Article

Treaty Interpretation and Constitutional

Transformation: Informal Change in
International Organizations
Julian Aratot

1. IN TR O DUCT ION ............................................................................................................................... 2 90

II. CONSTITUTIONS AND CONSTITUTIONAL CHANGE: AN APPROACH TO THE STUDY OF

INTERNATIONAL O RGANIZATIONS..............................................................................................296
A. T reaty vs. C onstitution....................................................................................................297
B. Formal vs. Material Constitution ............................ .......... 301
C. Juridical vs. Political Perspective ............................ ......... 303
D. A m endm ent vs. Transform ation ..................................................................................... 304

III. SUBSEQUENT PRACTICE AND THE PRIMACY OF CONSENT........................................................... 307

A. The Traditional Contours of Subsequent Practice.......................................................... 307
B. Subsequent Practice in the WTO: A Traditional Approach............................................ 311

IV. CONSTITUTIONAL INTERPRETATION IN THE UNITED NATIONS: "THE PRACTICE OF THE

ORGANIZATION" AS SUBSEQUENT PRACTICE IN THE ICJ ....................................................... 316
A. From "Subsequent Practice" to the "Practice of the Organization".................... 318
B. A Transformative Jurisprudence and Its Political Effects .............................................. 327

V. THE TRANSFORMATION OF THE EUROPEAN COURT OF HUMAN RIGHTS: "EUROPEAN

CONSENSUS" AS SUBSEQUENT PRACTICE................................................................................... 332
A. "European Consensus": The Weight of Majorities, Minorities, and Trends.................. 335
1. Establishing Consensus: The Subsequent Practice of (Most of) the Parties....... 337
2. Expansive Potential: Interpretation, Reinterpretation, and Modification...........343
B. "European Consensus" as a Transformation of Judicial Power...................................... 348

VI. CONCLUSION: LEGITIMATE AUTHORITY, ACCOUNTABILITY, AND THE AUTONOMY OF

INFORMA L TRANSFORM ATION .................................................................................................... 351

f Associate-in-Law, Columbia Law School. I am grateful to Professors Joseph Weiler,

Georg Nolte, and Jos6 Alvarez for their invaluable comments on drafts of this article. Further thanks are
due to Professors Seyla Benhabib, Lorenzo Casini, Grainne de Birca, Ryan Goodman, Robert Howse,
Mattias Kumm, Jonathan Nash, Stefan Oeter, Friedl Weiss and Neil Walker for their comments and
advice at various stages in the preparation of this piece. I am also very grateful to the participants in the
2012 NYU International Law & Human Rights Scholarship Conference for their questions and
comments, as well as the participants in the 2012 ASIL-ESIL-Rechtskulturen Workshop on
Transatlantic Debates in International Legal Theory, and the 2012 ASIL Research Forum. Thanks as
well to Davinia Abdul-Aziz, J. Benton Heath, Arie Rosen, and Guy Sinclair for numerous insightful
discussions related to this article, and finally to Will Smiley, Jacob Victor, and the Yale Journal of
International Law editorial team for their many thoughtful comments. All errors and omissions are of
course my own.
290 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

By constitutional amendment, I mean change in the text of the constitution

through a purposeful act of will; by constitutional transformation, I mean
change that allows the text to remain formally unchanged and is caused by
facts that need not be accompanied by an intention or awareness of the change.
I need hardly mention that the theory of transformation is much more
interesting than that of change.
-Georg Jellinek'

I. INTRODUCTION

Like all constituted bodies of government, international organizations

change over time-sometimes in profound and unexpected ways.2 Besides
developing through the obvious mechanism of formal amendment by the
constituent member states, these governance bodies can and do undergo a more
autonomous kind of constitutional development-what might be called
informal constitutional change or transformation. This type of quiet evolution
may occur on different levels-including the reordering of the organization's
internal architecture in terms of the relative competences of its various organs,
as well as the development of the powers of the organization as a whole vis-a-
vis the states parties. Unlike formal amendment, which occurs through the
express decision of the member states according to a certain procedure,
informal transformation occurs more subtly, through the practice of the
organization. Although the latter mode of change may attract less attention than
the former, the degree of change involved can be just as dramatic.
This Article is about one particular mode of informal change: the
transformation of an organization through the practice of its judicial organ in
the interpretation of its constituent instrument-in other words, transformation
through constitutional interpretation. Not all international organizations include
judicial organs, and not all judicial organs are constituted in the same way.4
Different institutions have been delegated different powers and different kinds
of jurisdiction. Their judges are chosen through widely varying procedures,

1. GEORG JELLINEK, CONSTITUTIONAL AMENDMENT AND CONSTITUTIONAL TRANSFORMATION

(1906), as reprinted in WEIMAR: A JURISPRUDENCE OF CRISIs 54, 54 (Arthur J. Jacobson & Bernhard
Schlink eds., Belinda Cooper et al., trans., Univ. of Cal. Press 2000) (footnote omitted).
2. See, e.g., Jost ALVAREZ, INTERNATIONAL ORGANIZATIONS AS LAW-MAKERS (2005). The
scope of this Article is limited to the analysis of international organizations established by treaty, and
charged with some degree of delegated public power.
3. Id at 415-516; JAN KLABBERS, AN INTRODUCTION TO INTERNATIONAL INSTITUTIONAL
LAW 160-61, 230-49 (2d ed. 2009).
4. Benedict Kingsbury, International Courts: Uneven Judicialization in Global Order I
(New York Univ. Sch. of Law, Working Paper No. 259, 2011), http://lsr.nellco.org/nyuplltwp/259; see
also Cesare Romano, The Proliferationof InternationalJudicialBodies: The Pieces of the Puzzle, 31
N.Y.U. J. INT'L L. & POL. 709, 711-23 (1999) (mapping the diverse field of international judicial and
quasi-judicial bodies).
5. Karen J. Alter, Delegating to International Courts: Self-Binding vs. Other Binding
Delegation, 71 L. & CONTEMP. PROBS. 37 (2008); Eyal Benvenisti & George W. Downs, Prospectsfor
the IncreasedIndependence of InternationalTribunals, 12 GER. L.J. 1057, I057-58 (2011).
2013] Treaty Interpretationand ConstitutionalTransformation 291

