‘Deliberative democratising’ of Australian treaty

making: Putting into context the significance of online
access to the treaty process
DAVID MASON*
Abstract
Some 20 years ago, a newly elected Australian Government announced in May 1996 its
intention to introduce reforms to existing Australian treaty-making processes in order,
it was claimed, to overcome a perceived ‘democratic deficit’ in treaty-making. Those
reforms sought to ensure enhanced accountability for, and increased transparency
about, all future treaties to which Australia intended becoming a party: the reforms
included inter alia the introduction of comprehensive parliamentary (and related civil
society) scrutiny of treaties prior to each treaty being ratified or acceded to. This was
achieved by establishing a Parliamentary Joint Stand Committee on Treaties (JSCOT)
tasked with publicly examining ‘National Interest Analyses’ of all treaties and
recommending whether to do so was in Australia’s national interest. A further pillar of
the suite of reforms introduced in 1996 was the establishment of online treaties service
giving online access (via AustLII) to all new treaties and related materials, in order to
make them immediately available to all members of civil society interested in examining
them. This last pillar has proved crucially important to enhancing the transparency of
Australia’s treaty making and is an ongoing process.
This article will place these developments in historical and theoretical context. It will
trace the long intellectual debate, fought over more than two centuries, dating back to
Sir William Blackstone’s ‘Commentaries on the Laws of England’ (1765-1769),
concerning whether that most quintessentially sovereign act – the making of a treaty
with another sovereign state – could or should ever be allowed to be subject to the
‘democratising’ influences of civil society. It will be argued that not only is that
theoretically possible, but that the 1996 reforms to the Australian treaty making process
in fact went a long way towards applying ‘deliberative democracy’ principles to how
Australia goes about making treaties. In so doing these reforms brought Australian
practices at least to the international norm, and in some respects well ahead of what
like-minded Westminster-style democracies have achieved.

1

Introduction

Traditionally, sovereign states have been the primary actors in a formalised
process of negotiating and concluding international treaties. While that still
remains the case it is also clear that other, non-state actors (NGOs) are

*

Executive Director, Treaties Secretariat, Department of Foreign Affairs and Trade,
Australia

2

Journal of Law, Information and Science

Vol 24(2) 2016

increasingly influencing treaty-making and the processes by which treaties
are constructed. Against that backdrop an important issue arises regarding
to what extent treaty-making could or should be influenced by actors outside
the executive branch of government, the body traditionally having the
exclusive prerogative power to make treaties. Indeed, that precise question
was the focus of an intense political and legal debate in 1990s Australia—and
to varying degrees in the United Kingdom, Canada and New Zealand—
concerning an alleged ‘democratic deficit’, or lack of proper accountability in
the treaty making processes of those countries. It was claimed at the time
that the ‘deficit’ arose because the subject matter covered in modern treaties
had expanded to such an extent that treaties were in effect becoming a form
of ‘global legislation’ for which the legislative branch of government was not
accountable. There was concern that no Parliamentarians—let alone civil
society (NGOs)—effectively were able to make any meaningful input into
the development of treaties.
Thus some 20 years ago, a newly elected Australian Government announced
in May 1996 its intention to introduce reforms to existing Australian treatymaking processes in order, it was claimed, to overcome this perceived
‘democratic deficit’ in treaty-making. Those reforms sought to ensure
enhanced accountability for, and increased transparency about, all future
treaties to which Australia intended becoming a party: the reforms included
inter alia the introduction of comprehensive parliamentary—and related
civil society—scrutiny of treaties prior to each treaty being ratified or
acceded to. This was to be achieved by establishing a Parliamentary Joint
Standing Committee on Treaties (JSCOT) tasked with publicly examining
‘National Interest Analyses’ of all treaties and recommending whether their
ratification was in Australia’s national interest. A further pillar of the suite of
reforms to be introduced was the establishment of online treaties service
giving online access to treaties and related materials, in order to make them
immediately available to all members of civil society interested in examining
them. This last pillar has proved crucially important to enhancing the
transparency of Australia’s treaty making, along with the introduction of
JSCOT and its associated national interest analyses (NIAs).
This article will place these developments in historical and theoretical
context. It will trace the long intellectual debate, fought over more than two
centuries, dating back to Sir William Blackstone’s Commentaries on the Laws of
England (1765–1769), concerning whether that most quintessentially
sovereign act – the making of a treaty with another sovereign state – could or
should ever be allowed to be subject to the ‘democratising’ influences of civil
society, however expressed. It will be argued that not only is that
theoretically possible, but that the 1996 reforms to the Australian treaty
making process in fact went a long way towards applying ‘deliberative
democracy’ principles to how Australia goes about making treaties. In so
doing these reforms brought Australian practices at least to the international

‘Deliberative democratising’ of Australian treaty making

3

norm, and in some respects well ahead of what like-minded Westminsterstyle democracies have achieved.

2

Treaties and Democracy

Treaties – written, legally binding agreements between governments – have
long been a feature of international relations. While their negotiation has
historically been seen as requiring secrecy—and consequently as only
capable of being negotiated by officials working in secret—the need for that
secrecy has been increasingly contested in Australia and elsewhere. Thus,
against the backdrop of (European) ‘secret treaties’ being widely regarded as
contributing to miscalculations leading to the First World War, United States
President Woodrow Wilson famously asserted in 1918, as the first of his
‘Fourteen Points for Peace’, that there should in future only be ‘open
covenants of peace, openly arrived at, after which there shall be no private
international understandings of any kind, but diplomacy shall proceed
always frankly and in the public view.’ Consistent with at least the spirit of
that principle, the Australian government in 1996 instituted treaty reforms
designed explicitly to address a perceived ‘democratic deficit’ in Australian
treaty making. Whether actors outside government (civil society) should
have some visibility of, and even input into, the treaty making process was
an issue barely addressed at the 1919 Versailles Peace Conference following
World War One: today it clearly has greater significance. In 1919, those
seeking access to the treaty making deliberations at Versailles were primarily
journalists: today thousands of individuals and organisations are potentially
affected and press to ensure that their views are heard. Moreover, notions of
what might constitute more ‘open and democratic’ forms of treaty making
have developed beyond simply ensuring that the public is informed by the
issuing of bland, formal, iterative communiques (as occurred at Versailles),
to establishing mechanisms which enable civil society potentially to have
some actual influence on the content of the treaties themselves.
1

Traditionally, democracy was held to be inimical to good foreign policy; but,
from the perspective of modern international relations, this view has little
salience. Indeed, this article examines the view that the concept and reach of
democracy needs to expand. It must be more than government by elected
representatives. It must moreover be more than mandatory consultations
between the executive and legislative branches on policy formulation. It is
argued that an additional attribute of democracy should be the right of

1

Woodrow Wilson, ‘Address to a Joint Session of Congress’ (Speech delivered at Joint Session of
Congress, 8 January 1918) quoted in Arthur S Link (ed), The Papers of Woodrow Wilson (Princeton
University Press, 1984) vol 45, 536. This passage comes from the first of Wilson’s famous
‘Fourteen Points’.

4

Journal of Law, Information and Science

Vol 24(2) 2016

ordinary constituents to be heard by law-makers on prospective laws,
including those emanating from treaties. This ‘opportunity for interlocution’
is depicted as fundamental ‘political capital’ for a liberal, pluralist society –
especially because it ensures that the views of minorities are heard.
Therefore, the argument is made that the structures of any liberal, pluralist
society are likely to be improved if they incorporate principles of what has
been characterised as ‘deliberative democracy’ in their law making,
including their treaty making processes.
2

Traditionally treaty making was not in any way connected with democracy.
Indeed the very concept of ‘democracy’ is semantically fraught. As John

3

2

See definition of ‘deliberative democracy’ by two of its major proponents, Amy
Gutmann and Dennis Thompson, who define it as ‘a form of government in which
free and equal citizens (and their representatives), justify decisions in a process in
which they give one another reasons which are mutually acceptable and generally
accessible, with the aim of reaching conclusions that are binding in the present on all
citizens but open to challenge in the future.’ Amy Gutmann and Dennis Thompson,
Why Deliberative Democracy (Princeton University Press, 2004) 7.

3

The modern definition of the term ‘democracy’ springs from the example of the
United States, which evolved representative liberal democracy where elected rulers’
power was alloyed by respect for individuals’ inalienable civil rights. (Cf Jurgen
Habermas, ‘Popular Sovereignty as Procedure’, in James Bohman and William Rehg
(eds), Deliberative Democracy: Essays on Reason and Politics (MIT Press, 1997) 35–65.)
The cornerstone for debates about the form that this new American democracy
would take was ‘classical democracy’ (that is, the ancient form of government
founded in Athens by Kleisthenes in 507BC). Nonetheless, classical democracy
differed from liberal democracy (where individuals’ civil rights are legally
protected). The demokratia of the ancient Greek republics established community rule;
but did not guarantee their citizens any legal rights against those communities. See:
Paul A Rahe, Republics Ancient and Modern (University of North Carolina Press, 1 ed,
1992) 19; John Uhr, Deliberative Democracy in Australia (Cambridge University Press,
1998) 21.
st

The ‘Founding Fathers’ of the United States were opposed to the tyrannical
majoritarianism of such ‘classical’ democracy, albeit their approaches differed on
how civil rights might trammel majoritarianism. Three examples will suffice:
a) Thomas Jefferson saw a government ruled by elected representatives as a
compromise between the ‘Leviathan’ of absolute monarchy and the majoritarianism
of pure democracy; but he feared that even this ‘middle way’ faced the pitfall of
elected politicians acting unjustly. Therefore, he supported a bill of rights as a
bulwark against congressional tyrants. Paul A Rahe, above n 3, 718–20; See also
Thomas L Pangle, The Spirit of Modern Republicanism (University of Chicago Press,
1988) 124–7.
b) Alexander Hamilton execrated ancient ‘direct’ democracy as follows:

‘It has been observed that a pure democracy if it were practicable would be the most
perfect government. Experience has proved that no position is more false than this.
The ancient democracies in which the people themselves deliberated never possessed
one good feature of government. Their very character was tyranny; their figure

‘Deliberative democratising’ of Australian treaty making

5

Dunn noted in his semantic history of democracy, Setting the People Free, the
term ‘democracy’ derives from the Greek word demokratia. Demokratia—
according to Dunn—described a particular form of government, which was,
for almost two millennia, overwhelmingly judged to be ‘grossly illegitimate
in theory and every bit as disastrous in practice.’
4

Even the great publicist for early American democracy Alexis de Tocqueville
considered that democracy vitiated foreign policy. Indeed, he saw
democracy as, systemically, quite deficient in matters of external affairs. To
quote him:
Foreign policy demands scarcely any of those qualities which are
peculiar to a democracy; on the contrary it calls for the perfect use of
almost all those qualities in which a democracy is deficient.
Democracy … fortifies the respect for law in all classes of society, but
it can only with great difficulty regulate the details of an important
undertaking, persevere in a fixed design, and work out its execution
in spite of serious obstacles. It cannot combine its measures with

deformity.’ Alexander Hamilton, ‘Speech Urging Ratification of the US Constitution)’
(Speech delivered in New York, 21 June 1788). It is received wisdom that Hamilton
objected to a bill of rights in the US Constitution for example Alexander Hamilton.
But, to be fair to him, he argued that ‘that the Constitution is itself, in every rational
sense, and to every useful purpose, A BILL OF RIGHTS.’ This was because some civil
rights—like habeas corpus—were guaranteed in the Constitution. Alexander Hamilton,
The Federalist No. 84 ‘Certain General and Miscellaneous Objections to the
Constitution Considered and Answered’ Independent Journal, July 16, July 26, August
9, 1788, <http://www.constitution.org/fed/federa84.htm>.
c) James Madison was even more disparaging of classical democracy than Hamilton.
Madison wrote: ‘Had every Athenian citizen been a Socrates … every Athenian
assembly would still have been a mob.’ Rahe, above n 3, 324. Madison initially
opposed a bill of rights for the United States because (like Hamilton) he believed the
US Constitution was a bill or rights. But (like Jefferson) he became a champion for
such a bill as an adjunct to that constitution. See generally Robert A Goldwin, From
Parchment to Power: How James Madison Used the Bill of Rights to Save The Constitution
(Aei Press, 1997).
Accordingly the United States adopted a ‘bill of rights.’ See generally Herbert J
Storing, ‘The Constitution and the Bill of Rights’, in Robert A Goldwin and William A
Schambra (eds), How Does the Constitution Secure Rights? (American Enterprise
Institute, 1985).
4

‘What is very strange indeed (in fact, quite bizarre) is the fact that this single term
[democracy] … should turn out to be the ancient Greek noun demokratia, which
originally meant … one particular form of government … overwhelmingly judged …
grossly illegitimate in theory and every bit as disastrous in practice.’ John Dunn,
Setting the People Free (Atlantic Books, 2005) 15.

6

Journal of Law, Information and Science

Vol 24(2) 2016

secrecy or await their consequences with patience. These are qualities
which are more characteristic of an individual or an aristocracy.
5

In short, to De Tocqueville, democracy was inimical to good foreign policy
making. He was by no means alone in this view. Over the centuries, howls
of derision erupted whenever diplomatic method took a seemingly
democratic turn. And much of this howling has come from ‘the best and the
brightest’ of the day. For example, the legendary 18 century lawyer Sir
William Blackstone in his legal classic Commentaries on the Laws of England
(1765-1769) expressed a common view of the time that legislative assemblies
should never be involved in the conclusion of treaties. Rather, such activities
should be left exclusively to the Executive Branch of government.
th

To put it in Sir William’s words:
It is impossible that individuals of a state, in their collective capacity,
can transact the affairs of that state with another community equally
numerous as themselves. Unanimity must be wanting to their
measures, and strength to the execution of their counsels. In the king
therefore, as in a centre, all the rays of his people are united, and
form by that union a consistency, splendor, and power, that make
him feared and respected by foreign potentates; who [Sir William
asked] would scruple to enter into any engagements, that must
afterwards be revised and ratified by a popular assembly.
6

Thereafter—as if to spite him—the American colonies revolted and installed
a constitution which gave their president the powers to make treaties ‘by and
with the Advice and Consent of the Senate … provided two thirds of
Senators present concur.’
7

5

Alexis de Tocqueville, Democracy in America (1835) Part 1, Chapter XIII. See also
Charles de Montesquieu, The Spirit of Laws (1748) Book XI (on ‘the separation of
powers’ to enforce a society’s security); Thomas Hobbes, Leviathan (1651) Chapters 17
and 18; John Locke, Two Treatises on Civil Government (1689) Vol. II, Chapter 12.

6

William Blackstone, Commentaries on the Laws of England (1765-1769) (Clarendon
Press,
Oxford,
UK),
Book
1,
Chapter
7,
<http://www.lonang.com/exlibris/blackstone/bla-107.htm>. Cf the view of John
Locke (the preceptor of rational liberalism) that continuous purely lawful rule was
impossible in foreign affairs (which were in ‘a state of nature’) and, accordingly,
executive governmental power should be broad. Pangle, above n 3, pp 255–6.
Further, Locke insisted that foreign policy should be conducted by a single man.
Rahe, above n 3, pp 474–5.

7

United States Constitution art II (2). For an analysis of the legal consequences of the US
President signing a treaty which remains unratified. See Curtis A Bradley, ‘Unratified

‘Deliberative democratising’ of Australian treaty making

7

As the commentator Hayden put it:
The United States, in fact, was introducing a new principle into the
diplomatic practice of the world. She had made her treaties a
supreme part of the law of the land and therefore had given the
[Senate] … a part in enacting them. … She was a democracy and as
such had declined to entrust the superlatively important function of
treaty-making to the executive alone … And in due time the world
consented to deal with the United States in the manner made
necessary by her form of government.
8

Actually, the Founding Fathers’ ideal was not easily realised. George
Washington (in keeping with the spirit of the new United States Constitution
(‘Constitution’)) entered the Senate in August 1789, presented a paper on his
proposed treaty with the southern Indians, and then asked Senators certain
questions regarding their advice and consent on the treaty. This was because
ideally the Senate was supposed to advise on the content of prospective
treaties. Ironically, Washington fretted over the Senators’ advice and their
possible delays in consent for his treaty. As John Quincy Adams recalls:
Washington left the Senate Chamber saying he would be damned if he ever
went there again. But Washington did return to the Senate and a new system
of treaty making evolved. The Senate never turned into the council of
advisers for proposed treaties that Washington and his compatriots had
hoped for. Such ‘deliberative micro-management’ proved impractical. But
the Constitution was ‘adapted’, so to speak, in a manner which meant that the
Senate was eventually regarded as having a power (under art II of the
Constitution) which was tantamount to that of a veto over treaties.
Conversely, the President was recognised as having the power to enter into
compacts with foreign governments without the advice and consent of the
Senate. Many such compacts were legally treaties on the international plane;
but were deemed not to be ‘treaties’ for the purposes of municipal American
9

10

Treaties, Domestic Politics, and the U.S. Constitution’ (2007) 48 Harvard International
Law Journal H307 <http://www.harvardilj.org/print/117>.
8

Ralston Hayden, The Senate and Treaties, 1789-1817 (Macmillan, 1920) 155–6.

9

United States Congress, Senate Committee on Foreign Relations, The Role of the Senate
in Treaty Ratification A Staff Memorandum to the Committee on Foreign Relations United
States Senate (University of Michigan Library, 1977) (Chair, J Sparkman) 34.

10

Ibid 1–4, 36–49.

8

Journal of Law, Information and Science

Vol 24(2) 2016

law. These treaties ‘which dared not speak their name’ became known as
‘Executive Agreements.’
11

In some ways, then, the vision of the American Founding Fathers prevailed.
Two centuries after their War of Independence it was unexceptional for
constitutions around the world to provide for intervention by the legislature
prior to the Head of State ratifying a treaty. Even back in 1919 Hayden
trumpeted:

12

13

England and practically every other democracy have now provided
some method by which the representatives of the people may have a
voice in determining what manner of treaties shall be made. The
success of the American experiment demonstrated the practicability
of such a system, and paved the way for similar democratic
developments in other nations.
14

It has been claimed that the US Senate’s participation in treaty making
epitomised a broader vision of the American Founding Fathers and a
recurrent purpose of their Constitution. The then US Senator, Barack Obama,
explained this ideal neatly:
What the framework of our Constitution can do is organize the way
by which we argue about our future. All of its elaborate machinery –
its separation of powers and checks and balances and federalist
principles and Bill of Rights – are designed to force us into a
conversation, a ‘deliberative democracy’ in which all citizens are
required to engage in a process of testing their ideas against an

11

See Amy Gilbert, Executive Agreements and Treaties (Thomas-Newell, 1 ed, 1973).
Chapter 1 for a taxonomy of Executive Agreements.

12

Alexander Hamilton wrote that the power to make treaties is neither a wholly
executive nor wholly legislative function and that the joint possession of that power
‘would afford a greater prospect of security, than the separate possession of either of
them.’ Alexander Hamilton, The Federalist No. 65 ‘The Powers of the Senate’ New York
Packet, March 7, 1788, <http://www.constitution.org/fed/federa64.htm>. The
American constitutional vision of the executive and legislative branches exercising
overlapping powers in the realm of foreign policy remains controversial to this day.
See Norman J Ornstein and Thomas E Mann, ‘When Congress Checks Out’ (2006)
85(6) Foreign Affairs 67–82.

13

‘A comparative study of the constitutions in force shows that in most countries the
intervention of parliament is required before final ratification by the Head of State.’
Jose Marie Ruda, The Final Acceptance of International Conventions (Stanley Foundation,
1976) 23.

14

Hayden, above n 8, 156

st

‘Deliberative democratising’ of Australian treaty making

9

external reality, persuading others of their point of view, and
building shifting alliances of consent. Because power in our
government is so diffuse, the process of making law in America
compels us to entertain the possibility that we are not always right
and to sometimes change our minds; it challenges us to examine our
motives and our interests constantly, and suggests that both our
individual and collective judgments are at once legitimate and highly
fallible.
15

In other words, American democracy is more than the right to elect a
government; it is, or should be, a process which makes people ‘have a
conversation’ about what is important. This ‘forced conversation’, this
institution of public deliberation, is intended to determine how ‘the majority’
should form in a democracy; which—by virtue of this defining process—
saves majority rule from its own base nature.
16

17

15

Barack Obama, The Audacity of Hope: Thoughts on Reclaiming the American Dream
(Crown, 2006) 92–3.

16

The relationship of liberal representative democracy to ‘deliberative democracy’ is
theoretically controversial. Russell Hardin argues that democratic representative
government of a large society ‘subverts deliberation.’ Russell Hardin, ‘Deliberation:
Method Not Theory’, in Stephen Macedo (ed), Deliberative Politics (Oxford University
Press, 1999) 103–19, 115. Cf Jurgen Habermas, ‘Popular Sovereignty as Procedure’, in
Bohman and Rehg (eds), above n 3, 67–91; Iris Marion Young, ‘Justice, Inclusion, and
Deliberative Democracy’, in Macedo (ed), Deliberative Politics (Oxford University
Press, 1999) 151–69.

