also generally: Lock (2010), p. 791.
56
An interesting suggestion of a European Parliament Resolution on this scenario says (although
one should not take it as being legally acceptable): ‘[. . .] for the purposes of complying with the
requirement set out in Article 35 ECHR for domestic remedies to have been exhausted, the
applicant shall have exhausted the judicial remedies of the State concerned including a reference
for a preliminary ruling to the Court in Luxembourg; the latter procedure shall be regarded as
having been complied with where following a request to that end by the applicant the national
court does not consider it appropriate for a reference for a preliminary ruling to be made.’
European Parliament resolution of 19 May 2010 on the institutional aspects of the accession of
the European Union to the European Convention for the Protection of Human Rights and
Fundamental Freedoms (2009/2241(INI)), para. 10.
57
On how easy it is for a constitutional court to refuse referring a question to the EU Court, see:
Pérez (2012), p. 123/4.
58
AG Jacobs Opinion in UPA, Case C-50/00 P, delivered on 21 March 2002, note 21, para. 42.
59
AG Jacobs Opinion in UPA, Case C-50/00 P, delivered on 21 March 2002, note 21, para. 42.
60
See: Alter (1996), p. 466/7.
61
AG Jacobs Opinion in UPA, Case C-50/00 P, delivered on 21 March 2002, note 21, para. 42.
8.2 A Background on the Subsidiary Nature of the Convention System from the. . . 307

EU Court gives the national court room to escape from many EU acts/provisions
which could have changed the answer of the EU Court on the specific case being
adjudicated by the national court. Such discretion of the national court puts in doubt
the scope of challenging that the preliminary reference procedure can supply to
answer the claimants’ allegations on EU law disputes. Apart from the above
arguments, indirect access to the EU Court through the national court would be
impossible if the latter calls valid the contested EU act. This situation does of
course, as explained in the previous subchapter, bar claimants from enjoying even
an indirect access to the EU Court if the quest for invalidity is rejected at the
national court.
From another point of view, it must be noted that the gateway for claimants to
have indirect access to the EU Court is for them to be able to access first of all a
national court. Accessing the latter in most cases needs to challenge an
implementing measure of the Member State, such implementing measure deriving
from an EU act. Although some national courts have recognized claimants’ right to
remedy even though there was no national implementing measure taken by that
Member State,62 normally there are little opened gateways to access a national
court if that Member State has not taken any measure which is challenged for its
unfavorable consequence on the claimants’ interests.63 So, the question is how to
assure that claimants can first access the national court if they are trying to attack an
EU act/measure which entails no implementing measure by the Member States. The
lack of such a national measure which forms the ground for the claimant to start the
procedure before the national court is a barrier that cannot be surpassed. On such
cases, the only way to get access to the national court if there is an EU act entailing
no implementing measure by the Member States is to violate that EU act
(e.g. regulation) and then request that such violation be reviewed in the request
for invalidity of the EU act.64 In all those cases, claimants therefore would need to
violate the EU act in order to get access to the national court; such precondition
makes it illogical to consider that individuals should first violate an EU act in order
to get access to the national court, all this being too vague and uncertain for the
applicants ‘right’ to an effective legal remedy. ‘Individuals clearly cannot be
required to breach the law in order to gain access to justice.’65 Making national
legal remedies effective on the cost of breaching the law would contradict the very
rationale of the ECHR angle of access to court and effective legal remedy as a
means to valuable judicial protection for individuals. Rightly, it should be noted,
the Court has also ruled in Traghetti del Mediterraneo that the failure of a national

62
E.g. see UK court at: Court of Justice of EU, Imperial Tobacco and others, Case C-74/99 [2000]
ECR I-8599, para. 2.
63
Cf.: Peers (2006), p. 454, who argues that by ruling on the equivalent protection presumption in
Bosphorus, the Court somehow recognized that equivalency with the accepted fact that the
preliminary reference procedure reaches a basic level of human rights protection as well.
64
AG Jacobs Opinion in UPA, Case C-50/00 P, delivered on 21 March 2002, note 21, para. 43.
65
AG Jacobs Opinion in UPA, Case C-50/00 P, delivered on 21 March 2002, note 21, para. 43; See
also: Eliantonio and Stratieva (2003), p. 4.
308 8 EU Prior-Involvement Review

court to refer the question to the EU Court, if it is later proved that such an action
was illegal, could lead to state liability under the Frankovic formula.66 This is not
that much the concern of this chapter, however.
Seen from the perspective of loyal cooperation and uniform application of the
EU law, one can rightly argue that the system of legal remedies which Member
States should provide on basis of Art. 19 TEU should at least be so as to assure that
all individuals in EU Member States have equal indirect access to the EU Court, in
order to fulfill the requirement of non-discrimination and fair access to court under
ECHR.67 As access to Art. 267 TFEU from different individuals of Member States
is not the same—seeing that no Member State offers exactly the same system and
scope of remedies which would lead to the preliminary reference procedure—that
‘would inevitably lead to inequality and a loss of legal certainty in an area of law
already marked by considerable complexity.’68 Certainly, this unequal treatment of
all individuals among all Member States’ system of legal remedies could amount to
the violation not only of the right to effective judicial protection under ECHR but
also could put into question the very lawfulness of application of the EU act
contested by claimants who were subject to dissimilar and/or asymmetrical systems
of legal remedies in Member States.69

66
Court of Justice of EU, Traghetti del Mediterraneo, Case C-173/03, judgement of 13 June 2006,
[2006] I-5177; Court of Justice of EU, K€ obler, Case C-224/01 [2003] ECR I-10139. There is a
tendency of the Luxembourg Court to penalize national courts if they misused the application of
the doctrine of acte clair as a means to escape the Luxembourg’s review while breaching the EU
law. See: Groussot and Minssen (2007), p. 385; This might sound positive as to the criticisms of
the huge discretion of national courts to decide whether to refer or not a question for preliminary
reference.; On the horizontal effect of human rights under Francovic, see also: Shelton
(2003), p. 112.
67
Cf.: Jacque (2011), p. 1019, who proposes that EU should ensure through internal rules that
everytime there is an EU issue the national court be forced to proceed with the preliminary
reference question to Luxembourg Court. This, he argues, would then better exhaust legal
remedies for that applicant if he/she chooses to file an application before the Strasbourg Court;
See also a general analysis, at: Craig and de Búrca (2003).
68
AG Jacobs Opinion in UPA, Case C-50/00 P, delivered on 21 March 2002, note 21, para. 53; Cf.:
Court of Justice of EU, Melki, joint cases C-188 and 189/10, judgment of 22 June 2010, [2010]
I-5667; To note, in SNUPAT the Luxembourg Court has noted that the principle of legal certainty
must be combined with the principle of legality, which makes really narrow the scope of a general
principle of law on legal certainty compared to democratic standards. See: Groussot and Minssen
(2007), p. 389.
69
This is further supported by the fact that the Court in Åkerberg Fransson case has followed a
narrow but negative legislating approach to defining the scope of the Charter of Fundamental
Rights application on Member States actions. Such narrow scope based on an abstract list of rights
makes the Charter as not very certain in guaranteeing some of the important human rights. Thym
seems to consider Fransson as rather narrow approach, for which I would rather disagree as
explained supra in the section on the Charter on Fundamental Rights (as, compared to the new
case-law on this issue, Fransson seem to have provided for a very broad approach on the scope of
Art. 51 ChFR). On the first argument above, see: Thym (2013b), p. 394; On the impaired system of
legal remedies, see also: van den Berghe (2010), p. 113.
8.2 A Background on the Subsidiary Nature of the Convention System from the. . . 309

It should be noted that even if there is an unequal or asymmetrical system of


legal remedies among EU Member States that give access to the preliminary
reference procedure before the EU Court, the latter cannot rule and/or oblige the
Member States to offer an identical and symmetrical system of remedies. It would
be, as AG Jacobs argues, even more problematic for the EU Court to monitor the
compliance of Member States with Art. 19 (1 & 2) TEU,70 adding here that the
Treaties give no explicit authority to the EU Court to penalize practically the
national courts if they fail to offer that as this is an issue that falls within the
‘realm of national procedural autonomy.’71 The EU Court, even when finding very
rational grounds to intervene on a national court like in Alassini,72 normally
chooses to skip over.73 National courts in the last instance may not be those holding
the blame for the system of remedies as such,74 but rather the national legislation on
legal remedies. This noted, the Luxembourg Court is gradually coming to the use of
the standard of ‘negative approximation’ in view of ruling on the efficiency of
Member States’ system of legal remedies. This means that ‘the Court applies a
presumption of national autonomy in the provision of judicial protection, and the
principle of effectiveness justifies more limited patterns of Community interven-
tion.’75 Such an approach could rather well limit and provoke the violation of Art
19 TEU, or make the latter merely a theoretical principle with almost no real
relevance in terms of the EU Court’s observance of EU law. This being the practice,
‘even if one admits that the Court has valid policy reasons for adopting a relatively
stringent approach to national remedial and procedural competence in cases involv-
ing challenges to the validity of [EU] action,’ this fact can become challenging for
the constitutional laws of Member States themselves by means of which they are
obliged nationally to provide efficient systems of legal remedies.76 In addition to
this, it is also argued that since the usual length for a preliminary answer from the
EU Court is 22 months, the length as such (e.g. Salesi v Italy) fails to meet Art. 6 &

70
AG Jacobs Opinion in UPA, Case C-50/00 P, delivered on 21 March 2002, note 21, para. 57; Cf.:
Court of Justice of EU, Alassini, joint cases C-317 to 320/08, judgement of 18 March 2010, [2010]
I-2213.
71
AG Jacobs Opinion in UPA, Case C-50/00 P, delivered on 21 March 2002, note 21, para. 58; Cf.:
Arnull (2011); See also: Court of Justice of EU, Unibet, case C-432/05, judgement of 13 March
2007, [2007] I-2271, para 41 et seq.
72
Court of Justice of EU, Alassini, joint cases C-317 to 320/08, judgement of 18 March 2010,
[2010] I-2213, para. 47 et seq.
73
E.g.: Groussot and Minssen (2007), p. 390.
74
Cf.: Lenaerts (2012), p. 382, who argues that national courts may be found liable under EU law if
they use procedural autonomy to narrow the effectiveness of applicants’ right to access the EU
courts.
75
Dougan (2004), p. 233/4; Cf.: Kakouris (1997).
76
Dougan (2004), p. 330; Cf.: Court of Justice of EU, Van Schijndel, Joint cases C-430 and
431/93, judgement of 14 December 1995, [1995] I-4705.
310 8 EU Prior-Involvement Review

Art. 13 ECHR standards,77 which certainly makes the problem even more complex
with the timing issue.78
Ex injuria jus non oritur. This understood, the preliminary reference procedure
stands in the border or within the border of violation of Art. 6 and 13 ECHR,79 as

77
Balfour (2005), p. 40.
78
Contra: Contrary to this argument, one may bring the Strasbourg’s case Pafitis and Others
v. Greece where the Court considered the lengthy procedures under EU judicature and their
compliance with the Convention standards. The case had pursued a process that costed around
almost 3 years due to the preliminary reference procedure triggered before the EU Court. Noting
this elongated proceeding, a partial blame of which should be put on the long preliminary reference
procedure before the Luxembourg Court, the Strasbourg Court ruled that it ‘cannot, however, take
this period into consideration in its assessment of the length of each particular set of proceedings:
even though it may at first sight appear relatively long, to take it into account would adversely
affect the system instituted by Article 177 of the EEC Treaty and work against the aim pursued in
substance in that Article.’ (ECtHR, Pafitis and Others v. Greece, Application no. 163/1996/782/
983, Judgment of 26 February 1998, para. 95). The Court in the end called this case ratione
personae inadmissible. One should say that, however, the Court does nowhere establish that the
elongated judicial proceedings before the EU Court comply with the Convention standards. By the
way, the Court itself noted that ‘it may at first sight appear relatively long’, hinting on the fact that
they go beyond the normal standards. The Court’s move with the above statement, I argue, is
simply intended to recognize the current system of preliminary reference without wishing to
disturb its functioning as long as EU does not accede to the ECHR (while deferring to the
Bosphorus formula). But one cannot imply from the foregoing statement that the Strasbourg
Court considers this elongated procedure as fully compliant with the Convention. Furthermore,
in his dissenting opinion of Judge Mifsud Bonnici, it is noted that ‘[t]hese three cases lasted for
such an unreasonable length of time because in all of them the court awaited the decision which
was to be given in case no. 10429/1986’ (para. 3). He went further to disagree with the Court’s rule,
by noting that ‘applicants in the three above-mentioned cases had a right to expect that their cases
would be heard within a reasonable time, irrespective of whether, in the opinion of the judges
hearing the cases, it was expedient to await the outcome of case no. 10429/1986. That case was not
heard within a reasonable time. The applicants in that case therefore suffered a breach of the basic
right guaranteed by Article 6 § 1 and I cannot agree that the applicants in the three subsequent
cases did not suffer the same kind of breach, just because their cases were made to depend on the
first one. Surely it follows – logically and juridically – that the breach of the reasonable time
requirement in the first case cannot be said to have purged the unreasonable delay in the three cases
which followed it simply because it was juridically convenient to tie them up with the previous
case which was already unreasonably long’ (para. 4). Demonstrating that the Court nowhere
implied that the entire idea of the elongated process before the EU Court was supported by it,
Judge Mifsud Bonnici further noted that ‘[t]he judgment on this point appears (at least) to imply
that the length of the proceedings in the first case serves as a justification for the length of
proceedings in subsequent cases. The violation suffered by the applicants in these three cases
cannot be so justified and these applicants should have been granted a remedy for the breach they
suffered even though it was due to the breach suffered by others before them’ (para. 5) (ECtHR,
Pafitis and Others v. Greece, Application no. 163/1996/782/983, Judgement of 26 February 1998,
Section on the Dissenting Opinion of Judge Mifsud Bonnici). Therefore, one may finally conclude
that it was not the intention of the Strasbourg Court to rule that the elongated procedure before the
Luxembourg Court is compliant with the Convention system.
79
It is argued that ECHR standard on Art. 6 and 13 is far higher than that of EU fundamental rights
law. See: Lenaerts and de Smijter (2001), p. 95; Cf.: Lawson (2005), p. 29, questioning why the
Charter did not provide for a better access to court system of rights; See also, interestingly and
8.2 A Background on the Subsidiary Nature of the Convention System from the. . . 311

there is no response on how to ensure that a contested EU act is reviewed for its
validity by the EU Court through the proceedings initiated before the national
court.80 Moreover, there is no legal remedy available against the rejection of the
national court to process a question for preliminary reference before the EU Court:
the rejection being in itself an act of impairing the right of the claimant to enjoy
uniform and accessible judicial protection under EU law.81 Besides being limited in
the use of an effective legal remedy through Art. 267 TFEU, individual applicants
are not entitled to any subsidiary right82 to contest the validity of an EU act
reviewed as valid and not pertinent for a preliminary reference procedure by the
national court.83 The EU Court itself nevertheless has no ‘authority for changing the
system of remedies and procedures established by the Treaty, which is designed to
give the Community judicature the power to review the legality of acts of the
institutions.’84 Of course, loyal cooperation—as interpreted by the Court—does not
prove to amount for a uniform horizontal application of EU law and an effective
communication between the EU court and the national courts of Member States, so
as to ensure that every individual in all Member States enjoys equal and full access
to a judicially protective system of remedies for interpretation of EU law. Opposed
to this background one adds the argument that even when the EU Court gives its
answer to the national court, it applies a certain margin of appreciation to leave the
national courts decide a bit on itself according to its own preferences the EU law
issue, which otherwise could be an exclusive competence of the EU Court.85 One
such example is Melloni where although the preliminary reference questions
submitted by the national court sought an answer as to which of the three alterna-
tives to apply, the Luxembourg Court ‘replied in terms of the absolute principle of

selectively, the Luxembourg Court deferring to the Strasbourg Court in case Brussels II-bis
Regulation. See on the latter: Morano-Foadi and Andreadakis (2011), p. 604; On the general
possibility for EU Court to violate Art. 6 & 13 ECHR, see e.g.: Balfour (2007), p. 222; Phelps
(2006), p. 277 (‘[. . .] although individuals may use national courts in an attempt to vindicate their
rights, their ability to bring an action before the ECJ is limited. Therefore, the contention that the
EC respects fundamental rights is a chimera, and a legal presumption of such should be considered
premature.’).
80
Landau (2008), p. 574 (‘Can one envisage the ECJ as a respondent in a claim, for instance, of a
violation of Article 6 of the ECHR on grounds of an unfair trial by the Community Court?’).
81
Court of Justice of EU, Heylens, case C-222/86, judgement of 15 October 1987, [1987] 4097.
82
Cf.: Court of Justice of EU, Rewe, case 33/76, judgement of 16 December 1976, [1976] 1989.
83
Tabarelli (2013), p. 352, to see how the discretion of national judges on applying the Community
law appears in practice.
84
Court of Justice of EU, Jégo Quéré, Case C-263/02 P, judgement of 1 April 2004, [2004] I-3425,
para 48.
85
Gerards (2011), p. 84; See also: Maduro (2003), p. 534, supporting the argument that this tactic
is often used to keep calm and tension-free relations of the EU Court with the national
constitutional laws.
312 8 EU Prior-Involvement Review

primacy and left many questions unanswered.’86 The said EU-internal margin of
appreciation87 could in itself limit even the indirect access of claimants to the EU
Court even when the national court chooses to utilize the preliminary reference
procedure. As a result, this may acutely add up to a ‘systemic deficiency’88 of the
rule of law in the EU, the latter being proclaimed as a Treaty-based ‘value’.
Certainly, the exhaustion of legal remedies for purposes of accessing Strasbourg
Court is almost too complex and is far from being well assured in terms of ECHR
guarantees for access to court and efficient legal remedies.
One should note that Biret is an important case wherein the Strasbourg Court
echoed on the compliance and situation of legal remedies at the EU level with
regard to Art. 6 & Art. 13 ECHR (fair trial and access to an effective legal remedy).
One may even bring Biret as a case to counter-argue the above analysis, by
maintaining that Biret has demonstrated that the EU fully complies with the
ECHR standards on effective legal remedies and fair trial.89 The author of this
book agrees that Biret is a perfect case to diminish the influence of the argument
presented above, however, there are some reasons why it is hereby argued that Biret
does not provide an answer on this issue. Let me then demonstrate what the
Strasbourg Court ruled in Biret. The Court first noted that the application’s com-
plaints come under Art. 6 and 13 ECHR, alleging that the EU judicature has
basically deprived the applicants from accessing the court through an efficient
legal remedy. To this extent, one needs to note that the claimant’s application
was addressed only against 15 Member States of the EU, but neither towards the
entire members of the EU nor against the EU itself.90 The latter is an indication that
the Court was not asked, and could have not, answered on the EU system of legal
remedies as that was not the main issue of the case. Having this in mind, the Court
called the case inadmissible by arguing that the alleged violations may not be
attributed to any of the 15 Member States since none of them intervened directly
or indirectly to that dispute. It then noted that as long as the Community has not

86
Komárek (2013), p. 433.
87
The Court held e.g. a differentiated approach in front of Member States also in Omega
Spielhallen- und Automatenaufstellungs-GmbH where it ruled: ‘The Court found that it is not
necessary that all member states apply the same standard, even if it concerns a restriction of one of
the fundamental economic freedoms on which the EU is based, as long as the restriction is
necessary and cannot be realised with less restrictive measures.’ Besselink and Reestman
(2008), p. 204 [emphasis added].
88
On the notion of ‘systemic deficiency’ in the rule of law in the EU, see: von Bogdandy and
Ioannidis (2014), p. 65 et seq; See also the Member States position on this, at: Report of the ECJ
submitted to the Inter-Governmental Conference of 1996, available at: http://europa.eu.int/comm/
eurostat/research/index.htm?http://europa.eu.int/en/comm/eurostat/research/viros/search3.htm&1;
Cf.: AIRE Centre et al. (2013), p. 2, which requests a better system of legal remedies in the EU, in
order to make the supervisory role of Strasbourg Court functional.
89
ECtHR, Biret, Application no. 13762/04, Decision as to admissibility of 09 December 2008,
para. 1.
90
ECtHR, Biret, Application no. 13762/04, Decision as to admissibility of 09 December 2008,
note 89, para. 1.
8.2 A Background on the Subsidiary Nature of the Convention System from the. . . 313

acceded to the Convention is may not be held liable under it. In that regard, the
Court concluded that the application is ratione personae inadmissible.91 Therefore,
with regard to the specific Art. 6 and Art. 13 ECHR, the Strasbourg Court neither
examined their compliance nor decided to include them under the roof of ‘equiv-
alent protection’ as it did with the second issue of the application (with regard to
Art. 1 Prot. 1 ECHR, on the peaceful enjoyment of property). The Court merely
deferred to the fact that the 15 Member States (the addressees) had no responsibility
as they took no act in the direct or indirect way, while also deferring to the fact that
as long as the Community has not acceded to the Convention it may not be held
liable under it. In this regard, one may find no convincing grounds to argue that
Biret basically ruled the EU system of legal remedies as compliant with the ECHR
standards on fair trial and effective legal remedies. The fact that the Court refused to
tackle that case in the substantive context may not be seen as arguing the contrary to
what is maintained above in this chapter. Roughly the same response would be
given also with regard to Bosphorus.92

91
ECtHR, Biret, Application no. 13762/04, Decision as to admissibility of 09 December 2008,
note 90, para. 1. This was the first issue addressed with this application. The second issue, that of
peaceful enjoyment under Art. 1 Protocol 1 ECHR, was decided as falling in the scope of
equivalent protection formula (Ibid, para. 2), therefore inadmissible as well.
92
In Bosphorus, the Strasbourg Court first clarified the EU system of legal remedies (para. 161).
Then, it made the assertion that ‘[i]t is true that access of individuals to the ECJ under these
provisions is limited: they have no locus standi under Articles 169 and 170; their right to initiate
actions under Articles 173 and 175 is restricted as is, consequently, their right under Article 184;
and they have no right to bring an action against another individual.’ (para. 162 [emphasis
added]). Then, in paragraph 163, the Court noted that nevertheless the current actions that may
be initiated before the EU Court by the community institutions represent important tools to control
the compliance with the Convention which may benefit individuals indirectly. It also noted that
individuals may ‘also bring an action for damages before the ECJ in respect of the non-contractual
liability of the institutions’ (para. 163). The Court afterwards (in para. 164) noted that the national
courts may use the preliminary reference procedure (as a complementary function) to address
certain questions on EU law to the Luxembourg Court. The Strasbourg Court was wrong in this
part to note that the preliminary reference procedure as a remedy of individuals, as this is in itself
not a right of individuals but rather a discretion of national courts (see supra AG Jacobs’
arguments in UPA). It was highlighted, however, that the EU Court (para. 164) may through
those proceedings also tackle the fundamental rights questions. In this regard, the Strasbourg Court
ruled that the system of protection of fundamental rights was ‘equivalent’ to that of the Convention
system (para. 165), therefore deferring to the ‘equivalent protection’ presumption (ECtHR,
Bosphorus Hava Yollari Turizm ve Ticaret Anonim Şirketi v. Ireland, Application No. 45036/95
(Judgment of 30 June 2005). Considering these arguments, one should first note that it is the
Strasbourg Court itself which considered the legal remedies’ system at the EU level as ‘limited’
(see supra para. 162). There seems no doubt that the Strasbourg Court nowhere recognized the
direct actions before the EU Court as being completely equivalent to the Convention system.
Basically, the Court tried to balance this ‘limited’ system of legal remedies with the fact that there
is still available the action for damages which may be used by individuals before the EU Court. On
the other hand, the Court nowhere overtly approved the preliminary reference procedure under the
EU Treaties as being in compliance with the Convention. It merely stated that such procedure
exists, and, amongst others, such procedure may consider fundamental rights questions posed by
the national court. Lastly, one may not say that the Strasbourg Court finally approved the EU
314 8 EU Prior-Involvement Review

To digress, in this regard, one needs to consider Ullens de Schooten where the
Strasbourg Court considered the question of whether it would be compliant with
Art. 6 ECHR if national courts did not pose the EU-law preliminary reference
question to the Luxembourg Court before deciding on the case before them.93 The
Strasbourg Court noted the procedure of the preliminary reference procedure under
the EU Treaties adding that it does not have an absolute nature (as noted supra). The
Court also commented on the scope of use of the preliminary reference procedure.94
By making this analysis, the Strasbourg Court ruled ‘that the Convention does not
guarantee, as such, any right to have a case referred by a domestic court to another
national or international authority for a preliminary ruling.’95 However, the Court
noted that if certain national courts have no jurisdiction to interpret a certain field of
law, like the EU law, then they must defer to the interpretation of that issue from the
court that holds exclusive jurisdiction for that (in this case the EU Court).96 This
being the issue, the Strasbourg Court ruled that ‘[t]he matter is not, however,
unconnected to Article 6 § 1 of the Convention which, in establishing that “every-
one is entitled to a . . . hearing . . . by [a] . . . tribunal established by law”, also leaves
to the competent court, in accordance with the applicable law, the task of hearing
any legal questions that may arise in the course of proceedings.’97 This interpreta-
tion, the Strasbourg Court ruled, is made to take into account the EU Treaties’
preliminary reference mechanism.98 Therefore, the Strasbourg Court ruled that

system of legal remedies as being compliant with the Convention: it merely deferred to the
‘equivalent protection’ presumption, which, as shown supra, is merely a systemic presumption
that applies as long as there is no serious ‘manifest deficiency’. One may not draw the argument
that as long as Strasbourg Court applied the ‘equivalent protection’ presumption it approved the
EU system as being compliant with the Convention, since this direction may lead to the fact that
EU will never be found liable before the Strasbourg Court post-accession. Therefore, the author of
this book continues to defer to the argument that the system of legal remedies at the EU level is
limited and not substantively in compliance with Art. 6 & 13 ECHR. Whether the Strasbourg Court
will recognize certain margin of appreciation on Luxembourg Court on these two rights and
therefore find the EU as not liable for violating the Convention is something that remains to be
seen. The fact that the EU system of legal remedies is problematic under the Convention system is
not doubtful at all from the substantive human rights perspective.
93
ECtHR, Ullens de Schooten and Rezabek v. Belgium, Applications nos. 3989/07 and 38353/07,
Judgement of 20 September 2011.
94
ECtHR, Ullens de Schooten and Rezabek v. Belgium, Applications nos. 3989/07 and 38353/07,
Judgement of 20 September 2011, note 93, para. 56.
95
ECtHR, Ullens de Schooten and Rezabek v. Belgium, Applications nos. 3989/07 and 38353/07,
Judgement of 20 September 2011, note 93, para. 57.
96
ECtHR, Ullens de Schooten and Rezabek v. Belgium, Applications nos. 3989/07 and 38353/07,
Judgement of 20 September 2011, note 93, para. 57.
97
ECtHR, Ullens de Schooten and Rezabek v. Belgium, Applications nos. 3989/07 and 38353/07,
Judgement of 20 September 2011, note 93, para. 58.
98
The Court further ruled that ‘[i]t should further be observed that the Court does not rule out the
possibility that, where a preliminary reference mechanism exists, refusal by a domestic court to grant
a request for such a referral may, in certain circumstances, infringe the fairness of proceedings – even
if that court is not ruling in the last instance. The same is true where the refusal proves arbitrary
8.2 A Background on the Subsidiary Nature of the Convention System from the. . . 315

‘Article 6 § 1 thus imposes, in this context, an obligation on domestic courts to give


reasons, in the light of the applicable law, for any decisions in which they refuse to
refer a preliminary question, especially where the applicable law allows for such a
refusal only on an exceptional basis.’99 This basically means that, in light of Art.
6 ECHR, the Strasbourg Court requires that national courts provide reasons if they
refuse to pose a question to the EU Court for interpretation of EU law. In the
negative, this may be a suggestion that the national court always refer questions to
the EU Court if there are no sufficient reasons to refuse making that request, as a
means to stand compliant with Art. 6 ECHR. The standard in Ullens de Schooten
therefore seems to suggest that the Strasbourg Court wishes to see the national
courts playing their complementary role of interpreting EU law and abiding to the
EU Court through a reasonable use of their discretionary competence. One must
point out that the Strasbourg’s argument in Ullens de Schooten is a direct interfer-
ence to EU law autonomy, and it should be the EU Court which should be the least
interested institution to see its relationship with national courts being delineated and
interpreted at the level of Strasbourg Court. By giving the jurisdiction to check
whether the national court has rightly reasoned whether the refusal to pose a
preliminary reference question to the EU Court, the Strasbourg Court will basically
engage with EU law and interpret its relationship with the national courts. Such
behavior would seem to be disliked by the EU judges. However, the mere fact that
the Strasbourg Court ruled in Ullens de Schooten that there is a duty of the national
court to reason decisions which refuse to pose preliminary reference questions to
the EU Court seems not sufficient to say that there is full certainty in the use of the
preliminary reference mechanism by national courts, and that for two simple
reasons: first, the Strasbourg Court merely requires that the decisions of national
courts refusing to pose preliminary reference questions to the EU Court be rea-
soned. This does not mean that the Strasbourg Court will engage with their
reasoning and interpret EU law obligations if such decisions are reasoned. Obvi-
ously, such fact does not mean, even if accepted, that the Strasbourg Court required
in the positive that the refusal to pose preliminary reference questions to the EU
Court amounts to an interference on Art. 6 ECHR. This was absolutely not the case;
and, second, the problem rests only partially on the national courts potential refusals
to pose questions. There are certain other loopholes with regard to the right to
access the EU Court even if such questions are made, as explicated supra. There-
fore, Ullens de Schooten does not seem to counter-argue the above analysis on the

(ibid.), that is to say where there has been a refusal even though the applicable rules allow no
exception to the principle of preliminary reference or no alternative thereto, where the refusal is
based on reasons other than those provided for by the rules, and where the refusal has not been duly
reasoned in accordance with those rules.’ (ECtHR, Ullens de Schooten and Rezabek v. Belgium,
Applications nos. 3989/07 and 38353/07, op. cit. note 228, para. 59).
99
ECtHR, Ullens de Schooten and Rezabek v. Belgium, Applications nos. 3989/07 and 38353/07,
Judgement of 20 September 2011, note 93, para. 60.
316 8 EU Prior-Involvement Review

limited scope of effective and fair trial standards that the preliminary reference
procedure provides currently under the EU Treaties.100
Finally, supposing that this legal situation harms the enjoyment of the right to
access the court and effective legal remedy on basis of Art. 6&13 ECHR, then the
question remains is there any duty of the EU to clean its domestic law barriers in
order to provide for a fully-fledged right to access the court and effective remedy
to applicants inside its legal order? There is of course an international obligation
of the EU to respect the Convention, thereby fully respect Art. 6 & 13 ECHR
as well—although there is no specific provision within the Convention which
prescribes how to do that. The American Convention on Human Rights—as a
comparative example—has a more sophisticated mechanism for ensuring that
gaps like this remain better sealed. Art. 2 ACHR for this purpose states ‘[w]
here the exercise of any of the rights or freedoms referred to [. . .in the
Convention. . .] is not already ensured by legislative or other provisions, the States
Parties undertake to adopt, in accordance with their constitutional processes and
the provisions of this Convention, such legislative or other measures as may be
necessary to give effect to those rights or freedoms.’101 If the latter provision of
ACHR would have been found in ECHR, the EU would have of course been sued
for omitting to legislate in a fundamental field of human rights law in order to
provide effective access to the court.102 However, the same rationale may be still
logically applied upon the EU103 if the latter continues to limit the right to its
citizens to access the court through an effective legal remedy. In that case, the EU
should have responded by taking active steps to ensure this appears in practice.

100
Cf.: Gragl (2013), p. 279, who advises that, post-accession, national judges should consider
from omitting from the practice of refusing to pose questions to the EU Court on matters of
EU law.
101
Emphasis added.
102
A similar construction of the duty to comply actively may be found also in Art. 1 of the African
Charter on Human and Peoples’ Rights, which reads: ‘The Member States of the Organization of
African Unity parties to the present Charter shall recognize the rights, duties and freedoms
enshrined in this Chapter and shall undertake to adopt legislative or other measures to give effect
to them.’ [emphasis added]; Accord.: Article 2 (1) of the Convention against Torture and Other
Cruel, Inhuman or Degrading Treatment of Punishment prescribes similarly this duty, by
establishing that: ‘Each State Party shall take effective legislative, administrative, judicial or
other measures to prevent acts of torture in any territory under its jurisdiction.’ [emphasis added].
103
One may also make the argument that it would not be only the EU the party responsible for this,
but also the Member States (or them alone) since, if one upholds the argument that the barriers to
access the court stem from the EU Treaties, then it is the Masters of the Treaties only which can
remedy that violation and therefore hold the responsibility for that barrier.
8.3 An Examination of the Prior Involvement Mechanism and Its Intended Function 317

8.3 An Examination of the Prior Involvement Mechanism


and Its Intended Function

Let us now turn to the prior involvement mechanism established by the DAA, having
observed that the EU system of legal remedies possesses certain deficiencies. Prior
involvement of the Luxembourg Court104 is foreseen as a core mechanism in the
Draft the DAA, although one may argue that it privileges the EU in face of other
Convention Member States’ court systems which have no such right.105 It may
however help build a better coherence of case-law of the two courts.106 Such prior
involvement would seem to serve the functional nature of Art. 19 (1) TEU whereupon
the Luxembourg Court is generally mandated to ‘ensure that in the interpretation and
application of the Treaties the law is observed’.107 This mechanism draws upon from
a common understanding that the presidents of the Luxembourg and Strasbourg
Court—although one may also describe the mechanism as such as an ‘anomaly
under international law’108—when they acknowledged that:
In order that the principle of subsidiarity may be respected also in that situation, a procedure
should be put in place, in connection with the accession of the EU to the Convention, which
is flexible and would ensure that the CJEU may carry out an internal review before the
ECHR carries out external review. The implementation of such a procedure, which does not
require an amendment to the Convention, should take account of the characteristics of the
judicial review which are specific to the two courts.109

104
Referring to the Luxembourg Court’s Discussion Document on this issue, Gragl points that the
Court basically argued that the principle of subsidiarity of the Convention system and the genuine
operation of the EU judicature are linked closely together. See: Gragl (2013), p. 234. It is rightly
maintained, therefore, the Convention’s subsidiarity would not operate if EU would continue to
function under the current limited access to legal remedies of applicants, as demonstrated supra by
this book.
105
See e.g.: Conforti (2010), p. 86; See also: Lock (2012), p. 182; See also: Eckes (2013), p. 268;
Craig (2013), p. 1127; Gaja (2013), p. 4.
106
Raba (2013), p. 560; See also: de Rivery and Chassaing (2013), p. 2, proposing that post-
accession Strasbourg Court will increase the standard of EU Court on human rights protection.
107
Art. 19 (1) TEU; See also a simple context of explanation on this, at: Commission Working
Document DS 1930/10, Council of the European Union, FREMP meeting (Brussels, 10 January
2011) on the topic ‘Previous involvement of the Court of Justice in the context of the accession of
the European Union to the European Convention for the protection of Human Rights and
Fundamental Freedoms’. Brussels, 22 December 2010, para. 12.
108
Odermatt (2014), p. 27.
109
Joint Communication from Presidents Costa and Skouris, European Court of Human Rights
(ECHR) and the Court of Justice of the European Union (CJEU), 17 January 2011, p. 2, available
at: http://curia.europa.eu/jcms/upload/docs/application/pdf/2011-02/cedh_cjue_english.pdf; See
also: O’Meara (2011), p. 1822; Odermatt (2014), p. 16; Besselink (2013), p. 308; Cf.: Balfour
proposes that—in spite of the prior involvement mechanism—in order to minimize conflicting
interpretations between Luxembourg Court and Strasbourg Court, there should be innovated a
procedure—named as ‘pre-decision interpretation questions’—in which case Luxembourg would
refer questions to Strasbourg Court asking it to issue guiding interpretative rulings in order to
harmonize their interpretations. See: Balfour (2005), p. 23. This proposal, however, would make
318 8 EU Prior-Involvement Review

It was clear from the very beginning that the characteristics of the Union’s
special features may only be saved if the Strasbourg Court would be retained
merely as a subsidiary court in relation to the Luxembourg Court’s position after
EU’s accession to ECHR, as also restated in the Interlaken Declaration.110 How-
ever, it is unclear whether the current result of the Draft Accession Treaty reaches
the objective of attaining the characteristics of the two courts and their respective
jurisdictions, as the two presidents of the courts have requested above. Hence, the
need to analyze the nature and scope of the prior involvement mechanism should
stand in the lines of the question whether such mechanism does in reality make
achievable better Convention rights’ protection while fully saving the EU law
autonomy.111 The Court of Justice of EU has further noted that:
It is true that the system established by the Convention does not lay down, as a condition of
admissibility of an application to the European Court of Human Rights, that in every case a
court of supreme jurisdiction must first have been asked to rule on the alleged violation of
fundamental rights by the act in question. However, what is at stake in the situation referred
to is not the involvement of the Court of Justice as the supreme court of the European
Union, but the arrangement of the judicial system of the Union in such a way that, where an
act of the Union is challenged, it is a court of the Union before which proceedings can be
brought in order to carry out an internal review before the external review takes place.112

As illustrious from the above, ‘[i]n the EU related cases [filed before the
Strasbourg Court] the exhaustion of local remedies principle does not entail that

the Strasbourg Court rather a constitutional or fourth-instance court than an international subsid-
iary tribunal of human rights.
110
Interlaken Declaration, High Level Conference on the Future of the European Court of Human
Rights, 19 February 2010, part E, para. 9; See also generally: Groussot et al. (2011), p. 6; Contra:
Gragl argues that it would not be possible in principle for the EU to find itself in the seat of a
co-respondent without its court having had first the opportunity to rule on that case. See: Gragl
(2013), p. 236. I fully disagree with Gragl, and maintain the opposite. In fact, the EU may apply for
the status of co-respondent only when its Member States—after the claimants had exhausted the
legal remedies before the Member State—caused a violation which the claimants filed before the
Strasbourg Court. Practically, this means that the EU Court has had no chance to review the legal
basis of the Member State(s) violation—excluding the situation when the Member State’s court
utilized the preliminary reference procedure before the EU Court, which, again, would not make it
possible for the EU to utilize the prior involvement—if the Member State’s violation was merely
exhausted before that state’s court system. Therefore, in this regard, a prior involvement mecha-
nism is absolutely necessary to preserve the EU law autonomy if one thinks from the perspective of
an EU Court judge.
111
Baratta (2013), p. 1316, on the EU law autonomy relevance of this mechanism; See also:
Quirico (2010), p. 48; Cf: Gragl makes the point that the prior involvement mechanism is in
principle not allowed to introduce and make any amendment to the EU Treaties, as that would then
go against both the EU Treaties’ limitations on this issue but also against the fact that the relevant
DAA provision on this mechanism explicitly says that it should not affect the competences of the
EU Court. See: Gragl (2013), p. 236.
112
Discussion document of the Court of Justice of the European Union on certain aspects of the
accession of the European Union to the European Convention for the Protection of Human Rights
and Fundamental Freedoms, Luxembourg, 5 May 2010, available at:http://curia.europa.eu/jcms/
upload/docs/application/pdf/2010-05/convention_en.pdf, para 11.
8.3 An Examination of the Prior Involvement Mechanism and Its Intended Function 319

the individual application should be dismissed by the [. . .Strasbourg Court if . . .]


the national judge fails to make a request for a preliminary ruling.’113 To go deeper,
as explained earlier, the lack of indirect access to many claimants could end up with
domestic remedies being exhausted without their chance to access the EU Court
through a preliminary ruling procedure, the main dispute of the case relying on an
EU legal act.114 Adding here that the presidents of the two European courts have
themselves argued that the preliminary reference procedure may not ‘normally’ be
even considered a remedy.115 To fill in the loophole wherein such claimants could
end up disputing an EU legal act whereas having had no possibility to challenge
such an act to the EU Court and gaining the right to access the Strasbourg Court, the
Draft the DAA has crafted the prior involvement of the EU Court in such cases.116
Such prior involvement is meant to save the EU Court’s external and internal
autonomy first of all, rather than fulfilling the claimants’ right to access the EU
Court. It was always argued in the discussions between the EU and the Council of
Europe that in order for the EU accession to the ECHR ‘to preserve this character-
istic of the Union’s system of judicial protection [the autonomy of EU law], the
possibility must be avoided of the European Court of Human Rights being called on
to decide on the conformity of an act of the Union with the Convention without the
Court of Justice first having had an opportunity to give a definitive ruling on the
point.’117 This is basically, as Thym puts it, made to ‘prevent[. . .] the unified patent
litigation system’ risk appear.118 ‘Therefore, from the EU perspective the matter is
not if the EU accedes to the ECHR—since the Union “shall accede”—but how (the
quomodo), when acceding, to preserve all the conditions laid down by the treaties

113
Baratta (2013), p. 1326; See also: Lock (2012), p. 181/2; Martı́n and De Nanclares (2013),
p. 11; Odermatt (2014), p. 108.
114
As appeared in Intertanko and Kadi, the EU had to hold itself an international responsibility for
its Member States’ international obligations. For this, see: van Rossem (2009), p. 184 et seq. By
contrast to these two examples, the prior involvement mechanism under Art. 3 (6) DAT aims to
make the EU itself liable before the Strasbourg Court for its own law, and not bypassing this
through the member states’ international obligations; On the need of Luxembourg for first
reviewing the EU act before it reaches Strasbourg Court, see also: Craig (2013), p. 1124; Odermatt
(2014), p. 27.
115
Jacque (2011), p. 1017/8; Adding here that prior involvement will most often be raised in the
context of domestic cases which were refused preliminary reference review by Luxembourg Court.
See on this: O’Meara (2011), p. 1823; In addition, the latter, in p. 1830, argues that post-accession
situation will force EU to request Member States’ courts to better make use of the preliminary
reference procedure, in order to make it easier for them to pass the admissibility procedure before
the Strasbourg Court.
116
See also: Lock (2012), p. 182 et seq; Odermatt (2014), p. 17, arguing that this is a core
mechanism that will convince the Luxembourg Court that DAT is aligned with the Treaties.
117
Discussion document of the Court of Justice of the European Union on certain aspects of the
accession of the European Union to the European Convention for the Protection of Human Rights
and Fundamental Freedoms, op. cit. note 11, para 9.
118
Thym (2013a), p. 3 [emphasis added].
320 8 EU Prior-Involvement Review

Fig. 8.1 A figurative explanation of the route via which prior involvement is initiated

and notably in Protocol No. 8 [of the Treaties].’119 Prior involvement mechanism is
one of those nucleus answers to that how.120
To address this concern of the EU, the negotiators have laid down Art. 3 (6) of
the Draft the DAA, which reads:
In proceedings to which the European Union is a co-respondent, if the Court of Justice of
the European Union has not yet assessed the compatibility with the rights at issue defined in
the Convention or in the protocols to which the European Union has acceded of the
provision of European Union law as under paragraph 2 of this article, sufficient time
shall be afforded for the Court of Justice of the European Union to make such an
assessment, and thereafter for the parties to make observations to the Court. The
European Union shall ensure that such assessment is made quickly so that the proceedings
before the Court are not unduly delayed. The provisions of this paragraph shall not affect
the powers of the Court.

Deconstructing Art. 3 (6) is very important for a thorough examination of this


mechanism (Fig. 8.1). In general, this seems to have been intentionally left as
‘flexible’.121 In this circumstance, there are four issues that need analyzing: first,

119
Baratta (2013), p. 1323.
120
Rieder (2005), p. 103, who asserts that: ‘If we consider the Schmidberger case we can imagine a
situation where the company goes to the ECJ in order to achieve free movement of goods but the
demonstrators go to the ECHR in order to preserve the right to demonstrate. Ultimately, at the end
of the day one side is going to lose. Which court should prevail?’ The preliminary reference
procedure would certainly try to prevent this situation from happening.
121
Raba (2013), p. 566; Gragl, in the same line, argues of this provision and the explanations
relating to it in the Explanatory Report have been written very generally and without details in
order to maintain the EU law autonomy and not to serve as a hidden amendment to the EU Treaties
(referring to Defrenne case). To this extent, Gragl basically argues that with this format it seems
8.3 An Examination of the Prior Involvement Mechanism and Its Intended Function 321

which conditions should be fulfilled to initiate the prior involvement of the


Luxembourg Court; second, how far could the origin of dispute on EU law be in
order for it to become reviewed by Luxembourg Court through prior involvement;
third, the scope of Luxembourg’s prior involvement jurisdiction, structure and
nature of that procedure, and, fourth, the constitutionally significant changes that
this new procedure may bring to the EU-internal system of jurisdictions. Before
getting on the examination of these issues, it is important to bring here the question
of Lock, who asks whether there is a real need to build this mechanism in order to
assure the autonomy of EU legal order.122 Lock goes on by arguing that, in fact, the
decisions of the Strasbourg Court may not invalidate an EU legal act even if they
find it being in contradiction with the Convention.123 This argument does not stand
entirely aligned with the real effects that prior involvement mechanism tries to rule
out. Disagreeing with Lock, it is here argued that although the Strasbourg Court has
no authority to invalidate or explicitly interpret any act of the EU law,124 it
nevertheless has the authority to issue internationally binding rulings on the EU,
whereas the latter must abide by those rulings in the context of its international law
obligations125 and Art. 216 (2) TFEU; adding here that EU is a rather appreciated
normative organization respecting the rule of law of international commitments as
well. Moreover, the standard ruled by the Inter-American Court of Human Rights in
the The Sawhoyamaxa Indigenous Community v. Paraguay case126 demonstrates
rather well that domestic justifications for noncompliance with an international duty
will not discharge an international law person like the EU from its international

that the prescription of this mechanism has been left to the EU law. In this regard, referring to
Lock, Gragl proposes that the internal accommodation of the prior involvement mechanism in EU
law should be addressed with a secondary law instrument (Council decision concluding the DAA)
rather than with primary law revisions. See: Gragl (2013), p. 239–240.
122
Consider in light of Court of Justice of EU, Opinion 1/09 (Creation of a unified patent litigation
system—European and Community Patents Court—Compatibility of the draft agreement with the
Treaties), 8 March 2011, para 66.
123
Lock (2011), pp. 21 & 22; See also: Baratta (2013), 1326/7.
124
See e.g. ECtHR, Slivenko v.Latvia, Application No. 48321/99, judgment of 9 Oct. 2003, para
105.
125
Accord: ‘Every internationally wrongful act of an international organization entails the inter-
national responsibility of that organization.’ Art. 3 Draft Articles on the Responsibility of
International Organizations; On the treaty obligation of EU to be bound to concluded international
agreements, see: Odermatt (2014), p. 12; On how the Court made it clear in Bosphorus that its
decisions are obligatory on all convention parties, see: Phelps (2006), p. 269; Moreover, a very
clear example is Judge v. Canada case of the Human Rights Committee, which ruled that ‘a treaty
should be interpreted in good faith and in accordance with the ordinary meaning to be given to the
terms of the treaty in their context and in the light of its object and purpose.’ (Human Rights
Committee, Judge v. Canada, Communication No. 829/1998, final views of 20 October 2003
(CC PR/C/78/D/829/1998 (2003)), para. 10.4).
126
Inter-American Court of Human Rights, The Sawhoyamaxa Indigenous Community
v. Paraguay (judgment of 29 March 2006), Series C No. 146, para. 137 (the ‘Court has ascertained
that the arguments put forth by the State to justify nonenforcement of the indigenous people’s
property rights have not sufficed to release it from international responsibility.’).
322 8 EU Prior-Involvement Review

responsibility. Such decisions of the Strasbourg Court may not however prejudice
the Luxembourg Court’s review; nonetheless, they may produce the same result-
effect in practice for the EU itself. Therefore a valid note is that ‘rejecting the prior
involvement mechanism would imply that the [. . .Strasbourg Court. . .] would be
called to rule in an EU related case without the EU judge being able to assess the
relevant issue.’127 That said, with no prior involvement mechanism in place, many
EU law provisions that have not been reviewed or interpreted by the Luxembourg
Court may become straightforwardly attacked before Strasbourg Court. This situ-
ation would have made the Strasbourg Court the first-instance court for the
Convention-legality of many EU law provisions. In such a scenario, the EU then
would have been internationally legally bound to implement such decisions of the
Strasbourg Court, in cases where the Luxembourg Court has had no chance to give
its verdict before and even after the Strasbourg’s ruling. Most of those Strasbourg
decisions would directly guide or legally prejudice the application of the contested
EU law provisions, which would have the same legal effect as interpreting or
invalidating directly the contested EU law provisions (many of which may have
also prejudiced the Member States’ application of the principle of primacy of EU
law). Practically, the Luxembourg Court would remain in such cases the only
unasked court for the contested EU law provisions. Save Luxembourg Court here,
the EU should have adhered to a practice wherein such Strasbourg decisions replace
the Luxembourg Court’s jurisdiction in cases where the latter had not been involved
at all. Such replacement would have made Luxembourg Court lose its control on the
indirect but binding interpretation of EU law, adding here that many of the
Strasbourg Court decisions may be of an interpretative nature which guides the
EU institutions how to implement and substantiate their own law; such decisions
may obviously take a form that interpret EU law indirectly but, practically, with
almost the same results. Therefore, for practical reasons, it must be argued that
without a prior involvement mechanism, the Luxembourg Court’s autonomy would
have been risked by many indirect reviews and interpretations which the Strasbourg
Court may have done as a court of first instance for many EU law provisions not
assessed before and after by the Luxembourg Court. In addition, with the prior
involvement mechanism the EU becomes in the equal footing with other state High
Contracting Parties, as Convention Member States may have almost never come to
a position where their cases become assessed at the Strasbourg Court before their
national court having had the possibility to rule on those cases. This mechanism
obviously makes the EU an equal high Contracting Party with the Member States.
Thym brings an interesting argument. He basically argues that the prior involve-
ment mechanism will motivate national courts to request preliminary ruling proce-
dure from Luxembourg less often, as the latter will be anyhow involved in the
adjudication of the case if it reaches the Strasbourg Court.128 Thym is generally
right with this argument, as it builds upon assumption that national courts always

127
Baratta (2013), p. 1326.
128
Thym (2013a), p. 3.
8.3 An Examination of the Prior Involvement Mechanism and Its Intended Function 323

try to find means to surpass the Luxembourg Court if there is room for that. The
author of this book, however, does not agree fully with Thym, having the support of
Breuer who asserts that ‘one has to distinguish between legal and psychological
aspects.’129 National courts, of course, may always seek ways to surpass
Luxembourg’s preliminary reference procedure, however they may rationally not
take the burden of surpassing it intentionally with the idea that if the case goes
before Strasbourg, the Luxembourg may still use the prior involvement mechanism,
and that for two basic reasons: first, the national court may never be assured that the
case before it will be filed by the claimant before the Strasbourg Court; secondly,
the national court should in principle not be interested to be ‘ashamed’ before the
Luxembourg and Strasbourg Court for having intentionally surpassed the prelimi-
nary ruling procedure even though the applicant had raised from the outset that
there is an EU law provision which is supposed to be the source of the human-right
violation. That said, one may not expect that national courts may use this route—at
least not so easily—as it will incur costs for the credibility of that national court.
Instead, it would be exactly the national court—one may put as the
counterargument—which would be now on more careful with the use of the
preliminary reference procedure, as, if it had not considered carefully utilizing
this procedure, it may become later on at the center of attack through the prior
involvement mechanism and the case at Strasbourg Court. The latter argument
would instead hold that national courts will now on rely more often on the
preliminary reference procedure. Therefore, one may also put this counterargument
as the supposed effect which prior involvement mechanism may generate in
relation to the use of the preliminary reference procedure by national courts.
As a final point of discussion, Torres Perez justifiably argues that prior involve-
ment mechanism may conceivably push the dialogue between the Luxembourg and
Strasbourg courts too comprehensively which may probably go to the detriment of
the human rights protection objective.130 The author of the book stands in the same
line with Torres Perez for two additional reasons. First, the precedent of improving
the flawed system of legal remedies in the EU may not—and should not—be
repaired through another advantaging mechanism, which, instead of pushing the
EU to change its system of legal remedies which is at the borders of violation of the
Convention, it does the contrary: offers privileges to a legal order that promotes
imperfect standards of human rights in the field of access to judicial remedies.
Second, there is no legitimate reason to trust that the prior involvement mechanism
will improve the current loophole with regards to the fact that certain cases become
judged—although based on an EU law origin—without having had the chance to go
through the lens of Luxembourg Court. The problem with the effectiveness of the
preliminary reference procedure is one of systemic deficiency that may be regulated
only if managed through a treaty reform—or at least, through a very rigid policy of
the Luxembourg Court on the level of procedural autonomy that member states

129
Breuer (2013), p. 1.
130
Perez (2014), p. 44.
324 8 EU Prior-Involvement Review

courts have with regard to the questions they pose—and their level of correctness—
on EU law issues. Providing for a prior involvement mechanism does not per se
improve this problem, as, still, there are many cases that may end up before the
Strasbourg Court—based on an EU-law origin—where the EU does not fulfill the
criteria for joining as a co-respondent, and, as a result, does not have any chance to
utilize the prior involvement mechanism. Hence, Torres Perez’s argument on not
pushing the dialogue too much is not only justifiable but also a core concern that
questions the level of credibility of external human rights control of Strasbourg
Court upon a system which it dialogues with on agreeable grounds.

8.4 Conditions to Initiate the Prior Involvement: A Rather


Complex Task That May Inhibit Jurisdictional
Allergies Between the Two Courts

Art 3 (6) of the Draft Accession Agreement does put two conditions whereupon the
prior involvement of the Court could be utilized. First, that the EU be a
co-respondent to the case before the Strasbourg Court, and second, that the
Luxembourg Court has not yet assessed the compatibility of the rights at issue
with the provision(s) of the EU law. Regarding the first condition, it is clear that the
EU must be a co-respondent to the case before the Strasbourg Court. This condition
allows that the prior involvement be possible for use only if EU is a co-respondent
to a case before Strasbourg Court (with the case having passed the admissibility test
at the latter; the admissibility test in itself having been a room for Strasbourg Court
to engage with EU law interpretation as a first-instance court131), wherein EU
Member State(s) are the original respondents. In this situation, the EU must have
first asked the Strasbourg Court to allow to become it a co-respondent with the other
respondents, as the EU law is involved somehow through an action which the
claimant challenged an EU Member State measure. This condition, procedurally,
requires that the EU first be added to the procedure in Strasbourg Court as a
co-respondent in order to gain the authority to bring the case for a prior ruling
before its Luxembourg Court. This said, this condition is supposed to be fulfilled
only if the EU is joined and accepted as a co-respondent to the case wherein one or
more Member States are the original respondents, as without that status the EU
would have no right on the prior involvement of its Luxembourg Court. The EU has
been assumingly given the co-respondent status in that proceeding before the
Strasbourg Court because the case being judged involves an EU act, which has
been transposed into practice through an implementing measure of the Member
State, the latter standing as the original respondent for the alleged Convention rights
violation. The EU therefore cannot utilize prior involvement if the EU was the

131
On the latter part of the argument, see also: Jacque (2011), p. 1021. Therefore clearly
interfering to EU law autonomy, if that scenario appears in practice.
8.4 Conditions to Initiate the Prior Involvement: A Rather Complex Task That. . . 325

original respondent. So, the additional question should be, in fact, whether there is
any chance for the EU to obtain prior involvement in a case standing before
Strasbourg Court if it has not first attained the status of co-respondent. Art
3 (6) makes it not possible for the prior involvement to exist if the EU is not
given the co-respondent status before Strasbourg Court. With this being the case,
the Luxembourg Court’s prior involvement is structurally dependent from the
procedure before Strasbourg Court that a case follows: should the latter refuse to
consider EU as having the right to join as a co-respondent, prior involvement
becomes merely a wish for the EU. This first condition, as written in Art.
3 (6) DAA seems to exclude the EU from having the right to use prior involvement
in cases where the claimant has named as original respondents both the EU and
Member States, and wherein the provisions of the EU law contested before Stras-
bourg Court may not be merely those which have gone through the direct review of
the Luxembourg Court but also those which may have remained unreviewed by the
latter as they have gone merely through the national court.
In order for the case to have arrived before Strasbourg Court, the claimant has
first exhausted all possible legal remedies before the national courts or tribunals,
with the latter having not utilized their right to refer the incidental question for a
preliminary reference procedure under Art 267 TFEU to the Luxembourg Court.
Practically, the claimant had not had any chance to be heard before the Luxembourg
Court, even if the issue involved any substance of validity of EU secondary law or
the interpretation of EU primary law.132 On this foundation rests the second
condition of Art. 3 (6) TFEU, which requires that the Luxembourg Court should
have not assessed before the compatibility with the Convention rights of the
provision or act of EU law, the latter including both primary and secondary law.
Therefore, the Luxembourg Court could only undertake a prior involvement review
if there is a contest on whether the ‘provision of EU law’ limits or infringes a right
of the Convention. The question here is who should examine whether there is such a
‘provision of EU law’ which needs be reviewed against the Convention rights? The
most logical answer here is that such a provision of EU should have been somehow
raised within the proceedings before Strasbourg Court. The second more important
question so is whether that ‘provision of EU law’ should have been the reason why
the alleged violation of the Convention right was caused. This brings the issue at the
first condition, which said that the EU should first have been accepted the status of
co-respondent by the Strasbourg Court wherein one of the Member States or more
are the original respondents. To become a co-respondent in this regard, the EU
should have joined as a co-respondent only if the alleged violation notified by the
Strasbourg Court ‘calls into question the compatibility with the rights at issue
defined in the Convention [. . .] of a provision of European Union law [. . .] notably
where that violation could have been avoided only by disregarding an obligation
under European Union law.’133 This said, the ‘provision of EU law’ which Art.