and entertain very different interpretive outlooks.6 Yet in the abstract the basic
fact remains clear: where they are established, such international courts and
tribunals tend to contribute substantially to the development of their larger
organizations. 7 Through the interpretation of the formal terms of their
constituent instruments, these constituted judicial bodies have proven capable
of transforming the material constitutions of the organizations to which they
belong.
The focus here will be on international organizations established by
treaty, charged with the exercise of certain elements of governmental authority
over the states parties-and specifically those with judicial organs. My goal is
to expose the transformative potential and effect of certain of these bodies'
interpretive practices when engaged in the interpretation of their own
constituent instruments. This Article is thus as much about treaty interpretation
as it is about constitutional transformation. Under the general law of treaties, all
interpretation is supposed to proceed according to a set of positive legal rules,
codified in general international law and external to any particular
organization. The rules of interpretation provide a legal framework for the
interpretation of all treaties, including the constituent instruments of
international organizations-as such the rules would appear to constrain
judicial bodies' interpretive discretion. Yet in the hands of certain judicial
organs engaged in constitutional interpretation, these codified interpretive
techniques have sometimes proven more empowering than limiting. In
confronting their own constituent instruments, the judicial organs of
international organizations have sometimes adopted remarkably broad
approaches to interpretation-even in the name of applying more modest
interpretive canons. I want to suggest that the use of broad interpretive methods
can materially contribute to the constitutional transformation of an international
organization: not only in terms of its internal architecture but, even more
fundamentally, in terms of its external autonomy and capacities vis-a-vis the
states parties.
Despite its justification in the voluntaristic language of the law of treaties,
I argue that such judicially driven transformation can fray the threads of state
consent binding the organization to its constituent member states. The same
informal process that contributes to the autonomy of an organization may at the
same time undermine the consensual grounds of its authority. Constitutional
interpretation in international organizations thus presents a familiar paradox: on
the one hand, in light of the important role of these organizations in
supranational governance, there is a perceived need for them to employ a
flexible approach to their functions in a changing legal and political

6. J.H.H. Weiler, The Interpretationof Treaties-A Reexamination: Preface,21 EUR. J. INT'L

L. 507, 507 (2010); Kingsbury, supranote 4, at 1-2.
7. ALVAREZ, supra note 2, at 65-109; Benvenisti & Downs, supra note 5, at 2-3; Ingo
Venzke, The Role of InternationalCourts as Interpreters and Developers of the Law: Working Out the
JurisgenerativePracticeofInterpretation,34 LOYOLA L.A. INT'L & COMP. L. REV. 99 (2011).
8. See Vienna Convention on the Law of Treaties art. 31, May 23, 1969, 1155 U.N.T.S. 331
[hereinafter VCLTJ.
292 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

environment. On the other hand, in light of the very same importance and
sensitivity of the areas regulated by these treaties, there is a countervailing
pressure to stick to the bargain struck. The evolution and adaptation of treaty-
based organizations may be desirable under the right conditions. Their
amendment procedures are often difficult to engage, making it difficult for
these organizations to respond appropriately to manifest changes in
international law and politics. Informal change may indeed yield desirable
results-any particular such development will have to be judged in its own
right, case-by-case. But the specter of consent always lurks in the background:
irrespective of its outcome, informal transformation carries in its wake the
potential to produce serious problems of legitimacy and accountability.
This Article is both conceptual and comparative in approach. I will
examine the interpretive practices of three very different international
organizations: the World Trade Organization (WTO) via the Appellate Body
(WTO-AB); the United Nations via its principal judicial organ, the
International Court of Justice (ICJ); and the Council of Europe (CoE) with a
special focus on the European Court of Human Rights (ECtHR). Two types of
questions will guide this analysis. On the one hand, much attention will be
given to doctrinal exegesis: How do the judicial organs of these organizations
approach the rules of interpretation when engaged in the interpretation of their
own constituent instruments? How do their approaches contrast with the more
general interpretation of treaties by neutral and independent third-party
international courts and tribunals? How do the contours of their interpretive
techniques compare to one another? On the other hand, the goal shall always be
to qualify the constitutional meaning of these organization's interpretive
practices: What, if any, are the constitutional effects of said judicial bodies'
approaches to their own constituent instruments? What effect does
interpretation have on the ordering of powers and competences within the
organization? What effect does it have on the powers of the organization as a
whole, or on its relationship to the parties? And how can their interpretive
practices be qualified in terms of state consent-the foundational principle of
the law of treaties, and the original basis of their authority?
To better compare the interpretive practices of such varied organizations,
I confine the analysis here to the use, by their judicial organs, of one particular
technique of interpretation: the interpretation of a treaty in light of the
subsequent practice of the parties ("subsequent practice").9 As codified at
Article 31(3)(b) of the Vienna Convention on the Law of Treaties (VCLT), all
treaty interpretation shall take into account "[a]ny subsequent practice in the
application of the treaty which establishes the agreement of the parties
regarding its interpretation." 10 This basic rule will reemerge throughout this
Article as a leitmotif In other words, I employ subsequent practice as an