17

Obama (above n 15, 92) does not claim that his insight about the US Constitution is
original to him. A strong case can be made that the herald of Obama’s concept of
‘deliberative democracy’ is John Dewey, who wrote:
the strongest point to be made on behalf of even such rudimentary political
forms as democracy has attained, popular voting, majority rule and so on, is
that to some extent they involve a consultation and discussion which concerns
social needs and troubles. …. De Tocqueville … pointed out in effect that
popular government is educative as other forms of political regulation are not.
It forces recognition that there are common interests, even though the
recognition of what they are is confused; and the need it enforces of discussion
and publicity brings about some clarification of what they are.’ John Dewey, The
Public and its Problems (Holt, 1936) 206–7.
But Dewey not only saw popular government as enforcing the need for deliberation,
(cf Obama, above n 15) he also argued that such deliberation gives minorities
purchase against the majority, and the masses leverage against technocrats. As
Dewey put it:

Majority rule, just as majority rule, is as foolish as its critics charge it with being.
But it never is merely majority rule. As a practical politician, Samuel J. Tilden,
said … ‘The means by which a majority comes to be a majority is the important
thing:’ antecedent debates, modifications of views to meet the opinions of

10

Journal of Law, Information and Science

Vol 24(2) 2016

Democracy’s genius is arguing about what matters. Foreign policy matters.
Therefore, it should not be surprising that democracies have seen
‘deliberative’ public opinion steadily increase its influence over their foreign
policy. Historically, not everyone has accepted this as a good thing. The
renowned British diplomatic theorist Sir Harold Nicholson—in the middle of
the twentieth century—wrote a polemic against both transparent diplomacy
and the ratification of treaties by legislatures. He argued that they were
blights on modern diplomatic method. This argument was largely premised
on how the ordinary masses, in matters of foreign policy, displayed a lack of
knowledge and understanding which cruelled their national interests. In
Nicholson’s lifetime, however, diplomacy’s remit moved from rudimentary
statecraft to complex law-making. Nicholson never saw how this shift would
capture—indeed demand—the involvement of popular opinion in
diplomacy. Presumably, he saw treaty making as a process of governmental
negotiation. Therefore, secrecy was desirable not only because it facilitated
frankness and compromises, but also because it excluded the glare of
uninformed public opinion which might skew objectives.
18

19

Of course, parliamentary democracy has many institutions which deliberate
secretly in order to govern. It is unremarkable for the important deliberations
of executive government to be confidential (Cabinet being a prime
20

minorities, the relative satisfaction given the latter by the fact that it has had a
chance and that next time it may be successful in becoming a majority. Think of
the meaning of the ‘problem of minorities’ in certain European states, and
compare it with the status of minorities in countries having popular
government. It is true that all valuable as well as new ideas begin with
minorities, perhaps a minority of one. The important consideration is that
opportunity be given that idea to spread and to become the possession of the
multitude. No government by experts in which the masses do not have the
chance to inform the experts as to their needs can be anything but an oligarchy
managed in the interests of the few. And the enlightenment must proceed in
ways which force the administrative specialists to take accounts of needs. The
world has suffered more from leaders and authorities than from the masses.’
John Dewey, The Public and its Problems (1936) 207–8.
Cf James Madison The Federalist No 10 ‘The Utility of the Union as a Safeguard
Against Domestic Faction and Insurrection’ Daily Advertiser, November 22 1787
<http://www.constitution.org/fed/federa10.htm> and James Bohman, Public
Deliberation: Pluralism, Complexity and Democracy (MIT Press, 1996) 1–2.
18

For an example of how ratification of treaties by a legislature can complicate
diplomacy see Lloyd E. Ambrosius, Woodrow Wilson and the American Diplomatic
Tradition: The Treaty Fight in Perspective (Cambridge University Press, 1990).

19

Harold Nicholson, The Evolution of Diplomatic Method: being the Chichele lectures
delivered at the University of Oxford in November 1953 (MacMillan, 1954) 89–90.

20

Hardin, above n 16, 103–19, 115–16 draws a distinction between deliberations on
public policy by officials and by ordinary citizens; concluding that the former is not

‘Deliberative democratising’ of Australian treaty making

11

example). So there is nothing peculiar in deliberations about the
quintessentially executive power of treaty making being done in secret. By
way of contrast, the parliamentary forum is distinctive in that its
‘deliberations’ are nearly always public. The assumption is that they are
public because law making is better done in public. A procedure for public
debate about law making makes for better laws.
21

22

23

An intriguing constitutional aspect of treaty making is that it fuses both law
making and executive power; which raises a question debated over the
centuries: whether or not such international law making should include
direct public scrutiny and debate as part of its deliberative process? In
considering this further, we must turn to the science of decision-making.
There is empirical evidence that groups of exclusive like-minded decision
makers deliberating among themselves do not reach equable compromises;
rather they radicalise. Ironically, they stray from ‘the Wisdom of the Crowd’
and tend towards the extremes of their ‘like-mindedness.’ Therefore,
engaging inclusive, unlike-minded individuals in deliberations who contest
24

25

democratic and the latter is impracticable as a method for governmental decision
making. However, in asserting that the theory of deliberative democracy focuses on
legislative and popular deliberation, he seemingly excludes from that ‘focus’
ordinary methods for giving citizens formal rights to a hearing by governmental
agencies.
21

Some parliamentary deliberations are done in camera.

22

Bohman argues that political deliberation is best carried out in an open public forum,
because the quality of the reasons for political decisions is likely to improve, which
naturally improves the outcomes of those decisions. James Bohman, above n 16, 27.

23

For a discussion of how discussion ‘adds value’ to decisions see James D Fearon,
‘Deliberation as Discussion’, in Jon Elster (ed), Deliberative Democracy (Cambridge
University Press, 1997) 44–68; Gutmann and Thompson, ‘Democratic Disagreement’,
in Stephen Macedo (ed), Deliberative Politics (Oxford University Press, 1999) 243–79;
Thomas Christiano, ‘The Significance of Public Deliberation,’ in Bohman and Rehg
(eds), above n 3, 243–77, 244–58 (who emphasises the intrinsic worth of public
deliberation); Christian F Rostboll, Deliberative freedom: deliberative democracy as critical
theory (State University of New York Press, 2008) 206–7, 151 (who posits the epistemic
value of public deliberation, which ‘frees’ us); Michael Rabinder James, Deliberative
Democracy and the Plural Polity (University Press of Kansas, 2004) (who argues that
‘plural’ deliberation provides ‘specifications for reforming processes and redesigning
institutions’); Stanley Fish, ‘Mutual Respect as a Device of Exclusion’, in Stephen
Macedo (ed), Deliberative Politics (Oxford University Press, 1999) 88–102.

24

James Surowiecki, The Wisdom of Crowds (Doubleday, 2004) 184–6.

25

Cass R Sunstein, ‘Misery and Company’ (22 October 2008) The New Republic 39–43,
<https://newrepublic.com/article/61718/misery-and-company>.

12

Journal of Law, Information and Science

Vol 24(2) 2016

each other’s views lessens polarisation, generates better options and fosters
the golden mean of compromise.
26

These examples vindicate the spirit of what President Obama has called
‘deliberative democracy’; but they fall well short of his definition of the term.
By ‘deliberative democracy’ we must surely be referring to a system where
any individual or group in civil society has the institutional right to be heard
by government and thereby to assist government explore morally justifiable
ways for solving problems that matter. This ‘deliberative system’ of
democracy rests upon two premises:
27

28

a) The first premise is philosophical, namely: that the essence of liberty
is trust in our fallibility. Democracy cherishes doubt. Therefore, a
democracy must aspire, philosophically, to a process that uses the

26

Where decisions are made in what have been properly assessed as simple contexts
(‘the realm of known knowns’) or, their polar opposite, chaotic contexts (‘the realm of
unknowables’) ‘command-and-control’ decision making generally works best: David
J Snowden and Mary E Boone, ‘A Leader’s Framework for Decision Making’, Harvard
Business Review (Watertown, Massachusetts, 2007) 68–76. Nonetheless, all decision
making is a process—which should include properly assessing the context in which a
decision must be made—and open inquisitive decision making processes usually
make for better decisions. David A Garvin and Michael A Roberto, 'What You Don't
Know About Making Decisions' (2001) 79(8) Harvard Business Review 108, 110–1. Cf
Jack Night, ‘Constitutionalism and Deliberative Democracy,’ in Stephen Macedo (ed),
Deliberative Politics (Oxford University Press, 1999) 15969; Amy Gutmann and Dennis
Thompson, ‘Democratic Disagreement’ (Belknap Press of Harvard University, 1996);
Stephen Macedo (ed), Deliberative Politics (Oxford University Press, 1999) 243–79 (see
especially at 245); cf Snowden and Boone, above n 26, 68–76, 74–5.

27

See generally Dewey, The Public and its Problems, above n 17, 166–219 on how true
liberty springs from the social communion of democracy (namely, individuals
cooperating democratically with each other to improve society and enjoying the
benefits of such cooperation). Cohen argues in a similar vein (albeit more prosaically
than Dewey) that when properly conducted ‘democratic politics involves public
deliberation focused on the common good.’ Joshua Cohen, 'Deliberation and Democratic
Legitimacy,' in James Bohman and William Rehg (eds), Deliberative Democracy: Essays
on Reason and Politics (MIT Press, 1997) 69.

28

James D Fearon notes that a major argument for ‘deliberation’ about political
decisions is that such deliberation is inherently moral: the right of all to participate in
the deliberation legitimises the cognate decision. Elster, above n 23, 44–68, 60–1.
However, according to Bohman, above n 16, at 27, ‘public deliberation’ is not so
much a form of discourse or argumentation as a joint, cooperative activity, which—
by definition—excludes much deliberation in public forums. (Cf Ian Shapiro ‘Enough
of Deliberation: Politics is about Interests and Power’ in Stephen Macedo (ed),
Deliberative Politics: Essays on Democracy and Disagreement (Oxford University Press,
1999) 28–38; Fish, above n 23, 88–102.)

‘Deliberative democratising’ of Australian treaty making

13

collective wisdom of ‘the public’ to test the rightness of our
decisions.
b) The second premise concerns practicalities. Good decision making
springs from ‘the Wisdom of the Crowd‘. But the empirical evidence
shows that this ‘wisdom’ is only induced when we aggregate the
decisions of disparate individuals who have the courage to say what
they really think because they are acting independently of each
other.
29

It is not just philosophy, but also practicalities which justifies democratic
decision making. In this context, James Surowiecki, the author of The Wisdom
of Crowds, epitomises the true wisdom and wise truth of democracy:
The decisions that democracies make may not demonstrate the
wisdom of the crowd. The decision to make them democratically
does.
30

A conclusion that may be drawn from the two aforementioned premises of
sceptical philosophy and epistemic practicalities is that in order to properly
express our doubts about ‘the people’s assembled wisdom we must amass
the ‘wisdom of the crowd.’ That is, the ‘majority opinion’ proposed by
representative government—or represented by government proposals—
must be challengeable by all the independent pluralities of ‘minority
opinions’ (which, in aggregate, are taken to be ‘the public’). In short, a
democracy must aspire to a system which not only assumes that its decisions
may be wrong, but also supplies ample opportunity for its public to
challenge and test their rightness. This conclusion is supported by a cognate
argument that fair procedures make for fair outcomes. The fairer laws are
perceived to be the more likely it is that they will be accepted by those whom
31

29

Surowiecki, above n 24, 40–65.

30

Ibid 271.

31

John Rawls posits that democracies’ constitutional ‘need for dubiety’ is an attribute of
their ‘reasonable pluralism.’ Rawls argues that, although there are no absolute truths,
as a matter of ‘public reason’ there must still be a peremptory acceptance of ‘a
constitutional democratic regime and its companion idea of legitimate law.’ John
Rawls, ‘The Idea of Public Reason Revisited,’ in John Rawls, The Law of Peoples: The
Idea of Public Reason Revisited (Harvard University Press, 2001) 131–2. Cf Peter G
Danchin ‘Whose public? Which Law? Mapping the internal/external distinction in
international law’, in Jeremy Farrall and Kim Rubenstein (eds), Sanctions,
Accountability and Governance in a Globalised World (Cambridge University Press, 2009)
27–52, 40–6.

14

Journal of Law, Information and Science

Vol 24(2) 2016

they govern. It is arguable that in this light, perhaps ‘deliberative
democracy’ attempts to do no more than make procedural systems of
governmental decision making fairer.
32

33

It is said that public participation in the deliberative processes of treaty
making tests and proves the utility of ‘deliberative democracy.’ Formal
parliamentary ‘Q&A‘ on foreign policy presents governments with new
demandeurs: ordinary constituents. This educates both governments and
constituents. Such education is argued to be a primary advantage of
democracy. The best education teaches us to ask the right questions; which,
in turn, so often averts the dearest education: missing the right answers.
Public deliberation draws out better questions and answers. From this it is
concluded that parliamentary deliberation about foreign policy increases
public policy making discipline. Logically, the more informed public opinion
is the better its contribution to good policy. But increasing social complexity
makes it harder for the public to be well informed. Nowadays, public
opinion needs to be apprised by more than the roar of a bullhorn soaring
above a soap-box. Those best positioned to inform the public about policy
are those who make policy. Public opinion empowers policy making; so, it is
argued, policy makers should empower public opinion.
34

35

36

3

Australia’s Changing Treaty Practice

Australia provides an interesting example of the general trend by
legislatures towards adopting more so called ‘democratic’ forms in the

32

Cf John Rawls’ ‘common good idea of justice’, which envisages ‘a family of
representative bodies’ whose role in society’s hierarchy is to participate in
consultation procedures. Rawls, above n 31, 71.

33

Cf Iris Marion Young, ‘Justice, Inclusion and Deliberative Democracy’, in Stephen
Macedo (ed), Deliberative Politics: Essays on Democracy and Disagreement (Oxford
University Press, 1999) 151–8; Night, above n 26, 159–69.

34

‘Expertise, the high degree of division of labour, new technologies, and many more
factors also seem to put many current issues beyond the grasp of even the bestinformed citizens. Popular sovereignty … seems to be the first casualty of
complexity.’ Bohman, above n 16, 151. Further, given that a single electoral vote
rarely effects an election result the rational voter cannot be expected to try that hard
to master such complexity, Macpherson, above n 3, 188–9.

35

‘There can be no public without full publicity in respect to all consequences which
concern it. Whatever obstructs and restricts publicity, limits and distorts public
opinion and checks and distorts thinking on social affairs.’ Dewey, above n 17, 167.

36

Cf Uhr, above n 3, 224–7.

‘Deliberative democratising’ of Australian treaty making

15

process of treaty making, including through the application of ‘deliberative
democracy’ principles to its treaty making processes in May 1996.
As with so much of our constitutional development, its origins lay in the
British experience. In 1924, the United Kingdom instituted a constitutional
policy known as ‘the Ponsonby Rule’: by which every treaty which required
ratification by the Crown would, after signature, be laid on the tables of both
Houses of Parliament for a period of 21 days prior to the treaty’s
ratification. Subsequently, in Australia in 1961, Prime Minister Menzies
followed the British lead by making a commitment that henceforth the texts
of treaties, which until then had not normally been brought to the attention
of Parliament, would be tabled in both Houses.
37

38

Still, many Australian parliamentarians regarded this mere tabling – without
provision for parliamentary consideration or debate – as an inadequate form
of review. So, they called for a more comprehensive system of parliamentary
scrutiny of treaties. In particular, the criticism of the treaty making process
began to grow among the States and Territories of Australia. A focus of this
criticism was that a treaty may expand the legislative power of the Federal
Parliament with respect to external affairs and thereby confine the States’
exercise of their executive and legislative powers. This developed with the
High Court’s interpretation of this legislative power in a way which the
States saw as a threat to their own powers.
39

40

Although the states had an important role in treaty making, prior to 1980
there was little judicial deliberation on the constitutional extent of this
capacity of the Federal Parliament to implement treaties and how that

37

For a history of the constitutional evolution of the Ponsonby rule see Ministry of
Justice, United Kingdom, The Governance of Britain: War Powers and Treaties: Limiting
Executive Powers (Consultation Paper CP26/07) (2007) [122–30]. See also Ministry of
Justice, United Kingdom, ‘Governance of Britain – Views Sought on Strengthening
the Role of Parliament’ (Media Release, 25 October 2007). A government paper The
Governance of Britain presented to Parliament (by Jack Straw MP, the Secretary of
State for Justice and Lord Chancellor) in July 2007 proposed consultations on
appropriate means to put the Ponsonby Rule on a statutory footing. See John Straw,
United Kingdom, The Governance of Britain (Paper presented to Parliament, July 2007).

38

Commonwealth, Parliamentary Debates, House of Representatives, 10 May 1961,
1693-4 (Robert Menzies, Prime Minister and Minister for External Affairs).

39

Brian Opeskin, ‘The Role of Government in the Conduct of Australia’s Foreign
Affairs’ (1994) 15 Australian Year Book of International Law 129, 138–9.

40

For example, Ninian Stephen, ‘Making Rules for the World’ (1995) 30(2) Australian
Lawyer, 13–14.

16

Journal of Law, Information and Science

Vol 24(2) 2016

implementation may encroach on states’ functions. Nonetheless, it was clear
enough—even back in the late 1930s—that the Australian Constitution
differed radically from the US and Canadian Constitutions in the way it
delimited powers over foreign policy. Unlike the US Constitution Australia’s
Constitution did not grant the federal government exclusive jurisdiction over
foreign policy.
41

42

Therefore, actions or laws of an Australian state government are not
unconstitutional—and consequently invalid—simply because they encroach
on matters of foreign policy. Unlike the Canadian Constitution Australia’s
Constitution did not grant the Australian states legislative powers which are
expressly defined and exclusively reserved to those states. Therefore, our
federal government does not have to rely on state laws to implement treaties
on certain subjects which are the constitutional preserve of the states.
Instead, Australia could, and did, follow a system of ‘co-operative
federalism’ in which our states voluntarily played an important role in
Australian treaty making. This co-operative system rested upon both a
constitutional fact and a political rationale: the constitutional fact is that only
a few legislative powers are vested exclusively in the Australian Parliament.
Consequently, legislation passed under the Federal Parliament's external
affairs may exist concurrently with state legislation unless the state
legislation is inconsistent. But federal legislation, if constitutionally valid,
43

44

45

41

For an historical summary of legal views on this capacity see David Solomon, The
Political Impact of the High Court (Allen & Unwin, 1992) 29–33.

42

See US v Belmont, 301 US 324 (1937).

43

Cf the opposite consequence in the United States in, for example, the case of
Bethlehem Steel Corporation v Board of Commissioners 276 Cal App 2d 299. See also
National Foreign Trade Council, Appellate Brief Filed by the National Foreign Trade
Council in its Case Against the State of Massachusetts' Burma Law (9 March 1999)
<http://archives.usaengage.org/archives/background/lawsuit/NFTCbrief.html>.

44

See Canada (AG) v Ontario (AG) [1937] AC 326 (‘Labour Conventions’). See also
MacDonald v Vapor Canada Ltd [1977] 2 SCR 134 at 168–9; C Wilfred Jenks, ‘The
Constitutional Capacity of Canada to Give Effect to International Labour
Conventions’ (1935) 17 (1) Journal of Comparative Legislation and International Law 12,
12–30. Friesen observes that ‘the Labour Conventions model’ of treaty-making creates
many difficulties and suggests (at 1450) that a ‘workable improvement’ would be to
give a federal government the power to pass laws to implement treaties, from which
the ‘sub-federal units’ (provinces) may ‘opt out.’ Jeffrey L Friesen, ‘The Distribution
of Treaty-implementing Powers in Constitutional Federations: Thoughts on the
American and Canadian Models’ (1994) 94 (4) Columbia Law Review 1415, 1415–50.

45

But cf Allan Gottlieb, Canadian Treaty-Making (Butterworths, 1968) 74–9.

‘Deliberative democratising’ of Australian treaty making

17

will prevail over inconsistent state legislation to the extent of such
inconsistency.
46

Accordingly, although the federal parliament has the power—through its
own federal legislation—to implement treaties and other matters concerning
‘external affairs’—Australia may also implement treaties through state laws.
The political rationale—obtaining for Australia’s Parliament as a corollary of
this constitutional fact—was that federal laws should not implement treaties
regarding those matters which were traditionally administered by the
states. This unsettled state of affairs ended in the 1980s, largely as a
consequence of the decisions in three key High Court cases. All three cases
were challenges to the Federal Parliament’s right to legislate in areas that
were traditionally reserved to the states. All three failed. In Koowarta v
Bjelke-Petersen the High Court upheld the Racial Discrimination Act 1975
(Cth) which implemented the International Convention on the Elimination of All
Forms of Racial Discrimination 1966. In Commonwealth v Tasmania and
Richardson v Forestry Commission of Tasmania the High Court upheld federal
Acts implementing the UNESCO Convention for the Protection of the World
Cultural and Natural Heritage 1972. After those cases, in the 1980s, complaints
about the treaty-making process in Australia began to grow – especially
among state governments.
47

48

49

50

51

52

That growing concern was exacerbated by the decision of the Australian
High Court in Minister of State for Immigration and Ethnic Affairs v Teoh where
it was held that the ratification of a treaty created a legitimate expectation
that the executive and its agencies would act in conformity with that treaty,
even when the treaty had not been implemented into domestic law. The
53

46

‘When a law of a State is inconsistent with a law of the Commonwealth, the latter
shall prevail, and the former shall, to the extent of the inconsistency, be invalid.’
Australian Constitution s 109.

47

Opeskin, above n 39, 142.

48

(1982) 153 CLR 168.

49

International Convention on the Elimination of All Forms of Facial Discrimination, open for
signature 21 December 1965, 660 UNTS 195 (entered into force 4 January 1969).

50

(1983) 158 CLR 1.

51

(1988) 164 CLR 261.

52

Convention for the Protection of the World Cultural and Natural Heritage, opened for
signature 16 November 1972, 1037 UNTS 151 (entered into force 7 December 1975).

53

(1995) 183 CLR 273 (‘Teoh’s Case’).

18

Journal of Law, Information and Science

Vol 24(2) 2016

Teoh’s Case principle was quickly repudiated by the Commonwealth
Government but nevertheless fuelled a passionate debate then taking place
about Australia’s involvement in international affairs and the lack of
parliamentary oversight. This debate had focused on the decision of Toonen v
Australia by the United Nations (UN) Human Rights Committee, which
found Australia in breach of the International Convention on Civil and
Political Rights because of a criminal ban on homosexuality in the state of
Tasmania. The case raised political issues, with some expressing concern
about the impact of UN bodies on Australian sovereignty, as well as
federation issues, given the Commonwealth ability to use the external affairs
power to over-ride Tasmanian law to achieve treaty compliance.
54

55

56

57

In 1995 the Federal Parliament’s Senate Legal and Constitutional References
Committee conducted hearings into the treaty making process, which
ventilated criticisms of that process. In November that year the Committee
recommended that legislation be enacted to establish a parliamentary
committee which would, among other things, report on proposals by
Australia to join any treaty. By then it was widely considered that there was
a ‘democratic deficit’ in the way that the Executive entered into treaties; so
some changes had to be made, and there was widespread and growing
political support for such changes.
58

In consequence, on 2 May 1996, Alexander Downer, the Minister for Foreign
Affairs in the newly elected Coalition Government, set out in Parliament the
government’s reform policy, to restore confidence in the Australian treaty

54

Joint Statement by the Minister for Foreign Affairs, Senator Gareth Evans QC, and
the Attorney-General Michael Lavarch MP: International Treaties and the High Court
in Teoh, 10 May 1995. See further W Lacey ‘In the Wake of Teoh: Finding an
Appropriate Government Response’ (2001) 29(2) Federal Law Review 219.

55

Communication No 488/1992, 50 sess, UN Doc CCPR/50/D/488/1992 (5 November
1992).

56

International Covenant on Civil and Political Rights, signed 16 December 1966, 999
UNTS 171, (1967) 6ILM368 (entered into force 23 March 1976).

57

See further Brian Opeskin and Donald Rothwell, ‘The Impact of Treaties on
Australian Federalism’ (1995) 27 Case Western Reserve Journal of International Law 1,
49–54.