132
Explanatory Report to the DAA, para. 65.
133
Art. 3 (2) DAA [emphasis added].
326 8 EU Prior-Involvement Review

3 (6) Draft Accession Agreement mentions does not unavoidably need to be related
to the case being judged before Strasbourg Court, and may also not be a type of
provision through which the specific alleged violation of the Convention right is
challenged by the claimant at Strasbourg. It may be every kind of provision of EU
law which may arise, needing no causal relationship with the alleged violation
challenged before Strasbourg Court. However, even though the reading of Art.
3 (6) Draft Accession Agreement suggests that prior involvement could be utilized
on any possible provision of EU law—independent from the fact whether it was
examined or proved as the cause of the alleged violation of the Convention right
before the Strasbourg Court—presented before Strasbourg, in fact, the EU cannot
use the prior involvement mechanism unless the Strasbourg Court itself accepts the
EU as co-respondent; a condition in which the Strasbourg Court should check
whether the alleged violation is directly linked to a provision of EU law, without
the implementation of which there could have been no alleged violation. That being
a precondition on the use of the prior involvement, practically, the Luxembourg
Court will come to the situation wherein it can only undertake the prior involvement
on provisions of EU law even mentioned in the case before Strasbourg Court if
there is first the co-respondent status obtained. Most practically, the EU could
obtain the co-respondent status even with a single provision of EU law which is
supposed to have been the direct cause for the alleged violation: however, when the
case comes to the Luxembourg Court for prior involvement review, the latter can
then undertake and review all possible provisions of EU law which might look to it
as infringing a Convention’s right within the context of the case resting at Stras-
bourg. The Luxembourg Court cannot be forced to review only the contested
provisions, but can on its own jurisdiction go ultra petitum and find other potential
EU-law provisions which it considers do in fact produce that alleged violation, as it
in fact needs to invalidate the EU legal basis for the act or omission complained at
Strasbourg.134 A test case in this regard is Kokkelvisserij,135 wherein the applicant
association had attacked before the Strasbourg Court the rejection of ECJ to grant
the possibility to reply to the Advocate General’s report. In filing the application,
the applicant association did attack merely the interpretation which ECJ made on its
Rules of Procedure act. However, the ECJ itself had made clear that if there is an
omission which is alleged by the applicant association, then that stems from the
Statute of the CJEU, that is in fact a substance of the Treaties. Hence, although the
claimant had appealed before the Strasbourg Court merely the alleged violation of
ECJ resulting from a misinterpretation of its Rules of Procedure, in a prior involve-
ment procedure in Kokkelvisserij the EU Court would necessarily need to
interpret also the Treaties (the Statute of CJEU) and the Charter of Fundamental
Rights’ right to access the court as the only way to judge the legal basis of the

134
Explanatory Report to the DAA, para. 66/67; Cf.: Lock (2011), p. 30.
135
ECtHR, Kokkelvisserij U.A. v Netherlands, Application no 13645/05, 20 January 2009, Part A,
the Applicant’s Submission.
8.4 Conditions to Initiate the Prior Involvement: A Rather Complex Task That. . . 327

alleged violation.136 Kokkelvisserij therefore serves as a good example of how the


EU Court may not undertake the review or interpretation of the EU law provision
compliance with the Convention by tackling merely those EU law provisions
‘attacked’ by the claimant in Strasbourg. This being the case, the second condition
of Art. 3 (6) of the Draft Accession Agreement gives very broad jurisdiction to the
Luxembourg Court to review the provisions of EU law supposed to have interfered
to a Convention’s right being reviewed before Strasbourg Court; this type of
jurisdiction may go far beyond the contested provisions of EU law on basis137 of
which the EU can gain the co-respondent status before the Strasbourg Court. The
Luxembourg Court does already use the practice, as in Gerritse,138 to reformulate
questions which have been posed by the national court through the preliminary
reference procedure if the legal basis which they posed is not that which it should
have been, therefore making its own analysis of where the illegal provision stands
and which legal basis should in fact be tackled to repair the alleged violation.
However, the Luxembourg Court may practice to use this jurisdiction on the same
margins as the Strasbourg Court when deciding on the provision of the co-respond
status to the EU: such a practice of the Luxembourg Court, however, may prove
problematic in relation to the thorough and systematic review of the alleged pro-
visions that it needs to undertake to examine whether there are other norms which
need be interpreted and reviewed in order to stop the alleged violation. A valid
argument here is that Luxembourg may not mirror Strasbourg, as the former has a
regular form of court whereas the latter is merely a subsidiary court.139 Another
doubt here is who decides whether there is a ‘provision of EU law’ considered to
have supposedly caused the alleged violation of the Convention right at issue. A
simple answer is that once the EU gains the co-respondent status in a case before
Strasbourg Court, it also automatically gains the full discretion to review the
provisions of EU law which are raised as involving the alleged violation, but also
those which the EU Court itself might consider as playing a role or in fact
implicating the Convention rights in that specific case. This mere fact suffices for
the EU to utilize the prior involvement mechanism as defined in Art. 3 (6) of the
DAA. However, the EU may also choose not to use the prior involvement

136
See supra, the chapter on co-respondent mechanism; Accord: ECtHR, Kokkelvisserij U.A. v
Netherlands, Application no 13645/05, 20 January 2009, Part A, the Applicant’s Submission, note
135, part on the ECtHR examination of the origin of the alleged violation.
137
Cf.: Commission Working Document, DS 1930/10, Council of the European Union, FREMP
meeting (Brussels, 10 January 2011) on the topic ‘Previous involvement of the Court of Justice in
the context of the accession of the European Union to the European Convention for the protection
of Human Rights and Fundamental Freedoms’. Brussels, 22 December 2010, note 107, para 9; The
argument presented there seems very unfounded and therefore may not be accepted as valid.
138
Court of Justice of EU, Gerritse, C-234/01 [2003] ECR I-5933, paras 23–24.
139
Cf.: Lock (2011), p. 30; See on the nature of the Strasbourg Court’s subsidiarity the Interlaken
Declaration, High Level Conference on the Future of the European Court of Human Rights,
19 February 2010, note 110, part E, para. 9.
328 8 EU Prior-Involvement Review

mechanism,140 adding that this mechanism is merely a right of the former and not a
duty under the DAA.
To this extent, Art. 3 (6) DAA requires that the ‘provision of EU law’ should
have never been reviewed against the alleged violation in order for the second
condition to be passed by the EU. It is to be examined the exact meaning of this
clause. Practically, this means that the Luxembourg Court should have never had
the chance to review the legality or interpret the ‘provision(s) of EU law’ which are
supposed to have caused somehow the alleged violation. Most simply, this means
that both a ‘never’ at the level of direct EU legal remedies and indirect remedy,
namely the preliminary reference procedure under Art. 267 TFEU. One needs
however to dig a bit deeper on this issue, as there is a huge difference between
the situation where the Luxembourg Court did some type of review on that
contested provision of EU law (either directly or indirectly) to the situation wherein
the Luxembourg Court did review the compliance of that contested provision with
the Convention’s rights. Art. 3 (6) DAA seems to suggest that the latter is the case,
as Art. 3 (6) DAA clearly sets that the Luxembourg Court should have ‘not yet
assessed the compatibility with the rights at issue defined in the Convention’ or its
protocols of the contested provision. Therefore, the big question here is whether
cases reviewed through direct and indirect actions before the Luxembourg Court
are of the nature excluded by Art. 3 (6) DAA conditions for prior involvement. One
can argue here that if the contested provision is reviewed through a direct action
before the Luxembourg Court, then the latter—while interpreting it or checking it
validity—should have also checked the validity in terms of the general principles of
law of the EU (wherein the Convention does form part at minimum) or the Charter
of Fundamental Rights, which does at minimum guarantee the same rights as the
Convention. This is rather clear, and suggests that cases/provisions which have
gone through the review of Luxembourg Court initiated by direct actions have all
been checked their legality against the EU law human rights standards, wherein
Convention rights form part of them. However, the problem is whether the EU law
provisions which have been checked through the preliminary reference procedure
(Art. 267 TFEU) could be considered as also assessed against the EU human rights
law. Here, the contestation is rather huge, for the mere reason that not every
question referred to Luxembourg Court by a national court is precise and entails
the examination of its compliance with the EU human rights standards. Such
questions referred may purely be of a type wherein the national court needs mere
guidance and not crosscheck with the EU human rights law. Besides, even though
they are referred as questions, most of the times such question do not have a
concrete nature which would make them look as putting into challenge a certain
provision against the EU human rights bulk; the EU Court may have rightly ignored
going beyond such a national court question if it did not put any type of contestation
on the human rights law side of it. On the other hand, the Explanatory Report of
DAA141 leaves the explicit impression that the prior involvement is only possible if

140
Lock (2011), p. 24.
141
Explanatory Report to the DAA, para. 65 et seq.
8.4 Conditions to Initiate the Prior Involvement: A Rather Complex Task That. . . 329

the preliminary reference procedure has not been used against a certain contested
EU law provision before the Strasbourg Court. Certainly, the Explanatory Report
seems to contradict our argument on the real picture that the examination of the
second condition of Art. 3 (6) DAA provides, however it is suggested that the
Luxembourg Court—for purposes of legal certainty—should better adhere to our
proposed interpretation rather than to the supposed impression that the Explanatory
Report of DAA provides. Finally, it is important to argue that a provision of EU law
has been assessed before by the Luxembourg Court may be considered only in light
of substantive/in meritum rulings of the Court, certainly excluding here decisions
on inadmissibility as grounds to call certain provisions assessed before. Another
point would be whether prior involvement may be recognized to the Luxembourg
Court on CFSP acts/measures, which naturally would have had no chance of being
reviewed by Luxembourg neither by direct nor indirect actions. A logical answer
would be that Luxembourg Court will have no competence to engage in prior
involvement on a CFSP measure, although it would be the obligation of Strasbourg
Court to permit referring it to Luxembourg for prior involvement. Then
Luxembourg would merely refuse to review it. If Strasbourg Court would automat-
ically reject to proceed such issue for prior involvement before Luxembourg with
the argument that the latter has no such treaty competence, that would be seen as
interpretation of EU Treaties by Strasbourg Court and effectively interference to the
autonomy of Luxembourg.
A further interesting issue is the scope of acts that can come under the jurisdic-
tional deconstruction the ‘provision of EU law’ which the Luxembourg Court can
review through the prior involvement mechanism of Art 3 (6) DAA. Logically, the
term ‘provision of EU law’ encompasses both primary and secondary law. How-
ever, a distinction must be drawn here in order to answer the question on the nature
of the decisions that the Luxembourg Court might bring under Art. 3 (6) DAA. It
should be noted that only the jurisdiction to review the legality/validity of a
provision of EU law might result in any effects that could change the direction of
proceeding before the Strasbourg Court. Otherwise, interpreting an EU law provi-
sion may not practically change and/or dismiss the victim status of the claimant
before the Strasbourg Court, as the legal effect of the measure which the Member
State has forced upon the claimant cannot result in the claimant losing the victim
status before the Strasbourg Court.
Another point to be examined is the question of who decides whether the EU law
provision concerned has been assessed before the Luxembourg Court or not, as a
precondition for the EU to fulfill the second condition. It is arguable that it is merely
the Luxembourg Court which will have the full discretion to decide whether a
certain challenged provision has been once assessed for its legality and/or
interpreted before, adding that the Strasbourg Court would have no authority to
examine such an issue as it would amount to an interference to the Luxembourg
Court autonomy. The Strasbourg Court could easily allow itself to assess whether a
certain ‘EU law provision’ has been reviewed against a Convention right before the
EU Court, by arguing that it is legally the Strasbourg Court which assesses the
admissibility criteria which apply on the prior involvement mechanism. Art.
330 8 EU Prior-Involvement Review

3 (6) DAA does certainly provide room to the Strasbourg Court to judge whether a
domestic legal remedy in the EU system has been used against a certain EU law
provision, as a means to assess the condition on whether the EU Court has anytime
judged that EU law provision. If Strasbourg Court would decide to use this space, it
would certainly amount to an issue which directly interprets the EU law. However,
if practically the Strasbourg Court refrains from using this space, then, in examining
whether an EU law provision has been assessed before or not by it, the Luxembourg
Court will remain the arbiter on deciding whether a potential question of prelimi-
nary reference procedure has been answered in a proper way as to fulfill the
criterion of a reviewed provision; this would certainly fortify the Luxembourg’s
Court autonomy on the issue.
Finally, it should be argued that these two conditions of Art. 3 (6) DAA seem to
have a cumulative and aggregate nature. The EU cannot pretend to assume the
authority on prior involvement without fulfilling both conditions. However, logi-
cally, if the EU was to fulfill the condition on becoming a co-respondent, that would
mean that there is also an EU law provision which should be checked against the
Convention’s rights. Obviously, it remains upon the Strasbourg Court to consider
these two conditions fulfilled, with the Luxembourg Court having no authority as to
the interpretation of Art. 3 (6) DAA for purposes of deciding when can a case be
sent for prior involvement of the Luxembourg Court. It should be noted, nonethe-
less, that both conditions may be interpreted in different ways by the two courts,
something which could make the prior involvement of the EU less likely to serve
the autonomy of EU law.

8.5 Prior Involvement of the Luxembourg Court: A


Procedural Assessment

If the two conditions are fulfilled, then the EU assumes the authority to use the prior
involvement of its Court in assessing the compatibility of the EU law provisions
with the Convention rights.142 A core question here is how is this procedure then
organized internally in the EU. Art. 3 (6) DAA does not name the actors which can
trigger this procedure, leaving it to the EU system of law to determine such
specifications.143 The intention of DAA was rather to ensure that the EU is kept
autonomous in deciding how to organize this procedure internally, as a step that
supports the autonomy of EU legal order. If DAA would in fact lay down the
internal procedure, that could have amounted to a real EU treaties’ amendment.144

142
Morano-Foadi and Andreadakis (2011), p. 609, as an example showing how the Luxembourg
Court does in fact consider human rights even beyond the jurisprudence of the Strasbourg Court. In
such a scenario, the prior involvement can really go beyond the Strasbourg standard, even before
the case reaching the Strasbourg Court practically.
143
Lock (2011), p. 25.
144
Lock (2011), p. 25.
8.5 Prior Involvement of the Luxembourg Court: A Procedural Assessment 331

The Council had preliminarily authorized the negotiations of the DAA under the
premise of ensuring ‘that the prior internal control by the Court of Justice of the
European Union, in accordance with primary law, is applicable also in cases where
the conformity with the [Convention] of an act of an institution, body, office or
agency of the Union is at stake in a case brought before the’ Strasbourg Court.145 It
is first of all problematic to think of such an internal procedure which is in line with
the primary law, as the Treaties allow almost no explicit room for the prior
involvement mechanism. By annexing a specific declaration to the DAA, ‘the
Council will unanimously adopt legally binding rules to the extent permitted by
the Treaties, on the prior involvement of the Court of Justice of the European
Union’ in proceedings before the Strasbourg Court as defined in Art. 3 (6) DAA.146
Such rules are meant to stand in line with the Treaties’ open spaces, which seem to
be rather minimal for allowing such a big mechanism to be established without
legal implications.
Such internal rules which will be adopted unanimously by the Council will aim
to seem similar to the preliminary reference procedure, at least so has been
suggested by the Commission’s Working Document.147 It would be rather awkward
to have this procedure organized internally similar to the preliminary reference
procedure, and that for two big reasons: first, Art. 267 TFEU does explicitly make
room only for preliminary reference questions posed by national courts or tribunals,
which are dealing with a specific case involving the need to interpret or review the
legality of an EU act. One would find almost no space within Art. 267 TFEU to

145
Commission Working Document, DS 1930/10, Council of the European Union, FREMP
meeting (Brussels, 10 January 2011) on the topic ‘Previous involvement of the Court of Justice
in the context of the accession of the European Union to the European Convention for the
protection of Human Rights and Fundamental Freedoms’. Brussels, 22 December 2010, note
107, p. 3; Refer also to the para 11 of the Negotiation Directives of the Council of 4 June 2010.
146
Commission Working Document, DS 1930/10, Council of the European Union, FREMP
meeting (Brussels, 10 January 2011) on the topic ‘Previous involvement of the Court of Justice
in the context of the accession of the European Union to the European Convention for the
protection of Human Rights and Fundamental Freedoms’. Brussels, 22 December 2010, note
107, p. 3; Referring to Council Declaration annexed to the Council Decision authorising the
negotiation of DAT (annex 3); See also: Gragl (2013), p. 240, who proposes that such rules could
be established with the Council’s decision concluding the DAA.
147
Commission Working Document, DS 1930/10, Council of the European Union, FREMP
meeting (Brussels, 10 January 2011) on the topic ‘Previous involvement of the Court of Justice
in the context of the accession of the European Union to the European Convention for the
protection of Human Rights and Fundamental Freedoms’. Brussels, 22 December 2010, note
107, para 5; Cf.: Another option—instead of the prior involvement mechanism, and somewhat
similar to the stages of the abovementioned proposal by the Commission’s working document—
proposed by CJEU Judge Timmermans, according to Gragl, designates the Commission as the
institution which may trigger the Luxembourg Court’s review in case it considers that the case
before the Strasbourg Court implicates an EU legal basis. Upon the Commission triggering this
procedure, the proceedings before the Strasbourg Court would be paused. If, in that scenario, the
Luxembourg Court would find that the case at hand violates the human rights of the claimant, then
the case before the Strasbourg Court should have been automatically dismissed. See: Gragl (2013),
p. 241/2.
332 8 EU Prior-Involvement Review

allow another institution e.g. the Commission, rather than a national court, to pose
questions to the Luxembourg Court; adding here that the logic of Art. 267 TFEU is
to allow the Luxembourg Court help resolve—substantiating its internal autonomy
towards the Member States—a case of a judicial nature pending before the national
court. If someone would argue that the Commission can be accommodated as an
institution that poses preliminary reference questions under Art. 267 TFEU, the
counter argument would be that the Commission does not review judicial cases and
therefore cannot use the incidental jurisdiction of the Luxembourg Court for
purposes of having a judicial pending case judged. The teleological interpretation
of Art. 267 TFEU seems to leave almost no room for the Commission. Second, the
preliminary reference procedure under Art. 267 TFEU is made to serve the concept
of EU law primacy and therefore to place the EU Court at the top of the vertical
integration of national courts in the EU law system (Patent case of ECJ). Therefore,
the prior involvement of the EU to a case pending before the Strasbourg Court does
not seem to accommodate any concept of vertical integration of the latter court to
the EU system, as the prior involvement procedure does merely serve the EU law
external autonomy rather than the internal autonomy. These being the two argu-
ments, legally speaking, it is rather problematic to consider prior involvement in a
similar pattern and logic with the preliminary reference procedure from the per-
spective of EU internal rules and organization.
One facet of prior involvement mechanism is similar to that of the preliminary
reference procedure, however: the fact that it must be initiated by someone in order
to trigger the EU Court’s jurisdiction. The Commission’s Working Document
suggests that such initiating actors of the prior involvement procedure could be
either the Commission or the Member States.148 This said, it is logically argued that
the prior involvement mechanism is not automatic, and cannot be triggered by the
EU Court on its own right. There needs to be another actor which should trigger the
procedure. The most plausible option seems for the Commission to trigger that
procedure, but not for the Member State(s) which are simultaneously respondent
(s) before the Strasbourg Court’s case. It is argued here that Member States have
had the option to refer any question of EU law through the preliminary reference
procedure in which the case has gone before the national court. Since such proce-
dure was not used in this scenario by the national court, that Member State should
have lost that right to refer the question after the case was res judicata before the
national courts. Further, being res judicata, the Member State should not be
allowed to have the chance to surpass their own courts’ discretion on whether to
use or not the preliminary reference procedure under Art. 267 TFEU. If, however,
Member States were allowed to initiate the prior involvement of the EU Court, that
would be seen as producing legal uncertainty in face the of claimants which have

148
Commission Working Document, DS 1930/10, Council of the European Union, FREMP
meeting (Brussels, 10 January 2011) on the topic ‘Previous involvement of the Court of Justice
in the context of the accession of the European Union to the European Convention for the
protection of Human Rights and Fundamental Freedoms’. Brussels, 22 December 2010, note
107, para 6.
8.5 Prior Involvement of the Luxembourg Court: A Procedural Assessment 333

had the cases resolved before their national courts as res judicata issue. Moreover,
given that practically the national courts have no chance to become those who
would trigger the use of the prior involvement—but rather a political body like the
Member State’s Government would do that—then it logically follows that such a
scenario would put the Member State’s government over their national court’s
independence and full discretion on having decided whether to refer or not a
question through Art. 267 TFEU.
The most likely scenario would be for the Commission to become the actor
which can trigger the prior involvement of the EU Court within the context of EU
internal rules. It is also argued that such a role does not excellently, but more
properly, fit to the Commission’s function rather than to the Member States.
Nevertheless, it is generally argued that this function of the Commission could
somehow be accommodated under Art. 263 (2) TFEU—although is clearly not in
this format—adding that the Commission would most thoroughly know the case
because it is Commission the institution that represents the EU before the Stras-
bourg Court.149 The Commission will therefore become the ‘filter’ of prior involve-
ment mechanism within the EU system.150 There is a big stone on the road in this
case, even if one tries to locate this jurisdiction of the EU Court and the Commis-
sion’s right to trigger proceedings on Art. 263 TFEU. The logical reason here is
that, although Art. 263 TFEU might stand as a location for the prior involvement
jurisdiction, if the Commission triggers this procedure to require the prior involve-
ment of the EU Court for a provision of EU law of a case pending before Strasbourg
Court, the Commission must attack that provision as being illegal. Art. 263(2) can
be utilized by the Commission on the ‘grounds of lack of competence, infringement
of an essential procedural requirement, infringement of the Treaties or of any rule
of law relating to their application, or misuse of powers.’151 That said, when the
Commission triggers the prior involvement of the EU Court, it may do so only by
arguing that the certain EU law provision is considered as touching one of those
grounds of illegality, otherwise it would not be possible for the Commission to
utilize Art. 263 (2) TFEU only by saying that it wishes to see the EU Court
reviewing those contested provisions before Strasbourg. Although this is a very
big stone in the attemption to locate the prior involvement of the Court and the
Commission’s right thereon within Art. 263 TFEU, it still remains the most possible
space for the prior involvement of the EU Court within the Treaties.152 This being

149
Lock (2011), p. 26.
150
Lock (2011), p. 27.
151
Art. 263 (2) TFEU.
152
Cf.: One may rightly question why should the competence of the Commission to trigger the
prior involvement be rooted in a treaty provision (or why should the treaties change to accommo-
date this mechanism). Four arguments denounce the latter and support the main-text reasons. First,
to establish the Commission’s competence to trigger the prior involvement procedure, one has to
amend Protocol 3 to the Treaties (Statute of the CJEU) to define the Commission’s authority under
this procedure when it appears before the Court. The Protocol itself—although it may be changed
through an ordinary legislative procedure for most of its parts—comprises a segment of the
334 8 EU Prior-Involvement Review

the case, as long as it is presumed that the DAA does not revise the Treaties, we
must take this assumption as accepting that the prior involvement be done by the
Commission through Art. 263 TFEU. All these important gaps, by the way, were
left uncovered by the Luxembourg Court’s Opinion 2/13.
The argument behind this assumption would be that the Commission can play
that role in its capacity to serve the EU law’s aims of uniformity and being the
guardian of the Treaties.153 However, with the Commission being the actor that has
the discretion of triggering or not the prior involvement of the EU Court,154 one

Treaties (primary law). Second, most importantly, one has to consider Art. 1 (3) DAA, which
reads: ‘Nothing in the Convention or the protocols thereto shall require the European Union to
perform an act or adopt a measure for which it has no competence under European Union law.’ In
addition to this, Protocol 8 to the EU Treaties, amongst others, establishes that ‘accession
agreement should ensure that accession does not affect the competences of the EU or the powers
of its institutions, or the situation of the Member States in face of the Convention.’ This means not
only accession agreement but also ‘accession’ at-large. The same has been reiterated in Opinion
2/13 where the Court noted several times that accession as such should not disturb the institutional
balance of the EU or its Member States’ position in face of the Convention system. This said, one
should argue that the prior involvement jurisdiction which the Luxembourg Court will be given
with the DAA is rather different from the current jurisdiction which is possesses under the
Treaties. Such new jurisdiction may not be described in another way but as a new jurisdiction to
the Luxembourg Court. Moreover, the prior involvement jurisdiction will rather well reshape the
position of the Commission in face of the Luxembourg Court with regard to the potential
violations upon the Convention. This altogether means that the DAA is extending new compe-
tences and touching upon the already-existing treaty balances that exist in the EU institutional
structure (including the CJEU). Such fact quire well affects the authorities of the EU institutions,
something which the DAA and Protocol 8 explicitly prohibit. Third, not only the prior involvement
procedure seems problematic in face of the treaty structures but also with regard to the Protocol
8 prohibition that the accession should not affect the situation of the Member States in face of the
Convention. One basic fact, that needs consideration, is a scenario in which the Member State as
respondent will be submitted for an alleged violation before the Strasbourg Court—a violation
which the Member State committed by implementing an EU-law obligation that left no state
discretion. Should in that scenario the EU Court invalidate (through the prior involvement
mechanism) the legal basis that pushed the Member State to conduct that violation, that would
still not change the victim status of the applicant and the status of respondent of the Member State
in question. This scenario will basically produce a situation in which the respondent Member State
may become liable before the Strasbourg Court for implementing an EU-law obligation which was
invalidated by the EU Court, and does not even exist anymore. Such form of liability for the
Member State for which it holds no original responsibility factually changes the position of the
Member States in face of the Convention. These four arguments prove that there is a need to
introduce primary law amendments to address these concerns.
153
On the Commission as the guardian of the treaties, see e.g.: Von Bogdandy et al. (2012), p. 490.
154
Especially if this role is considered under case: Court of Justice of EU, Star Fruit v Commis-
sion, Case 247/87 [1989] ECR 291; See also: Lock (2011), p. 27; Cf.: Gragl (2013), p. 242/3, who
basically argues that it would be in the Commission’s discretion to trigger or not the prior
involvement of the EU Court: such decision of the Commission being based, amongst others, on
the criterion of whether the same EU legal basis has been reviewed once by the Luxembourg Court
for its legality. That would certainly, as argued supra, push the Commission to play a sort of filter
function as well; Gragl in this regard argue that should the Commission refuse to trigger the prior
involvement procedure, that would not be seen as interference to EU law autonomy, as the
8.5 Prior Involvement of the Luxembourg Court: A Procedural Assessment 335

may argue that this procedural environment might produce a conflict of interest
between the Commission’s function and the EU Court’s prior involvement function.
In its original constitutional mandate, the EU Court exists to observe and review the
application of the EU law, such application being mostly a competence of the
Commission and Member States. In that constitutional context, the Luxembourg
Court is there to ensure a democratic rule of law control upon the Commission and
Member States (but not only) when they apply and make the EU law. In that natural
role, the EU Court is made to counterbalance institutions which produce and
implement the EU law, as a means to provide with judicial protection every
possible party concerned. It must be noted here that much of the EU law that
would be attacked in Strasbourg may originate either from the initiative of the
Commission in the legislative procedure or from the Council’s delegated powers to
the Commission under Art. 290 TFEU and Art. 291 TFEU wherein the Commission
passes alone certain EU legal acts. Found in this situation, the Commission would
remain the body which will have the right to trigger the prior involvement proce-
dure, the latter being a mechanism that will likely review and potentially invalidate
the Commission’s acts or acts which were initiated by the Commission itself. This
normally leaves room for calling into question whether the Commission ‘act
[s] within the powers conferred on [it] in the Treaties’, as established by Art.
13 (2) TEU.155 In this scenario, the Commission would be the least actor interested
in having EU law invalidated before the EU Court. Therefore, the argument that the
Commission has a subjective conflict of interest in being given the authority to
trigger this procedure stands objectively valid, moreover the Commission may not
be considered as being legitimate in triggering the fair reviewability of the law in
the adoption of which it has itself participated. This conflict of interest might
therefore compromise and prejudice the Commission’s credibility at being the
core initiator of the prior involvement mechanism under Art. 3 (6) DAA.156
Regarding the nature of the decisions that could be taken by the prior involve-
ment, the Commission’s Working Document foresees that: ‘The Court of Justice
would decide on the merits of the request by a judgment or possibly, in certain

argument would be that the EU has no interest in that case to be considered first by the
Luxembourg Court. In that scenario, the Strasbourg Court would decide without the Luxembourg’s
decision. To this point, Gragl rightly argues that the Commission should not be obliged to trigger
this mechanism, and that decision would be in line with the subsidiary character of the Convention
system (Ibid, p. 243), although the Luxembourg’s Opinion 2/13 follows a different line of
argument.
155
See also: Court of Justice of EU, Commission v. Ireland, C-459/03, [2006] ECR-I 4635, para
123.
156
Contra: Gragl sees no implication from accommodating the prior involvement mechanism in
this legal basis, and therefore proposes that it would be in line with the current treaty prescriptions
on this issue. See: Gragl (2013), p. 240. Although Gragl does not examine this in-depth, it seems
even from a very superficial position hard to capture such argument as being valid.
336 8 EU Prior-Involvement Review

circumstances, by an order (but not merely by an opinion).’157 This said, it is


suggested that prior involvement of the Luxembourg Court will produce legally
binding results of the same effect that may be produced e.g. under a direct or
indirect action. Those legally binding effects may interpret, confirm or nullify any
EU law provision contested before the Strasbourg Court.158 In undertaking that
action, the Luxembourg Court will be bound by the rule set out in the Draft
Accession Agreement, which reads: ‘The CJEU will not assess the act or omission
complained of by the applicant, but the EU legal basis for it.’159 A potential
invalidation of the EU legal basis for the measure or omission of the Member
State based on that will automatically produce the effect of having that measure or
omission considered without legal basis: but not effectually invalidating also the
measure or omission of the Member State. Therefore, although the decision of the
Luxembourg Court will judge in meritum the legal basis within the EU law for the
act or omission of the Member State, it is not said that such a decision may
invalidate Member State act or omission which has caused the alleged violation.
An important and legitimate question here would be whether such an ‘EU legal
basis’ is mentioned by the claimant in Strasbourg Court, or whether the Strasbourg
Court would tell which that legal basis is? To ensure fortification of the EU law

157
Commission’s Working Document DS 1930/10, Council of the European Union, FREMP
meeting (Brussels, 10 January 2011) on the topic ‘Previous involvement of the Court of Justice
in the context of the accession of the European Union to the European Convention for the
protection of Human Rights and Fundamental Freedoms’. Brussels, 22 December 2010, note
107, para. 8.
158
A legitimate question raised by Gragl in this regard is the grounds of review which the
Luxembourg Court may consider in relation to the prior involvement mechanism. Gragl basically
argues that the Luxembourg Court would not need to be limited merely to the fundamental rights’
ground of review, or to the grounds proposed by the Commission with the decision to trigger the
prior involvement mechanism, but may go beyond the grounds raised by the Commission and
decide on its own right on the scope of review that it wishes to undertake on the EU act concerned.
Gragl bases this argument also in the practice of preliminary reference questions—e.g. Finanzamt
Kassel v. Viessmann—wherein the Luxembourg Court has ruled that it may reformulate or even go
beyond the questions posed by the national courts, by recognizing to itself the right to delimit the
scope of review. See: Gragl (2013), p. 249–251.
159
Explanatory Report to the DAA, para. 67; Cf.: Gragl (2013), p. 246/7, who makes an interesting
question and argument, by asking whether providing to the Commission a ‘comprehensive
competence to request the CJEU to review legislation on regular basis (regardless of the concrete
circumstances with respect to the applicant’s standing) would modify the functional nature of
CJEU proceedings’ and therefore interfere to the EU law autonomy? In connection to this
argument, I would not agree with its general line of reasoning, as the Commission currently
possesses the same competence with regard to the action for annulment. However, more prob-
lematic than this, I would say—something Gragl does not mention at all—is the question of
whether bringing the Commission to the position of an institution which—based in the circum-
stances of the case before the Strasbourg Court—becomes a player in the process of validation of
interpretation of rights and law on which a judicial case is based and depended from. This aspect
would be far more problematic, even within the context of the right to fair trial and independent
judicial proceedings, and certainly leads to the conclusion that with this new role, the primary law
function of CJEU and its proceedings may see substantive change.
8.6 Refraining from Undue Delay: How Can This Be Met? 337

autonomy, the Strasbourg Court may not be allowed to name which legal basis in
the EU law is the legal source for the act or omission complained thereon, as that
would amount to an infringement of the Luxembourg Court’s exclusive jurisdic-
tion. Most logically, the ‘EU legal basis’ concerned should have been somehow
mentioned by the claimants, and then exclusively judged by the Luxembourg Court
on whether it is in fact the EU legal basis for the alleged violation, or, moreover, if
there is another EU legal basis for that alleged violation. In identifying this legal
basis, the Luxembourg Court should be given freedom to go beyond the claimants’
allegations, as it has exclusive jurisdiction to interpret and observe the law of the
EU from its own jurisdictional autonomy.

8.6 Refraining from Undue Delay: How Can This Be Met?

The Draft Accession Agreement does require that the prior involvement procedure
before the Luxembourg Court ‘is made quickly so that the proceedings before the
[Strasbourg] Court are not unduly delayed.’160 Certainly, the Draft Accession
Agreement provides that Luxembourg Court should be given the chance to review
the EU legal act concerned, ‘it nonetheless remains ambiguous as to when the
“pause button” is pushed in Strasbourg.’161 Paul Craig puts a very grounded doubt
on the consequence of this delay by arguing that: ‘It nonetheless remains paradox-
ical that the EU courts should be able to use Article 3(6) of the Draft Agreement in
order to “pause” the case at Strasbourg in order for the CJEU to adjudicate on the
substance of Convention rights, where it was the CJEU’s very own restrictive
standing criteria that prevented the EU courts from doing so before the case was
taken to Strasbourg.’162 As it is highly important to acknowledge the ‘maxim
“justice delayed is justice denied”’163 before Strasbourg Court, one needs to
carefully examine this procedural guarantee. There is no formula of interpreting
the ‘undue delay’ from Art. 3 (6) DAA.164 However, prior involvement procedure

160
Art. 3 (6) DAA; See also: O’Meara (2011), p. 1825; Undue delay, in this regard, should also be
considered within the context of the extra-length of proceedings if one sees the period from the
moment the claimant files the application to the Strasbourg Court until it reaches the final decision
of the latter. This may seriously harm the right to efficient access to the Court. For the latter part of
the argument, see e.g.: Rieder (2005), p. 105; On the latter, see also: Wetzel (2003), p. 2849; See
also: Gragl (2013), p. 243.
161
Craig (2013), p. 1126; Cf.: White (2010), p. 444, who questions whether the time-delay issue
may push accession to the notable fact of ‘justice delayed is justice denied’.
162
Craig (2013), p. 1131.
163
O’Meara (2011), p. 1826.
164
See generally also: Groussot et al. (2011), p. 15; Cf.: Gragl (2013), p. 247, who—referring to
Tulkens—argues that the prior involvement mechanism will ‘unquestionably’ exceed the normal
timeline which an application follows to become processed before the Strasbourg Court, therefore
complicating the position of the applicant in an ‘unacceptable manner’.
338 8 EU Prior-Involvement Review

should be used in a way which does not render the Strasbourg Court’s function as
extremely not efficient because of the Luxembourg Court’s incidental review.165 If
that would be the case, certainly, the Strasbourg human rights system would be
harshly attacked with the accession of the EU to ECHR.
A core question would be whether the ‘undue delay’ refers mainly to the time
before which the prior involvement mechanism is triggered by the EU, or also to the
Luxembourg Court’s proceedings as well. It is argued here that undue delay refers
to both: namely, the EU should trigger prior involvement very fast and its Court
may not take too long to adjudicate it. Art. 3 (6) DAA notion on undue delay is
made to ensure that the final response from the EU to the Strasbourg Court is made
as fast as possible,166 in order for the latter to resume proceedings in the case before
it. It should also be noted that a pressurized behavior of the Strasbourg Court on the
Luxembourg Court to interpret the ‘undue delay’ according to its prescription might
get inside the borders of the Luxembourg’s autonomy,167 with Gragl basically
arguing that with the mere fact that the DAA provides for the obligation of
‘undue delay’ it basically regulates an internal issue of EU law (therefore interfer-
ing to its autonomy).168 If the EU would practice to use Art. 263 (2) TFEU as the
ground whereupon the Commission would base its decision to trigger the prior
involvement mechanism, then the time limit from Art. 263 (6)TFEU should be
followed, which reads: ‘The proceedings provided for in this Article shall be
instituted within two months of the publication of the measure, or of its notification
to the plaintiff, or, in the absence thereof, of the day on which it came to the
knowledge of the latter, as the case may be.’ The Court has ruled that the core
element here is the publication of the act, and that the ‘criterion of the day on which

165
One must say that the overall delay within the Luxembourg Court’s review may per se
constitute a violation of Art. 6 and 13 ECHR, not only in this type of procedure but also in all
others. See e.g.: Balfour (2007), p. 222; On the latter, see also: AIRE Centre et al. (2013), p. 4; On
this delay, see also: Wetzel (2003), p. 2851.
166
Gragl argues that with the accelerated procedure which the Luxembourg Court may use—the
so-called PPU procedure—to address a prior involvement request, it will take 6–8 months to
deliver a ruling on the prior-involvement issue. See: Gragl (2013), p. 247. This means that an
application before Strasbourg Court who was made part of the prior involvement procedure will
consume at least 6 more months to become adjudicated.
167
Cf.: Lock (2011), p. 29; Cf. also with: Court of Justice of EU, Rules of Procedure Art 104 a and
104 b; See also on a similar argument: Gragl (2013), p. 244, who argues that the time-limit within
the EU legal order is something that will be observed strictly by the EU Court, and a case may
become certainly inadmissible before it (TWD case of CJEU being an example for cases involving
individual decisions subject to an action before the latter), which would then open spaces for
interference to the EU law autonomy.
168
Gragl (2013), p. 247; Although (Ibid, p. 248), Gragl argues that if this procedure is located
under the CJEU’s accelerated procedure in accordance with Art. 23a to the CJEU Statute that
would not amount to interference to the EU law autonomy. I would fully agree with this point from
a general perspective, although, according to Gragl, such accelerated procedure may be utilized
only in cases of ‘exceptional urgency’, which may not still be the case in view of the prior
involvement procedure. This being the case, it would be more certain to argue that the ‘undue
delay’ will certainly have a cost for the EU law autonomy.
8.6 Refraining from Undue Delay: How Can This Be Met? 339

a measure came to the knowledge of an applicant, as the starting point for the period
prescribed for instituting proceedings, is subsidiary to the criteria of publication or
notification of the measure’.169 Notification of a plaintiff, on the other hand, would
be described as ‘the operation by which the author of a decision of individual
relevance communicates the latter to the addressees and thus puts them in a position
to take cognizance of it’.170
Of course, the element of Art. 263 (6) TFEU that is important in this regard is the
publication of the act, as the Commission under Art. 263 (2) TFEU would proceed
the action for annulment against an act of general nature (as opposed to a decision).
The time relevance of such an act is therefore the moment when it was published in
the Official Gazette of the EU. Practically, the question would be whether the
Commission could trigger prior involvement on a case pending before Strasbourg
Court, an element of that case being an EU legal act which should have practically
been published far before the 2-month period started to count. First, it is argued that
surpassing the time limit set in Art. 263 (6) TFEU for acts of general nature does not
constitute a breach of legal certainty, as such acts have never had the form of
individual decisions which individual applicants have had the chance to attack
under the action for annulment and did not do it.171 This argument applies only to
cases on which the prior involvement procedure has been triggered. Lock argues
that the 2-month time limit should not be interpreted as applying over acts of
general nature, such as those that are usually reviewed by the prior involvement
procedure. To make his argument based, Lock suggests that Accrington Beef is one
of the cases wherein the Luxembourg Court left the space to think that the 2-month
period does not apply on acts of general nature.172 Examining Atzeni173 leaves the
same impression. Therefore, it is suggested that the 2-month time-limit would not
apply in the prior involvement procedure if Commission was to use Art.
263 (2) TFEU as the legal basis for triggering this involvement. However, one
must argue here that Art. 263 (6) TFEU does textually very clearly read that the
2-month time-limit does apply also upon acts of general nature of the EU, from the
moment they become published. That being the case, if the Luxembourg Court
would reject the Commission from considering the 2-month time-limit under Art.
263 (6) TFEU as inapplicable upon acts of general nature, then Art. 263 (2) cannot
be practically used by the Commission as a legal basis to trigger the prior involve-

169
Court of Justice of EU, Alitalia v Commission, Case T-296/97 [2000] ECR II-3871, para. 61.
170
Court of Justice of EU, Olsen v Commission, Case T-17/02 [2005] ECR II-2031, para. 74.
171
Lock (2011), p. 28; See also: Court of Justice of EU, Textilwerke Deggendorf, Case C-188/92
[1994] ECR I-833, para. 13 et seq.
172
Lock (2011), p. 29; See also the case: Court of Justice of EU, Accrington Beef, Case C-241/95
[1996] ECR I-6699; Cf.: Gragl (2013), p. 245–247 (Gragl stands in the same line of argument).
173
Court of Justice of EU, Atzeni and others, Joined Cases C-346/03 and 529/03 [2006] ECR
I-1875, para. 30 et seq.
340 8 EU Prior-Involvement Review

ment procedure. With that practice, a new legal basis must then be added or
supposed for the Commission.174
From the moment the case reaches the Luxembourg Court, in whichever manner
or legal basis, the latter might use Art. 23a of the Statute of CJEU to provide an
accelerated procedure in order to satisfy the ‘undue delay’ prescribed in Art.
3 (6) DAA. Art. 23a of the Statute of CJEU for this reason provides that: ‘The
Rules of Procedure may provide for an expedited or accelerated procedure and, for
references for a preliminary ruling relating to the area of freedom, security and
justice, an urgent procedure. Those procedures may provide, in respect of the
submission of statements of case or written observations, for a shorter period [. . .
and the possibility. . .] for the case to be determined without a submission from the
Advocate General.’ Having this legal basis on the Statute of CJEU, even if the
Luxembourg Court would use this accelerated procedure to satisfy the ‘undue
delay’ requested from the DAA and the Strasbourg Court, that would not touch
upon the principle of EU law autonomy, as clearly this category of procedure has an
EU primary-law basis. On the other hand, making the prior involvement a lengthy
process might have diminished the ‘fruitful’ cooperation that must appear between
the two courts.175

8.7 Legal Effects of the Prior Involvement Procedure


on the Reviewed EU-Law Provisions

Art. 3 (6) DAA reads that the decision of the Luxembourg Court on the prior
involvement procedure must be taken into account by the Strasbourg Court when it
resumes the adjudication of the case before it. That legal prescription must be
interpreted as meaning that the contested EU law provisions in the case before the
Strasbourg Court must be read by the latter in the way in which they were
interpreted or legally decided by the Luxembourg Court. This means that the
prior involvement decision on the EU law provision(s) which the Luxembourg
Court rules are part of the renewed facts in the case before the Strasbourg Court. If
the prior involvement procedure would have not produced legally effective decision
which must be taken into consideration by the Strasbourg Court as constituting a
new fact to the case before it, the existence of this mechanism would not have been
rationalized with the shield to the EU law autonomy that it offers. Therefore the

174
Cf. Lock (2011), p. 29; See also: Gragl (2013), p. 246/7, who points that the Luxembourg Court
would not interpret the timelimit standards contra legem.
175
On the ‘fruitful’ cooperation that must exist under the preliminary reference procedure, see:
Dauses (1986), p. 577; Cf.: However, Gragl and Lock rightly point out that if the prior involvement
undue delay principle would push the Luxembourg Court to relax the time limits on the current
actions, that would qualify as interference to the EU law autonomy. The case-law of the Luxem-
bourg Court, Gragl and Lock argue, speaks for a rigid application of the principle of legal certainty
on the existing time-limits as well. See: Gragl (2013), p. 244.
8.7 Legal Effects of the Prior Involvement Procedure on the Reviewed EU-Law. . . 341

Strasbourg Court would need to wait until the Luxembourg Court gives its ruling on
the prior-involvement EU law provision.176
Two possible scenarios might result from the prior involvement of the Luxem-
bourg Court: first, that the Luxembourg Court find valid, interpret differently or
invalidate an EU secondary law provision (paradoxically, Opinion 2/13—para.
247—does understand this merely as review of validity of secondary law, therefore
it requests that the DAA include the word ‘interpretation’ as well; see subchapter
supra on the Opinion 2/13 for a broader examination of this), and second, that the
Luxembourg Court make no change or interpret differently a provision of EU
primary law.177 In the first case, if the Luxembourg Court finds the provision of
the secondary law valid and needing no further interpretation, the Strasbourg Court
will continue its proceedings with no new fact in the case before it.178 If the
Luxembourg Court would interpret the secondary law provision differently from
the way it was interpreted by the Member State, then a renewed fact would be
attached to the case pending before Strasbourg Court. If the Luxembourg Court
finds the secondary law provision invalid, the Strasbourg Court might decide that
the EU legal basis for that violation never existed and continue the case with this
new fact (A very legitimate question would be whether the decision of Luxembourg
Court in prior involvement procedure may affect the res judicata decision of the
Member State’s judiciary now pending before the Strasbourg Court, and how would
that then effect that individual case.179 This all remains to be seen how it will work
in reality). The same logic would also apply in the second scenario, excluding here
the possibility for the Luxembourg Court to invalidate any part of the primary
law.180 Although primary law provisions may not be invalided by Luxembourg
Court, there may appear situations wherein the latter would need to solve a conflict
e.g. between a treaty provision of competence and a right under the Charter of
Fundamental Rights. Luxembourg Court may choose to interpret the Charter’s right
as having primacy over that treaty provision, therefore producing the effect of
‘invalidation’ of that treaty provision.181 In this exceptional scenario, nevertheless,

176
See also: Lock (2011), p. 31; See also Court of Justice of EU, Roquette Frères SA, Case C-94/
00, [2002] ECR I-9039, para 29, where the CJEU changes the context under which it judged the
case in the first instance; this showing an example on how prior involvement could be used by the
Luxembourg Court itself to change its own position on certain issues of EU law.
177
See a similar explanation to the preliminary reference procedure, at: Vink et al. (2009), p. 4.
178
Cf.: in a similar approach of EU Court: Court of Justice of EU, Da Costa, Joined Cases 28-30/
62, 1963 E.C.R. 31.
179
On the latter, see: Jacque (2011), p. 1022.
180
On the question of the scope of effect deriving from this supposed procedure, see: Dauses
(1986), p. 574, wherein it is suggested that the EU Court does usually try to keep a narrow scope of
effect of the decisions it takes under the preliminary reference procedure, silently omitting from
giving an erga omnes effect to such decisions. It is suggested that the same narrow scope of legal
effect might be followed in the prior involvement procedure before the EU Court as well.
181
An example innovatively mirrored from: Jacque (2011), p. 1005–1007.
342 8 EU Prior-Involvement Review

Luxembourg’s prior involvement may be to the same effect as if the review would
be undertaken on a provision of EU secondary law.
The new facts which can be produced by the prior involvement procedure—and
which must be taken into account by the Strasbourg Court, but may not bind the
latter in any respect182—may change the position of the claimant before the
Strasbourg Court but may also deem it not admissible for the EU to stand as a
co-respondent if any of its provisions of law are not anymore alleged to violate the
Convention’s rights in the case being judged. Besides these two new situations that
may apply, most likely the claimant would not lose the victim status before the
Strasbourg Court. The reason after this stands in the fact that the direct alleged
violation which the claimant has attacked before the Strasbourg Court is an act or
omission of the Member State(s), the latter having not been removed even though
the Luxembourg Court may have nullified some or all EU legal basis for such acts
or omissions.183 Even if the EU legal bases for such an act or omission have been
removed, the Member State has sovereign authority on those acts or omissions in
face of the Strasbourg Court. They may merely be removed, for purposes of the
claimant’s victim status before the Strasbourg Court, by the Member State’s
national court or tribunal only,184 or any other national-constitutional authorized
body. This being the case, the prior involvement mechanism, even if it would
invalidate the EU legal basis, may not be likely to dismiss the claimant from the
victim status. The largest possible effect would be that of EU stopping to fulfill
the conditions to stand as co-respondent to that case. Even though the prior
involvement may not produce a larger effect, certainly, its effect within the EU
legal system may be as large as that stemming from a direct action before the
Luxembourg Court.
One may need to point out that although the EU Court would invalidate the EU
legal basis that provoked the Convention’s violation, the claimant will continue to
consume the effects of the illegal act of the Member State’s court.185 This situation
will contradict to the standard devised by the Permanent Court of International
Justice’s Chorz ow case decades ago, which had asserted that: ‘The essential prin-
ciple contained in the actual notion of an illegal act . . . is that reparation must, as far

182
On the latter part of the argument, see: O’Meara (2011), p. 1827.
183
Cf.: Lock (2011), p. 31.
184
Cf.: for a general note on this: Baratta (2013), p. 1315.
185
If one would have been based in the principle of ICJ’s La Grand, then the EU Court should have
been considered as capable of offering to the applicant a remedy to his/her victim status as well. In
La Grand, ICJ has ruled that ‘where jurisdiction exists over a dispute on a particular matter, no
separate basis for jurisdiction is required by the Court to consider the remedies a party has
requested for the breach of the obligation.’ International Court of Justice, Germany v United
States of America (LaGrand), Merits, Judgment, 2001, para 48 (available at icj-cij.org/icjwww/
idocket/igus/igusframe.htm). It follows clearly from this logic—if applied upon the prior involve-
ment mechanism—that the EU Court should be considered as competent in intervening and
changing the legal situation of the applicant affected by its decision on basis of the prior
involvement procedure.
8.8 Is Prior Involvement a New Remedy: Defending the Non-defendable? 343

as possible, wipe out all the consequences of the illegal act and reestablish the
situation which would, in all probability, have existed if that act had not been
committed.’186 The EU Court’s jurisdiction under the prior involvement mecha-
nism would not be able of wiping out the illegal effect which the national court had
produced, thereby leaving the claimant still under a victim status.
Finally, if Strasbourg Court would practice to consider victim status of the
claimant dismissed if the EU Court revokes the EU law basis whereupon Member
State’s act or omission that caused the alleged violation derives, that would seem as
a practice that interferes to the EU law autonomy; because Strasbourg Court would
be seen as prejudicing the relationship and interpretation of EU law with regard to
EU Member States’ obligations under it, something that amounts to an interpreta-
tion of EU law.

8.8 Is Prior Involvement a New Remedy: Defending


the Non-defendable?

A core question regarding the prior involvement mechanism is whether it in fact


produces a new remedy, capable of affecting the autonomy of EU law or at least
making uncertain the current legal remedy system based on the Treaties. To
examine this issue, it is important to mention that Art. 3 (6) DAA provides the
opportunity that once a case is engaged through the prior involvement procedure by
the Luxembourg Court, after the procedure being triggered inside the EU, it is
‘thereafter for the parties to make observations to the Court.’187 This shows that
there is an opportunity for the parties involved to have their arguments heard before
the Luxemburg Court,188 therefore making the prior involvement procedure not
anymore a Luxembourg Court-only assessment but rather an arena in which parties
will be given the chance to be heard within the Luxembourg Court’s prior involve-
ment procedure. Therefore, it is argued that as long as the DAA provides for the
possibility for the parties to be heard through giving their own observations to the
case, the prior involvement provides a new remedy at least as private parties are
given access to the Court by providing observations.189

186
Permanent Court of International Justice, Factory at Chorzow, Judgment No. 13, 13 September
1928, Series A, No. 17, p. 47.
187
Art. 3 (6) DAA.
188
Cf.: AIRE Centre et al. (2013), p. 4, promoting the idea that claimant appearing before
Strasbourg Court—in case the prior involvement mechanism is utilized—should be given the
chance to be heard before Luxembourg Court as well (including a package of legal aid, due to the
fact that they become represented twice in a single process before two courts).
189
Cf.: Lock (2012), p. 185, arguing that ‘[t]he wording of Article 6(2) TEU does not suggest that
it also grants competence to the EU’s institutions to create a new procedure for the prior
involvement of the ECJ.’ This may not be taken for granted, however, as the EU Treaties may
not have prohibited that competence while simultaneously asking that the DAA uphold the EU
344 8 EU Prior-Involvement Review

A more complex question in this regard, therefore, is how to accommodate the


‘observations’ before the Luxembourg Court within the EU Treaties current rem-
edies. As assessed above, it is argued that the best possible Treaty space for
accommodating the Commission’s authority to trigger the prior involvement, and
the Court’s jurisdiction in this regard, is through Art. 263 TFEU, in which capacity
the EU Court judges an action for annulment. However, under Art. 263 TFEU
jurisdictions, it is not possible for private parties to make any observations to the
EU Court if the action for annulment comes from the Commission as a privileged
applicant. There is no such space even in the Statute of CJEU for private parties to
make observations to the Court within the context of an action for annulment
initiated by the Commission under Art. 263 TFEU. Therefore there is an
indispensible conflict in this case between the rooms offered by Art. 263 TFEU
and the Statute of CJEU and Art. 3 (6) DAA, the latter establishing the possibility
for parties to make observations before the EU Court. For purposes of Art.
3 (6) DAA, the term parties is logically and should reasonably be understood as
covering both privileged but also non-privileged applicants, as the very procedure
of prior involvement does first of all assess the compatibility of an EU law provision
with an alleged violation of a right claimed by a private party, the latter’s interest
and legal position being the very core focus of the entire procedure.190
There being an open conflict between Art. 263 TFEU and the role argued to have
been assigned to the Commission in that regard with Art. 3 (6) TFEU, there needs to
be considered another solution. The Commission’s Working Document, on the
other hand, suggests that the prior involvement procedure and the observations of
the parties should be accommodated through Art. 23 of the Statute of CJEU. Art.
23 of the Statute of CJEU foresees the possibility for written observations of the
parties through Art. 267 TFEU, the latter being the preliminary reference procedure
initiated by national courts. It should be maintained here that Art. 267 TFEU may
not serve at all as the basis for accommodating the prior involvement of the
Luxembourg Court, as has been extensively argued above. However, of course,
Art. 23 of the Statute of CJEU might serve as a better Treaty basis to accommodate
the issue of parties’ observations before the Luxembourg Court as prescribed by
Art. 3 (6) DAA. This is so because Art. 23 (2) of the Statute of CJEU establishes
that: ‘Within two months of this notification, the parties, the Member States, the
Commission and, where appropriate, the institution, body, office or agency which

peculiarities, one of such core peculiarities being preserved merely with the prior involvement
mechanism’.
190
Cf.: Commission’s Working Document DS 1930/10, Council of the European Union, FREMP
meeting (Brussels, 10 January 2011) on the topic ‘Previous involvement of the Court of Justice in
the context of the accession of the European Union to the European Convention for the protection
of Human Rights and Fundamental Freedoms’. Brussels, 22 December 2010, note 107, para. 10;
Cf.: also: Lock (2011), p. 28 et seq, for the logic on how to receive the idea and legal nature of the
prior involvement mechanism.
8.8 Is Prior Involvement a New Remedy: Defending the Non-defendable? 345

adopted the act the validity or interpretation of which is in dispute, shall be entitled
to submit statements of case or written observations to the Court.’191 Art. 23 of the
Statute therefore remains the only space within the Treaties wherein one could
make room for private parties which are claimants before the case in Strasbourg
Court, to submit observations to the Luxembourg Court in its prior involvement
procedure instigated as a consequence of the case pending before Strasbourg Court.
Although Art. 23 of the Statute of CJEU may stand as a valid point, it is argued that
such observations cannot be made based on this legal basis if the prior involvement
procedure was itself initiated under Art. 236 TFEU. Hence, either Art. 3 (6) should
be accepted as making a hidden amendment to the Treaties or merely as attaching
new powers to the Luxembourg Court (of course not touching its existing powers
under the Treaties), or as a means that somehow reconciles the current loopholes
between Art. 263 TFEU and Art. 23 of the Statute of CJEU for purposing of
accommodating the prior involvement mechanism and the right of the parties to
make observations before the EU Court under that procedure.
Another important issue is whether this practice—the right to observations of the
parties before the EU Court in cases of prior involvement—amounts to the estab-
lishment of a new legal remedy. A core fact here, as discussed above, is that Art.
3 (6) DAA allows the possibility for parties to provide observations to the EU Court
in proceedings wherein the EU law provision is being reviewed. If the private
parties are allowed to make observations, then a new legal remedy is being made.
The founding argument behind this is that since a private party has not had the
opportunity to be heard somehow before the EU Court in such a case neither
through the preliminary reference procedure nor through the direct actions, then
the opportunity of the private to submit observations through the prior involvement
procedure institutes room for his/her arguments to be heard before the EU Court.
Although in the prior involvement procedure the proceedings are not instituted by
the private party—but rather by the e.g. the Commission—when the Luxembourg
Court assesses the validity or interprets the EU law provision it gives equal access
to the arguments of the private party. Hence, in terms of the possibility to present
the claimants’ argument, the possibility to submit observations to the EU Court
amounts to a hearing instrument for the claimants’ access to judicial review for that
contested EU law provision.192
In addition to the previous argument, the insistence to have the contested EU law
provision assessed before the Luxembourg Court through Art. 263 TFEU and Art.
267 TFEU does not comply with the jurisdiction which the Luxembourg Court
acquires from Art. 3 (6) DAA. In fact, neither Art. 263 TFEU nor Art. 267 TFEU
offer a fundamental rights review jurisdiction; instead, they both offer very broad
jurisdiction encapsulated within the context of the review of legality and broad-

191
Emphasis added.
192
Contra.: AIRE Centre et al. (2013), p. 2 (‘It does not appear that such a ruling would have any
consequences for the applicant’s case, so as to provide an effective remedy in accordance with
Article 13 ECHR.’).
346 8 EU Prior-Involvement Review

scope interpretation. Whereas, Art. 3 (6) TFEU does merely provide the EU Court
with the jurisdiction to review the compliance of the contested EU law provisions
with the Convention rights (all of which are also part of the Charter of Fundamental
Rights under the Treaties). This Convention-based fundamental rights jurisdiction
crafted by Art. 3 (6) DAA seems to be rather narrow and may not be portrayed as
already existing within the Treaties in that shape. Hence, the fundamental rights
jurisdiction which the EU Court obtains as a result of the prior involvement under
Art. 3 (6) DAA proves the second distinctive nature of this assumingly new remedy.
Both abovementioned arguments taken into account, the prior involvement proce-
dure seems to introduce a new legal remedy, rather restricted and not entirely on the
hands of the private parties; however, certainly leading to a new form of access to
the EU Court in a way not prescribed exactly so by the Treaties. This semi-access to
the EU Court193 could be considered as somehow supporting the indirect access to
the EU Court for those private parties that have reached Strasbourg Court without
having had the opportunity to benefit from the preliminary reference procedure.
Therefore, a valid argument that should be made here is that this new form of
accessing the EU Court stands as a superficial aim of the prior involvement
mechanism, the latter’s main aim being the need to fortify and ensure the EU
Court’s external autonomy.194 However, this can seriously touch upon the red line
of Protocol 8 of the Treaties. It would not be assured, however, whether Art.
3 (6) DAA as interpreted here would allow that the EU institutions ‘act within the
powers conferred on [them] in the Treaties’, as established in Art. 13 (2) TEU.195
Luxembourg Court’s Opinion 2/13 does not observe any of these extremely impor-
tant gaps that may pose threats to the EU law autonomy.
Finally, an important question of curiosity was raised by Jacque, who questions
whether prior involvement could ruin the concept of res judicata of cases that were
judged before the national courts.196 The author of this book argues that the prior
involvement mechanism does not in itself lead to ruining the concept of res
judicata, however it may make its application rather legally uncertain. Let us
suppose that case X that appeared before an EU Member State court has been
based in an EU law provision. The national court—either was not asked by the
claimant or simply did not accept to refer a preliminary question to the EU Court
regarding the validity of the EU law basis on which the case depends—has ruled
against the claimant in X. That decision has gone through all stages of legal review
in that Member State’s judicial system. Once the claimant of X submits the national
court’s judgment to the Strasbourg Court, the EU Court automatically acquires the

193
For a note on this argument, Cf.: Baratta (2013), p. 1327.
194
On external autonomy, see: Wessel and Blockmans (2013), p. 1; See also: Court of Justice of
EU, Opinion 1/91 (1991) ECR I-6079, paras 12–13; See the newest approach on the external
autonomy, at: Court of Justice of EU, Kadi and Al Barakaat v. Council and Commission, Joined
Cases C-402/05 P and C-415/05 P [2008] ECR I-6351.
195
Cf. Court of Justice of EU, Commission v. Ireland, C-459/03, [2006] ECR-I 4635, para 123.
196
Jacque (2011), p. 1022.
8.9 Does Prior Involvement Produce the Effect of a Hidden Amendment to the Treaties? 347

right to prior involvement (after an EU institution triggers that procedure). The EU


Court may find the legal basis whereupon case X depended as invalid. In this
scenario, the claimant of X would still continue to be a victim of the violation, as
his/her case is already res judicata, however the EU legal basis for that violation has
been ruled as invalid. Of course, this will complicate the res judicata principle
before the national courts, as the victimized claimant will be confronted with a res
judicata case which was based in an invalidated EU legal basis—adding that the EU
Court under the prior involvement mechanism would not have any legal capacity to
order the national court concerned to apply the restitutio inintegrum procedure.197
Basically, this will lead to a scenario where the certainty on res judicata would be
seriously harmed whereas restitutio inintegrum would remain inapplicable. This
said, it is argued that prior involvement may seriously harm res judicata certainty
before national courts for claimants whose cases were based in an EU legal basis,
obviously harming the relationship between the EU Court and national courts
as well.