9. This venerable technique of interpretation has long been recognized in international law.
Id. art. 31(3)(b). However, its use significantly predates the Vienna Convention. See, e.g., Russian Claim
for Interest on Indemnities (Russ. v. Turk.), II R.I.A.A. 421, 433 (Perm. Ct. Arb. 1912); Air Transport
Services Agreement Arbitration (U.S. v. Fr.), 38 I.L.R. 182, 248-55 (1963).
10. VCLT, supra note 8, art. 31(3)(b).
2013] Treaty Interpretationand ConstitutionalTransformation 293

analytical device, to illuminate patterns of change within the WTO, the United
Nations, and the CoE.
The doctrine of subsequent practice is a venerable technique of
interpretation oriented above all toward state consent. The traditional idea of
the technique is that the interpretation of any treaty should take into account the
practice of the parties in applying the agreement. Such practice, properly
defined, can be a valuable guide to the meaning of the treaty in the eyes of the
parties." Because the goal is establishing the extent of the parties' mutual
consent to be bound by an agreement, an authentic practice must entail the
consistent practice of all of the parties. Although active participation by all is
most probative, it is generally recognized that the mere acquiescence of some
will suffice.12 Where it is properly applicable, however, the doctrine has a
highly expansive potential. Insofar as it evidences their common intentions, the
parties' consistent and concordant practice may establish a far-reaching
interpretation, and even successive reinterpretations.' 3 Though rooted in the
consent of the parties, the doctrine can potentially establish significant treaty
change over time.
From the perspective of the law of treaties, scholars often acknowledge
that subsequent practice works somewhat differently in international
organizations where judicial organs are charged with the interpretation of the
organization's charter on a standing (rather than an ad hoc) basis.14 In
interpreting their own constituent instruments, these courts and tribunals
occasionally resolve questions about the requirements and effects of subsequent
practice idiosyncratically. 5 Some have taken radically broad views of
subsequent practice, both in terms of evidence (i.e., whose practice is relevant
and with what degree of consistency) and expansive potential (i.e., how much
change the doctrine can support). These idiosyncrasies tend to be addressed as
a problem relating to the much-vaunted consistency of interpretive doctrine in
the law of treaties.16 In general, the issue tends to be bracketed as the province

11. Russian Indemnities, II R.I.A.A. at 433.

12. Rec. of the Int'l Law Comm'n, 18th Sess., May 4-July 19, 1966, U.N. Doc. A/6309/Rev.1,
reprinted in, 2 Y.B. INT'L L. COMM'N, 187, 222 [hereinafter ILC DALT]; see also MARK E. VILLIGER,
COMMENTARY ON THE 1969 VIENNA CONVENTION ON THE LAW OF TREATIES 431 (2009) (confirming
that this provision requires the active practice of only some of the parties).
13. See ANTHONY AUST, MODERN TREATY LAW AND PRACTICE 191-95 (2000).
14. See RICHARD GARDINER, TREATY INTERPRETATION 11-12 (2008); TETSUO SATO, THE
EVOLVING CONSTITUTIONS OF INTERNATIONAL ORGANIZATIONS: A CRITICAL ANALYSIS OF THE
INTERPRETIVE FRAMEWORKS OF THE CONSTITUENT INSTRUMENTS OF INTERNATIONAL ORGANIZATIONS
xiii (1996); Salo Engel, Living Constitutions and the World Court (The Subsequent Practice of
InternationalOrganizations Under Their Constituent Instruments), 16 INT'L & COMP. L.Q. 865 (1967).
It is worth bearing in mind that some judicial organs enjoy substantially more opportunities to interpret
their constituent instruments than do others. The ICJ only hears a handful of cases a year (and even more
rarely enjoys the opportunity to interpret the U.N. Charter or the Statute of the Court), a far cry from the
overwhelming influx of cases facing the ECtHR (over 50,000 a year, with a backlog of nearly 150,000).
15. See C.F. AMERASINGHE, PRINCIPLES OF THE INSTITUTIONAL LAW OF INTERNATIONAL
ORGANIZATIONS 24 (2d rev. ed. 2005); GARDINER, supra note 14, at 247..
16. According to the ILC, in its commentaries to the DALT, the various elements of Article 31
constitute a single rule of interpretation. ILC DALT, supra note 12, at 219-20 ("All the various
elements, as they were present in any given case, would be thrown into the crucible, and their interaction
294 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

of special regimes,' 7 or addressed case-by-case by appeal to other features of

the particular treaty in question.
From the perspective of constitutional theory, however, the liberality of
approaches to subsequent practice in certain international organizations takes
on a very different color. What matters at this altitude is that these judicial
organs expand the contours of the doctrine when interpreting their own
constituent treaties. Some organizations, like the WTO, have taken a relatively
static approach to the technique. But others have felt less restrained. In the
cases of the United Nations and CoE, the expansive use of subsequent practice
in the context of constitutional interpretation has had important transformative
effects.
In the comparative portion of the Article I trace the use of subsequent
practice by the WTO-AB, the ICJ, and the ECtHR. The first body represents a
control, as an exemplar of a strict and traditionalistic approach to subsequent
practice. The WTO-AB has consistently maintained a rigidly narrow approach
to the technique in the interpretation of the complex web of agreements
constituting the WTO regime. 19 The second and third have adopted very
particular approaches to subsequent practice in the interpretation of their
constituent instruments-quite different in their contours, but similar in their
remarkable liberality.20 In broad strokes I argue that in contrast to the WTO-
AB, the expansive approaches of the ICJ and ECtHR have materially
transformed the constitutions of their respective organizations. The