58

For example, Stephen, above n 40, 13–14. Cf Daryl Williams, ‘Australia’s TreatyMaking Processes: The Coalition’s Reform Proposals’, in Philip Alston and
Madelaine Chiam (eds), Treaty-Making and Australia: Globalisation versus Sovereignty
(Federation Press, 1995) 185–95. The term ‘democratic deficit’ was reportedly coined
in the context of European Community institutions (see Australian Senate Legal and
Constitutional References Committee, Trick or Treaty? Commonwealth Power to Make
and Implement Treaties (1995) 229).

th

‘Deliberative democratising’ of Australian treaty making

19

making process by eliminating ‘the democratic deficit’ in that process. These
implemented many of the reforms recommended by the Senate inquiry,
introducing five key reforms to Australia’s treaty-making processes:
59

Tabling in Parliament of all treaty actions proposed by the
government for at least 15 joint sitting days before binding treaty
action is taken.

Preparation of a National Interest Analysis (NIA) for each treaty,
outlining information regarding including the obligations contained
in the treaty and the benefits for Australia of entering into the treaty.
The NIA must be tabled in Parliament and published on the Internet.

Establishment of the parliamentary JSCOT comprising 16 members
from government, opposition and minority parties, to inquire into
and make recommendations in relation to Australia’s entry into
treaties.

Establishment of the Treaties Council comprising the Prime Minister,
Premiers and Chief Ministers and an enhanced role for the
Commonwealth/State and Territories Standing Committee on
Treaties (SCOT) to improve the quality of State and Territory
participation in the treaty-making process.

Establishment of the Australian Treaties Library and the provision of
an online treaties service (via AustLII).
60

This last reform was particularly significant insofar as it made possible the
addressing of the ‘lack-of-information’ side of the ‘democratic deficit.’ The
stated aim of the government’s treaty process reform policy was to restore
confidence in the process by eliminating the ‘democratic deficit’ inherent in
it. Specifically, the new policy provided that the arrangements for
parliamentary scrutiny of treaties, through tabling them in Parliament,
would apply to all treaty actions, multilateral as well as bilateral, and extend
not only to new treaties but to all actions to amend, terminate or withdraw
from treaties.

59

Department of Foreign Affairs and Trade (Cth) ‘Government Announces Reform of
Treaty-Making (Media Release, FA29, 2 May 1996
<http://foreignminister.gov.au/releases/1996/fa29.html>.

60

Ibid.

20

Journal of Law, Information and Science

Vol 24(2) 2016

The effect of this policy—which has been reaffirmed by successive incoming
Australian governments since 1996—is that parliamentarians make
considered reports on proposed treaties before the government decides
whether Australia should join them. Nonetheless, the government does not
have to follow these reports; it can take a treaty action even where JSCOT
recommends against that. This parliamentary approval is not formally
required before the executive branch of government agrees to take a treaty
action. In this respect, the prerogative power of the executive under s 61 of
the Constitution has been regulated through the adoption of the 1996
procedures, but has not been formally limited in scope.
So, put simply: Australia's treaty making process normally means that
before Australia ratifies a treaty, that treaty will be tabled in Parliament with
an explanation from the relevant department on why the treaty will serve the
national interest. JSCOT—the parliamentary committee responsible for
treaties—considers the treaty, invites views from the general public, holds
public hearings. Importantly, it also considers the views of states and
territories. JSCOT then writes a report recommending whether (or not) the
treaty should be ratified—which is submitted to parliament and published.
The government then takes a decision whether to proceed with ratification of
the treaty.
61

Initially, before the new committee had even been set up, these sweeping
proposals were subjected to a degree of criticism. There were objections to
opening up the government’s executive treaty making responsibility to
parliamentary scrutiny and public consultation in this way. There were
suggestions that the Coalition was seeking to limit international
engagements through new treaty obligations because it believed that such
commitments somehow diminished Australian sovereignty. It was said,
moreover, that a cloud of uncertainty and inefficiency would hang over our
treaty negotiations; that Australian diplomacy would be hamstrung by the
legislature; that JSCOT would simply add a further layer of useless
bureaucracy to the treaty-making process; and so on.
None of those criticisms were in fact borne out, and have not been repeated
in the last twenty years. Rather, over the twenty years that the treaty reforms
have been in place, a range of different criticisms have been made,
culminating in those aired before a Senate Foreign Affairs, Defence and
Trade References Committee enquiry in 2015 into Australia’s treaty making

61

Where the Foreign Minister considers that delaying a treaty action would be
detrimental to the national interest then that treaty action may be taken before it is
tabled.

‘Deliberative democratising’ of Australian treaty making

21

process. That Committee’s Report entitled Blind Agreement: Reforming
Australia’s Treaty Making Process was tabled in Parliament on 25 June 2015.
Its conclusion acknowledged the successes of implementation of the five
pillars of reform but argued more was now needed, particularly with respect
to trade treaties. As its Report put it:
62

63

Much was made 20 years ago of a so-called 'democratic deficit'
surrounding treaty-making. The reforms introduced in the mid1990s, following the landmark Trick or Treaty? Report, strengthened
the treaty-making process and gave parliament a greater say through
the establishment of the Joint Standing Committee on Treaties
(JSCOT) and the mandatory tabling of treaties in both houses of
parliament. However, a ‘democratic deficit’ has remained a feature of
the process, albeit with a different complexion today as the scope
and reach of trade agreements into domestic law is unlike anything
previously seen. While the 1996 reform package was undoubtedly
ground-breaking at the time, twenty years on the global environment
in which trade agreements are negotiated and community
expectations of transparency and accountability have changed to
such an extent that the case for review and further reform is
compelling ... . Three key points were raised in evidence to the
inquiry. First, that there needs to be a significantly higher level of
consultation in treaty-making before agreements are signed and that
more information should be communicated to stakeholders and the
public about how agreements will affect them. Second, that
parliament should have opportunities to play a constructive role
during negotiations that goes beyond rubber-stamping agreements
after they are signed. Third, that proposed treaty action should be
subject to independent assessment at the commencement of
negotiations and monitoring and evaluation after implementation, to
ensure that mistakes and unintended consequences are not
repeated.
64

Unsurprisingly, Coalition Senators took a notably divergent view about the
need for further reform of the treaty-making process. In its dissenting report

62

Foreign Affairs, Defence and Trade References Committee (Cth), Defence and Trade
Inquiry into the Commonwealth’s Treaty-making Process, (Senate, 18 June 2015)
<http://www.aph.gov.au/Parliamentary_Business/Committees/Senate/Foreign_A
ffairs_Defence_and_Trade/Treaty-making_process>.

63

Foreign Affairs, Defence and Trade References Committee (Cth), Blind Agreement:
Reforming Australia’s Treaty-making Process (Senate, 2015).

64

Ibid ch 6, 6.2, 6.5.

22

Journal of Law, Information and Science

Vol 24(2) 2016

to the majority findings—released under the name of the Committee Deputy
Chairman Senator Chris Back—the Government Senators stated the
following:
Coalition members of the committee disagree with all of the findings
and recommendations of the majority report. We do not believe that
the evidence received by the Committee during the inquiry leads to
the argument that Australia’s treaty making process is in need of
reform ... . Since its introduction by the Coalition government in
1996, Australia’s treaty-making process has been subject to only
minor alterations. Government of either persuasion has made use of
the systems accepting the balance between the respective role of
parliament and the executive which is mandated by the Australian
Constitution… The process in place in Australia closely resembles
those operating in countries with comparable political system, such
as Canada and New Zealand ... . The Coalition agrees that effective
consultation is essential to getting the best outcomes from
negotiations, but considers that opposition criticism of DFAT’s
consultation process is over blown and borderline insulting. DFAT
has convened over 1000 briefing sessions with stakeholders on the
Trans-Pacific Partnership (TPP) alone since May 2011 ... . In short, the
Coalition members of the Committee see no reason to proceed with
an extensive reform agenda when the current treaty making system
is working well.
65

The divergence of views reflected in the above quotations is a relatively
recent phenomenon. They contrast in tone with the flavour of the 1999
Government review of the reforms which was widely advertised and
attracted 66 submissions. That review found that the 1996 reforms had
greatly increased scrutiny, transparency and consultation in the treatymaking process and community awareness of treaties. Later some further
reforms were made to the process. In reforms announced on 20 August 2002
the tabling period for treaty texts and NIAs was increased from 15 to 20
sitting days for a subset of treaties—those of major political, economic or
social significance. As well, NIAs were required to be accompanied by
additional background reports and an enhanced treaties database was
66

65

Ibid ‘Dissenting Report by Coalition Senators.’

66

The reforms were reviewed in 1999. See Department of Foreign Affairs and Trade,
Review of the Treaty-making Process (Department of Foreign Affairs and Trade,
Canberra, 1999).

‘Deliberative democratising’ of Australian treaty making

23

launched. A further refinement came in 2008 with the introduction of a
mechanism by which minor treaty actions with negligible financial or legal
effect would be referred to JSCOT without tabling unless JSCOT deemed
tabling necessary and with an Explanatory Statement rather than a National
Interest Analysis. Other improvements since 1996 have included the
requirement for each National Interest Analysis to include a comprehensive
attachment on consultation.
67

68

4

Conclusions

Evaluating the success (or otherwise) of the five key reforms of 1996, it
should be noted that over the past 20 years, JSCOT has considered over 800
treaties and produced over 160 Reports and, as was asserted in a 2001
Australian International Law Journal article by Glen Cranwell:
69

70

The depth and effectiveness of the consultative process appear[ed] to
have [been] improved by allowing interested organisations and
individuals access to a process that was previously difficult to
influence and perceived to be quite off-limits. In a debate in the
House of Lords in 1999, Lord Lester of Herne Hill noted that
Australia was well ahead of the United Kingdom in the scrutiny of
the treaty making process because the ‘Australian Senate (sic) has a
well-developed treaty scrutiny committee.’
71

Cranwell also cited Professor James Crawford as assessing ‘the changes
introduced in 1996 were no doubt useful and they certainly entail a greater

67

Department of Foreign Affairs and Trade, Submission No 59 to Senate Standing
Committee on Foreign Affairs Defence and Trade, Inquiry into the Commonwealth’s
Treaty-making Process, 25 February 2015, 9.

68

Ibid.

69

Joint Standing Committee on Treaties (Cth), ‘A History of the Joint Standing Committee
on Treaties: 20 Years’ Forward (Australian Government Publishing Service, Canberra,
2016) vi.

70

Glen Cranwell, ‘The Treaty Making Process in Australia’ (2001) 8 Australian
International Law Journal 177.

71

Ibid 206, citing Lord Lester of Herne Hill statement in the House of Lords, United
Kingdom, Hansard, House of Lords, 12 January 1999, Column 131.

24

Journal of Law, Information and Science

Vol 24(2) 2016

level of formal consultation and communication with the federal Parliament
and state governments then previously obtained.’
72

It therefore seems reasonable to conclude that most of the goals of the 1996
treaty reforms have been achieved, particularly those relating to consultation
and transparency. In respect of the former, the workings of JSCOT—and the
use of improved NIAs—have been of significance. Regarding the latter—
transparency—the ‘fifth’ pillar of the reforms (provision of an online treaties
via AustLII) has led to the establishment of a world-leading and uniquely
Australian Treaties database. As the Canadian academic, Joanna
Harrington, has commented:
73

to bolster these reforms, Australia also created an excellent online
treaty database, providing free access to treaty texts, their ratification
records, NIAs and detailed information on multilateral treaty actions
currently under negotiation, consideration or review by the
Australian government.
74

It is concluded, therefore, that to the extent implementation of the five
‘pillars of reform’ have contributed to a measure of ‘deliberative
democratising’ of Australian treaty-making, providing online access (via
AustLII) to the treaty process has been of particular significance.
All of which compels us to return to part of that quote of Sir William
Blackstone cited earlier in this article, namely his cri de coeur who would
scruple to enter into any engagements, that must afterward be revised and
ratified by a popular assembly?
75

The answer today can be gleaned from the fact that intervention by
legislatures and input by civil society into the treaty-making process is
increasingly the norm around the world, and that such intervention to
achieve, among other things, an opportunity for civil society representatives

72

James Crawford, ‘International Law and Australian Federalism: past, present and
future’, in Brian Opeskin and Donald Rothwell (eds), International Law and Australian
Federalism (Melbourne University Press, 1997) 325, 336–7.

73

See Department of Foreign Affairs and Trade (Cth) The Australian Treaties Database
<http://www.info.dfat.gov.au/treaties> (via DFAT) and Australasian Legal
Information Institute AustLII <http://www.austlii.edu.au/dfat/> (via AustLII).

74

Joanna Harrington, ‘Redressing the Democratic Deficit in Treaty Law Making: (Re)
Establishing a Role for Parliament’ (2005) 50(3) McGill Law Journal 465. See also
Harrison’s suggestion that JSCOT should be used as a model for other
Commonwealth countries, ibid 491.

75

See Blackstone, above n 6.

‘Deliberative democratising’ of Australian treaty making

25

to have their views heard about the content of treaties before they become
binding, reflects an application of ‘deliberative democratic’ principles to
treaty-making processes around the world, including Australia. And the
development in Australia of online access to the treaty-making process (via
AustLII) has contributed significantly to that achievement.

Court Decisions on the Internet:
Development of a Legal Framework in Europe
MARC VAN OPIJNEN*
Abstract
This paper discusses the gradual development of a legal framework in Europe regarding
the publication and accessibility of court decisions on the internet. The evolving
doctrine of the European Court of Human Rights on the public pronouncement of
judgment is assessed, as are developments in national legislations with regard to the full
or selected publication of case law. Specific attention is paid to data protection, Open
Data and the European Case Law Identifier.

1

Introduction

In most European countries with a civil law tradition, the dissemination of
(important) court decisions had been a monopoly of commercial legal
publishers for decades. As in many other societal areas the internet changed
everything: courts started to publish decisions themselves, thereby not only
disrupting the legal information market, but also contributing significantly
to the visibility of the judiciary and the transparency of justice.
While the pioneering started in a vacuum, a legal framework now is
gradually evolving. In this paper the four most important—and intertwined
—aspects of this legal framework will be reviewed. The first question is to
what extent there is any legal obligation to publish decisions on the internet.
To this end in Part 1 the evolving case law of the European Court of Human
Rights (ECHR) on the public pronouncement of the judgment will be
evaluated, as well as practice and legal framework regarding the publication
of decisions from European courts and the legal frameworks at the national
level, paying specific attention to the problem of selection. The overarching
problem of data protection is discussed in Part 2 and the rapid developments
with regard to Open Data are the subject of Part 3. To increase the usability
of all these data published, standardisation of identifiers and metadata is of

*

Mr. dr. M van Opijnen is adviser legal infomatics at the Publications Office of the
Netherlands (UBR|KOOP). This article is partly based on his PhD thesis On and In
the Web. How the Accessibility of Case Law can be Improved (Op en in het web. Hoe de
toegankelijkheid van rechterlijke uitspraken kan worden verbeterd), University of
Amsterdam, Boom Juridische Uitgevers, 2014). The text was concluded in September
2016.

Court decisions on the Internet

27

crucial importance. This will be discussed in Part 4, with a special focus on
the European Case Law Identifier. Some conclusions will be drawn in Part 5.

2

An Obligation to Publish

2.1 The Doctrine of the European Court of Human Rights

For the legitimacy of the judiciary it is of crucial importance that everybody
can see how justice is done. Therefore, art 6, para 1 of the European
Convention on Human Rights and Fundamental Freedoms reads:
Judgment shall be pronounced publicly but the press and public may
be excluded from all or part of the trial in the interests of morals,
public order or national security in a democratic society, where the
interests of juveniles or the protection of the private life of the parties
so require, or to the extent strictly necessary in the opinion of the
court in special circumstances where publicity would prejudice the
interests of justice.
So, while access to the hearing may be restricted, the judgment itself always
has to be pronounced in public – for the scrutiny of justice and to prevent
secret trials. Comparable provisions can be found in national constitutions
throughout Europe, but of course the question arises what exactly is meant
by this ‘public pronouncement of judgment.’
The European Court of Human Rights has developed an interesting doctrine
on this topic, which has to be outlined in some detail to comprehend how
internet databases fit in. In 1983 the Court formulated its basic rule in the
often-cited judgment in Pretto v Italy (‘Pretto’):
1

Many member States of the Council of Europe have a long-standing
tradition of recourse to other means, besides reading out aloud, for
making public the decisions of all or some of their courts, and
especially of their courts of cassation, for example deposit in a
registry accessible to the public. The authors of the Convention
cannot have overlooked that fact. … The Court therefore does not
feel bound to adopt a literal interpretation. It considers that in each
case the form of publicity to be given to the ‘judgment’ under the
domestic law of the respondent State must be assessed in the light of

1

(European Court of Human Rights, ECLI:CE:ECHR:1983:1208JUD000798477, 8
December 1983).

28

Journal of Law, Information and Science

Vol 24(2) 2016

the special features of the proceedings in question and by reference
to the object and purpose of Article 6 para. 1.
2

Making decisions available in a publicly available register as a way of public
pronouncement was sanctioned by the Court in the Pretto itself, and in Sutter
v Switzerland the Court sanctioned a practice whereby interested outsiders
have to establish an interest to consult or obtain a copy from full text
judgments from such a register and whereby only a selection of (important)
decisions is published. In their dissenting opinion Cremona, Ganshof van
der Meersch, Walsh and MacDonald JJ argued that ‘[p]ublic knowledge of
court decisions cannot be secured by confining that knowledge to a limited
class of persons,’ and the Court receded from Sutter in its decision in Szücs v
Austria of 1997, establishing a violation of art 6 by Austria on making access
to the full texts in the court registry dependent upon leave granted at the
discretion of the courts.
3

4

5

To meet the requirements of art 6, not only the individual judgment has to be
assessed. In Axen v Germany the absence of a public hearing, as well as a
public pronouncement of the decision of a court of highest instance rejecting
an appeal on formal grounds, was no breach of art 6 since:
6

[T]he object pursued by Article 6 para. 1 ... – namely, to ensure
scrutiny of the judiciary by the public with a view to safeguarding
the right to a fair trial – was achieved during the course of the
proceedings taken as a whole.
7

2

Ibid [26]. Because the European Case Law Identifier (ECLI), discussed below (in Part 4),
is intended as a neutral citation and suffices to find judgments online, case references
in this article deviate from the Australian Guide to Legal Citation.

3

(European Court of Human Rights, ECLI:CE:ECHR:1984:0222JUD000820978, 22
February 1984) (‘Sutter’).

4

(European Court of Human Rights, ECLI:CE:ECHR:1997:1124JUD002060292, 24
November 1997).

5

Moser
v
Austria
(European
Court
of
Human
Rights,
ECLI:CE:ECHR:2006:0921JUD001264302, 21 September 2006) led to a comparable
conclusion.

6

(European Court of Human Rights, ECLI:CE:ECHR:1983:1208JUD000827378, 8
December 1983) (‘Axen’).

7

Ibid [32].

Court decisions on the Internet

29

While in Axen the requirements were met by a preceding court, in Lamanna v
Austria it was ruled that the full public pronouncement can also be done in
appeal, even if the final decision is rendered many years later.
8

9

The Biryukov v Russia decision of 2008 is consistent with previous
jurisprudence. In this case only the operative part had been pronounced in
public, the full-text was only available to the parties and not to the public,
and there was no preceding decision in which the full text was publicly
pronounced: the European Court of Human Rights concluded to a violation
of art 6. The situation in Welke v Poland was comparable, but was sanctioned
by the Court because of a legislative provision allowing both hearings and
pronouncement of the court’s reasoning to be held in camera because
otherwise state secrets would be revealed. Keeping a judgment completely
from the public for reasons of state security—even if it concerns terrorist
activities—is not allowed.
10

11

12

It should be stressed that while the European Court of Human Rights
considers public access to a register to be a sufficient means for public
pronouncement, it does not oblige to keep such a register, leaving aside the
question whether such a register should be accessible via the internet. Also,
the Court does not make an explicit distinction between the judgment in the
act of pronouncement (establishing a valid judicial decision) and the
judgment as a document that can be consulted after it has been pronounced.
Not only do they differ in an ontological sense, one can easily argue that the
(long-term) availability of decisions in an internet repository enhances the
possibilities for public scrutiny.

8

(European Court of Human Rights, ECLI:CE:ECHR:2001:0710JUD002892395, 10 July
2001).

9

Cf
Shmushkovych
v
Ukraine
(European
Court
of
ECLI:CE:ECHR:2013:1114JUD000327610, 14 November 2013).

10

(European Court of Human Rights, ECLI:CE:ECHR:2008:0117JUD001481002, 17
January 2008) confirmed in Malmberg v Russia (European Court of Human Rights,
ECLI:CE:ECHR:2015:0115JUD002304505, 15 January 2015).

11

(European Court of Human Rights, ECLI:CE:ECHR:2011:0301JUD001592405, 1 March
2011).

12

Raza
v
Bulgaria
(European
Court
of
Human
Rights,
ECLI:CE:ECHR:2010:0211JUD003146508, 11 February 2010) [53]; Fazliyski v Bulgaria
(European Court of Human Rights, ECLI:CE:ECHR:2013:0416JUD004090805, 16 April
2013) [69]; Nikolova v Bulgaria (European Court of Human Rights,
ECLI:CE:ECHR:2013:1217JUD002068804, 17 December 2013) [85].

Human

Rights,

30

Journal of Law, Information and Science

Vol 24(2) 2016

An obligation to publish (all) rendered judgments on a website cannot be
read in art 6 itself, nor in the case law of the Court. Since such an obligation
would entail major legal, technical, societal and financial consequences it
would arguably require an explicit decision by the state parties to the
European convention. On the other hand, the question whether publication
of a judgment on an internet site is to be regarded as public pronouncement
is not answered explicitly yet, but in Nikolova v Bulgaria, the Court seems to
hold that view by listing this possibility on an equal level as other recognised
ways of public pronouncement:
13

[T]he judgments given were not delivered in public and were not
available at the registry of the court or on its Internet site, nor could
the first applicant herself obtain a copy.
14

Publishing a judgment on the internet as a method of public pronouncement
has been suggested already by the late president of the Dutch Supreme
Court in 1999. Although it might be a very efficient method, data protection
issues—to be discussed below—create an insurmountable barrier to turning
this idea into practice.
15

Although implicitly, the technological ease with which court decisions can be
published on the internet also played a role in Ramsahai v the Netherlands.
The case was about a citizen who died by a police bullet. After investigations
the public prosecutor decided to drop charges against the police officer
concerned. Having an established interest the family of Ramsahai contested
this decision in a special complaints procedure before the Court of Appeal.
The order of this Court (confirming the Prosecutor’s decision) was not
pronounced in public. The European Court of Human Rights considered art
6 of the Convention not to be applicable, because it is confined to civil and

16

13

(European Court of Human Rights, ECLI:CE:ECHR:2013:1217JUD002068804, 17
December 2013).

14

Ibid [84].