8.9 Does Prior Involvement Produce the Effect of a Hidden


Amendment to the Treaties?

As known from Protocol 8 to the Treaties, the DAA may not amount to something
that changes or affects the Treaties198; however, it can for sure amount to something
that supplements the Treaties, as the notion ‘affects’ used in Protocol 8199 is
something that should be read as changing in the negative sense or infringing the
powers of the EU institutions, but of course not as something that would make such
powers simply more functional and stable towards an instrument wherein the EU
accedes to. The more realistic position in this regard would be to qualify the Draft
Accession Agreement as making certain ‘good’ but hidden amendments to the
Treaties, in a form which would merely make more stable and functional the EU
institutions and their powers, but clearly not affect negatively them. Baratta argues

197
See an example on the application of restitutio in integrum in international jurisdictions, at:
Inter-American Court of Human Rights, Aloeboetoe et al. v. Suriname, Judgment of September
10, 1993, (Ser. C) No. 15 (1994), para. 96.
198
ICJ’s Judge Gaja, for example, argues that in order for the prior involvement mechanism to
function it may need to be accommodated to the EU Treaties through an amendment to the latter.
See: Gaja (2013), p. 4; Cf.: Gragl maintains that the mere idea of introducing the prior involvement
mechanism through the DAA interferes to the EU law autonomy as it basically introduces a hidden
amendment to the EU Treaties. See: Gragl (2013), p. 237.
199
The same question remains unsolved at: Conforti (2010), p. 84; Cf.: Lock (2012), p. 184/5, who
argues that there may be no way to affect the powers of the institutions set forth in the Treaties,
something which does not contradict this book’s argument; See also: Ritleng (2012), which seems
putting doubts on whether the prior involvement mechanism will work without treaty
amendments.
348 8 EU Prior-Involvement Review

that a logical reading of Opinion 1/92 and Opinion 1/00 allow the understanding
that the Luxembourg Court has ruled that international treaties to which the EU
becomes a party may provide additional powers to the EU institutions, in a way
which would not affect the Treaties’ prescription negatively.200 That valid point
does very excellently serve the arguments presented below, but it may not be taken
as accepting that the DAA introduces a new layer of legal remedies in the EU
System, something being suggested by Baratta in the abovementioned note. The
new semi-remedies introduced do merely serve as making the Treaties’ system of
remedies working better in essence with what has already been written in the
Treaties themselves.201 That means that the Luxembourg Court’s powers are not
changed, adding here that ‘the different manner of seizing it does not entail the prior
involvement mechanism altering the essential character of the powers attributed to
[. . .it. . .] by the Treaties.’202
Hence, it must be accepted that the reconciliation203 of Art. 263 TFEU and Art.
267 TFEU with Art. 3 (6) DAA for purposes of making the prior involvement
mechanism work204 do in fact entail at least an casual hidden amendment to the
Treaties system of remedies and configuration of the EU Court’s jurisdiction, but
such reconciliation does in no way interfere or affect negatively the Court’s role
assigned by Art. 19 (1) TEU, namely ensuring that in the ‘interpretation and
application of the Treaties the law is observed.’ That said, the hidden amendment
to the Treaties does not seem to generate substantive changes that would affect the
powers of the EU Court and the system of legal remedies, rather such a hidden
amendment does simply make the Court’s function under Art. 19 (1) TEU more
functional205 to guarantee the rights under the Convention and the Charter of
Fundamental Rights, the latter being a Treaties’ core objective itself. Structural
changes in the form of a hidden amendment to the Treaties for purposes of making
Art. 3 (6) DAA work do not also seem to infiltrate an external influence that would
ruin the full autonomy of the EU law and the exclusive competence of the EU
Court,206 rather only the contrary is the result when interpreting Art. 3 (6) DAA
from a teleological and systemic perspective. Protocol 8 to the Treaties therefore is
fully maintained in the sense that the hidden amendment to the Treaties ensure an
even more functional exercise of powers assigned by the Treaties to the EU
institutions, although certain formalistic and structural changes do appear at the

200
Baratta (2013), p. 1329, referring to: Opinion 1/92, paras. 32 and 41 & Opinion 1/00, paras.
20 and 21.
201
Cf.: Lock (2011), p. 22.
202
Baratta (2013), p. 1330.
203
Compare this notion with the arguments presented in: Eckes (2012).
204
This contestation has also been questioned at: Conforti (2010), p. 84.
205
And clearly not amounting to something that would diminish the standard set in: Court of
Justice of EU, Costa v ENEL, Case 6/64 [1964] ECR 585, at 601.
206
Which according to the author stands also in line with the: Court of Justice of EU, Commission
v. Ireland (Mox Plant case), Case C-459/03 [2006] ECR I-4635, para. 123.
8.10 Chapter’s Summary of Conclusions 349

level of many other case-law amendments done on the Treaties by the EU Court
itself several times. The latter may be well rationalized with the soft law concept,207
certainly Art. 3 (6) DAA might be described as having that teleological nature.

8.10 Chapter’s Summary of Conclusions

The chapter examined the design, nature and legal effects of the prior involvement
mechanism, in light of the specificities of the Draft Accession Agreement and the
estimated consequences which it will produce in practice. The chapter offered
specialized analysis relating to the mechanics that underlying this mechanism,
and their legal understanding from both an EU constitutional law and Convention
law perspective, the latter being served with international law arguments as well. In
general, one may conclude that the prior involvement mechanism is an appreciated
innovation of the DAA that will sufficiently protect the EU law autonomy from its
own gaps in face of Strasbourg Court’s probable interferences.
The chapter first examined the legal procedural safeguards on basis of which this
mechanism justifies its application, offering certain arguments on the ways and
methods that may fulfill the criteria to make its application eligible. In
deconstructing the criteria that make its use eligible, the chapter argued that the
prior involvement mechanism may allow room for the Luxembourg Court to go far
further with its jurisdiction and not limit itself merely on the supposed violating EU
law provision. In veing offered this room, it is concluded that the prior involvement
mechanism may give Luxembourg Court far wider jurisdictional capacity. On the
other hand, the chapter offered an in-depth analysis as to the means and ways to
make the use of prior involvement mechanism accommodated within the EU
Treaties. The chapter concludes that the EU Treaties will see themselves positively
adjusted—if not formally amended—by the DAA as regards the prior involvement
mechanism. Such adjustments to the EU Treaties may be an unintended effect of
the DAA, but that remains the only possible route to make the application of the
prior involvement mechanism possible.
The chapter also carefully analyzed the legal mechanics on basis of which the
Luxembourg Court will utilize the prior involvement jurisdiction, therefore dem-
onstrating the legal effects that may come out as a result of the use of this
mechanism. The chapter therefore concludes that the Luxembourg Court’s effects
from the use of the prior involvement may either lead to adjusting the meaning of
certain primary law provisions or invalidation or validation of certain secondary
law provisions. On the other hand, the chapter also questions whether prior
involvement introduces certain amendments to the Treaties and if it becomes a
new legal remedy for EU-related applicants, therefore the chapter concludes that

207
E.g.: Abbott and Snidal (2000).
350 8 EU Prior-Involvement Review

there is a ‘yes’ to both of these questions: each of these ‘yes’ have different levels of
support from arguments.
Finally, the chapter concludes that the prior involvement mechanism will be a
highly valuable mechanism in international law as regards the means that may be
used to keep certain competing jurisdictions of international courts institutionally
relaxed and undisturbed. Therefore, the prior involvement mechanism will likely
keep the relationship between Strasbourg and Luxembourg courts legally certain,
ruling our allergic tendencies from Luxembourg on potential arguments that it is
not the court that normally first checks and interprets the EU law.

References

Abbott KW, Snidal D (2000) Hard and soft law in international governance. Int Organ 54
(3):421–456
AIRE Centre, AI, ICJ (2013) European Union accession to the European Convention on Human
Rights: briefing note on the accession agreement. Brussels, 6 September 2013
Albors-Llorens A (2012) Remedies against the EU institutions after Lisbon: an era of opportunity?
Camb Law J 71(3):507–536
Alter KJ (1996) The European court’s political power. West Eur Polit 19(3):458–487
Analysis (1997) The European community cannot accede to the European Convention on Human
Rights. Eur Law Rev 1:235–249
Arnull A (2011) The principle of effective judicial protection in EU law: an unruly horse. Eur Law
Rev 36(1):51–70
Balfour ADJ (2005) Application of the European Convention on Human Rights by the European
Court of Justice. Harvard Law School Student Scholarship Series. Paper 4, pp 1–57
Balfour ADJ (2007) Eliminating conflicting interpretations of the European Convention on Human
Rights by the European Court of Justice and the European Court of Human Rights: the PDIQ
system as a preventative solution. Intercult Hum Rights Law Rev 2:183–247
Baratta R (2013) Accession of the EU to the ECHR: the rationale for the ECJ’s prior involvement
mechanism. Common Market Law Rev 50(5):1305–1332
Besselink L, Reestman J-H (2008) The relative autonomy of the EU human rights standard. Eur
Constitut Law Rev 4(2):199–204
Besselink LFM (2013) Should the European Union ratify the European Convention on Human
Rights? Some remarks on the relations between the European Court of Human Rights and the
European Court of Justice. In: Føllesdal A et al (eds) Constituting Europe: the European Court
of Human Rights in a national, European and global context. Cambridge University Press,
Cambridge
Besson S (2009) European legal pluralism after Kadi. Eur Constitut Law Rev 5(2):237–264
Breuer M (2013) No Donum Danaorum! A reply to Daniel Thym’s “A Trojan Horse?”, VerfBlog,
2013/9/16, http://www.verfassungsblog.de/en/no-donum-danaorum-a-reply-to-daniel-thymsa-
trojan-horse/
Broberg M, Fenger N (2013) Variations in Member States’ preliminary references to the Court of
Justice—are structural factors (part of) the explanation? Eur Law J 19(4):488–501
Callewaert J (2014) The accession of the European Union to the European Convention on Human
Rights. Council of Europe, Strasbourg
Carrubba CJ, Murrah L (2005) Legal integration and use of the preliminary ruling process in the
European Union. Int Organ 59(2):399–418
Conforti B (2010) Comments on the accession of the European Union to the ECHR. Ital Yearb Int
Law 20:83–86
References 351

Craig P (2013) EU accession to the ECHR: competence, procedure and substance. Fordham Int
Law J 36(5):1114–1150
Craig P, de Búrca G (2003) EU law, 3rd edn. Oxford University Press, New York, chapter 6
Craig P, De Búrca G (2008) EU law: text, cases and materials, 5th edn. Oxford University Press,
Oxford
Dauses MA (1986) Practical considerations regarding the preliminary ruling procedure under
article 177 of the EEC treaty. Fordham Int Law J 10(3):538–577
de Rivery EM, Chassaing E (2013) EU accession to the ECHR: bringing about an aggiornamento
for EU case law? Overview of ECHR and EU case law. e-Competitions 1–16, N 61327
Dougan M (2004) National remedies before the Court of Justice: issues of harmonisation and
differentiation. Hart Publishing, Oxford
Eckes C (2012) Protecting supremacy from external influences: a precondition for a European
constitutional legal order? Eur Law J 18(2):230–250
Eckes C (2013) EU accession to the ECHR: between autonomy and adaptation. Mod Law Rev 76
(2):254–285
Eliantonio M, Stratieva N (2003) From Plaumann, through UPA and Jégo-Quéré, to the Lisbon
Treaty: The Locus Standi of Private Applicants under Article 230(4) EC through a political
lens. Maastricht Faculty of Law Working Paper 2009/13
Gaja G (2013) The “Co-Respondent Mechanism” according to the draft agreement for the
accession of the EU to the ECHR. ESIL Reflect 2(1):1–6
Gerards J (2011) Pluralism, deference and the margin of appreciation doctrine. Eur Law J 17
(1):80–120
Gormley LW (2006) Judicial review: advice for the deaf? Fordham Int Law J 29(4):655–689
Gragl P (2013) The accession of the European Union to the European Convention on Human
Rights. Hart Publishing, Oxford
Groussot X et al (2011) EU accession to the European Convention on Human Rights: a legal
assessment of the Draft Accession Agreement of 14th October 2011. Fondation Robert
Schuman/European Issues N 218/7 November 2011
Groussot X, Minssen T (2007) Res Judicata in the Court of Justice case-law. Eur Constitut Law
Rev 3(3):385–417
Harding C (2000) The identity of European law: mapping out the European legal space. Eur Law J
6(2):128–147
Hart JW (2010) The European human rights system. Law Libr J 102(4):553–559
Heringa AW, Verhey L (2011) The EU Charter: text and structure. Maastricht J Eur Comp Law 8
(1):11–32
Jacque JP (2011) The accession of the European Union to the European Convention on Human
Rights and fundamental freedoms. Common Market Law Rev 48(4):995–1023
Kakouris CM (1997) Do the Member States possess procedural “autonomy”? Common Market
Law Rev 34(6):1389–1412
Koch C (2005) Locus standi of private applicants under the EU constitution: preserving gaps in the
protection of individuals’ rights to an effective remedy. Eur Law Rev 30(4):511–527
Komárek J (2013) The place of constitutional courts in the EU. Eur Constitut Law Rev 9
(3):420–450
Krüger HC (2002–2003) Reflections concerning accession of the European Communities to the
European Convention on Human Rights. Pa State Int Law Rev 21(1):89–99
Landau EC (2008) A new regime of human rights in the EU? Eur J Law Reform 10(4):557–575
Lawson R (2005) Human rights: the best is yet to come. Eur Constitut Law Rev 1(1):27–37
Lenaerts K (2012) Exploring the limits of the EU Charter of fundamental rights. Eur Constitut Law
Rev 8(3):375–403
Lenaerts K, de Smijter E (2001) The Charter and the role of the European Courts. Maastricht J Eur
Comp Law 8(1):90–101
Lock T (2010) EU accession to the ECHR: implications for judicial review in Strasbourg. Eur Law
Rev 35(6):777–798
352 8 EU Prior-Involvement Review

Lock T (2011) Walking on a tightrope: the Draft Accession Agreement and the autonomy of the
EU legal order walking on a tightrope. A pre-edited version; forthcoming with CMLR 2011.
Available at: http://papers.ssrn.com/sol3/papers.cfm?abstract_id¼1825227
Lock T (2012) End of an epic? The draft agreement on the EU’s accession to the ECHR. Yearb Eur
Law 31(1):162–197
Maduro MP (2003) Contrapunctual law: Europe’s constitutional pluralism in action. In: Walker N
(ed) Sovereignty in transition. Essays in European law. Hart Publishing, Oxford
Martı́n J, De Nanclares P (2013) The accession of the European Union to the ECHR: more than
just a legal issue. Working Papers on European Law and Regional Integration WP IDEIR n
15. Instituto de Derecho Europeo e Integracion Regional (IDEIR), Universidad Complutense
Miiller CA (2007) Fundamental rights in multi-level legal systems: recent developments in
European human rights practice. Interdiscip J Hum Rights Law 2(1):33–60
Morano-Foadi S, Andreadakis S (2011) Reflections on the architecture of the EU after the Treaty
of Lisbon: the European judicial approach to fundamental rights. Eur Law J 17(5):595–610
O’Meara N (2011) “A More Secure Europe of Rights?” The European Court of Human Rights, the
Court of Justice of the European Union and EU accession to the ECHR. German Law J 12
(10):1813–1832
Odermatt J (2014) The EU’s accession to the European Convention on Human Rights: an
international law perspective. Leuven Center for Global Governance Studies. Working Paper
No. 136
Payandeh M (2011) Constitutional review of EU law after Honeywell: contextualizing the
relationship between the German Constitutional Court and the EU Court of Justice. Common
Market Law Rev 48(1):9–38
Peers S (2006) Bosphorus – European Court of Human Rights. Eur Constitut Law Rev 2
(3):443–455
Pérez AT (2012) Spanish Constitutional Court, constitutional dialogue on the European arrest
warrant: the Spanish Constitutional Court knocking on Luxembourg’s door; Spanish Consti-
tutional Court, Order of 9 June 2011, ATC 86/2011. Eur Constitut Law Rev 8(1):105–127
Perez AT (2014) Too many voice? The prior involvement of the Court of Justice of the European
Union. In: Kosta V et al (eds) The EU accession to the ECHR. Hart Publishing, Oxford
Phelps J (2006) Comment: reflections on bosphorus and human rights in Europe. Tulane Law Rev
81:251–279
Quirico O (2010) Substantive and procedural issues raised by the accession of the EU to the
ECHR. Ital Yearb Int Law 20:31–53
Raba K (2013) The accession of the European Union to the European Convention for the
Protection of Human Rights and fundamental freedoms—overview of the accession agree-
ment. ERA Forum 14:557–572
Ragolle F (2003) Access to justice for private applicants in the community legal order: recent (R)
evolutions. Eur Law Rev 28(1):90–101
Ribble MR (2010–2011) I don’t trust your judgment: the European Convention on Human Rights
meets the European Union on new grounds? Pa State Int Law Rev 29(1): 211–232
Rieder C (2005) Protecting human rights within the European Union: who is better qualified to do
the job-the European Court of Justice or the European Court of Human Rights? Tulane Eur
Civil Law Forum 20:73–107
Ritleng D (2012) The accession of the European Union to the European Convention on Human
Rights and fundamental freedoms: a threat to the specific characteristics of the European Union
and Union Law? University of Uppsala, Faculty of Law Working Paper 1. Available at: http://
uu.diva-portal.org/smash/record.jsf?pid¼diva2:526830
Shelton D (2003) The boundaries of human rights jurisdiction in Europe. Duke J Comp Int Law
13:95–153
Tabarelli M (2013) The influence of the EU and the ECHR on “Parliamentary Sovereignty
Regimes”: assessing the impact of European Integration on the British and Swedish Judi-
ciaries. Eur Law J 19(3):340–363
References 353

Thym D (2013a) A Trojan Horse? Challenges to the primacy of EU law in the draft agreement on
accession to the ECHR, VerfBlog, 2013/9/11, http://www.verfassungsblog.de/en/atrojan-
horse-challenges-to-the-primacy-of-eu-law-in-the-draft-agreement-on-accession-to-the-echr/
Thym D (2013b) Separation versus fusion – or: how to accommodate national autonomy and the
Charter? Diverging visions of the German Constitutional Court and the European Court of
Justice. Eur Constitut Law Rev 9(3):391–419
Usher JA (2005) Direct and individual concern – an effective remedy or a conventional solution.
Eur Law Rev 28(4):575–600
van den Berghe F (2010) The EU and issues of human rights protection: same solutions to more
acute problems? Eur Law J 16(2):112–157
van Gerven W (2005) The European Union – a polity of states and people. Hart Publishing, Oxford
van Rossem JW (2009) Interaction between EU law and international law in the light of Intertanko
and Kadi: the dilemma of norms binding the Member States but not the community. Neth
Yearb Int Law 40:183–227
Vink M et al (2009) Explaining the use of preliminary references by domestic courts in EU
Member States: a mixed-method comparative analysis. Paper presented in Panel 6B ‘Judicial
Politics in the EU and Beyond’, 11th Biennial Conference of the European Union Studies
Association, Friday April 24, 2009
Von Bogdandy A et al (2012) Reverse Solange – protecting the essence of fundamental rights
against EU Member States. Common Market Law Rev 49(2):489–519
von Bogdandy A, Ioannidis M (2014) Systemic deficiency in the rule of law: what it is, what has
been done, what can be done. Common Market Law Rev 51(1):59–96
Ward A (2001) Judicial architecture at the cross-roads: private parties and challenge to EC
measures post Jégo-Quéré. Cambr Yearb Eur Leg Stud 4:413–444
Wessel RA, Blockmans S (eds) (2013) Between autonomy and dependence: the EU legal order
under the influence of international organisations. T.M.C. Asser Press/Springer, The Hague
Wetzel JR (2003) Improving fundamental rights protection in the European Union: resolving the
conflict and confusion between the Luxembourg and Strasbourg Courts. Fordham Law Rev
71:2823–2862
White S (2010) The EU’s accession to the Convention on Human Rights: a new era of closer
cooperation between the Council of Europe and the EU? New J Eur Crim Law 1(4):433–446
Part III
A Strasbourg Perspective on Applications
of EU-Law Origin
Chapter 9
Testing the Co-respondent Mechanism from
the Strasbourg Court’s Perspective: Three
Distinctive Cases with Three Distinctive
Scenarios

9.1 Introduction

This chapter examines the functionality and efficacy of the co-respondent mecha-
nism, applied to three seminal cases of relevance to an EU-Member State relation-
ship before the Strasbourg Court in a post-accession scenario. The chapter responds
to the research sub-question: ‘How functional and legally certain is the
co-respondent mechanism if applied to individual and real cases?’ The
co-respondent mechanism therefore is tested adjacent to Mathews, Bosphorus,
and Kokkelvisserij cases, examined cautiously, testing the functionality of mechan-
ics, overall procedural process and implications that this mechanism will likely
demonstrate post-accession. The chapter offers an original and provocative analysis
of the loopholes that this mechanism may demonstrate, keeping a special focus on
the EU law autonomy question and the problems relating to the application of a
joint liability model for the EU and its Member States. The chapter also questions
how the Strasbourg Court would normally carry out the co-respondent procedures
upon these cases, investigate the origins of the violation, and then share the burden
based on the complex duty of non-interference. While examining modes as a result
of which co-respondents may join the proceedings before the Strasbourg Court, this
chapter also proposes arguments in relation to the facts which need to appear in
order for co-respondents to fulfill the eligibility criteria. Obviously, a specific
analysis of these three cases identifies practical loopholes within the mechanics
of the co-respondent procedures wherein the Strasbourg Court may touch upon EU
law autonomy, the latter being an issue which is examined not only from a bottom-
up perspective but also from a top-down outlook that the Strasbourg Court may
conceivably apply. The chapter concludes that the test upon these three seminal
cases demonstrates the lack of full functionality of the co-respondent mechanism,
and the legal implications that may arise therefrom. A similar vagueness—under
the DAA—may also be concluded for the part of referrals to the Grand Chamber of
the Strasbourg Court.

© Springer International Publishing Switzerland 2015 357


F. Korenica, The EU Accession to the ECHR, DOI 10.1007/978-3-319-21759-8_9
358 9 Testing the Co-respondent Mechanism from the Strasbourg Court’s. . .

9.2 A Starting Note to the Three Tests

As noted above, it is important to consider the co-respondent mechanism in light of


the previous practical cases, and evaluate the potential implications that might
derive therefrom. The Explanatory Report of the Accession Treaty points that
only three cases to date have been filed before the Strasbourg Court which would
have most likely activated the co-respondent mechanism, namely Bosphorus,
Mathews and Kokkelvisserij. Tulkens supports the same view.1 Therefore, to assess
the suitability, design and certainty of the co-respondent mechanism, one needs to
test it in each of these three cases individually, and measure its functionality against
the above-proclaimed standards in each of them. Let us start with Bosphorus first.

9.2.1 Testing Bosphorus with the Co-respondent Mechanism

Bosphorus had concerned the allegation of an applicant, Bosphorus Hava Yolları


Turizm Airline Company, which had claimed that the company’s right of peaceful
enjoyment of property was violated by Ireland. In that setting, the applicant
company had filed an application against the Republic of Ireland before the
Strasbourg Court, alleging that Republic of Ireland has violated the applicant
company’s right guaranteed by the Convention.2 As a result of the United Nations
Security Council sanctions on Federal Republic of Yugoslavia, legally effected
through Resolution 820 (1993) of the United Nations Security Council, the
European Union adopted Regulation (EEC) no. 990/93 to concretely implement
the UNSC Resolution,3 Article 8 of the EU Regulation being the legal basis for the
Republic of Ireland’s measure attacked before the Strasbourg Court for its alleged
non-compliance with the Convention rights.4 The applicant, most concretely,
‘alleged that the impounding of its leased aircraft by the respondent State had
breached its rights under Article 1 of Protocol No. 1.’5 Such impoundment is in fact
a measure of the EU Member State which claims to have taken such a measure on
the basis of an EU law obligation.6 The case has been adjudicated by the Grand
Chamber, with Ireland suspected to have violated the company’s right by
impounding applicant’s leased aircraft. The case involved a Member State, the

1
Tulkens (2013).
2
E.g.: van den Berghe (2010), p. 117; See also: Douglas-Scott (2006), p. 243 et seq; Miiller
(2007), p. 38 et seq; Škugor (2012), p. 115.
3
E.g.: Papangelopoulou (2006), p. 133 et seq; See also: Gragl (2013), p. 70.
4
ECtHR, Bosphorus Hava Yollari Turizm ve Ticaret Anonim Şirketi v. Ireland, Application
No. 45036/95 (Judgment of 30 June 2005), para. 22 et seq.
5
ECtHR, Bosphorus, Application No. 45036/95 (Judgment of 30 June 2005), note 4, para. 3; See
also: Peers (2006), p. 446; See also: Defeis (2007–2008), p. 8.
6
E.g.: Albi (2009), p. 62; See also: Licková (2008), p. 481.
9.2 A Starting Note to the Three Tests 359

Republic of Ireland, which had taken the action of implementing an EU Regulation.


The “measure” of Ireland, in fact, was the final-step act that was alleged to have
made the applicant a victim. In this scenario, the alleged violating provision was
located in EU law, whereas the action that implemented the said EU Regulation was
undertaken by the EU Member State, Ireland. This said, in order for the applicant to
fulfill the admissibility criteria before the Strasbourg Court, besides proving that
he/she is a victim of such an act, the domestic legal remedies need be exhausted.
Most practically, the applicant in the Bosphorus scenario should have exhausted the
Ireland’s national domestic remedies, in order to be able to submit Ireland as the
respondent before the Strasbourg Court.
Since the implementing measure of the EU Regulation in question was under-
taken by the Republic of Ireland, the applicant company first used the domestic
legal remedies to seek judicial protection of its rights. The applicant filed (on 21 June
1994) an application first to the High Court, asking whether the Republic of Ireland
Ministry of Transport’s measure of impounding its aircraft was taken within the
context of UNSC and EU obligations.7 The High Court ruled that was not an issue
of the UNSC Resolution, as that would not be automatically incorporated into
Ireland’s legal order, but rather a duty under EU law.8 Even though the High
Court accepted that the measure of impoundment of the Ireland’s Ministry of
Transport derives from an obligation of the EU law, the measure as such was
ultra vires of Art. 8 of Regulation (EEC) no. 990/93. This being a negative result
for the Ministry of Transport, the latter filed an application to the Republic of
Ireland Supreme Court, requesting the latter to submit a preliminary reference
question to the EU Court on whether Article 8 of Regulation (EEC) no. 990/93
on this issue is valid.9 On 30 July 1996, the EU Court answered to the preliminary
reference question by ruling that Regulation (EEC) no. 990/93 applied exactly to
the type of the impounded aircraft.10 On 29 November 1996, being based on the
preliminary reference answer from the EU Court, the Irish Supreme Court issued its
judgment accepting the appeal of the Minister of Transport, ruling that the Ministry
of Transport’s measure of impoundment was one merely resulting from the obli-
gation of Article 8 of Regulation (EEC) no. 990/93.11
The question of the origin of the legal basis of the alleged violation—most
practically, the impoundment of the applicant company’s aircraft derives from

7
ECtHR, Bosphorus, Application No. 45036/95 (Judgment of 30 June 2005), note 4, para. 35; See
also: Hoffmeister (2006), p. 443.
8
ECtHR, Bosphorus, Application No. 45036/95 (Judgment of 30 June 2005), note 4, para. 35.
9
ECtHR, Bosphorus, Application No. 45036/95 (Judgment of 30 June 2005), note 4, para. 43; See
also: Papangelopoulou (2006), p. 133; De Londras and Kingston (2010), p. 372; Hoffmeister
(2006), p. 443.
10
ECtHR, Bosphorus, Application No. 45036/95 (Judgment of 30 June 2005), note 4, para. 52; De
Londras and Kingston (2010), p. 372; Hoffmeister (2006), p. 443.
11
ECtHR, Bosphorus, Application No. 45036/95 (Judgment of 30 June 2005), note 4, para 58.
360 9 Testing the Co-respondent Mechanism from the Strasbourg Court’s. . .

UNSC Resolution 820 (1993), adopted on 17 April 1993, which per se would not
make sense neither for the EU nor for the Republic of Ireland.12 However, such an
origin of law was made legally binding when the Council of the EU transposed the
UNSC 820 (1993) into the Regulation (EEC) no. 990/93 which came into force on
28 April 1993.13 A core question here would be whether the Member State,
respectively the Republic of Ireland, had any discretion in applying Art. 8 of this
Regulation when impounding the applicant company’s aircraft.14 Art. 8 of the
Regulation read: ‘All vessels, freight vehicles, rolling stock and aircraft in which
a majority or controlling interest is held by a person or undertaking in or operating
from the [FRY] shall be impounded by the competent authorities of the Member
States.’ It was clear that the measure of the Republic of Ireland of impounding the
applicant company’s aircraft had originated directly from Art. 8 of the EEC
Regulation15—the latter being directly effective vertically and horizontally—and
those Irish authorities had absolutely no discretion on whether to impound or not
the applicant company’s aircraft.16 The Member State has no chance of departing
from this EU law obligation. Let us now test the co-respondent mechanism in
this case.
Are basic conditions for the co-respondent mechanism fulfilled by the EU under
Bosphorus—If Bosphorus case would be filed before the Strasbourg Court once the
EU accedes to the ECHR, then Art. 3 (2) DAA would serve as the basis for the EU to
join as a co-respondent to the case. In practice, the applicant company would
challenge the Republic of Ireland’s measure of impoundment before the Strasbourg
Court, with such impoundment originating in an EU directly-applicable regulation.
Therefore, the original respondent which the applicant company would have named
in this scenario would be the Republic of Ireland, adding here that the latter’s
domestic remedies would have been exhausted. Besides, the Supreme Court of
Republic of Ireland would refer the question for preliminary reference—with the

12
ECtHR, Bosphorus, Application No. 45036/95 (Judgment of 30 June 2005), note 4, para. 64.
13
ECtHR, Bosphorus, Application No. 45036/95 (Judgment of 30 June 2005), note 4, para. 65; See
also: De Londras and Kingston (2010), p. 395; On the legal origin of the violation, see: Peers
(2006), p. 447; See also: Papangelopoulou (2006), p. 133; Cf.: Gragl rightly points out that in
Bosphorus the Strasbourg Court did not make it clear—and did not rule in a straightforward
manner—where the legal origin of the infringement derived from (Gragl refers to Brohmer on this
argument). See: Gragl (2013), p. 70.
14
Cf.: Cardwell et al. (2009), p. 231, compared to the Advocate General argument in Kadi, who
maintained that although the Community may be bound by an international obligation leaving no
discretion, by contrast to the standard ruled by Bosphorus, its court may not apply that obligation
unconditionally and with ‘complete acquiescence’, requesting that it be reviewed against basic ius
cogens on human rights. If the discretion of Ireland on Bosphorus would be seen in this line, than
Ireland would have probably needed not to make the application of the Regulation unconditional,
which would then have changed the direction of this case; See also: Kokott and Sobotta (2012),
p. 1018 et seq.
15
Douglas-Scott (2006), p. 246.
16
E.g.: van den Berghe (2010), p. 115; See also: Phelps (2006), p. 270; Hoffmeister (2006), p. 444;
Peers (2006), p. 447.
9.2 A Starting Note to the Three Tests 361

request of the Ministry of Transport rather than the applicant company—the answer
of the EU Court being one which clearly rules that Art. 8 of the contested Regulation
is binding on Ireland. Basically, this being the situation, since the original respondent
is an EU Member State and the proceedings before the Strasbourg Court would
involve an allegation which calls into question the compatibility with the Convection
rights of an EU law provision, then the EU would be more than interested to join the
proceedings before the Court. Still, another more important subcondition would need
to be fulfilled before the EU could obtain the right to co-respondent status.
Whether the alleged violation could have been avoided simply by disregarding an
obligation under EU law—Obviously, although Bosphorus involves an EU law
provision as the basis for the alleged violation of the Convention rights, the EU
could only join as a co-respondent if it is proved that the EU law provision (Art. 8 of
the Regulation concerned) could have only been avoided by disregarding an obliga-
tion under EU law. To assess whether the Republic of Ireland could have disregarded
Art. 8 of the Regulation (EEC) no. 990/93, one needs to assess first, whether the
regulation concerned could have been disregarded by the Republic of Ireland while
not breaching any obligation under the EU law, and, second, whether the contested
Article 8 of the contested regulation leaves room for the Republic of Ireland to escape
from its duties by using state discretion. As to the first point, the regulation concerned
is directly applicable and binding upon all Member States according to Art.
288 (2) TFEU. Second, Article 8, the contested EU law provision in the proceedings
before the Strasbourg Court, leaves absolutely no state discretion to the Republic of
Ireland, other than impounding mandatorily the aircraft. Therefore, it is argued here
that the Republic of Ireland would have had no chance to escape from the impound-
ment of the aircraft without breaching the obligation under the Treaties, respectively
Art. 288 (2) TFEU. These being the two criteria passed, the EU would have standing
to join as a co-respondent before the Strasbourg Court in the Bosphorus case.
Who would become liable if the Court would decide that Bosphorus violates the
Convention, and would this amount to an interpretation of the EU law by the
Strasbourg Court—Since the Republic of Ireland has had absolutely no discretion
on implementing Art. 8 of the Regulation, the impoundment of the aircraft of the
applicant company was merely a causal result of the EU Regulation. This being the
case, although the original respondent was the Republic of Ireland, in fact, the
alleged violation of the applicant company originated merely from EU law. Once
the EU has joined as a co-respondent, the Court’s decision would be one which
would find both co-respondents (the EU and the Republic of Ireland) as the liable
parties for the violation, although in their join statement the EU and Ireland would
agree that in practice, only the EU would have the liability to remedy that alleged
violation. On the other hand, the doubt is whether the Strasbourg Court would pass
its own subsidiary borders, and interpret the EU law either directly or indirectly.
The Strasbourg Court prescribed that:
362 9 Testing the Co-respondent Mechanism from the Strasbourg Court’s. . .

For these reasons, the Court finds that the impugned interference was not the result of an
exercise of discretion by the Irish authorities, either under Community or Irish law, but
rather amounted to compliance by the Irish State with its legal obligations flowing from
Community law and, in particular, Article 8 of Regulation (EEC) no. 990/93.17

The above paragraph of Bosphorus demonstrates that the Court went to assess
and find whether the impugned interference to the applicant company rights was an
exercise of state discretion of the Republic of Ireland or merely a duty under which
Ireland had to fulfill its obligation under EU law. Such assessment of the Court was
obviously conducted by interpreting the internal relationship between the Republic
of Ireland and Art. 288 (2) TFEU.18 Without having assessed such relationship the
Strasbourg Court would have never argued that the Republic of Ireland has in fact
no state discretion in implementing the measure from Art. 8 of the contested
Regulation. That behavior certainly amounts to an indirect interpretation of EU
law, as an internal relationship between the EU Treaties and the Republic of Ireland
national legal order was interpreted and assessed as a precondition for the case to be
resolved by the Court.19 Hence, interference with EU law autonomy had existed in
Bosphorus,20 but such interference could become more limited if the co-respondent
mechanism is established, as the Court would not delve into interpreting the internal
relationship between the co-respondents if they are required to agree beforehand on
the share of their responsibility through a friendly settlement (although the risk still
would have existed with this example).21

17
ECtHR, Bosphorus, Application No. 45036/95 (Judgment of 30 June 2005), note 4, para. 148.
18
Gragl, referring to Lock and others, rightly argues in general that the Strasbourg Court often
turns to the interpretations of national constitutional courts and to the domestic law and sometimes
even interprets it. Gragl rightly argues that the national law of a contracting party is not always
considered by the Strasbourg Court as mere ‘fact’. See: Gragl (2013), p. 116. This line of argument
has be found to exist also in the above examined case relating to EU law, therefore concluding that
it is not always safe to consider that Strasbourg Court would never engage with the interpretation
of EU law for the needs of its jurisdictional activity.
19
Cf.: ECtHR, Bosphorus, Application No. 45036/95 (Judgment of 30 June 2005), note 4, para.
151, wherein the Court went even further, by asking: ‘The question is therefore whether, and if so
to what extent, that important general interest of compliance with Community obligations can
justify the impugned interference by the Irish State with the applicant company’s property rights.’.
20
Hoffmeister (2006), p. 447 (If the Strasbourg Court ‘were to fully review an EC act, it would, in
effect, be putting itself in a hierarchically superior position over the ECJ – a court that is, by
design, intended to have exclusive competence over matters of EC law.’).
21
See e.g.: Schimmelfennig (2006), p. 1248, who argues that there is an increasing trend of the
primacy of Strasbourg over Luxembourg, as the latter tries to take more influence in face of its
member states’ legal orders; On this ‘normal’ behavior of Strasbourg Court, see: Letsas (2004),
p. 279/80.
9.2 A Starting Note to the Three Tests 363

9.2.2 Testing Mathews with the Co-respondent Mechanism

In Mathews case,22 the applicant (Ms. Denise Matthews) submitted an application


against the United Kingdom of Great Britain and Northern Ireland on 18 April
1994. The application concerned an alleged violation of the respondent state under
Article 3 of Protocol No. 1 taken alone or mutually with Article 14 of the Conven-
tion. The first alleged violation therefore being the breach of the Convention’s right
to free and equal elections seen under the Convention’s prohibition of discrimina-
tion for persons residing in Convention High Contracting Parties’ jurisdictions.23
Basically, the Court had to judge whether24 ‘Article 3 of Protocol 1 did apply to the
situation even though what was at issue was the right to elect members of a
supranational legislative organ within an organisation which was not a party to
the Convention or its Protocols.’25 More practically, the applicant was a citizen of
Gibraltar, the latter being a dependent territory of the United Kingdom.26 As a
dependent territory, Gibraltar was not recognized as being part of the United
Kingdom’s legal territory but rather as a personal dominion of the United King-
dom’s Queen. Such dominionship constitutes a territory which is under the factual
possession of the United Kingdom, but can produce problems as to the legal
authority that must be exercised over it if UK were to implement the law of an
international organization. The United Kingdom, however, had extended the appli-
cation of the Convention under its international legal authority to the territory of
Gibraltar as well.27 Mathews, most concretely, involved the request of Ms. Denise
Matthews that, as a citizen of Gibraltar on which the UK law applied, she should be
given equal access to the European Parliament elections in the same basis as all UK
citizens. To reach this objective, Ms. Mathews had applied to the Electoral Regis-
tration Officer for Gibraltar to become a registered voter for the European Parlia-
ment elections under the UK’s authority. The Electoral Registration Officer for
Gibraltar replied to the applicant by arguing that28:

22
ECtHR, Matthews v. United Kingdom, Application no. 24833/94, D. 18 February 1999. (Judge-
ment, Grand Chamber); See how this was interpreted without the EU being a party to the ECHR,
at: Govaere (2009), p. 11.
23
ECtHR, Matthews v. United Kingdom, Application no. 24833/94, D. 18 February 1999. (Judge-
ment, Grand Chamber), note 22, para. 1; See also: Licková (2008), p. 479; See also: Miiller (2007),
p. 37; See also: Rieder (2005), p. 79/80.
24
Peers (2006), p. 444.
25
Harding (2000), p. 143; See also: Peers (2006), p. 446.
26
ECtHR, Matthews v. United Kingdom, Application no. 24833/94, D. 18 February 1999. (Judge-
ment, Grand Chamber), note 22, para. 8; See also: Licková (2008), p. 480.
27
ECtHR, Matthews v. United Kingdom, Application no. 24833/94, D. 18 February 1999. (Judge-
ment, Grand Chamber), note 22, para. 19; See also a general background of facts, at: Gragl
(2013), p. 68.
28
ECtHR, Matthews v. United Kingdom, Application no. 24833/94, D. 18 February 1999. (Judge-
ment, Grand Chamber), note 22, para. 7.
364 9 Testing the Co-respondent Mechanism from the Strasbourg Court’s. . .

The provisions of Annex II of the EC Act on Direct Elections of 1976 limit the franchise for
European parliamentary elections to the United Kingdom. This Act was agreed by all
member States and has treaty status. This means that Gibraltar will not be included in the
franchise for the European parliamentary elections.29

Therefore, the answer of the Gibraltar Electoral Office revealed that the territory
and citizens of Gibraltar were not included in the EU law as being part of the United
Kingdom for purposes of the European Parliamentary elections and right to vote
thereon. The answer of the Gibraltar Electoral Office therefore is the alleged
omission of the United Kingdom which the applicant considers as violating her
Convention’s right to vote and the prohibition of discrimination. On basis of this
omission, the applicant acquired the chance to apply before the Strasbourg Court by
challenging the United Kingdom for having not guaranteed a positive measure
allowing Gibraltar citizens to enjoy their right to vote for the European Parliament
elections. However, the omission which the applicant considers as violating her
Convention’s right does seem not to originate in the UK law, because UK in fact has
no authority as such to determine the legislation on the registered EU citizens who
are recognized as having the right to vote in the European Parliament elections.
The question of the origin of the legal basis of the alleged violation—The
alleged UK’s omission to guarantee to Gibraltar citizens the right to vote under
its authority for the European Parliament elections does not in fact originate from
within the UK’s legal order. Instead, such an omission is an obligation which the
UK had to apply in respect to the EU law. More practically, the European Parlia-
ment elections were regulated with Article 138(3) of the EEC Treaty, which had
established that the Council had the authority to determine the proper legal pro-
visions which were to be recommended to the Member States and adopted in
accordance with their domestic constitutional procedures.30 On this legal ground,
Council Decision 76/787 had established such provisions,31 and then concretized
them further with the Act Concerning the Election of the Representatives of the
European Parliament by Direct Universal Suffrage of 20 September 1976,32
attached to the Council Decision 76/787. Annex II of the 1976 Act (forming part
of Council Decision 76/787) read: ‘The United Kingdom will apply the provisions
of this Act only in respect of the United Kingdom’.33 Therefore, it is argued that the

29
ECtHR, Matthews v. United Kingdom, Application no. 24833/94, D. 18 February 1999. (Judge-
ment, Grand Chamber), note 22, para. 7.
30
ECtHR, Matthews v. United Kingdom, Application no. 24833/94, D. 18 February 1999. (Judge-
ment, Grand Chamber), note 22, para. 17.
31
See e.g.: Lenaerts and de Smijter (2001), p. 93.
32
To note, ‘the 1976 Act cannot be challenged before the European Court of Justice for the very
reason that it is not a ‘normal’ act of the Community, but is a treaty within the Community legal
order.’ ECtHR, Matthews v. United Kingdom, Application no. 24833/94, D. 18 February 1999.
(Judgement, Grand Chamber), note 22, para. 33; Cf.: Peers (2006), p. 445 et seq; See also: van den
Berghe (2010), p. 122/3.
33
ECtHR, Matthews v. United Kingdom, Application no. 24833/94, D. 18 February 1999. (Judge-
ment, Grand Chamber), note 22, para. 18; See also: European Commission Of Human Rights,
9.2 A Starting Note to the Three Tests 365

United Kingdom omission challenged originates from the Council Decision 76/787,
which excludes Gibraltar from the voting rights for the European Parliament
stemming from Article 138(3) of the EEC Treaty. In the practical context, the
challenged United Kingdom omission—which the applicant alleges violates her
Convention rights—originates not from the United Kingdom’s law but rather from
the EU law, namely a Council Decision. The latter seems clearly not to give any
room to the United Kingdom to depart from that obligation, without having to
breach the EU Treaties. Let us now test the co-respondent mechanism in this case.
Are basic conditions for the co-respondent mechanism fulfilled by the EU under
Mathews—As examined above, it is clear that the United Kingdom caused an
alleged violation, and that the applicant challenged that omission before the Stras-
bourg Court as belonging to the United Kingdom; the United Kingdom being the
mere respondent which the applicant would have named even after the EU accedes
to ECHR. Hence, in this scenario, the original respondent on basis the of Art.
3 (2) DAA whose omission is attacked before the Strasbourg Court is the United
Kingdom. An EU law provision being the legal basis from which the UK’s omission
had resulted, the EU would be interested and have the space to utilize Art.
3 (2) DAA to protect its Council Decision 76/787, the latter forming part of
secondary law stemming from the legal obligation under Article 138(3) of the
EEC Treaty. Therefore, the legal origin of the UK’s omission is at the EU law
and the United Kingdom had no state discretion in implementing this obligation or
not. Without Council Decision 76/787 being revised, there would be no way of
remedying the alleged violation of the applicant’s right to vote for the European
Parliament and the prohibition of discrimination under the Convention. Since the
EU Court was neither involved in any way in the review of this case neither through
direct nor the preliminary reference procedure,34 the EU would also fulfill the
conditions of Art. 3 (6) DAA to utilize the prior involvement mechanism once
the case becomes admissible before the Strasbourg Court; Mathews would then get
hold until the EU Court takes the decision from the prior involvement procedure.
Whether the alleged violation could have been avoided simply by disregarding
an obligation under EU law—It is clear that the United Kingdom’s omission to
guarantee the right to vote for the European Parliament elections to Ms. Mathews
originated from a Council Decision. The Council Decision was, first, very clear in
establishing that the right to vote for the European Parliament with regard to the UK
belongs only to citizens legally attached to the territory of it. Here, the UK had no
state discretion in applying this duty as there was no room for the UK to choose
whether to give this right to Gibraltar citizens which were explicitly excluded from
the Council Decision. Such a decision was legally binding on the UK. On the other
hand, the Council Decision was a Treaty obligation for the Member States, includ-
ing here the UK; Not respecting Annex II of the Council Decision 76/787 would

Mathews, Application No. 24833/94, Report of the Commission, adopted on 29 October 1997,
para. 21.
34
See also: Vranes (2003), p. 10.
366 9 Testing the Co-respondent Mechanism from the Strasbourg Court’s. . .

have amounted to a breach by the United Kingdom of Article 138(3) of the EEC
Treaty. There was no way for the UK to omit to take the action of providing the
citizens of Gibraltar with a right to vote for the European Parliament without having
to breach the EEC Treaty. Thus, the alleged violation challenged by the applicant
before the Strasbourg Court could have not been caused without disregarding an
obligation under the EEC Treaty. These being the two criteria passed, the EU is
tested as having standing to have joined as a co-respondent before the Strasbourg
Court in the Mathews case.
Who would become liable if the Court would decide that Mathews violates the
Convention, and would this amount to an interpretation of the EU law by the
Strasbourg Court—First, as the alleged violation in Mathews originates in EU
law—namely the Council Decision 76/787—and the UK had no discretion on
whether to omit the territory of Gibraltar from the right to vote for the European
Parliament, then the origin of the alleged violation is EU secondary law, which can
merely be revised by the EU institutions (as opposed to EU primary law).35 Hence,
it would seem practical that it should be the EU which should take the liability for
this violation. However, as both the EU and the UK are (co-)respondents, they
should have agreed together through a friendly settlement that such liability be held
merely by the EU, whereas the Strasbourg Court would find both the EU and the
UK jointly responsible for having violated the Convention’s right to vote and
having discriminated against Ms. Mathews in this case.36 Second, Mathews
seems the prime example to demonstrate how the Strasbourg Court can in fact
enter the borders of EU law autonomy, and, by ruling on the compliance of EU law
with the Convention’s rights, interpreting EU law on top of the EU Court. By
finding that the origin of the Convention’s rights violation stems from the Council
Decision 76/787, the Strasbourg Court had a different interpretation of Article 138
(3) of the EEC Treaty. Since the Strasbourg Court found the Council Decision
76/787 being in violation of the Convention, it automatically and effectively
interprets Article 138(3) of the EEC Treaty as having a territorial application
upon Gibraltar as well.37 This interpretation would limit the discretion of the
Council to decide whether or not to include Gibraltar territory under the application
of Article 138(3) of the EEC Treaty, thereby implicitly forcing an interpretation of
EU primary law in a way which can no longer be denied even by the Council. To
prove this, the Strasbourg Court e.g. ruled in Mathews that:
According to the case-law of the European Court of Justice, it is an inherent aspect of EC
law that such law sits alongside, and indeed has precedence over, domestic law (see, for
example, Costa v. ENEL, 6/64 [1964] ECR 585, and Amministrazione delle Finanze dello

35
Cf. Garcı́a (2009), p. 196, who shows the circular way which UK and EU followed to remedy
this violation.
36
In the original Mathews, the Court did not apply the joint responsibility formula, as there was no
legal basis for that. See generally: Larsen (2008), p. 523/4.
37
This was argued to make the Convention apply to the full-range of the Contracting Party’s
jurisdiction and effectively-controlled territory. See: Callewaert (2007), p. 512.
9.2 A Starting Note to the Three Tests 367

Stato v. Simmenthal SpA, 106/77 [1978] ECR 629). In this regard, Gibraltar is in the same
position as other parts of the European Union.38

As exemplified above, the Strasbourg Court in Mathews went into evaluating the
meaning of EU law in the face of the domestic law of the Member States, by
extending such interpretation not only to EU law but to the seminal cases of the
EU Court.39 Not only that such an interpretation made explicitly, but the Court went
further to interpret that such an internal Treaty-based relationship between the EU
and Member States should be applied upon the territory of Gibraltar as well. The
same was done with the interpretation of the European Parliament’s scope of
representation, wherein the Strasbourg Court held that the European Parliament
should be interpreted as the ‘legislature’ for Gibraltar as well within Convention
terms.40 Such an interpretation amounts to guiding how the European Parliament’s
representing scope should be intended within the context of the Treaties.41 In
undertaking this interpretation,42 the Strasbourg Court ruled on the territorial scope
of the EU Treaty and its internal material regarding how the relationship between the
EU and Member States should be read.43 With this logic, the Strasbourg Court ‘could
[have] even “intervene[d]” concerning the hotly disputed political question of how
many representatives in Parliament is one Member State entitled.’44 This should
explicitly be seen as encroaching on EU law autonomy. The co-respondent mecha-
nism could however minimize this chance, as the EU and UK could have narrowed
this space through the friendly settlement mechanism set forth in the DAA (although
not exclude the possibility for such interference in this example).

9.2.3 Testing Kokkelvisserij with the Co-respondent


Mechanism

In Kokkelvisserij case, the applicant association, Cooperatieve Producentenor-


ganisatie van de Nederlandse Kokkelvisserij U.A., had filed an application against

38
ECtHR, Matthews v. United Kingdom, Application no. 24833/94, D. 18 February 1999. (Judge-
ment, Grand Chamber), note 22, para. 41.
39
Cf.: Cardwell et al. (2009), p. 240, wherein it is argued that in Kadi the EU Court insisted that
international law is inferior to the EU law, something that might be the next debate not anymore
with the UN law but with the Convention law after EU accession to ECHR.
40
See: Licková (2008), p. 480.
41
For the nature of the European Parliament as a ‘legislature’ for Gibraltar as well, see: O’Boyle
(2008), p. 2.
42
Rieder (2005), p. 80 (‘Because the European Convention is a living instrument it is of little
concern that the body of the European Parliament was not explicitly envisaged by the drafters of
the Convention.’) [emphasis added].
43
On the behavior of Strasbourg Court to ‘jump’ on issues which touch upon national legal orders,
see also: Letsas (2004), p. 279/80.
44
Rieder (2005), p. 83.
368 9 Testing the Co-respondent Mechanism from the Strasbourg Court’s. . .

the Netherlands in relation to the EU law compliance with the Convention rights.
Originally, the Netherlands Deputy Minister of Agriculture, Nature Conservation
and Fisheries had approved a licence to the applicant association enabling its
members to fish cockle meat from the Wadden Sea for a certain period of time.45
The Wadden Sea Society had then rejected the Deputy Minister’s decision to entitle
Kokkelvisserij that amount of fishing permission.46 The Wadden Sea Society filed
that decision to the Council of State, the respective court of Netherlands for
administrative judicial review cases. Since the application of Article 6 § 3 of
Council Directive 92/43/EEC of 21 May 1992 was to be considered within the
context of the legality of the permitted amount of fishing in the Wadden Sea, the
Netherlands Council of State referred a preliminary question to the EU Court for the
meaning of terms ‘plan or project’.47 The EU Court involved an Advocate General
to make the assessment, and the latter presented his arguments before the Court.
The applicant association in a letter sent to the EU Court on 1 February 2004 sought
permission to submit written response to the Advocate General’s opinion, as a
means of presenting its own arguments within the context of the rights of equality of
arms and fair access to court. In making such a request, the applicant association
also requested that the procedure before the EU Court be reopened as a means for
the Advocate General’s opinion to be revisited in the context of an adversarial
procedural examination before it. The EU Court rejected the applicant association’s
request to reopen the proceedings on the basis of Art. 59 (2) of the Rules of
Procedure of the EU Court,48 and further provided the Netherlands Council
of State with the requested answer on the interpretation of Article 6 (3) of
Council Directive 92/43/EEC.49 Based on the preliminary reference answer, the
Netherlands Council of State then took the parties opinion on the EU Court’s
answer. The applicant association contended that the EU Court’s procedure of
preliminary reference violated its rights under the Convention. However, the
Council of State dismissed that objection and also revoked the licenses provided
to it for fishing in the Wadden Sea.50
The question of the origin of the legal basis of the alleged violation—the original
alleged violation for which the applicant association filed the application to the
Council of State of Netherlands concerned the interpretation of the terms ‘plan or
project’ within the context of Article 6 (3) of Council Directive 92/43/EEC. For the

45
ECtHR, Kokkelvisserij U.A. v Netherlands, Application no 13645/05, 20 January 2009, para. 2.
46
ECtHR, Kokkelvisserij U.A. v Netherlands, Application no 13645/05, 20 January 2009, note
45, para 2.
47
ECtHR, Kokkelvisserij U.A. v Netherlands, Application no 13645/05, 20 January 2009, note
45, para. 3.
48
Cf.: Court of Justice of EU, Emesa Sugar, judgement, Case C-17/98 [2000] ECR I-665, para. 5 et
seq.
49
ECtHR, Kokkelvisserij U.A. v Netherlands, Application no 13645/05, 20 January 2009, note
45, para. 4.
50
ECtHR, Kokkelvisserij U.A. v Netherlands, Application no 13645/05, 20 January 2009, note
45, para. 5.
9.2 A Starting Note to the Three Tests 369

same point of interpretation, the Council of State of Netherlands had referred a


preliminary question to the EU Court. However, although the applicant association
had first claimed to have suffered a harm due to the Council Directive concerned,
once the latter became interpreted by the EU Court through the preliminary
reference procedure, the applicant association claimed that it had also suffered
from a violation of the right to adversarial proceedings, as during the preliminary
reference procedure the EU Court did not give the right to reply to the Advocate
General’s report51 and therefore refused its claim to reopen the preliminary refer-
ence proceedings. Hence, the applicant association did not in fact attack the Council
Directive before the Strasbourg Court but rather the refusal of the EU Court to grant
the applicant association the right to respond to the Advocate General’s report.52
Therefore, the EU law provision which the applicant association challenged at the
Strasbourg Court was not the substantive issue of interest stemming from the
Council Directive on which the case had started before the national court but rather
the rejection of the EU Court to its claim to reply to the Advocate General. This
being the case, the contested EU law provision to assess their compliance with the
Convention are the legal acts which establish the organization and functioning of
the EU Court procedures, namely the Statute of the Court of Justice, which is now a
Treaty-level protocol, and the Rules of Procedure of the now Court of Justice of EU
(in the original case, of ECJ), which is an internal act of the EU Court. In practice,
these two legal acts would have to be assessed for their compliance with the
Convention at Strasbourg—which the applicant association alleges to have violated
its right under Art. 6 of the Convention—adding that in the application of those two
acts the Netherlands had taken no implementing measure and in effect had abso-
lutely no legal and factual authority for deciding whether the applicant association
may reply to the Advocate General’s report within the context of the observations
of parties submitted under a preliminary reference procedure, now regulated by Art.
267 TFEU.
Are basic conditions for the co-respondent mechanism fulfilled by the EU under
Kokkelvisserij—One should examine this if Kokkelvisserij was to arrive before the
Strasbourg Court once the EU has acceded to ECHR. In the real scenario, although
the claims of the applicant association directly challenge an act of the EU wherein
the Member States have had absolutely no involvement, they would have arrived at
the Strasbourg Court by exhausting the domestic remedies of the Netherlands. In

51
It is argued that the Strasbourg’s case-law has a far higher standard compared to Luxembourg’s
fundamental rights regime when it comes to the right to reply to AGs. See: Lenaerts and de Smijter
(2001), p. 95.
52
Cf.: ECtHR, Kokkelvisserij U.A. v Netherlands, Application no 13645/05, 20 January 2009, note
45, Part A, wherein the applicant association argued its right to adversarial proceedings including
AG’s reports this in line with the following assertion: ‘In Mantovanelli v. France (18 March 1997,
§ 36, Reports of Judgments and Decisions 1997-II) and Krčm arˇ and Others v. the Czech Republic
(no. 35376/97, § 45, 3 March 2000), violations of Article 6 § 1 had been found on the ground that
the applicants had not had the opportunity to comment on documentary evidence.’ This argument
seems rather grounded and suitable for application in Kokkelvisserij.
370 9 Testing the Co-respondent Mechanism from the Strasbourg Court’s. . .

doing so, the applicant association would factually claim that the original contested
provision is the Council Directive, which, taken together with the rejection of the
EU Court to give the applicant association the right to reply to the Advocate
General’s report, would constitute a breach of Art. 6 of the Convention. Therefore,
in that scenario, the applicant association would have been given the chance to
exhaust the domestic legal remedies for another EU law provision, and then to
submit the contested rejection of the EU Court subsidiary to the issue for which the
domestic legal remedies were exhausted, and which in fact resulted in the applicant
association becoming a victim. Therefore, to arrive before the Strasbourg Court, in
this real scenario, the applicant association would have to challenge a measure of
the Netherlands, which factually does not relate directly to the contested EU law
provision. The applicant association would therefore, in this real scenario, chal-
lenge the omission of the Netherlands judicial system to offer the protection of Art.
6 ECHR while judging whether the applicant association suffered from the misin-
terpretation of Article 6 (3) of Council Directive 92/43/EEC. In a second scenario,
the applicant association could have used an action for annulment (Art.
263 (4) TFEU) before the EU Court to directly attack the act of the EU law
which the applicant considers as violating its right to adversarial proceedings
under the Convention. Such two acts which would be attacked by an action for
annulment are the Statute of CJEU or the Rules of Procedure of CJEU, or both
of them.
However, in order to specify how these two scenarios would lead the legal
situation of the applicant association—the real and the supposed scenario—one
needs to evaluate once more the nature of the EU acts which would potentially
become the contested EU law provisions before the Strasbourg Court.
In assessing the claim of the applicant association during the preliminary
reference proceedings, the EU Court refused its request to reply to the Advocate
General, by ruling that:
6. It should be pointed out that the Statute of the Court of Justice and the Court’s Rules of
Procedure make no provision for parties to submit observations in reply to the opinion of
the Advocate General (see the decision of 4 February 2000 in Case C-17/98 Emesa Sugar
[2000] ECR (European Court Reports) I-665, point 2). This circumstance, however, does
not violate a party’s right to adversarial proceedings flowing from Article 6 § 1 of the
Convention as construed by the European Court of Human Rights (see the above-
mentioned Emesa Sugar decision, points 3–16).53

The EU Court therefore explicitly asserted that if the fact why the applicant
association is not allowed to reply to the Advocate General’s report is that there is
no room within the Statute of the EU Court that would allow claimants to rely
thereupon. However, the EU Court simultaneously interpreted the concerned Stat-
ute’s lack of the specific provision as not violating Art. 6 of the Convention. It must
be noted here that the Statute of the EU Court is a segment of EU primary law,

53
ECtHR, Kokkelvisserij U.A. v Netherlands, Application no 13645/05, 20 January 2009, note
45, para. 4.
9.2 A Starting Note to the Three Tests 371

namely the Treaties. Hence, if the applicant association was to claim that the
contested EU law provision violating its right to adversarial proceedings is a Treaty
omission, then the EU Court would have no chance of reviewing the legality of it
but merely providing an interpretation of the Statute, as it in fact did. However,
there is another chance to base the alleged violation within the EU law, that of
considering as if the violation stems from the Rules of Procedure of the European
Court of Justice, which in Article 59 read:
1. The Advocate General shall deliver his opinion orally at the end of the oral procedure.
2. After the Advocate General has delivered his opinion, the President shall declare the oral
procedure closed.54

The Rules of Procedure of ECJ also made it clear that there was no room for the
applicant association to reply to the Advocate General’s report,55 as the latter report
is given at the very end of the oral procedure, thereby claimants may not no longer
provide observations on that procedure. Essentially, if this alleged violation would
have been viewed as well founded, one would have to consider that it is an omission
of the Treaties and the Rules of procedure to guarantee Art. 6 of the Convention
where parties provide observations in a preliminary reference procedure.
That said, in the scenario of Kokkelvisserij (scenario no. 1), the applicant
association would have attacked before the Strasbourg Court both the Statute of
the EU Court and the Rules of Procedures of the latter, having utilized a case
exhausted before the national courts. In this scenario, the applicant association
would have attacked the Netherlands before the Strasbourg Court for having failed
to ensure the full respect of Art. 6 of the Convention when reviewing the applicant’s
case under Council Directive 92/43/EEC. That being the case, the original respon-
dent would have been the Netherlands, although it did not cause the alleged
violation with any implementing measure of EU law. Then, on the basis of Art.
3 (2) DAA, the EU would be interested to become a co-respondent as its law is
being called into question for its compliance with the Convention.
A contrary scenario (scenario no. 2) would have been that wherein the applicant
association used a remedy to directly attack the omission deriving from the Statute
and Rules of Procedure before the EU Court. In such a scenario, the applicant
association would have needed to bring the Statute of the Court before the EU
Court, which is a segment of the Treaties, therefore impossible to be attacked in any
way (including the action for annulment, which can be used to attack merely acts of
secondary law). Hence, a direct action would not exist against the Statute of the EU
Court. However, since the Rules of Procedure of the EU Court are neither a part of
the primary law nor constitute a legislative act, then the applicant association could
have attacked those rules directly through the action for annulment, using the

54
ECJ’s Rules of Procedure (Official Journal of the European Communities) L 176 of 4 July
1991, p. 7.
55
See also: Court of Justice of EU, Emesa Sugar, judgement, Case C-17/98 [2000] ECR I-665,
para. 5 et seq.
372 9 Testing the Co-respondent Mechanism from the Strasbourg Court’s. . .