would give the legally relevant interpretation."); see also GARDINER, supra note 14, at 9 (invoking the
ILC's "crucible" approach).
17. See, e.g., GARDINER, supra note 14, at 247.
18. Some explain certain differences in approach by appeal to the substantive law of the
treaties or treaty provisions being interpreted. For example, it is sometimes said that treaty norms
protecting the human person form a special category. See THEODOR MERON, THE HUMANIZATION OF
INTERNATIONAL LAW 193-201 (2006). The report of the International Law Commission (ILC) on
fragmentation notes that, "[i]n State practice, and the practice of international tribunals, particular
approaches to interpretation have of course developed. Thus it has become a practice of human rights
bodies to adopt readings of human rights conventions that look for their effet utile to an extent perhaps
wider than regular treaties." Rep. of the Study Group of the Int'l Law Comm'n, Fragmentation of
International Law: Difficulties Arising from the Diversification and Expansion of International Law,
58th Sess., May I-June 9, July 3, Aug. 11, 2006, 1j428, U.N. Doc. A/CN.4/L.682 (Apr. 13, 2006)
[hereinafter Fragmentation Report]. More recently, one hears that treaties protecting the environment
have a special place. See id. T 493(1). Some commentators have argued that these substantive norms are
different (and subject to differential treatment in interpretation) because they represent interests of the
international community as opposed to the mere bilateral interests of the parties. See Bruno Simma,
From Bilateralism to Community Interest in International Law, 250 RECUEIL DES COURS 217 (1994);
Santiago Villalpando, The Legal Dimension of the International Community: How Community Interests
Are Protected in International Law, 21 EUR. J. INT'L L. 387 (2010). Others suggest that inter-state
treaties that create rights or benefits for third party non-state actors should be treated differently for
purpose of interpretation, by qualifying the weight given to the evolving intentions of the parties. See,
e.g., Anthea Roberts, Power and Persuasion in Investment Treaty Interpretation: The Dual Role of
States, 104 AM. J. INT'L L. 179, 202 (2010) ("Treaties that grant rights to nonstate actors, such as human
rights and investment treaties, do not share the symmetry between those who hold the rights and those
who can interpret them . . . .The treaty parties may still be the masters of the treaty, but one cannot
assume no harm, no foul in accepting their interpretations of nonstate actors' rights.").
19. See Alexander Feldman, Note, Evolving Treaty Obligations: A Proposal for Analyzing
Subsequent Practice Derivedfrom WTO Dispute Settlement, 41 N.Y.U. J. INT'L L. & POL. 655, 676
(2009).
20. See AMERASINGHE, supra note 15, at 49-56; GARDINER, supra note 14, at 244-49.
2013] Treaty Interpretationand ConstitutionalTransformation 295

transformation of these organizations through the interpretive practices of their

judicial bodies has had dramatic effects, enhancing these constituted bodies'
general powers and rendering them significantly more autonomous vis-A-vis the
states that created them.
The ICJ and ECtHR deploy subsequent practice in different ways, and
with different kinds of effects. The ICJ has held that the practice of the organs
(and other bodies) of the U.N. Organization provides an authoritative guide to
the interpretation of the U.N. Charter-as a proxy for the subsequent practice
of the parties.21 However, the Court lacks the jurisdictional competence to
22
weigh in on the interpretive value of the organs' practice most of the time. As
a result, the power of its jurisprudence lies in its role as a signal to the more
competent organs that they may establish an authentic interpretation of the
Charter through their consistent practice. In particular, the Court's approach
empowers the Security Council (UNSC) and the General Assembly (UNGA).
At the same time, it also empowers some states parties against others-
especially the five permanent members of the Security Council (P5). By
contrast, the ECtHR's approach to interpreting its constituent instrument, the
European Convention on Human Rights (ECHR),23 is more directly court-
empowering. That Court does look to the actual practice of the states parties to
establish authentic subsequent practice. However, it has proven willing to rely
on the practice of merely most states under its jurisdiction, even in the face of
blatant contradictory conduct by the others. What is essential, in the view of the
ECtHR, is that the practice of the parties is sufficient to establish a "European
consensus." It remains within the province of the Court to determine just how
much practice will suffice in particular cases.24 Both the ICJ and the ECtHR
adopt very different approaches to subsequent practice, with different
constitutional consequences. What the practices of these Courts have in
common is their broader transformative effect of empowering each
organization as a whole, over and above the states parties.
The following Part lays the conceptual groundwork for the rest of the
Article. I seek to both elaborate a narrow concept of the "constitution" of an
international organization, and develop a theoretical apparatus for qualifying
constitutional change over time. In Part III, I examine traditional approaches to
interpretation on the basis of subsequent practice. I first develop the contours of
subsequent practice as it is typically applied outside the context of international
organizations, to indicate the strong consensual basis of the technique. Second,
I examine the use of subsequent practice by the WTO-AB, as an exemplar of
the traditional approach, even in the context of constitutional interpretation. In

21. Certain Expenses of the United Nations, Advisory Opinion, 1962 I.C.J. 151, 157 (July 20);
Legal Consequences for States of the Continued Presence of South Africa in Namibia (South West
Africa) Notwithstanding Security Council Resolution 276 (1970), Advisory Opinion, 1971 I.C.J. 16, T
21-22 (June 21); Legal Consequences of the Construction of a Wall in the Occupied Palestinian
Territory, Advisory Opinion, 2004 I.C.J. 136, T 27-28 (July 9).
22. See infra Part IV.
23. Convention for the Protection of Human Rights and Fundamental Freedoms (Rome), Nov.
4, 1950, 213 U.N.T.S. 221 [hereinafter ECHR].
24. See, e.g., Ocalan v. Turkey [GC], App. No. 46221/99, Eur. Ct. H.R. (May 12, 2005).
 296 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

Part IV, I examine the ICJ's evolving approach to interpretation on the basis of
subsequent practice qua the "practice of the organization." 25 I attempt to draw
out the transformative effects of its approach from a constitutional perspective,
as well as expound its prospective transformative potential. In Part V, I turn to
the ECtHR, with a focus on its use of "European consensus" as a stand-in for
subsequent practice. Here too, though in a different way, the Court's expansive
assertion of competence to interpret the Convention on the basis of practice has
had significant constitutional effects. Finally, in Part VI, I sketch a foundation
for the informed critique of informal change in international organizations. I
suggest that informal change is not necessarily a bad thing, and it is important
to proceed on a case-by-case basis. Nevertheless, I emphasize certain general
problems of legitimacy and accountability that are always likely to follow in
the wake of transformation through constitutional interpretation, especially in
the context of international organizations.