15

S Martens, ‘With Rechtspraak.nl the Judiciary is in a Better Position to Comply with
the International Obligation Regarding the Public Pronouncement of Court
Decisions’ (Met Rechtspraak.nl kan de rechterlijke macht beter voldoen aan internationale
verplichting tot openbaarmaking vonissen) (2002) 1 Trema 1, also discussed in W Davids
and W Thomassen, 'Publication of Judgments on the Internet' in S Breitenmoser, B
Ehrenzeller and M Sassoli (eds), Human Rights, Democracy and the Rule of Law, Liber
amicorum Luzius Wildhaber (Baden-Baden, Nomos, 2007) 1185.

16

(European Court of Human Rights, ECLI:CE:ECHR:2005:1110JUD005239199, 10
November
2005)
and
(European
Court
of
Human
Rights,
ECLI:CE:ECHR:2007:0515JUD005239199, 15 May 2007).

Court decisions on the Internet

31

criminal proceedings and does not cover such administrative procedures.
Still though, with a reference to Finucane v United Kingdom, it ruled:
17

Where it is decided that a person vested with public authority at
whose hands a human being has died should not face criminal
proceedings, Article 2 [for example, ‘right to life’] requires the
decision to be open to public scrutiny … .
18

However, on appeal this decision was overthrown by the Grand Chamber:
The applicants were allowed full access to the investigation file and
were enabled to participate effectively in the Court of Appeal’s
hearing; they were provided with a reasoned decision. ... In addition,
given that the applicants were not prevented from making the decision
public themselves, the Court takes the view that the requirement of
publicity was satisfied to an extent sufficient to obviate the danger of
any improper cover-up by the Netherlands authorities.
19

If applied to art 6 as well, this decision would contradict all previous
jurisprudence. It would mean that as long as a party to a case has acquired a
full copy of the judgment, the judge would be dismissed from his obligation
of public pronouncement, since in these modern times (nearly) anybody has
the opportunity to make a judgment ‘public.’ Although decided on a very
specific type of proceeding, the reasoning of the European Court of Human
Rights is startling; to ensure the scrutiny of justice ‘[a]n obligation to make
the decision public cannot be placed on the applicants.’
20

Meanwhile, at the national level many Member States of the Council of
Europe have adopted legislation that obliges (all or a selection of) court
decisions to be published on the internet, although nowhere as a way of
publicly pronouncing judgment. These national practices will be reviewed in
Part 2.3, after having discussed the publication of decisions of the European
Court of Human Rights itself and other judicial decisions of European
relevance.

17

(European Court of Human Rights, ECLI:CE:ECHR:2003:0701JUD002917895, 1 July
2003).

18

Ramsahai
v
the
Netherlands
(European
Court
of
Human
ECLI:CE:ECHR:2005:1110JUD005239199, 10 November 2005) [422].

19

Ramsahai
v
the
Netherlands
(European
Court
of
Human
Rights,
ECLI:CE:ECHR:2007:0515JUD005239199, 15 May 2007) [354] (emphasis added).

20

Jointly partly dissenting opinion of Jočienė and Popović to the Grand Chamber
decision, 9.

Rights,

32

Journal of Law, Information and Science

Vol 24(2) 2016

2.2 Dissemination of European Case Law

As established by art 44 para 3 of the European Convention ‘judgments’ of the
European Court of Human Rights are to be published. The Convention has no
comparable provision for ‘decisions’ (declaring an application inadmissible),
but such an obligation was introduced in 2002 by Resolution of the
Committee of Ministers, in which it was also decided that publication had
to be effectuated via an internet database.
21

Apart from the publication of its case law by the European Court of Human
Rights itself, the 47 member states of the Council of Europe themselves have
to publish the most relevant case law of the Court, alongside the text of the
European Convention; a 2002 recommendation of the Committee of Ministers
asks the member states to:
[E]nsure that judgments and decisions which constitute relevant
case-law developments, or which require special implementation
measures on their part as respondent states, are rapidly and widely
published ... in their entirety or at least in the form of substantial
summaries or excerpts … in the language(s) of the country … in,
where appropriate, the Internet sites.
22

The number of member states complying with this recommendation is rather
limited; the European Court of Human Rights recently took responsibility
itself by actively disseminating (non-authorised) translations of its most
relevant judgments.
Within the European Union, the legal framework regarding the publication
of the decisions of the Court of Justice of the EU (‘CJEU’) is more
complicated and less explicit on publication on the internet. All decisions
23

21

Council of Europe Committee of Ministers, Resolution Res(2002)58 On the publication
and dissemination of the case-law of the European Court of Human Rights.

22

Council of Europe Committee of Ministers, On the publication and dissemination in the
member states of the text of the European Convention on Human Rights and of the case-law
of the European Court of Human Rights (Recommendation R(2002)13, ii).

23

For an overview see Presentation of the European Court Reports (2016) Court of Justice
of the European Union <http://curia.europa.eu/jcms/jcms/P_106308/en/>.

Court decisions on the Internet

33

though are published on the website of the CJEU itself, as well as on the
EUR-Lex website.
24

25

Especially of interest in the EU is the legal framework regarding the crossborder accessibility of EU-related case law. Since EU law is (partially)
applicable within the member states, the national judge has to implement
and interpret it. Already in 1982 the CJEU ruled in Srl CILFIT and Lanificio di
Gavardo SpA v Ministry of Health that national judges, when deciding cases
in which specific questions of European law are raised, have an obligation to
consult not only national and CJEU jurisprudence, but also decisions of other
member states’ courts. This obligation to consult the case law of European
colleagues implies a joint obligation to make national case law accessible, as
was reinforced by a resolution of the European Parliament in 2008. There
have been various initiatives by European judiciary networks and EU
funded programmes to live up to this obligation.
26

27

28

An even more explicit obligation to publish case law of European relevance
via the internet can be found in art 3 para 1 of the Second Protocol to the
2007 Lugano Convention on Jurisdiction and the Recognition and Enforcement of
Judgments in Civil and Commercial Matters and related instruments: member
states are obliged to send to the European Commission all decisions from the
highest courts and relevant decisions from lower courts concerning the
Convention, while the Commission has to publish those decisions in a public
29

24

See Info Curia – Case-law of the Court of Justice: Search Form (2016) Court of Justice of
the European Union <http://curia.europa.eu/juris/recherche.jsf?language=en>.

25

See Access to European Law (2016) EUR-Lex <http://eurlex.europa.eu>.

26

(European Court of Justice, ECLI:EU:C:1982:335, 6 October 1982).

27

European Parliament, The Role of the National Judge in the European Judicial System
(2007/2027(INI), CELEX:52008IP0352, 9 July 2008). In deviation of the Australian
Guide to Legal Citation most EU materials are (additionally) cited with their CELEX
number, since this is currently the most reliable method to find documents on the
EUR-Lex website (above n 21).

28

For an overview: M van Opijnen, ‘Finding Case Law on a European Scale Current
Practice and Future Work’, in E Francesconi, G Sartor and D Tiscornia (eds), Legal
Knowledge and Information Systems JURIX 2008: The Twenty-First Annual Conference
(IOS Press, 2008) 43. The project ‘EU-cases’ can be added, EUCases (2016) EUCases
<http://www.eucases.eu>.

29

[2007] OJ L 339/3 (‘Lugano Convention’).

34

Journal of Law, Information and Science

Vol 24(2) 2016

database. With the Publications Office taking over responsibility from the
Commission, this JURE-database is now integrated in EUR-Lex.
30

A comparable obligation to send judicial decisions of legal interest to the
Commission can be found in art 15 para 2 of Council Regulation (EC) No
1/2003 of 16 December 2002 on the Implementation of the Rules on Competition
Laid Down in Arts 81 and 82 of the Treaty, but this provision is not
accompanied by an obligation for the Commission to disseminate these
decisions.
31

32

2.3 Legal Frameworks at the National Level and the Problem of Selection

Two objectives are served by publishing court decisions on the internet. The
first is to serve the rule of law, encompassing for example, the scrutability of
the judiciary and the equitable application of the law. The second objective is
to inform the public—lawyers and laymen alike—about the continuous
development of the law.
Theoretically the first goal requires all case law to be published—which
might converge with using the internet as a method of public
pronouncement —the second goal is best served by publication of a limited
number of cases, because, in the wording of the still only international legal
instrument on this topic, The Selection, Processing, Presentation, and Archiving
of Court Decisions in Legal Information Retrieval Systems: ‘[S]election should
ensure ... that the accumulation of useless information is avoided.’
33

34

35

This Recommendation tries to give guidance on the question of which court
decisions should be selected. One of the most important distinctions it makes

30

See JURE – Jurisdiction, Recognition and Enforcement of Judgments in Civil and
Commercial Matters (2016) EUR-Lex <http://eur-lex.europa.eu/collection/nlaw/jure.html>.

31

[2003] OJ L 1/1, CELEX: 32003R0001.

32

Although
the
(limited)
database
is
online:
Application of Competition Rules by National Courts – Article 15(2) (2016) European
Commission:
Competition
<http://ec.europa.eu/competition/elojade/antitrust/nationalcourts>.

33

Davids and Thomassen, above n 15.

34

Committee of Ministers of the Council of Europe, ‘The Selection, Processing,
Presentation, and Archiving of Court Decisions in Legal Information Retrieval
Systems’ (Recommendation No R(95)11, 11 September 1995).

35

Ibid introductory text to para II of app II.

Court decisions on the Internet

35

is between ‘negative’ and ‘positive’ selection criteria. Negative criteria are—
at least—to be applied on the highest courts: all of their decisions should be
published, except:
[I]f the grounds on which they are based are stated according to a
standard formula of formula clause … [or] if they concern questions
of evidence which are in agreement with existing case-law.
36

Positive criteria are to be applied to the decisions of lower courts. These
decisions should be published if they possibly are of interest for the
development of law, but at the same time the selection ‘must be generally
representative,’ a criterion quite hard to make operational.
37

Most European countries nowadays apply the negative criterion for their
highest courts. In some countries this is based on explicit legislation. For
example, in Austria both negative and positive criteria have a legal basis: the
Minister of Justice is obliged to develop an internet database in which all
decisions of the Supreme Court have to be published, unless they are
declared ill-founded without extensive motivation. The database should
also contain the for the Supreme Court most important decisions of other
courts, which have to be published: ‘In as far as they are of general interest,
exceeding the individual case.’
38

39

40

Also in France all highest courts’ decisions have to be published, while from
the lower courts those decisions should be published: ‘[w]hich are to be
selected on criteria suitable for the jurisdiction in question.’ In Spain those
41

36

Ibid s 5 of para II of app II.

37

Ibid s 2-2 of para III of app II.

38

Bundesgesetz über den Obersten Gerichtshof 1968 [Supreme Court of Justice Act]
(Austria) 28 August 1968, ss 15, 15a.

39

Bundesgesetz über den Obersten Gerichtshof 1968 [Supreme Court of Justice Act]
(Austria) 28 August 1968, ss 14–1, 15–2.

40

Gerichtsorganisationsgesetz 1896 [Organization of the Courts Act] (Austria) 27
November 1896, s 48a.

41

Décret n° 2002-1064 du 7 août 2002 relatif au service public de la diffusion du droit par
l'internet [Decree 2002-1064 on the Public Service of Dissemination of Law via the
Internet] (France) 7 August 2002.

36

Journal of Law, Information and Science

Vol 24(2) 2016

decisions from lower courts have to be published: ‘[o]f which relevance and
academic interest justify publication.’
42

In some countries it is completely left to the judiciary to make a selection, in
some others more detailed rules are formulated, for example, in Belgium.
But even then these criteria are often formulated in a very general wording,
leaving—too—much room for debate and interpretation, for example:

43

[D]ecisions in which completely new arguments are brought
forward’ … [or] decisions in which a legal term is explained, in other
words, in which a legal rule is being formulated or changed.
44

The absence of more detailed guidelines can easily be used as an excuse for a
very limited selection.
In the Netherlands, a generous selection of decisions from district and
appellate courts has been a policy goal of the judiciary portal website from
its start in 1999. It now contains a balanced mix: 169,000 decisions from the
four highest courts, 72,000 courts of appeal judgments and 139,000 cases
from district courts. In the Netherlands there is no legislation on the internet
publication of judicial decisions, but the selection criteria—fully redrafted in
2012—are quite elaborated. Apart from legal importance, also societal
impact is an important selection criterion and by grouping negative, positive,
objective and subjective criteria as well as by putting emphasis on
45

46

42

‘Cuya trascendencia e interés doctrinal justifique su difusión’ [Of which relevance and
academic interest justify publication]. See Acuerdo de 7 de mayo de 1997, del Pleno del
Consejo General del Poder Judicial, por el que se aprueba el Reglamento 1/1997, del Centro de
Documentación Judicial, art 8.

43

See Selectieregels voor de rechtspraak opgenomen in de externe gegevensbank voor de
rechtspraak [Selection Rules for Case Law in the External Database of the Judiciary]
(Belgium), 5 October 2007, NUMAC 2007009841. See also H van Bossuyt and B de
Groote, ‘Free Access to Law and Judicial Decisions: Solutions and Challenges from a
Belgian Viewpoint’, in G Peruginelli and M Ragona (eds), Law via the Internet. Free
Access, Quality of Information, Effectiveness of Rights (Florence, 30/31-10-2008) (European
Press Academic Publishing, 2008) 371.

44

Selection Rules, above n 43.

45

De Rechtspraak, Zoeken in uitspraken <http://uitspraken.rechtspraak.nl>.

46

Raad voor de rechtspraak/ [Council for the Judiciary], /Besluit selectiecriteria
uitsprakendatabank Rechtspraak.nl 2012 /[Decision on Selection Criteria case law
database
Rechtspraak.nl
2012]
(Netherlands),
2012,
Rechtspraak.nl
<https://www.rechtspraak.nl/Uitspraken-ennieuws/Uitspraken/Paginas/Selectiecriteria.aspx>

Court decisions on the Internet

37

operational wordings, these criteria give clear guidance to judges and offer
transparency to the public. After some introductory provisions art 3 states
the negative selection criterion for all (four) highest courts and specialised
chambers of three other courts. For district and appellate courts arts 4–6
define positive criteria, starting in art 4 with the most objective ones. Some of
those are of a procedural nature— for example, decisions starting or
following a preliminary reference proceeding at the CJEU or decisions within
the framework of the 2007 Lugano Convention —other objective criteria are of
a more substantive nature: for example, criminal cases in which the accused
is sentenced to at least four years’ imprisonment (regardless of the crime) or
cases in which someone is prosecuted for murder, manslaughter or a related
crime (regardless of the decision). Article 5 contains positive criteria of a
more subjective nature: cases that have attracted media attention, decisions
that are published or reviewed in legal periodicals or are of particular
relevance for special interest groups, will affect the interests of natural or
legal persons not involved in the case itself, or can be expected to be
influential on legal doctrine. Finally, art 6 determines that decisions not
specifically meeting one of the positive criteria listed should nevertheless be
published as much as possible, as long as they do not merely consist of
standard formula.
47

There are also a substantial number of European countries though where
negative criteria are also applied to lower courts, because scrutability of the
judiciary seems to have been the most determining motive in drafting
legislation on the publication of judicial decisions. In Bulgaria publication of
all judicial decisions is imposed by art 64 para 1 of the Закон за съдебната
власт [Judiciary System Act], and in the Former Yugoslav Republic of
Macedonia art 10 of the Court Case Management Act stipulates that every
decision has to be published on the internet within two days after it has been
rendered. Obstructing the case law database to function as a collective
memory of legal development, para 4 prescribes that decisions have to be
removed from the repository after five years. In Lithuania art 39 of the
Lithuanian Court Act requires publication of all decisions, with the
48

49

50

51

47

Lugano Convention [2007] OJ L 339/3.

48

Закон за съдебната власт [Judiciary System Act 2007] (Bulgaria).

49

Закон за управување со движењето на предметите во судовите [Court Case
Management Act] (Former Yugoslav Republic of Macedonia).

50

In criminal cases other terms apply.

51

Lietuvos Respublikos teismų įstatymo [The Republic of Lithuania Court Act] (Lithuania).

38

Journal of Law, Information and Science

Vol 24(2) 2016

exception of a few categories. In Estonia the Public Information Act lays
down the basic principle that all decisions have to be published, but the,
hierarchically higher, procedural laws allow exceptions; family cases are
hardly published, but most other civil as well as criminal cases are,
although in sensitive cases it is allowed to confine publication to the
operative part. Also in Russia the full texts of court decisions are to be
published on the internet, with an exception for cases relating to state
security and personal and family cases.
52

53

54

55

56

Publishing all—or most—decisions on the internet might be counterproductive though. It is illusory to propound that the judiciary can be
effectively monitored by just putting hundreds of thousands of anonymised,
untagged and unstructured decisions online. In addition, publishing these
enormous volumes means that the second function of case law publication is
also endangered: the law-shaping voice of the judge in potentially landmark
decisions is muted by a tsunami of legally irrelevant cases.
As a consequence, if it is considered essential for reasons of judicial
transparency to stick to the negative criteria also for the lower courts, some
kind of a tagging mechanism is absolutely required to meet also the second
objective of case law publication.
The European Court of Human Rights has developed a workable solution.
As prescribed by the Committee of Ministers, the Court publishes all of its
decisions and judgments, but uses a classification scheme with four
importance levels. The highest category contains all decisions which have
been published (since 1998) in the Case Reports, the other three are labelled
(from low to high importance): 3, 2, 1. To give an indication of the
57

52

With more detailed provisions in a decision of the Council for the Judiciary
(Nacionalinė teismų administracija). See Teismų tarybos 2005 m. rugsėjo 9 d. nutarimu
Nr. 13P-378” Change to: “/Nutarimas dėl teismų sprendimų, nuosprendžių,
nutarimų ir nutarčių skelbimo internete tvarkos patvirtinimo/ [Ruling on publication
of judgments, sentences and orders on the Internet] (Lithuania)

53

Avaliku teabe seadus [Public Information Act] (Estonia).

54

Due to the discretionary powers given to the judge based on Tsiviilkohtumenetluse
seadustik [Code of Civil Procedure] (Estonia), art 462.

55

Kriminaalmenetluse seadustik [Code of Criminal Procedure] (Estonia) art 408–1.

56

Federal Law No. 262-FZ of 22 December 2008 on the Provision of Access to Information on
the Activities of Courts in the Russian Federation (Russia) art 15.

57

Committee of Ministers of the Council of Europe, ‘The Publication and
Dissemination of the Case-law of the European Court of Human Rights’ (Resolution
Res(2002)58, 18 December 2002).

Court decisions on the Internet

39

distribution of decisions over these categories, from the highest to the lowest
level: 3%, 4%, 11%, 82%.
58

At the European Court of Human Rights the adjudication of cases to these
categories is done manually. With their limited number of cases this is
doable, but for a court system with a yearly production of hundreds of
thousands of decisions it is not. Smart algorithms might come to the rescue.
Citation analysis is the most obvious method, since authority can be
established by counting the number of incoming citations. Notwithstanding
its usability this method fails on decisions recently rendered or if one wants
to discriminate between decisions that have not been cited in other decisions
(as is the case with many decisions from lower courts). Hence additional
variables have to be taken into account; the Model for Automated Rating of
Case law (MARC) uses for example, outgoing citations (to legislation as
well as to case law), selection for publication on a judiciary website or in
commercial periodicals, incoming citations from magazine articles and
treatises, type of court and length of the decision. It turns out that data which
are only available on the date of pronouncement or shortly afterwards offer
reliable predictors for future citations.
59

60

3

Data protection

At the turn of the century there was debate on the balance between the need
for transparency and scrutiny—as expressed in art 6 of the European
Convention—and the protection of privacy as laid down in art 8 of
Convention. Protagonists of full and unselected publication held that personal
data in judgments ‘[c]an be necessary for the exercise of public scrutiny,’
and therefore should not be rendered anonymous. One argument was drawn
from the fact that the European Court of Human Rights itself did not

61

58

As of September 2016. Until June 2012 there were three levels (high, medium and low
importance) with then a distribution of 7%, 11% and 82%. As can be learned from
comparing the two distributions the Case Reports documents have been in category 1
before.

59

See, eg, A Mowbray, P Chung, and G Greenleaf, ‘Free-access Case Law
Enhancements for Australian Law’, in G Peruginelli and M Ragona (eds), Law via the
Internet. Free Access, Quality of Information, Effectiveness of Rights (Florence, 30/31-102008) (European Press Academic Publishing, 2009) 285–98; J H Fowler and S Jeon,
‘The Authority of Supreme Court precedent’ (2008) 30 Social Networks 16.

60

M van Opijnen, ‘A Model for Automated Rating of Case Law’ (Paper presented at
Fourteenth International Conference on Articifial Intelligence and Law, Rome, 13
June 2013) 140.

61

Davids and Thomassen, above n 15, 1194.

40

Journal of Law, Information and Science

Vol 24(2) 2016

anonymise its judgments; at that time rule 47 para 3 of the Rules of Court
read:
Applicants who do not wish their identity to be disclosed to the
public shall so indicate and shall submit a statement of the reasons
justifying such a departure from the normal rule of public access to
information in proceedings before the Court. The President of the
Chamber may authorise anonymity in exceptional and duly justified
cases.
62

However, art 8 won the debate gloriously. Nowadays in (nearly) all
European continental law countries court decisions are anonymised before
publication, and in many countries this anonymisation is regulated in the
same statutory provisions that prescribe the (massive or selected)
publication. Which data exactly have to be anonymised is sometimes
outlined in great detail, more often it is left to the discretion of the courts.
Even if more precise anonymisation instructions are drafted, they are hardly
published, and clear rules on how to object to the way a document is
anonymised are often lacking. In most countries the names of professionals
involved are not rendered anonymous; an exception can be found in Russia,
where names of lawyers and even of the hearing judges have to be
anonymised. A remarkable provision is to be found in Estonia, stating that
in criminal cases the name of the convict is not to be anonymised. To avoid
confusion with namesakes the personal identification number has to be
included.
63

64

65

66

Also the European Court of Human Rights itself followed the evolving state
practice: in 2008 it changed the cited provision of the Rules of Court, of

62

Registry of the European Court of Human Rights, Rules of Court (July 2006)
<http://www.echr.coe.int/Documents/Library_2006_RoC_ENG.pdf>.

63

The very detailed anonymization rules in the Netherlands are an exception. See de
Rechtspraak,
Anonimiseringsrichtlijnen
<http://rechtspraak.nl/Uitspraken-enRegisters/Uitspraken/Anonimiseringsrichtlijnen/Pages/default.aspx>.

64

In Belgium, Wet tot oprichting van het informatiesysteem Phenix [Act on the
Development of the Phenix Information System 2005] (Belgium), art 9 imposes the
drafting of a Royal Decree regarding both anonymisation guidelines and a
complaints procedure, but it has not been drafted yet.