Microban56 scenario. As the Rules of Procedure of the EU Court constitute a


non-legislation act entailing no implementing measures from the Member States,
they constitute a regulatory act under Art. 263 (4) which can be attacked by an
action for annulment by a private party if they prove that such a regulatory act is of
direct concern to it; the direct concern being proved if that regulatory act automat-
ically and mandatorily modifies the legal position of the applicant in a certain
way.57 Therefore, Kokkelvisserij could have utilized Art. 263 (4) TFEU to attack
the EU the Rules of Procedure before the Court of Justice, claiming that they omit
to provide to the applicant a fully-fledged right to fair trial including the right to
reply to the Advocate General’s report, while proving that such omission in those
rules makes that regulatory act illegal. The EU Court could have then returned
again to the argument that it is the Treaties which do not permit the Court of Justice
to extend the right to adversarial proceedings also in the context of the Advocate
General’s report. With the EU Court decision in place, than the applicant associa-
tion would have exhausted remedies in the EU to be able to file the application
before the Strasbourg Court. When Kokkelvisserij would have reached the Stras-
bourg Court under this scenario, the original respondent would have been the EU,
whereas the Member States (including the Netherlands) could not have joined as
co-respondents on the basis of Art. 3 (2) DAA but on basis of Art. 3 (3) DAA.
However, the Member State(s) could have joined as co-respondents only if the
Statute of CJEU as part of primary law would have been considered as the legal
basis for not permitting the Rules of Procedure of CJEU (the regulatory act) to
extend the right of applicants to adversarial proceedings against the AG’s report. If
the EU primary law was not the attacked EU law provision for that alleged violation
of Kokkelvisserij in this scenario, the Member States would have not fulfilled the
basic criteria of Art. 3 (3) DAA to join as co-respondents in the proceedings before
the Strasbourg Court.
Whether the alleged violation could have been avoided simply by disregarding an
obligation under EU law—A dissimilar examination, as seen from the above, needs to
be made for the two scenarios. In scenario no. 1 of Kokkelvisserij, it is rather complex
and problematic to assess whether the Netherlands—which took no measure in the
direction of the applicant association position in the preliminary reference proceedings
in the EU Court—could have avoided the alleged violation without breaching the EU
law. Although the Netherlands had taken no measure in that regard, the applicant
association’s claim was to be channeled through its basic claim the Netherlands
violated its rights by not applying Council Directive 92/43/EEC in line with the
Convention’s Art. 6. Such an awkward argument would serve as a bypass claim
whereupon the applicant association would base the second, more substantial claim

56
Court of Justice, Microban International and Microban (Europe) v Commission, Case T-262/10
[2011] ECR II-0000, para. 26 et seq.; See supra, the chapter on the exhaustion of direct remedies
before the EU Court.
57
See supra, the chapter on the exhaustion of direct remedies before the EU Court, and the section
on the regulatory acts.
9.2 A Starting Note to the Three Tests 373

for the EU Court failing to respect its right to an adversarial proceeding against the
AG’s report. If the Strasbourg Court would accept Kokkelvisserij as admissible on basis
of this premise—as there seems no other way for it to become admissible in Stras-
bourg—then the one would argue that such an alleged violating ‘measure/omission’ of
the Netherlands may have not been remedied without the latter breaching an obligation
under EU law: the obligation of the Netherlands to respect the EU Court’s jurisdiction
as by the definition of the Treaties. Hence, the requirement of Art. 3 (2) DAA to prove
the linkage between the alleged violation and the obligation that the Netherlands had
under EU law would seem to be fulfilled for the EU to join as co-respondent. In
scenario no. 2 of Kokkelvisserij, the Member States would have needed to prove that
the EU Court could have only provided the applicant association with the right to
adversarial proceedings as regards the AG’s report if it was to breach the Treaty
provisions on its jurisdictional authorizations. It is clear that, in this scenario, that
linkage would have been clearly demonstrated by the Member States, as it is relatively
easy to prove that the EU Court may not introduce a new provision which has not been
written or at least meant in the Statute of the CJEU (the latter being part of the Treaties)
without breaching the Treaties themselves. Hence, even in this scenario, Member State
(s) would have passed the test of Art. 3 (3) DAA and would gain the co-respondent
status in the Kokkelvisserij proceedings before the Strasbourg Court. In Kokkelvisserij
the possibility exists for the applicant association to have exhausted the legal remedies
both in Netherlands and directly in the EU—in Netherlands through contesting the
interpretation of its right to fish in the Wadden Sea in the context of Council Directive
92/43/EEC, whereas in the EU through contesting the validity of the Rules of Proce-
dure of CJEU as regards the adversarial proceedings against AG’s reports—and in that
case Art. 3 (4) DAA would allow for the applicant association to attack both the EU and
Netherlands before the Strasbourg Court as joint and original respondents jointly
notified by the Court for the alleged violation.
Who would become liable if the Court would decide that Kokkelvisserij violates
the Convention, and would this amount to an interpretation of the EU law by the
Strasbourg Court—The examination of the liability issue would again be very
complex in Kokkelvisserij. In both scenario no. 1 and scenario no. 2, the EU
would be singularly liable if the contested provision is merely one stemming
from the Rules of Procedure of CJEU. However, were the applicant association
and the Strasbourg Court to consider that the Statute of CJEU is the main legal basis
for the violation, not merely the Rules of Procedure of CJEU, then it would need for
both the EU and the Member States as the Masters of the Treaties of the EU to
become jointly liable: the first to change the Rules of Procedure of CJEU after the
Statute has given room for adversarial proceedings in AG’s reports; the second to
revise the Treaties (Statute of CJEU) to make room for adversarial proceedings in
AG’s reports. This internal liability, even in the first scenario would have to be
agreed upon by the EU and Member State(s) through a friendly settlement, as the
Strasbourg Court would find in both scenarios the EU and Member State(s) jointly
responsible, even though if in scenario no. 1 the alleged violation would have been
proven as merely stemming from the Rules of Procedure of CJEU, it would be
inaccurate to find the Member States liable for that act. If both co-respondent and
374 9 Testing the Co-respondent Mechanism from the Strasbourg Court’s. . .

the original respondent would agree on the basis of Art. 3 (7) DAA, then the
Strasbourg Court could have also ‘on the basis of the reasons given by the
respondent and the co-respondent, and having sought the views of the applicant,
decide that only one of them be held responsible.’ With regard to whether
Kokkelvisserij leaves room for the interpretation of the EU law by the Strasbourg
Court (consuetudo pro lege servatur seems to apply even to EU cases), it must be
said that if one of the abovementioned scenarios would appear, with the Statute of
the EU Court being a contested act as well, the Strasbourg Court would be given the
chance to interpret that Statute and to extend its scope by indirectly finding that the
EU violated the right to adversarial proceedings of the applicant company by not
providing what in fact the Treaties do not prescribe in writing.58 Should that appear,
the Strasbourg Court would be seen as effectively interpreting the spirit of the
Treaties, and forcing the EU Court to transpose that model of interpretation to it
Rules of Procedure; that certainly amounting to a doorway for Strasbourg to
impinge on EU law autonomy. To briefly digress, ICJ’s Judge Gaja reasonably
questions ‘how can one expect an external control over compliance by the EU with
its obligations under the Convention if the European Court was prevented from
interpreting the content of provisions of EU law? The concern that matters relating
to EU law should be left to be resolved by the EU and its member States cannot be
carried too far.’59 The author of this book agrees with the latter argument, adding
that it would be premature to take it for granted that the Strasbourg Court never
engages with the interpretation of municipal law when assessing the human rights
violations. If that would have been an issue, it would have been impossible for the
Strasbourg Court to locate the responsibility without looking into—and accordingly
interpreting to the degree needed (Fig. 9.1)—a certain municipal norm. Therefore
one needs to disagree with Gragl who argues that the co-respondent mechanism is
‘the perfect instrument in preventing potential pitfalls for the Union’s legal auton-
omy in the form of external interferences by Strasbourg.’60 Kokkelvisserij pushes
forward a strong counter-argument to Gragl, as analyzed above.

58
E.g.: The Strasbourg Court normally uses to consider the interpretation of the Convention as
‘autonomous’ from national legal order terms, thereby establishing its right to make its own
consideration of national law when it is assessed its compliance with the Convention. See: Letsas
(2004), p. 282/3; Contra: Gragl (2013), p. 156, who argues that the co-respondent mechanism
allows no room for the Strasbourg Court to intervene into EU internal law and its autonomy.
59
Gaja (2013), p. 5.
60
Gragl (2013), p. 157.
9.3 Chapter’s Summary of Conclusions 375

Fig. 9.1 Figurative


explanation of the potential
backdoor access of
Strasbourg Court to
Luxembourg Court tested
with the three examples

9.3 Chapter’s Summary of Conclusions

The chapter offered a technically-oriented examination against the mechanics of


the co-respondent mechanism of the three landmark cases that implicate joint
responsibility of the EU and its Member States, namely Bosphorus, Mathews, and
Kokkelvisserij. This chapter was devoted to testing these cases if they were applied
in a post-accession scenario, with the co-respondent mechanism utilized by the
relevant parties. Testing the co-respondent mechanism against these three cases
proved not only several ramifications that exist with regard to its functionality,
nature of legal effect and reach, but also with regard to the latent gaps that may
permit Strasbourg’s access to the EU law autonomous territory. The chapter
concludes that the test on the functionality and applicability of the co-respondent
mechanism shows that the latter will obviously offer better support to the protection
of EU law autonomy, however, it still leaves many gaps open and several blurred
practices that seem to have been omitted intentionally.
The chapter first examined and practically tested the identification of the legal
origin of the alleged violation, the latter being one of the criteria to utilize the
co-respondent mechanism. The chapter concludes that identifying the legal origin
of the alleged violation is often followed by a certain level of discretion left to the
Strasbourg Court—something that may be either abused or prove vagueness to the
interface between the laws of the two regimes. The chapter then tested the criterion
on whether the obligation could have been avoided simply by avoiding an obliga-
tion under the relevant law, therefore identifying the modes through which this test
should be tracked technically while proving the level of uncertainty that may
subsist in this regard. This is especially blurred with regard to omissions. The
376 9 Testing the Co-respondent Mechanism from the Strasbourg Court’s. . .

chapter concludes that the criterion on the nature of the obligation that led to the
alleged violation seems not so inflexible to be applied in practice, but it still leaves
certain discretion to the Strasbourg Court to select among some ways which it may
utilize to perform this test. This practical indistinctness does per se seem with
potential to interfere to the EU law autonomy. Then, the chapter tested the issue of
attribution of liability against the co-respondent mechanism. The chapter concludes
that although there is still possibility for the Strasbourg Court to interpret EU law
while attributing the responsibility on it and its Member States, the co-respondent
mechanism does critically recover the identified gaps in this regard. To digress with
this conclusion, referring to Connolly, Kokkelvisseri and MSS cases, Gragl gener-
ally argues that the Strasbourg Court may sometimes need to engage with the
domestic law provisions of a high contracting party in order to find an answer for
the alleged violation of the Convention. Gragl argues that this may be especially the
case when the Convention itself refers to domestic law.61 Gragl’s point in this
regard is valid, and further supports the findings of this chapter.
The chapter finally concludes that—at the technical level—the application of the
co-respondent mechanism may seem a complex job to the Strasbourg Court,
especially if there appear cases involving omissions. This tough job may turn into
room for aversion from the Luxembourg Court; however, it is still argued that this
mechanism will improve the previously identified loopholes with regard to the
protection of EU law autonomy in a pre-accession scenario. Such enhancement
which the co-respondent mechanism will bring in practice will likely make the
Luxembourg Court more comfortable in its reception of Convention law tone, while
also institutionalizing the cooperation and competition between the two courts.

References

Albi A (2009) Ironies in human rights protection in the EU: pre-accession conditionality and post-
accession conundrums. Eur Law J 15(1):46–69
Callewaert J (2007) The European Court of Human Rights and the area of freedom, security and
justice. ERA Forum 8(4):511–518
Cardwell PJ, French D, White N (2009) I. European Court of Justice, Yassin Abdullah Kadi and Al
Barakaat International Foundation v Council and Commission (JOINED CASES C-402/05 P
AND C-415/05 P) Judgment of 3 September 2008. Int Comp Law Q 58(1):229–240
De Londras F, Kingston S (2010) Rights, security, and conflicting international obligations:
exploring inter-jurisdictional judicial dialogues in Europe. Am J Comp Law 58:359–414
Defeis E (2007–2008) Dual system of human rights: the European Union. ILSA J Int Comp Law
14:1–8
Douglas-Scott S (2006) ‘Bosphorus Hava Yollari Turizm Ve Ticaret Anonim Sirketi v. Ireland,
(2006)’ Casenote. Common Market Law Rev 43(1):243–254
Gaja G (2013) The “Co-Respondent Mechanism” according to the draft agreement for the
accession of the EU to the ECHR. ESIL Reflect 2(1):1–6

61
Gragl (2013), p. 121.
References 377

Garcı́a IG (2009) Gibraltar and the European Parliament elections before the European Court of
Justice. Bull Transilvania Univ Braşov 2(51):195–202
Govaere I (2009) The importance of international developments in the case-law of the European
Court of Justice: Kadi and the autonomy of the EC legal order. Research Papers in Law 1/2009,
College of Europe
Gragl P (2013) The accession of the European Union to the European Convention on Human
Rights. Hart Publishing, Oxford
Harding C (2000) The identity of European law: mapping out the European legal space. Eur Law J
6(2):128–147
Hoffmeister F (2006) International decisions: Bosphorus Hava Yollari Turizm Ve Ticaret Anonim
Sirket v. Ireland. App. No. 45036/98. Am J Int Law 100:442–449
Kokott J, Sobotta C (2012) The Kadi case – constitutional core values and international law –
finding the balance? Eur J Int Law 23(4):1015–1024
Larsen KM (2008) Attribution of conduct in peace operations: the ‘Ultimate Authority and
Control’ test. Eur J Int Law 19(3):509–531
Lenaerts K, de Smijter E (2001) The Charter and the role of the European Courts. Maastricht J Eur
Comp Law 8(1):90–101
Letsas G (2004) The truth in autonomous concepts: how to interpret the ECHR. Eur J Int Law 15
(2):279–305
Licková M (2008) European exceptionalism in international law. Eur J Int Law 19(3):463–490
Miiller CA (2007) Fundamental rights in multi-level legal systems: recent developments in
European human rights practice. Interdiscip J Hum Rights Law 2(1):33–60
O’Boyle M (2008) Electoral disputes and the ECHR: an overview, Strasbourg, 12 November 2008,
CDL-UD(2008)010
Papangelopoulou E (2006) Introductory note to European Court of Human Rights (ECHR) Grand
Chamber: Bosphorus Hava Yollari Turizm Ve Ticaret Anonim Sirketi v. Ireland. Int Leg Mater
45(1):133–135
Peers S (2006) Bosphorus – European Court of Human Rights. Eur Constitut Law Rev 2
(3):443–455
Phelps J (2006) Comment: reflections on bosphorus and human rights in Europe. Tulane Law Rev
81:251–279
Rieder C (2005) Protecting human rights within the European Union: who is better qualified to do
the job-the European Court of Justice or the European Court of Human Rights? Tulane Eur
Civil Law Forum 20:73–107
Schimmelfennig F (2006) Competition and community: constitutional courts, rhetorical action,
and the institutionalization of human rights in the European Union. J Eur Public Policy 13
(8):1247–1264
Škugor EM (2012) European Union and the European Convention on Human Rights – the
accession of a giant. Pravnik 46(1):105–118
Tulkens F (2013) EU accession to the European Convention on Human Rights. Speech at: National
School of Judiciary and Public Prosecution (KSSIP)—Krakow (Poland), Friday 1 March 2013
van den Berghe F (2010) The EU and issues of human rights protection: same solutions to more
acute problems? Eur Law J 16(2):112–157
Vranes E (2003) The final clauses of the Charter of fundamental rights – stumbling blocks for the
first and second convention. European Integration online Papers (EIoP) 7 (2003) N 7. Avail-
able at: http://eiop.or.at/eiop/texte/2003-0007a.htm
Chapter 10
Admissibility Before the Strasbourg Court:
An Outlook on the EU-Law-Originated
Applications

10.1 Introduction

This chapter offers an examination of the procedure that Strasbourg Court will
pursue once an EU-related application reaches its jurisdictional gates in a post-
accession scenario, contextualized against a background of exceptional rules which
Strasbourg may apply on an exceptional contracting party like the EU. The chapter
responds to the research sub-question: ‘How should Strasbourg Court perform its
admissibility jurisdiction on Luxembourg/EU-related cases, with an eye also on the
exceptional admissibility standards? How may these decisions of Strasbourg Court
look into EU legal order?’ The chapter watchfully examines Strasbourg’s admissi-
bility standards applied on EU-related cases, playing with the flexibility clauses
which the former may apply on EU-originated cases, while offering likely perspec-
tives on such scenarios. Such flexibility standards are looked from the perspective
of Strasbourg Court’s likely tendencies to become a court of first instance in cases
when there are extremely difficult circumstances for the EU applicant to become
reviewed by the Luxembourg Court first. The EU–Member States problems in
relation to the parallel systems of available remedies are also considered in light
of the Strasbourg’s admissibility rules. The chapter also addresses whether and how
EU primary law violations may be filed at Strasbourg, in addition to a special
analysis on the latter’s decisions nature of effect within EU legal order. The chapter
concludes with the remark that Strasbourg Court may undoubtedly pass over some
EU-law-autonomy safeguards and find cases admissible using numerous flexibility
standards, which may definitely appear as interfering to EU law autonomy. The
internal reception of Strasbourg decisions in EU law follows the same logic.

© Springer International Publishing Switzerland 2015 379


F. Korenica, The EU Accession to the ECHR, DOI 10.1007/978-3-319-21759-8_10
380 10 Admissibility Before the Strasbourg Court: An Outlook on the EU-Law. . .

10.2 A Notional Start: What Does One Need to Do to Reach


the Strasbourg Court?

As stated above, in order for an EU-related application to reach Strasbourg Court,


the claimant should have exhausted all domestic legal remedies.1 However, there
remain a set of other locus standi and ratione materie issues that need be fulfilled
before the claimant fulfils the criteria to get heard at Strasbourg. From the perspec-
tive of the Convention, it is established in Art. 34 ECHR2 that:
The Court may receive applications from any person, non-governmental organisation or
group of individuals claiming to be the victim of a violation by one of the High Contracting
Parties of the rights set forth in the Convention or the Protocols thereto. The High
Contracting Parties undertake not to hinder in any way the effective exercise of this right.

Art. 34 ECHR therefore allows all possible legal or natural persons to challenge
a national measure before the Strasbourg Court, on the condition that they be the
victim of an alleged violation by one of the High Contracting Parties. The Stras-
bourg Court has made it clear in Mamatkulov that the individual complaint mech-
anism provided for in Art. 34 ECHR is made to serve the ‘fundamental guarantees
of the effectiveness of the Convention system of human rights protection [. . .]’3
adding that the Convention should be seen ‘as a treaty for the collective enforce-
ment of human rights.’4 The application of Art. 34 ECHR, and the effective
possibility of claimants to utilize it, including here the claimants from the EU, is
profoundly the very rationale and effectiveness of the Convention system. That
noted, it was stated in Van Der Tang that it would suffice for the claimant to claim
that the act or omission has been caused by one of the High Contracting Parties,
such as the EU, one of its Member States, or the EU and its Member State(s) jointly,
for it to have the right to access the Strasbourg Court.5
As explained supra in the inter-party chapter of this book, one may steadily
argue that the inter-party application seems to allow room for abstract claims
(similar to an actio popularis). However, the Strasbourg Court has made it clear
in Klass that the individual complaint mechanism may not serve as an actio
popularis for any claimant, unless they are a victim of a violation. Individual
complaint mechanism therefore ‘does not permit individuals to complain against

1
Cf.: Conforti (2010), p. 83, who argues that if the Court would continue to uphold the principle of
equivalent protection, then accession would make no sense in practice.
2
On these conditions, see also: Ribble (2010–2011), p. 214.
3
ECtHR, Mamatkulov and Askarov v. Turkey [GC]—46827/99, Judgment 4.2.2005, para. 100.
4
ECtHR, Mamatkulov and Askarov v. Turkey [GC]—46827/99, op. cit. note 3, para. 100.
5
ECtHR, Van Der Tang v. Spain, Application no. 19382/92, Judgment, Strasbourg, 13 July 1995,
para. 53.
10.2 A Notional Start: What Does One Need to Do to Reach the Strasbourg Court? 381

a law in abstracto simply because they feel that it contravenes the Convention.’6
The same was reiterated in Burden wherein the Court ruled that the need for the
claimant to be a victim must mean that the claimant has been ‘directly affected by
the impugned measure’ which is being challenged by utilizing Art. 34 ECHR.7
EU-related applications would have to testify this either at the level of Member
States’ measures originating from EU law or merely at EU-direct measures or
omissions originating from EU law. Two issues therefore remain important to be
assessed in this regard: first, the victim status under Art. 34 ECHR and, second, the
possible exceptions on the rule of exhaustion of legal remedies.

10.2.1 The Victim Status of the Claimant

Strasbourg Court has established that the notion of victim is something that needs
be interpreted ‘autonomously and irrespective of domestic concepts such as those
concerning an interest or capacity to act.’8 It is for that reason absolutely irrelevant
how the EU legal order would consider the concept of victim, thereby making it
very irrelevant the nature of the domestic considerations on whether there has been
a violation of the Convention or not. In order for one to be considered a victim, as
stated in Gorraiz Lizarraga, ‘there must be a sufficiently direct link between the
applicant and the harm which they consider they have sustained on account of the
alleged violation.’9 The need to prove this link would automatically need to
associate the claimant with the act or omission which the claimant challenges,
adding here that such linkage may not be derogated as the entire victim status would
then be considered lost. The Court does apply this concept rather substantively,
aiming to observe whether in practice the claimant is a victim of a violation, rather
than applying excessive bureaucratic tests.10 The Court however normally requests
that the claimant be a direct victim of the EU law,11 respectively that the claimant is

6
ECtHR, Klass and Others v. Germany, Application no. 5029/71, JUDGMENT, STRASBOURG,
6 September 1978, para. 33.
7
ECtHR, Burden v. The United Kingdom, Application no. 13378/05, JUDGMENT, STRAS-
BOURG, 29 April 2008, para. 33.
8
ECtHR, Gorraiz Lizarraga and Others v. Spain, Application no. 62543/00, Judgment, Stras-
bourg, 27 April 2004, para. 35.
9
ECtHR, Gorraiz Lizarraga and Others, op. cit. note 8, para. 35; See also: ECtHR, Tauira and
Others v. France, Application no. 28204/95, Commission decision of 4 December 1995, Decisions
and Reports (DR) 83-B, p. 112.
10
ECtHR, Monnat v. Switzerland, Application no. 73604/01, Judgment, Strasbourg, 21 September
2006, para. 30 et seq.
11
A general argument, at: Raba (2013), p. 560.
382 10 Admissibility Before the Strasbourg Court: An Outlook on the EU-Law. . .

‘directly affected by the act or omission in issue, the existence of a violation of the
Convention being conceivable even in the absence of prejudice.’12 However, ‘a
decision or measure favourable to [. . .the claimant. . .] is not in principle sufficient
to deprive him of his status as a “victim” unless the national authorities have
acknowledged, either expressly or in substance, and then afforded redress for, the
breach of the Convention.’13 The Court has also ruled that this ‘criterion cannot be
applied in a rigid, mechanical and inflexible way throughout the whole proceed-
ings.’14 Compared to the direct concern test which is applied under the action for
annulment and action for failure to act in EU Court’s jurisdiction, one must say that
the ‘directly affected’ test which the Strasbourg Court applies for the notion of
victim is far more flexible and substantive in the result that it is looked from rather
than the mechanical standards which it must fulfill. The same can be said about the
individual concern.15 The Strasbourg Court does allow the bar to be passed even if
the measure or omission has a temporal and not a final effect, as set in Monnat
v. Switzerland,16 allowing also persons not directly but indirectly affected by a
measure or omission to be recognized as victims (as set in McCann).17 Interest-
ingly, as well, in Bowman, the Strasbourg Court allowed the claimant to challenge a
law of general application if it of direct adverse effect to him or her. This must be
seen in the context of the possible lack of domestic legal remedies which the
EU-related claimants may not raise against an act of general application, but
which may end up being considered for their compliance with the Convention
before the Strasbourg Court.18 All said, once an EU-related application is checked
on this basis, then it may proceed further before the Strasbourg Court.

12
ECtHR, Amuur v. France, Application no. 19776/92, Judgment, Strasbourg, 25 June 1996, para.
36; From a comparative perspective, the UN Human Rights Committee, in Disabled and
handicapped persons in Italy v. Italy, has ruled that in order to get access to the Committee, one
should claim ‘to be the victim of a violation by the State party concerned. It is not the task of the
Human Rights Committee [. . .] to review in abstracts national legislation as to its compliance with
obligations imposed by the Covenant. It is true that, in some circumstances, a domestic law may by
its mere existence directly violate the rights of individuals under the Covenant.’ (UN Human
Rights Committee, Disabled and handicapped persons in Italy v. Italy, Communication
No. 163/1984, U.N. Doc. CCPR/C/OP/2 at 47 (1990), para. 6.2); Being a victim of an abstract
norm—theoretically—is therefore possible; hence, EU general acts—such as regulations—may
directly make an applicant a victim of a human rights violation.
13
ECtHR, Amuur v. France, op. cit. note 12, para. 36; See also: ECtHR, L€ udi v. Switzerland,
judgment of 15 June 1992, Series A no. 238, p. 18, para. 34.
14
ECtHR, Karner v. Austria, Application no. 40016/98, Judgment, Strasbourg, 24 July 2003, para.
25.
15
Disappointed with this situation, see e.g.: Arnull (2003), p. 792.
16
ECtHR, Monnat v. Switzerland, Application no. 73604/01, op. cit. note 10, para. 33.
17
ECtHR, McCann and Others v. The United Kingdom, Application no. 18984/91, Judgment,
Strasbourg, 27 September 1995.
18
ECtHR, Bowman v. The United Kingdom, 141/1996/760/961, Judgment, Strasbourg,
19 February 1998, para. 47.
10.3 Exceptions to the Rule on Exhaustion of Domestic Legal Remedies 383

10.3 Exceptions to the Rule on Exhaustion of Domestic


Legal Remedies

It is important to note that the Convention’s system primary function is to serve as


subsidiary international mechanism of human rights protection.19 ‘The machinery
of complaint to the Court is thus subsidiary to national systems safeguarding human
rights. This subsidiary character is articulated in Articles 13 and 35 § 1 of the
Convention.’20 For this purpose, Art. 35.1 ECHR reads:
The Court may only deal with the matter after all domestic remedies have been exhausted,
according to the generally recognized rules of international law, and within a period of six
months from the date on which the final decision was taken.

The Court has ruled in Kudła v. Poland that the purpose of the rule on ‘exhaus-
tion of domestic remedies, is to afford the Contracting States the opportunity of
preventing or putting right the violations alleged against them before those allega-
tions are submitted to the Court.’21 The Court went further to rule in Takis
Demopoulos by establishing that the rule on exhaustion forms an indispensable
part of the Convention’s system of human rights protection, because ‘[s]tates are
dispensed from answering before an international body for their acts before they
have had an opportunity to put matters right through their own legal system and
those who wish to invoke the supervisory jurisdiction of the Court as concerns
complaints against a State are thus obliged to use first the remedies provided by the
national legal system.’22 This being the primary intention of the Convention system
itself, one may argue that there is no tendency from the side of the Strasbourg Court
to invoke instruments via which the Luxembourg Court’s autonomy is defied unless
there appear EU-circumstantial problems which would make the protection of
human rights ineffective and useless. Therefore, the recognized position of the
Strasbourg Court on the rule on exhaustion is one which normally requests that
cases be first given the chance to have the Luxembourg Court’s rule before they
embark on the Strasbourg jurisdictional review. However, it would not be

19
On the importance of subsidiarity in international law, see: Carozza (2003), p. 79; The Court has
ruled in Edwards v. United Kingdom that in this regard that in principle it ‘is not within the
province of the European Court to substitute its own assessment of the facts for that of the
domestic courts and, as a general rule, it is for these courts to assess the evidence before them.
The Court’s task is to ascertain whether the proceedings in their entirety, including the way in
which evidence was taken, were fair [. . .]’. ECtHR, Edwards v. United Kingdom, App. No. 13071/
87, 15 Eur. Ct. H.R. (ser. A) 417, 431 (1992), para. 34/5.
20
ECtHR, Kudła v. Poland, Application no. 30210/96, Judgment, Strasbourg, 26 October 2000,
para. 152; See also: Costa (2003), p. 456.
21
ECtHR, Kudła v. Poland, Application no. 30210/96, op. cit. note 20, para. 152; See also:
ECtHR, Selmouni v. France [GC], Application no. 25803/94, ECHR 1999-V.
22
ECtHR, Takis Demopoulos and Others v. Turkey, Grand Chamber Decision As To The
Admissibility of Application nos. 46113/99, 3843/02, 13751/02, 13466/03, 10200/04, 14163/04,
19993/04, 21819/04, para. 69; See also: ECtHR, Akdivar and Others v. Turkey, 16 September
1996, § 65, Reports 1996-IV.
384 10 Admissibility Before the Strasbourg Court: An Outlook on the EU-Law. . .

problematic for the Strasbourg Court if the national courts or authorities decide
themselves that they need not the use of domestic exhaustion of remedies as an
advantage to restore the possible human rights violations, as set in De Wilde.23 One
should also point in this regard the argument that by having the Convention
authority to check whether domestic remedies have been exhausted or not, the
Strasbourg Court ‘would not simply become involved in adjudicating on substan-
tive questions of human rights but would also decide whether the Court of Justice’s
jurisdiction is satisfactory from the point of view of ensuring that individuals can
exhaust their domestic remedies.’24 As the EU system of remedies is rather narrow
and too restrictive, one may say that the Strasbourg Court should keep the reference
to Art. 32 (1) DARIO which clearly sets that: ‘The responsible international
organization may not rely on its rules as justification for failure to comply with
its obligations [. . .],’ as a means to surpass the rule of exhaustion on EU-related
cases if the latter may seriously undermine Art. 6 & 13 ECHR. This latter concern
remains essential to the presentation of the below loopholes that currently exist.

10.3.1 Strasbourg’s Elastic Approach to the Rule


on Exhaustion: The EU System of Remedies May
Become Surpassed

Although Strasbourg Court and the Convention system as a whole have predomi-
nantly held the argument that they serve as a subsidiary human rights protection
space, therefore the national courts should first be given the chance to rule on the
individual cases of human rights protection, the former applies some exceptions
when it comes to the application of the rule on exhaustion of legal remedies. Since
Art. 35.1 ECHR establishes that the rule on exhaustion of domestic remedies is one
that must be interpreted in light of international law—as opposed to the domestic
law—the EU law is rather irrelevant on whether it considers a case exhausted
domestically or not. The Strasbourg Court has ruled in Ringeisen v. Austria, in
1971, a landmark decision, that ‘international courts have on various occasions held
that international law cannot be applied with the same regard for matters of form as
is sometimes necessary in the application of national law. [. . .Art. 35.1 ECHR. . .]
of the Convention refers expressly to the generally recognised rules of international
law.’25 This said, there is a clear position of the Strasbourg Court that the rule on
exhaustion may not be applied in a way which recognizes the formal criteria as

23
ECtHR, De Wilde, Ooms and Versyp (“Vagrancy”) v. Belgium, Application no. 2832/66; 2835/
66; 2899/66, Judgment, Strasbourg, 18 June 1971, para. 55.
24
Analysis (1997), p. 241.
25
ECtHR, Ringeisen v. Austria, Application no 2614/65, Judgment, Strasbourg, 16 July 1971,
para. 89; See also: ECtHR, Lawless v. Ireland, decision of 30th August 1958 on the admissibility
of Application No. 332/57, Yearbook of the Convention, Vol. 2, pp. 324–326.
10.3 Exceptions to the Rule on Exhaustion of Domestic Legal Remedies 385

supreme to the substantial position of the applicant and the case, respectively. ‘This
means amongst other things that [. . .the Court. . .] must take realistic account not
only of the existence of formal remedies in the legal system of the Contracting Party
concerned but also of the general legal and political context in which they operate
as well as the personal circumstances of the applicants.’26 There being no need to
consider the domestic law excessive formalism—such as e.g. in the EU direct
actions formal restrictions—it would be ‘therefore quite right in declaring in
various circumstances that there was a need for a certain flexibility in the applica-
tion of the rule.’27 All this is made with the mere intention to make the human rights
protection the higher aim, as opposed to the respect for formalistic procedures
which national courts might bring as against a substantive human rights law
review.28
With the human rights protection jurisdiction of the Strasbourg Court being
empowered continually, the latter went even further in 2009 Kozacio glu v. Turkey
case, wherein it explicitly ruled that ‘[. . .] Article 35 § 1 must be applied with some
degree of flexibility and without excessive formalism’.29 Further, the Court has
gone to establish that ‘the rule of exhaustion is neither absolute nor capable of being
applied automatically; in monitoring compliance with this rule, it is essential to
have regard to the circumstances of the individual case. This means amongst other
things that it must take realistic account not only of the existence of formal
remedies in the legal system of the Contracting Party concerned but also of the
context in which they operate as well as the personal circumstances of the appli-
cant.’30 Therefore, the question remains whether the direct actions allowed in the
EU system and the indirect action of preliminary reference suffice not to pass the
bar of the need to be exhausted under the above standard of Kozacio glu v. Turkey. It
must be argued that whenever the applicant in the EU is in need to pass the tests of
individual and direct concern—meaning in cases of actions of annulment and
actions for failure to act, when the act or omission is not formally addressed to
them—one would argue that such tests would render the application of the rule on

26
ECtHR, Akdivar and Others v. Turkey, op. cit. note 22, para. 69.
27
ECtHR, Ringeisen v. Austria, op. cit. note 25, para. 89; See also: ECtHR, Lawless v. Ireland,
Application No. 332/57, op. cit. note 25, pp. 324–326; In addition, it is important to mention that
Art. 46.2.a ACHR, rather similarly, installs an identical standard prescribing that the rule on
exhaustion does not apply if ‘the domestic legislation of the state concerned does not afford due
process of law for the protection of the right or rights that have allegedly been violated.’ The EU
system of remedies may easily fit with this standard’s test.
28
The Court applied a similar argument in D.H. and Others v. The Czech Republic, wherein it
ruled that ‘[i]n these circumstances, it considers that it would be unduly formalistic to require the
applicants to exercise a remedy which even the highest court of the country concerned had not
obliged them to use.’ See: ECtHR, D.H. and Others v. The Czech Republic, Application no. 57325/
00, Judgment, Strasbourg, 13 November 2007, para. 118.
29
ECtHR, Kozacio glu v. Turkey, Application no. 2334/03, Judgment, Strasbourg, 19 February
_
2009, para. 40; See also: ECtHR, Ilhan v. Turkey [GC], no. 22277/93, § 59, ECHR 2000-VII.
30
ECtHR, Kozacio glu v. Turkey, op. cit. note 29, para. 40 [emphasis added]; See also: ECtHR,
_
Ilhan v. Turkey [GC], Application no. 22277/93, ECHR 2000-VII.
386 10 Admissibility Before the Strasbourg Court: An Outlook on the EU-Law. . .

exhaustion as rather problematic and excessively formalistic. Two core arguments


prove this: first, since both the direct and individual concern tests offer no space for
substantive human right protection unless the tensioned excessive formal proce-
dures are applied, and second, the personal circumstances of the applicant—recog-
nized in Kozacio glu v. Turkey as forming part of the evaluation on whether the rule
on exhaustion of domestic remedies needs be applied without exceptions—show
that he/she may have not found another means to directly attack the act or omission
of the EU without the two tests being passed first. It would suffice if the claimant
did everything ‘that could reasonably be expected of him or her to exhaust domestic
remedies.’31 Logically, if the applicant has done everything to pass the direct and
individual concern tests—even though adversely affected from the EU act or
omission—but ended up being refused by the Luxembourg Court, then the Stras-
bourg Court could very well call the case admissible under the flexibility standards
established in Kozacio glu v. Turkey. The same flexibility clause can apply when an
applicant challenged a certain Member State measure implementing EU law, in
which case the national court refused to send a preliminary reference question to the
EU Court even though the violation was rooted in an EU law provision.
The flexibility standards of the Strasbourg Court may cover the time limits as
well. As the 2-month time-limit of Art. 263 TFEU and Art. 265 TFEU may be
considered as too short for an applicant to make use of them, the Strasbourg Court
may also rely on the argument ruled in Financial Times Ltd and others, wherein it
ruled: ‘The Court observes that the timetable for the proceedings before the High
Court was tight and that the deadline for lodging written arguments before the Court
of Appeal was short.’32 Therefore, the time limits set in the action for annulment
and action for failure to act may be considered as meeting the benchmarks set in
Financial Times Ltd and others, therefore released from the duty to be exhausted
before arriving at Strasbourg Court. If the Strasbourg Court were to accept that
parties be given this elastic position as regards their standing,33 then the Strasbourg
Court would proceed with the admissibility stage of the EU applicant without the
EU Court being heard its voice. Such voice may, however, be heard only if the case
if found admissible, where then the Luxembourg Court would gain the jurisdiction
to prior involvement under Art. 3 (6) DAA.

31
ECtHR, Kozacioglu v. Turkey, op. cit. note 30, para. 40.
32
ECtHR, Financial Times Ltd and others v. The United Kingdom, Application no. 821/03,
Judgment, Strasbourg, 15 December 2009, para. 43.
33
Something not accepted in Pafitis relating to the EU’s assumed violation of Art. 6 ECHR, see:
Lawson (2005), p. 32.
10.3 Exceptions to the Rule on Exhaustion of Domestic Legal Remedies 387

10.3.2 There Are Several EU Remedies in Place: Which One


to Exhaust? A Question of Rationality
and Effectiveness

A very interesting and important scenario which could usually happen in the EU
system is that a claimant could have had his/her case heard in more than one level/
system of remedies, as was the case e.g. in Uni on de Peque~ nos v Council before the
EU Court where the applicant was reminded that there were two options for him to
choose: either the EU remedies or the national court.34 In case there is more than
one remedy possible—as is the case in most of the EU-related complaints—then the
question would be whether using an unsuccessful remedy by the applicant within
the EU-two-level system of remedies renders the application before the Strasbourg
Court inadmissible?
The Strasbourg Court has ruled in Karak o v. Hungary ‘that, where several
remedies are available, the applicant is not required to pursue more than one and it
is normally that individual’s choice as to which.’35 It is therefore not imperative for
the claimant to have exhausted both remedies, such as both the direct remedy like the
action for annulment and the indirect preliminary reference procedure. It would
suffice in practice for the claimant to exhaust one of them, in principle. However,
the bigger question in this regard is whether the Strasbourg Court would accept as
admissible a case wherein the claimant has in fact exhausted or at least tried a
domestic remedy which was thought to have been not successful. An example
would be an attempt like the applicant in ECJ Apple Computer International36 case
which used an action for annulment knowing that, in a logical way, the attacked EU
act was not of individual concern to it. Therefore, the question remains whether the
Strasbourg Court would accept Apple Computer International as being exhausted
even though the latter did try a remedy which in fact was thought to be insufficient to
meet the criteria of Art. 263 TFEU? Or, if the applicant has intentionally tried a legal
remedy, like the action for annulment or action for failure to act, which was known
from the very beginning that it would not pass the admissibility criteria? Would the
applicant be obliged the try the other available remedy, in case he/she tried first one
that was thought to have been inaccessible for such an action?
The Strasbourg Court has ruled in Kozacio glu v. Turkey ‘that an applicant must
have made normal use of domestic remedies which are likely to be effective and
sufficient and that, when a remedy has been pursued, use of another remedy which
has essentially the same objective is not required.’37 Furthermore, the Court ruled in

34
Court of Justice of EU, Union de Peque~ nos v Council, Case C-50/00, para. 40 et seq.
35
ECtHR, Karak o v. Hungary, Application no. 39311/05, Judgment, Strasbourg, 28 April 2009,
para. 14; See also: ECtHR, Granger v. the United Kingdom (no. decision of 9 May 1988); ECtHR,
Aquilina v. Malta, Application no. 25642/94, Judgment, Strasbourg, 29 April 1999, para. 39.
36
Court of Justice of EU, Apple Computer International v Commission, Case T-82/06, order of
19 February 2008.
37
ECtHR, Kozacio glu v. Turkey, op. cit. note 29, para. 40 [emphasis added].
388 10 Admissibility Before the Strasbourg Court: An Outlook on the EU-Law. . .

Jasinskis v. Latvia that ‘in the event of there being a number of domestic remedies
which an individual can pursue, that person is entitled to choose a remedy which
addresses his or her essential grievance.’38 This being the case, what Strasbourg
Court requests are three criteria which must be passed in order to an applicant such
as Apple Computer International not to be in need of exhausting the national-court
remedy before reaching Strasbourg: to be exact, first, that the remedy tried should
have been likely effective, second, that the remedy tried should have been sufficient,
and third, that the remedy tried has had essentially the same objective with the
untried remedy. Let us consider e.g. whether Apple Computer International passes
these three criteria: first, it is too difficult to know whether a certain remedy, like the
action for annulment or action for failure to act is effective, in a case like Apple
Computer International. Whether the action for annulment is effective in this case
would lead to the question of what were the likely scenarios and how many of such
scenarios were positive for Apple Computer International to pass the tests on direct
and individual concern. Logically, most of the scenarios would have been negative
in this regard. Second, Apple Computer International would have passed the test on
sufficiency, as the action for annulment tried by it would have sufficiently addressed
its claims, because, if the EU Court would find it admissible, the claimant had not
been in need because of the jurisdictional competence of the EU to send this case
before the national court as well. Third, for Apple Computer International it would
have had essentially the same objective if the case was submitted through a national
court, therefore passing the third test as well. This told, Apple Computer Interna-
tional would not have likely passed the first test of Kozacio glu v. Turkey, in which
case the Strasbourg Court would likely not accept it as exhausted in accordance
with Art. 35.1 ECHR, unless it is exhausted before a national court as well.
However, if it would have been likely that Apple Computer International be
found admissible, even though later found inadmissible, the Strasbourg Court
would have considered the case exhausted and admissible to get into its review
even though it was not exhausted in substance before the Luxembourg Court.
The Strasbourg Court has further clarified more flexibly the three above-
mentioned criteria39 in Akdivar and Others v. Turkey, by ruling that to ‘satisfy
the Court that the remedy was an effective one available in theory and in practice at
the relevant time, that is to say, that it was accessible, was one which was capable of
providing redress in respect of the applicant’s complaints and offered reasonable

38
ECtHR, Jasinskis v. Latvia, Application no. 45744/08, Judgment, Strasbourg, 21 December
2010, para. 50 [emphasis added]; See also: ECtHR, T.W. v. Malta [GC], Application no. 25644/94,
29 April 1999.
39
A similar line of argument was presented by the Court in D.H. and others v. The Czech Republic,
by ruling that: ‘It must then examine whether, in all the circumstances of the case, the applicant did
everything that could reasonably be expected of him or her to exhaust domestic remedies.’ See:
ECtHR, D.H. and others v. The Czech Republic, Application no. 57325/00, Judgment, Strasbourg,
13 November 2007, para. 116.
10.3 Exceptions to the Rule on Exhaustion of Domestic Legal Remedies 389

prospects of success.’40 In Akdivar and Others v. Turkey, the Court therefore


lowered a bit the threshold, and made the test a bit more substantive as opposed
to formal in nature. The mere requirement being the need that the tried remedy was
not only effective41 in practice but also in theory, meaning something that could
provide redress to the claimant’s submission and be of reasonable hope of success.
Certainly, one may say that Apple Computer International’s action was at least
available in theory, because the Luxembourg Court does often supersede that very
narrow objection, and the action for annulment which the applicant raised might
have both provided redress to it and be of reasonable success thereafter. Therefore,
had the flexible standard of Akdivar and Others v. Turkey been accepting as
applying to Apple Computer International case, the latter would have been consid-
ered as passing the domestic remedy exhaustion rule. Were the Strasbourg Court,
however, to apply in this case the standard of Kozacioglu v. Turkey, of course Apple
Computer International would have needed to bring the case before the national
court initially in order to become exhausted. The Strasbourg and Luxembourg
courts relations being rather rigid, and the need being to protect the EU law
autonomy with certain level of care, the former would be expected to apply the
more rigid standard of Kozacio glu v. Turkey as regards the application of the rule on
exhaustion of legal remedies on EU-related cases. Otherwise, a clear space for
interference to EU law autonomy would be available to Strasbourg Court.

10.3.3 The Complaint Must Have Been Substantively Raised


Through the Domestic Legal Remedies: Does This
Apply to the Preliminary Reference Procedure?

Another important condition which a case must fulfill under the context of the rule
on exhaustion is that it must have been raised before the domestic courts in a way or
another. However, the extent to which the alleged violation submitted before the
Strasbourg Court should have been raised in the domestic court is an ongoing
difficulty (not in the case of EU direct actions). It becomes an even more problem-
atic difficulty of one applies it in the context of the EU, wherein cases might have

40
ECtHR, Akdivar and Others v. Turkey, Application no. 21893/93, Judgment, Strasbourg,
16 September 1996, para. 68.
41
It is important to note the fact that the Strasbourg Court established in K.S. and K.S. AG
v. Switzerland that the claimants have requested the reopening of proceedings, in which case the
Court has ruled that ‘requests for the reopening of proceedings do not as a rule constitute an
effective remedy within the meaning of Article 26 (Art. 26) of the Convention.’ See: ECtHR, K.S.
and K.S. AG v. Switzerland, Application No. 19117/91, 12 January 1994, decision as to admissi-
bility; Compared to the circumstances in Kokkelvisserij, the applicant’s request to reopen the
proceedings were not something which they needed to do in order to become eligible to enter the
gates of the Strasbourg Court. This speaks for the fact that only available remedies were needed to
be used, but not those which are not in principle available or are extraordinary in nature.
390 10 Admissibility Before the Strasbourg Court: An Outlook on the EU-Law. . .

not merely been decided through direct actions but also through the indirect action
before the Luxembourg Court. One such problem is especially relevant in the
application of the preliminary reference procedure,42 in which case the Luxem-
bourg Court has merely dealt with the question of the national court as regards the
validity or interpretation of the EU law rather than the applicant’s claim as such.
For this purpose, the Strasbourg Court has established in G€ afgen v. Germany that
as regards whether and how should the case have been raised through the domestic
legal remedy, that: ‘It normally requires also that the complaints intended to be
brought subsequently before the Court should have been made to those same courts,
at least in substance and in compliance with the formal requirements and time-
limits laid down in domestic law.’43 Therefore, the need to have a case first
exhausted domestically should be accompanied with these two tests: first, that it
should have been raised in substance, without the need for it to be the first or the
main claim before the national court, and second, that such claim should have been
raised by respecting the procedural requirements of the domestic legal remedies.
Therefore, not only should the matter that is being brought before the Strasbourg
Court have been somehow raised as a matter of fact, but it should have also been
done in line with the domestic laws in the sense that an applicant may not contend to
have raised something in substance without having done that in accordance with the
procedural routes that make it possible to be heard by the domestic court. As
regards this, the Court has gone far further in Ahmet Sadik v. Greece by arguing
that: ‘Even if the Greek courts were able, or even obliged, to examine the case of
their own motion under the Convention, this cannot have dispensed the applicant
from relying on the Convention in those courts or from advancing arguments to the
same or like effect before them, thus drawing their attention to the problem he
intended to submit subsequently, if need be, to the institutions responsible for
European supervision.’44 What the Court practically asked in Ahmet Sadik
v. Greece is that the applicant should have substantively asked the national court
to review a certain measure against the Convention; otherwise it would not suffice
to pass the threshold.
Although the said threshold may depend also on the personal circumstances of
the applicant, the prevalent question related to the EU after accession is whether it
would be considered sufficient for a case to be considered substantively exhausted
before the EU court if it has merely gone through the preliminary reference
procedure. An example to test this is provided below. In C.I.L.F.I.T. v Ministry of
Health a national Italian court had referred a question for preliminary ruling to the
Luxembourg Court—in a case where the textile industry firms contested the

42
Cf.: Jacque (2011), p. 1017/8, noting that the presidents of the two European courts have argued
that the preliminary reference procedure under EU law may not ‘normally’ be considered a legal
remedy.
43
ECtHR, G€ afgen v. Germany, Application no. 22978/05, Judgment, Strasbourg, 1 June 2010,
para. 142 [emphasis added].
44
ECtHR, Ahmet Sadik v. Greece, Application no. 18877/91, Judgment, Strasbourg, 15 November
1996, para. 33.
10.3 Exceptions to the Rule on Exhaustion of Domestic Legal Remedies 391

application of their financial duties to the state authorities—by asking the following
question:
Does the third paragraph of Article 177 of the EEC Treaty, which provides that where any
question of the same kind as those listed in the first paragraph of that article is raised in a
case pending before a national court or tribunal against whose decisions there is no judicial
remedy under national law that court or tribunal must bring the matter before the Court of
Justice, lay down an obligation so to submit the case which precludes the national court
from determining whether the question raised is justified or does it, and if so within what
limits, make that obligation conditional on the prior finding of a reasonable interpretative
doubt?45

As seen, in C.I.L.F.I.T. v Ministry of Health, the national court has referred a


question to the Luxembourg Court merely as relating to the interpretation of the
treaty, but not as relating to the application of the law for which the applicant does
in fact submitted the claim to the Italian court.46 The original issue in C.I.L.F.I.T. v
Ministry of Health for which the claim was brought before the national court was
the applicant’s request to have a different interpretation of the financial duties of the
industrial firms, however the ‘substantive’ issue for which the national court
referred the preliminary question was one that merely covered the interpretation
of a Treaty provision on judicial remedies, which is in fact a subsidiary issue to the
main claim in C.I.L.F.I.T. v Ministry of Health. Applying the Strasbourg’s standard
of G€afgen v. Germany upon C.I.L.F.I.T. v Ministry of Health would render that the
only exhausted issue in substance here is the one on whether there is a right to legal
remedy under the EU law, and not the one whether there is any violation of the
Convention by the act of the Ministry of Health to apply a discriminated levy on the
textile industry. Therefore, the substantively exhausted issue in this case is the
subsidiary issue which rose throughout the proceedings in C.I.L.F.I.T. v Ministry of
Health, but not the main claim which the applicant raised therein. On the other
hand, even if the applicant would have wished to raise his main claim in the
preliminary reference question sent to the Luxembourg Court, there would have
been no right to him on doing that, as explained infra. Strasbourg Court conse-
quently would not accept C.I.L.F.I.T. v Ministry of Health as admissible if it attacks
an EU law provision, as it has not been exhausted properly before the EU Court.
This said, it is argued that fulfilling the standard of substantive exhaustion does
stand as a very difficult task in the cases of preliminary reference procedure invoked
by a national court to the Luxembourg Court. One should also add here that the
majority of EU-law cases are those that have been raised indirectly through the
Luxembourg Court. Therefore, a need to apply the effective standards of protection
on this category of cases stands as definite; otherwise the Strasbourg filter would be
so narrow that only a few cases may arrive there from the bulk of those that have

45
Court of Justice of EU, C.I.L.F.I.T. v Ministry of Health, Case 283/81, Judgment OF 6. 10. 1982,
part I.
46
Many of the preliminary questions may be indirect to the very claim raised by the applicant
before the national court. See e.g. d’Oliveira et al. (2011), p. 139, in a Bundesverwaltungsgericht
indirect preliminary reference question.
392 10 Admissibility Before the Strasbourg Court: An Outlook on the EU-Law. . .

gone through the preliminary reference procedure. This would seriously harm the
human right protection effectiveness from the Strasbourg side towards EU-related
applications.

10.4 The Nature of the Strasbourg Court’s Rulings


on EU-Related Matters: Is There Space
for the Supremacy of Strasbourg on Luxembourg?