II. CONSTITUTIONS AND CONSTITUTIONAL CHANGE: AN APPROACH TO THE

STUDY OF INTERNATIONAL ORGANIZATIONS

The concept of a constitution is employed in innumerable ways. 26 If not

adequately cabined, its deployment in the context of international organizations
can cause conceptual mischief.27 In this section, I first attempt to avoid a
misleading dichotomy between the concepts of an international treaty and a
constitution. I then develop the idea of constitutions and constitutional change
in the context of international organizations by drawing three more meaningful
distinctions: between the formal and material constitution of an international
organization; between two methodological perspectives on the constitution

25. See generally Certain Expenses, 1962 I.C.J. 151; Namibia, 1971 I.C.J. 16; Wall, 2004
I.C.J. 136.
26. At the outset I want to disclaim an especially broad use of the concept of a constitution on
the global stage: i.e., the claim that the constituent instrument of a public international organization
amounts to, or represents part of, some kind of global constitution. The idea occasionally surfaces that
the U.N. Charter represents the constitution of the world. See Bardo Fassbender, The United Nations
Charter as Constitution of the International Community, 36 COLUM. J. TRANSNAT'L L. 36 (1998);
Thomas Franck, Is the UN Charter a Constitution?, in VERHANDELN FOR DEN FRIEDEN 95 (Jochen
Frowein et al. eds., 2003). Others highlight the Charter as the best candidate for developing such a
constitution in the future. See, e.g., Jirgen Habermas, The Constitutionalization of International Law
and the Legitimation Problems of a Constitution for World Society, 15 CONSTELLATIONS 444 (2008).
Another line of argument suggests that the Charter, along with other major universal agreements,
constitute fragmented islands in a global constitutional archipelago. See JAN KLABBERS, ANNE PETERS
& GEIR ULFSTEIN, THE CONSTITUTIONALIZATION OF INTERNATIONAL LAW (2009). It has been
suggested, in this vein, that the World Trade Agreements (including the Global Agreement on Tariffs
and Trade) play the part of a world trade constitution. J.H. Jackson, The WTO "Constitution" and
Proposed Reforms: Seven "Mantras" Revisited, 4 J. INT'L ECON. L. 67, 67-78 (2001); Joel Trachtman,
The Constitutions of the WTO, 17 EuR. J. INT'L L. 623 (2006). Unsurprisingly, these views have their
detractors. However I want to side-step the issue of the global constitution altogether, foregoing any
position on the relationship between these organizations and any constitution of the world. The focus
here is not the global constitution or lack thereof. Admittedly, organizations like the United Nations,
WTO, and ECtHR are interesting largely because they are charged with the exercise of public power on
the international stage. But the point of talking about their constitutions is to illuminate something about
each organization's internal composition as a discrete entity, and their power over their members.
27. Neil Walker, Postnational Constitutionalism and the Problem of Translation, in
EUROPEAN CONSTITUTIONALISM BEYOND THE STATE 27 (Joseph Weiler & Marlene Wind eds., 2003).
2013] Treaty Interpretationand ConstitutionalTransformation 297

(juridical and political); and between constitutional amendment (formal

change) and constitutional transformation (informal change).

A. Treaty vs. Constitution

Certain scholars insist on drawing particularly stark lines between

"constitutions" and mere "treaties." This proclivity tends to arise in the
literature on federalism and, especially, in the context of the European Union-
often by reference to the American experience with federation in the late
28
eighteenth century. Certain voices distinguish between treaties and
constitutions in order to stratify phases of federal integration. 29 Treaty and
constitution here become analogues for two ideal types: a confederation,
meaning a loose political organization of sovereign states (e.g., the United
States as envisioned by the Articles of Confederation of 1781)30 ; and a veritable
federal state, meaning a single state composed of a combination of sub-units
(e.g., the United States as established by the U.S. Constitution of 1789).31 In
this milieu, constitution comes to be inexorably associated with the notion of a
sovereign federal state, while treaty is associated with the weaker confederation

28. The issue is particularly sharp in the debates surrounding the proposed "Treaty
Establishing a Constitution for Europe," Oct. 29, 2004, 2004 O.J. (C 310) 47, available at http://eur-lex
.europa.euILexUriServ/LexUriServ.do?uri=OJ:C:2004:310:FULL:EN:PDF. Even in spite of the label
"Constitution," certain eminent scholars insisted that the new treaty cannot be understood as a true
constitution. See, e.g., Dieter Grimm, Integration by Constitution, 3 INT'L J. CONST. L. 193, 193-98
(2005); see also DEBORAH Z. CASS, THE CONSTITUTIONALIZATION OF THE WORLD TRADE
ORGANIZATION: LEGITIMACY, DEMOCRACY, AND COMMUNITY IN THE INTERNATIONAL TRADING
SYSTEM 3 (2005); Jeffrey Dunoff, Constitutional Conceits: The WTO's 'Constitution' and the
Discipline of International Law, 17 EUR. J. INT'L L. 647, 672 (arguing "that there is no WTO
constitution, and that scholarly invocation of constitutional discourse represents an effort to invest
international legal structures with the power and authority that domestic constitutional entities possess").
See generally SYMPOSIUM ON THE PROPOSED EUROPEAN CONSTITUTION, 3 INT'L J. CONST. L. 163, 163-
515 (2005). See even the hesitation of a constitutionalist like Miguel Poiares Maduro, in writing about
the proposed "Treaty for a Constitution of Europe" in 2005. Miguel Poiares Maduro, The Importance of
Being Called a Constitution:ConstitutionalAuthority and the Authority of Constitutionalism,3 INT'L J.
CONST. L. 332, 352 (insisting that "the choice of the legal form of a treaty, and the subjection of the text,
for the most part, to the traditional mechanisms of treaty ratification make clear that we are not
confronting a constitution in the classical sense").
29. Grimm, supra note 28, at 193-94, 208; Michel Rosenfeld, The European Treaty-
Constitution and Constitutional Identity: A View from America, 3 INT'L J. CONST. L. 317, 317-18
(2005). But see J.H.H. Weiler, On the Power of the Word: Europe's ConstitutionalIconography, 3 INT'L
J. CONST. L. 173, 174 (2005) (referring to the proposed EU Constitution as a "treaty masquerading as a
constitution," in the interest of impugning what he takes to be an empty gesture at formal
constitutionalism). Weiler does, however, insist that the European Union is a constitutional entity in
spite of the "constitutional" paucity of its formal instruments; materially speaking, "in its quotidian
existence Europe is constitutional and accepted as such." Id. at 183. This "run-of-the-mill treaty
amendment" only more or less modifies an already constituted EU polity, established at first by earlier
treaties but only truly constitutionalized by the praxis of the European Court of Justice in tandem with
the national courts of the Member States. See id. at 173; J.H.H. WEILER, The Transformation ofEurope,
in THE CONSTITUTION OF EUROPE, 10 (1999); see also infra Section II.B (discussing the distinction
between a formal and a material constitution).
30. See ARTICLES OF CONFEDERATION of 1781.
31. See U.S. CONST.; see generally HANS KELSEN, GENERAL THEORY OF LAW AND STATE
316-19 (1945) (classically distinguishing between a fundamentally unitary (but decentralized) federal
state, and a confederation-which Kelsen calls a "purely international union of states").
298 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