65

Federal Law No. 262-FZ of 22 December 2008 on the Provision of Access to Information on
the Activities of Courts in the Russian Federation (Russia), art 15 para 3.

66

Kriminaalmenetluse seadustik [Code of Criminal Procedure] (Estonia) art 408.

Court decisions on the Internet

41

which the last sentence now reads: ‘The President of the Chamber may
authorise anonymity or grant it of his or her own motion.’
67

Also the CJEU changed its rules of procedure. With regard to the
preliminary rulings proceedings art 104 of the former Rules of Procedure
stated explicitly that the original (hence non-anonymised) version of the
national judgment had to be communicated to the various stakeholders. In
the new Rules of Procedure, in force since 1 January 2013, a specific art (95),
titled ‘Anonymity’ has been introduced. It reads:

68

69

1.
Where anonymity has been granted by the referring court or
tribunal, the Court shall respect that anonymity in the proceedings
pending before it.
2.
At the request of the referring court or tribunal, at the duly
reasoned request of a party to the main proceedings or of its own
motion, the Court may also, if it considers it necessary, render
anonymous one or more persons or entities concerned by the case.
In general, one may conclude that with a growing awareness of the risks of
data mining and the enactment of the EU General Data Protection
Regulation, the debate on easing anonymisation rules has silenced.
70

4

Open Data

At the global level the re-usability of public data has been recognised as a
stimulus for economic growth and governmental transparency. Soft law
instruments are gradually shaping a legal framework, like the G8 Open Data

67

Registry of the European Court of Human Rights, Rules of Court (14 November 2016)
<http://www.echr.coe.int/Documents/Rules_Court_ENG.pdf>.

68

Rules of Procedure of the Court of Justice of the European Communities of 19 June 1991
[1991] OJ L 176/7, CELEX:31991Q0704(02).

69

Rules of Procedure of the Court of Justice [2012] OJ C 337/1, CELEX:32012Q1106(01).

70

Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016
on the protection of natural persons with regard to the processing of personal data and on the
free movement of such data, and repealing Directive 95/46/EC [2016] OJ L 119/1,
CELEX:32016R0679.

42

Journal of Law, Information and Science

Vol 24(2) 2016

Charter and, focussed on legal data, the Guiding principles of the Hague
Conference on Private International Law.
71

72

The more firm legal framework within the EU is governed by Directive
2003/98/EC on the re-use of public sector information (‘PSI directive’) as
amended by Directive 2013/37/EU. For the re-usability of case law
documents this amendment, that had to be transposed into national
legislation by all member states on 18 July 2015, entails two important
changes. First, governmental institutions (judiciaries included) do not have a
choice anymore whether to make public documents available for re-use; they
now have an obligation, though without prejudice to for example, data
protection rules. The delicate balance between re-usability versus data
protection has been discussed by, inter alia, the European Data Protection
Supervisor and the Article 29 Data Protection Working Party, which
formulated ‘purpose limitation’ as a leading principle:
73

74

75

The mere fact that personal data are publicly available for a specific
purpose does not mean that such personal data are open for re-use
for any other purpose.
76

One could argue very well that if court decisions are not rendered
anonymous before publishing them on the web (like in many common law
countries) and additional measures have been taken to protect the interests
77

71

Cabinet Office, UK Government, G8 Open Data Charter and Technical Annex (18 June
2013) <http://www.gov.uk/government/publications/open-data-charter/g8-opendata-charter-and-technical-annex>.

72

Permanent Bureau, Hague Conference on Private International Law, Accessing the
Content of Foreign Law and the Need for the Development of a Global Instrument in this
Area a Possible Way Ahead, Preliminary Document No 11 A (March 2009)
<http://www.hcch.net/upload/wop/genaff_pd11a2009e.pdf>.

73

[2003] OJ L 345/90, CELEX:32003L0098.

74

[2013] OJ L 175/1, CELEX:32013L0037.

75

European Data Protection Supervisor, Opinion on the 'Open-Data Package' of the
European Commission including a Proposal for a Directive amending Directive 2003/98/EC
on Re-use of Public Sector Information (PSI), a Communication on Open Data and
Commission Decision 2011/833/EU on the Reuse of Commission Documents [2012] OJ C
335, CELEX:52012XX1101(04).

76

Article 29 Data Protection Working Party, Open Data and Public Sector Iinformation
('PSI')
Reuse,
Opinion
06/2013
(5
June
2013)
<http://ec.europa.eu/justice/data-protection/article-29/documentation/opinionrecommendation/files/2013/wp207_en.pdf>.

77

See, eg, Apata, R v Secretary of State for the Home Department [2016] EWVA Civ 802.

Court decisions on the Internet

43

of named litigants—for example, by the use of the Robots Exclusion
Protocol —purpose limitation might be a valid reason for not making these
documents available for reuse, notwithstanding the possible breach of data
protection rules by maintaining such databases in the first place. In any case,
purpose limitation should and cannot be used as an argument against
making available for reuse any public case law database that is already fully
anonymised.
78

The second important amendment of the PSI Directive regards the technical
usability of the data. The original text of art 5 para 1 read:
Public sector bodies shall make their documents available in any preexisting format or language, through electronic means where
possible and appropriate.
After the amendment it reads:
Public sector bodies shall make their documents available in any preexisting format or language, and, where possible and appropriate, in
open and machine-readable format together with their metadata.
Both the format and the metadata should, in so far as possible,
comply with formal open standards.
More specific guidelines are published by the European Commission,
recommending, inter alia, the highest possible level of granularity,
publication at a stable location, publishing in formats like XML, JSON and
RDF, using rich metadata (for example, Dublin Core), controlled
vocabularies and APIs as well as proper documentation. Apart from these
general instructions, equally applicable to all types of public data, guidance
for improving reusability of case law documents can be found in the Council
Conclusions on the European Case Law Identifier, which will be discussed in
Part 5 below.

79

There are substantial differences across Europe regarding the extent to which
case law repositories are actually available for reuse. In the Netherlands for
example, all published decisions have been available for unlicensed reuse in
XML-format, via FTP since 2004 and via a RESTful web service since 2013,
but in many other countries re-users are confined to screen scraping and in

78

According to this protocol a small text file (‘robots.txt’) is placed on a webserver,
specifying which web crawlers are allowed to index which data. Although the
protocol is generally observed, it can be ignored easily.

79

European Commission, Guidelines on recommended standard licences, datasets and
charging for the reuse of documents [2014] OJ C 240/1, CELEX:52014XC0724(01).

44

Journal of Law, Information and Science

Vol 24(2) 2016

some other even legal restrictions apply – being at odds with the PSI
Directive.
An interesting court case on this topic can be found in Germany where
LexXpress GmbH, a legal publisher, wanted to have equal access to the XML
formatted judgments of the Constitutional Court (Bundesverfassungsgericht),
as well as to the accompanying metadata, which were till then exclusively
sent to Juris GmbH, a company in which the state holds a majority stake.
These metadata—like head notes, summaries and references—are written by
the staff of the Constitutional Court. While LexXpress held that, based on the
(by then unamended) PSI directive and the way it was transposed into
national German law, it was equally entitled to the information as Juris
GmbH; the defendant (the State) held that the metadata were copyright
protected. In its decision of 3 November 2011 the Administrative court in
Karlsruhe ruled for the defendant, but that decision was radically
overthrown by the Appellate Administrative Court (Verwaltungsgerichtshof,
‘VGH’) Baden-Württemberg. The VGH ruled that not only court decisions
themselves are free from copyright, but also the metadata concerned, since
they are produced by court staff. Hence, the exception for copyrighted
material, made in the national legislation implementing the PSI directive was
not applicable. Also, the argument of the Constitutional Court that the State
had a specific interest in an exclusive contract with Juris GmbH was waved
aside by the VGH. The final decision had to come from the Federal High
Administrative Court (Bundesverwaltungsgericht, ‘BVerwG’). Because the
BVerwG also has an exclusive contract with Juris GmbH, a motion to recuse
was filed against all judges of the BVerwG, which was dismissed (an
unappealable decision of the BVerwG itself). Disappointingly from a legal
perspective a decision on the merits will not be rendered, since the case has
been withdrawn in 2015 after reaching a friendly settlement. The launch of
the website ‘Court decisions on the internet’ (Rechtsprechung im Internet) on
which the decisions of the Constitutional Court and the other Supreme
80

81

82

83

80

Verwaltungsgericht Karlsruhe, 3 K 2289/09, 2 March 2011.

81

Verwaltungsgerichtshof Baden-Württemberg, 10 S 281/12, 7 May 2013.

82

(Bundesverwaltungsgericht, ECLI:DE:BVerwG:2014:290114B7C13.13.0, 29 January
2014).

83

Bundesministerium der Justiz und für Verbraucherschutz, Rechtsprechung im Internet
<http://www.rechtsprechung-im-internet.de>.

Court decisions on the Internet

45

Courts can be searched and downloaded seems to have played a major role
in reaching this settlement.
84

5

Identifiers and metadata

One of the reasons for the only moderate success of some of the initiatives
for improving access to EU related national case law was the lack of proper
identifiers and common metadata. To provide for standardisation the EU
Council of Ministers in December 2010 established the Council conclusions
on ECLI. The ECLI framework consists of five core components.
85

86

87

The first component is the identifier itself, which can be assigned to every
court decision rendered by a European or Member State’s court. The fixed
format of five elements—ECLI:{country code}:{court code}:{year of
decision}:{unique code}—makes this identifier easily recognised and read by
lawyers and computers alike, while at the same leaving room for flexibility
in national implementation.
Currently ECLI has been introduced by – and made available in public case
law repositories of—the CJEU, the European Court of Human Rights, the
Boards of Appeal of the European Patent Office (EPO), all courts in the
Netherlands, Spain and Austria, and a growing number of courts in
Germany, France, Slovenia, the Czech Republic, Slovakia, Finland and
Greece. In Romania, Italy, Cyprus, Belgium, Latvia, Estonia, Bulgaria,
Denmark and Croatia the implementation of ECLI is work in progress.
88

89

84

T Podolski, ‘Juris: Lexxpress gegen BVerfG: Sieg über mächtige Gegner’, Legal Tribune
(online), 10 February 2015 <http://www.lto.de/persistent/a_id/16771/>.

85

See above n 28.

86

Council of the European Union, Council Conclusions Inviting the Introduction of the
European Case Law Identifier (ECLI) and a Minimum Set of Uniform Metadata for Case Law
[2011] OJ C 127/1, CELEX:52011XG0429(01).

87

See also M van Opijnen, ‘European Case Law Identifier: indispensable asset for legal
information retrieval’ in M A Biasiotti and S Faro (eds), From Information to Knowledge.
Online Access to Legal Information: Methodologies, Trends and Perspectives (IOS Press,
2011) 91.

88

The ECLI of the CJEU is available at the Court’s website as well as on EUR-Lex.

89

M van Opijnen and A Ivantchev, ‘Implementation of ECLI State of Play’ in A Rotolo
(ed), Legal Knowledge and Information Systems JURIX 2015: The Twenty-Eighth Annual
Conference, Braga, 10/11-12-2015 (IOS Press, 2015) 165.

46

Journal of Law, Information and Science

Vol 24(2) 2016

The second component of the ECLI framework is a unified set of mandatory
and optional metadata, based on the Dublin Core metadata standard.
Thirdly, Member State implementation should be guided by a national ECLI
co-ordinator, which is, inter alia, responsible for documenting the national
technical specifications on the ECLI pages of the European e-justice portal,
the fourth component of the ECLI framework.
90

Finally, the Council’s conclusions assign the Commission with the task to
develop the ECLI Search Engine on the European e-justice portal (‘ESEEEJP’), giving access to possibly all European and national court decisions
having an ECLI. One of the advantages would be that one can discover easily
whether for example, summaries in other languages are available for a
specific judgment. The ESE-EEJP went live on 4 May 2016, and contains
nearly four million decisions from the CJEU, the EPO, Spain, the
Netherlands, Czech Republic, Slovenia, Finland, Germany, France as well as
from the JuriFast database of the Association of Councils of State and
Supreme Administrative Jurisdictions of the European Union.
91

92

93

To even further improve case law accessibility at the European level, in
October 2015 ‘Building on ECLI’ has started, a project with sixteen partners
from ten EU Member States. A first goal is to (further) implement ECLI in
Belgium, Croatia, Greece, Estonia, Germany, Czech Republic, Italy and the
Netherlands, and to connect their case law repositories to the ESE-EEJP. A
second goal is to improve the accessibility of published case law, inter alia
within the ESE-EEJP, by creating computer readable—and hence
searchable—legal references within judicial decisions. Thirdly, a 2.0 version
of the ECLI standard is to be developed, which should be aligned with other
evolving semantic web standards. A fourth objective is to have EU wide
policy guidelines on the publication of case law, specifically addressing the
issues of selection criteria, data protection and Open Data. Last but not least,
94

90

European Justice, European Case Law Identifier (4 May 2016) <https://ejustice.europa.eu/content_european_case_law_identifier_ecli-175-en.do>.

91

European
Justice,
ECLI
Search
Engine
(4
May
justice.europa.eu/content_ecli_search_engine-430-en.do>.

92

Association of the Councils of the State and Supreme Administrative Jurisdictions of
the
European
Union
i.n.pa.,
Home
Page
<http://www.acaeurope.eu/index.php/en/jurifast-en>.

93

J C van Haersolte, ‘A Wheel Within a Wheel: the Association of the Councils of State
and Supreme Administrative Jurisdictions of the European Union’ (2010) 3 Review of
European Administrative Law 1, 131.

94

Building on ECLI, Building on ECLI <http://www.bo-ecli.eu>, the project is cofunded by the Justice Programme of the European Union (2014–2020).

2016)

<https://e-

Court decisions on the Internet

47

the fifth objective is to promote the use of ECLI for interoperability in (legal)
IT applications and for case law citations in all legal and academic writings.

6

Conclusions

It is a positive development that the publication of court decisions is viewed
as an essential ingredient of modern judicial transparency. Although some
European countries are more advanced—quantitatively or qualitatively—
than others, the trend is irreversible. The gradual development of a legal
framework—as outlined in this paper—is encouraging for the (wide variety
of) end-users, although more common ground would be welcome.
Because transparency or re-usability should never take precedence over the
primary function of the judicial decision—often dealing with the most
intimate and embarrassing details of the private lives of citizens —data
protection should be a cornerstone of any legal framework. For that reason
publication on the internet can and should never be accepted as a means of
official public pronouncement; if anonymised, the scrutiny of justice would
be at stake, if not anonymised the interests of litigants would be violated.
And although the jurisprudence of the European Court of Human Rights
does not set alarm bells ringing yet, some of its recent judgments are
worrisome: will the Court really accept internet publication as a means of
public pronouncement, and if yes, how does that relate to art 8 of the
Convention? Will the Court persevere in its opinion that the responsibility for
openness of a governmental institution can be waived by the argument that
relevant information can nowadays be made public by anybody?
95

The amended PSI Directive is very much welcomed by re-users of case law
databases. Many improvements can be expected in the coming years,
although two issues require attention. First, also with respect to reuse data
protection should be the guiding principle, especially for those countries not
yet having a very strict regime regarding anonymisation of court decisions.
To have the best of both worlds, more drastic anonymisation should be
preferred over limitations of reuse.
A second issue regards the problem of information overload. Data scientists
and academics with a hard-core vision on judicial transparency make us
believe there is no problem with just making everything available. By doing

95

See, eg, P A Winn, ‘Online Court Records: Balancing Judicial Accountability and
Privacy in an Age of Electronic Information’ (2004) 99 Washington Law Review 307; I
Currie, ‘Problems of Privacy in Online Court Materials’ in G Peruginelli and M
Ragona (eds), Law via the Internet. Free Access, Quality of Information, Effectiveness of
Rights, Florence, 30/31-10-2008 (European Press Academic Publishing, 2008) 73.

48

Journal of Law, Information and Science

Vol 24(2) 2016

so—as experienced in many countries having such rigorous policies—the
interests of average users—legal professionals looking for relevant cases and
interested citizens alike—are disregarded completely. Without any policies
assuring that the qualitative accessibility of the information holds pace with
the quantitative growth of the repositories the net balance will be negative.

Applying the Akoma Ntoso XML Schema
to Japanese Legislation
KAWACHI GEN , NAGAI AKIRA , NAKAMURA MAKOTO ,
OGAWA YASUHIRO , OHNO TOMOHIRO , TOYAMA KATSUHIKO
*

§

**

††

Abstract
This paper presents a case study for avoiding ambiguous annotation when applying
Akoma Ntoso, an XML schema for parliamentary documents, to Japanese statutes,
which follow strict drafting rules. The Japanese statutory schema is designed to reflect
the rules, while the Akoma Ntoso schema has underlying structural ambiguity due to
its flexibility. We propose a method to convert from the former to the latter schema and
provide a subset of Akoma Ntoso that retains a strict annotation approach.

1

Introduction

Japanese statutes have maintained a common descriptive style following a
Japanese legislation drafting manual for more than one hundred years. This
manual includes strict rules about the hierarchical structure of statutory
documents, the typefaces of sequential numbers, grammatical expressions
peculiar to statutory documents, the notation of characters (letters), and so
on. Noteworthy is that these rules have been maintained for all statutes,
which facilitates their understanding.
Some of the statutes are provided in extensive markup language (XML)
following a schema designed exclusively for the structure of Japanese

*

KAWACHI Gen, final year Master’s student at Graduate School of Information
Science, Nagoya University.

NAGAI Akira, final year student at School of Engineering, Nagoya University.

NAKAMURA Makoto, Doctor of Information Science, Designated Assistant
Professor at Japan Legal Information Institute, Graduate School of Law, Nagoya
University.

§

OGAWA Yasuhiro, Doctor of Engineering, Associate Professor at Information
Technology Center and Graduate School of Information Science, Nagoya University.

**

OHNO Tomohiro, Doctor of Information Science, Assistant Professor at Information
Technology Center and Graduate School of Information Science, Nagoya University.

††

TOYAMA Katsuhiko, Doctor of Engineering, Professor at Information Technology
Center and Graduate School of Information Science, Nagoya University.

50

Journal of Law, Information and Science

Vol 24(2) 2016

statutes. This schema is so strict that it basically does not allow a document
to have ambiguous structures. In other words, this schema can be regarded
as a digitised version of the Japanese legislation drafting manual. In this
paper, we consider how to disseminate Japanese legislation while
maintaining this strict annotation approach. The key issue is how to adapt an
international standard for XML schemata to Japanese legislation.
1

Akoma Ntoso (Architecture for Knowledge-Oriented Management of Any
Normative Texts using Open Standards and Ontologies), is a jurisdictionindependent XML standard that can be used for interchange between public
organisations or business enterprises and as a platform for generic legal
software. Although Akoma Ntoso was originally created to share legislative
documents among African countries, it is now widely used as the OASIS
(Organization for the Advancement of Structured Information Standards)
LegalDocumentML which was developed based on Akoma Ntoso’s
specifications. In fact, a number of government bodies in several countries,
such as EUCases, LexML Brazil, and the Serbian judiciary, apply this
schema to legal documents. Flexibility is a desirable trait of Akoma Ntoso,
allowing many jurisdictions to employ this schema. However, this flexibility
can cause ambiguous annotation. For example, there are no restrictions
among the elements for hierarchical structures because the Akoma Ntoso
2,

3

4

5

6

7

1

Katsuhiko Toyama, Daichi Saito, Yasuhiro Sekine, Yasuhiro Ogawa, Tokuyasu
Kakuta, Tariho Kimura, and Yoshiharu Matsuura, ‘Design and Development of
Japanese Law Translation Database System’ (Paper presented at the Law via the
Internet Conference 2011, University of Hong Kong, 9 June 2011).

2

Monica Palmirani and Fabio Vitali, ‘Akoma-Ntoso for Legal Documents’ in Giovanni
Sartor, Monica Palmirani, Enrico Francesconi, and Maria Angela Biasiotti (eds),
Legislative XML for the Semantic Web (Springer Netherlands, 2011) 75; Akoma
Ntoso, Akoma Ntoso <http://www.akomantoso.org/>.

3

Alexander Boer, Radboud Winkels, and Fabio Vitali, ‘Proposed XML Standard for
Law: MetaLex and LKIF’ in Arno R Lodder and Laurens Mommers (eds), Legal
Knowledge and Information Systems: JURIX 2007: The Twentieth Annual Conference (IOS
Press, 2007) 19.

4

OASIS, OASIS LegalDocumentML (LegalDocML) TC <https://www.oasisopen.org/committees/tc_home.php?wg_abbrev=legaldocml>.

5

6
7

Guido Boella, Loredana Cupi, Luigi di Caro, Monica Palmirani, Livio Robaldo, and
Andrea Violato, D2. 2 Legal XML-schema (XSD) (Integration, 2014).
Palmirani and Vitali, above n 2, 75.
Marko Marković, Stevan Gostojić, Zora Konjović, and Mart Laanpere, ‘Machinereadable Identification and Representation of Judgments in Serbian Judiciary’ (2014)
44(1) Novi Sad Journal of Mathematics 165.

51

Applying the Akoma Ntoso XML Schema to Japanese Legislation

schema validates many kinds of structures by itself to support various legal
document formats around the world.
Our purpose in this paper is to create a seamless connection from Japanese
statutes following strict drafting rules to Akoma Ntoso as an international
standard for jurisdiction-independent XML schema. We present a case study
for avoiding ambiguous annotation when applying the Akoma Ntoso XML
schema to Japanese legislation and propose a method to convert statutory
documents structured by our XML schema to that of Akoma Ntoso, enabling
us to show other jurisdictions how to overcome this problem.
Our paper is organised as follows: in Part 2, we introduce the characteristics
of Japanese statutes and a sophisticated proofreading process for the
Japanese legislative system. We then provide examples of ambiguity in
Akoma Ntoso annotation in Part 3. In Part 4, we explain the XML schema for
Japanese statutes and our approach to convert the schema’s annotation to
that of Akoma Ntoso. We summarise our findings in Part 5.