A fundamental question is also whether the Luxembourg47 would accept as domes-


tically binding the rulings of the Strasbourg Court—and if yes, how?48 This
relationship between the rulings of the two courts would reflect not only their
formal and hierarchic positions but also their substantive model of dialogue, if
there will be one. Of course, it is clearly established that the Strasbourg Court’s
decision in EU-related cases will be internationally binding on the EU with an inter
partes effect.49 However, the question should be a bit deeper, that is to say whether
such decisions of the Strasbourg Court may become domestically binding in the EU
law as well? A paramount answer on this issue is the illustrious Opinion 1/91, which
amongst other builds the principle that:
Where, however, an international agreement provides for its own system of courts, includ-
ing a court with jurisdiction to settle disputes between the Contracting Parties to the
agreement, and, as a result to interpret its provisions, the decisions of that court will be
binding on the Community institutions, including the Court of Justice. Those decisions will
also be binding in the event that the Court of Justice is called upon to rule, by way of
preliminary ruling or in a direct action, on the interpretation of the international agreement,
in so far as that agreement is an integral part of the Community legal order.50

Opinion 1/91—as a self-proclaimed principle by the Luxembourg Court51—


makes it possible for certain international tribunals’ decisions to become binding on

47
These two cases demonstrate the rather opened position of Luxembourg Court to recognizing
the duties under international law as well: Court of Justice of EU, Poulsen and Diva, Case C-286/
90 [1992] ECR 1-6019; Court of Justice of EU, Racke, Case C-162/96 [1998] ECR 1-3655.
48
A jurisdictional conflict on this issue between the two courts may not be ruled out, however. See:
Olsen (2009–2010), p. 56.
49
Accord.: Rieder (2005), p. 83, who argues that since Mathews there were no more dispute on the
supremacy of Strasbourg Court over Luxembourg, and the hierarchy produced therein; On the fact
that EU will be obliged under international law to implement the Convention post-accession, see
also Odermatt (2014), p. 14; The EU Court has generally held that EU is bound to international law
‘in the exercise of its powers’. See e.g.: Court of Justice of EU, Anklagemyndigheden v Peter
Michael Poulsen and Diva Navigation Corp, Case C-286/90 [1992] ECR I-6019, para. 9.
50
Court of Justice of EU, Opinion 1/91, para 39; See also: Analysis (1997), p. 241; A similar
argument was pushed by ECJ also in: Court of Justice of EU, Deutsche Shell AG v Hauptzollamt
Hamburg-Harburg, Case C-188/91 [1993] ECR I-363, para. 17.
51
On the worries of the Court with this assertion, see: Eckes (2013), p. 259.
10.4 The Nature of the Strasbourg Court’s Rulings on EU-Related Matters:. . . 393

the EU-law level if two formal and substantive conditions are fulfilled: first, that the
international agreement be part of the EU legal order, and second, that the interna-
tional agreement provide for its own court system with jurisdiction to interpret its
provisions. The Convention, once EU accedes to it, will completely fulfill both of
these conditions,52 therefore the question then remains whether there is any specific
consequence that will be conveyed in this regard? It is first of all argued that the
principle set forth in Opinion 1/91 produces a huge effect on the post-accession
relationship between the Luxembourg and Strasbourg courts. It is for that reason
argued that the rulings of the Strasbourg Court will become internally binding on
the EU and the Luxembourg Court, which stands as a separate obligation compared
to the EU respect the Convention in terms of its international law obligations. With
the Strasbourg Court’s decisions on EU-related matters becoming even domesti-
cally accepted as binding on all institutions, including the Luxemburg Court, a clear
positioning in the hierarchy between the two courts is also established: namely, the
Strasbourg Court becoming the supreme instance in face of the Luxembourg Court
as the subordinated instance (obviously only in the context of Convention funda-
mental rights jurisdiction). This relationship will prove also as having effect on
inter-party issues, as Opinion 1/91 makes it clear that in the last part of the
abovementioned paragraph.
Given that the decisions of the Strasbourg Court are meant as penetrating
instantaneously in the EU legal order and are placed above the Luxembourg’s
chain of authority, then the dispute would be whether such decisions of the
Strasbourg Court can invalidate EU law directly and on their own right. A comment
of Edinburgh Law Review makes a strong point on this, by arguing that:
Although the obiter comments in Opinion 1/91, quoted above, open the way for the Court of
Justice to accept the binding nature of an international court’s rulings on the provisions of
an international agreement to which the Community is a party, they do not clarify the
position where those binding decisions have the effect of abrogating other provisions of
Community law.53

It is true, as claimed in the above quoted paragraph, that this has not been made
clear explicitly with Opinion 1/91, however a logical interpretation of the latter
gives a rather clear understanding on this. The argument is that since the decisions
of the Strasbourg Court on EU-related matters become internally binding on EU
institutions and the Luxembourg Court as well, they of course may directly inval-
idate conflicting EU law norms. The argument goes, however, that Opinion 1/91
makes it clear that such decisions become binding only on EU institutions,54 and,
correspondently, they may invalidate only those conflicting acts which have been
adopted under the competence of the EU institutions—of course excluding primary
law here. Therefore, the argument should be that as long as Opinion 1/91 is

52
See a similar argument at: Lock (2012), p. 195.
53
Analysis (1997), p. 241; Contra generally: Groussot et al. (2011), p. 5.
54
Cf.: Quirico (2010), p. 45, who considers that a scenario like this may put in danger the
autonomy of EU law (especially Art. 19(1) TEU).
394 10 Admissibility Before the Strasbourg Court: An Outlook on the EU-Law. . .

sustained on its original concept, then the Strasbourg Court rulings are supposed to
get incorporated and stand above EU secondary law, ipso iure invalidating
conflicting EU secondary law norms. Although this scenario might seem too
activist on the Strasbourg Court and rather harmful on the subordinated position
of the EU secondary law and Luxembourg Court, one must remind that such a
relationship may be so only within the remit of EU-related inter partes rulings
decided at Strasbourg55 and only within the Convention fundamental rights’ law
context.56 Therefore, the prospects for a hierarchical relationship57 between the
Strasbourg and Luxembourg courts seem to produce the effect that the former’s
rulings become not only international but also EU-domestic in nature, of course
making the disclaimer that the practice remains to test whether this will become real
or not. Whether the above practice would amount to an open interference to the EU
law autonomy58—a question made validly—seems not sound as an argument, as it
is the Luxembourg Court itself which has devised the nature of such tribunals’
decisions in the EU legal order, adding that such decisions may normally deal
merely with the Convention rights and not the EU law as such (the latter being a
distinction which may not be made practically at the EU Treaties’ level, as the
Convention is part of EU law). Ultimately, one must also keep in motion the Kadi
presumption which the Luxembourg Court could utilize against a binding Stras-
bourg Court decision which may be thought as interfering to a ius cogens human
right enshrined in the EU Treaties59—adding that e.g. ICJ’s Democratic Republic of
the Congo v. Rwanda confirms that fundamental rights comprise jus cogens
norms60—although this may seem not realistic to appear in practice.61

55
‘It is unlikely that the ECJ envisaged a bindingness akin to a doctrine of stare decisis since such
a doctrine does not exist anywhere in EU law or international law.’ Lock (2012), p. 195.
56
Cf.: Lock (2012), p. 195/6, who argues that this nature of bindingness may not go beyond the
accepted international law standard.
57
Franklin (2010–2011), p. 158 (‘[. . .] accession to the Convention might lead to a surrendering of
its jurisdiction with regards to fundamental rights protection in the Community, with its decisions
being made subject to review by a “higher” judicial authority - the ECtHR in Strasbourg.’).
58
Rieder (2005), p. 100 (‘Prioritizing the [. . .Strasbourg Court . . .] over the ECJ would weaken the
integrative authority of the ECJ at least in the field of human rights [. . .]’).
59
The logical path for this argument flows from: Eckes (2013), p. 279/10; See also a general
approach on this case, at: De Londras and Kingston (2010), pp. 409 et seq.
60
Accord: ICJ considers that human rights under international law comprise jus cogens norms
which may not be derogated. See an exemplary case on this: International Court of Justice, Armed
Activities on the Territory of the Congo (Democratic Republic of the Congo v. Rwanda), judgment
of 3 February 2006 (Jurisdiction of the Court and Admissibility of the Application), para. 64; On
the latter principle, see also: Inter-American Court of Human Rights, Advisory Opinion OC-18/03
of 17 September 2003 (requested by the United Mexican States), para. 97.
61
On the other hand, Yusuf seems rather contradictory to the previous argument, as the Court there
noted the fact that EU is bound to the hierarchy of international law, which would mean that the
EU Court does also accept the supremacy of Strasbourg Court in that light. See a similar argument
on Yusuf, at: Defeis (2007–2008), p. 6 et seq; One should also mention the argument that Kadi had
practically dealt with the lack of access to a court based on law, which may not be invoked as an
argument before EU Court if a case has already gone through the filter of Strasbourg Court (as the
10.5 Violations Originating in EU Primary Law: A Question of Possibility to. . . 395

From another point of view, Art. 52 (3) of the Charter seems to suggest that not
only ECHR but also the case-law of the Strasbourg Court stand as the minimum
standard for interpreting the Charter rights.62 It is of course arguably assumed that
then the Strasbourg Court’s position (erga omnes) is also supreme on Luxembourg
as long as the latter has to subsume to it for the minimum standard that it sets with
its case-law.63 This position is one that derives from the Treaties themselves, and
not from the international law obligations of the Union. Of course, it is finally
important to note that the EU Court constantly insists that the interpretation of the
nature of rights be done autonomously within the EU law (Internationale
Handelsgesellschaft),64 something these abovementioned arguments demonstrate
as contradicting the latter’s autonomy.

10.5 Violations Originating in EU Primary Law: A


Question of Possibility to Challenge the Treaties at
Strasbourg?

As violations originating in EU secondary law might be better addressed either


through direct or indirect actions as a means to fulfill the rule on exhaustion of
domestic remedies, it is far more problematic to exhaust ‘any’ remedy—either
direct or indirect—against an EU Treaty provision, as a means to become eligible to
stand before the Strasbourg Court. It is first of all important that, as regards the
Strasbourg Court jurisdiction, it has no limits on the scope of human rights review
protection that it can use to apply the Convention. One such landmark example of
the Strasbourg Court is Sejdic´ and Finci v. Bosnia and Herzegovina, wherein the
applicant has substantively challenged the Bosnia and Herzegovina constitution for
its failure to guarantee the Convention standards. The Strasbourg Court felt no

latter would have already offered access to court to that case, which makes it totally contrary to the
Kadi scenario).
62
Heringa and Verhey (2011), p. 17; Lenaerts and de Smijter (2001), p. 99; See the following three
cases showing a clear deference of Luxembourg to Strasbourg Court: Court of Justice of EU,
Bauer Verlag, C-368/95 [1997] ECR I-3689, para. 26; Court of Justice of EU, Grant v South West
Trains, C-249/96 [1998] ECR I-621, para. 34; Court of Justice of EU, Roquette Freres SA v
Directeur General de la Concurrence, de la Consommation et de la Repression des Fraudes,
C94/00 [2002] E.C.R. I-9011, para. 29; Cf.: Balfour (2005), p. 16 (‘It is not surprising that the ECJ
has, on occasion, adopted conflicting interpretations of the Convention that fail to meet the
‘minimum standards’ set by the ECtHR.’).
63
Wetzel (2003), p. 2853 (‘[. . .] each Court will draw upon the same Strasbourg jurisprudence,
minimizing the possibility of continuing grossly divergent ECHR interpretations.’).
64
Court of Justice of EU, Internationale Handelsgesellschaft mbH v. Einfuhr-und Vorratsstelle fur
Getreide und Futtermittel, Case 11/70, [1970] E.C.R. 1125, paras. 3–4.
396 10 Admissibility Before the Strasbourg Court: An Outlook on the EU-Law. . .

dilemma at finding that Bosnian constitution did not comply, therefore violated, the
Convention rights.65 Hence, the Sejdic´ and Finci v. Bosnia and Herzegovina seems
to demonstrate that in principle there would be no obstacle or reluctance from the
Strasbourg Court to engage with a direct review of compatibility of EU Treaties
with the Convention, one such case having been Mathews but also more or less
Kokkelvisserij in the past. As to the question whether in fact Treaties may become
illegal if they are found to violate the Convention, it is argued that the ‘Kadi has
elevated human rights (together with other core principles of EU law) to the status
of being the ‘very foundations’ of constitutional law that rank above ‘ordinary’ EU
primary law, a breach of the ECHR would logically make the EU Treaties unlawful
under EU law.’66 Of course, the EU has no instrument to remedy violations
stemming from the Treaties as it has no competence to change them on its own
right, although that reason is not proposed to exclude the EU from such responsi-
bility.67 Hence, the question remains which remedy—if the flexibility standard is
not applied—should an applicant exhaust at the EU level to become eligible to
attack a violation stemming from the EU Treaties before Strasbourg Court?
One must reiterate here that using the direct actions against the EU Treaties is
not possible, as there is no jurisdiction for the Luxembourg Court to check the
validity of an alleged violation of human rights situated in the EU Treaties68; the
same applies to the preliminary reference procedure under Art. 267 TFEU, wherein
the Luxembourg Court may do nothing else then interpreting the Treaties,69 but can
do nothing to review the legality of a supposed violation stemming from the
Treaties. The most probable route in this case would be to bring the case before
the national constitutional court, and ask it to review the constitutionality of the EU
Treaties70 such as in Bundesverfassungsgericht Lisbon case71 or the Czech

65
ECtHR, Sejdic´ and Finci v. Bosnia and Herzegovina, Applications nos. 27996/06 and 34836/06,
Judgment, Strasbourg, 22 December 2009, para. 39 et seq.
66
Eckes (2013), p. 282.
67
E.g.: Lock (2010), p. 783.
68
See also: Lock (2010), p. 790; Eckes (2013), p. 282; Rieder (2005), p. 82.
69
Competence-competence on this still remains on Member States. See e.g.: Břı́za (2009), p. 151.
70
Lock (2010), p. 790; See also: Payandeh (2011), p. 9; Cf.: Editorial (2013), p. 473, which
questions whether Member States constitutional courts may find unconstitutional if the EU Charter
provides more extensive human rights protection, and whether the primacy of EU law would apply
unreservedly; See also more on national constitutional courts sometimes blocking positions on EU
law primacy at the Member States’ legal orders, at: Martinico (2012), p. 419/20.
71
Federal Republic of Germany Bundesverfassungsgericht, Second Senate of 30 June 2009,
2 BvE 2/08, 2 BvE 5/08, 2 BvR 1010/08, 2 BvR 1022/08, 2 BvR 1259/08, 2 BvR 182/09; See:
Thym (2013), p. 391, on how this strategy of the German constitutional court had resulted in
Luxembourg Court being self-restrained not to make national constitutional laws irrelevant,
something which may change passively with EU becoming a party on its own right in the
ECHR system.; See also: Olsen (2009–2010), p. 62.
10.5 Violations Originating in EU Primary Law: A Question of Possibility to. . . 397

Constitutional Court’s Lisbon Case.72 In its Maastricht case,73 as an example, the


German Constitutional Court ‘acknowledged its task to review whether EU legal
acts remain within the limited competences of the EU and to declare acts that
transgress these borders to be ultra vires and hence inapplicable in Germany (ultra
vires review).’74 Although the legality of a national constitutional court75 to check a
posteriori the constitutionality of an international treaty,76 such as the EU Treaties,
might be in violation of Art. 46 of the Vienna Convention of the Law of Treaties,77
it may nevertheless appear as an exceptional practice.78,79 Lock takes the example
of TRNC case80—wherein the Grand Chamber of the Strasbourg Court made no
exception to the application of the rule on exhaustion and where the principle of
effective control of the territory was dismissed—as an example to demonstrate how
the Strasbourg Court might refuse to deal with cases which are not exhausted
substantively in the EU system for their question of the human rights legality of
EU Treaties.81

72
See: Břı́za (2009), p. 143, who points to the court when it was found acting ex ante by finding the
Lisbon Treaty constitutional but did not give up to control its constitutionality.
73
Hinarejos (2009), p. 113, showing how the Court played the role of a negative legislator on the
third pillar’s constitutionality, not willing to engage very directly and actively into the review of
the treaty.
74
Payandeh (2011), p. 9.
75
In the Lisbon case, although the decision of Germany on Lisbon Treaty was found constitu-
tional, the Extending Act was declared partially unconstitutional. See: Bieber (2009), p. 391 et seq,
to understand the argument that national constitutional courts might still be a very powerful
standing point for the authorship and origin of EU primary law.
76
Cf.: Bossuyt and Verrijdt (2011), p. 363 et seq, wherein one can see that some EU Member
States do tend to keep recognizing their constitutions (in hierarchical terms) as either softly in the
same position with the EU Treaties or sometimes even above them; This was not the case,
however, with the Czech example, at: Břı́za (2009), p. 143/4; See also: van Rossem (2009),
p. 203, which argues that Member States may not invoke the EU law—as established by VCLT—
as a reason for violations of ECHR. The same argument must apply on the EU as well.
77
See also: Suchocka (2007).
78
Lock (2010), p. 790; Cf.: to the European Arrest Warrant Framework Decision reviewed by the
Polish Constitutional Court, wherein the latter—interestingly enough—found no constitutional
means to rationalize it but decided that the Framework Decision be applied as long as the national
constitution is not reformed, at: Hinarejos (2009), p. 104.
79
Accord: Jackson (1992), pp. 310–340.
80
ECtHR, Demopoulos v Turkey (App. Nos 46113/99, 3843/02, 13751/02, 13466/03, 10200/04,
14163/04, 19993/04 & 21819/04) (2010).
81
Cf.: Gragl notes that it was initially thought to exclude the primary law from the accession
agreement and Strasbourg’s jurisdiction. In view of Gragl, there seem three arguments against
excluding primary law from the agreement: first, due to the fact that the EU would become highly
privileged against other contracting parties, second, the Luxembourg Court is well capable of
addressing problems in relation to the primary law issues that raise fundamental rights implica-
tions as there is now the legally-binding Charter, and, third, such exclusion would invite the
Strasbourg Court to review the division of borders between primary and secondary law the only
way to portray its jurisdiction on EU-originated cases, therefore interfering to the EU law
autonomy. See: Gragl (2013), p. 136/7.
398 10 Admissibility Before the Strasbourg Court: An Outlook on the EU-Law. . .

Although Lock’s argument might seem probable, there is another layer of


argumentation that needs be mentioned. First of all, there is a history of the
Strasbourg Court willing to engage with violations of the Convention stemming
from the EU primary law: both Mathews (more explicitly) and Kokkelvisserij
(implicitly) provide good examples,82 and the DAA does not exclude this possibil-
ity.83 Secondly, the EU Treaties’ potential violations might be well exhausted
through contesting their validity before the national courts, which may refer
preliminary questions to the Luxembourg Court for interpretation of Treaties;
although the majority of such proceedings will end up not considering the EU
Treaties against to the human rights law, they still do exhaust the issue domestically
on its substance. Thereafter, the applicant can have his/her claim heard at Stras-
bourg Court. Finally, the Strasbourg Court would most likely accept that the
flexibility clauses that it has developed on the application of the rule on exhaustion
if there are specific legal and systemic circumstances of the applicant—one such
specific legal circumstance being the fact that the EU Treaties may not be reviewed
their legality by the Luxembourg Court.84 A prime example of this fact—adding
here that also no constitutional court would have the jurisdiction to review the
legality of a violation stemming from the constitution—is Sejdic´ and Finci
v. Bosnia and Herzegovina, wherein the Court does flexibly apply the rule on
exhaustion and takes the constitutional violation as something which may have
not been exhausted by any domestic legal remedy at it very substantive point. The
Luxembourg Court, on the other hand, seems to be very allergic on such potential
situations, and would argue that the principle pacta sunt servanda applies first to the
Member States’ obligations towards the EU Treaties and excludes any other
normative conflict that may fracture this link.85

10.6 The New Life of Bosphorus Post-accession: A Rational


Viewpoint from the Strasbourg’s Lens

To start, De Schutter, on this inquiry, argues that—post-accession—it ‘would be


neither legally justified nor politically opportune to maintain the Bosphorus doc-
trine in its current form’ wherein Luxembourg Court is privileged compared to its

82
Some Member States had originally proposed excluding the responsibility of the Union for
primary law violations during the DAA negotiations, at: Groussot et al. (2011), p. 4.
83
Groussot et al. (2011), p. 9; See also: Gragl (2013), p. 136, who mentions the initial ideas that
had pushed for excluding the primary law from the accession agreement, which would have
immunized the EU from its obligation to have its treaties on its own capacity complied with the
Convention standards. Gragl in this regard argues that it would have been a wrong decision to
exclude primary law from the jurisdiction of the accession agreement, as that would have produced
an unequal treatment for those persons subject to primary law violations compared to those subject
to the secondary law violations.
84
On the latter point, see e.g.: van den Berghe (2010), p. 113.
85
See e.g.: Licková (2008), p. 473/4.
10.6 The New Life of Bosphorus Post-accession: A Rational Viewpoint from. . . 399

normal position as a constitutional court of a typical contracting party.86 De


Schutter in this regard says that—if for certain political reasons Bosphorus needs
to continue in force for another period—then it needs to be adjusted into a more
modern and theoretically grounded framework. Such transformation needs to
construe Bosphorus into an objective criteria-based framework that applies across
all contracting parties if they are able to justify that they have met procedurally the
required justifiable standards. Such application on the horizontal level—as De
Schutter argues—would then make the accession process relevant not only from
the perspective of the changed nature of Bosphorus with regard to EU legal order
but also state High Contracting Parties’ legal orders. Such objective standards—De
Schutter argues—need to derive from across all legal orders of Europe, and, by
taking that direction, the freshened Bosphorus doctrine will basically reflect better
the European standard of rights which will be then compulsory among all
contracting parties, therefore altering the quality of European common human
rights law into a wealthier and better standardized across the jurisprudence of
Strasbourg Court.87
Basically, the proposal of De Schutter seems rather well promoting equality
between Convention’s contracting parties—instead of a privileged status for the
EU. Such redefinition of Bosphorus would therefore apply an equal-footing stan-
dard among all contracting parties. As to the nature of standard of assessment, this
new format of Bosphorus proposed by De Schutter seems reasonable and more
notionally grounded. However, the scope of equality provided for the renewed
Bosphorus does not per se justify its application even though De Schutter’s
proposal is to keep Bosphorus as an objectively-based set of criteria reflecting the
best standards of European human rights law. The author of this book gives two
reasons for opposing this proposal and supporting the entire dismissal of Bosphorus
post-accession. First, the Convention system is designed—both originally and from
the context of its intended function—as a system of human rights protection to
allow full access of claimants to its review filter when it comes to individual human
rights complaints. If the Convention system would become merely a set of high
standards not practically allowing individual access and assessment of each

86
De Schutter (2014), p. 198; See also: Gragl (2013), p. 74, who, referring to Heer-Reismann,
argues that post-accession the risk of potentially conflicting obligations between the two treaty
regimes will not exist anymore, as there is will an institutional system in place to coordinate and
potentially resolve conflicts.
87
De Schutter (2014), p. 198; Cf.: Gragl rightly brings the question of unequal treatment that the
EU compared to other high contracting parties would maintain if Bosphorus will remain in force
post-accession. Furthermore, Gragl points out that it is not the aim of accession to subject the EU
to unequal treatment compared to other contracting parties in the Convention system. In this
regard, Gragl takes the example of MSS case in which the Strasbourg Court basically demonstrated
that it does not wish to apply the equivalent protection standard on national law even if the latter
comes under the umbrella of EU law (if there are possibilities for the member state to depart from
that law exceptionally). In this regard, Gragl notes that it would be wise even to revoke entirely this
presumption post-accession, as this step would finally remedy the ‘anomaly’ of the EU being a
privileged contracting party to the Convention system. See: Gragl (2013), p. 75.
400 10 Admissibility Before the Strasbourg Court: An Outlook on the EU-Law. . .

claimant’s application it would depart from the logic of serving a full and effective
external human rights control mechanism as it serves currently to EU Member
States’ claimants. Second, it would be too speculative for the Strasbourg Court to
engage with developing the standards of human rights protection from the best
national examples for two basic reasons: one, the Convention system should not
aim to demise the traditional models of human rights protection of many national
legal orders; and, two, it is not—and it should not be—the intention of the
Convention system to become the maximum human rights standard of human rights
law in Europe: instead, it should be the minimum standard. The latter argument
makes it not possible for the Strasbourg Court to take the role which De Schutter
proposes for it within the redefined modeling of Bosphorus. And finally, it would
not be practically possible—for the Strasbourg Court, being a subsidiary and
negative legislator-oriented court—to take over devising and updating actively
the standard of human rights within the De Schutter’s proposal if it does not play
actively and mostly not in the positive-legislative attitude of jurisprudence. There-
fore, for the benefits of effective human rights protection in Europe—although De
Schutter’s proposal is sound and rational to the degree of political needs which the
Convention system may need to push forward—there is the necessity for complete
demise of the Bosphorus doctrine in order to make EU a regular and legitimate
contracting party and its citizens equal claimants before the Strasbourg Court’s
review mechanisms.

10.7 Chapter’s Summary of Conclusions

The chapter analyzed the course of actions on admissibility that the Strasbourg
Court performs on an EU-related application once it reaches its gates. The chapter
offered a number of assessments to demonstrate the practical and conceptual
problems and situations that may arise once an EU-related application comes before
the review of Strasbourg Court. The chapter concludes that there are several
exceptional standards which the Strasbourg Court may apply upon an
EU-originated application, that being a fact that shows the potential room for
disagreement between the two orders.
The chapter first offered an analysis of the admissibility criteria applied by the
Strasbourg Court on EU-related applications, considering specifically the victim
status of the claimant. The chapter concludes that Strasbourg Court may apply a far
more appropriate standard on the victim status of the claimant compared to the EU
law, this being an important fact that may push the EU Court to change the criteria
of admissibility for its unprivileged applicants. On the other hand, the manner the
victim status of the applicant is tested by the Strasbourg Court is rather
unproblematic and will not put in motion any concern from any side. Having
considered the victim status issue, the chapter then embarks on the very important
issue of exceptional admissibility criteria which Strasbourg Court applies in rela-
tion to the High Contracting Parties’ applications. The chapter argues that there is
References 401

sufficient set of standards that may allow the Strasbourg Court to call an EU-related
case admissible even though it may have not been exhausted internally in EU law: if
that scenario would happen upon the EU, with the latter having no right to join as
co-respondent, then there would be no right on the Luxembourg Court to utilize the
prior involvement mechanism, thereby allowing that the Strasbourg Court become
the first-instance court for such cases. The effect of exceptional rules on admissi-
bility—this chapter argues—may well hinder the shielding measures which Proto-
col 8 requests in face of the DAA.
The chapter then embarks on the issue of legal effect of Strasbourg Court
decisions in the EU legal order. The chapter concludes that, being based in
Luxembourg’s previous case law, it is likely that the Strasbourg’s decisions will
have direct application in the EU legal order, with the final effect of making the
Luxembourg Court subsumed to the Strasbourg Court. On the other hand, the
chapter questioned whether Strasbourg Court would accept to board in the direct
review of EU Treaties, and whether it would consider itself authorized to label them
as contrary to the Convention. The chapter concludes that there is an established
practice of Strasbourg Court engaging directly in the review of constitutional law of
other High Contracting Parties, therefore no barrier may be though to exclude
Strasbourg’s potential to reign directly on the EU Treaties’ legality under the
Convention. Such attitude would seriously promote the Strasbourg’s rule in Europe
as the constitutional court for the EU as well, hindering the nature and position of
Luxembourg’s reign over the EU Treaties.
Finally, the chapter concludes that the review of Strasbourg Court upon
EU-related applications is one which may sometime follow the exceptional proce-
dural routes, the latter empowering Strasbourg Court even further in face of
Luxembourg regime of law. Such attitude does of course—the chapter con-
cludes—further shape the fact that accession will contribute to making the rela-
tionship with Strasbourg Court institutionalized at a certain likely cost for the EU
law system, nevertheless.

References

Analysis (1997) The European community cannot accede to the European Convention on Human
Rights. Eur Law Rev 1:235–249
Arnull A (2003) From charter to constitution and beyond: fundamental rights in the new. European
Union. Public Law (Winter Issue) (4):774–793
Balfour ADJ (2005) Application of the European Convention on Human Rights by the European
Court of Justice. Harv Law Sch Stud Scholarship Ser Paper 4:1–57
Bieber R (2009) An association of sovereign states. Eur Const Law Rev 5(3):391–406
Bossuyt M, Verrijdt W (2011) The full effect of EU law and of constitutional review in Belgium
and France after the Melki Judgment. Eur Const Law Rev 7(3):355–391
Břı́za P (2009) The Czech Republic: the constitutional court on the Lisbon Treaty Decision of
26 November 2008. Eur Const Law Rev 5(1):143–164
Carozza PG (2003) Subsidiarity as a structural principle of international human rights law. Am J
Int Law 97(1):38–79
402 10 Admissibility Before the Strasbourg Court: An Outlook on the EU-Law. . .

Conforti B (2010) Comments on the accession of the European Union to the ECHR. Italian Yearb
Int Law 20:83–86
Costa J-P (2003) The European Court of Human Rights and its recent case law. Tex Int Law J
38:455–468
d’Oliveira HUJ, de Groot GR, Seling A (2011) Court of Justice of the European Union: Decision
of 2 March 2010, Case C-315/08, Janko Rottman v. Freistaat Bayern Case Note 1 Decoupling
Nationality and Union Citizenship? Case Note 2 The Consequences of the Rottmann Judgment
on Member State Autonomy – The European Court of Justice’s Avant-Gardism in Nationality
Matters. Eur Const Law Rev 7(1):138–160
De Londras F, Kingston S (2010) Rights, security, and conflicting international obligations:
exploring inter-jurisdictional judicial dialogues in Europe. Am J Comp Law 58:359–414
De Schutter O (2014) Bosphorus post-accession: redefining the relationships between the
European Court of Human Rights and the parties to the Convention. In: Kosta V et al (eds)
The EU accession to the ECHR. Hart Publishing, Oxford
Defeis E (2007–2008) Dual system of human rights: the European Union. ILSA J Int Comp Law
14:1–8
Eckes C (2013) EU accession to the ECHR: between autonomy and adaptation. Mod Law Rev 76
(2):254–285
Editorial (2013) Human rights: Member State, EU and ECHR levels of protection; P.S. Catalonia;
why does it take so long for my article to be published? In this issue: human rights: Member
State, EU and ECHR levels of protection. Eur J Int Law 24:471–475
Franklin C (2010–2011) The legal status of the EU charter of fundamental rights after the Treaty of
Lisbon. Tilburg Law Rev 15:137–162
Gragl P (2013) The accession of the European Union to the European Convention on Human
Rights. Hart Publishing, Oxford
Groussot X et al (2011) EU accession to the European Convention on Human Rights: a legal
assessment of the Draft Accession Agreement of 14th October 2011. Fondation Robert
Schuman/European Issues No 218/7 November 2011
Heringa AW, Verhey L (2011) The EU charter: text and structure. Maastrich J Eur Comp Law 8
(1):11–32
Hinarejos A (2009) The Lisbon Treaty versus standing still: a view from the Third Pillar. Eur
Const Law Rev 5(1):99–116
Jackson JH (1992) Status of treaties in domestic legal systems: a policy analysis. Am J Int Law 86
(2):310–340
Jacque JP (2011) The accession of the European Union to the European Convention on Human
Rights and Fundamental Freedoms. Common Market Law Rev 48(4):995–1023
Lawson R (2005) Human rights: the best is yet to come. Eur Const Law Rev 1(1):27–37
Lenaerts K, de Smijter E (2001) The charter and the role of the European Courts. Maastrich J Eur
Comp Law 8(1):90–101
Licková M (2008) European exceptionalism in international law. Eur J Int Law 19(3):463–490
Lock T (2010) EU accession to the ECHR: implications for judicial review in Strasbourg. Eur Law
Rev 35(6):777–798
Lock T (2012) End of an epic? The draft agreement on the EU’s accession to the ECHR. Yearb Eur
Law 31(1):162–197
Martinico G (2012) Is the European Convention going to be ‘Supreme’? A comparative-
constitutional overview of ECHR and EU Law before National Courts. Eur J Int Law 23
(2):401–424
Odermatt J (2014) The EU’s accession to the European Convention on Human Rights: an
international law perspective. Leuven Center for Global Governance Studies. Working Paper
No. 136
Olsen JB (2009–2010) Protecting fundamental rights and the evolving roles of the Court of Justice
of the European Union and the European Court of Human Rights; Europeanisation in action at
the supranational level. Edinb Stud Law Rev 1(3):54–73
References 403

Payandeh M (2011) Constitutional review of EU law after Honeywell: contextualizing the


relationship between the German Constitutional Court and the EU Court of Justice. Common
Market Law Rev 48(1):9–38
Quirico O (2010) Substantive and procedural issues raised by the accession of the EU to the
ECHR. Italian Yearb Int Law 20:31–53
Raba K (2013) The accession of the European Union to the European Convention for the
Protection of Human Rights and Fundamental Freedoms—overview of the accession agree-
ment. ERA Forum 14:557–572
Ribble MR (2010–2011) I don’t trust your judgment: the European Convention on Human Rights
meets the European Union on new grounds? Pa State Int Law Rev 29(1):211–232
Rieder C (2005) Protecting human rights within the European Union: who is better qualified to do
the job-the European Court of Justice or the European Court of Human Rights? Tulane Eur
Civil Law Forum 20:73–107
Suchocka H (2007) Opinion no. 405/2006 on the Constitution of Serbia (Parts V, 7–9, VI and
VIII). CDL(2007)005rev. European Commission for Democracy Through Law (Venice Com-
mission), Strasbourg, 6 March 2007
Thym D (2013) Separation versus Fusion – or: how to accommodate national autonomy and the
charter? Diverging visions of the German Constitutional Court and the European Court of
Justice. Eur Const Law Rev 9(3):391–419
van den Berghe F (2010) The EU and issues of human rights protection: same solutions to more
acute problems? Eur Law J 16(2):112–157
van Rossem JW (2009) Interaction between EU law and international law in the light of Intertanko
and Kadi: the dilemma of norms binding the Member States but not the Community. Nether-
lands Yearb Int Law 40:183–227
Wetzel JR (2003) Improving fundamental rights protection in the European Union: resolving the
conflict and confusion between the Luxembourg and Strasbourg Courts. Fordham Law Rev
71:2823–2862
Part IV
Approaching the Final ‘Station’
Chapter 11
Before the Conclusion: Luxembourg Court’s
Opinion 2/13 on the DAA’s Compatibility
with the EU Treaties

11.1 Introduction

At the time when this book was almost completely finalized, the Luxembourg Court
issued its Opinion 2/13 on the compatibility of the DAA with the EU Treaties. The
author of the book was advised by his supervisors to examine this opinion in a
subchapter and to compare its stances with the book’s core arguments (where
applicable). This said, the book has been devised from a position where there was
no opinion in place—and it still continues to have its academic relevance with those
findings—however, it is important to give some taste to the book on the conclusion
of the Luxembourg Court which should at least be considered from a comparative
perspective against this book’s most important findings. As of the time of finalizing
this book, there was no other development except Opinion 2/13 which would have
added upon this policy perspective. This said, one should not read the Opinion as
negating the accession process or its intention and the current accession agreement:
it merely should serve as a list of requirements that will finally add and revise some
of the minor issues in the Accession Agreement to make it compliant with the
‘allergic’ wish of the Luxembourg Court. In this subchapter, the book examines
each substantive part of the Opinion and suggests certain arguments which either
criticize or support the Court’s findings. In the final part of this subchapter, the book
proposes a list of nine concrete amendments (either in the form of revisions to the
current provisions or in the form of supplementary provisions that need be added to
the DAA) to make the current DAA compliant with the Opinion 2/13.
It is important to restate that on basis of Art. 218 (11) TFEU:
A Member State, the European Parliament, the Council or the Commission may obtain the
opinion of the Court of Justice as to whether an agreement envisaged is compatible with the
Treaties. Where the opinion of the Court is adverse, the agreement envisaged may not enter
into force unless it is amended or the Treaties are revised.

© Springer International Publishing Switzerland 2015 407


F. Korenica, The EU Accession to the ECHR, DOI 10.1007/978-3-319-21759-8_11
408 11 Before the Conclusion: Luxembourg Court’s Opinion 2/13 on the. . .

Based on this legal basis, the Commission had requested from the Luxembourg
Court to pronounce on the compliance of the DAA package with the EU Treaties.
The specific question which the Commission submitted to the Court was:
Is the draft agreement providing for the accession of the European Union to the Convention
for the Protection of Human Rights and Fundamental Freedoms [, signed in Rome on
4 November 1950 (“the ECHR”),] compatible with the Treaties?1

The Commission’s question was rather simple and straightforward, as it was


built upon the premise that none of the negotiating parties of the Draft Accession
Agreement—including the EU Member States—had raised any doubt as to the
legality of it under the EU Treaties. Roughly the same positive view was provided
by AG Kokkot, who had although argued that there exist loopholes which may be
used to interfere to the EU law autonomy, nevertheless, the agreement was consid-
ered to be in general in compliance with the EU Treaties.2 The Court responded in
Opinion 2/13 that the DAA and its package was not in compliance with the EU
Treaties (Art. 6 TEU and Protocol 8 to the Treaties), something that will force the
negotiators to add another layer of revisions/additions to the DAA to finally make it
compliant with the Luxembourg Court’s requisites. The Opinion of the Court is
written in a very simple language, it is fairly short in its argumentative part and tries
to say unilateral propositions rather than engage with and provide in-depth and
sound arguments. In spite of its language, it must be taken legally into account in
order to make the accession project realizable. As a general overview, the author of
this book finds the Opinion poorly argued—although its key findings do have
academic relevance, if not practical, the majority of them standing in the same
lines which this book has proved far before in time—which may certainly give the
impression that the Luxembourg Court is ‘jealous’ in witnessing accession taking
place with so much pride.
Two broad-spectrum comments need be made on the opinion considered against
the findings of this book. First, the Opinion confirms many of the loopholes which
this book has examined profoundly. It is in this sense a proof that the book’s
findings were correct, although this book had found far more and far deeper
problems than just those declared by the Court. The Opinion itself therefore does
not destabilize or counter any of the findings of this book: it merely serves/supports
the arguments which have been devised in this book far earlier in time. Second, the
opinion nevertheless confirms the very allergic tendency of the Luxembourg Court
to recognize external control from an international court. Such aversion, as shown
in this Opinion, goes far beyond the likely situations that may emerge in practice.
With this mood of allergy, one may properly say that Opinion 2/13 is an excellent

1
Court of Justice of EU, Opinion 2/13 (Accession of the European Union to the European
Convention for the Protection of Human Rights and Fundamental Freedoms—Compatibility of
the draft agreement with the EU and FEU Treaties). Opinion of the Court (Full Court) of
18 December 2014, para. 1.
2
Opinion of AG Kokkot, in case: Court of Justice of EU, Opinion 2/13, delivered on 13 June 2014,
para. 280.
11.1 Introduction 409

illustration to show that Luxembourg Court judges—not the EU Treaties—have


been pained by the fact that accession of EU to ECHR will bring another layer of
judicial control to the EU: this being a requirement of the European constitutional
democracy which the EU should normally have no luxury to refuse. This strange
frame of mind of the Luxembourg Court confirms the fact that its judges wish to
continue to be reigning over the kingdom of EU law alone, without any form of
external democratic control3 in the format of an international human rights con-
vention system. One must finally mention here that academic doubts over the EU
law autonomy should not be the language which the Luxembourg Court—which
should referee from the standpoint of practical and probable situations—may use
(although not constructed in a sophisticated language) to opine on a constitutional
matter which will produce systemic positive effects for the EU’s participation in
democratic international law.
To start, one should mention that Opinion 2/13 starts with the Court’s general
observation of courtesy—not necessarily fitting with the final findings of the
Court—which states that:
The Court of Justice has admittedly already stated in that regard that an international
agreement providing for the creation of a court responsible for the interpretation of its
provisions and whose decisions are binding on the institutions, including the Court of
Justice, is not, in principle, incompatible with EU law; that is particularly the case where, as
in this instance, the conclusion of such an agreement is provided for by the Treaties
themselves. The competence of the EU in the field of international relations and its capacity
to conclude international agreements necessarily entail the power to submit to the decisions
of a court which is created or designated by such agreements as regards the interpretation
and application of their provisions.4

By reiterating this, the Court desires to reveal the fact that it is not against EU
accession to ECHR as such—or the Draft Accession Agreement—but it wishes to
see additional safeguards which will protect its autonomy, both internally and
externally. It answers this concern with a ‘Lotus’ language, by showing ‘what is
not incompatible’ with the EU Treaties rather than ‘what is compatible’ with them.
This ‘prohibitive’ language which is a reflection of the ICJ rhetoric seems to
demonstrate the sensitive disposition of the Luxembourg Court to the jurisdictions
of international courts. It is nevertheless contradictory to rule that, on the one hand,

3
By naming the Convention system as an ‘external democratic control’, the book does not intend
to make a difference between the Luxembourg and Strasbourg in terms of democratic credentials,
or even name the latter as more democratic than the former. The intention is rather to explicate the
fact that in a European democracy landscape, there is always the need to formally—at least—
comply with external human rights law systems, like the Convention system. Such fact—although
it may have formal value in face of the EU, as many would say—amounts to the idea that even
though a certain policy has sufficient internal democratic mechanisms (e.g. the EU), it still needs to
become exposed to an external human rights control system like the Convention system in order to
tie up the an overarching approach to the protection of human rights. Such external democratic
control is therefore a must, even though in certain cases such as the EU it may have merely a
formal relevance.
4
Court of Justice of EU, Opinion 2/13, op. cit. note 1, para. 182.
410 11 Before the Conclusion: Luxembourg Court’s Opinion 2/13 on the. . .

it is possible for the EU to accede to treaty regimes which possess their own
international court mechanisms, and on the other, to seek additional (sometimes
unusual) safeguards and privileges to participate there. There is something which
the EU Court needs to learn: public international law may not become so soft as to
remove every possible ‘tooth’ of external review towards the EU, since that would
result into an international ‘lack of law’. On this, Steve Peers engagingly argues
that: ‘Quite frankly, EU accession to the Convention, in the terms defined today by
the CJEU, could only appeal to those who don’t like human rights very much.’5 Let
us now examine each substantive issue dealt with in the Court’s Opinion 2/13.

11.2 Accession Shall (Should Not?!) Bring Significant


Constitutional Changes to the Treaty System

The Court starts its substantive arguments in Opinion 2/13 by referring to its earlier
Opinion 2/94 in which it had ruled that, as the treaty system had stood at that time,
then Communities had no competence to accede to ECHR. Now, with Art. 6 TEU,
the Court recalls that the basis for that competence has been formed already. One
basic distinction needs be made, nevertheless. If one reads carefully Opinion 2/94,
it centrally argues that the core problem with the EU accession to ECHR at that time
was the lack of competence for the then Communities to achieve it. That argument
was founded on the rationale that, if it were the Masters of the Treaties to form that
basis, they would do it with the understanding that accession as such will bring
significant constitutional changes to the EU constitutional regime. The contradic-
tory statement which one finds now in paragraph 153–4 of Opinion 2/13 tries to say
that basically it is not anymore important whether there is any constitutional basis
and competence for the EU to accede to the ECHR—which contradicts with
Opinion 2/94—bur rather the nature of changes which the utilization of that
competence will produce in the EU legal order. The Court is therefore changing
the scheme of its argument, and as a result using a double standard. Although this
argument in itself further confirms the basic Hypothesis of this book (see supra
Hypothesis 1 at Chap. 1), it stands rather illogical in the way the Court has
substantiated it in this recent opinion.
The Court reiterates that the accession of EU to ECHR should preserve—or,
otherwise, should not harm—the specific characteristics of EU law. ‘It should be
borne in mind that these characteristics include those relating to the constitutional
structure of the EU, which is seen in the principle of conferral of powers referred to
in Articles 4(1) TEU and 5(1) and (2) TEU, and in the institutional framework
established in Articles 13 TEU to 19 TEU.’6 It then insists that there is a clear
internal and external autonomy principle in EU law which needs be preserved

5
Peers (2014).
6
Court of Justice of EU, Opinion 2/13, op. cit. note 1, para. 165.
11.3 Art. 53 of the Charter (Un)coordinated with Art. 53 of the Convention:. . . 411

consistently (see supra the chapter on EU law autonomy).7 That principle, the Court
argues, reflects upon the fact that the judicial system of the EU should be kept
isolated from external interferences, and that the rights and freedoms of the EU law
should be ensured within the objectives of the EU law on its own right (referring to
Internationale Handelsgesellschaft and Kadi, as examined supra in Chaps. 2 and
3).8 It is argued at this point that the Court itself proclaimed in Opinion 2/94 that EU
accession to ECHR will be of significant constitutional impact to the EU treaty
system. By accepting this fact, the Court had logically accepted that should there be
a competence to allow that effect, it would not be prohibited that accession generate
such significant constitutional impact in the EU treaty system. This said, the general
line of the argument in Opinion 2/13 seems to be that as long as accession is of
constitutional significant impact for the EU treaty system, it will not comply with
the EU treaties, although it is the treaties themselves which have authorized that
accession process. This contradictory approach—although this book argues that in
any way accession will be of significant constitutional impact to the EU treaty
system—seems to say that EU accession to ECHR should not disturb any consti-
tutional mechanism in order for it to comply with the EU Treaties: if this condition
shall prevail in the coming renegotiation process, it may end up with EU accession
to ECHR being done merely for formalistic reasons.

11.3 Art. 53 of the Charter (Un)coordinated with Art.


53 of the Convention: Fighting for Internal Primacy?

A first specific substantive issue which the Court emphasizes in its Opinion is the
interaction between the Charter of Fundamental Rights and the ECHR, and the
possible external impact from the Strasbourg Court to the EU law interpretation of
rights. It started this battle by recalling that ‘[. . .]any action by the bodies given
decision-making powers by the ECHR, as provided for in the agreement envisaged,
must not have the effect of binding the EU and its institutions, in the exercise of
their internal powers, to a particular interpretation of the rules of EU law.’9 There
are two basic misunderstandings of the EU Court with this statement (although this
book confirms that it is precisely this effect which will be reached through the

7
Court of Justice of EU, Opinion 2/13, op. cit. note 1, para. 166 et seq.
8
Court of Justice of EU, Opinion 2/13, op. cit. note 1, para. 170; Cf.: Gragl (2013), p. 57, who in
the same line argues that the EU Court has followed a path of argument that basically says that
fundamental rights derive from Union law as opposed to an external source of obligation. This
being an issue, Gragl rightly points out that the fact that ECHR will remain a minimum standard in
terms of its Art. 35 may bring collisions with the EU law primacy when Member States act under
the scope of EU fundamental rights obligations. I agree with this stance to the extent that such
conflicts may arise, but not to the extent that such conflicts would remain unmanageable under the
current set-up of EU law.
9
Court of Justice of EU, Opinion 2/13, op. cit. note 1, para. 184.
412 11 Before the Conclusion: Luxembourg Court’s Opinion 2/13 on the. . .

current DAA and EU internal rules on this matter). First, as well known, ECHR is
an international law instrument which does not seek to bind any institution of EU
from the EU internal law perspective. It merely requests compliance from the
external perspective. For the ECHR system it would suffice that the EU merely
respect its judgements and standards. The fact that there exists the possibility that
the Luxembourg Court be obliged to follow the line of the Strasbourg Court when
interpreting ECHR rights which correspond with the Charter rights is an issue of EU
law not of ECHR law. It is the Charter—which forms part of the EU Treaties—and
the declaration attached to it (explained supra in Chap. 2, Sect. 4) which makes it
obligatory for the EU Court to interpret the Charter rights corresponding to the
ECHR rights at least not below the standard of protection provided by Strasbourg
Court’s case-law. This is an internal EU law obligation which the EU Court needs to
fulfil within the scope of its internal obligations at any time, be it before or after the
accession. When the EU Treaties refer to ECHR they refer to it in both times, before
and after accession. Therefore, for the Luxembourg Court to lay blame on Stras-
bourg Court and the DAA for a stipulation which forms part of the EU internal law
is very unreasonable and beyond the points raised in the question posed by the
Commission. So, the point then is this: the Luxembourg Court wants to change the
effect of an EU Treaty obligation by requiring that the DAA limit the scope of effect
which the Strasbourg Court’s interpretation of rights may have on Luxembourg
Court, something that originates from an EU law obligation and no-one else. This is
may be described merely as an intentional circumvention of the EU Treaties from
the Luxembourg Court in order to lock the loophole which may interfere to its
external or disbalance its internal autonomy. Let us, however, take that as it appears
in the Opinion 2/13.
Most concretely, the Luxembourg Court sees this problem with regard to Art.
53 of the Charter, which ‘provides that nothing therein is to be interpreted as
restricting or adversely affecting fundamental rights as recognised, in their respec-
tive fields of application, by EU law and international law and by international
agreements to which the EU or all the Member States are party, including the
ECHR, and by the Member States’ constitutions.’10 To this end, the Court further
argues that:

10
Court of Justice of EU, Opinion 2/13, op. cit. note 1, para. 187; Cf.: Referring to this provision,
Gragl basically notes that this is a conflict resolution norm, which Lenaerts and De Smijter argue
represent an indication for the law that should always take precedence over the other. See: See:
Gragl (2013), p. 61. The argument of Lenaerts and De Smijter is something I fully agree, adding
that such reasoning may lead to the assumption that the Convention may be considered part of the
primary law of EU as well (something which has been argued extensively supra in the chapter on
the status of ECHR and DAA in EU legal order). Contrary to this, Gragl argues that Art. 53 ChFR
also forms part of the principle of EU law primacy, therefore the above argument seems to be
rejected by him. However, the mere fact that the Court seeks a sort of coordination between Art.
53 ChFR and the Convention indicates that the Court is somehow accepting a horizontally equal
level between the Charter and the Convention, otherwise there would have been no need to seek
‘coordination’ but rather seek to enforce the ‘subordination’ of one to the other.
11.3 Art. 53 of the Charter (Un)coordinated with Art. 53 of the Convention:. . . 413

In so far as Article 53 of the ECHR essentially reserves the power of the Contracting Parties
to lay down higher standards of protection of fundamental rights than those guaranteed by
the ECHR, that provision should be coordinated with Article 53 of the Charter, as
interpreted by the Court of Justice, so that the power granted to Member States by Article
53 of the ECHR is limited — with respect to the rights recognised by the Charter that
correspond to those guaranteed by the ECHR — to that which is necessary to ensure that the
level of protection provided for by the Charter and the primacy, unity and effectiveness of
EU law are not compromised.11

It therefore requires a provision in the DAA to make this coordination possible


and manageable. One big argument on this is that a provision in the DAA which
would make this coordinated is intended to bind the Member States—when they
implement EU law—to the Luxembourg Court rather than to their own possible
higher standards of protection. In the view of this book, this coordination already
exists although the intention of the Luxembourg Court is to make this coordination
sealed explicitly, and to finally bring to a standstill any type of solange II decisions
from member states’ courts when they implement EU law. The problem is, how-
ever, that this coordination is something that should belong to the EU law rather
than ECHR or DAA. The Luxembourg Court may not request ECHR bodies or the
Council of Europe to negotiate something which does not appear to be a problem of
ECHR system. Whether EU Member States may provide a higher standard of
protection compared to the Charter—and the effectiveness of the Charter from
that internal perspective—is an ongoing debate for which there may be no blame
put on ECHR or DAA. Should the Masters of the EU Treaties have wished to block
this possibility, they may have very easily established a provision that would say
that no right that is enshrined by the Charter may be interpreted by Member States’
courts differently from the interpretation of the Court of Justice. By the way, even in
the current system, Member States are forced to apply the Charter standard—and no
one may find them liable under EU law if they provide higher standards of
protection—when their implement EU law. Furthermore, requiring the DAA to
coordinate something of this kind leads to an interference to EU law autonomy.
Why? Because it is the Luxembourg Court itself which considers EU law an
independent source of law, which should not be influenced from external sources
such as ECHR system. Making this coordination in DAA—and requesting the
ECHR bodies to provide for this recognition—is equal with an external interference
to the EU legal order, because EU Member States are being bound through the DAA
for something which they are meant to manage under EU law. The author therefore
opines that the Court went beyond the question posed by the Commission—which
merely requested an opinion on the compatibility of DAA with the EU Treaties and
not on how should EU law be legislated through DAA to regulate matters of EU
internal law—this being in itself a mismanagement of the opinion-making process
for the purposes of considering accession’s potential problems from the EU
Treaties’ perspective.

11
Court of Justice of EU, Opinion 2/13, op. cit. note 1, para. 189.
414 11 Before the Conclusion: Luxembourg Court’s Opinion 2/13 on the. . .

11.4 Uniform Interpretation of EU Law and Mutual Trust


Between EU Member States May Not Be Jeopardized
by the Convention System

In its second specific issue, the Court tackles the issue of mutual trust between the
EU Member States when it comes to implementing EU law, especially in problem-
atic policy areas such as freedom, home and justice affairs.12 This trust, the Court
argues, should go as far as each EU Member State should trust that all other member
states have provided the same level of human rights protection, and may not call
that into question.13 Such presumption, the Court argues, ensures that each Member
State puts no doubts as to whether the other member states are under the same roof
when it comes especially to the issue of standard of human rights protection under
EU law. This said, the Court, as regards to the DAA, further argues that:
The approach adopted in the agreement envisaged, which is to treat the EU as a State and to
give it a role identical in every respect to that of any other Contracting Party, specifically
disregards the intrinsic nature of the EU and, in particular, fails to take into consideration
the fact that the Member States have, by reason of their membership of the EU, accepted
that relations between them as regards the matters covered by the transfer of powers from
the Member States to the EU are governed by EU law to the exclusion, if EU law so
requires, of any other law. [. . .] In so far as the ECHR would, in requiring the EU and the
Member States to be considered Contracting Parties not only in their relations with
Contracting Parties which are not Member States of the EU but also in their relations
with each other, including where such relations are governed by EU law, require a Member
State to check that another Member State has observed fundamental rights, even though EU
law imposes an obligation of mutual trust between those Member States, accession is liable
to upset the underlying balance of the EU and undermine the autonomy of EU law.14

With this being the point, the Opinion requires that this development should
somehow be barred from happening, therefore a safeguard needs be put into the
DAA. This point, nevertheless, remains rather weakly argued. The Court needs
merely to see EU and its Member States having one seat—rather than independent
seats—in the ECHR system of contracting parties. It is not, however, clearly argued
how this may be achieved in practice in order to stop potential situations when one
EU Member State may be required to see if another EU Member State has respected
ECHR with the same degree of protection. By the way, in the current pre-accession
stage, this risk stands far more probable than in a post-accession stage with the
current DAA.
With this requirement, the Court wishes to make the EU and its Member States
one contracting party to the Convention.15 Two reasons exist to show why this is

12
Court of Justice of EU, Opinion 2/13, op. cit. note 1, para. 191.
13
Court of Justice of EU, Opinion 2/13, op. cit. note 1, para. 192.
14
Court of Justice of EU, Opinion 2/13, op. cit. note 1, paras. 193–194.
15
Cf.: One may wonder why the EU and its Member State should become one contracting party to
the Convention. This is implied from the Opinion 2/13, which, in para. 194, established that ‘[i]n
so far as the ECHR would, in requiring the EU and the Member States to be considered
11.4 Uniform Interpretation of EU Law and Mutual Trust Between EU Member. . . 415

not feasible: first, EU Member States act as agents of EU law and under its human
rights roof (including the ECHR) only as long as they implement EU law. To bring
the EU Member States into one roof for all human rights competences—possibly
including those not transferred to the EU by its Masters of the Treaty—seems
against the principle of conferred powers. Second, there may exist the possibility
for the EU and its Member States to stand as one contracting party before the ECHR
as regards the EU human rights competences. However, this would go to the
detriment of EU external and internal autonomy of law. In that scenario, the EU
would be barred from demonstrating its legal personality in international law,
moreover that would seem as infringing the capacity of EU to stand as a sovereign
agent under international law with its own regime of law. This situation would in
fact reflect the contrary to the ‘independent source of law argument’ which the EU
Court uses with so much empathy: instead, the EU would be portrayed as a product
of its Member States, and under the roof and control of its Member States.
Therefore, implementing this criterion in practice is rather easy, however it
would merely lead to decreasing the capacity of the EU law to stand as an
independent source of law and the EU holding its legal personality in international
law. By the way, the mere fact that the TEU provides EU with legal personality
contradicts this finding of the Court, or, otherwise, this finding of the Court pushes
Member States to violate the legal personality of the EU to their benefit.
One may also argue that this stance of the Court—as enlightened in paragraphs
193–194—stands rather contrary to the EU’s contractual status in other interna-
tional treaties. One such example is WTO, wherein EU holds the same contracting
status as in the DAA. Moreover, the same risk which the Court refers to in the above
paragraphs may seem as relevant for the EU and Member States’ positions in the
WTO system as well. It is clear therefore that the Court establishes a double
standard in this regard. To this extent, one may legitimately argue that the above
principle of mutual trust—which is also established by the EU Treaties as the
principle of sincere cooperation (see supra Chap. 6 for this risk and the related

Contracting Parties not only in their relations with Contracting Parties which are not Member
States of the EU but also in their relations with each other, including where such relations are
governed by EU law, require a Member State to check that another Member State has observed
fundamental rights, even though EU law imposes an obligation of mutual trust between those
Member States, accession is liable to upset the underlying balance of the EU and undermine the
autonomy of EU law.’ This basically points to a system of accession wherein the EU and its
Member States are not independent contracting parties but rather one contracting party when it
comes to the EU law obligations under the Convention. One may legitimately question why should
that be the case in view of the above statement of the Court? As one can read from the above
statement, the Court simply does not wish EU to be a normal high contracting party in the
Convention system, neither in relation to its members nor in relation to non-EU members as
long as that may imply EU’s standing as independent from its member states. Building upon this
concern, the only means to address this concern would be to provide to the EU and its Member
States one seating in the Convention system wherein neither the EU nor its Member States may
stand as independent contracting parties inter se. There seems no other way to address this
concern.
416 11 Before the Conclusion: Luxembourg Court’s Opinion 2/13 on the. . .

arguments examined), formerly known as loyal cooperation—may be regulated by


internal rules of the EU. By the way, the book clearly argues in many instances
(e.g. see supra the subchapter on the outline of the DAA) that the Court has devised
several safeguards to keep the EU and its Member States into one seat when it
comes to the participation in international treaty systems. The same could have
easily—by analogy—applied in the participation of EU and its Member States in
the DAA as well. That said, it is not rational to make all this textual change in the
DAA—which, by the way, seriously harms the legal personality of the EU—merely
to ensure something that may be simply managed with an internal rule in the
EU law.

11.5 Risks from the Application of Protocol 16 ECHR: Too


Many Doubts Being Raised?

Further to the above points, the Opinion embarks on the potential risks of Protocol
16 ECHR, although DAA contains no provision to accede EU to that separate
protocol. This being the case, the Court goes beyond the question posed by the
Commission in this regard, as it was neither asked nor the DAA foresees any
intention or possibility for the EU to become a contracting party to Protocol
16 ECHR. Tacking the compliance of Protocol 16 ECHR with the EU Treaties
seems rather vague and illogical from the perspective of EU accession to ECHR at
this stage. Two big reasons exist for this. First, the DAA neither establishes nor
allows any space for the EU to become a contracting party to Protocol 16 ECHR.
This question therefore seems brought by the Court to itself on its own initiative.
Secondly, accession to that individual protocol—even after EU would have
acceded to ECHR through the current DAA—should have been done with a
separate accession protocol, which could have again been brought to the Luxem-
bourg Court for an opinion. So, the risk of surpassing the Luxembourg Court after
many years by acceding to Protocol 16 ECHR is absolutely not realistic. Moreover,
it seems rather not unserious to engage with the review of a protocol—and, here,
one must add that there may be an indefinite number of protocols which may be
signed but wherein the EU may not accede without a separate accession proce-
dure—for which the Luxembourg Court has no sign of intention or act for which the
EU would pave the way towards that protocol.
The Court first shows what is Protocol 16 ECHR made for, namely to allow
higher courts of ECHR contracting parties to request the Strasbourg Court to
provide advisory opinions on certain matters relating to the interpretation of
ECHR. According to the Court, the same procedure is also established under EU
law (Art. 267 TFEU) wherein national courts may refer preliminary questions to the
Luxembourg Court relating to EU law issues.16 In the Luxembourg Court’s view

16
Court of Justice of EU, Opinion 2/13, op. cit. note 1, para. 196.
11.5 Risks from the Application of Protocol 16 ECHR: Too Many Doubts. . . 417

the procedure under Protocol 16 ECHR and the one under Art. 267 TFEU overlap,
therefore such development should be of high concern. The Court tries to rational-
ize this view with the following interesting statement:
It is indeed the case that the agreement envisaged does not provide for the accession of the
EU as such to Protocol No 16 and that the latter was signed on 2 October 2013, that is to
say, after the agreement reached by the negotiators in relation to the draft accession
instruments, namely on 5 April 2013; nevertheless, since the ECHR would form an integral
part of EU law, the mechanism established by that protocol could — notably where the
issue concerns rights guaranteed by the Charter corresponding to those secured by the
ECHR — affect the autonomy and effectiveness of the preliminary ruling procedure
provided for in Article 267 TFEU.17

It is of course not likely that Protocol 16 ECHR could in any way produce effects
towards the EU as long as the latter does not accede to it. It is too surprising to read
the Court saying that it may—even though it does not say ‘how’—affect the
autonomy of EU law if EU itself is not a party to the protocol. In addition, the
advisory opinions under Protocol 16 ECHR (Art. 5) are not obligatory and therefore
have no legal effect (neither erga omnes nor inter partes). The comparison between
the two procedures made by the court is moot, because under the preliminary
reference procedure the Luxembourg Court adopts legally effective opinions.
However, should the EU 1 day accede to Protocol 16 ECHR, then, the argument
should be that it may interfere to the EU Court’s exclusive jurisdiction to guide with
abstract opinions on the application of ECHR. The argument for that, however, may
not be of the form which the Court shows above. The interference to EU law
autonomy would not be far higher in that case than merely a competition with the
preliminary ruling procedure under EU law. By the way, it would not change
anything even if EU does not accede to ECHR, as EU Member States will still
accede to Protocol 16 ECHR on their right and therefore make their courts available
to the Strasbourg Court. Even with Opinion 2/13 in force and no accession in place,
the EU Member States’ courts will still have the perfect opportunity to surpass the
preliminary reference procedure under EU law by requesting advisory opinions
from the Strasbourg Court. Hence, the Luxembourg Court tries to stop with this
statement something it may not, even if it finds DAA violating the EU Treaties.
Such strategy of the Luxembourg Court seems rather poor and not realistic in the
target it tries to meet.
To mention, there are many international courts which do have advisory mech-
anisms. International Court of Justice is one of them. To prohibit the EU to
participate in international treaties because such treaties may provide for an inter-
national court with advisory functions is not realistic either. With this reasoning, the
Luxembourg Court will 1 day argue that EU may not accede to the UN Charter
because ICJ may be asked by the General Assembly (where EU Member States
participate) on certain advisory opinion, which then may overlap with the prelim-
inary ruling procedure under EU law. This reasoning leads nowhere. Therefore,

17
Court of Justice of EU, Opinion 2/13, op. cit. note 1, para. 197.
418 11 Before the Conclusion: Luxembourg Court’s Opinion 2/13 on the. . .

even though that may be a sort of competition, the Luxembourg Court would have
better strategized if it banned EU Member States requesting opinions relating to EU
law from the Strasbourg Court under Protocol 16 ECHR (where EU may accede one
day), rather than making it fully incompliant with the EU Treaties. The argument
goes, it is the result—making it not practical for the EU Member States to request
advisory opinions from the Strasbourg Court under Protocol 16 ECHR—which
should be the intention of the Luxembourg Court (for which I would agree), rather
than the formal construction of a system which is normal under international law.