form.32 In a similar vein, some of these scholars add an ideational element to

the concept of a constitution, as opposed to a mere treaty, such as the presence
of an authentic demos.33 Irrespective of the level of integration achieved by
agreement between states, by this view, a compact can only be understood as
being a true constitution when it constitutes (and is constituted by) a political
community of individuals who share in a single constitutional identity. 34 In
broad strokes, the name "constitution" is reserved for only those treaties or
instruments that establish a unitary sovereign power, reflecting the identity of a
constituent (and in turn reflexively constituted) demos; anything less remains a
mere treaty-based organization among sovereigns. To many in the world of
scholarship on federalism, and especially European integration, the conflation
of the concepts of treaty and constitution amounts to a cardinal analytical sin.36
Most international organizations, however, make no pretensions to
federalism, nor do they lay claim to sovereignty as a telos. And it would be
difficult indeed to argue that many (if any) reflect a true demos with a unitary
constitutional identity.37 Organizations like the WTO and ECtHR do not fall
neatly on an axis of federal-confederal integration. Even the United Nations
cannot be easily conceived in such terms, despite the very forward-looking
aspirations of certain eminent scholars. It too is ultimately a functionally
delimited organization to which states have delegated significant powers in

32. Grimm, supra note 28, at 208. But see JEAN L. COHEN, GLOBALIZATION AND
SOVEREIGNTY, 80-158 (2012) (problematizing the hard distinction between federal state and
confederation as ideal types, especially in the context of understanding the European Union and the
United Nations).
33. Grimm, supra note 28, at 193; Rosenfeld, supra note 29, at 317.
34. See Grimm, supra note 28, at 198 ("[I]t is not the [proposed European Union]
constitution's legal functions that are of central importance-these could be fulfilled by treaties. Rather,
the constitution is viewed as crucial because of its anticipated collective and emotive benefits.").
35. See id. at 208 ("[T]he desire for a transition from treaty to constitution ... still lacks a
solid basis in reality. At present, there is little reason to expect that the [Treaty for a Constitution for
Europe] will make up for the European Union's legitimacy deficit, anchor the Union in the hearts of
Europe's citizens, and, in so doing, play an integrative and identificatory role.").
36. In Europe, the "constitutional" status of the European treaties is hotly contested even in the
courts, with a palpable linguistic divergence in the jurisprudence of the Court of Justice of the European
Union (CJEU) and the supreme courts of the Member States. See Joined Cases C-402/05 P & C-415/05
P, Kadi v. Council, 2008 E.C.R. 1-06351, S 285 ("[T]he obligations imposed by an international
agreement [UNSC Res. 1267] cannot have the effect of prejudicing the constitutional principles of the
EC Treaty, which include the principle that all community acts must respect fundamental rights, that
respect constituting a condition of their lawfulness which it is for the Court to review in the framework
of the complete system of legal remedies established by the Treaty."). But see the opinion of the German
Constitutional Court in its Decision on the Treaty of Lisbon, Bundesverfassungsgericht (BVerfG)
(Federal Constitutional Court), June 30, 2009, 2 be 2/08, TT 233, 235, http://www.bverfg.de
/entscheidungen/es20090630_2bve000208en.html, which insists that Europe is not at present a true
federal State, and stating moreover that the German Constitution would prohibit any degree of
integration going so far as to supplant "the constituent power of the Member States as the masters of the
Treaties." In other words, in the view of the Court, the constituent power of the peoples of Europe
remains within the national Member States; and at least in Germany it must remain so.
37. I bracket the European Union because it is too state-like, too integrated, and too much like
a federal constitution; for present purposes it would only muddy the analysis. There is a serious
argument that the European Union is a fully constituted system where the locus of sovereign power
remains relatively undecided. See COHEN, supra note 32, at 123, 136. It is further very much up for
discussion whether it satisfies the ideational requirement, proposed by some scholars, of a demos with a
true constitutional identity. See Grimm, supra note 28, at 208; Rosenfeld, supra note 29, at 317.
38. See Habermas,supranote 26, at 444.
2013] Treaty Interpretationand Constitutional Transformation .299