2

Japanese Legislation

In this part, we briefly explain the process and characteristics of Japanese
legislation and its strict rules for the structure, format, and expression of all
Japanese statutes.
2.1 Basic Organisation of Japanese Laws

Laws are generally divided into written and unwritten categories. Although
unwritten laws include local customs and judicial precedents, we address
only written laws in this paper. Written laws are also called statutes, which
are further divided into acts and bylaws. In Japan, while acts are enacted by
the Diet (Parliament), bylaws consist of orders enacted by the Cabinet and
ordinances and regulations enacted by various governmental organisations,
such as ministries and administrative commissions.
In this paper, we focus on statutory texts. A statute consists of a number of
articles, each of which may be further subdivided into paragraphs or items.
Articles are integrated into a chapter and chapters are integrated into a part
based on their content. Articles, paragraphs, and items have sequential
numbers with different typefaces. A provision denotes an individual article
or a paragraph.
2.2 Characteristics of Japanese Legislation

The logic of the legislative system in Japan maintains the notation of the
Japanese statute expressions. Figure 1 outlines the legislative process in the
national government from drafting to promulgation. Although both Diet
members and the Cabinet can submit a bill to the Diet, most are introduced

52

Journal of Law, Information and Science

Vol 24(2) 2016

by the Cabinet. For the 10,164 acts enacted from 1947 to 2014, about 80%
were presented by the Cabinet.
Before the Cabinet submits a bill to the Diet, it sends a draft to the Cabinet
Legislation Bureau, which scrutinises it for consistency with other statutes,
expressions, formats, etc. The Bureau's mission is described on its website as
follows:
8

During the examination by the Bureau, the bill is examined from all
angles, legally and technically. The points examined include the
following:

The relationship between the proposed bill on one hand and
the Constitution and other existing laws on the other, as well
as the legal appropriateness of the contents of the bill;

Whether or not the intentions of the proposed bill are
accurately expressed in the text;

Whether or not the structure of the bill (for example, the
order of articles) is appropriate;

Whether the usage of letters or words is correct.

Even the usage of such punctuation as commas and periods is maintained.
When a Diet member submits a bill, it is reviewed by the Legislation Bureau
of the House of Representatives or Councillors.
The Cabinet Legislation Bureau has a manual for drafting Japanese
legislation. Although the original version has never been published, many
derivations are available for drafters of local governments and other
organisations.

8

Cabinet Legislation Bureau, Until the Law Can Be Done
<http://www.clb.go.jp/english/process.html>.

53

Applying the Akoma Ntoso XML Schema to Japanese Legislation

The Cabinet

Diet members

Drafting legislative bills

Planning ideas of bills

Cabinet Legislation Bureau

Legislative Bureau of the
House of Representatives/
House of Councillors

Revision
Examination of legislative bills w.r.t.
• Legal appropriateness
• Usage of letters or words

Drafting legislative bills
Examination of legislative bills

Submission

The Diet
The House of Representatives
Approval

The House of Councillors
Approval

Enactment
Signature
by the Cabinet

Promulgation

In the name of the Emperor

Figure 1: Outline of the Japanese legislative process.

9

Not every country’s legislative system resembles Japan’s. In the United
Kingdom, the verification of legislation is less strict, since in most cases the
bill is drafted outside of the ministry. In the United States, no organisation or
system verifies legislation. In Asian countries, other than Japan and Korea,
often each ministry independently prepares a draft of a bill without
coordinating with other ministries. As a result, the notation of bills differs
among ministries. In some countries, bills are often modified during
deliberation in the national assembly, which may result in inconsistency of
descriptive styles, while bills generally pass the Diet in Japan as drafted.
10

9

In some cases, bills are sent up for consideration to the House of Councillors first,
rather than the House of Representatives.

10

This paragraph is based on a discussion with Prof. Yoshiharu Matsuura at Graduate
School of Law, Nagoya University. Ministry of Justice, Japan, Japanese Law
Translation Database System (2016) <http://www.japaneselawtranslation.go.jp/>.

54

Journal of Law, Information and Science

Vol 24(2) 2016

2.3 Dissemination of Japanese Legal Information

The Japanese government publishes its statutes online in the Japanese Law
Translation Database System (JLT). JLT has provided an English version of
important statutes since 2009. These statutes are digitised in XML to simplify
their management and to convert their format. Since the main purpose of JLT
is to offer English translation of Japanese statutes, we can only access
statutory documents that have been translated into English. Since JLT's
release, the small number of statutes that have actually been translated has
been problematic. As of July 2015, only 524 statutes were included in JLT,
which is merely 6.5% of all statutes currently in effect.
11

12

The Japanese government has been preparing a more sophisticated scheme,
called the e-Legislative Activity and Work Support System (e-LAWS), to
support drafting bills in government and to disseminate all Japanese statutes
currently in effect. It will follow the XML schema based on Document Type
Definition (DTD) in JLT. We refer to this new schema as the Japanese
Statutory Schema (JSS). It is designed exclusively for the structure of
Japanese statutes and does not allow a document to have ambiguous
structures. Thus far, Akoma Ntoso has never been considered in the
development of JSS. This study is the first attempt to apply Akoma Ntoso to
Japanese statutes.

3

Ambiguity in Akoma Ntoso Annotation

In this part, we highlight problematic annotations for two types of structures
that are often seen in legal documents—hierarchical structures and
provisions in amendment acts.

11

Toyama et al, above n 1. Note that, despite an official website by the Ministry of
Justice, all of the translations contained in JLT are unofficial. Since Japanese laws and
regulations are written in Japanese, only their original Japanese texts have legal
effect; translations are to be used solely as reference materials to aid in the
understanding of Japanese laws and regulations. The government of Japan is not
responsible for the accuracy, reliability, or currency of the legislative material
provided in the website, or for any consequence resulting from use of the
information on the website.

12

Yasuhiro Sekine, Yasuhiro Ogawa, Katsuhiko Toyama, and Yoshiharu Matsuura,
‘The Development of Translation Memory Database System for Law Translation’
(Paper presented at the Law via the Internet Conference 2012, Cornell University, 9
October 2012).

55

Applying the Akoma Ntoso XML Schema to Japanese Legislation

3.1 Ambiguity in Annotating Hierarchical Structures

The Akoma Ntoso XML Schema validates many kinds of elements and
various structures to deal with the legal documents of different nations that
follow different formats. In other words, this flexibility allows the schema to
validate unintended structuralisation to the same document. We regard
flexibility in annotation as ambiguity in formal language theory .
13

14

For example, Akoma Ntoso has difficulty annotating hierarchical structures.
We use the article in Figure 2 to discuss how to annotate an article of a
statute. This article about term definitions has two paragraphs, one of which
consists of two items. We need to determine an appropriate hierarchical
element for the article's children. Akoma Ntoso defines 27 elements, such as
‘article’ and ‘paragraph’, to describe the hierarchical structure of legal
documents. The problem is that there are no restrictions on the hierarchical
relationships among the elements. Therefore, this schema validates both
structures
shown
in
Figure 3. The children of ‘article’ can be ‘paragraph,’ as shown in Candidate
1, while ‘rule’ can be used for the same role, as shown in Candidate 2.
15

16

The annotation for the content in each article is also ambiguous. There are
multiple possible annotations to the ordered items with parenthesised
Roman numerals, (i) and (ii) in Figure 3. In Candidate 1, these items are
regarded as a part of the hierarchical structure and are expressed with the
tag ‘point.’ On the other hand, in Candidate 2, the tag ‘tblock’ is used. The
schema validates both candidates, which we consider ambiguous.

13

Fabio Vitali, Monica Palmirani, Roger Sperberg, and Véronique Parisse, Akoma
Ntoso Version 1.0. Part 2: Specifications (14 January 2015) <http://docs.oasisopen.org/legaldocml/akn-core/v1.0/akn-core-v1.0-part2-specs.html>.

14

John E Hopcroft and Jeffrey D Ullman, Introduction to Automata Theory, Languages, and
Computation (Addison-Wesley, 1979).

15

Japanese statutes are originally written in Japanese. The article in Figure 2 is the
English translation.

16

In Figure 3, attributes are omitted for simplification.

56

Journal of Law, Information and Science

Vol 24(2) 2016

(Definitions, etc.)
Article 2 (1) The term “Bank” as used in this Act means a person who
engages in Banking under the license from the Prime Minister
prescribed in Article 4, paragraph (1).
(2) The term “Banking” as used in this Act means commercial pursuits
carried out through any of the following acts:
(i)

Acceptance of deposits or Installment Savings, in addition to
loans of funds, or the discounting of bills and notes; or

(ii) Carrying out exchange transactions.

Figure 2: Example of a Japanese statute (Article 2 of Banking Act (Act No. 59 of
1981))
Candidate 1
<article>
<heading>(Definitions, etc.)</heading>
<subheading>Article 2</subheading>
<paragraph>
<num>(1)</num>
<content>
<p>The term "Bank" as used in this Act means a person who engages
in Banking under the license from the Prime Minister prescribed in Article 4,
paragraph (1).</p>
</content>
</paragraph>
<paragraph>
<num>(2)</num>
<list>
<intro>
<p> The term "Banking" as used in this Act means commercial
pursuits carried out through any of the following acts:</p>
</intro>
<point>
<num>(i)</num>
<content>
<p>Both acceptance of deposits or Installment Savings, and loans
of funds or discounting of bills and notes; or </p>
</content>
</point>
<point>
<num>(ii)</num>
<content>
<p> Carrying out exchange transactions.</p>
</content>

57

Applying the Akoma Ntoso XML Schema to Japanese Legislation

</point>
</list>
</paragraph>
</article>
Candidate 2
<article>
<heading>(Definitions, etc.)</heading>
<subheading>Article 2</subheading>
<rule>
<num>(1)</num>
<content>
<p>The term "Bank" as used in this Act means a person who
engages in Banking under the license from the Prime Minister prescribed in
Article 4, paragraph (1).</p>
</content>
</rule>
<rule>
<num>(2)</num>
<content>
<p>The term "Banking" as used in this Act means commercial
pursuits carried out through any of the following acts:</p>
<tblock>
<num>(i)</num>
<p>Acceptance of deposits or Installment Savings, in addition to
loans of funds, or the discounting of bills and notes; or</p>
</tblock>
<tblock>
<num>(ii)</num>
<p>Carrying out exchange transactions.</p>
</tblock>
</content>
</rule>
</article>
Figure 3: Possible annotation of Figure 2 validated by Akoma Ntoso
3.2 Inconsistent Annotation of Amendment Acts

The issue of inconsistency in Akoma Ntoso’s annotation is also seen in the
provisions in amendment acts. In this part, we demonstrate that there are at
least two ways to use the element ‘mod.’ Figures 4 and 5 show a comparison
of the element between amendment acts that were enacted in (A) Kenya and
(B) the United Kingdom (UK), respectively. Table 1 details these acts. Both
provisions include the new provisions to be inserted into the target act. The
provision in Figure 4 explains how s 50 of the existing act is to be amended,

58

Journal of Law, Information and Science

Vol 24(2) 2016

while the provision in Figure 5 explains where to insert sub-s (8A) into s (2)
of the existing act.
Note that statutes are written in two types of languages: an object language
for new enactments and metalanguage, which re-writes the description in
object language, for amendments, rearrangements, and to repeal acts. In
Figure 4, the first three lines are written in the metalanguage, while the rest
is in the object language. In Figure 5, the first two lines are written in the
metalanguage, while the rest is in the object language.
7. Section 50 of the principal act is amended -

(a) by deleting subsection (2) and inserting the following (2) Where anything other than a trophy is seized and detained under
section 49 of this Act it shall be forfeited to the government after a
period of three months unless within that period some person is
charged with a forfeiture offence under the Act and it is alleged
that -

(a) the offence was committed in relation to or in connection with
that thing; or

(b) the thing was used in, or for the purpose of, the commission
of such an offence.
Figure 4: Amending provision in an act enacted in Kenya
In section 2 of the Regulation of Investigatory Powers Act 2000 (meaning of
“interception” etc), after subsection (8) insert —
“(8A) For the purposes of the definition of “telecommunications
service” in subsection (1), the cases in which a service is to be
taken to consist in the provision of access to, and of facilities for
making use of, a telecommunication system include any case
where a service consists in or includes facilitating the creation,
management or storage of communications transmitted, or that
may be transmitted, by means of such a system.”

Figure 5: Amending provision in an act enacted in the UK
Table 1: Information on the acts shown in Figures 4 and 5
Label

Corresponding

Nation

Name

Figure
(A)

Figure 4

Kenya

The Wildlife

Name of the act to

position

be amended

s7

The Wildlife

(Conservation and

(Conservation and

Management)

Management) Act

(Amendment) Act,

(Cap. 376)

1989 (No. 16 of 1989)

Excerpted

59

Applying the Akoma Ntoso XML Schema to Japanese Legislation

(B)

Figure 5

the UK

Data Retention and

ss 2,

Regulation of

Investigatory Powers

s4

Investigatory

Act 2014 (2014 Chapter

Powers Act 2000

27)

(2000 Chapter 23)

Figures 6 and 7 show examples of the Akoma Ntoso annotation of the
provisions in (A) and (B), respectively. Both are downloadable.
17,

Here, we focus on the element ‘mod,’ which is an XML tag defined for
modification. This element is used for separating the metalanguage and the
object language. However, ‘mod’ is used in a different way in each of these
two examples. For (A), the instruction ‘... and inserting the following—...’ is
included within ‘mod.’ On the other hand, the instruction ‘... after subsection
(8) insert’ in (B) is not included within ‘mod,’ even though both elements
play the same role. In other words, the region of the object language includes
a part of the metalanguage in (A).
The gap between the above two examples implies an underlying problem of
the Akoma Ntoso XML schema, under which annotations can differ
depending on the annotator because details in the usage are not published.
Furthermore, such an inconsistency may occur not only among different
articles in a single act, but also different acts in a single nation and different
nations in a single XML standard, which violates the intended
interchangeability of Akoma Ntoso. Considering that JSS is designed
exclusively for the structure of Japanese statutes and does not allow a
document to have ambiguous structures, we need to carefully determine
how to implement Akoma Ntoso to Japanese legislation.
<section>
<num>7.</num>
<content>
<blockList>
<listIntroduction>Section 50 of the principal act is amended </listIntroduction>
<item>
<num>(a)</num>
<p>by deleting
<mod>
<ref
href="/ke/act/1980-01-01/1/main#sec50-sub2">subsection

17

The official web site of Akoma Ntoso for (A): Akoma Ntoso Example Documents
<http://examples.akomantoso.org/>. The repository of OASIS for (B):
<http://docs.oasis-open.org/legaldocml/akn-core/v1.0/csd02/part2specs/examples/uk_pga-2014-27-enacted-data.xml>.

60

Journal of Law, Information and Science

Vol 24(2) 2016

(2)</ref> and inserting the following –
<quotedStructure>
<subsection>
<num>(2)</num>
<content>
<blockList>
<listIntroduction>Where anything other than a trophy is seized
and detained under section 49 of this Act it shall be forfeited to the
government after a period of three months unless within that period some
person is charged with a forfeiture offence under the Act and it is alleged
that -</listIntroduction>
<item>
<num>(a)</num>
<p> the offence was committed in relation to or in connection
with that thing; or</p>
</item>
<item>
<num>(b)</num>
<p>the thing was used in, or for the purpose of, the commission
of such an offence.</p>
</item>
</blockList>
</content>
</subsection>
</quotedStructure>
</mod>
</p>
</item>
<item>
<num>(b)</num>
<p>by deleting
<mod>
<ref href="/ke/act/1980-01-01/1/main#sec50-sub3-itmc">paragraph
(c) of subsection (3)</ref>
</mod>.
</p>
</item>
</blockList>
</content>
</section>
Figure 6: Annotation for the provision in Figure 4
<section>
<num>5</num>
<heading>Meaning of “telecommunications service”</heading>
<content>

61

Applying the Akoma Ntoso XML Schema to Japanese Legislation

<p>In
<ref
href="/akn/uk/act/2000-0728/chapter23/main#sec_2__subsec_8"> section 2 of the Regulation of
Investigatory Powers Act 2000 (meaning of “interception” <abbr title="Et
cetera" xml:lang="la">etc</abbr>), after subsection (8)</ref>insert—</p>
<p class="BlockAmendment">
<mod for="#ref_12">
<quotedStructure class="primary main" startQuote="“" endQuote="”">
<subsection>
<num>8A</num>
<content>
<p>For the purposes of the definition of “telecommunications
service” in subsection (1), the cases in which a service is to be taken to consist
in the provision of access to, and of facilities for making use of, a
telecommunication system include any case where a service consists in or
includes facilitating the creation, management or storage of communications
transmitted, or that may be transmitted, by means of such a system.</p>
</content>
</subsection>
</quotedStructure>
</mod>
</p>
</content>
</section>
Figure 7: Annotation for the provision in Figure 5

4

Akoma Ntoso Annotation for Japanese Statutes

In this Part, we consider how to apply Akoma Ntoso to Japanese statutes.
We first explain the Japanese Statutory Schema in Part 4.1. We then propose
a method to convert JSS to Akoma Ntoso in Part 4.2 and report the
implementation of an automatic XML converter from JSS to Akoma Ntoso in
Part 4.3. In Part 4.4, we address the problem of unnatural annotation
resulting from Akoma Ntoso’s restrictions.
Note that the Akoma Ntoso Schema requires a ‘meta’ section, which
indicates what the resource is about and how it can be accessed. Since the
content of the ‘meta’ section in JSS is the same as that of Akoma Ntoso, we
omit further discussion of it in our paper.
18

18

Fabio Vitali and Flavio Zeni, ‘Towards a Country-independent Data Format: The
Akoma Ntoso Experience’ in Giovanni Sartor, Carlo Biagioli, and Enrico Francesconi
(eds), Proceedings of the V Legislative XML Workshop (European Press Academic
Publishing, 2007) 67.

62

Journal of Law, Information and Science

Vol 24(2) 2016

4.1 The Japanese Statutory Schema

In the Japanese Statutory Schema (JSS), which is based on DTD in JLT, many
elements and structures are defined to fit Japanese statutes. This schema is so
strict that the documents validated by it are guaranteed to conform to the
descriptive Japanese statute styles in document structure. Since JSS is
designed based on a thorough study of the styles of Japanese statutes, we
can annotate all statutes in accordance with the schema without ambiguity.
19

20

In this paper, we discuss the DTD used in JLT, which is basically the same as
the current version of JSS, while JSS conforms to W3C (World Wide Web
Consortium) XML schema. JSS will be updated for the launch of e-LAWS.
Under JSS, the structure shown in Figure 8 is the only valid annotation for the
article in Figure 2. Based on the hierarchical structure of actual Japanese
statutes, this schema does not allow elements other than ‘Paragraph’ to be
children of ‘Article,’ or those other than ‘Item’ or ‘Class’ to be children of
‘Paragraph.’
In JSS, the 12 hierarchical elements shown in Table 2 are defined. This
hierarchy is strictly ordered except for the specified skipped levels.
Therefore, ‘Chapter’ cannot be a child of ‘Section.’
JSS concretely defines not only the hierarchical structures, but also the type
and order of their content. For example, no element can be used as the
children of ‘Chapter’ other than ‘ChapterTitle’ and ‘Section,’ and only
‘Sentence’ and ‘Table’ can be contents of ‘Item.’ Since Japanese statutes are
described by strict rules, their hierarchical structures are achieved by a strict
schema.
<Article Num="2">
<ArticleCaption>(Definitions, etc.)</ArticleCaption>
<ArticleTitle>Article 2</ArticleTitle>
<Paragraph Num="1">
<ParagraphNum>(1)</ParagraphNum>
<ParagraphSentence>
<Sentence>The term "Bank" as used in this Act means a person who
engages in Banking under the license from the Prime Minister prescribed in
Article 4, paragraph (1).</Sentence>
</ParagraphSentence>

19

Japanese Law Translation, Related Information
<http://www.japaneselawtranslation.go.jp/rel_info/rel_info_dtd?re=02>.

20

Toyama et al, above n 1.

63

Applying the Akoma Ntoso XML Schema to Japanese Legislation

</Paragraph>
<Paragraph Num="2">
<ParagraphNum>(2)</ParagraphNum>
<ParagraphSentence>
<Sentence>The term "Banking" as used in this Act means commercial
pursuits carried out through any of the following acts:</Sentence>
</ParagraphSentence>
<Item Num="1">
<ItemTitle>(i)</ItemTitle>
<ItemSentence>
<Sentence>Acceptance of deposits or Installment Savings, in addition
to loans of funds, or the discounting of bills and notes; or</Sentence>
</ItemSentence>
</Item>
<Item Num="2">
<ItemTitle>(ii)</ItemTitle>
<ItemSentence>
<Sentence>Carrying out exchange transactions.</Sentence>
</ItemSentence>
</Item>
</Paragraph>
</Article>
Figure 8: The only valid JSS annotation for the article in Figure 2
Table 2: Hierarchical elements defined in JSS
Level
1
2
3
4
5
6
7
8
9
10
11
12

Elements in JSS
Part
Chapter
Section
Subsection
Division
Article
Paragraph
Class
Item
Subitem1
Subitem2
Subitem3

4.2 Converting JSS to Akoma Ntoso

Our aim is to create Japanese statute documents structured by Akoma Ntoso.
Our approach is to convert the documents structured in JSS into those in
Akoma Ntoso through the following two steps:

64

Journal of Law, Information and Science

Vol 24(2) 2016

1.

Create rules to replace the JSS tags with those of Akoma Ntoso.

2.

Design a custom schema of Akoma Ntoso to validate only the
structures that are suitable for Japanese statutes.