11.6 Art. 344 TFEU in Risk from the DAA: What About
Excluding Inter-Party Mechanism?

The fourth substantive issue which the Luxembourg Court tackles in Opinion 2/13
is the inter-party mechanism under the DAA. It chooses a tactic of ruling that
simply confirms what this book argues in Chap. 7 (see supra chapter on inter-state
mechanism). However, for practical reasons the way the Luxembourg Court pro-
nounces on this issue is rather strange and not well substantiated. As well known,
Art. 344 TFEU establishes that EU Member States undertake not to submit any
dispute relating to the interpretation of EU law to an external court or tribunal. As
explained in Chap. 7, infra, this may contravene with the ECHR inter-party
mechanism. This is so because post-accession ECHR becomes part of EU law,
therefore bringing a dispute for the interpretation of ECHR to the Strasbourg Court
under the inter-party mechanism would amount to violating Art. 344 TFEU. As
Protocol 8 to the EU Treaties specifically states that the DAA should not interfere to
Art. 344 TFEU, the Court builds upon this with a rather firm approach although this
book argues that there exists a constructive scenario which may surpass this
problem (see infra Chap. 7). The Court argues in Opinion 2/13 that:
[. . .] the fact that Member States or the EU are able to submit an application to the ECtHR is
liable in itself to undermine the objective of Article 344 TFEU and, moreover, goes against
the very nature of EU law, which, as noted in paragraph 193 of this Opinion, requires that
relations between the Member States be governed by EU law to the exclusion, if EU law so
requires, of any other law. [. . .]In those circumstances, only the express exclusion of the
ECtHR’s jurisdiction under Article 33 of the ECHR over disputes between Member States
or between Member States and the EU in relation to the application of the ECHR within the
scope ratione materiae of EU law would be compatible with Article 344 TFEU.18

The requirement of the Luxembourg Court in this regard seems to be of a high


bar. It requires that the DAA call inadmissible ratione materie all ECHR-EU
related applications. The problem is wider than the DAA itself. First of all, Art.
344 TFEU may not be read as something that needs to be reflected in international
agreement wherein EU becomes a party. It is merely a prohibition of result, rather

18
Court of Justice of EU, Opinion 2/13, op. cit. note 1, paras. 212–213.
11.7 Two Small Threats from the Co-respondent Mechanism That Need Be Addressed 419

than a prohibition of EU becoming part of treaty systems which provide for such
mechanisms. It would suffice for Art. 344 TFEU if the result of Member States not
embarking to another international court is reached, independent of the fact whether
there exist or not formal routes that may in theory be used. Should it have been so,
the EU should have not become member of e.g. the WTO (which has a similar
dispute-resolution—inter se—mechanism). Art. 344 TFEU is merely made to serve
as an internal safeguard that prevents Member States—from an internal constitu-
tional point of view—to initiate external dispute settlement procedures inter se.
Therefore, the Luxembourg Court’s pronunciation tries to make Art. 344 TFEU as
something that prohibits EU to accede to instruments of international law which
possess inter-party complaint mechanisms. With this logic, EU should have been
prevented to accede to UNCLOS as well, but should also be prohibited to accede to
UN (whose court, the ICJ possesses an inter-state jurisdiction). Second, to request
that the DAA call all these cases inadmissible amounts to an external interference to
EU law as well. The argument goes, the Court is requesting that the DAA support
the broad-scope interpretation of Art. 344 TFEU, something that would amount to
an influence of DAA in the interpretation of the EU Treaties.

11.7 Two Small Threats from the Co-respondent


Mechanism That Need Be Addressed

The fifth substantive issue raised in Opinion 2/13 is the co-respondent mechanism,
this being the most important innovation of the DAA. The Court does not become
so unenthusiastic on this mechanism.19 The claims of the Court expressed in the
Opinion, as regards the co-respondent mechanism, do merely confirm some of the
arguments presented by this book in Chap. 6 (infra). By the way, this book finds
fare more loopholes compared to the Court. The Court identifies two problems
which the co-respondent mechanism may provoke in face of the autonomy of EU
law. By analyzing the modes via which the co-respondent mechanism may be
triggered, the Court basically argues that it has no problem with the design of the
procedure whereby a High Contracting Party is invited by the Strasbourg Court to
join as co-respondent.20 However, the Court is concerned when this procedure
involves an application of the EU or its Member States to join as co-respondent
(s), in which moment the Strasbourg Court is provided with jurisdiction to assess
whether such applications are ‘plausible’. Such jurisdiction may pull the Strasbourg
Court into EU law, as it need to assess whether the claims of the co-respondent
applicants—which are based in EU law—are real or not (see supra Chap. 6 for a far

19
Cf.: Gragl (2013), p. 279, who rightly advises that there need be adopted detailed rules in EU law
to address the loopholes and potential gaps that may harm the EU law autonomy with regard to the
operation of the co-respondent mechanism.
20
Court of Justice of EU, Opinion 2/13, op. cit. note 1, para. 222.
420 11 Before the Conclusion: Luxembourg Court’s Opinion 2/13 on the. . .

deeper examination).21 This finding of the Court confirms the book’s arguments in
relation to the ambiguities of the co-respondent mechanism, which are examined
profoundly in infra Chap. 6.
To this end, the Court argues that:
Admittedly, in carrying out such a review, the ECtHR is to ascertain whether, in the light of
those reasons, it is plausible that the conditions set out in paragraphs 2 and 3 of Article 3 are
met, and that review does not relate to the merits of those reasons. However, the fact
remains that, in carrying out that review, the ECtHR would be required to assess the rules of
EU law governing the division of powers between the EU and its Member States as well as
the criteria for the attribution of their acts or omissions, in order to adopt a final decision in
that regard which would be binding both on the Member States and on the EU.22

Therefore, the Court concludes that with this loophole in place the DAA is liable
to interfere with the EU law autonomy. The author of this book stands in the same
line of reasoning. However, for practical reasons, the only mechanism to address
this concern would be to provide High Contracting Party with the right of automatic
intervention as co-respondent(s) whenever they consider that the criteria have been
met. This approach may lead to the possibility that every High Contracting Party
join as co-respondent(s) when they wish, something which makes it not possible to
check whether the criteria for joining as co-respondent(s) have been met. This said,
it remains problematic in practice to exclude the Strasbourg Court from the logic
that co-respondent(s) should be reviewed if they have satisfied certain criteria to
join the procedure.
The second concern of the Court as regards the co-respondent mechanism is the
model of mutual liability which the DAA installs in cases when co-respondent(s)
have joined the procedure. The Court argues that the DAA relevant:
[. . .] provision does not preclude a Member State from being held responsible, together
with the EU, for the violation of a provision of the ECHR in respect of which that Member
State may have made a reservation in accordance with Article 57 of the ECHR. [. . .] Such a
consequence of Article 3(7) of the draft agreement is at odds with Article 2 of Protocol No
8 EU, according to which the accession agreement is to ensure that nothing therein affects
the situation of Member States in relation to the ECHR, in particular in relation to
reservations thereto.23

Here, the author of this book argues that the Court makes a big mistake. Why?
The fact that EU Member States are subject to a Strasbourg Court decision which
attributes to them mutual liability—and this argument applies for the majority of
cases—reflects the EU Member States participation therein from their capacity as
agents of EU law—in which case they represent the jurisdiction of the EU compe-
tences upon which the ECHR applies;24 therefore, their reservations made at their

21
Court of Justice of EU, Opinion 2/13, op. cit. note 1, paras. 221–225.
22
Court of Justice of EU, Opinion 2/13, op. cit. note 1, para. 224.
23
Court of Justice of EU, Opinion 2/13, op. cit. note 1, paras. 227–228.
24
Cf.: One may rightly question whether the opposite may be true as well, given that the above
argument would push the Member States outside their sovereign territory even when they act as
agents of EU law and therefore outside their constitutional requirements as well. I tend to disagree
11.7 Two Small Threats from the Co-respondent Mechanism That Need Be Addressed 421

non-EU capacity are not relevant when they are subject to a decision on mutual
liability for a violation originating in EU law. To make this even simpler, let me
make a question: Do EU Member States—which are subject to a Strasbourg
decision on mutual liability for a violation originating in EU law—need to hold
responsibility for human rights under the ECHR protocols to which the EU is not a
party? The answer is no, because they have joined therein merely in their capacity
as EU-law implementers (agents), not as individual High Contracting Parties
independent of the EU (on their own, non-EU individual ECHR jurisdiction).
Therefore, this point of the Court seems legally and rationally moot.
The third concern of the Court is the possibility that exists for the Strasbourg
Court to depart from the general norm of mutual responsibility, and to find only one
of the co-respondent(s)—after having sought their views—liable.25 The Court
opines that Strasbourg Court will be given the opportunity to enter into the EU
law borders whilst it needs to check the requests of the co-respondent(s) for finding
only one of them as liable. The Court argues that it is not clearly written in the DAA
whether there needs that both parties make such request to the Court or if it suffices
for only one of them to make that request.26 To that end, the Court further argues
that:
In any event, even it is assumed that a request for the apportionment of responsibility is
based on an agreement between the co-respondent and the respondent, that in itself would
not be sufficient to rule out any adverse effect on the autonomy of EU law. The question of
the apportionment of responsibility must be resolved solely in accordance with the relevant
rules of EU law and be subject to review, if necessary, by the Court of Justice, which has
exclusive jurisdiction to ensure that any agreement between co-respondent and respondent
respects those rules. To permit the ECtHR to confirm any agreement that may exist between
the EU and its Member States on the sharing of responsibility would be tantamount to
allowing it to take the place of the Court of Justice in order to settle a question that falls
within the latter’s exclusive jurisdiction.27

with such claim and defer to the above statement made in the main text. First, in principle, Member
States would never find themselves in a decision of the Strasbourg Court placing mutual respon-
sibility (resulting from the use of the co-respondent mechanism) on them and the EU unless they
had acted in furthering an EU law obligation without exercising state discretion. In that capacity,
the Member State(s) have been acting as agents of EU law and undertaking actions on behalf of the
EU (and at its cost; DARIO may help here as well). Second, I also argue that if one would follow
the Court’s argument and install a system that addresses that difference, it would run counter to the
principle of conferred competences in EU law. If Member States were to be considered as acting
within their constitutional sovereign competence even when they act as agents of EU law in
implementing obligations without state discretion that would basically go to the detriment of the
principle of conferred competences (which stands as a core concept of EU law as an independent
source of law). Third, if Member States were recognized the right stemming from the Court’s
concern, then they would even violate their constitutional principles on international responsibil-
ity, which basically prohibit that member states respond for the conduct and wrongdoings of other
entities of international law (like the EU). This line of reasoning may be also drawn with regard to
the ILC Articles on State Responsibility. This being said, it is noted here that the arguments
presented in the main text seem valid.
25
Court of Justice of EU, Opinion 2/13, op. cit. note 1, para. 229.
26
Court of Justice of EU, Opinion 2/13, op. cit. note 1, para. 233.
27
Court of Justice of EU, Opinion 2/13, op. cit. note 1, para. 234.
422 11 Before the Conclusion: Luxembourg Court’s Opinion 2/13 on the. . .

Therefore, the Court considers that the co-respondent mechanism—as it stands


in the DAA—may be operated in a scenario that could contravene the EU Treaties.
This finding of the Court merely confirms the arguments of this book in Chap. 6 (see
supra), although the book finds many more reasons to prove this finding. Therefore,
there is no need for further analysis on this finding in this subchapter.

11.8 Prior Involvement Mechanism: The Need to Add


Another Layer of Safeguard

The next substantive issue of concern to the Luxembourg Court’s Opinion 2/13 is
the operation of the prior involvement mechanism. It generally agrees with this
mechanism, but is concerned with the fact that the DAA wording of ‘“[a]ssessing
the compatibility of the provision” [does] mean, in essence, to rule on the validity of
a legal provision contained in secondary law or on the interpretation of a provision
of primary law.’28 The Court considers that it has been given jurisdiction to
interpret EU primary law—for which it agrees—and merely to check the validity
of EU secondary law, but not to interpret the latter. To that extent, it further argues:
If the Court of Justice were not allowed to provide the definitive interpretation of secondary
law, and if the ECtHR, in considering whether that law is consistent with the ECHR, had
itself to provide a particular interpretation from among the plausible options, there would
most certainly be a breach of the principle that the Court of Justice has exclusive jurisdic-
tion over the definitive interpretation of EU law. [. . .] Accordingly, limiting the scope of the
prior involvement procedure, in the case of secondary law, solely to questions of validity
adversely affects the competences of the EU and the powers of the Court of Justice in that it
does not allow the Court to provide a definitive interpretation of secondary law in the light
of the rights guaranteed by the ECHR.29

Therefore, the Court requests that the provision regulating this mechanism
include the term ‘interpretation’ for secondary law as well. The author of this
book opines that the Court has misunderstood the terms of the prior involvement
mechanism, and there exist two huge reasons why the Court should have not
followed this line of interpretation. First, when the DAA Explanatory Report refers
to the fact that the prior involvement mechanism allows the Luxembourg Court
either to interpret EU primary law or to check the validity of EU secondary law, it
does that with the intention of making the difference—for the general reader—of
the degree of jurisdictional difference that the Luxembourg Court has in relation to
the two types of EU law: primary and secondary law. Therefore, by advising that
the Luxembourg Court will check the ‘validity’ of EU secondary law, the DAA
does not prohibit the latter to interpret it as well. Of course, a court that has the
jurisdiction to check the validity of a certain norm it does also possess the

28
Court of Justice of EU, Opinion 2/13, op. cit. note 1, para. 242.
29
Court of Justice of EU, Opinion 2/13, op. cit. note 1, paras. 246–247.
11.9 CFSP Measures Before the Strasbourg Court: Luxembourg Court in a Panic 423

jurisdiction to interpret that norm. The fact that the DAA mentions the jurisdiction
of the Luxembourg Court to check the validity of an EU secondary law norm is
merely to show what it can do at maximum, because, of course, it can do everything
else below that bar. By the way, it would not be possible for the Luxembourg Court
to decide on the validity of a certain EU secondary law norm without having
interpreted it first. Second, and a bigger mistake than the previous one, the Lux-
embourg Court sees the DAA as something that may attach or shape its jurisdiction.
Accepting this feature to the DAA would be tantamount to an interference to EU
law external autonomy. If the Luxembourg Court insists that there is no interna-
tional agreement that may fix or change the competences of EU institutions, then,
how come that it considers that the DAA is limiting to it a competence that it
already has on basis of the Treaties? Therefore, the Court’s argument in this regard
seems to propose that the DAA should give to the Luxembourg Court a jurisdiction
which it does not have and that it should intervene into EU law to fix this problem
because—otherwise—the Luxembourg Court will remain without that competence.
Most logically, the Luxembourg Court should have merely pronounced that—in
order to retain its consistency and to keep preserving its autonomy—the DAA
merely guides the public on the type of jurisdictions which the Luxembourg Court
possesses, as the DAA may not—even if it would wish—prescribe and/or limit a
competence which the Luxembourg Court possesses on basis of the Treaties. This is
why the way the Court has made the argument for this part of the Opinion seems
both poor and lacks credence.

11.9 CFSP Measures Before the Strasbourg Court:


Luxembourg Court in a Panic

The last substantive point which the Court dealt with in Opinion 2/13 is the
reviewability of CFSP acts by the Strasbourg Court. It started the analysis by taking
note of the fact that—as the current treaties stand—it has no jurisdiction to review
certain (the large majority) of acts and omissions undertaken under the CFSP policy
area.30 This being noted, the Court argues that it does not wish to see someone else
having jurisdiction on this area without it having had the chance to review it first.
Therefore, the Court argued:
[. . .] on the basis of accession as provided for by the agreement envisaged, the ECtHR
would be empowered to rule on the compatibility with the ECHR of certain acts, actions or
omissions performed in the context of the CFSP, and notably of those whose legality the
Court of Justice cannot, for want of jurisdiction, review in the light of fundamental rights.
[. . .] Such a situation would effectively entrust the judicial review of those acts, actions or

30
Court of Justice of EU, Opinion 2/13, op. cit. note 1, para. 252.
424 11 Before the Conclusion: Luxembourg Court’s Opinion 2/13 on the. . .

omissions on the part of the EU exclusively to a non-EU body, albeit that any such review
would be limited to compliance with the rights guaranteed by the ECHR.31

Therefore, the requirement put forth by the Luxembourg Court in this regard is
that the Strasbourg Court may not have jurisdiction on certain aspects of CFSP acts
as long as it is not itself authorized to do so under the Treaties. This finding merely
supports the arguments of this book on the status of CFSP acts which become
reviewed by the Strasbourg Court as a court of first instance (see supra the chapters
on attribution of liability and inter-party complaint procedure). Therefore, there is
nothing new in this regard. However, the Court should take note of the fact that,
nevertheless, the Strasbourg Court’s review is of an international nature, as it
originates and derives from international law. It has no direct internal effect in
EU law, in principle. Therefore, making the argument that the Strasbourg Court
should not possess an external competence to review certain acts of EU law as long
as Luxembourg Court itself has not been authorized to review them on internal
basis seems as a means of conditioning the nature of international law. On the other
hand, this claim of the Luxembourg Court supports the call for interference to EU
external autonomy: because, if the DAA will exclude CFSP acts from the jurisdic-
tion of the Strasbourg Court, then, the latter would need to enter into EU law—
interpret it—every time it reviews and act or omission originating in EU law, in
order to check if that act or omission derives from CFSP—which it is a precondition
for it to declare on its admissibility. Therefore, the proposal of the Luxembourg
Court made in this paragraph does encourage the interference to EU law autonomy
instead of seeking to seal it.

11.10 The Technical Changes That Need Be Addressed


in the Draft Accession Agreement to Make It
Compliant with Opinion 2/13 Requirements

As the Court’s opinion makes it still possible for the vast majority of the DAA
provisions to remain in compliance with the EU Treaties, the author opines that
there are some inconsequential further add-ons and revisions that need be addressed
to make the DAA compliant with the Opinion 2/13 requirements, as follows:
1. The DAA must provide for an additional provision to guarantee the coordination
of Art. 53 of the Charter with Art. 53 of the Convention. In light of the arguments
presented by the Court in two paragraphs, such concerns may be isolated if DAA
were to be added this kind of provision in Art. 1 (8): ‘The EU Member States
undertake not to use the Convention minimum standard of human rights protec-
tion as means to refuse or omit from recognizing the primacy of Charter rights
and its effectiveness in their internal legal orders. Should Member States use any

31
Court of Justice of EU, Opinion 2/13, op. cit. note 1, paras. 254–255.
11.10 The Technical Changes That Need Be Addressed in the Draft Accession. . . 425

kind of such argument, the Court of Justice of EU is provided with jurisdiction to


hear a case and decide upon it by directing the Member State(s) concerned.’
2. The DAA must provide for an additional provision to put the EU and its Member
States into one seat of contracting status when they represent the EU law. This
may be addressed with the following additional provision that would be attached
to Art. 1 (6) of the DAA: ‘The EU and its Member States have the status of one
contracting party in the ECHR system when their acts or omissions originating
from EU are in question’.
3. The DAA must add a specific provision to prohibit the potential clash of the EU
Treaties with Protocol 16 ECHR. Such specific provision in the DAA may read
as following: ‘The EU may not accede to Protocol 16 ECHR’. This type of
provision, on the contrary, will be read as something delimiting from outside the
contracting capacity which the EU Treaties provide to the EU, therefore inter-
fering to the competences which have already been conferred to the EU. The
latter is a big reason why the Court’s argument on Protocol 16 ECHR was
faultily managed and strategized.
4. The DAA must add to Art. 5 DAA the following provision to comply with the
requirement of the Opinion on Art. 344 TFEU. Such provision may seem as
follows: ‘The Strasbourg Court will, after an application passes the locus standi
admissibility test, call inadmissible ratione materie all applications submitted
under the inter-party complaint mechanism which involve an issue of ECHR that
reflects EU law as well.’
5. To address the Opinion 2/13 requirements with regard to the co-respondent
mechanism, Art. 3 DAA should revise some of its mechanisms, as following:
(a) in the first place, DAA should be included a sort of provision that makes the
desire to join as co-respondent(s) before Strasbourg Court as legally effectual
without any review jurisdiction of the latter. Such provision may be of this kind:
‘The EU Member States will join, according to their own assessment whether the
criteria have been met, as co-respondent(s) in a case in which the EU is the
original respondent. The decision of the EU Member State(s) binds the Stras-
bourg Court.’ (b) in the second place, to address the problem of asymmetrical
reservations, the DAA may be added this kind of provision: ‘Strasbourg Court
will take due account of cases of mutual responsibility if one of the parties
therein may become liable for a right which is protected by a protocol to which it
has put a reservation. In such cases, the Strasbourg Court will invite the relevant
co-respondent to withdraw from the procedure for the part of the right for which
it does not hold any kind of liability due to the submitted reservation.’ (c) finally,
to address the possibility that two or more co-respondent(s) divorce in the
procedure, the DAA must change its current provision on this issue either by
deleting it or revising it to appear in this format: ‘Once two or more parties to a
case (as co-respondents) join the procedure, they may not—even if one of them
is or may seem not liable for the act or omission on basis of which the case was
filed to the Court—withdraw from the procedure. The Strasbourg Court will find
all of them mutually responsible for the alleged violation, whereas it is an issue
426 11 Before the Conclusion: Luxembourg Court’s Opinion 2/13 on the. . .

of internal EU law and EU Court how they will apportion their responsibility
internally on that matter.’
6. To address the concern with the prior involvement mechanism, the Explanatory
Report to DAA—in its paragraph 66—should add the word ‘interpretation’ of
EU secondary law along the word ‘rule on validity’. Again, as argued above, this
type of provision would instead prove as interfering to the division of compe-
tences within the EU, something which Protocol 8 to the EU Treaties has
prohibited. Therefore, one should note that it is immature to see the Luxembourg
Court asking for such kind of provision.
7. Finally, to address the concern on CFSP acts, the DAA in Art. 1 (3) should add
another sentence, reading: ‘Notwithstanding this, the Strasbourg Court is
excluded from the jurisdiction to review EU acts and omissions deriving from
its CFSP policy area’. This will mean that the ECHR jurisdiction will still cover
CFSP acts, however the Strasbourg Court will not have any jurisdiction to
review them. This kind of provision, one should repeat, opens way for the
Strasbourg Court to interfere to EU law autonomy, as it will need to assess
every act and omission originating in EU law whether it derives from CFSP or
not—in order to declare its jurisdiction on it. Such manoeuvre—although
strangely requested by the Luxembourg Court—would amount to obvious inter-
ference to EU external autonomy.

11.11 A Closing Summary

The chapter carefully examined the latest legal development in this regard, namely
Luxembourg Court’s Opinion 2/13; the latter mainly supports the findings of this
book in many regards. However, it is here concluded that the Opinion 2/13
demonstrates that the Luxembourg Court is not very pragmatic when it comes to
the need for the EU to penetrate more in-depth in international law structures,
especially those of a human rights law nature. This ‘jealousy’ of the Luxembourg
Court may pose certain costs to the human rights position of the EU. Several other
issues were examined in the chapter, therefore demonstrating that the language
expressed by the Luxembourg Court was not only harsh but may bring practical
problem to the accession project in general.

References

Gragl P (2013) The accession of the European Union to the European Convention on Human
Rights. Hart Publishing, Oxford
Peers S (2014) The CJEU and the EU’s accession to the ECHR: a clear and present danger to
human rights protection. Blog Article. Available at: http://free-group.eu/2014/12/18/steve-
peers-the-cjeu-and-the-eus-accession-to-the-echr-a-clear-and-present-danger-to-human-
rights-protection/
Chapter 12
An Overall Conclusion

12.1 A General Overview

The book examined intensely the core implications resulting from the competition
or cooperation between the Strasbourg and Luxembourg courts in light of the EU
accession to ECHR. Answering to the big and subsidiary research questions, it is
generally concluded that the EU accession to ECHR will affect significantly the
current dialogue between the two systems of law and their model of coexistence. It
is therefore generally concluded that the EU accession to ECHR will prove impor-
tant not only for symbolic reasons, but also for legal and practical reasons that affect
and determine the human right position of persons in the EU and its Member States.
The in-depth examination of the effects that this process will bring proves the
implicative legal relationships that exist between competing international courts’
jurisdictions in general, and EU and Convention autonomous approaches to
their laws.
While the EU accession to ECHR involves certain modifications on the laws of
both sides, it is important to mention that the accession as such is meant not only to
provide for a normal accession of another High Contracting Party to the Convention
system but also to provide for a special model of accession, one which is intended—
at least for symbolic reasons—to keep the EU as ‘peculiar’ as it is currently. The
intention to uphold the peculiar nature of the EU law sometimes—if not often—was
traded off with certain mechanisms which feature the Convention’s nature of
effectiveness in human rights protection. Such tradeoffs, the book concludes,
sometime question the primary meaning of the accession process, while trying to
preserve the key normative elements for which the Convention does in fact exist. It
is therefore generally concluded that the DAA will prove for a peculiar relationship
between the EU and Convention regimes of law—including the peculiar nature of
relationship between the Strasbourg and Luxembourg courts—a conclusion which
certainly makes the point that nevertheless the EU becomes a bit more ‘equal’ than
the other state High Contracting Parties to the Convention. Whether there is any

© Springer International Publishing Switzerland 2015 427


F. Korenica, The EU Accession to the ECHR, DOI 10.1007/978-3-319-21759-8_12
428 12 An Overall Conclusion

reason for that unequal equality, one should say that it is concluded that there is no
such reason excluding the need of the EU to prove its external autonomy with
robust characterizations.
The book also points to the fact that the accession process as such will experi-
ence many theoretical innovations which have never been tested in practice not
only in the EU relationship with international tribunals, but also between the
international tribunals at large. The theoretical nature of the many of the DAA’s
mechanisms being yet untested, the book concludes generally that the accession of
EU to ECHR will provide for another layer of experience for the Luxembourg
Court to either redesign its external autonomy or to improve its legitimacy so as to
be able to stand firmly against the Strasbourg regime of law. In this context, it is
concluded that the EU would need to improve many of its current judicial practices
in order to make itself better equipped to consume the substantive and procedural
safeguards which the Convention system envisages.
On the other hand, the book concludes generally that a sounder system of
dialogue between the two courts need be introduced, something that will not
anymore put the burden on the Strasbourg Court to make the ‘back steps’. In a
new legal setting, it is concluded that the Strasbourg Court will be given the
opportunity to improve its positions in face of the EU with the substantive human
rights standards which it should apply, clearly paying the consideration to the
Luxembourg Court as a domestic constitutional court for EU but nothing beyond
this. As such, the general conclusion is that the upcoming relationship between the
Strasbourg and Luxembourg courts will be one which will give more products on
the substance of rights adjudicated rather than provide food for symbolic deference
to each other.

12.2 Conclusion on the Overall Functionality


of the Accession Agreement and Its Outputs

EU accession to ECHR ought to be packed as a particularistic development in


international law, not harming the hierarchy of international law per se. To this
extent, the book concludes that accession will unquestionably fill in the gaps in the
present human rights protection architecture in Europe. It will ensure that all EU
citizens are assigned to the same degree of human rights protection although they
may be under different layers of state and supranational systems of governance. The
gap-filling function of the accession will also build upon the needed coherence of
law in Europe, but also contribute to positioning the EU law into a non-fragmenting
seat in the Pan-European law setting. This will altogether harmonize the human
rights protection between European-international layer and EU constitutional layer
of human rights law, integrated within the framework of EU Member States’
contracting stipulations as well. The amalgamation of these two regimes of law
will be integrated through the attitude of Convention being considered a universal
12.2 Conclusion on the Overall Functionality of the Accession Agreement and Its. . . 429

human rights instrument of law whereas EU law being considered a particular law
in that regard. This development will likely prove right Helfer & Anne-Marie
Slaughter’s theoretical observation that if this kind of European project would be
replicated in other continents it may well ‘enhance the compliance opportunities for
international law in general and for international adjudication in particular.’ (see
supra Chap. 1)1
Bruno de Witte—in this context—essentially argues that EU accession to ECHR
will also bring the question of EU’s participation and accession to other interna-
tional human rights instrument.2 Such ‘new perspective’ will likely make the EU
gradually better embark to other international human rights law instruments, but in
the same time will also open ways to untie conventional human rights treaties from
offering place to regional international organizations like the EU. This double-way
development will likely contribute to the advancement of international law through
the lens of a growing institutionalized legal pluralism in Europe and beyond.
From a more focused perspective—in the context of specific effects of the
accession process—the book basically embraced the line of reasoning that demon-
strated that the adoption of the Charter does not make EU accession to ECHR
irrelevant and unwelcome. This held, the book basically proved that AG Jacobs
argument that EU accession to ECHR should be taken for symbolic reasons3 as
inaccurate.4 Instead, the book concludes that EU accession to ECHR—from the

1
Helfer and Slaughter (2005), p. 8.
2
de Witte (2014), p. 351/2.
3
Jacobs (2007).
4
One may legitimately question whether there is any practical case that would demonstrate that
there are divergencies in the scope of protection between the ECHR standards and EU law
fundamental rights standards (or merely accession is being undertaken for symbolic reasons).
The author would like to bring two examples: first, the right to vote and the prohibition of
discrimination. The right to vote under EU law is projected to guarantee one’s articulation of
will only with regard to the European Parliament. Considering the Sejdic & Finci case of the
Strasbourg Court, the EU may likely fall within the border of violation of Art. 3 Protocol 1 to the
ECHR (in conjunction with the prohibition of discrimination). Under the latter case standards, the
right to vote should be guaranteed in face of the EU legislature, which also includes the
Commission and the Council. The right to vote under the EU law seems not only limited but
also supporting certain member state-related affiliations, which, in light of the Sejdic case, may
seem as violating the Convention (on this, see: Korenica and Doli 2015). Second, one may take the
example of the compliance with ECHR standards of competition fines which the Commission
applies on enterprises. Such fines, which in light of the Convention standards fall under Art.
6, should in principle have to be imposed by an independent court. This fact alone may bring
claimants under the competition law procedures in the EU outside the guarantees of the latter
ECHR provision. ‘The uncertainty alluded to above as to whether EC competition law proceedings
could be considered as involving a “criminal charge” within the meaning of Article 6 ECHR has
stemmed from the fact that the EC law’s domestic classification of sanctions imposed by the
Commission for breaches of Articles 101 and 102 TFEU is explicitly non-criminal. The wording of
Article 23(5) of Regulation 1/2003 sets out that the decision by which the Commission imposes a
fine on undertakings “shall not be of a criminal nature”. However, according to the case-law of the
ECtHR, such classifications under domestic law as to the criminal nature of the offence have only a
“relative value”’ (Talbot 2014). With the EU accession to ECHR, these two substantive human
430 12 An Overall Conclusion

structural point of view—may largely improve not only the current structural
deficiencies which exist in the human rights protection in the EU but also the
substantive situations regarding human rights protection of EU-related claimants
appearing before the Strasbourg Court. The valid conclusion therefore is that EU
accession to ECHR will prove rather substantive for the human rights interaction
and integration between the laws of the two legal regimes—although not at the
desired level. The accession, and the upcoming integrative dialogue between the
two courts, will likely prove right Barber’s theoretical conclusion that ‘law has
played a powerfully integrating role in Europe’.5
It is also concluded that accession formally pays respect for EU competences,
but this may not always be practically the case. In this regard, the book concludes
that it was the intention of the Lisbon policy-makers to allow some sort of
penetration of Convention law directly to the EU Treaties in the positive sense.
This being concluded, it must be asserted that the coexistence of EU law and
ECHR—and the latter’s incorporation to the EU Treaties—does not per se harm
the autonomy of EU law, rather makes it more compliant and accommodated with
the EU’s international human rights law obligations. At the end of the day, the book
concludes that this has been the will of the Masters of Lisbon Treaty—so at least the
teleological interpretation of the relevant provisions demonstrates. In this regard,
building upon Han’s theoretical observation (examined supra Chap. 1), one may
conclude that the accession project may seem a good opportunity to address the
resolution of normative tensions in the European continent from the perspective of
integration through the definition of roles between universalism and particularism.
The book, on the other hand, concludes that the formal equal position of the EU
as a High Contracting Party to the Convention will make the picture of Convention
parties more appropriate as to the intended objectives of securing rights. However,
substantively, the book clearly demonstrates that EU is rather privileged within the
Convention system, one with regard to the model of attribution of liability when
both the EU and its Member States appear as (co-)respondents, and, two, regarding
the use of the prior involvement mechanism by the EU Court. External supervision
from the side of the Strasbourg Court—from the position of the claimant—will be
one of equal treatment compared to the respondent positions of other state High
Contracting Parties. In this regard, the external control of human rights perfor-
mance of the EU from the Strasbourg Court will convey ‘real added value in
relation to purely national supervision exercised “from within”.’6 Acceptance of
the external control from the Strasbourg Court will also demonstrate EU’s credi-
bility to undertake internationally supervised obligations on human rights, which
are nowadays fundamental conditions for developed states.

rights’ issues may start gradually improving. The same may be said with regard to the right to
access the court and effective remedy under EU law.
5
Barber (1993), p. 137.
6
Callewaert (2014), p. 15.
12.2 Conclusion on the Overall Functionality of the Accession Agreement and Its. . . 431

Callewaert—for instance—rightly noted that there was a practice of ‘cross


references between Luxembourg and Strasbourg [that] were legion and bore clear
witness to the existence of a common heritage of fundamental rights shared by the
“two Europes”.’7 Accession will fuse such two ‘Europes’ into one human rights law
‘Europe’. Therefore, regarding coherence of human rights in Europe, it is con-
cluded that accession will certainly make both legal orders more coherent in terms
of their minimum human rights protection standard. Such coherence, the book
concludes, does not necessarily mean full uniformity of human rights protection
between the two legal orders, as the EU is provided the authority through its Charter
to offer broader protection than the minimum standard of the Convention.
The book concludes that the principle of subsidiarity will start functioning in
regards to the EU as a High Contracting Party as well. Its nature and scope,
however, will considerably vary from the context and point of view. With a certain
degree of clarity, it is concluded that the nature and scope of subsidiarity of the
Strasbourg Court will differ in face of EU compared to other state High Contracting
Parties. This is just another signal of the substantive distinctiveness of the EU’s
‘equal’ position in the Convention system. On the other hand, the book concludes
that there is a rather complex set of human rights law instruments in Europe, such
density of law needing special consideration and organizational management to
become ordered and coordinated determinedly. EU accession to ECHR makes the
structure of human rights law in Europe even denser. This being the problem, the
book concludes that the regulation of the relationships between these types of law
which subsist in the same environment needs vigilant analysis and is of an immense
need. The book therefore argues that although the European human rights law
structure becomes denser it nevertheless is ordered better with the accession
project, such ordering having a core role in institutionalizing the relationship
between previously rather competing regimes of human rights law. This proves
valid Han’s advice that it ‘is necessary to cope with the increasing trend of
decentralization and the growing complexity of an international society.’8 It is
therefore concluded that Han’s theoretical conclusion has been rather vigilantly
addressed by the DAA and the accession project at-large.
With regard to the Protocol 8 intentions and limitations, the book concludes that
the original intention of Lisbon Treaty was to attribute responsibility merely to the
responsible parties—and not to those not substantively responsible. This is the only
way to properly identify the concept of ‘correct addressees’. If this was the original
intention of Protocol 8, then the DAA’s co-respondent mechanism deviates from
that meaning rather incredibly. As the co-respondent mechanism establishes a joint-
responsibility model for attributing liability when both EU and its Member States
are (co)-respondents, the latter is basically a model of attributing mutual liability
both on the author of the violation and on the party that merely implemented a duty
under international law that tolerated no state discretion. This being the case, the

7
Callewaert (2014), p. 11.
8
Han (2006), p. 108.
432 12 An Overall Conclusion

book concludes that it is exactly the contrary to the intention of Protocol 8 which is
practically being achieved, namely that the incorrect addressee—the one that was
not the original author of the violation—is becoming liable for something it did not
rule upon. Hence, the co-respondent mechanism—it is concluded—undertakes to
do quite the opposite of Protocol 8 purpose.
On the other hand, from a more normative development point of view, the book
concludes that accession will likely broaden the scope of jurisdiction of the
Luxembourg Court in two directions. First, the EU Court will be forced to apply
more stringent rules with regard to the use of the preliminary reference procedure
by national courts, and provide more substantive answers as to the issues raised by
national courts. And, second, by being forced to use the prior involvement mech-
anism, the EU Court would need to engage with a broader human rights jurisdiction
over larger policy areas in order to retain its position as the ultimate reviewer of EU
law. The panic that EU Court may loose territory towards Strasbourg Court will
force the latter to consider its jurisdictional scope with more flexibility while
providing more substantive reviews of human rights law in order to avoid itself
from being humiliated before Strasbourg Court.
The book concludes that, post accession, the Strasbourg Court will develop into
a supreme court towards Luxembourg Court, however, only within the remit of
specialized Convention law jurisdiction. Luxembourg Court’s President, Judge
Skouris, had stood in the same line of argument, however, he had argued in addition
that ‘there is nothing shocking in this: the position is the same when the constitu-
tional or supreme courts of Member States test the constitutionality or legality of
acts within their legal systems.’9 The book—by contrast to Judge Skouris—con-
cludes that although the Luxembourg Court will now on become more or less an
ordinary domestic court in face of Strasbourg, this is not that normal from the
perspective of the principle of autonomy of EU law. One needs to disagree here
with Judge Skouris—as the book concludes—given that the autonomy of EU law
has been built on very harsh lines of sensitivity and aversion towards external
courts, something which one may never find to such degree at national courts. This
being said, the book argues that the mere fact—if not going further—that the
Luxembourg Court becomes positioned at the level of a domestic court of a state
clearly opens ways to rearrange the principle of autonomy of EU law from the
perspectives of Kadi, Opinion 1/91, Mox Plant, etc. This will demise Lavranos
theoretical point that both courts appear to have a ‘horizontal jurisdictional com-
petition’.10 Therefore, the book concludes that—in principle—the rearranged posi-
tions between the Strasbourg and Luxembourg courts to a vertical line do as such
open ways for the principle of autonomy in the EU to change. Whether the mediated
dialogue will allow that to happen, it may only be speculated at this stage, however
the book concludes that structurally the possibility for it will be achievable from a
post-accession perspective. Therefore, it is concluded that Buergenthal’s theoretical

9
Quoted from: Barbera (2012), p. 9.
10
Lavranos (2009), p. 2.
12.2 Conclusion on the Overall Functionality of the Accession Agreement and Its. . . 433

advice for competing courts that they need ‘to recognize that [. . .they and all other
international courts. . .] are all part of the same legal system and that this fact
imposes certain obligations [. . .]’11 will finally take an answer by the DAA.
Buergenthal’s theoretical suggestion seems to have been at least formally addressed
by this accession project.
The book concludes that the DAA has responded to the EU’s need to shield its
autonomy with two mechanisms: the co-respondent mechanism and the prior
involvement mechanism. It has not responded in the same way with the rule on
exhaustion and the inter-party mechanism, however. With regard to the
co-respondent mechanism—which the book concludes is an innovation in interna-
tional law of multilevel responsibility—seems to relatively well support the EU law
autonomy, both in the internal and external context. However, there remain too
many loopholes which either endanger its functionality and effectiveness or provide
room for interference to EU law autonomy. Regarding the use of the co-respondent
mechanism, the book has enumerated four specific ambiguities which make espe-
cially its nature and scope blurred in several perspectives. The use of the
co-respondent mechanism—the book concludes—is especially problematic with
regard to omissions, as the Strasbourg Court, in order to verify which is the party
competent to deliver on a certain positive obligation, will need to inquire about the
division of competences between the EU and its Member States within EU law. The
prior involvement mechanism—the book concludes—is one which especially
ensures that autonomy of EU Court’s exclusive jurisdiction, by setting up a system
to ensure that the EU Court will always be the first and ultimate reviewer of EU law
in case the EU judicial architecture skips over from having cautiously provided for
that. The book concludes, on the other hand, that when the EU does not qualify to
use the prior involvement mechanism, its autonomy may seriously become at risk,
as its rule on exhaustion does not make any exception in those circumstances. The
same is concluded about the inter-party mechanism, wherein the EU law is not
prevented from the potential interference to its autonomy, with the book concluding
that DAA and the EU Treaties may be reconciled to guarantee that the EU become
procedurally submitted to the inter-party complaint mechanism.
Finally, although not very well connected to the above conclusions, one needs to
mention that Luxembourg’s Opinion 2/13 may play a core role in either weakening
the accession project (and changing the functions of the current DAA formulations)
or ruining it at all. It seems not possible to address the Luxembourg Court’s
concerns while simultaneously maintain the nature of the Convention as an instru-
ment of law offering equal-footing protection to all its contracting parties. This
seems especially problematic with regard to the Court’s request to exclude the
jurisdiction of the Strasbourg Court from the inter-party complaint procedure
between EU and its Member States and also exclude its jurisdiction over CFSP
acts under EU law. These two specific points seem problematic to be addressed,
therefore endangering the entire accession route. Although the decision to address

11
Buergenthal (2001), p. 274.
434 12 An Overall Conclusion

the Court’s concerns needs be addressed first at a political level, the only way to
make the ‘original’ accession intention achievable seems to be treaty amendments
via which the EU Court’s roots of concerns would be explicitly removed. The latter
seems a moot point considering the political momentum in EU policy-making
today. Therefore, one has to conclude that Luxembourg’s Opinion 2/13 seems to
seriously harm the idea of a unified European human rights law architecture, with a
vertically integrated but cooperative EU law and Convention treaty regimes. Such
move of the Luxembourg Court may seem to protect the EU law autonomy in the
short run, but may turn into a move that will seriously hamper the EU human rights
external policy and legitimacy later on. This development will soundly reject
Callewaert’s theoretical conclusion that there is the need to control centrifugal
tendencies between the two courts (examined supra in Chap. 1). In addition, if one
would take into account Alford’s observation that ‘[t]he effectiveness of a particular
international court depends in large measure on the theoretical rationale for why
nations obey its mandates [. . .]’,12 then Opinion 2/13 would demonstrate that the
Luxembourg Court distrusts the theoretical rationale that requires submission to the
mandate of the Strasbourg Court from the perspective of a Pan-European human
rights landscape. This seems a step that would hugely fragment the European
human rights law architecture. Not least, but last, it is the EU citizens who will
suffer at the end of the day from this status quo where no external human rights
control over EU conduct is possible.

12.3 Searching for a Theoretical Model to Explicate


the Accession Output

As regards H1, the book concludes that EU accession to ECHR will generate fairly
sizeable changes in the EU internal legal order, both in terms of the concepts
whereupon EU has typically based its autonomy jealously and in terms of the
structural circumstances which will make the EU operate from a more
Convention-esteemed position. H1, therefore, is only partially tested as being
positive, as the book has constantly argued that it will remain with the practical
interaction between the two courts to define the level of the ‘substantive’ changes
which it may push in reality. The space for such ‘substantive’ changes in the EU
system of law may be argued as present with the DAA, nevertheless whether such
spaces will be used or abused has not been satisfactorily tested. This said, the book
concludes that in the structural sense the accession will bring significant changes to
the EU legal system.
As regards H2, the book concludes that the EU accession to ECHR will sub-
stantially root EU into a rather privileged—not only equal—position with the other
state High Contracting Parties, something that will altogether increase EU’s

12
Alford (2000), p. 164.
12.4 Post-accession (Forthcoming) Perspectives: What About a New Normative. . . 435

capacity to stand as a state party to international law treaties. Such development—it


is hereby concluded—will increase EU’s stateness attitude in international law,
clearly reflecting EU’s increasing external sovereignty features. With that devel-
opment taking place rather vigorously, accession will likely start the EU’s journey
of becoming a normal entity of international law capable of undertaking interna-
tional obligations in the strategic and conventional state-domains of law. That
tendency, therefore, will clearly prove H2 as being a core effect of the accession,
therefore concluding that EU and ECHR regimes of law will build a sort of
interaction that will benefit EU’s external portray of ‘stateness’.
As regards H3, the book concludes that EU accession to ECHR will clearly make
Strasbourg Court a higher court in face of Luxembourg Court, therefore their
relations will in the structural sense change extensively. Whether there will be a
mediated dialogue that will keep the vertical line of relationship frozen practically,
that has not been sufficiently tested since without a practice on the ground it was not
possible to conclude on that. Such new relationship will seem of substantive
nature—at least for the Luxembourg’s conventional role as the only valid adjudi-
cator of rights for EU-related claims—and will be expected to promote the
Luxembourg Court as a domestic court liable towards the specialized human rights
law court of international law, the Strasbourg Court. This vertical line will be of
substantive effect for the EU internally, as the Luxembourg Court will likely see
itself excluded from the exclusive primacy on the human rights law jurisdiction—
internally and externally (especially with the Bosphorus doctrine not in place
anymore). The headship on the European human rights law jurisdiction will
unquestionably belong to the Strasbourg regime of law post-accession.

12.4 Post-accession (Forthcoming) Perspectives: What


About a New Normative Order in Europe?

A core difference which EU accession to ECHR will pose to the current architecture
of legal pluralism and multilevel court system in Europe will be the fact that
Luxembourg and Strasbourg courts will not any longer have a horizontal line of
positions. Accession will—at least from a legal point of view—hierarchize their
relationship into a vertical line, with Strasbourg Court standing as a court of
international law on top of Luxembourg Court standing as a domestic court of the
European Union, the latter being a contracting party to the Convention. This
changed map of positions between the treaty regimes of EU and ECHR on the
one hand, and Luxembourg and Strasbourg courts on the other hand, posits a new
outlook for the upcoming architecture of legal pluralism in Europe, with both
regimes of law and their courts needing definite and accommodated functions to
exercise. This will also cure Lavranos theoretical observation that the divergences
between the case-law of the two courts may have been caused by a lack of
hierarchical positioning of the two courts (see supra Chap. 1).
436 12 An Overall Conclusion

With the hierarchized map, there is a need to assign each court and legal regime
with a certain theoretical and normative role which—from the interactive point of
view—would make the overall picture of European legal pluralism more effective
and sound. To that end, the book proposes that the Luxembourg and Strasbourg
courts—reflecting on the nature of their regimes of law—need each play a certain
specific role which would make the Europe’s court system more integrated although
plural. As Strasbourg Court is a specialized court of human rights—situated in an
international human rights law background—but also a subsidiary one, then its role
should be one which sets the general abstract guidelines of the European human
rights constitutional architecture, hence carefully preserving its subsidiary role in
face of Luxembourg regime of law. On its side, the Luxembourg Court may further
develop its constitutional role regarding the interpretation of EU law, however, in
order to offer full access to court to potential claimants, Luxembourg is proposed to
keep its nature as a regular court for EU law claims. Having two constitutional
courts in Europe would not be that contradictory, as Strasbourg Court would have a
constitutionally specialized role for human rights doctrine in Europe, whereas
Luxembourg would have this regular-function for all issues relating to EU law.
Basically, the Strasbourg Court would fashion a Europe-wide constitutional level of
law that applies from an international law perspective, whereas the Luxembourg
Court would play its function from the perspective of EU-domestic constitutional
law level. This architecture would basically connote that Strasbourg Court would
become a court for European-international constitutional law, whereas Luxem-
bourg Court would become a regular constitutional court for EU-domestic consti-
tutional law: both orders proposed to engage in a monist relationship. Strasbourg
Court—one needs to evoke—is not only international in terms of the origin of its
authority, but also in terms of its position in face of Luxembourg and other national
courts.
With the Strasbourg Court becoming the constitutional court for the European
continent—as opposed to the Luxembourg Court as a constitutional court for EU
law13—the former would need to adopt a rather well arranged model of margin of
appreciation to apply on the EU: the latter being a supranational organization with
28 Member States, each of them having specific legal circumstances on human
rights law. The margin of appreciation that Strasbourg Court would need to
formulate for the EU should therefore be far more complex and multilevel in
nature, in order to allow for the peculiarities of EU Member States to find their
place therein. The degree of that margin of appreciation, however, will all depend
on the trust which Strasbourg Court will have towards the Luxembourg regime of
law. If Strasbourg Court will follow to apply the Bosphorus standard as a standard
of merits—rather than an admissibility standard—it would be rather difficult for the

13
Contra: Gragl comes to the conclusion that the Luxembourg Court will become a ‘quasi-
constitutional or supreme court’, whereas the Strasbourg Court will merely have the final say on
fundamental rights in Europe. See: Gragl (2013), p. 278. My argument presented above goes
beyond Gragl, proposing that both courts undertake a more structured function with regard to their
role in the increasing legal pluralism but also human rights law architecture in Europe as a whole.
12.4 Post-accession (Forthcoming) Perspectives: What About a New Normative. . . 437

Convention system to exercise its role effectively towards the EU. It is suggested
here that Bosphorus should neither stand as an admissibility standard—as it was up
to now—nor as a standard of merit. The degree of margin of appreciation should
therefore be extensively reduced, and applied with variant grades even upon EU
legal acts. As EU is a complex institutional system, it demonstrates certain levels of
difference between its own credibility of human rights in certain policy areas: EU
may be rather credible in its human rights approach on issues of employment, but
rather unconvincing with regard to the e.g. right of persons to access efficiently the
court. To this end, the vertical relationship between the Strasbourg and Luxembourg
courts will very much depend on the scale and design of the margin of appreciation,
which needs to reflect on the fact that both courts have different constitutional
functions to play.
On the international plan, accession will seriously strengthen the position of EU
and its Court in face of their Member States but also in relation to its external
sovereignty portrait. In this regard, with EU fortifying its external portray of
sovereignty, accession will likely produce an output that fortifies EU’s external
competences and authority. Such output will possibly also adjust the context under
which global law offers room to the EU, with a constant increase of international
law becoming open to the EU’s seating. With the EU increasing relentlessly its
external capacity to penetrate in conventionally state-reserved domains of interna-
tional law, one will likely observe a gradual increase of EU’s constitutional
capacity due to its increasing role in global law.
With the EU—as a supranational organization until today—becoming a
contracting party to the Convention system, many other international organizations
will start considering—and also becoming pressured—to join international human
rights treaty regimes. EU accession to a conventionally state-oriented domain of
international law, the Convention, will open way for international law to better
develop its multilevel structures to allow for the integration of other international
organizations to treaty regimes for human rights. This will likely be one of the spill-
over effects of the EU accession to ECHR, as this novelty will likely push
especially organizations which are very delicate and untouched by human rights
law obligations to carefully find strategies to address this issue as well. This spill-
over potential effect of the accession will also gradually transform the international
human rights law with states as the only liable agents—including for the liability of
international organizations—for international human rights law violations, to inter-
national organizations being assigned directly with that sort of obligations as long
as they effectively control their own actions. This enlightment—although in line
with DARIO—will bring the point of debate on the answerability of international
organizations for human rights breaches into a far more sophisticated and advanced
level; and this should be recognized as a novel global law development as regards
the law of international responsibility of international organizations.
438 12 An Overall Conclusion

References

Alford RP (2000) The proliferation of international courts and tribunals: international adjudication
in ascendance. Scholarly Works Paper 9. Available at: http://scholarship.law.nd.edu/law_
faculty_scholarship/9, 160–165
Barber BR (1993) Global democracy or global law: which comes first? Indiana J Global Legal
Stud 1(1):119–137
Barbera M (2012) Diritti sociali e crisi del costituzionalismo europeo. WP CSDLE “Massimo
D’Antona”. INT – 95/2012
Buergenthal T (2001) Proliferation of international courts and tribunals: is it good or bad? Leiden J
Int Law 14(2):267–275
Callewaert J (2014) The accession of the European Union to the European Convention on Human
Rights. Council of Europe, Strasbourg
de Witte B (2014) Beyond the accession agreement: five items for the European Union’s Human
Rights Agenda. In: Kosta V et al (eds) The EU accession to the ECHR. Hart Publishing, Oxford
Gragl P (2013) The accession of the European Union to the European Convention on Human
Rights. Hart Publishing, Oxford
Han SWD (2006) Decentralized proliferation of international judicial bodies. J Transnatl Law
Policy 16(1):101–122
Helfer LR, Slaughter A-M (2005) Why states create international tribunals: a response to Pro-
fessors Posner and Yoo. Calif Law Rev 93:1–58
Jacobs FG (2007) Accession of the European Union to the European Convention on Human
Rights. Hearing organised by the Committee on Legal Affairs and Human Rights in Paris on
11 September 2007. Available at: http://www.statewatch.org/news/2007/sep/jacobs-eu-echr.
pdf
Korenica F, Doli D (2015) The EU’s parliamentary representation in light of Strasbourg Court’s
Sejdic and Zornic´ standards: is there tendency for a new parliamentary order in the EU? In
Kużelewska E et al (eds) European judicial systems as a challenge for democracy. Intersentia
(forthcoming)
Lavranos N (2009) On the need to regulate competing jurisdictions between International Courts
and Tribunals. EUI Working Paper MWP 2009/14
Talbot C (2014) Human rights and competition law in a European context. BlogPost. Available at:
http://humanrights.ie/uncategorized/human-rights-and-competition-law-in-a-european-context
Bibliography

Books, Journal Articles and Academic Reports

Abbott KW, Snidal D (2000) Hard and soft law in international governance. Int Organ 54(3):421–
456
Ahmed T, Butler I (2006) The European union and human rights: an international law perspective.
Eur J Int Law 17(4):771–801
AIRE Centre, AI & ICJ (2013) European Union Accession to the European Convention on Human
Rights: briefing note on the accession agreement. Brussels, 6 September 2013
Akehurst M (1981) The application of the general principles of law by the Court of Justice of the
European communities. Br Yearb Int Law 52(1):29–51
Albi A (2009) Ironies in human rights protection in the EU: pre-accession conditionality and post-
accession conundrums. Eur Law J 15(1):46–69
Albors-Llorens A (2012) Remedies against the EU institutions after Lisbon: an era of opportunity?
Camb Law J 71(3):507–536
Alegre S, Leaf M (2004) Mutual recognition in European judicial cooperation: a step too far too
soon? Case study—the European arrest warrant. Eur Law J 10(2):200–217
Alemanno A (2010) Opinion 1/08, Community competence to conclude with certain members of
the World Trade Organization Agreements modifying the schedules of specific commitments
of the community and its Member States under the General Agreement on Trade in Services.
Am J Int Law 104(3):467–474
Alemanno A, Stefan O (2014) Openness at the Court of Justice of the European Union: toppling a
taboo. Common Market Law Rev 51(1):97–140
Alford RP (2000) The proliferation of international courts and tribunals: international adjudication
in ascendance. Scholarly Works Paper 9, pp 160–165. Available at: http://scholarship.law.nd.
edu/law_faculty_scholarship/9
Alter KJ (1996) The European court’s political power. West Eur Polit 19(3):458–487
Alter KJ (1998) Who are the “Masters of the Treaty”? European Governments and the European
Court of Justice. Int Organ 52(1):121–147
Alter KJ, Helfer LR (2010) Nature or nurture? Judicial lawmaking in the European Court of Justice
and the Andean Tribunal of Justice. Int Organ 64(4):563–592
Analysis (1997) The European Community cannot accede to the European Convention on Human
Rights. Eur Law Rev 1:235–249
Arnull A (1989) The use and abuse of article 177. Modern Law Rev 52(5):622–639

© Springer International Publishing Switzerland 2015 439


F. Korenica, The EU Accession to the ECHR, DOI 10.1007/978-3-319-21759-8
440 Bibliography

Arnull A (1995) Private applicants and the action for Annulment under Art. 173 of the EC Treaty.
Common Market Law Rev 32(1):7–49
Arnull A (2001) Private applicants and the action for annulment since Codorniu. Common Market
Law Rev 38(1):7–52
Arnull A (2003) From charter to constitution and beyond: fundamental rights in the new European
Union. Public Law (Winter Issue) (4):774–793
Arnull A (2011) The principle of effective judicial protection in EU law: an unruly horse. Eur Law
Rev 36(1):51–70
Arold Lorenz N-L et al (2013) The European human rights culture – a paradox of human rights
protection in Europe? Martinus Nijhoff Publishers, Leiden
Balfour ADJ (2005) Application of the European Convention on Human Rights by the European
Court of Justice. Harvard Law School Student Scholarship Series. Paper 4:1–57
Balfour ADJ (2007) Eliminating conflicting interpretations of the European Convention on Human
Rights by the European Court of Justice and the European Court of Human Rights: the PDIQ
system as a preventative solution. Intercult Hum Rights Law Rev 2:183–247
Balthasar S (2010) Locus Standi rules for challenges to regulatory acts by private applicants: the
new article 263(4) TFEU. Eur Law Rev 35(4):542–550
Baratta R (2013) Accession of the EU to the ECHR: the rationale for the ECJ’s prior involvement
mechanism. Common Market Law Rev 50(5):1305–1332
Barber BR (1993) Global democracy or global law: which comes first? Indiana J Global Legal
Stud 1(1):119–137
Barbera M (2012) Diritti sociali e crisi del costituzionalismo europeo. WP CSDLE “Massimo
D’Antona”. INT – 95/2012
Barents R (2004a) The autonomy of community law. Kluwer Law International, The Hague
Barents R (2004b) The court of justice in the draft constitution. Maastricht J Eur Comp Law 11
(2):121–142
Barents R (2009) The precedence of EU law from the perspective of constitutional pluralism. Eur
Const Law Rev 5(3):421–446
Barents R (2010) The Court of Justice after the Treaty of Lisbon. Common Market Law Rev 47
(3):709–728
Berman PS (2002) The globalization of jurisdiction. University of Connecticut School of Law
Articles and Working Papers. Paper 13. Available at: http://lsr.nellco.org/uconn_wps/13
Bernhard W (2011) Art 344 AEUV. In: Calliess C, Ruffert M (eds) EUV/AEUV (Das
Verfassungsrecht der Europäischen Union mit Europäischer Grundrechtecharta Kommentar).
C.H.Beck (Auflage: 4)
Besselink LFM (2013) Should the European Union ratify the European Convention on Human
Rights? Some remarks on the relations between the European Court of Human Rights and the
European Court of Justice. In: Føllesdal A et al (eds) Constituting Europe: the European Court
of Human Rights in a national, European and global context. Cambridge University Press,
Cambridge
Besselink L, Reestman J-H (2008) The relative autonomy of the EU human rights standard. Eur
Const Law Rev 4(2):199–204
Besson S (2009) European legal pluralism after Kadi. Eur Const Law Rev 5(2):237–264
Bieber R (2009) An association of sovereign states. Eur Const Law Rev 5(3):391–406
Biehler G (2008) Procedures in international law. Springer, Berlin
Bjoerklund M (2001) Responsibility in the EC for Mixed Agreements—should non-member
parties care? Nordic J Int Law 70(3):373–402
Bossuyt M, Verrijdt W (2011) The full effect of EU law and of constitutional review in Belgium
and France after the Melki Judgment. Eur Const Law Rev 7(3):355–391
Bradley CA (1998) The Charming Betsy canon and separation of powers: rethinking the interpre-
tive role of international law. Georgia Law J 86:479–537
Bibliography 441