relatively specific spheres; it can boast of no high level of integration (let alone
federation) and cannot easily be said to reflect any global demos.39
Yet international organizations wield substantial power over their
creators; they are in essence creatures of governance. And after all, the notions
of constitution and constitutionalism are about the analysis, regulation, and
critique of public power. These organizations have been delegated substantial
governmental powers by the parties. Most importantly, in many cases (and to
varying degree) the states have ceded significant control over their affairs to the
organization by agreeing in advance to submit to the decisions of its organs
without their unanimous consent.40 These organizations enjoy substantial
autonomy from, and power over, their constituent states parties. There would
be something very artificial about forgoing the tools of analysis associated with
the concept of a constitution simply because these treaty-based bodies have no
pretensions to sovereignty, or because they rarely act upon-let alone
capture-the hearts and minds of individuals across the borders of their
member states.
While the treaty frame is indispensable to the study of international
organizations, constitutional theory can also offer valuable insight into the
structure of law and power within these supranational bodies. Three distinct but
related reasons stand out: first, the conceptual and analytical power of the
constitutional frame; second, its value as a platform for assessment and
critique; and third, its historical specificity.
The first and most immediate benefit of the constitutional frame is its
power for analyzing institutions of governance, and how they change over time.
The idea of a constitution presents a rich frame for analyzing the capacities of a
public governance organization, emphasizing both law and power. The typical
questions are: What powers does the organization possess? How are these
powers organized? How might they be changed? Thinking about an
organization in constitutional terms draws particular attention to the various
modes through which it exercises power through the promulgation, execution,
and interpretation of legal norms.4 1 It further directs attention to the related
questions of how powers are separated, checked, and balanced among the
constituted bodies, and whether power is divided in such a way as to ensure
accountability. And of course such institutional features are likely to be found
in some form in different organizations; the comparative value of the language
lies in the ubiquity of constitutional forms.

39. See Neil Walker, ConstitutionalPluralism in Global Context, in CONSTITUTIONALISM IN

EUROPE AND BEYOND (M. Avbelj & J. Kormek, eds., forthcoming 2013).
40. Within the U.N. system, amendments of the Charter, as well as decisions and resolutions
of the UNSC, ICJ, and UNGA, are taken not on the basis of unanimity but by majority (or super-
majority) vote. Similarly the decisions of the ECtHR are binding upon states parties as respondents,
even when they evidently disagree with the result. Finally, within the WTO system, the opinions
("Reports") of the Panels and WTO-AB comprising the Dispute Settlement Mechanism (DSM) become
binding on losing States, unless said States can convince the entire membership (including the victor in
the particular suit in question) to reject the report by consensus. As Kelsen rightly suggests, the
possibility of action by majority vote renders an international organization independent and non-
identifiable with its member states. See HANS KELSEN, PEACE THROUGH LAw 20-21, 43-44 (1944).
41. See infra Section II.B.
300 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

Secondly, the constitutional frame invites normative reflection and

evaluation. It is not only a powerful analytical and comparative tool, but also
connects to modem ideas about how public power should be organized. When
linked to the construction and organization of public power, the idea of a
constitution is not just about analysis of institutions but also inherently about
their evaluation. 42 The concept is inexorably tied to the ethos of
constitutionalism: any constitution creates powers, but only the good
constitution constrains them in appropriate ways. The devices of
constitutionalism-e.g., the separation of powers, checks and balances, and
judicial review-are just as valuable for the constraint of power on the
international stage as within the domestic state.43
Finally, the idea of a constitution draws strength from its deep historical
roots, reaching back to the dawn of political modernity." Since the eighteenth
century, the idea of a constitution has connoted an instrument that can create
public power, but also organize and constrain its use by law.45 Though nurtured
in the cradle of the modem state, there is nothing necessarily state-centric about
the ideal of constitutional government. Indeed the power of the constitutional
idea is manifest throughout the history of international organizations. 46 The
individual officials at the centers of these organizations tend to speak about
their institutions in constitutional terms, especially in emphasizing the
autonomy of these bodies from their creators. 47 Most tellingly, the language of

42. See Julian Arato, Constitutionality and Constitutionalism Beyond the State: Two
Perspectives on the Material Constitution of the United Nations, 10 INT'L J. CONST. L. 627 (2012)
(distinguishing between a purely juridical perspective on a material constitution, concerned with
analysis of the legal system of a constituted order, and a political-theoreticalperspective more oriented
toward the analysis of power; the first may be relativistic, but the second is inherently value-driven and
encourages evaluation).
43. See Mattias Kumm, The Legitimacy of InternationalLaw: A ConstitutionalistFramework
ofAnalysis, 15 EUR. J. INT'L L., 907, 929 n.55 (2004) (defending constitutionalism as an approach to the
assessment and design of national and transnational institutions in light of a wide range of features that
affect legitimation); Walker, supra note 27, at 42 (portraying constitutionalism as "an exercise in
practical reasoning" with the aim of working out a viable and legitimate framework for the organization
of political community, within the state and beyond).
44. THOMAS JEFFERSON, NOTES ON THE STATE OF VIRGINIA ch. 13 (1781); E.J. Sidyds, Dire
sur la question du veto royale (1789), in ECRITS POLITIQUES 236, 239 (Roberto Zapperi ed., 1985);
MARQUIS DE CONDORCET, PROJET DE CONSTITUTION FRANCAISE, tit. 3, 8, 9 (1793).
45. See THE FEDERALIST, NO. 51 (James Madison); THE FEDERALIST NO. 67 (Alexander
Hamilton); JEFFERSON, supra note 44, ch. 13; Sidyds, supra note 44; CONDORCET, supra note 44; A.V.
DICEY, THE LAW OF THE CONSTITUTION, 110 (8th ed. 1915) (naming the absence of institutional
discretion as an essential feature of constitutional government, i.e., in his view, the rule of law).
46. Already in 1784, Immanuel Kant argued that the most difficult problem of constitutional
government on the global scale is to constrain the governors by law and thus stop slippage from the rule
of law toward arbitrary personal rule. IMMANUEL KANT, IDEA FOR A UNIVERSAL HISTORY FROM A
COSMOPOLITAN POINT OF VIEW, thesis 5, 6, and 7 (Lewis White Beck trans., 1963).
47. See ALVAREZ, supra note 2; Jan Klabbers, ConstitutionalismLite, I INT'L ORG L. REV. 31,
32-33 (2004). For example, then Director-General Renato Ruggiero generated significant controversy by
referring to the WTO in constitutional terms in a 1998 speech. Renato Ruggiero, WTO Director General,
Address to the Royal Institute of International Affairs in London: The Multilateral Trading System at
Fifty (Jan. 16, 1998), http://www.wto.org/english/news e/sprr ellondon e.htm; see Trachtman, supra
note 26, at 628 (discussing the controversy arising out of Ruggiero's remarks). By contrast, and perhaps
with said controversy in mind, current Director-General Pascal Lamy studiously avoids the use of
explicit constitutional language, even when emphasizing the autonomy of the WTO regime. See, e.g.,
2013] Treaty Interpretationand ConstitutionalTransformation 301