We determined 131 rules for 103 elements defined in JSS. Some JSS elements
have several replacement rules. These rules are applied properly according
to the structure around the element. Appendix A lists the rules to replace the
JSS tags with those of Akoma Ntoso. The replacement rules follow some
basic policies. First, for easy access to the semantic information of each
statute, we basically replace a JSS tag with an Akoma Ntoso tag whose
meaning corresponds to that of the JSS tag. For example, we replace the tag
‘LawTitle’
in
JSS
with
‘docTitle,’
as
shown
in
Figure 9, because the meanings of these tags are similar. Table 3 shows some
of the tags replaced in accordance with this policy.
Annotation in JSS
<LawTitle>Act on Special Measures for Enhancement of the Conservation
and Management of Tuna Resources</LawTitle>
Akoma Ntoso conversion
<docTitle>Act on Special Measures for Enhancement of the Conservation
and Management of Tuna Resources</docTitle>
Figure 9: Replacement of JSS tag with corresponding Akoma Ntoso tag
Table 3: Elements in JSS and Akoma Ntoso with similar meanings
JSS
Part
Chapter
Section
Article
Paragraph
Signature
LawNum
LawTitle
ChapterTitle

Akoma Ntoso
part
chapter
section
article
paragraph
signature
docNumber
docTitle
heading

Annotation in JSS
<ParagraphSentence>
<Sentence>The term "Bank" as used in this Act means a person who
engages in Banking under the license from the Prime Minister prescribed in
Article 4, paragraph (1).</Sentence>

65

Applying the Akoma Ntoso XML Schema to Japanese Legislation

</ParagraphSentence>
Akoma Ntoso conversion
<blockContainer class="ParagraphSentence">
<p>
<inline name="Sentence">The term "Bank" as used in this Act means a
person who engages in Banking under the license from the Prime Minister
prescribed in Article 4, paragraph (1).</inline>
</p>
</blockContainer>
Figure 10: Replacement of tags unique to JSS
The second policy is that, when replacing tags unique to JSS, we give an
attribute with the name of the element to the replaced elements in Akoma
Ntoso. For example, as shown in Figure 10, the tags ‘ParagraphSentence’ and
‘Sentence’ are defined in JSS. The former is to contain character-based
content, such as sentences and tables, and the latter is to contain just one
sentence. Since Akoma Ntoso does not have any corresponding tags, we
replace them with ‘blockContainer’ and ‘inline’ with the attributes
‘class="ParagraphSentence"’ and ‘name="Sentence".’
We can replace the structured document shown in Figure 8 with that in
Figure 11. The tables of contents, the supplementary provisions, and the
appendixes in the Japanese statutes structured in JSS can also be converted
into Akoma Ntoso based on our rules.
We also designed a custom schema that is a subset of Akoma Ntoso for the
elements and structure of Japanese statutes. The custom schema validates
only the annotation shown in Figure 11, but does not validate the annotations
shown in Figure 3. Appendix B shows the definitions related to ‘chapter’ in
Akoma Ntoso and our custom schema to validate the Japanese statutes, and
Appendix C illustrates these definitions in terms of their children. The
definition in the custom schema is simplified and only two hierarchical
elements are allowed to be children of ‘chapter.’
<article eId="chp1-art2">
<num>(Definitions, etc.)</num>
<heading>Article 2</heading>
<paragraph eId="chp1-art2-par1">
<heading>(1)</heading>
<content>
<blockContainer class="ParagraphSentence">
<p>
<inline name="Sentence">The term "Bank" as used in this Act means a
person who engages in Banking under the license from the Prime Minister
prescribed in Article 4, paragraph (1).</inline>
</p>

66

Journal of Law, Information and Science

Vol 24(2) 2016

</blockContainer>
</content>
</paragraph>
<paragraph eId="chp1-art2-par2">
<heading>(2)</heading>
<intro>
<blockContainer class="ParagraphSentence">
<p>
<inline name="Sentence">The term "Banking" as used in this Act
means commercial pursuits carried out through any of the following
acts:</inline>
</p>
</blockContainer>
</intro>
<subparagraph eId="chp1-art2-par2-itm1">
<heading>(i)</heading>
<content>
<blockContainer class="ItemSentence">
<p>
<inline name="Sentence">Acceptance of deposits or Installment
Savings, in addition to loans of funds, or the discounting of bills and notes;
or</inline>
</p>
</blockContainer>
</content>
</subparagraph>
<subparagraph eId="chp1-art2-par2-itm2">
<heading>(ii)</heading>
<content>
<blockContainer class="ItemSentence">
<p>
<inline
name="Sentence">Carrying
out
exchange
transactions.</inline>
</p>
</blockContainer>
</content>
</subparagraph>
</paragraph>
</article>
Figure 11: The only valid annotation for the Article 2 outcome of the replacement of
the structure in Figure 2 validated by our custom schema
4.3 Automatic Converter

We also implemented an automatic converter for Japanese statutes from JSS
to Akoma Ntoso.

67

Applying the Akoma Ntoso XML Schema to Japanese Legislation

The converting process consists of the following two steps:
1.

Analyse the structure of JSS documents; and

2.

Re-write the name of the elements and attributes and reorganize the
document structure to adapt the description rules of Akoma Ntoso.

These operations are based on the replacement rules determined in Part 4.2.
We employed Racc, a Ruby-implemented Look-ahead LR (LALR) parser
generator, for (1) and REXML, an XML parser, for (2).
We examined the converter with an input set consisting of 1,276 documents
in JSS format that are Japanese and English versions of 638 Japanese
statutes. We obtained the same number of outputs in Akoma Ntoso format,
which were thrown into the XML processor xmllint (a component of XML
toolkit, LibXML2) for the validation check.
21

22

As a result, we found that 2.7% of the input in JSS format, which is
automatically annotated, includes incorrect annotations, resulting in
conversion failure. In other words, we confirmed that all of the outputs were
validated, as long as the inputs were correctly annotated.
Figure 12 shows an example of a JSS document including incorrect
annotations and its Akoma Ntoso conversion. The attribute ‘Num’ of the
elements ‘Article’ for Articles 1 and 3 is given an identical value ‘1,’ although
the correct values for the former and latter are ‘1’ and ‘3,’ respectively. JSS
validates this annotation despite a semantically incorrect attribute for Article
3.
This kind of incorrect annotation creates incorrect values of attribute ‘eId’ in
the Akoma Ntoso documents, which is generated from the attribute ‘Num’ in
the JSS documents. In the Akoma Ntoso document in Figure 12, the values of
attribute ‘eId’ overlap between two ‘article’ elements, which are converted
from ‘Article’ in the JSS document. The Akoma Ntoso XML schema does not
validate this structure because the schema requires that attribute ‘eId’ of
every element in a single document have a unique value.
After modification of the incorrect annotations of inputs, they were properly
converted and validated in the Akoma Ntoso XML schema.

21

All
of
the
statutes
used
in
this
paper
are
downloadable
at:
Ministry
of
Justice,
Japan,
Japanese
Law
Translation
(2016)
<http://www.japaneselawtranslation.go.jp>.

22

Daniel Veillard, The XML C Parser and Toolkit of Gnome <http://xmlsoft.org/>.

68

Journal of Law, Information and Science

Vol 24(2) 2016

Annotation in JSS
<SupplProvision>
<SupplProvisionLabel>Supplementary Provisions (Act No. 92 of June 20,
2007) (Excerpts)</SupplProvisionLabel>
<Article Num="1" >
<ArticleCaption>(Effective Date)</ArticleCaption>
<ArticleTitle>Article 1</ArticleTitle>
(snip)
</Article>
(snip)
<Article Num="1" >
<ArticleTitle>Article 3</ArticleTitle>
(snip)
</Article>
(snip)
</SupplProvision>
Akoma Ntoso conversion
<hcontainer eId="sup6" name="SupplProvision">
<heading>Supplementary Provisions (Act No. 92 of June 20, 2007)
(Excerpts)</heading>
<article eId="sup6-art1">
<num>(Effective Date)</num>
<heading>Article 1</heading>
(snip)
</article>
(snip)
<article eId="sup6-art1">
<num/>
<heading>Article 3</heading>
(snip)
</article>
</hcontainer>
Figure 12 JSS document including incorrect annotations and its Akoma Ntoso
conversion (Supplementary Provisions in Act for Implementation of the Mutual
Recognition between Japan and Foreign States in Relation to Results of Conformity
Assessment Procedures of Specified Equipment (Act No. 111 of 2001))
(a) Annotation in JSS
<ArticleCaption>(Definitions, etc.)</ArticleCaption>
<ArticleTitle>Article 2</ArticleTitle>
(b) Akoma Ntoso conversion
<num>(Definitions, etc.)</num>
<heading>Article 2</heading>
(c) More desirable conversion (invalid in Akoma Ntoso)
<heading>(Definitions, etc.)</heading>

69

Applying the Akoma Ntoso XML Schema to Japanese Legislation

<num>Article 2</num>
Figure 13: Possible and desirable replacements of JSS elements into those of Akoma
Ntoso
4.4 Unnatural Annotations Due to Restrictions of Akoma Ntoso

As mentioned in Part 4.2, Akoma Ntoso seems applicable to Japanese
statutes. Since JSS is stricter than Akoma Ntoso, it is not difficult to convert
from the former to the latter. However, some constraints on Akoma Ntoso
prevent us from converting the tags in JSS into conventional ones in Akoma
Ntoso.
As shown in Figure 2, articles in Japanese statutes have a caption above their
titles that describes their content. It is impossible to annotate the caption and
the title of articles in Japanese statutes with the relevant tags provided by
Akoma Ntoso because Akoma Ntoso does not allow any elements other than
‘num’ before ‘heading’ in a hierarchical structure, as shown in the definition
of ‘xsd:complexType name="basehierarchy"” in Appendix B. Therefore,
‘ArticleCaption’
and
‘ArticleTitle’
in
Figure
13(a)
are
replaced
with
‘num’
and
‘heading’
in
Figure 13(b). If Akoma Ntoso allowed the order to be shifted between ‘num’
and ‘heading,’ we could replace ‘ArticleCaption’ and ‘ArticleTitle’ with
‘heading’
and
‘num,’
as
shown
in
Figure 13(c).
The Akoma Ntoso XML schema has great flexibility in annotating
hierarchical structure. The same could be true for captions and titles.
Enhancing the flexibility would enable Japanese legislation to adapt Akoma
Ntoso to enhance the schema as a platform for the international exchange of
legal information.
23,

5

Conclusion

We introduced an approach to applying Akoma Ntoso to Japanese statutes.
We identified an underlying problem of the structural ambiguity of Akoma
Ntoso, that is, that the flexibility of annotations allows the schema to validate

23

This issue on the restrictions was modified in the latest version of Akoma Ntoso
(WD17 AN 3.0) in response to our previous paper presented at the Law via the
Internet Conference 2015: see Gen Kawachi, Makoto Nakamura, Yasuhiro Ogawa,
Tomohiro Ohno, and Katsuhiko Toyama, ‘Applying the Akoma Ntoso XML Schema
to Japanese Legislation’ (Paper presented at the Law via the Internet Conference
2015, The University of New South Wales, Australia, 11 November 2015).

70

Journal of Law, Information and Science

Vol 24(2) 2016

unintended structuralisation to the same document. On the other hand, we
can annotate all the statutes in accordance with JSS without ambiguity.
Therefore, we proposed a method to convert JSS to Akoma Ntoso. We
determined the rules to replace JSS tags with Akoma Ntoso and designed a
custom schema for Japanese statutes. Our approach made it possible to cope
with consistency of the strict rules in the descriptive styles of Japanese
statutes and achieve the advantage of Akoma Ntoso documents in
exchanging legal information.
The Japanese legislation system applies strict rules for the styles of statutes,
as reflected in JSS. Namely, we confirmed that the Akoma Ntoso XML
schema does not fit Japanese statutes at the native level, taking processing
overhead into account. On the other hand, as we revealed in this paper, it is
not difficult to convert JSS documents into Akoma Ntoso documents, which
has an advantage in terms of sharing legal information internationally.
Therefore, we can easily obtain structured Japanese legal documents in
Akoma Ntoso format whenever needed.
For these reasons, e-LAWS, which is the new legal information system in
Japan mentioned in Part 2, is expected to employ a JSS-based XML schema,
instead of the direct use of Akoma Ntoso for structured documents of
Japanese statutes. Since we have continuously supported the development of
Japanese legal schemata from JLT, it is not an exaggeration to say that our
work greatly contributed to this decision.
24

6

Acknowledgements

The authors would like to give special thanks to Associate Professor Monica
Palmirani of CIRSFID, University of Bologna, for her valuable comments.
This research was partly supported by the Japan Society for the Promotion of
Science KAKENHI Grant-in-Aid for Scientific Research (S) No.23220005, (A)
No.26240050, and (C) No.15K00201

24

Toyama et al, above n 1.

Collaboration between the state and the LII community
in improving access to legislation in a national
jurisdiction: a New Zealand example
DAVID NOBLE*
Abstract
After a brief and necessarily truncated review of some of the historical examples of state
provision of access to, and promulgation of, legislation (with a particular focus on the
Magna Carta given its 800 anniversary in the year of the conference) this paper
examines the approach taken in New Zealand since January 2008. From that date, when
the government’s legislation drafting and publishing system (LENZ) went live, the
New Zealand state institution with responsibility for the publication of legislation, the
Parliamentary Counsel Office (PCO) began making legislation freely available via the
web as well as via the traditional, and charged-for, printed paper version.
th

The relationship between the government-provided free access website
<http://www.legislation.govt.nz> and that provided by the New Zealand Legal
Information Institute <http://www.nzlii.org> is examined as an example of the type of
collaboration between the state and the LII community which can provide improved
access to legislation by the public in one national jurisdiction (and internationally).
The paper considers the scope for state institutions such as the PCO in New Zealand to
further improve access to, and the accessibility of, historic legislation of all kinds
through such collaboration with local and regional Legal Information Institutes such as
NZLII, AustLII and PacLII.
1

Introduction

Thomas Hobbes’ much quoted conclusion on the need for the promulgation
as well as the passing of legislation: ‘For to the nature of Lawes belongeth a
sufficient and clear Promulgation, such as may take away the excuse of
Ignorance’ remains as apposite now as it was in 16 century England. In the
20 Century in New Zealand the then Chief Justice, Sir Richard Wild said
much the same thing in a case concerning the future of the governmentowned print service:
1

th

th

*

1

Chief Parliamentary Counsel and Chief Executive of the Parliamentary Counsel
Office, New Zealand. October 2007 May 2016. This article is professional, non-peerreviewed expert commentary.
Thomas Hobbes, Leviathan (Scholar P, 1969).

72

Journal of Law, Information and Science

Vol 24(2) 2016

People must be told what Parliament is doing and must be able to
read the letter of the law.
2

In a seminal 1989 report, the New Zealand Law Commission emphasised
their view of the constitutional significance of the promulgation of law
within the jurisdiction:
Our purpose here is to stress again the central constitutional function
of the Government in making the law… known to the public.
3

2

A Selected History of Promulgation of Legislation

It seems axiomatic that those who produce laws, and in particular legislation
(as distinct from the growth of judge-made law through the development of
the common law or civil law processes) bear an equivalent burden or duty to
make those laws known to the subjects or citizens of that jurisdiction.
Looking briefly at ancient history in this respect there is evidence of early
promulgation in the form of the Code of King Hammurabi from the ancient
kingdom of Babylonia (18 Century BCE). Partial copies of the code exist,
carved into a 2.25 metre tall diorite stele column from Sippar and baked into
clay tablets (which can be viewed in the Louvre Museum, Paris, France). The
Code consists of 282 laws, with scaled punishments, adjusting ‘an eye for an
eye, a tooth for a tooth’ depending on social status. Interestingly however,
nearly one-half of the Code deals with matters of contract law, establishing,
for example, the wages to be paid to a driver of oxen, or a surgeon. Other
provisions specify the necessary terms of a transaction and establish the
liability of a builder for a house that collapses.
th

4

The Louvre records that over 30 copies, or partial copies of the Code have
been found on separate clay tablets (or stela) at Susa and in the wider
Mesopotamian region, indicating a clear understanding of the need for the
State to disseminate the content of laws to the subjects of those laws.
Interestingly for those charged now with promulgating legislation on behalf
of parliaments and other legislators, differences in the texts between those
stela shows that there were various versions of the Code.

2

VUWSA v Government Printer [1973] 2 NZLR 21.

3

Law Commission, Legislation and its Interpretation (NZLC R11, 1989).

4

Louvre Museum, The Code of Hammurabi (2016) Louvre
<http://musee.louvre.fr/oal/code/indexEN.html>.

Collaboration between the state and the LII community in improving access to legislation

73

By the 13 Century CE in England, dissemination of laws had moved on from
stone pillars and clay tablets to something more portable – high quality
calfskin made into vellum. This has more immediately obvious similarities to
the printed paper legislation of more recent times, although clearly still from
the pre-printing press era of mass production. It was this medium that was
used for the original and the subsequently transcribed copies of the Magna
Carta (1215). The Australian government is fortunate to hold, in the
Parliament House in Canberra, a good example of the 1297 reissue by King
Edward I, of the infamous bad King John’s Charter from 800 years ago.
th

5

It is worth noting, in passing, that depictions of the King signing the charter
(and its accompanying Charter of the Forests) are faulty as shown by the
copy in the Australian parliament: the Royal Assent was evidenced by the
attachment of the Great Seal rather than by a handwritten signature.
Promulgation (and drafting probably) depended a great deal on the Church
since, to an extent, King John had no interest in ensuring that the rights of
Barons and Freemen and the restrictions on his Royal Sheriffs, to which he
had agreed, were widely known. For this reason it seems likely that the so-

6

7

5

Research Section, Department of the Senate, Magna Carta (Great Charter), 1297,
Parliament House Canberra <http://www.magnacarta.senate.gov.au/>.

6

John appears seated on a throne, holding a sword in his right hand and a sceptre in
his left, and with the legend ‘+ IOHANNES DEI GRACIA REX ANGLIE DOMINVS
HIBERNIE’ (John, by the grace of God King of England and Lord of Ireland). On the
reverse is an equestrian portrait of John, riding with his visor open. The legend on
this side gives the king’s other titles, ‘+ IOHANNES DUX NORMANNIE ET
AQVITANNIE COMES ANDEGAVIE’ (John, Duke of Normandy and Aquitaine,
Count of Anjou).

7

Only four copies of Magna Carta dating from 1215 have survived the ravages of time
and Salisbury Cathedral is home to the best-preserved original manuscript. Elias of
Dereham, priest and steward of the archbishop of Canterbury is thought to have
brought Salisbury’s copy to Old Sarum in the days following the events at
Runnymede and it has remained in the Cathedral’s care ever since. A second copy of
the 1215 Magna Carta owned by Lincoln Cathedral is displayed at Lincoln Castle. Its
particular quality lies in the fact that it is written in an ‘official’ hand since Hugh of
Wells, then Bishop of Lincoln, was present at the sealing of the original charter at
Runnymede. This, together with the role of Stephen Langton, the Archbishop of
Canterbury at the time, and a significant mediator in the dispute between King John
and his Barons, who had been educated in Lincoln (a seat of radical thinking and
learning in the 12th and 13th Centuries) ensured that Lincoln was one of the
Cathedral cities to receive a copy of the charter. Interestingly the Lincoln copy,
received only four days after its sealing, contains numerous errors indicating the
haste with which the ‘official’ scribes were working. The other two surviving copies
of the 1215 version of Magna Carta are housed in the British Library. See Salisbury
Cathedral, Magna Carta Exhibition <http://www.salisburycathedral.org.uk/visit-whatsee/magna-carta-exhibition>;
Lincoln
Castle,
Magna
Carta

74

Journal of Law, Information and Science

Vol 24(2) 2016

called ‘Cathedral copies’ were transcribed by church scribes and despatched
to the various dioceses of England so that the contents of the charter could be
read out or at the very least made known more widely.
Finally, in this very brief and selective review, moving hemispheres and a
further 600 years to New Zealand during the period of early European
settlement (before the establishment of full self-governing assemblies or
Parliaments in 1854) great reliance was still being placed initially on the
Church and (later) the private sector to print, publish and promulgate laws.
William Colenso, originally from Penzance in Cornwall, by 1833 was
working in the London firm of Richard Watts, printers to the Church
Missionary Society (CMS). The CMS was seeking someone to set up and run
a printing press in the mission settlement at Paihia in the Bay of Islands.
Colenso was recruited and left England bound for Australia on the Prince
Regent in June 1834. He finally arrived at Paihia on board the schooner
Blackbird from Sydney on the 30 December 1834. Despite defective
equipment supplied by the CMS Colenso was highly productive as a printer.
Te Ara - The Encyclopedia of New Zealand records that, in addition to
printing the first pamphlet printed in New Zealand (a 16 page translation
into Maori of the Epistles of Paul to the Philippians and to the Ephesians on
17 February 1835) and the production of 5,000 copies of William Williams's
356 page Maori New Testament books, Colenso was also employed
producing government publications:
In October 1835 the first tract produced in English was printed by
order of the British Resident, James Busby, warning settlers about the
imperialist ambitions of the French Baron Charles de Thierry. Over
the following nine years other official notices and publications
appeared, including the first New Zealand government Gazette on
30 December 1840. Colenso’s most memorable work of this sort was
the printing of the Maori text of the Treaty of Waitangi on 17
February 1840. At the signing his cautious representations to
Lieutenant Governor William Hobson that many Maori were
unaware of the meaning of the treaty were brusquely set aside. His
observations recorded at the time were published as ‘The authentic

<https://www.lincolncastle.com/content/magna-carta>; British Library, Magna Carta,
<http://www.bl.uk/magna-carta>.

Collaboration between the state and the LII community in improving access to legislation

75

and genuine history of the signing of the Treaty of Waitangi (1890)’,
the most reliable contemporary European account of the signing.
8

The official Government Printing Office was not formally established until
1864, its purpose being to ensure the publication of legislation and Hansard
(10 years after the parliament first met in Auckland). During the 19 and 20
Century Parliament and government did not stand still in terms of
legislating for, and considering their role and duties in relation to, the
provision of access to the legislation that they were rapidly introducing and
passing. Similarly, considerable effort was put into producing compiled and
consolidated versions of legislation, particularly in the early 20 Century.
However, the resources and commitment to keep the statute book (let alone
the subordinate legislation ‘book’) up-to-date were never sufficient in a
printed paper-based environment. So, increasingly the late 20 and early 21
Century reports and legislation look more to developing technologies, the
internet and electronic/digital sources to solve that problem.
th

th

th

th

3

st

State Provision of Free Access to Legislation in New Zealand
2007 Onwards

The Parliamentary Counsel Office in 2007 was however, still supplying huge
amounts of printed hard copy to the House of Representatives, holding a
significant appropriation from Parliament each year for printing copies of
primary and subordinate legislation and then making it available through
designated bookshops across the country. Given the size of some of these
individual statutes (at Royal Assent, the Income Tax Act 2007 was 2855
pages long) change was long overdue.
But there was a slight ‘local difficulty’ in this move from paper to web-based
publishing and promulgation. The government no longer owned the
intellectual property in the marked-up database of old and existing
legislation having (inadvertently) sold it with the deregulation of
government printing and the associated sale of the Government Printing
Office (GPO).
9

8

David Makay, Colenson, William (30 October 2012) Dictionary of New Zealand
Biography;
TEARA
Encyclopaedia
of
New
Zealand,
<http://www.teara.govt.nz/en/biographies/1c23/colenso-william>.

9

To misquote former UK Prime Minister Harold Macmillan ‘… and so I thought the
best thing to do was to settle up these little local difficulties, and then turn to the
wider vision of the Commonwealth.’ See Harold Macmillan, ‘Mr Macmillan sets out’,
The Times (UK) 8 January 1958, 8

76

Journal of Law, Information and Science

Vol 24(2) 2016

In 1987 the GPO had revenues of just under $100 million with pre-tax profits
of $14.5 million and, as one commentator noted: ‘Its parliamentary
publications kept citizens informed in great and sometimes crushing detail
of their government's activities; its legislative publications guided citizens
(or at least their lawyers) on their rights and duties.’ The consequences of
the sale were noted at the time by some commentators, for example an article
by Warren Berryman in the leading business newspaper in New Zealand
was titled ‘Sovereignty for sale: the law to be privatised in Printing Office
move.’
10

11

12

Consequently, the move to web-based publishing and access to legislation
was both lengthy and costly for New Zealand. In total it has taken 10 years
and $30 million (against early predictions/estimates given in the early 2000s
of 2 years and $5 million) to achieve the digital legislative drafting and webpublishing systems and environments that are currently provided in New
Zealand. Since 2008 the government (through the medium of PCO’s
legislation website) has been making available in-force legislation (primary
and secondary) and Bills in both internet and hard copy formats. And in
2014 the PCO finally completed the long process of checking and correcting
the database that had been purchased from the commercial publishers and
was able to declare the PDF output an official source of legislation.
13

14

This is critical for a properly functioning democracy; to have a reliable,
authoritative and official source of the laws which apply to its citizens and
others. To have that body of legislation available free to all is also a
significant improvement over the previous charged-for paper, based product

10

Report of the Government Administration Committee on the Inquiry into the Sale of the
Government Printing Office (1992) House of Representatives, New Zealand.