Breuer M (2013) No Donum Danaorum! A reply to Daniel Thym’s “A Trojan Horse?”, VerfBlog,
2013/9/16. http://www.verfassungsblog.de/en/no-donum-danaorum-a-reply-to-daniel-thymsa-
trojan-horse/
Břı́za P (2009) The Czech Republic: the constitutional court on the Lisbon Treaty Decision of 26
November 2008. Eur Const Law Rev 5(1):143–164
Broberg M, Fenger N (2013) Variations in Member States’ preliminary references to the Court of
Justice—are structural factors (part of) the explanation? Eur Law J 19(4):488–501
Brown C (2002) The proliferation of international courts and tribunals: finding your way through
the maze. Review essay. Melb J Int Law 3(2). Available at: http://www.law.unimelb.edu.au/
files/dmfile/downloadb15e1.pdf
Brownlie I (2003) Principles of public international law, 6th edn. Oxford University Press, Oxford
Buergenthal T (2001) Proliferation of international courts and tribunals: is it good or bad? Leiden J
Int Law 14(2):267–275
Busby N, Zahn R (2013) The EU’s accession to the ECHR: conflict or convergence of social
rights? Paper to be presented at the labour law research network’s inaugural conference
Barcelona, 13–15th June 2013
Buxbaum RM (2005) A legal history of international reparations. Berkeley J Int Law 23:314–346,
Available at: http://scholarship.law.berkeley.edu/bjil/vol23/iss2/5
Caldwell RG (1920) The settlement of inter-state disputes. Am J Int Law 14(1/2):38–69
Callewaert J (2007) The European Court of Human Rights and the area of freedom, security and
justice. ERA Forum 8(4):511–518
Callewaert J (2014) The accession of the European Union to the European Convention on Human
Rights. Council of Europe, Strasbourg
Cardwell PJ, French D (2007) Who decides? The ECJ’s judgment on jurisdiction in the MOX Plant
dispute. J Environ Law 19(1):121–129
Cardwell PJ, French D, White N (2009) I. European Court of Justice, Yassin Abdullah Kadi and Al
Barakaat International Foundation v Council and Commission (JOINED CASES C-402/05 P
AND C-415/05 P) Judgment of 3 September 2008. Int Comp Law Q 58(1):229–240
Carozza PG (2003) Subsidiarity as a structural principle of international human rights law. Am J
Int Law 97(1):38–79
Carrubba CJ, Murrah L (2005) Legal integration and use of the preliminary ruling process in the
European Union. Int Organ 59(2):399–418
Carrubba CJ et al (2008) Judicial behavior under political constraints: evidence from the European
Court of Justice. Am Polit Sci Rev 102(4):435–452
Carruthers S (2009) The Treaty of Lisbon and the reformed jurisdictional powers of the European
Court of Justice in the field of justice and home affairs. Eur Hum Rights Law Rev 6:784–804
Cartabia M (2009) Europe and rights: taking dialogue seriously. Eur Const Law Rev 5(1):5–31
Cassese A (1985) Modern constitutions and international law. Martinus Nijhoff, Hague
Casteleiro AD (2014) United we stand: the EU and its Member States in the Strasbourg Court. In:
Kosta V et al (eds) The EU accession to the ECHR. Hart Publishing, Oxford
Chalmers D, Tomkins A (2007) European Union Public Law. Cambridge University Press,
Cambridge
Chalmers D, Davies G, Monti G (2010) European Union Law. Text and materials, 2nd edn.
Cambridge University Press, Cambridge
Chatháin CN (1999) The European Community and the Member States in the dispute settlement
understanding of the WTO: united or divided? Eur Law J 5(4):461–478
Chinkin CM (1986) Third-party intervention before the International Court of Justice. Am J Int
Law 80:495–531
Churchill R, Scott J (2004) The Mox Plant litigation: the first half-life. Int Comp Law Q 53
(3):643–676
Comment (1978) The international human rights treaties: some problems of policy and interpre-
tation. Univ Pa Law Rev 126(4):886–929
442 Bibliography

Conforti B (2010) Comments on the accession of the European Union to the ECHR. Italian Yearb
Int Law 20:83–86
Coppel J, O’neill A (1992) The European Court of Justice: taking rights seriously? Common
Market Law Rev 29(4):669–692
Cormick NM (1995) The Maastricht Urteil: sovereignty now. Eur Law J 1(3):259–266
Corthaut T (2002) Jégo-Quéré v. Commission, T-177/01 and C-50/00 P ECJ July 25, 2002, Uni on
de Peque~ nos Agricultores v. Council’. Columbia J Eur Law 9(1):141–166
Costa J-P (2003) The European Court of Human Rights and its recent case law. Texas Int Law J
38:455–468
Council of Bars and Law Societies of Europe (CCBE) Statement on the European Union accession
to the European Convention on Human Rights, July 2013
Council of Europe, doc. CDDH-UE (2010) 06rev (1st Working Meeting of the CDDH Informal
Working Group on the Accession of the European Union to the European Convention on
Human Rights (CDDH-UE) with the European Commission)
Craig P (2013) EU accession to the ECHR: competence, procedure and substance. Fordham Int
Law J 36(5):1114–1150
Craig P, de Búrca G (2003) EU law, 3rd edn. Oxford University Press, New York, chapter 6
Craig P, de Búrca G (2007a) EU law: text, cases and materials, 4th edn. Oxford University Press,
Oxford
Craig P, de Búrca G (2007b) EU Law, 4th edn. Oxford University Press, New York
Craig P, De Búrca G (2008) EU law: text, cases and materials, 5th edn. Oxford University Press,
Oxford
Cremona M (2003) The draft constitutional treaty: external relations and external action. Common
Market Law Rev 40(6):1347–1366
Crnic-Grotic V (1997) Object and purpose of treaties in the Vienna Convention on the Law of
Treaties. Asian Yearb Int Law 7:141–174
Cuyvers A (2011) The Kadi II Judgment of the General Court: the ECJ’s predicament and the
consequences for Member States. Eur Const Law Rev 7(3):481–510
d’Aspremont J (2007) Abuse of the legal personality of international organisations and the
responsibility of member states. Int Organ Law Rev 4(1):91–119
d’Aspremont J (2014) A European law of international responsibility? The articles on the
responsibility of international organizations and the EU. In: Kosta V et al (eds) The EU
accession to the ECHR. Hart Publishing, Oxford
d’Oliveira HUJ, de Groot GR, Seling A (2011) Court of Justice of the European Union: Decision
of 2 March 2010, Case C-315/08, Janko Rottman v. Freistaat Bayern Case Note 1 Decoupling
Nationality and Union Citizenship? Case Note 2 The Consequences of the Rottmann Judgment
on Member State Autonomy – The European Court of Justice’s Avant-Gardism in Nationality
Matters. Eur Const Law Rev 7(1):138–160
Dauses MA (1986) Practical considerations regarding the preliminary ruling procedure under
article 177 of the EEC treaty. Fordham Int Law J 10(3):538–577
de Búrca G (2005) Limiting EU powers. Eur Const Law Rev 1(1):92–98
de Búrca G (2009) The European Court of Justice and the international legal order after Kadi. Jean
Monnet Working Paper 1/09. Available at: http://www.jeanmonnetprogram.org/papers/09/
090101.pdf
de Búrca G (2010) The European Court of Justice and the international legal order after Kadi.
Harvard Int Law Rev 51(1):1–49
de Búrca G (2011) The road not taken: the European Union as a global human rights actor. Am J
Int Law 105(4):649–693
de Búrca G (2013) After the EU Charter of Fundamental Rights: the Court of Justice as a human
rights adjudicator? Maastricht J Eur Comp Law 20(2):168–184
De Hert P (2005) Crafting European Integration with a Constitution containing an abundance of
substantive human rights. Are freedom, democracy and pluralism served? In: Van Thiel S, De
Bibliography 443

Gucht K, Lewis R (eds) Understanding the new European Constitutional Treaty. VUB Press,
University Press, Brussels
De Hert P, Korenica F (2012) The Doctrine of Equivalent Protection: its life and legitimacy before
and after the European Union’s accession to the European Convention on Human Rights.
German Law J 13(7):874–895
De Londras F, Kingston S (2010) Rights, security, and conflicting international obligations:
exploring inter-jurisdictional judicial dialogues in Europe. Am J Comp Law 58:359–414
de Rivery EM, Chassaing E, EU accession to the ECHR: bringing about an aggiornamento for EU
case law? Overview of ECHR and EU case law. e-Competitions | No 61327:1–16
De Schutter O (2007) Accession of the European Union to the European Convention on Human
Rights, p 13. Available at: http://www.statewatch.org/news/2007/sep/decchutte-contributin-
eu-echr.pdf
De Schutter O (2014) Bosphorus post-accession: redefining the relationships between the Euro-
pean Court of Human Rights and the parties to the Convention. In: Kosta V et al (eds) The EU
accession to the ECHR. Hart Publishing, Oxford
de Vries SA (2013) Editorial: EU and ECHR: conflict or harmony? Utrecht Law Rev 9(1):78–79
de Waele H (2011) The EU, the Member States and international law. In: Layered global player.
Springer, Berlin, Chapter (141–152)
De Witte F (2008) The European judiciary after Lisbon. Maastricht J Eur Comp Law 15 (1):43–54
de Witte B (2014) Beyond the accession agreement: five items for the European Union’s Human
Rights Agenda. In: Kosta V et al (eds) The EU accession to the ECHR. Hart Publishing, Oxford
Defeis EF (2000–2001) Human Rights and the European Union: Who decides? Possible conflicts
between the European Court of Justice and the European Court of Human Rights. Dickinson J
Int Law 19(2):301–331
Defeis E (2007–2008) Dual system of human rights: The European Union. ILSA J Int Comp Law
14:1–8
den Heijer M (2013) Shared responsibility before the European Court of Human Rights. Nether-
lands Int Law Rev 60(3):411–440
Dijk V (2007) The accession of the European Union/European Community to the European
Convention on Human Rights. Available at: http://www.statewatch.org/news/2007/sep/van-
dijk-eu-echr.pdf
Dimopoulos A (2011) The validity and applicability of international investment agreements
between EU Member States under EU and International Law. Common Market Law Rev 48
(1):63–93
Donahue DE (1980) Human rights in Northern Ireland: Ireland v. the United Kingdom. Boston
Coll Int Comp Law Rev 3(2):377–432
Dougan M (2004) National remedies before the Court of Justice: issues of harmonisation and
differentiation. Hart Publishing, Oxford
Douglas-Scott S (2006) Bosphorus Hava Yollari Turizm Ve Ticaret Anonim Sirketi v. Ireland,
(2006) Casenote. Common Market Law Rev 43(1):243–254
Dyevre A (2013) European Integration and National Courts: defending sovereignty under institu-
tional constraints? Eur Const Law Rev 9(1):139–168
Eckes C (2012) Protecting supremacy from external influences: a precondition for a European
Constitutional legal order? Eur Law J 18(2):230–250
Eckes C (2013) EU accession to the ECHR: between autonomy and adaptation. Mod Law Rev 76
(2):254–285
Editorial (2013) Human rights: Member State, EU and ECHR levels of protection; P.S. Catalonia;
why does it take so long for my article to be published? In this issue: human rights: Member
State, EU and ECHR levels of protection. Eur J Int Law 24:471–475
Editorial Comments (1995) The aftermath of Opinion 1/94 or how to ensure the Unity of
Representation for Joint Competences. Common Market Law Rev 32:385–90
Editorial Comments (2009) Karlsruhe has spoken: “Yes” to the Lisbon Treaty, but. . .. Common
Market Law Rev 46(4):1023–1033
444 Bibliography

Eeckhout P (2004) External relations of the European Union. Oxford University Press, Oxford
Eliantonio M, Kas B (2010) Private parties and the annulment procedure: can the gap in the
European system of judicial protection be closed? J Polit Law 3(2):121–133
Eliantonio M, Stratieva N (2003) From Plaumann, through UPA and Jégo-Quéré, to the Lisbon
Treaty: the Locus Standi of Private Applicants under Article 230(4) EC through a political lens.
Maastricht Faculty of Law Working Paper 2009/13
European Union commits to joining European Human Rights Court. International Law Update
13:182–183 (2007)
Farrall JM (2007) United Nations sanctions and the rule of law. Cambridge University Press,
Cambridge
Fassbender B (1998) The United Nations Charter as constitution of the international community.
Columbia J Transnat Law 36:529–619
Fitzmaurice G (1957) The general principles of international law considered from the standpoint
of the rule of law. Collected Courses of the Hague Academy of International Law, vol 92. Brill|
Nijhoff, Boston
Franck TM (1988) Legitimacy in the international system. Am J Int Law 82(4):705–759
Franklin C (2010–2011) The legal status of the EU Charter of Fundamental Rights after the Treaty
of Lisbon. Tilburg Law Rev 15:137–162
Friendly Settlement and Strike Out (Articles 37-38). Article 3 of the European convention on
human rights: a practitioner’s handbook. Available at: http://www.omct.org/files/2006/11/
3633/handbook1_eng_08_part8.pdf
Gaja G (2013) The “Co-Respondent Mechanism” according to the Draft Agreement for the
accession of the EU to the ECHR. ESIL Reflect 2(1):1–6
Garcı́a RA (2002) The general provisions of the charter of fundamental rights of the European
Union. Eur Law J 8(4):492–514
Garcı́a IG (2009) Gibraltar and the European Parliament elections before the European Court of
Justice. Bull Transilvania Univ Braşov 2(51):195–202
Gerards J (2011) Pluralism, deference and the margin of appreciation doctrine. Eur Law J 17
(1):80–120
Gondek M (2005) Extraterritorial application of the European Convention on Human Rights:
territorial focus in the age of globalization? Netherlands Int Law Rev 52(3):349–387
Gormley LW (2006) Judicial review: advice for the deaf? Fordham Int Law J 29(4):655–689
Govaere I (2009) The importance of international developments in the case-law of the European
Court of Justice: Kadi and the autonomy of the EC legal order. Research Papers in Law 1/2009,
College of Europe
Grabenwarter C (2009) Europäische Menschenrechtskonvention. Verlag C.H. Beck, Munchen
Gragl P (2011) Accession revisited: will fundamental rights protection trump the European
Union’s Legal Autonomy? In: Benedek W, Benoı̂t-Rohmer F, Karl W, Nowak M (eds)
European Yearbook on human rights. Neuer Wissenschaftlicher Verlag (NWV), Wien
Gragl P (2013) The accession of the European Union to the European Convention on Human
Rights. Hart Publishing, Oxford
Greer S, Williams A (2009) Human rights in the Council of Europe and the EU: towards
‘Individual’, ‘Constitutional’ or ‘Institutional’ justice? Eur Law J 15(4):462–481
Grossman C, Bradlow DD (1993) Are we being propelled towards a people-centered transnational
legal order? Am Univ Int Law Rev 9(1):1–25
Groussot X, Minssen T (2007) Res Judicata in the Court of Justice case-law. Eur Const Law Rev 3
(3):385–417
Groussot X et al (2011) EU accession to the European Convention on Human Rights: a legal
assessment of the draft accession agreement of 14th October 2011. Fondation Robert Schuman/
European Issues N 218/7 November 2011
Halberstam D, Stein E (2009) The United Nations, the European Union, and the King of Sweden:
economic sanctions and individual rights in a plural world order. Common Market Law Rev 46
(1):13–72
Bibliography 445

Han SWD (2006) Decentralized proliferation of international judicial bodies. J Transnatl Law
Policy 16(1):101–122
Hancox E (2013) The meaning of “implementing” EU law under Article 51(1) of the Charter:
Åkerberg Fransson. Common Market Law Rev 50(5):1411–1431
Hannan D (2010) The EU demands recognition as a state by the United Nations. The Telegraph
Blog. Available at: http://blogs.telegraph.co.uk/news/danielhannan/100053681/the-eu-wants-
to-be-treated-as-a-state-by-the-united-nations/
Harding C (2000) The identity of European law: mapping out the European legal space. Eur Law J
6(2):128–147
Harmsen R (2014) The (geo-)politics of EU accession to the ECHR: democracy and distrust in the
wider Europe. In: Kosta V et al (eds) The EU accession to the ECHR. Hart Publishing, Oxford
Hart JW (2010) The European human rights system. Law Libr J 102(4):533–559
Hartley T (1999) Constitutional problems of the European Union. Hart Publishing, Oxford
Heffernan L (1997) A comparative view of individual petition procedures under the European
Convention on Human Rights and the international covenant on civil and political rights. Hum
Rights Q 19(1):78–112
Heifer LR, Slaughter A-M (1997) Toward a theory of effective supranational adjudication. Yale
Law J 107:273–392
Helfer LR (1993) Consensus, coherence and the European Convention on Human Rights. Cornell
Int Law J 26:133–165
Helfer LR, Slaughter A-M (2005) Why states create international tribunals: a response to Pro-
fessors Posner and Yoo. Calif Law Rev 93:1–58
Heliskoski J (1996) The “Duty of Cooperation” between the European Community and its
Member States within the World Trade Organization. Finnish Yearb Int Law 7:59–133
Heliskoski J (2000) The jurisdiction of the European court of justice to give preliminary rulings on
the interpretation of mixed agreements. Nord J Int Law 69:395–412
Heringa AW, Verhey L (2011) The EU Charter: text and structure. Maastricht J Eur Comp Law 8
(1):11–32
Higgins R (2003) The ICJ, the ECJ, and the integrity of international law. Int Comp Law Q 52
(1):1–20
Hinarejos A (2009) The Lisbon Treaty versus standing still: a view from the Third Pillar. Eur
Const Law Rev 5(1):99–116
Hoffmeister F (2006) International decisions: Bosphorus Hava Yollari Turizm Ve Ticaret Anonim
Sirket v. Ireland. App. No. 45036/98. Am J Int Law 100:442–449
Hoffmeister F (2007) Outsider or frontrunner? Recent developments under international and
European law on the status of the European Union in international organizations and treaty
bodies. Common Market Law Rev 44:41–68
Hofmann HCH, Mihaescu BC (2013) The relation between the Charter’s fundamental rights and
the unwritten general principles of EU law: good administration as the test case. Eur Const Law
Rev 9(1):73–101
Human Rights Act: How it works. BBC, 29 September 2000. Available at: http://news.bbc.co.uk/
2/hi/uk_news/946390.stm
International Law Commission, (‘The unity or diversity of the juridical regime for reservations’),
Preliminary Conclusions of 1997
Jackson JH (1992) Status of treaties in domestic legal systems: a policy analysis. Am J Int Law 86
(2):310–340
Jacobs FG (2007a) Accession of the European Union to the European Convention on Human
Rights. Hearing organised by the Committee on Legal Affairs and Human Rights in Paris on 11
September 2007. Available at: http://www.statewatch.org/news/2007/sep/jacobs-eu-echr.pdf
Jacobs FG (2007b) The sovereignty of law: the European way. Cambridge University Press,
Cambridge
Jacobs FG (2008) Direct effect and interpretation of international agreements in the recent case
law of the European Court of Justice. In: Dashwood A, Maresceau M (eds) Law and practice of
446 Bibliography

EU external relations: salient features of a changing landscape. Cambridge University Press,


Cambridge
Jacque JP (2011) The accession of the European Union to the European Convention on Human
Rights and fundamental freedoms. Common Market Law Rev 48(4):995–1023
Jayasuriya K (1999) Globalization, law, and the transformation of sovereignty: the emergence of
global regulatory governance. Indiana J Global Leg Stud 6(2):425–455
Joint Communication from Presidents Costa and Skouris, European Court of Human Rights
(ECHR) and the Court of Justice of the European Union (CJEU), 17 January 2011. Available
at: http://curia.europa.eu/jcms/upload/docs/application/pdf/2011-02/cedh_cjue_english.pdf
Jones C (2012) StateWatch Analysis: ‘The EU’s accession to the European Convention on Human
Rights: a cause for celebration or concern?’ Available at: http://www.statewatch.org/analyses/
no-187-echr.pdf
Joris T, Vandenberghe J (2008–2009) The Council of Europe and the European Union: natural
partners or uneasy bedfellows? Columbia J Eur Law 15:1–43
Kaczorowska A (2013) European Union law, 3rd edn. Routledge, London
Kakouris CM (1997) Do the Member States possess procedural “autonomy”? Common Market
Law Rev 34(6):1389–1412
Kapteyn PJG (1974) The “Domestic” law effect of rules of international law within the European
community system of law and the question of the self-executing character of GATT rules. Int
Lawyer 8(1):74–82
Kelemen RD, Sibbitt EC (2004) The globalization of American law. Int Organ 58(1):103–136
Kerson DLA (1961) The European Convention for the protection of human rights and fundamental
freedoms. Calif Law Rev 49(1):172–186
Kinley D (2005) The universalizing of human rights and economic globalization. What roles for
the rule of law? In: Zifcak S (ed) Globalization and the rule of law. Routledge, London
Klabbers J (2002) International law in community law: the law and politics of direct effect. Yearb
Eur Law 21:263–298
Klabbers J (2009) Treaty conflict and the European Union. Cambridge University Press,
Cambridge
Koch C (2005) Locus standi of private applicants under the EU constitution: preserving gaps in the
protection of individuals’ rights to an effective remedy. Eur Law Rev 30(4):511–527
Kokott J, Sobotta C (2012) The Kadi case – constitutional core values and international law –
finding the balance? Eur J Int Law 23(4):1015–1024
Komárek J (2013) The place of constitutional courts in the EU. Eur Const Law Rev 9(3):420–450
Korenica F, Doli D (2015) The EU’s parliamentary representation in light of Strasbourg Court’s
Sejdic and Zornic´ standards: is there tendency for a new parliamentary order in the EU? In:
Kużelewska E et al (eds) European judicial systems as a challenge for democracy. Intersentia,
Cambridge (forthcoming)
Koroma JA (1996) The interpretation of international law. In: Contemporary International Law
Issues: Proceedings of the Third ASIL/NVIR Joint Conference
Koutrakos P (2001) Trade, foreign policy and defence in EU constitutional law: the legal
regulation of sanctions, exports of dual-use goods and armaments. Hart Publishing, Oxford
Koutrakos P (2006) EU international relations law. Hart Publishing, Oxford
Koutrakos P (2010) I. External relations. Int Comp Law Q 59(2):481–489
Krüger HC (2002–2003) Reflections concerning accession of the European Communities to the
European Convention on Human Rights. Pa State Int Law Rev 21(1):89–99
Kuiper PJ (1995) The conclusion and implementation of the Uruguay Round results by the
European Community. Eur J Int Law 6(1):222–244
Kuiyper PJ, Paasivirta E (2004) Further exploring international responsibility: the European
Community and the ILC’s project on responsibility of international organisations. Int Organ
Law Rev 1:111–138
Kumm M (2005) The jurisprudence of constitutional conflict: constitutional supremacy in Europe
before and after the Constitutional Treaty. Eur Law J 11(3):262–307
Bibliography 447

La Pergola A (1993) The relationship between international and domestic law: traditional prob-
lems and new trends. In: ‘European Commission for Democracy Through Law (Venice
Commission),’ ‘The Relationship Between International and Domestic Law,’ proceedings of
the UniDem Seminar organized in Warsaw on 19–21 May 1993 in cooperation with the
University of Wroclaw and the Poznan Human Rights Centre. Science and Technique of
Democracy No. 5. CDL-STD(1993)005. Strasbourg, 15 September 1993
Ladenburger C (2012) FIDE 2012—Session on ‘Protection of Fundamental Rights post-Lisbon’
Report available at: www.fide2012.eu/index.php?doc_id=88
Landau EC (2008) A new regime of human rights in the EU? Eur J Law Reform 10(4):557–575
Lang JT (1986) The ozone layer convention: a new solution to the question of community
participation in “Mixed” International Agreements. Common Market Law Rev 23(1):157–176
Larsen KM (2008) Attribution of conduct in peace operations: the ‘Ultimate Authority and
Control’ Test’. Eur J Int Law 19(3):509–531
Larsen H (2009) A distinct FPA for Europe? Towards a comprehensive framework for analysing
the foreign policy of EU Member States’. Eur J Int Relat 15(3):537–566
Lavender N (1997) The problem of the margin of appreciation. Eur Hum Rights Law Rev 4
Lavranos N (2006a) MOX Plant Dispute – Court of Justice of the European Communities. Eur
Const Law Rev 2(3):456–469
Lavranos N (2006b) The MOX plant and IJzeren Rijn disputes: which court is the supreme arbiter?
Leiden J Int Law 19(1):223–246
Lavranos N (2007) The scope of the exclusive jurisdiction of the court of justice. Eur Law Rev 32
(1):83–94
Lavranos N (2008) Regulating competing jurisdictions among international courts and tribunals.
Za€oRV 68:575–621
Lavranos N (2009) On the need to regulate competing jurisdictions between international courts
and tribunals. EUI Working Paper MWP 2009/14
Law DS (2008) Globalization and the future of constitutional rights. Northwest Univ Law Rev 102
(3):1278–1349
Lawson R (2005) Human rights: the best is yet to come. Eur Const Law Rev 1(1):27–37
Leach P (2011) Taking a case to the European Court of Human Rights. Oxford University Press,
New York
Leal-Arcas R (2001) The European Community and mixed agreements. Eur Foreign Aff Rev 6
(4):483–513
Lenaerts K (2003) Interlocking legal orders in the European Union and comparative law. Int Comp
Law Q 52(4):873–906
Lenaerts K (2012) Exploring the limits of the EU Charter of fundamental rights. Eur Const Law
Rev 8(3):375–403
Lenaerts K, Corhaut T (2006) Of birds and hedges: the role of primacy in invoking norms of EU
law. Eur Law Rev 31(3):287–315
Lenaerts K, de Smijter E (2001) The Charter and the role of the European Courts. Maastricht J Eur
Comp Law 8(1):90–101
Lenaerts K, Gutiérrez-Fons JA (2010) The constitutional allocation of powers and general
principles of EU law. Common Market Law Rev 47(6):1629–1669
Lenart J (2012) ‘Fortress Europe’: compliance of the Dublin II Regulation with the European
Convention for the protection of human rights and fundamental freedoms. Utrecht J Int Eur
Law 28(75):04–19
Letelier R (2005) The competing jurisdictions of international courts and tribunals. Revista De
Estudios Internacionales 38(150):125–133
Letsas G (2004) The truth in autonomous concepts: how to interpret the ECHR. Eur J Int Law 15
(2):279–305
Lewis X (2006/2007) Standing of private plaintiffs to annul generally applicable European
Community measures: if the system is broken, where should it be fixed? Fordham Int Law J
30(5):1496–1544
448 Bibliography

Licková M (2008) European exceptionalism in international law. Eur J Int Law 19(3):463–490
Liisberg JB (2002) Does the EU Charter of fundamental rights threaten the supremacy of
Community law? Common Market Law Rev 38(5):1171–1199
Lixinski L (2014) Taming the fragmentation monster through human rights? International consti-
tutionalism, ‘Pluralism Lite’, and the common territory of the two European Legal Orders. In:
Kosta V et al (eds) The EU accession to the ECHR. Hart Publishing, Oxford
Lock T (2009) The ECJ and the ECtHR: the future relationship between the two European courts.
Law Pract Int Courts Tribunals 8:375–398
Lock T (2010) EU accession to the ECHR: implications for judicial review in Strasbourg. Eur Law
Rev 35(6):777–798
Lock T (2011a) Walking on a tightrope: the Draft Accession Agreement and the autonomy of the
EU legal order walking on a tightrope. A pre-edited version; forthcoming with CMLR 2011.
Available at: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1825227
Lock T (2011b) Walking on a tightrope: the Draft ECHR Accession Agreement and the autonomy
of the EU legal order. Common Market Law Rev 48:1025–1054
Lock T (2012) End of an epic? The draft agreement on the EU’s Accession to the ECHR. Yearb
Eur Law 31(1):162–197
o€
L€ of R (2006) Shooting from the hip: proposed minimum rights in criminal proceedings through-
out the EU. Eur Law J 12(3):421–430
Lüsberg JB (2006) The EU Constitutional Treaty and its distinction between legislative and non-
legislative acts – Oranges into apples? Jean Monnet Working Paper 01/06. Available at: http://
www.jeanmonnetprogram.org/papers/06/060101.pdf
Maduro MP (2003) Contrapunctual law: Europe’s constitutional pluralism in action. In: Walker N
(ed) Sovereignty in transition. Essays in European law. Hart Publishing, Oxford
Maniruzzaman MAF (2001) State contracts in contemporary international law: Monist versus
Dualist controversies. Eur J Int Law 12(2):302–329
Marsden S (2009) MOX Plant and the Espoo Convention: can Member State disputes concerning
mixed environmental agreements be resolved outside EC law? Rev Eur Commun Int Environ
Law 18(3):312–327
Martin FF (2003) The international human rights aspects of the forum non conveniens doctrine.
Univ Miami Inter-Am law Rev 35:101–123
Martin FF (2004) Our constitution as federal treaty: a new theory of United States constitutional
construction based on an originalist understanding for addressing a new world. Hastings Const
Law Q 31:268–354
Martı́n J, De Nanclares P (2013) The accession of the European Union to the ECHR: more than
just a legal issue. Working Papers on European Law and Regional Integration WP IDEIR no
15. Instituto de Derecho Europeo e Integracion Regional (IDEIR), Universidad Complutense
Martinico G (2012) Is the European Convention going to be ‘Supreme’? A comparative-constitu-
tional overview of ECHR and EU law before national courts. Eur J Int Law 23(2):401–424
Mendez M (2010) The enforcement of EU agreements: bolstering the effectiveness of treaty law?
Common Market Law Rev 47(6):1719–1756
Merrill JG (1993) The development of international law by the European Court of Human Rights.
Manchester University Press, Manchester
Miiller CA (2007) Fundamental rights in multi-level legal systems: recent developments in
European human rights practice. Interdiscip J Hum Rights Law 2(1):33–60
Millet F-X (2014) How much lenience for how much cooperation? On the first preliminary
reference of the French Constitutional Council to the Court of Justice. Common Market Law
Rev 51(1):195–218
Morano-Foadi S, Andreadakis S (2011) Reflections on the architecture of the EU after the Treaty
of Lisbon: the European judicial approach to fundamental rights. Eur Law J 17(5):595–610
Morijn J (2006) Balancing fundamental rights and common market freedoms in union law:
Schmidberger and Omega in the light of the European Constitution. Eur Law J 12(1):15–40
Bibliography 449

Muir E (2014) The fundamental rights implications of EU legislation: some constitutional


challenges. Common Market Law Rev 51(1):219–246
Nanopoulos E (2013) Trust issues and the European Common Asylum System: finding the right
balance. Camb Law J 72(2):276–280
Neframi E (2010) The duty of loyalty: rethinking its scope through its application in the field of EU
external relations. Common Market Law Rev 47(2):323–359
Neuman GL (2008) Import, export, and regional consent in the Inter-American Court of Human
Rights. Eur J Int Law 19(1):101–123
O’Boyle M (2008) Electoral disputes and the ECHR: an overview. Strasbourg, 12 November 2008,
CDL-UD(2008)010
O’Cinneide C (2009) Human rights and within multi-layered systems of constitutional gover-
nance: rights cosmopolitanism and domestic particularism in tension. UCD Working Papers in
Law, Criminology & Socio-Legal Studies. Research Paper No. 12/2009
O’Meara N (2011) “A More Secure Europe of Rights?” The European Court of Human Rights, the
Court of Justice of the European Union and EU accession to the ECHR. German Law J 12
(10):1813–1832
Odermatt J (2014) The EU’s accession to the European Convention on Human Rights: an
international law perspective. Leuven Center for Global Governance Studies. Working Paper
No. 136
Olsen JB (2009–2010) Protecting fundamental rights and the evolving roles of the court of justice
of the European Union and the European Court of human rights; Europeanisation in action at
the supranational level. Edinburgh Student Law Rev 1(54):54–73
Orakhelashvili A (2003) Restrictive interpretation of Human Rights Treaties in the recent juris-
prudence of the European Court of Human Rights. Eur J Int Law 14(3):529–568
Orakhelashvilli A (2009) Peremptory norms in international law. Oxford University Press, New
York
Papangelopoulou E (2006) Introductory note to European Court of Human Rights (ECHR) Grand
Chamber: Bosphorus Hava Yollari Turizm Ve Ticaret Anonim Sirketi v. Ireland. Int Legal
Mater 45(1):133–135
Pasqualucci JM (1994/1995) The Inter-American Human Rights System: establishing precedents
and procedure in human rights law. Univ Miami Inter Am Law Rev 26(2):297–361
Pauwelyn J (2003) Conflict of norms in public international law. Cambridge University Press,
Cambridge
Pauwelyn J (2004) Bridging fragmentation and unity: international law as a universe of inter-
connected islands. Mich J Int Law 25:903–927
Payandeh M (2011) Constitutional review of EU law after Honeywell: contextualizing the
relationship between the German Constitutional Court and the EU Court of Justice. Common
Market Law Rev 48(1):9–38
Peers S (2006) Bosphorus – European Court of Human Rights. Eur Const Law Rev 2(3):443–455
Peers S (2012) The future of EU treaty amendments. Yearb Eur Law 31(1):17–111
Peers S (2013) Towards a new form of EU law? The use of EU Institutions outside the EU legal
framework. Eur Const Law Rev 9(1):37–72
Peers S (2014) The CJEU and the EU’s accession to the ECHR: a clear and present danger to
human rights protection. Blog Article. Available at: http://free-group.eu/2014/12/18/steve-
peers-the-cjeu-and-the-eus-accession-to-the-echr-a-clear-and-present-danger-to-human-
rights-protection/
Pérez AT (2012) Spanish Constitutional Court, constitutional dialogue on the European arrest
warrant: the Spanish Constitutional Court knocking on Luxembourg’s door; Spanish Consti-
tutional Court, Order of 9 June 2011, ATC 86/2011. Eur Const Law Rev 8(1):105–127
Perez AT (2014) Too many voice? The prior involvement of the Court of Justice of the European
Union. In: Kosta V et al (eds) The EU accession to the ECHR. Hart Publishing, Oxford
Pernice I, Kanitz R (2004) Fundamental rights and multilevel constitutionalism in Europe. White
Paper 7/04, Humboldt University, p 20
450 Bibliography

Petersmann E-U (2002) Constitutionalism and WTO law: from a state-centered approach towards
a human rights approach in international economic law. The political economy of international
trade law, 1st edn. Cambridge University Press, Cambridge, pp 32–67
Phelps J (2006) Comment: reflections on bosphorus and human rights in Europe. Tulane Law Rev
81:251–279
Press Release, ‘European Commission and Council of Europe kick off joint talks on EU’s
accession to the Convention on Human Rights’. European Commission, Strasbourg, 7 July
2010. IP/10/906
Quinn G (2001) The European Union and the Council of Europe on the issue of human rights:
twins separated at birth? McGill Law J 46(849–874):853
Quirico O (2010) Substantive and procedural issues raised by the accession of the EU to the
ECHR. Italian Yearb Int Law 20:31–53
Raba K (2013) The accession of the European Union to the European Convention for the
protection of human rights and fundamental freedoms—overview of the accession agreement.
ERA Forum 14:557–572
Ragolle F (2003) Access to justice for private applicants in the community legal order: recent (R)
evolutions. Eur Law Rev 28(1):90–101
Reestman JH (2005) Primacy of union law. Eur Const Law Rev 1(1):104–107
Report of the ‘Study Group on Fragmentation of International law: Difficulties arising from the
Diversification and Expansion of International Law’. Session of 18 July 2003, ILC, 55th A/
CN.4/L.644
Report of the ECJ submitted to the Inter-Governmental Conference of 1996. Available at: http://
europa.eu.int/comm/eurostat/research/index.htm?http://europa.eu.int/en/comm/eurostat/
research/viros/search3.htm&1
Report to the Committee of Ministers on the elaboration of legal instruments for the accession of
the European Union to the European Convention on Human Rights. Steering Committee for
Human Rights (CDDH), (CM/Del/Dec(2011)1126/4.1, CM(2011)149). Committee of Minis-
ters of Council of Europe, 1145th meeting – 13 June 2012. Available at: http://www.coe.int/t/
dghl/standardsetting/hrpolicy/Accession/Accession_documents/2012_CM_Decision_acces
sion_en.pdf
Ribble MR (2010–2011) I Don’t Trust Your Judgment: the European Convention on Human
Rights meets the European Union on new grounds? Pa State Int Law Rev 29(1):211–232
Rieder C (2005) Protecting human rights within the European Union: who is better qualified to do
the job-the European Court of Justice or the European Court of Human Rights? Tulane Eur
Civil Law Forum 20:73–107
Ritleng D (2012) The accession of the European Union to the European Convention on Human
Rights and fundamental freedoms: a threat to the specific characteristics of the European Union
and Union Law? University of Uppsala, Faculty of Law Working Paper 1. Available at: http://
uu.diva-portal.org/smash/record.jsf?pid=diva2:526830
Roberts A, Sivakumaran S (2012) Lawmaking by nonstate actors: engaging armed groups in the
creation of international humanitarian law. Yale J Int Law 37(1):108–152
Robertson AH (1959) The European Court of Human Rights. Int Comp Law Q 8(2):396–403
Rodley NS (1989) Human rights and humanitarian intervention: the case law of the world court.
Int Comp Law Q 38(2):321–333
Romano C (2007) Commission of the European Communities v. Ireland. Am J Int Law 101
(1):171–179
Ryngaert C (2011) The European Court of Human Rights’ approach to the responsibility of
member states in connection with acts of international organizations. Int Comp Law Q 60
(4):997–1016
Sabel CF, Gerstenberg O (2010) Constitutionalising an overlapping consensus: the ECJ and the
emergence of a coordinate constitutional order. Eur Law J 16(5):511–550
Sampford C (2005) Reconceiving the rule of law for a globalizing world. In: Zifca S (ed)
Globalization and the rule of law. Routledge, London
Bibliography 451

Sarvarian A (2014) The EU accession to the ECHR and the law of international responsibility. In:
Kosta V et al (eds) The EU accession to the ECHR. Hart Publishing, Oxford
Scheuner U (1975) Fundamental rights in European Community Law and in National Constitu-
tional Law. Common Market Law Rev 12:171–191
Schilling T (1996) The autonomy of the community legal order: an analysis of possible legal
foundations. Harvard Int Law J 37:389–435
Schimmelfennig F (2006) Competition and community: constitutional courts, rhetorical action,
and the institutionalization of human rights in the European Union. J Eur Public Policy 13
(8):1247–1264
Schreuer C (2008) What is a legal dispute? Available at: http://www.univie.ac.at/intlaw/
wordpress/pdf/95.pdf
Schütze R (2004) Parallel external powers in the European Community: from “Cubist” perspec-
tives towards “Naturalist” constitutional principles? Yearb Eur Law 23:225–274
Sellers M (2007) Autonomy in the law. Springer, Dordrecht
Sera JM (1996) The case for accession by the European Union to the European Convention for the
protection of human rights. Boston Univ Int Law J 14:151–186
Shany Y (2004) The first MOX Plant award: the need to harmonize competing environmental
regimes and dispute settlement procedures. Leiden J Int Law 17(4):815–827
Shapiro M (1993) The globalization of law. Indiana J Global Leg Stud. Symposium: The
Globalization of Law, Politics, and Markets: Implications for Domestic Law Reform 1
(1):37–64
Shaw M (2008) International law, 6th edn. Cambridge University Press, Cambridge
Shelton DL (1989) Private violence, public wrongs, and the responsibility of states. Fordham Int
Law J 13:1–34
Shelton D (2003) The boundaries of human rights jurisdiction in Europe. Duke J Comp Int Law
13:95–153
Simma B (1985) Self-contained regimes in international law. Netherlands Yearb Int Law 16:111–
136
Škugor EM (2012) European Union and the European Convention on Human Rights – the
accession of a giant. Pravnik 46(1):105–118
Smith M (2007) The European Union and International Order: European and Global Dimensions.
Eur Foreign Aff Rev 12(4):437–456
Sohn L (1995) How American international lawyers prepared for the San Francisco Bill of Rights.
Am J Int Law 89(1):540–553
Statements by the President, 26 October 2000. Address by H.E. Judge Gilbert Guillaume,
President of the International Court of Justice to the United Nations General Assembly.
Available at: http://www.icj-cij.org/court/index.php?pr=84&pt=3&p1=1&p2=3&p3=1
Streinz T (2013) Avoiding an Odyssey – the EU’s accession to the ECHR. VerfBlog, 2013/9/30.
Available at: http://www.verfassungsblog.de/en/avoiding-an-odyssey-the-eus-accession-to-
theechr/
Study Of Technical And Legal Issues Of A Possible EC/EU Accession To The European
Convention On Human Rights, Report adopted by the Steering Committee for Human Rights
(CDDH) at its 53rd meeting (25–28 June 2002). DG-II(2002)006. [CDDH(2002)010 Adden-
dum 2]
Suchocka H (2007) Opinion no. 405/2006 on the Constitution of Serbia (Parts V, 7–9, VI and
VIII). CDL(2007)005rev. European Commission for Democracy Through Law (Venice Com-
mission), Strasbourg, 6 March 2007
Sweet AS (2009) Constitutionalism, legal pluralism, and international regimes. Indiana J Global
Legal Stud 16(2):621–645
Tabarelli M (2013) The influence of the EU and the ECHR on “Parliamentary Sovereignty
Regimes”: assessing the impact of European integration on the British and Swedish judiciaries.
Eur Law J 19(3):340–363
452 Bibliography

Talbot C (2014) Human rights and competition law in a European context. BlogPost. Available at:
http://humanrights.ie/uncategorized/human-rights-and-competition-law-in-a-european-
context
Tamanaha BZ (2008) Understanding legal pluralism: past to present, local to global. Sydney Law
Rev 30:374–411
The European Court of Human Rights (2014) The ECHR in 50 Questions, p 8. Available at: http://
www.echr.coe.int/Documents/50Questions_ENG.pdf
Thym D (2009) In the name of sovereign statehood: a critical introduction to the Lisbon judgment
of the German Constitutional Court. Common Market Law Rev 46(6):1795–1822
Thym D (2013a) A Trojan Horse? Challenges to the Primacy of EU Law in the Draft Agreement
on Accession to the ECHR. VerfBlog, 2013/9/11. http://www.verfassungsblog.de/en/atrojan-
horse-challenges-to-the-primacy-of-eu-law-in-the-draft-agreement-on-accession-to-the-echr/
Thym D (2013b) Separation versus Fusion – or: how to accommodate national autonomy and the
charter? Diverging visions of the German Constitutional Court and the European Court of
Justice. Eur Const Law Rev 9(3):391–419
Timmermans C (2001) The constitutionalization of the European Union. Yearb Eur Law 21(1):1–
11
Timmermans C (2014) The accession of the EU to the ECHR and the Charter of Fundamental
Rights: enlarging the field of protection of human rights in Europe. In: Kosta V et al (eds) The
EU accession to the ECHR. Hart Publishing, Oxford
Tomuschat C (1983) Liability for mixed agreements. In: O’Keeffe D, Schermers HG (eds) Mixed
agreements. Kluwer, Deventer
Tomuschat C (2002) The international responsibility of the European Union. In: Cannizzaro E (ed)
The European Union as an actor in international relations. Kluwer Law International, The
Hague
Toth AG (1997) The European Union and human rights: the way forward. Common Market Law
Rev 34(3):491–529
Tridimas T (2006) The general principles of EU law, 2nd edn. Oxford University Press, New York
Tridimas T, Eeckhout P (1995) The external competence of the community and the case-law of the
court of justice: principle versus pragmatism. Yearb Eur Law 14(1):143–177
Tridimas T, Gutierrez-Fons JA (2008) EU law, international law and economic sanctions against
terrorism: the judiciary in distress? Fordham Int Law J 32(2):660–730
Tridimas T, Poli S (2008) Locus Standi of individuals under Article 230 (4): the return of
Euridice? In: Arnull A, Eeckhout P, Tridimas T (eds) Continuity and change in EU law –
essays in honour of Sir Francis Jacobs. Oxford University Press, Oxford
Tulkens F (2013) EU accession to the European Convention on Human Rights. Speech at: National
School of Judiciary and Public Prosecution (KSSIP) – Krakow (Poland), Friday 1 March 2013
Uerpmann R (2003) International Law as an Element of European Constitutional Law: Interna-
tional Supplementary Constitutions. Jean Monnet Working Paper 9/03. Available at http://
www.jeanmonnetprogram.org/archive/papers/03/030901-02.pdf
Usher JA (2005) Direct and individual concern – an effective remedy or a conventional solution.
Eur Law Rev 28(4):575–600
van den Berghe F (2010) The EU and issues of human rights protection: same solutions to more
acute problems? Eur Law J 16(2):112–157
Van Dooren MI (2009) The European Union and human rights: past, present future. Merkourios
70:47–52
van Gerven W (2005) The European Union – a polity of states and people. Hart Publishing, Oxford
van Rossem JW (2009) Interaction between EU law and international law in the light of Intertanko
and Kadi: the dilemma of norms binding the Member States but not the Community. Nether-
lands Yearb Int Law 40:183–227
Villiger ME (2009) Commentary on the 1969 Vienna Convention on the law of treaties. Brill/
Martinus Nijhoff Publishers, Leiden
Bibliography 453

Vink M et al (2009) Explaining the use of preliminary references by domestic courts in EU


Member States: a mixed-method comparative analysis. Paper presented in Panel 6B ‘Judicial
Politics in the EU and Beyond’, 11th Biennial Conference of the European Union Studies
Association, Friday April 24, 2009
Volbeda MB (2006–2007) The MOX Plant case: the question of supplemental jurisdiction for
international environmental claims under UNCLOS. Texas Int Law J 42:211–240
Von Bogdandy A (2000) The European Union as a human rights organization? Human rights and
the core of the European Union. Common Market Law Rev 37(6):1307–1338
von Bogdandy A (2012) The European lesson for international democracy: the significance of
Articles 9 to 12 EU Treaty for International Organizations. Eur J Int Law 23(2):315–334
von Bogdandy A, Ioannidis M (2014) Systemic deficiency in the rule of law: what it is, what has
been done, what can be done. Common Market Law Rev 51(1):59–96
Von Bogdandy A et al (2012) Reverse Solange – protecting the essence of fundamental rights
against EU Member States. Common Market Law Rev 49(2):489–519
Von Papp K (2013) Clash of “Autonomous Legal Orders”: can EU Member State courts bridge the
jurisdictional divide between investment tribunals and the ECJ? A plea for direct referral from
investment tribunals to the ECJ. Common Market Law Rev 50(4):1039–1082
Vranes E (2003) The final clauses of the Charter of fundamental rights – stumbling blocks for the
first and second convention. European Integration online Papers (EIoP) 7 (2003) N 7.
Available at: http://eiop.or.at/eiop/texte/2003-0007a.htm
Ward A (2001) Judicial architecture at the cross-roads: private parties and challenge to EC
measures post Jégo-Quéré. Cambr Yearb Eur Leg Stud 4:413–444
Weber GS (2007) Who killed the friendly settlement? The decline of negotiated resolutions at the
European Court of Human Rights. Pepperdine Disp Resolut Law J 7(2). http://digitalcommons.
pepperdine.edu/drlj/vol7/iss2/4
Weiler J (1981) The community system: the dual character of supranationalism. Yearb Eur Law 1
(1):267–306
Weiler J (1986) Eurocracy and distrust: some questions concerning the role of the European Court
of Justice in the protection of fundamental human rights within the legal order of the European
communities. Wash Law Rev 61:1103–1142
Weiler JHH (1992) Thou shalt not oppress a stranger: on the judicial protection of the human rights
of non-EC nationals—a critique. Eur J Int Law 3(1):65–91
Weiler J (1999) The constitution of Europe. Cambridge University Press, Cambridge
Weiler J (2008) Editorial: Kadi – Europe’s Medellin? EJIL Talk, Blog of the European Journal of
International Law. Available at: http://www.ejiltalk.org/letters-to-the-editor-respond-to-ejil-
editorials-vol-195/
Weiler JHH, Haltern U (1996) The autonomy of the community legal order through the looking
glass. Harvard Int Law J 37:411–448
Weiler JHH, Lockhart N (1995a) “Taking rights seriously” seriously: the European Court and its
fundamental rights jurisprudence – part I. Common Market Law Rev 32(1):51–94
Weiler JHH, Lockhart NJS (1995b) “Taking rights seriously” seriously: the European Court and its
fundamental rights jurisprudence – part II. Common Market Law Rev 32(2):579–627
Weiß W (2011) Human Rights in the EU: rethinking the role of the European Convention on
Human Rights after Lisbon. Eur Const Law Rev 7(1):64–95
Wessel RA (2000) The inside looking out: consistency and delimitation in EU external relations.
Common Market Law Rev 37(5):1135–1171
Wessel RA (2008) The EU as a party to international agreements: shared competences, mixed
responsibilities. In: Dashwood A, Maresceau M (eds) Law and practice of EU external
relations: salient features of a changing landscape. Cambridge University Press, New York
Wessel RA, Blockmans S (eds) (2013) Between autonomy and dependence: the EU legal order
under the influence of international organisations. T.M.C. Asser Press/Springer, The Hague
454 Bibliography

Wetzel JR (2003) Improving fundamental rights protection in the European Union: resolving the
conflict and confusion between the Luxembourg and Strasbourg Courts. Fordham Law Rev
71:2823–2862
White S (2010) The EU’s accession to the convention on human rights: a new era of closer
cooperation between the Council of Europe and the EU? New J Eur Crim Law 1(4):433–446
Wildhaber L (2002) A constitutional future for the European Court of human rights. Hum Rights
Law J 23:161–165
Williams A (2004) EU human rights policies: a study in Irony. Oxford University Press, Oxford
Wilson RR (1964) International law in new national constitutions. Am J Int Law 58(2):432–436
Woolcock S (2005) European Union trade policy: domestic institutions and systemic factors. In:
Kelly D, Grant W (eds) The politics of international trade in the twenty-first century: actors,
issues and regional dynamics. International political economy series. Palgrave, Basingstoke

Legal Acts and Documents

African Charter on Human and Peoples’ Rights. Adopted on 27 June 1981, OAU Doc. CAB/LEG/
67/3 rev. 5, 21 I.L.M. 58 (1982). Entered into force on 21 October 1986
American Convention on Human Rights. Adopted in José, Costa Rica, on 22 November 1969.
Entered into force on 18 July 1978
Amsterdam Treaty, Official Journal of the European Communities, OJ C 340 of 10.11.1997
CDDH-UE (2011)16fin. ‘Draft legal instruments on the accession of the European Union to the
European Convention on Human Rights’. Adopted on the 8th Working Meeting of the CDDH
Informal Working Group On The Accession Of The European Union To The European
Convention On Human Rights (CDDH-UE) with the European Commission. Strasbourg,
Monday 20 June–Friday 24 June 2011
Commission Working Document DS 1930/10, Council of the European Union, FREMP meeting
(Brussels, 10 January 2011) on the topic ‘Previous involvement of the Court of Justice in the
context of the accession of the European Union to the European Convention for the protection
of Human Rights and Fundamental Freedoms’. Brussels, 22 December 2010
Convention against Torture and Other Cruel, Inhuman or Degrading Treatment of Punishment.
Adopted and opened for signature, ratification and accession by General Assembly resolution
39/46 of 10 December 1984. Entry into force on 26 June 1987, in accordance with article 27 (1)
Council Common Position 2001/869/CFSP [2001] OJ L 325/1
Council Dec 98/685 [1998] OJ L 326/1
Council Decision 2002/628/EC [2002] OJ L 201/48
Council Decision of 24 June 2010 on the conclusion of the Statute of the International Renewable
Energy Agency (IRENA) by the European Union (2010/385/EU)
Council Regulation (EC) No 881/2002
Decision No. CM/882/2605201, Adopted by the Ministers’ Deputies at their 1085th meeting
(26 May 2010)
Directive 2010/64/EU of the European Parliament and of the Council of 20 October 2010 on the
right to interpretation and translation in criminal proceedings
Directive 2012/13/EU of the European Parliament and of the Council of 22 May 2012 on the right
to information in criminal proceedings
Directive 2012/29/EU of the European Parliament and of the Council of 25 October 2012
establishing minimum standards on the rights, support and protection of victims of crime
Discussion document of the Court of Justice of the European Union on certain aspects of the
accession of the European Union to the European Convention for the Protection of Human
Rights and Fundamental Freedoms, Luxembourg, 5 May 2010, para 9, available at: http://curia.
europa.eu/jcms/upload/docs/application/pdf/2010-05/convention_en.pdf
Bibliography 455

Document 47+1(2013)008rev2, Final Report to CDDH. Fifth Negotiation Meeting Between The
CDDH Ad Hoc Negotiation Group And The European Commission On The Accession Of The
European Union To The European Convention On Human Rights. Strasbourg, Wednesday
3 April – Friday 5 April 2013
Document Technical and legal issues of a possible EC/EU accession to the European Convention
on Human Rights, CDDH(2002)010
Draft Agreement on a European Common Aviation Area, Re (Opinion 1/00) [2002] E.C.R. I-3493
Draft Articles On The Law Of Treaties Between States And International Organizations Or
Between International Organizations With Commentaries 1982. Text adopted by the Interna-
tional Law Commission at its thirty-fourth session, in 1982, and submitted to the General
Assembly as a part of the Commission’s report covering the work of that session (at para. 63).
The report, which also contains commentaries on the draft articles, appears in Yearbook of the
International Law Commission, 1982, vol. II, Part Two
Draft Articles On The Law Of Treaties With Commentaries 1966. Text adopted by the Interna-
tional Law Commission at its eighteenth session, in 1966, and submitted to the General
Assembly as a part of the Commission’s report covering the work of that session (at para.
38). The report, which also contains commentaries on the draft articles, appears in Yearbook of
the International Law Commission, 1966, vol. II
Draft Articles On The Responsibility Of International Organizations (DARIO). ‘Adopted by the
International Law Commission at its sixty-third session, in 2011, and submitted to the General
Assembly as a part of the Commission’s report covering the work of that session (A/66/10,
para. 87).’
ECJ’s Rules of Procedure (Official Journal of the European Communities) L 176 of 4 July 1991
European Agreement relating to Persons Participating in Proceedings of the European Court of
Human Rights of 5 March 1996 (ETS No. 161)
European Arrest Warrant Framework Decision. Council Framework Decision 2002/584/JHA of
13 June 2002 on the European arrest warrant and the surrender procedures between Member
States
European Convention on Human Rights and its Protocols 1950 (Rome), as amended by Protocols
No. 11 and No. 14. Available at: http://conventions.coe.int/Treaty/en/Treaties/Html/005.htm
European Council decision on the drawing up of a Charter of Fundamental Right of the European
Union, available at: http://www.consilium.europa.eu/ueDocs/cms_Data/docs/pressData/en/ec/
kolnen.htm
European Court of Human Rights, Unilateral declarations: policy and practice, 2012, available at:
http://www.echr.coe.int/Documents/Unilateral_declarations_ENG.pdf
European Parliament Committee on Constitutional Affairs Resolution (May 2010) (P7_TA(2010)
0184). ‘Institutional aspects of accession by the European Union to the European Convention
for the Protection of Human Rights and Fundamental Freedoms’ in ‘European Parliament
resolution of 19 May 2010 on the institutional aspects of the accession of the European Union
to the European Convention for the Protection of Human Rights and Fundamental Freedoms
(2009/2241(INI))’. Adopted on Wednesday, 19 May 2010 – Strasbourg
European Parliament resolution of 19 May 2010 on the institutional aspects of the accession of the
European Union to the European Convention for the Protection of Human Rights and Funda-
mental Freedoms (2009/2241(INI))
Explanations Relating to the Charter of Fundamental Rights, OJ C 303, 14.12.2007
FAO Rules of Procedure, available at: http://www.fao.org/docrep/006/y4800e/y4800e0m.htm)
Fifth Negotiation Meeting Between the CDDH Ad Hoc Negotiation Group and the European
Commission for the EU accession to ECHR, on Strasbourg 3 – 5 April 2013
General Agreement on Privileges and Immunities of the Council of Europe of 2 September 1949
(ETS No. 2)
ILC ‘Fragmentation Of International Law: Difficulties Arising From The Diversification And
Expansion Of International Law’. Report of the Study Group of the International Law
456 Bibliography

Commission, Fifty-eighth session, Geneva, 1 May-9 June and 3 July-11 August 2006.
U.N. doc. A/CN.4/L.702 of 18 July 2006
ILC Draft Articles on Responsibility of International Organizations. Adopted by the International
Law Commission at its sixty-third session, in 2011, and submitted to the General Assembly as a
part of the Commission’s report covering the work of that session (A/66/10, para. 87). The
report will appear in Yearbook of the International Law Commission, 2011, vol. II, Part Two
ILC Responsibility of States for Internationally Wrongful Acts 2001 (ILC Articles on State
Responsibility). Text adopted by the Commission at its fifty-third session, in 2001, and
submitted to the General Assembly as a part of the Commission’s report covering the work
of that session. The report, which also contains commentaries on the draft articles, appears in
Yearbook of the International Law Commission, 2001, vol. II (Part Two). Text reproduced as it
appears in the annex to General Assembly resolution 56/83 of 12 December 2001, and
corrected by document A/56/49(Vol. I)/Corr.4
Interlaken Declaration, High Level Conference on the Future of the European Court of Human
Rights, 19 February 2010
Internal Agreement between Representatives of the Governments of the Member States, meeting
within the Council, on the Financing and Administration of Community Aid under the
Financial Protocol to the Partnership Agreement between the African, Caribbean and Pacific
States and the European Community and its Member States signed in Cotonou on 23 June 2000
and the allocation of financial assistance for the Overseas Countries and Territories to which
Part Four of the EC Treaty applies (2000/770/EC)’. OJ L 317, 15.12.2000
Joint Declaration, 27. 4. 77 Official Journal of the European Communities No C 103/1
Maastricht Treaty, Official Journal of the European Communities, OJ C 191 of 29.07.1992
Marrakesh Agreement Establishing the World Trade Organization. Signed in Marrakesh,
Morocco, on 15 April 1994, and officially came into force on January 1, 1995
Memorandum of Understanding Between the Council of Europe and the European Union, 10 May
2007. Available at: http://cor.europa.eu/en/about/interinstitutional/Documents/5fe3aa86-d3c2-
4ac2-a39c-c2ea21618ffe.pdf
Memorandum on the accession of the European Communities to the Convention for the Protection
of Human Rights and Fundamental Freedoms. COM (79) 210 final, 2 May 1979. Bulletin of the
European Communities, Supplement 2/79. [EU Commission - COM Document]
Montevideo Convention on the Rights and Duties of States 1933. This treaty was signed at the
International Conference of American States in Montevideo, Uruguay on December 26, 1933.
It entered into force on December 26, 1934
Official Commentary on DARIO. Adopted by the International Law Commission at its sixty-third
session, in 2011, and submitted to the General Assembly as a part of the Commission’s report
covering the work of that session (A/66/10). The report, which also contains commentaries to
the draft articles (para. 88), will appear in Yearbook of the International Law Commission,
2011, vol. II, Part Two
Protocol 6 to the General Agreement on Privileges and Immunities of the Council of Europe of
5 March 1996 (ETS No. 162)
Robert Schuman Declaration of 9 May 1950 (Paris)
Rules of Procedure of the Court of Justice of EU. 29 September 2012 [Official Journal L 265 of
29.9.2012]
Secretariat of the European Convention, Final report of the discussion circle on the Court of Justice
of 25 March 2003 (Document CONV 636/03) and Cover note from the Praesidium of 12 May
2003 (Document CONV 734/03)
Single European Act, Official Journal of European Communities, N L 169/2, d. 29.06.1987
Statute of International Court of Justice, available at: http://www.icj-cij.org/documents/index.
php?p1¼4&p2¼2&p3¼0&#CHAPTER_III
Statute of the Council of Europe, London, 5.5.1949. Available at: http://conventions.coe.int/
Treaty/en/Treaties/Html/001.htm
Bibliography 457

Statute of the International Renewable Energy Agency. Signed in Bonn, on Monday, 26 January
2009. World Conference Center (IRENA/FC/Statute)
Steering Committee For Human Rights (CDDH), Ad hoc terms of reference concerning accession
of the EU to the Convention given to the CDDH by the Ministers’ Deputies during their 1085th
meeting (26 May 2010). Strasbourg, 3 June 2010 CDDH(2010)008. Available at: http://www.
coe.int/t/dghl/standardsetting/hrpolicy/accession/Working_documents/ToR_CDDH-UE_en.
pdf
The 1970 Declaration on Principles of International Law Concerning Friendly Relations and
Co-operation among States (UN General Assembly resolution 2625 (XXV) - October 1970)
The European Convention, The Secretariat, Brussels, 22 October 2002, CONV 354/02 WG II
16 Report, Chairman of Working group II “Incorporation of the Charter/ accession to the
ECHR”, Final report of Working Group II
The International Convention on the Elimination of All Forms of Racial Discrimination. Adopted
and opened for signature and ratification by General Assembly resolution 2106 (XX) of
21 December 1965. Entered into force on 4 January 1969, in accordance with Article 19
Travaux préparatoires of ECHR. Official Version Of The Travaux Preparatoires Of The European
Convention On Human Rights ‘Collected edition of the “Travaux préparatoires” of the
European Convention on Human Rights / Council of Europe ¼ Recueil des travaux
préparatoires de la Convention européenne des droits de l’homme / Conseil de l’Europe’.
The Hague: Martinus Nijhoff, 1975–1985
Treaty on the European Union. Consolidated versions of the Treaty on European Union and the
Treaty on the Functioning of the European Union - Consolidated version of the Treaty on
European Union - Protocols - Declarations annexed to the Final Act of the Intergovernmental
Conference which adopted the Treaty of Lisbon, signed on 13 December 2007 - Tables of
equivalences. Official Journal C 326, 26/10/2012 P. 0001 – 0390
Treaty on the Functioning of the European Union. Consolidated versions of the Treaty on
European Union and the Treaty on the Functioning of the European Union - Consolidated
version of the Treaty on European Union - Protocols - Declarations annexed to the Final Act of
the Intergovernmental Conference which adopted the Treaty of Lisbon, signed on 13 December
2007 - Tables of equivalences. Official Journal C 326, 26/10/2012 P. 0001 – 0390
United Nations Charter. Signed on 26 June 1945, in San Francisco, at the conclusion of the United
Nations Conference on International Organization. Entered into force on 24 October 1945
United Nations Convention on the Law of the Seas. Signed at Montego Bay in Jamaica, on
10 December 1982
Vienna Convention on the Law of Treaties between States and International Organizations or
between International Organizations. Done at Vienna on 21 March 1986. Not yet in force.
Official Records of the United Nations Conference on the Law of Treaties between States and
International Organizations or between International Organizations, vol. II (United Nations
publication, Sales No. E.94.V.5)
Vienna Convention on the Law of Treaties. Signed at Vienna, on 23 May 1969 (UN Doc.
No. 18232). Entered into force on 27 January 1980. Available at: https://treaties.un.org/doc/
Publication/UNTS/Volume%201155/volume-1155-I-18232-English.pdf

Court Cases

2 BvE 2/08 Gauweiler v Treaty of Lisbon, judgment of 30 June 2009, available at http://www.
bundesverfassungsgericht.de/entscheidungen/es20090630_2bve000208en.html
AG Jacobs Opinion in Court of Justice of EU, Case 5/88, H. Wachauf v. Germany, of 13 July 1989
Bundesverfassungsgericht [BVerfG] [Federal Constitutional Court] Oct. 18, 1967,
22 Entscheidungen des Bundesverfassungsgerchts [BVerfGE] 223 (1967) (F.R.G)
458 Bibliography

BVerfGE, Solange II - Wünsche Handelsgesellschaft decision of 22 October 1986, BVerfGE


73, 339, case number: 2 BvR 197/83, Europäische Grundrechte-Zeitschrift)
BVerfGE 37, 271 2 BvL 52/71 Solange I-Beschluß (Internationale Handelsgesellschaft v Einfuhr-
und Vorratstelle für Getreide und Futtermittel)
Committee on Economic, Social and Cultural Rights, [The Right to Adequate Food (Art. 11 of the
International Covenant on Economic, Social and Cultural Rights)], General Comment
No. 12 (1999), E/C.12/1999/5
Council Decision 98/392/EC. OJ 1998 L 179
Constitutional Court of the Czech Republic, ÚS 19/08 Treaty of Lisbon, Judgment of 26 November
2008
Court of Justice of EU, Case 16/59, Geitling Ruhrkolhlen-Verkaufsgesellschaft Mbh vs. High
Authority of the European Coal and Steel Community, Judgment, Court of Justice of the
European Communities
Court of Justice of EU, ACAVand Others v Council, Case T 138/98 [2000] ECR II 341, para. 64
Court of Justice of EU, Accrington Beef, Case C-241/95 [1996] ECR I-6699
Court of Justice of EU, Agrover Srl v Agenzia Dogane Circoscrizione Doganale di Genova, Case
C-173/06 [2007] ECR I-8783
Court of Justice of EU, Åklagaren v Hans Åkerberg Fransson, Case C-617/10 [2013], 26 February
2013
Court of Justice of EU, Alassini, joint cases C-317 to 320/08, judgement of 18 March 2010, [2010]
I-2213
Court of Justice of EU, Alitalia v Commission, Case T-296/97 [2000] ECR II-3871
Court of Justice of EU, Andersson and Wåkerås-Andersson, Case C-321/97 [1999] ECR I-3551
Court of Justice of EU, Anklagemyndigheden v Peter Michael Poulsen and Diva Navigation Corp,
Case C-286/90 [1992] ECR I-6019
Court of Justice of EU, Annibaldi, Case C-309/96 [1997] ECR I-7493
Court of Justice of EU, Apple Computer International v Commission, Case T-82/06, order of
19 February 2008
Court of Justice of EU, Apple Computer International v Commission, Case T-82/06, order of
19 February 2008
Court of Justice of EU, Aprile v Amministrazione delle Finanze dello Stato, Case C-125/94 [1995]
ECR I-2919
Court of Justice of EU, Arizona Chemical and Others v Commission, Case T-369/03 [2005] ECR
II-5839, para. 65
Court of Justice of EU, Arizona Chemical and Others v Commission, Case T-369/03 [2005] ECR
II-5839
Court of Justice of EU, Asia Motor France and others v Commission, Case T-28/90 [1992] ECR
II-2285
Court of Justice of EU, Asocarne v Council, C-10/95 P [1995] I-4149
Court of Justice of EU, Associazione delle Cantine sociali venete v Ombudsman and Parliament,
Case T-103/99 [2000] ECR II-4165
Court of Justice of EU, Asteris and Others v Commission, Joined Cases 97/86, 99/86, 193/86 and
215/86 [1988] ECR 2181
Court of Justice of EU, Athinaı̈ki Techniki v Commission, Case C-521/06 P, judgment of 17 July
2008
Court of Justice of EU, Athinaı̈ki Techniki v Commission, para. 44
Court of Justice of EU, Atzeni and Others v Regione autonoma della Sardegna, Joined Cases
C-346/03 and C-529/03 [2006] ECR I-1875
Court of Justice of EU, Audiolux SA and Others v Groupe Bruxelles, Case C-101/08, Judgement of
the Court of 15 October 2009
Court of Justice of EU, Bakers of Nailsea, Case C-27/95 [1997] ECR I-1847
Court of Justice of EU, Bauer Verlag, C-368/95 [1997] ECR I-3689
Court of Justice of EU, Baumbast, Case C-413/99 [2002] ECR I-7091
Bibliography 459

Court of Justice of EU, Baustahlgewebe GmbH v Commission of the European Communities,


Case C-185/95, [1998] ECR 8417
Court of Justice of EU, Beauport v Council and Commission, Cases 103–109/78 [1979] ECR 17
Court of Justice of EU, Bessin et Salson, Case 386/87 [1989] ECR 3551
Court of Justice of EU, Binderer v Commission, Case 147/83 [1985] ECR 257, para. 13
Court of Justice of EU, Booker Aquaculture Ltd. v. The Scottish Ministers, Joined Cases C-20/00
& C-64/00, [2003] E.C.R. 1-7411
Court of Justice of EU, Borker, Case 138/80 [1980] ECR 1975
Court of Justice of EU, Bosman, Case C-415/93 [1995] ECR I-4921
Court of Justice of EU, Bosman, Case C-415/93 [1995] ECR I-4921
Court of Justice of EU, Bostock, Case C-2/92, [1994] ECR I-955
Court of Justice of EU, Branco v Commission, Joined Cases T 194/97 and T 83/98 [2000] ECR II
69
Court of Justice of EU, Buckl and Others v. Commission, Joined Cases C-15/91 and C-108/91
[1992] ECR I-6061
Court of Justice of EU, Buralux v Council, Case C-209/94 P ˛I1996) ECR 1-615
Court of Justice of EU, C.I.L.F.I.T. v Ministry of Health, Case 283/81, Judgment OF 6. 10. 1982
Court of Justice of EU, C-402/05 P and C-415/05 P Kadi and Al Barakaat International Foundation
v Council and Commission [2008] ECR I-6351
Court of Justice of EU, C-97/85 Deutsche Lebensmittelwerke v Commission [1987] ECR 2265
Court of Justice of EU, Case 1/58, Friedrich Stork & Co., Kohlengroßhandlung vs. High Authority
Of The European Coal And Steel Community, Court of Justice of the European Communities,
Judgment
Court of Justice of EU, Case 1/59, Macchiorati Dalmas e figli v. High Authority [1959]
Court of Justice of EU, Case 104/81 Hauptzollamt Mainz ν Kupferberg [1982] ECR 3641,
judgment
Court of Justice of EU, Case 11/82 A.E. Piraiki-Patraiki v Commission [1985] ECR 207
Court of Justice of EU, Case 11/82, Piraiki-Patraiki v. Commission [1985] ECR 207
Court of Justice of EU, Case 12/86 Demirel [1987] ECR 3719
Court of Justice of EU, Case 181/73 Haegeman [1974] ECR 449, judgment of 30 April 1974. Para
5
Court of Justice of EU, Case 22/70, Commission of the European Communities v Council of the
European Communities [1971] ECR 263
Court of Justice of EU, Case 222/84, Johnston v. Chief Constable of Royal Ulster Constabulary
[1986], judgement of 15 May 1986
Court of Justice of EU, Case 247/87 Star Fruit Company v Commission [1989] ECR 291
Court of Justice of EU, Case 25/62 Plaumann & Co v Commission [1963] ECR 95
Court of Justice of EU, Case 6/64 Costa v ENEL [1964] ECR 585
Court of Justice of EU, Case C 208/03 P Le Pen v Parliament [2005] ECR I 6051
Court of Justice of EU, Case C 263/02 P Commission v Jégo-Quéré [2004] ECR I 3425
Court of Justice of EU, Case C-12/86 Demirel [1987] ECR 3719
Court of Justice of EU, Case C-13/00 Commission v. Ireland, Judgment of the Court, 19 March
2002
Court of Justice of EU, Case C-159/90 Society for the Protection of Unborn Children Ireland
[1991] ECR I-4685
Court of Justice of EU, Case C-209/94P Buralux and Others v Council [1996] ECR I-615
Court of Justice of EU, Case C-25/94 Commission v Council (FAO) [1996] ECR I-1469
Court of Justice of EU, Case C-266/03 Commission v Luxembourg [2005] ECR I-4805
Court of Justice of EU, Case C-299/95, Kremzow v. Austria [1997] ECR I-2629
Court of Justice of EU, Case C-309/89 Codorniu SA v Council [1994] ECR I-1853
Court of Justice of EU, Case C-319/06 Commission v Luxembourg [2008] ECR 1-4323
Court of Justice of EU, Case C-354/87 Weddel v Commission [1990] ECR I-3847
Court of Justice of EU, Case C-355/95 P TWD v Commission [1997] ECR 1-2549
460 Bibliography

Court of Justice of EU, Case C-358/89 Extramet Industrie SA v. Council [1991] ECR I-2501
Court of Justice of EU, Case C-39/93 P SFEI and Others v Commission [1994] ECR I-2681
Court of Justice of EU, Case C-424/99 Commission v Austria [2001] ECR I-9285
Court of Justice of EU, Case C-459/03 Commission v Ireland (Mox Plant) [2006] ECR I-4635
Court of Justice of EU, Case C-57/95 France v Commission [1997] ECR I-1627
Court of Justice of EU, Case C-94/00 Roquette Freres [2002] ECR I-9011 and Case C-276/01
Steffensen [2003] ECR I-3735
Court of Justice of EU, Case T-112/98, Mannesmannr€ohren-Werke v Commission [2001] ECR
II-729
Court of Justice of EU, Case T-177/01 Jégo-Quéré v Commission [2002] ECR II-2365
Court of Justice of EU, Case T-18/10 Inuit Tapiriit Kanatami and Others v Parliament and Council
[2011] ECR II-0000
Court of Justice of EU, Case T-262/10 Microban International and Microban (Europe) v Com-
mission [2011] ECR II-0000
Court of Justice of EU, Case T-277/94 Associazione Italiana Tecnico Economica del Cemento
(AITEC) v Commission of the European Communities [1996] ECR II-351
Court of Justice of EU (CFI), Yassin Abdullah Kadi v Council of the European Union &
Commission of the European Communities, Case T-315/01, 21 September 2005
Court of Justice of EU, Case T-387/94 Asia Motor France SA v Commission [1996] ECR II-961
Court of Justice of EU, Cases 10 and 18/68 Societa ‘Eridania’ Zuccherifici Nazionali v Commis-
sion [1969] ECR 459
Court of Justice of EU, Chalkor v Commission, Case C-386/10 P [2011] ECR I-13085
Court of Justice of EU, Chevalley v Commission, Case 15/70 [1970] ECR 975
Court of Justice of EU, Cinetheque S.A. and Others v. Federation Nationale des Cinemas Francais,
Cases 60-61/84, 1985 E.C.R. 2627
Court of Justice of EU, Cinéthèque v Fédération nationale des cinémas français [1985] ECR 2605,
Judgment of 11 July 1985 in Joined Cases 60 and 61/84
Court of Justice of EU, Commission of the European Communities v Federal Republic of Germany
(International Dairy Arrangement), Case C-61/94 [1996] ECR I-3989
Court of Justice of EU, Commission of the European Communities v. Federal Republic of
Germany, Case C-61/94, Judgment
Court of Justice of EU, Commission v Austria, Case C-424/99 [2001] ECR I-9285
Court of Justice of EU, Commission v Council (FAO), Case C-25/94 [1996] ECR I-1469
Court of Justice of EU, Commission v Council, Case 22/70 [1971] ECR 263
Court of Justice of EU, Commission v Council, Case 45/86 [1987] ECR 1493
Court of Justice of EU, Commission v Council, Case C-25/94 [1996] ECR I-1469
Court of Justice of EU, Commission v Ireland (C-459/03) [2006]
Court of Justice of EU, Commission v. France (Case C-239/03, judgment, CJEU)
Court of Justice of EU, Commission v. Ireland (Mox Plant case), Case C-459/03 [2006] ECR
I-4635
Court of Justice of EU, Confederation Francaise Democratique du Travail v. the European
Communities, no. 8030/77 (D. 10.07.1978)
Court of Justice of EU, Costa v ENEL, Case 6/64 [1964] ECR 585
Court of Justice of EU, Costa/ENEL, 15 July 1964, Case 6/641
Court of Justice of EU, Court of First Instance of the European Communities, Yassin Abdullah
Kadi v. Council of the EU and Commission of the European Communities, Case T-315/01,
judgment of 21 September 2005
Court of Justice of EU, Court of First Instance, Yusuf v. Council and Commission, Case T-306/01
[2005] ECR II-3353
Court of Justice of EU, Cruciano Siragusa v. Regione, Case C-206/13, Judgement of the Court of
6 March 2014
Court of Justice of EU, Da Costa, Joined Cases 28-30/62, 1963 E.C.R. 31
Court of Justice of EU, Defrenne v Sabena, Case 43/75 [1976] ECR 455
Bibliography 461

Court of Justice of EU, Demirel V Stadt Schwäbisch Gmünd, Case 12/86, Judgment Of The Court,
30 September 1987, E.C.R. 3719
Court of Justice of EU, Demirel v. Stadt Schwäbisch Gmünd, Case 12/86, Judgment, CJEU,
30 September 1987
Court of Justice of EU, Demirel, Case 12/86 [1987] ECR 3719
Court of Justice of EU, Dereci, C-256/11, 15 November 2011
Court of Justice of EU, Deutsche Shell AG v Hauptzollamt Hamburg-Harburg, Case C-188/91
[1993] ECR I-363
Court of Justice of EU, Deutz und Geldermann v Council, Case 26/86 [1987] ECR 941
Court of Justice of EU, Dorsch Consult, Case C-54/96 [1997] ECR I-4961
Court of Justice of EU, Dreyfus v Commission, Case C-386/96 P [1998] ECR I-2309
Court of Justice of EU, DSTV v Commission, Case T-69/99 [2000] ECR II-4039, at para. 24
Court of Justice of EU, EKWand Wein & Co, Case C-437/97 [2000] ECR I-1157
Court of Justice of EU, Emesa Sugar, judgement, Case C-17/98 [2000] ECR I-665
Court of Justice of EU, ENU v Commission, Case C-107/91 [1993] ECR I-599
Court of Justice of EU, ERT Case, C-260/89, Reference for Preliminary Ruling
Court of Justice of EU, Europese Gemeenschap v Otis NV and Others, Case C-199/11 [2012]
Court of Justice of EU, Eurotunnel SA and Others v Sea France, Case C-408/95 [1997] ECR
I-6315
Court of Justice of EU, Express Dairy Foods, Case 130/79 [1980] ECR 1887
Court of Justice of EU, FMC and Others v Intervention Board For Agricultural Produce And
Another, Case C-212/94 [1996] ECR I-389
Court of Justice of EU, Foster and others v British Gas plc, Case C-188/89 [1990] ECR I-3313
Court of Justice of EU, Foto-Frost v Hauptzollamt Lübeck-Ost, Case 314/85 [1987] ECR 4199
Court of Justice of EU, France and Others v Commission, Joined Cases C-68/94 and C-30/95
[1998] ECR I-1375
Court of Justice of EU, France v Commission, Case C-327/91 [1994] ECR I-3641
Court of Justice of EU, France v Commission, Joined Cases 15/76 and 16/76 [1979] ECR 321
Court of Justice of EU, France v UK, Case 141/78 [1979] ECR 2923
Court of Justice of EU, France v. Commission, Case C-327/91 [1994] ECR I-3641 & CJEU,
Germany v. Council, Case C-122/95 [1998] ECR I-973
Court of Justice of EU, Gaston Schul Douane-expediteur, Case C-461/03 [2005] ECR I-10513
Court of Justice of EU, Geotronics v Commission, Case C-395/95 P [1997] ECR I-2271
Court of Justice of EU, Germany v. Council, Case C-280/93, Judgment of the Court, 5 October
1994
Court of Justice of EU, Gerritse, C-234/01 [2003] ECR I-5933
Court of Justice of EU, Gestetner Holdings v Council and Commission, Case C-156/87 [1990]
ECR I-719
Court of Justice of EU, Gestevision Telecinco v Commission, Case T-95/96, Judgment of 15.
9. 1999
Court of Justice of EU, Granaria v Council and Commission, Case 90/78 [1979] ECR 1081
Court of Justice of EU, Grant v South West Trains, C-249/96 [1998] ECR I-621
Court of Justice of EU, Greis Unterweger, Case 318/85 [1986] ECR 955
Court of Justice of EU, Grimaldi v Fonds Des Maladies Professionnelles, Case C-322/88 [1989]
ECR 4407
Court of Justice of EU, Guérin automobiles v Commission, Case C-282/95 P [1997] ECR I-1503
Court of Justice of EU, Guérin Automobiles v Commission, Case T-186/94 [1995] ECR II-1753
Court of Justice of EU, Gunella v Commission, Case 33/72 [1973] ECR 475
Court of Justice of EU, Hake, Case 75/69 [1970] ECR 535
Court of Justice of EU, Hauer v. Land Rheinland-Pfalz, Judgment of the Court of 13 December
1979, case 44/79 [1979] ECR 3727
Court of Justice of EU, Hauptzollamt Mainz v C.A. Kupferberg & Cie KG a.A., Case 104/81
[1982] ECR 3641
462 Bibliography

Court of Justice of EU, Hauptzollamt Mainz v CA Kupferberg & Cie KG aA, Case C-104/81
[1982] ECR 3641
Court of Justice of EU, Hermès International v FHT Marketing Choice BV, Case C-53/96 [1998]
ECR I-3603
Court of Justice of EU, Hermès International v FHT Marketing Choice BV, Case C-53/96 [1998]
ECR I-3603
Court of Justice of EU, Het Vlaamse Gewest v Commission, Case T-214/95 [1998] ECR II-717
Court of Justice of EU, Heylens, case C-222/86, judgement of 15 October 1987, [1987] 4097
Court of Justice of EU, IATA and ELFAA, Case C-344/04 [2006] ECR I-403
Court of Justice of EU, Iida, Case C-40/11 [2012] ECR
Court of Justice of EU, IBM v Commission, Case 60/81 [1981] ECR 2639
Court of Justice of EU, Imperial Tobacco and others, Case C-74/99 [2000] ECR I-8599
Court of Justice of EU, Infortec v Commission, Case C-12/90 [1990] ECR I-4265
Court of Justice of EU, International Chemical Corporation v Amministrazione Delle Finanze
Dello Stato, Case 66/80 [1981] ECR 1191
Court of Justice of EU, International Fruit Company v Produktschap voor Groenten en Fruit,
Joined Cases 21 to 24/72 [1972] ECR 1219
Court of Justice of EU, Internationale Handelgesellschaft mbh v. Einfuhr und Vorratsstelle für
Getreide und Futtermittel (case 11/70), 1970, ECR 1125
Court of Justice of EU, Internationale Handelsgesellschaft mbH v. Einfuhr-und Vorratsstelle fur
Getreide und Futtermittel, Case 11/70, [1970] E.C.R. 1125
Court of Justice of EU, Intertanko, C-308/06, [2008] ECR I-4057
Court of Justice of EU, Iposea v Commission, Case T-49/00, order of the Court
Court of Justice of EU, Irish Cement Ltd v Commission, Joined Cases 166 and 220/86 [1988] ECR
6473
Court of Justice of EU, Italia Testa, Case C-101/96 [1996] ECR I-3081
Court of Justice of EU, Japan Tobacco Inc and JT International SA v European Parliament and
Council, Case T-223/01 [2002] ECR II-3259
Court of Justice of EU, Japan Tobacco Inc and JT International SA v European Parliament and
Council
Court of Justice of EU, Jégo Quéré, Case C-263/02 P, judgement of 1 April 2004, [2004] I-3425
Court of Justice of EU, Johnston, Case 222/84 [1986] ECR 1651
Court of Justice of EU, Joined Cases 106-7/63 Alfred Toepfer and Getreide-Import Gesellschaft v
Commission [1965] ECR 405
Court of Justice of EU, Joined cases 166 and 220/86 Irish Cement Limited v Commission of the
European Communities [1988] ECR 6473
Court of Justice of EU, Joined Cases 21 to 24/72, International Fruit Company et al 1972
Court of Justice of EU, Joined Cases C-128/97 and C-137/97 Testa and Modesti [1998] ECR
I-2181
Court of Justice of EU, Joined Cases C-402/05 P and C-415/05 P, Yassin Abdullah Kadi and Al
Barakaat International Foundation v Council of the European Union and Commission of the
European Communities, [2008] ECR I-06351
Court of Justice of EU, Joined Cases T-213/95 and T-18/96 SCK and FNK v Commission [1997]
ECR II-1739
Court of Justice of EU, Kadi and Al Barakaat v. Council and Commission, Joined Cases C-402/05
P and C-415/05 P [2008] ECR I-6351
Court of Justice of EU, K€obler, Case C-224/01 [2003] ECR I-10139
Court of Justice of EU, Kolpinghuis Nijmegen, Case 80/86 [1987] ECR 3969
Court of Justice of EU, Korkmaz and Others v Commission, Case T-2/04 [2006] ECR II-32
Court of Justice of EU, Kremzow, Case 299/95, preliminary ruling, Court of Justice, 1997, p. 2645
Court of Justice of EU, Kremzow, Case C-299/95, [1997] ECR I-2629
Court of Justice of EU, KSH v Council and Commission, Case 101/76 [1977] ECR 797
Court of Justice of EU, Kücükdeveci, Case C-555/07, [2010] ECR I-00365
Bibliography 463

Court of Justice of EU, Lagardère SCA & Canal+ v. Commission, Case T-251/00 [2002] ECR
II-4825
Court of Justice of EU, Les Verts v European Parliament, C-294/83 [1986] ECR 1339
Court of Justice of EU, Lord Bethell v Commission, Case 246/81 [1982] ECR 2277
Court of Justice of EU, Lugano Convention, Opinion 1/03 [2006] ECR I-1145
Court of Justice of EU, Makhteshim-Agan and others v Commission, Case C-258/05, judgment of
28
Court of Justice of EU, Maurin, Case C-144/95 [1996] ECR I-2909
Court of Justice of EU, Mayr-Melnhof Kartongesellschaft mbH v Commission of the European
Communities, Case T-347/94 [1998] ECR II-1751
Court of Justice of EU, Meilicke v ADV/ORGA, Case C-83/91 [1992] ECR I-4871
Court of Justice of EU, Melki, joint cases C-188 and 189/10, judgment of 22 June 2010, [2010]
I-5667
Court of Justice of EU, Merck Genéricos – Produtos Farmacêuticos v Merck, Case C-431/05
[2007] ECR I-7001
Court of Justice of EU, Meroni, Joined Cases 21 to 26/61 [1962] ECR 73
Court of Justice of EU, Meryem Demirel v Stadt Schwäbisch Gmünd, Case 12/86 [1987] ECR
3719
Court of Justice of EU, N.M. v Commission and Council, order of the Court of Justice of
17 October 1984 in Joined Cases 83/84 and 84/84 [1984] ECR 3571
Court of Justice of EU, N.S. and Others, C-411/10 and C-493/10, 21 December 2011
Court of Justice of EU, National Farmers’ Union v Secrétariat général du gouvernement, Case
C-241/01 [2002] ECR I-9079
Court of Justice of EU, Nederlandse Antillen v Council, Case C-452/98 [2001] ECR 1-8973
Court of Justice of EU, Netherlands v Commission, Case C-164/02 [2004] ECR I-1177
Court of Justice of EU, Netherlands v Commission, Case C-308/95 [1999] ECR I-6513
Court of Justice of EU, Nold KG v. Commission, Case 4/73 [1974] ECR 491, judgment
Court of Justice of EU, Nordgetreide v Commission, Case 42/71 [1972] ECR 105
Court of Justice of EU, Olsen v Commission, Case T-17/02 [2005] ECR II-2031
Court of Justice of EU, Omega Spielhallen, Case C-36/02, [2004] ECR 1-9609
Court of Justice of EU, Opinion (on the Draft Convention of the International Atomic Energy
Agency on the Physical Protection of Nuclear Materials, Facilities and Transports) Opinion
1/78 [1978] ECR 2151
Court of Justice of EU, Opinion 1/00 [2002] ECR I-3493
Court of Justice of EU, Opinion 1/00 [2002] I-3493
Court of Justice of EU, Opinion 1/00 ECAA Agreement [2002] ECR I-349
Court of Justice of EU, Opinion 1/09 (Creation of a unified patent litigation system – European and
Community Patents Court – Compatibility of the draft agreement with the Treaties), 8 March
2011
Court of Justice of EU, Opinion 1/09 [2011] Agreement on the European and Community Patents
Court. ECR I-1137
Court of Justice of EU, Opinion 1/75 (OECD Local Cost Standard) [1975] ECR 1355
Court of Justice of EU, Opinion 1/78 International Agreement on Natural Rubber [1979] ECR
2871
Court of Justice of EU, Opinion 1/91 (1991) ECR I-6079
Court of Justice of EU, Opinion 1/91 (Draft Agreement Relating to the Establishment of the
European Economic Area) [1991] ECR I-6079
Court of Justice of EU, Opinion 1/91 [1991] E.C.R. I-6079
Court of Justice of EU, Opinion 1/91 [1991] ECR I-6079
Court of Justice of EU, Opinion 1/92 [1992] E.C.R. I-2825
Court of Justice of EU, Opinion 1/94, [1994] ECR I-5267
Court of Justice of EU, Opinion 2/00 [2001] ECR I-9713
Court of Justice of EU, Opinion 2/91 (Convention No. 170 ILO on safety . . .) [1993] ECR I-1061
464 Bibliography

Court of Justice of EU, Opinion 2/94 (Accession by the Community to the European Convention
for the Protection of Human Rights and Fundamental Freedoms), 28 March 1996
Court of Justice of EU, Opinion 2/94 [1996] ECR I-1759
Court of Justice of EU, Opinion 2/13 (Accession of the European Union to the European
Convention for the Protection of Human Rights and Fundamental Freedoms — Compatibility
of the draft agreement with the EU and FEU Treaties), 18 December 2014
Court of Justice of EU, Order of 9 January 2007 in Case T-127/05 Lootus Teine Osaühing v
Council
Court of Justice of EU, Orkem SA v. Commission, Case 374/87, [1989] ECR 3283
Court of Justice of EU, Orkem v Commission, Case 374/87 [1989] ECR 3283
Court of Justice of EU, P Athinaı̈ki Techniki v Commission, Case C-521/06, judgment of 17 July
2008
Court of Justice of EU, P Gestoras Pro Amnistı́a and Others v Council, Case C-354/04 [2007] ECR
I-1579
Court of Justice of EU, P Sodima v Commission, Case C-44/00 [2000] ECR I 11231
Court of Justice of EU, Parfums Christian Dior SA and Tuk Consultancy BV and Assco Gerüste
GmbH, Rob van Dijk, and Wilhelm Layher GmbH & Co. KG, Joined Cases 300/98 and 392/98
[2000] ECR I-11307
Court of Justice of EU, Parliament v Council and Commission, Joined Cases C-181/91 and C-248/
91 [1993] ECR I-3685
Court of Justice of EU, Parliament v Council, Case C-189/97 [1999] ECR I-4741
Court of Justice of EU, Parliament v Council, Case C-316/91 [1994] ECR I-625
Court of Justice of EU, Parliament v Council, Case C-540/03 [2006] ECR I-5769
Court of Justice of EU, Parti écologiste ‘Les Verts’ v European Parliament, Case 294/83 [1986]
ECR 1339
Court of Justice of EU, Parti Ecologiste ‘Les Verts’ v. European Parliament, Case 294/83, [1986]
ECR 1339
Court of Justice of EU, Parti Ecologiste les Verts v. European Parliament, Case 294/83 [1986]
ECR I-1368
Court of Justice of EU, Pfizer v Commission, Order of 2 June 2004 in Case T-123/03 [2004] ECR
II-1631
Court of Justice of EU, Piraiki-Patraiki and Others v Commission, Case 11/82 [1985] ECR 207
Court of Justice of EU, Piraiki-Patraiki v Commission
Court of Justice of EU, Plaumann v Commission, Case 25/62 [1963] ECR 95
Court of Justice of EU, Portugal v. Council, Case C-149/96, Judgment of the Court, 23 November
1999
Court of Justice of EU, Poulsen and Diva, Case C-286/90 [1992] ECR 1-6019
Court of Justice of EU, PreussenElektra, Case C-379/98 [2001] ECR 1-2099
Court of Justice of EU, Producteurs de Fruits v Council, Joined Cases 16 and 17/62 [1962] ECR
471
Court of Justice of EU, Promusicae, Case C-275/06, [2008] ECR I-271
Court of Justice of EU, R & V Haegeman v Belgian State, Case 181/73 [1974] ECR 449
Court of Justice of EU, R & V Haegeman v Belgian State, Case 181/73 [1974] ECR 449
Court of Justice of EU, Racke v Hauptzollamt Mainz, Case C-162/96, judgment of 16 June 1998
Court of Justice of EU, Racke, Case C-162/96 [1998] ECR 1-3655
Court of Justice of EU, Rewe, Case 33/76, judgement of 16 December 1976, [1976] 1989
Court of Justice of EU, Rewe-Zentralfi nanz and Others v Landwirtschaftskammer für das
Saarland, Case 33/76 [1976] ECR 1989
Court of Justice of EU, Rheinmühlen-Düsseldorf v Einfuhr- und Vorratstelle für Getreide, Case
166/73 [1974] ECR 33
Court of Justice of EU, Roland Rutili v. The Minister for the Interior, Case 36/75, Reference for
Preliminary Ruling
Bibliography 465

Court of Justice of EU, Roquette Freres SA v Directeur General de la Concurrence, de la


Consommation et de la Repression des Fraudes, C-94/00 [2002] E.C.R. I-9011
Court of Justice of EU, Salamander & Others v Parliament & Council, Joined Cases T-172/98 and
T-175/98 to T-177/98;
Court of Justice of EU, Salonia v Poidomani and Giglio, Case 126/80 [1981] ECR 1563
Court of Justice of EU, Schecke GbR v. Land Hessen, Cases C-92-93/09. [2010] E.C.R. I-11117
Court of Justice of EU, Schmidberger v. Austria, Case C-112/00, [2003] ECR 1-5659
Court of Justice of EU, Schwarze v Einfuhr und Vorratsstelle fir Getreide und Füttermittel Case
16/65 [1965] ECR 877
Court of Justice of EU, Scottish Sofi Fruit Growers v Commission, Order in Case T-22/98 [1998]
ECR II-4219
Court of Justice of EU, Secretary of State for the Home Department v Akrich, Case C-109/01
[2003] ECR I-9607
Court of Justice of EU, Secretary of State for the Home Department v. Akrich, Case C-109/01,
[2003] E.C.R. I-9607
Court of Justice of EU, Sevince, Case C-192/89 [1990] ECR 3461
Court of Justice of EU, Sgarlata v Commission, Case 40/64 [1965] ECR 215
Court of Justice of EU, Simutenkov, Case C-265/03 [2005] ECR I-2579
Court of Justice of EU, Société des Usines de Beauport v Council, Joined Cases 109/78 [1979]
ECR 17, (part III)
Court of Justice of EU, Sonito and Others v Commission, Case C-87/89 [1990] ECR I-1981
Court of Justice of EU, Sony Computer Entertainment Europe v Commission, Case T-243/01
Court of Justice of EU, Spain v Commission, Case C-443/97 [2000] ECR I-2415
Court of Justice of EU, Spain v. UK and Northern Ireland, Case 145/04, Judgment of 12 September
2006, ECR, 2006
Court of Justice of EU, Standesamt Stadt Niebüll, Case C-96/04 [2006] ECR I-3561
Court of Justice of EU, Standesamt Stadt Niebüll, Case C-96/04 [2006] ECR I-3561
Court of Justice of EU, Star Fruit v Commission, Case 247/87 [1989] ECR 291
Court of Justice of EU, Stauder v. City of Ulm, case 29/69 [1969] ECR 419, request for
preliminary ruling
Court of Justice of EU, Sulvida v Commission, Case T-184/04 [2005] ECR II-85
Court of Justice of EU, Sulvida v Commission, para. 14
Court of Justice of EU, T. Port v Bundesanstalt für Landwirtschaft und Ernährung, Case C-68/95
[1996] ECR I-6065
Court of Justice of EU, T. Port, Case C-68/95 [1996] ECR I-833, para. 53
Court of Justice of EU, Texaco A/S v Middelfaert Havn and others, Joined Cases C-114/95 and
C-115/95 [1997] ECR I-4263
Court of Justice of EU, Textilwerke Deggendorf, Case C-188/92 [1994] ECR I-833
Court of Justice of EU, TF1 v Commission, Case T-17/96 [1999] ECR II-1757
Court of Justice of EU, The Queen ex parte Centro-Com v. HM Treasury, Case C-124/95 [1997]
ECR I-81
Court of Justice of EU, The Queen V Intervention Board For Agricultural Produce, Ex Parte
Accrington Beef And Others, [1996] 1-6699
Court of Justice of EU, Tielorassi v. Pliroforissis, Case C-260/89, [1991] E.C.R. 1-2925
Court of Justice of EU, Tillack v Commission, Case T-193/04 [2006] ECR II-3995
Court of Justice of EU, Toyo Bearing Company and Others v Council, Case 113/77 NTN [1979]
ECR 1185
Court of Justice of EU, Traghetti del Mediterraneo, Case C-173/03, judgement of 13 June 2006,
[2006] I-5177
Court of Justice of EU, UK v Commission, Case C-46/03 [2005] ECR I-10167
Court of Justice of EU, Unibet v Justitiekanslern, Case C-432/05 [2007] ECR I-2271
Court of Justice of EU, Unibet, case C-432/05, judgement of 13 March 2007, [2007] I-2271
Court of Justice of EU, UNICME v Council, Case 123/77 [1978] ECR 845
466 Bibliography

Court of Justice of EU, Union de Peque~nos Agricultores v Council, Case C-50/00 P [2002] ECR
I-6677
Court of Justice of EU, Union De Peque~nos Agricultores v Council, Case C-50/00
Court of Justice of EU, Usinor ν Commission, Joined Cases 81/85 and 119/85 [1986] ECR 1777
Court of Justice of EU, Vaassen (neé G€obbels), Case 61/65 [1966] ECR 261
Court of Justice of EU, Van Eick v Commission, Case 35/67 [1968] ECR 329
Court of Justice of EU, Van Gend en Loos v. the Netherlands, case 26/62 Judgment, CJEC
Court of Justice of EU, Van Schijndel, Joint cases C-430 and 431/93, judgement of 14 December
1995, [1995] I-4705
Court of Justice of EU, Wachauf v. Forstwirtschaft, Case 5/88, [1989] E.C.R. 2609
Court of Justice of EU, Weber v European Parliament, Case C-314/91 [1993] ECR I-1093
Court of Justice of EU, Weddel v Commission, Case C-354/87 [1990] ECR I-3847
Court of Justice of EU, Wünsche, Case 69/85 [1986] ECR 947
Court of Justice of EU, Zuckerfabrik Watenstedt v Council, Case 6/68 [1968] ECR 409
Decision on Admissibility of Czech Constitutional Court, 2008/12/02 - Pl. ÚS 12/08:
Non-Applicability of Contested Provision, available at: http://www.usoud.cz/en/decisions/?
tx_ttnews%5Btt_news%5D¼483&cHash¼6b8830d1a16e4ba14dcbc3b678bb7bb2
ECtHR, Ahmet Sadik v. Greece, Application no. 18877/91, JUDGMENT, STRASBOURG,
15 November 1996
ECtHR, Akdivar and Others v. Turkey, 16 September 1996, § 65, Reports 1996 IV
ECtHR, Akdivar and Others v. Turkey, Application no. 21893/93, JUDGMENT, STRASBOURG,
16 September 1996
ECtHR, Akdivar and Others v. Turkey, para. 69
ECtHR, Amuur v. France, Application no. 19776/92, Judgment, Strasbourg, 25 June 1996
ECtHR, Aquilina v. Malta, Application no. 25642/94, Judgment, Strasbourg, 29 April 1999
ECtHR, Bankovic and others v. Belgium and 16 Other States
ECtHR, Bankovic v Belgium and others, Application no. 52207/99, Grand Chamber, Decision as
to the Admissibility
ECtHR, Behrami & Saramati, Grand Chamber, Decision As To The Admissibility, Application
no. 71412/01 Agim BEHRAMI and Bekir BEHRAMI v. France and Application no. 78166/01
by Ruzhdi Saramati v. France, Germany and Norway
ECtHR, Bekerman v. Liechtenstein (no. 15994/10), 29 November 2011
ECtHR, Biret, Application no. 13762/04, Decision as to admissibility of 09 December 2008
ECtHR, Bosphorus Hava Yollari Turizm ve Ticaret Anonim Şirketi v. Ireland, Application
No. 45036/95 (Judgment of 30 June 2005)
ECtHR, Bowman v. The United Kingdom, 141/1996/760/961, Judgment, Strasbourg, 19 February
1998
ECtHR, Burden v. The United Kingdom, Application no. 13378/05, JUDGMENT, STRAS-
BOURG, 29 April 2008
ECtHR, Cantoni v. France, No. 17862/91, D. 15/11/1996 (Grand Chamber)
ECtHR, Confédération Française Démocratique du Travail (CFDT) v European Communities
[1979]
ECtHR, Connolly v 15 Member States of the European Union (App. No. 73274/01) (Section V),
December 9, 2008
ECtHR, Cyprus v. Turkey (IV), Appl. No. 25781/94, Judgment of the Court of 10 May 2001
ECtHR, Cyprus v. Turkey, Application no. 25781/94, Decision on the Admissibility of 28 June
1996
ECtHR, D.H. and Others v. The Czech Republic, Application no. 57325/00, JUDGMENT,
STRASBOURG, 13 November 2007
ECtHR, De Wilde, Ooms and Versyp (“Vagrancy”) v. Belgium, Application no. 2832/66; 2835/
66; 2899/66, Judgment, Strasbourg, 18 June 1971
ECtHR, Demopoulos v Turkey (App. Nos 46113/99, 3843/02, 13751/02, 13466/03, 10200/04,
14163/04, 19993/04 & 21819/04) (2010)
Bibliography 467

ECtHR, Denmark v. Turkey, Appl. No. 34382/97, Judgment (Friendly Settlement) of the Court of
5 April 2000
ECtHR, Edwards v. United Kingdom, App. No. 13071/87, 15 Eur. Ct. H.R. (ser. A) 417, 431
(1992)
ECtHR, Financial Times Ltd and others v. The United Kingdom, Application no. 821/03, JUDG-
MENT, STRASBOURG, 15 December 2009
ECtHR, Gäfgen v. Germany, Application no. 22978/05, JUDGMENT, STRASBOURG, 1 June
2010
ECtHR, Gorraiz Lizarraga and Others v. Spain, Application no. 62543/00, Judgment, Strasbourg,
27 April 2004
ECtHR, Granger v. the United Kingdom (no. decision of 9 May 1988)
ECtHR, İlhan v. Turkey [GC], no. 22277/93, § 59, ECHR 2000-VII
ECtHR, Ilse Hess v. United Kingdom (Appl. No. 6231/73), Decision of 28 May 1975 on the
admissibility of the application
ECtHR, Ireland v. United Kingdom, Appl. No. 5310/71, Judgment of the Court of 18 January 1978
ECtHR, Jasinskis v. Latvia, Application no. 45744/08, Judgment, Strasbourg, 21 December 2010
ECtHR, K.S. and K.S. AG v. Switzerland, Application No. 19117/91, 12 January 1994, decision as
to admissibility
ECtHR, KAOS-GL v Turkey Application No. 4982/07
ECtHR, Karak o v. Hungary, Application no. 39311/05, JUDGMENT, STRASBOURG, 28 April
2009
ECtHR, Karner v. Austria, Application no. 40016/98, JUDGMENT, STRASBOURG, 24 July
2003
ECtHR, Klass and Others v. Germany, Application no. 5029/71, Judgment, Strasbourg,
6 September 1978
ECtHR, Kokkelvisserij U.A. v Netherlands, no 13645/05, 20 January 2009
ECtHR, Kozacio glu v. Turkey, Application no. 2334/03, Judgment, Strasbourg, 19 February 2009
ECtHR, Krčmář and Others v. the Czech Republic (no. 35376/97, § 45, 3 March 2000)
ECtHR, Kudła v. Poland, Application no. 30210/96, Judgment, Strasbourg, 26 October 2000
ECtHR, Lawless v. Ireland, decision of 30th August 1958 on the admissibility of application
No. 332/57, Yearbook of the Convention, Vol. 2, pp. 324–326
ECtHR, LCB v UK, Application no. 14/1997/798/1001, Judgment, Strasbourg, 9 June 1998
ECtHR, Liptay v. Hungary (no. 12144/09), 22 May 2012
ECtHR, Loizidou v. Turkey (preliminary objections), judgment of 23 March 1995, Series A
no. 310
ECtHR, Lüdi v. Switzerland. judgment of 15 June 1992, Series A no. 238, p. 18
ECtHR, M. & Co. v. Federal Republic of Germany, No. 13258/87, Dec. 9 February 1990
ECtHR, M.S.S. v. Belgium and Greece, Application No. 30696/09, 21 January 2011
ECtHR, Mamatkulov and Askarov v. Turkey [GC] - 46827/99, Judgment 4.2.2005
ECtHR, Mantovanelli v. France (18 March 1997, § 36, Reports of Judgments and Decisions 1997
II)
ECtHR, Mathews, European Commission Of Human Rights, Application No. 24833/94, Report of
the Commission, adopted on 29 October 1997
ECtHR, Matthews v. United Kingdom, Application no. 24833/94, D. 18 February 1999. (Judge-
ment, Grand Chamber)
ECtHR, McCann and Others v. The United Kingdom, Application no. 18984/91, Judgment,
Strasbourg, 27 September 1995
ECtHR, Micallef v. Malta, (Application no. 17056/06), Grand Chamber, Judgment, Strasbourg,
15 October 2009
ECtHR, Michaud v. France, Application No. 12323/11, 6 December 2012
ECtHR, Monnat v. Switzerland, Application no. 73604/01, Judgment, Strasbourg, 21 September
2006
468 Bibliography

ECtHR, M.S.S. v. Belgium and Greece, Application no. 30696/09, Judgment, Strasbourg,
21 January 2011
ECtHR, Nada v. Switzerland, Grand Chamber, Application no. 10593/08, Judgment, Strasbourg,
12 September 2012
ECtHR, Paksas v. Lithuania, Application No. 34932/04, 6 January 2011
ECtHR, Pafitis and Others v. Greece, Application no. 163/1996/782/983, Judgment of 26 February
1998
ECtHR, Ringeisen v. Austria, Application no 2614/65, JUDGMENT, STRASBOURG, 16 July
1971
ECtHR, Scordino v. Italy (No. 1), Application No. 36813/97, 29 March 2006
ECtHR, Sejdić and Finci v. Bosnia and Herzegovina, Applications nos. 27996/06 and 34836/06,
Judgment, Strasbourg, 22 December 2009
ECtHR, Selmouni v. France [GC], no. 25803/94, ECHR 1999-V
ECtHR, Slivenko v. Latvia, Appl. No. 48321/99, judgment of 9 Oct. 2003
ECtHR, Soering v. United Kingdom, ECtHR, Series A No. 161 (1989)
ECtHR, T.W. v. Malta [GC], no. 25644/94, § 34, 29 April 1999
ECtHR, Takis Demopoulos and Others v. Turkey, Grand Chamber, Decision As To The Admis-
sibility Of Application nos. 46113/99, 3843/02, 13751/02, 13466/03, 10200/04, 14163/04,
19993/04, 21819/04
ECtHR, Tauira and Others v. France, no. 28204/95, Commission decision of 4 December 1995,
Decisions and Reports (DR) 83-B
ECtHR, Tyrer v UK, No 5856/72, 25 April 1978
ECtHR, Ullens de Schooten and Rezabek v. Belgium, Applications nos. 3989/07 and 38353/07,
Judgement of 20 September 2011
ECtHR, Van Der Tang v. Spain, Application no. 19382/92, Judgment, Strasbourg, 13 July 1995
European Commission of Human Rights, Ilse Hess v. United Kingdom (Appl. No. 6231/73),
Decision of 28 May 1975 on the admissibility of the application
European Commission of Human Rights, N 235/56, Dec. 10.6.1958, Yearbook 2 p. 256 (300)
European Commission on Human Rights, Austria v. Italy (the ‘Pfunders’ Case), Appl. No. 788/60,
Decision as to Admissibility (11th January 1961)
European Committee on Social Rights, International Federation for Human Rights (FIDH)
v. France, Collective Complaint No. 14/2003. Decision on the merits, 8 September 2004
European Court of Human Rights (3d sect.), T.I. v. United Kingdom (Appl. No. 43844/98),
decision (inadmissibility) of 7 March 2000
European Court of Human Rights (GC), Assanidze v. Georgia (Appl. No. 71503/01), judgment of
8 April 2004
European Court of Human Rights (GC), Bankovic and others v. Belgium and 16 Other States
(Appl. No. 52207/99), decision (inadmissibility) of 12 December 2001
European Court of Human Rights (GC), Beer and Regan v. Germany (Appl. No. 28934/95),
judgment of 18 February 1999
European Court of Human Rights (GC), Ilascu and others v. Moldova and Russia (Appl.
No. 48787/99), judgment of 8 July 2004
European Court of Human Rights (plen.), Soering v. United Kingdom, judgment of 7 July 1989,
Series A, No. 161
Federal Republic of Germany Bundesverfassungsgericht, Second Senate of 30 June 2009, 2 BvE
2/08, 2 BvE 5/08, 2 BvR
1010/08, 2 BvR
1022/08, 2 BvR
1259/08, 2 BvR
182/09
Foto-frost, para. 14
Bibliography 469

Human Rights Committee, (Issues Relating to Reservations made upon Ratification or Accession
to the Covenant or the Optional Protocols thereto, or in Relation to Declarations under Article
41 of the Covenant), General Comment No. 24, 4 November 1994 (CC PR/C/21/Rev.1/Add. 6)
Human Rights Committee, Judge v. Canada, Communication No. 829/1998, final views of
20 October 2003 (CC PR/C/78/D/829/1998 (2003)
Human Rights Committee, H. v. The Netherlands, Communication No. 217/1986, U.N. Doc.
CCPR/C/OP/2 at 70 (1990)
ICJ Advisory Opinion on the Competence of the General Assembly for the Admission of a State to
the United Nations, I.C.J. Reports 1950
ICJ, Cameroon v Nigeria, Preliminary Objections Judgment, ICJ Reports, 1998
ICJ, Gabcı́kovo-Nagymaros Project (Hungary/Slovakia) case, Judgment, ICJ Reports, 1997
ICJ, Germany v United States of America (LaGrand), Merits, Judgment, 2001, available at icj-cij.
org/icjwww/idocket/igus/igusframe.htm
ICJ, Interhandel Case (Switz. v. U.S.), [1959] I.C.J. 6, 27 (Mar. 21)
ICJ, Nicaragua v United States of America, Merits, Judgment, ICJ Reports, 1986
ICJ, North Sea Continental Shelf (Federal Republic Of Germany/Denmark; Federal Republic Of
Germany/Netherlands), Judgment, ICJ Reports 1969, 20 February 1969
ICJ, Reservations to the Convention on the Prevention and Punishment of the Crime of Genocide,
ICJ Reports, 1951
ICJ, United Kingdom v Albania (Corfu Channel), ICJ Reports, 1949
ICTY Appeals Chamber, Prosecutor v Dusko Tadic. Decision on the Defence Motion for Inter-
locutory Appeal on Jurisdiction, Judgment of the Appeals Chamber. Case No IT-94-1-AR72
(2 October 1995)
Inter-American Commission of Human Rights, Coard v. United States, Report No 109/9 of
29 September 1999 (DC 215-6)
Inter-American Court of Human Rights, (‘The Effect of Reservations on the Entry into Force of
the American Convention on Human Rights (Arts. 74 and 75)’), Advisory Opinion OC 2/82
of 24 September 1982, Inter-American Court of Human Rights (Ser. A, no. 2) (1982)
Inter-American Court of Human Rights, (on the effect of reservations on the entry into force of the
American Convention on Human Rights) Advisory Opinion OC 2/82 of 24 September 1982
(Arts. 74 and 75), Series A, No. 2
Inter-American Court of Human Rights, Advisory Opinion OC 18/03 of 17 September 2003
(requested by the United Mexican States)
Inter-American Court of Human Rights, Aloeboetoe et al. v. Suriname, Judgment of September
10, 1993, (Ser. C) No. 15 (1994)
Inter-American Court of Human Rights, Exceptions to the Exhaustion of Domestic Remedies in
Cases of Indigency or Inability to Obtain Legal Representation Because of a Generalized Fear
Within the Legal Community (Arts. 46(1), 46(2)(a) and 46(2)(b) of the American Convention
on Human Rights), Advisory Opinion OC-11/90 of August 10, 1990
Inter-American Court of Human Rights, Hilaire v. Trinidad and Tobago, Preliminary Objections,
judgment of 1 September 2001, Series C, No. 80
Inter-American Court of Human Rights, Hilaire v. Trinidad and Tobago, Preliminary Objections,
judgment of 1 September 2001, Series C, No. 80
Inter-American Court of Human Rights, The Sawhoyamaxa Indigenous Community v. Paraguay
(judgment of 29 March 2006), Series C No. 146
Inter-American Court of Human Rights, Velasquez Rodrıguez v. Honduras, Judgment of July
29, 1988, Ser. C No. 4
International Court of Justice, (Reservations to the Convention on the Prevention and Punishment
of the Crime of Genocide), Advisory Opinion, I.C.J. Reports 1951, 19 (28 May 1951)
International Court of Justice, Armed Activities on the Territory of the Congo (Democratic
Republic of the Congo v. Rwanda), judgment of 3 February 2006 (Jurisdiction of the Court
and Admissibility of the Application)
470 Bibliography

International Court of Justice, Order of 8 April 1993 on the request for the indication of provisional
measures in the case of the Application of the Convention on the Prevention and Punishment of
the Crime of Genocide (Bosnia and Herzegovina v. Serbia and Montenegro). Separate opinion
of Judge ad hoc Elihu Lauterpacht (I.C.J. Reports 1993)
Ireland v United Kingdom 41 ILM 405 (2002)
Italian Constitutional Court, Frontini v Ministero delle Finanze, Case 183/73
ITLOS, Advisory Opinion of the Seabed Disputes Chamber of the International Tribunal for the
Law Of the Sea on Responsibilities and Obligations of States sponsoring persons and entities
with respect to activities in the Area, Available at: http://www.itlos.org/fileadmin/itlos/docu
ments/cases/case_no_17/adv_op_010211.pdf
Opinion by Advocate General Jacobs in Case C-50/00 P Uni on de Pequeňos Agricultores v
Council [2002] ECR I-6677
Opinion of Advocate General Capotorti Delivered On 30 May 1978, Case 149/77
Opinion of Advocate General Jacobs in Case C-358/89 Extramet Industrie v Council [1991] ECR
2501
Opinion of Advocate General Kokott, delivered on 17 January 2013, Court of Justice of EU, Case
T-18/10 Inuit Tapiriit Kanatami and Others v Parliament and Council [2011] ECR II-0000
Opinion of AG Kokkot, delivered on 13 June 2014, in case: Court of Justice of EU, Opinion 2/13
Opinion of Advocate General La Pergola delivered on 6 February 1997, P. 2635, Case. 299/95
Opinion of Advocate General Poiares Maduro delivered on 18 January 2006 in Case C-459/03
Commission v Ireland (Mox Plant) [2006] ECR I-4635
Opinion of Advocate General Warner in CAM v Commission, Case 792/79 R., delivered on
9 January 1980
Opinion of Advocate-General Lagrange, Case 1/58 Stork
Opinion of Advocate-General Trabucchi Delivered On 28 March 1974, Case. Nold KG 4/73
Opinion of AG Cruz Villalon (delivered on 17 February 2011), Case C-120/10, European Air
Transport SA v. Collège d’Environnement de la Région de Bruxelles-Capitale
Opinion of AG Darmon in Borelli, Case C-97/91 [1992] ECR I-6313
Opinion of AG Jacobs, Union De Peque~nos Agricultores v Council, Case C-50/00
Opinion of AG Mischo, Court of Justice of EU, Parliament v Council, Case 377/87 [1988] ECR
4017
Opinion of AG Ruiz-Jarabo Colomer in Case C-142/95 P Associazione agricoltori della provincia
di Rovigo and others [1996] ECR 1-6669
Opinion of AG Slynn in Case 246/81 Bethell [1982] ECR 2277
Opinion of AG Tesauro delivered on 13 November 1997, CJEU, Hermès International v FHT
Marketing, Case C-53/96 [1998] ECR I-3603
Opinion of Mr Advocate General Capotorti, Delivered On 8 November 1979, Case 44/79 Hauer
Opinion of AG Cruz Villalon delivered on 12 June 2012, on Case C-617/10, Åklagaren v Hans
Åkerberg Fransson
Opinion of Mr Advocate-General Dutheillet De Lamothe Delivered On 2 December 1970, Case
11/70
Opinion of Mr Advocate-General Karl Roemer delivered on 12 December 1962
Opinion of Mr. Advocate General Darmon, Demirel v. Stadt Schwäbisch Gmünd. Delivered on
19 May 1987
Opinion of Mr. Tesauro in Commission of the European Communities v Federal Republic of
Germany, Case C-61/94, delivered on 7 May 1996
Opinion of the Advocate-General Mischo (Berne Convention for the Protection of Literary and
Artistic Works), Commission v Ireland, Case C-13/00 [2001] ECR I-2943
Permanent Court of International Justice, Advisory Opinion on the Competence of the ILO to
Regulate Agricultural Labour, P.C.I.J. (1922), Series B, No. 2/3
Permanent Court of International Justice, Advisory Opinion on the Competence of the ILO to
Regulate Agricultural Labour, P.C.I.J. (1922), Series B, No. 2/3
Bibliography 471

Permanent Court of International Justice, Factory at Chorz


ow, Judgment No. 13, 13 September
1928, Series A, No. 17
Permanent Court of International Justice, Greco-Bulgarian Communities Advisory Opinion PCIJ
(1930), No. 17, Available at: http://www.worldcourts.com/pcij/eng/decisions/1930.07.31_
greco-bulgarian.htm
Polish Constitutional Court, Polish Membership of the European Union (Accession Treaty),
Judgment K18/04 of 11 May 2005
US Supreme Court, Hilton v. Guyot, 159 U.S. 113, 163/4 (1895)
US Supreme Court, Medellin v. Texas, 552 U.S. 491 (2008)
US Supreme Court, Murray v. The Charming Betsy, 2 Cranch 64, 118 (1804)

You might also like