11

James Bruce Ringer, Book and Print in New Zealand: A Guide to Print Culture in Aotearoa
(Victoria University Press, 1997).

12

Warren Berryman, ‘Sovereignty for sale: the law to be privatised in Printing Office
Move’, National Business Review (New Zealand) 8 December 1988.

13

New
Zealand
Legislation
(2016)
<http://www.legislation.govt.nz>.

14

Christopher Finlayson, ‘Official legislation online today’ (Press Release, 6 January
2014) <https://www.beehive.govt.nz/release/official-legislation-online-today>. See
also ‘Legislation (Official Electronic Versions) Notice 2014’ (2014) 1 New Zealand
Gazette 32. From 5 January 2016 HTML content of the website was declared also to be
an official source of legislation. See ‘Legislation (Official Electronic Versions) Notice
2015’ (2015) 141 New Zealand Gazette 7126.

Parliamentary

Counsel

Office

Collaboration between the state and the LII community in improving access to legislation

77

which inevitably was always out-of-date given the ability to produce
compiled copies of legislation sufficiently quickly.
15

The declaration of ‘official’ status is also essential for the operation of
‘judicial notice’ in court proceedings under the law of evidence in New
Zealand.
16

And, it discharges Hobbes’ requirement for there to be a ‘sufficient and clear
promulgation’ of the legislation passed by Parliament or made by the
Executive.
Finally, in relation to the digital database output and its official status it
should be noted that printed copies of assented-to legislation are required by
the Standing Orders of the House of Representatives to be kept by the Clerk
of the House and the Registrar of the High Court at Wellington. So, there
will always be authoritative hard copies to refer to and reprint if necessary
and the PCO itself continues to maintain a stock of hard copy printed
legislation for reasons of security and longevity. However, the reality in
terms of public access to legislation (including access by legal professionals
and librarians) is that the official website output is fast replacing the official
printed paper version as Figures 1 and 2 below indicate.

15

For more discussion see ‘Law News gets the heads up on official electronic
legislation’, ADSLI LAWNEWS (13 December 2013), <http://www.adls.org.nz/forthe-profession/news-and-opinion/2013/12/13/law-news-gets-the-heads-up-onofficial-electronic-legislation/>.

16

Legislation Act 2012 (NZ) s 16.

78

Journal of Law, Information and Science

Vol 24(2) 2016

Figure 14 Source: Parliamentary Counsel Office, Wellington, New Zealand

Figure 15 Source: Parliamentary Counsel Office, Wellington, New Zealand

4

Collaboration between Free Access to Law Movement
members in New Zealand and the wider Pacific region

The PCO (as the State provider of free access to legislation) began its
collaboration with the New Zealand Legal Information Institute (NZLII) for
primarily practical reasons. First, NZLII already had a substantial body of
both legislation and case law freely accessible via its website and the

Collaboration between the state and the LII community in improving access to legislation

79

government website was designed in such a way to allow anyone to
download content by ATOM feed so collaboration between two free-access
providers made sense in a small jurisdiction where there are already three or
four main commercial retailers of legislation and add-on services. Second,
and rather more pragmatically, applicable government web standards
required the PCO to publish documents in both PDF and HTML formats
which the official legislation website conforms to. However, the PCO
holding of older databases was only available in PDF or indeed hard copy
formats which required OCR scanning prior to PDF production. The costs
and time involved in converting these historic (but valuable) research
collections into HTML format and then incorporating them into the NZL
website databases did not represent value for money for the PCO given its
appropriation and other priority functions.
However, the PCO was tasked with, and wanted to, improve access to
legislation for New Zealanders, so again, collaboration with NZLII made
sense in terms of publishing the PDF historic legislation content.
Consequently the PCO, working with its free access partner (NZLII), is now
making significant historical legislative documents freely available online.
The latest results of this collaboration, Bills from 1949 to 2008, are now on the
NZLII website in PDF format. The volumes will be added to over time and
will extend back to 1854 when complete. The collection includes Bills that
became Acts as well as those that failed to make it to the statute book.
17

The Historical Bills Collection sits alongside the New Zealand Acts 1841–
2007 As-Enacted Collection of historical Acts and the recently completed
Statutory Regulations 1936–2007 as-made collection.
18

As an example of cost-effective collaboration, the additions to the existing
NZLII databases of the Regulations as-made and the Acts as-enacted
collections together with the current project to produce a comprehensive
collection of historic Bills will have cost the PCO less than $200,000 in

17

New Zealand Parliamentary Counsel Office, New Zealand Historical Bills (11 March
2016)
New
Zealand
Legal
Information
Institute
<http://ww.nzlii.org/nz/legis/hist_bill/>. The collection has been put together
from scans of bound volumes provided by the PCO and the Office of the Clerk to the
House of Representatives, and the documents hosted by NZLII. While most volumes
were taken apart to enable scanning, some of the earliest (not yet available online)
were too rare to treat in this way. Instead they were scanned by the Australasian
Legal Information Institute (AustLII in Sydney) using a non-destructive scanner that
produces high-quality images from books open to only 30°.

18

New Zealand Parliamentary Counsel Office, New Zealand Regulations As Made (3 May
2016)
New
Zealand
Legal
Information
Institute
<http://www.nzlii.org/nz/legis/num_reg/>.

80

Journal of Law, Information and Science

Vol 24(2) 2016

scanning and processing costs. This is a fraction of the charges it had
previously paid to commercial providers of the interim website offerings
before the official legislation website went live.
In terms of the wider Pacific region the PCO has, since 2009, been providing
a legislative drafting, training and mentoring function for Pacific Island
nations with financial support from the New Zealand ministry of Foreign
Affairs and Trade (starting with those of the NZ Realm States, Cook Islands,
Niue and Tokelau but since expanded to include work for Fiji’s reestablishment of Parliamentary democracy).
19

Access to legislation in the Pacific island states is often difficult, particularly
in hard copy form which makes the regional LII – (The Pacific Islands Legal
Information Institute (PacLII) a facility of the University of the South Pacific
School of Law) all the more important for the New Zealand legislative
counsel when drafting amending legislation on instructions from
jurisdictions in the Pacific.
20

In Australasia state legislation publishing institutions in the Australian
jurisdictions have different and varied relationships with local free-access
providers – but in New Zealand the PCO has committed to developing a
collaborative relationship with NZLII (its free-access partner) and indeed
encouraging other free-access providers, combiners and developers of the
official output.

19

For more information see the latest evaluation report available from MFAT at: New
Zealand Foreign Affairs and Trade, Evaluation Reports 2015 (August 2015) New
Zealand
Foreign
Affairs
and
Trade/Manatu
Aorere
<https://www.mfat.govt.nz/en/aid-and-development/our-approach-toaid/evaluation-and-research/evaluation-reports-2015/>.

20

Pacific Islands Legal Information Institute (July 2016) PacLII <http://www.paclii.org/>.

Publishing Family Court judgments: problems and
solutions
LYN NEWLANDS*
Abstract
Why does the Family Court publish its judgments in anonymised form? What are the
practical options for anonymisation? How are pseudonyms allocated? How does the
Court ensure that judgments are sufficiently anonymised to avoid identification of
parties and their children without impacting on the integrity of the Judge’s reasons?
How does the Court balance the competing needs for public access and parties’ privacy?
This paper provides an overview of judgment anonymisation and publication by the
Family Court, and examines some of the associated problems.

1

Introduction

The Family Court of Australia is a superior court of record, tasked to assist
Australians to resolve their most complex legal family disputes. As a matter
of interest, it is not a ‘closed’ Court and anyone is able to observe its
hearings; only rarely are hearings held in camera. Judgments should ideally
be made widely available to promote access to justice for precedent and
public information purposes. This is particularly important, given that up to
22 per cent of litigants in the Court are self-represented at some stage during
the course of their matter.
Historically, until 2007, limited numbers of judgments were published online
based solely on their precedential value. This had the effect of severely
limiting the amount of publicly available family law case law and promoted
the perception of the Court as a secretive organisation, exempt from public
scrutiny. It also meant that first instance judgments, which provide a
window into the conduct of proceedings and contribute to transparency,
were inaccessible to the wider community.
In preparing family law decisions for publication extra resourcing, over and
above that required in other jurisdictions to undertake the same task, must
be provided to anonymise them. Anonymisation is the deceptively simple
task of removal and/or replacement of any identifying information in the
judgment – whether that information is factual or contextual.

*

Judgments Publication Coordinator, Family Court of Australia. This article is
professional, non-peer-reviewed expert commentary.

82

Journal of Law, Information and Science

Vol 24(2) 2016

2

Family Law Act 1975 (Cth) s 121 – Restriction on publication
of court proceedings

Under the Act, publication by any means of information that would identify
parties and witnesses is a criminal offence. Section 121(1) reads:
A person who publishes in a newspaper or periodical publication, by
radio broadcast or television or by other electronic means, or
otherwise disseminates to the public or to a section of the public by
any means, any account of any proceedings, or of any part of any
proceedings, under this Act that identifies:
(a) a party to the proceedings;
(b) a person who is related to, or associated with, a party
to the proceedings or is, or is alleged to be, in any other way
concerned in the matter to which the proceedings relate; or
(c) a witness in the proceedings;
is guilty of an offence punishable, upon conviction by imprisonment
for a period not exceeding one year.
Fortunately, s 121(3)(a) provides an effective starting point for
anonymisation. It lists the sort of ‘particulars’ that should be taken to be
identifying:
(3) Without limiting the generality of subsection (1), an account of
proceedings, or of any part of proceedings, referred to in that
subsection shall be taken to identify a person if:
(a) it contains any particulars of:
(i) the name, title, pseudonym or alias of the person;
(ii) the address of any premises at which the person resides
or works, or the locality in which any such premises are situated;
(iii)

the physical description or the style of dress of the

person;
(iv) any employment or occupation engaged in, profession
practised or calling pursued, by the person or any official or
honorary position held by the person;
(v) the relationship of the person to identified relatives of the
person or the association of the person with identified friends or

Publishing Family Court judgments

83

identified business, official or professional acquaintances of the
person;
(vi) the recreational interests, or the political, philosophical
or religious beliefs or interests, of the person; or
(vii) any real or personal property in which the person has
an interest or with which the person is otherwise associated;
being particulars that are sufficient to identify that person to a
member of the public, or to a member of the section of the public to
which the account is disseminated, as the case requires;
This list is obviously not exhaustive and comprises largely factual
information. Online publication of judgments increases the complexity of
anonymisation because the use of search engines such as Google (depending
on the expertise of the user) makes ‘joining the dots’ to identify
parties/children much easier than in the ‘old days’ of print and paper.

3

Family Court publication policy post-2007

In 2007, the Chief Justice introduced a new judgments publication policy
whereby almost all of the Family Court’s judgments would be published on
AustLII, including both substantive and interim judgments at first instance
and on appeal. To give an idea of the scope of that change of policy, prior to
2007 approximately 200–250 judgments were published annually; since 2007,
that figure has increased to around 1500 fully-anonymised judgments
published per annum.
The Judgments Publication Office (‘JPO’) was created to develop, manage
and undertake the publication process. Under the guidance of senior judicial
officers, a standard judgments template, judgments production style guide,
anonymisation guidelines and a family law specific catchwords list were
developed and implemented.
The JPO, with a permanent staff of two supplemented with occasional
contract assistance, depending on workload requirements, carries out the
anonymisation and publication of judgments. Exception to the publication
policy may be made where a suppression or non-publication order is made
by the Judge – in which case, the JPO must be notified to avoid ‘accidental’
publication. Such orders are relatively rarely made.

4

Anonymisation principles

When anonymising judgments, three principles need to be borne in mind –
privacy of the parties and children; integrity of the Judge’s reasons; and

84

Journal of Law, Information and Science

Vol 24(2) 2016

ensuring the finished product is intelligible. These principles may at times
conflict with each other, requiring a delicate balancing act to achieve
effective anonymisation of the judgment at hand.

5

Anonymisation – which option?

The general practice of the Family Court has been to use generic language in
judgment writing. In family law judgments, using terms such as ‘the
mother’, ‘the father’, ‘the child’, ‘the paternal grandparents’ etc, or legal
classifications (for example, ‘applicant’, ‘respondent’), can actually be more
easily read and understood than the use of names ad infinitum, particularly
if the matter involves a large extended family with intergenerational namesharing. Writing in generic language has the additional benefit of minimising
the amount of anonymisation required to make the judgment publishable.
For courts undertaking routine anonymisation of judgments, there are two
options from which to choose.
Partially automated anonymisation uses software
replace pre-defined words from text. This process
subsequent quality control check to identify and
anonymisation required. This is the option chosen
Court of Australia

designed to find and
is supplemented by a
undertake any further
by the Federal Circuit

Manual anonymisation is a traditional ‘word by word, line by line’ editing
process where individual words and phrases are substituted by less
identifiable terms/phrases, for example, an ‘electrician’ might become
‘skilled tradesman.’ As an incidental consequence of the anonymisation
process, the judgments officer undertakes an informal proof reading role
whereby minor punctuation, grammatical and/or formatting errors may be
corrected, and more serious problems referred back to chambers for
resolution, prior to external publication of the judgment. This is the option
selected by the Family Court of Australia.

6

Why use pseudonyms?

Prior to 2007, sensitive Family Court judgments were usually anonymised by
the use of the first letter of the parties’ surnames or their initials place of
names. With the increased publication, that practice fell out of favour –
within 1500 judgments per annum there could potentially be a hundred or
more ‘B & B [2015] FamCA … ’, most of which would be unrelated matters.
The decision to replace party names with pseudonyms was taken for several
reasons:

Publishing Family Court judgments

85

It is easier for litigants and lawyers to remember actual names rather
than letters/initials for citations.

It is an effective way to protect the parties’ privacy.

It enables the media to report on family law cases without risk of
breaching s 121.
1

7

Pseudonyms are allocated to the matter, rather than the individual
decision, which enables the reader to track back across multiple
decisions in the same matter.

Factors to consider when allocating a pseudonym

Publication of its judgments provides a window into the client base and
operation of a court. To facilitate that transparency for the Family Court it is
desirable the pseudonym that is allocated to a matter is as akin to reality as is
possible within the constraints of s 121.
2

This is achieved by using pseudonyms that:

8

Begin with same letter as the party’s actual name.

Be ethnically appropriate.

Be culturally sensitive.

Be respectful to the parties.

Be phonetically dissimilar to the party’s actual name.

Anonymisation – how to anonymise a Family Court judgment

The document’s Advanced Properties fields are checked and any personal
metadata about judges and staff is removed. The Microsoft Word user profile
for JPO staff is set up so that all documents are saved as ‘Family Court of
Australia.’
The first step in anonymisation of a judgment is the identification of an
existing pseudonym or the allocation of the pseudonym if a new matter.

1

Family Law Act 1975 (Cth) s 121.

2

Ibid.

86

Journal of Law, Information and Science

Vol 24(2) 2016

Only parties identified on the cover sheet of the judgment as applicant(s) and
respondent(s) are given a pseudonym, all other names used in the judgment
are anonymised by letters or generic terminology.
As each letter used stands in place of a word, it is important that the same
letter should not be used for different words. Of necessity, particularly when
anonymising complex judgments, there may be times where this can only be
achieved by using combinations of letters.
Government agencies and large business entities do not usually require
anonymisation, unless it is apparent that failure to do so would lead to the
identification of a party or witness. For example, the fact that the parties hold
accounts with the Commonwealth Bank would not be identifying unless
account numbers were included.
Private business entities and smaller companies with which the parties have
some involvement are anonymised where possible using generic language.
However, in complex property matters, for example, it will often be
necessary to replace company names with letters. In that case it is important
to ensure that only the name of the company is edited and that the
appropriate indicator of the business type (Pty Ltd, Limited, PLC, LLC, etc.)
is retained.
Significant dates which are likely to be in the public domain, such as dates of birth,
marriage and death, should be reduced to the year only where possible, or
month and year if necessary. It is not usually necessary to anonymise dates
of separation, court-related dates, dates of personal conflicts, etc.
Geographical information such as residential and business addresses can be
anonymised using such terms as ‘the former matrimonial home’, ‘the rural
property’, ‘overseas.’ Again, particularly in more complex decisions, it may
be necessary to use letters combined with contextual descriptors. In this era
of global business and internet romance, it is important to note that the
anonymisation of overseas countries is very much dependent on context –
for example, the requesting country in a Hague Convention judgment about
international child abduction is not anonymised, given that the decision
relates to international jurisdictional issues.
Unique numerical identifiers (vehicle registration, credit card number, phone
number, property folio number, ABN numbers etc) are highly identifiable
online, and should always be either replaced by an ellipsis, or an ellipsis and
the last digit(s) of the number.
Other identifying details (occupation, sports clubs, religion, volunteering,
politics, recreational pursuits etc) are anonymised as required using generic
descriptions or letters.

Publishing Family Court judgments

87

Quotes from other proceedings involving the same parties, including from
proceedings in other jurisdictions, are anonymised, with any changes
enclosed in square brackets. Particular care needs to be taken where quotes
are from other jurisdictions – it may not be possible to anonymise such
quotes sufficiently to prevent identification by a skilled internet searcher. In
that case, permission may be sought from the judge to omit part or all of the
quoted text with an appropriate notation (for example, ‘Sentencing reasons
omitted to comply with Family Law Act 1975 (Cth) s 121’).

9

Recognising and anonymising contextual information risks

Contextual information in the form of apparently innocuous facts that in and
of themselves would not be identifying but that can, in combination with
each other, become identifying, may be difficult to recognise when
anonymising judgments. This problem is compounded by the power of
internet search engines to crawl across diverse unrelated websites and
harvest relevant information.
Contextual information challenges may arise simply out of the use of
language that is specific to a particular occupation or circumstance. For
example, there is no point anonymising an occupation related to the military
where one party has been dismissed from the service, without then
anonymising the term ‘dishonourable discharge.’
A further challenge is faced where there is media coverage, unrelated to the
family law proceedings, which enables the astute online searcher (be he/she
media reporter, extended family member, child’s school friend, or local
neighbourhood gossip) to identify the parties and their children. One
example will suffice – an interlocutory judgment was published by the
Family Court wherein one of the parties was an international sportsman who
was the recipient of some lucrative sponsorship deals. The Judge was
contemplating how sponsorship deals should be dealt with in the context of
property settlement and that piece of information could not be anonymised
without impacting on the integrity of the reasons. Unfortunately,
‘sportsman’ and ‘sponsorship’ were sufficient for a well-known media outlet
to generate stories about the case which, while they complied with the letter
of the law by not actually naming them, provided sufficient detail for their
readership to identify the parties.
The onus of recognising and resolving contextual information risk lies with
the judgments officer undertaking the anonymisation. Recognition is
facilitated by maintaining awareness of diverse current events and the
application of lateral thinking. Resolution may involve editing the
information to make it less identifiable, or it may involve discussions with
the relevant judicial officer to negotiate a solution.

88

Journal of Law, Information and Science

Vol 24(2) 2016

10 Anonymisation keys
Consistent anonymisation promotes the ability to read multiple judgments in
the same matter without having to constantly revise ‘who’s who’ and ‘what’s
what’, thus making the judgments more user friendly and accessible for
research purposes.
To facilitate consistent anonymisation, the JPO compiles and maintains
running sheets (aka anonymisation keys) to record words and phrases that
are edited in the course of anonymisation. Essentially each key is a Microsoft
Word document containing a two-column table, linking changes made from
the original to the anonymised version. The keys are stored on a shared
drive, making them accessible to all JPO staff and can be provided to other
Court staff on request.
The keys ensure anonymisation is applied consistently throughout the life of
a matter, regardless of the number of judgments delivered in that matter.
Once the key is created, it can be added to as necessary when later
judgments in that matter are anonymised.

11 Transparency – public access v parties’ and children’s privacy
The internet is a powerful platform for courts to publish their judgments,
providing access and transparency in decision-making to the profession and
the wider community. For courts with family law jurisdiction, however, the
internet brings with it some unique challenges.
‘Jigsaw identification’ is the identification of a person, achieved by
combining two or more pieces of information (for example, a party’s
occupation and his/her conviction of a criminal offence) from two or more
sources (for example, a judgment and a Google search). It poses a particular
problem where either or both of the parties have been involved in activities
that have resulted in them acquiring a high level media profile. That profile
may result from involvement in criminal proceedings but may equally arise
from involvement in philanthropy, high society, elite sports, or any other
circumstance that leads to media reporting.
3

Use of social media such as Facebook or Twitter, either deliberately (as a
means of disparaging the other party) or inadvertently (recording personal
news and events), can also put highly identifiable personal information in
the public domain in a way that was not previously possible. And it is not

3

H v A (No 2) [2015] EWHC 2630 (Fam).

Publishing Family Court judgments

89

unknown for an unhappy litigant to seek validation by creating a personal
website to vent their dissatisfaction. Such online activity can render
successful anonymisation of judgments extremely difficult, especially where
reference to that activity itself features within the judgment.
In the family law online environment the demand for public access to
judgments competes with the statutory requirement for privacy. In the
anonymisation process, securing the parties’ privacy vies with the need to
maintain the integrity of the reasons—if anonymising the judgment
sufficiently to prevent identification undermines or destroys the integrity of
the judicial officer’s reasons, then publication is a futile exercise.
In applying an approach of ‘must retain if ratio, may omit if obiter’, it is
essential that JPO staff are able to recognise when anonymisation threatens
to encroach on the integrity of the reasons, at which stage the judicial officer
will need to be informed/consulted. In that situation, the Judge will decide
whether or not the offending information can be properly omitted.

12 Conclusion
Since 2007 the Family Court of Australia has endeavoured to publish as
many of its judgments as possible. It has provided the extra resources
required to enable it to do so in compliance with the requirements of s 121 of
the Act. When unforeseen problems have arisen, they have been recognised
and resolved. The learning thus acquired has added to the knowledge base
of the JPO and is applied constantly in the complex and challenging exercise
of anonymisation.
Online publication of its judgments enables the Family Court to make them
widely accessible to the public and the profession, and logically extends the
transparency inherent in its operation as an ‘open’ court. At the same time,
parties can be reassured that the Court makes every effort to protect their
privacy. Many problems have been addressed since 2007 but, given the
constantly evolving online environment, I am sure there will always be
more!