constitutions and constitutionalism tends to arise in the opinions of their

judicial bodies and individual judges in justifying or advocating a dynamic
approach to the interpretation of their constituent instruments. The ECtHR has
gone so far as to justify its expansive approach to interpretation on the basis of
the special nature of the ECHR as a "constitutional instrument of European
public order." 48
There are thus good reasons to refer to the architecture of public
international organizations in constitutional terms.49 In essence the
constitutional analytic draws our attention to important facts about the structure
of power and authority in international organizations, while at the same time
providing a familiar language for assessment, comparison, and critique.
The concepts of treaty and constitution operate on different planes and
illuminate different things. We may provisionally define a treaty as a juridical
instrument, reflecting a binding agreement between states and apportioning
rights and obligations between them. As elaborated below, a constitution
reflects a much more complex juridical, political, and social reality-an
architecture that creates and regulates public power through law. But there is no
reason that an international treaty cannot provide the basis for a constitutional
system. This Article is concerned with treaties that constitute public
international organizations (constituted bodies), by formal agreement among
contracting sovereign states (constituent power). Such instruments should be
understood both as treaties and constitutions.
B. Formalvs. Material Constitution

A constitution is more than a paper document. It is a complex architecture

of norms, articulating the composition and competences of an organization
charged with the exercise of public power. Public international organizations
are generally constituted by formal treaty; but any such constituted body will
likely be much more than the dry parchment that created it. This is not to say
that the solemn instrument is not important. Rather the point is not to succumb

Pascal Lamy, The Place of the WTO and Its Law in the InternationalLegal Order, 17 EUR. J. INT'L L.
969, 970 (2007) (referring to the WTO as a "distinctive organization" comprising a "true legal order").
48. Loizidou v. Turkey (preliminary objections) [GC], App. No. 15318/89, Eur. Ct. H.R. 75
(Mar. 23, 1995); see also Voting Procedure on Questions Relating to Reports and Petitions Concerning
the Territory of South-West Africa, Advisory Opinion, 1955 I.C.J. 67, 106 (June 7) (separate opinion of
Judge Lauterpacht) ("A proper interpretation of a constitutional instrument must take into account not
only the formal letter of the original instrument, but also its operation in actual practice and in the light
of the revealed tendencies in the life of the Organization.").
49. The analytical framework of constitutions and constitutionalism is, of course, only one
among many-I do not suggest that it is the only useful lens for the study of international organizations.
There is much to be gained from alternative scholarly approaches to the phenomenon of global
institutions and governance, such as the "Global Administrative Law" project most closely associated
with the N.Y.U. School of Law. See Benedict Kingsbury, Nico Krisch & Richard B. Stewart, The
Emergence of Global Administrative Law, 68 LAW & CONTEMP. PROBS. 15, 17 (2005). The "Public
Law" project associated with the Max Planck Institute represents another fruitful school of analysis. See
THE ExERCISE OF PUBLIC AUTHORITY BY INTERNATIONAL INSTITUTIONS: ADVANCING INTERNATIONAL
INSTITUTIONAL LAW (Armin von Bogdandy et al. eds., 2010).
302 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

to textual formalism in assessing the constitutional order of an organization, if

the goal is to understand its real capabilities.50
The founding document may be called the formal constitution of an
organization.5 1 Necessarily in writing, it shapes and establishes the functions,
powers, and limits of an organization in the first cut. Further, it attempts to
entrench aspects of this structure and, in some cases, other important norms, 52
by making them particularly difficult to change (usually via an onerous
amendment rule).53 Yet a formal constitution is neither a necessary nor a
sufficient component of a constitution in the full sense of the term. Even in the
domestic context, states like the United Kingdom have no solemn document at
all, and yet clearly possess a normative structure that articulates how and by
whom laws shall be passed, interpreted, executed and enforced. If the goal of
constitutional analysis is to understand how a state or organization is
"constituted"-how a system provides for the creation, interpretation, and
application of legal norms, and how powers are delegated, divided and
delimited-it would seem farcical to ignore foundational norms simply because
they are not expressed in a solemn charter. Likewise, even where there is some
kind of founding document, no state's formal constitution really articulates the
full constitutional structure, especially over time. Some norms in the document
may fall into desuetude, while others are expanded by legislative, executive,
and judicial bodies to mean all sorts of things-often totally unanticipated by
the text and sometimes at cross-purposes with other aspects of the document.
Hans Kelsen thus classically distinguishes the material constitution from
the purely formal document. The material constitution, he explains, is that set
of norms that dictate the methods through which norms are created, interpreted
and applied at the highest level of the legal system.54 It may consist of a wide
array of laws and customs, some perhaps enshrined in a document, and others
developed through legislation, judgment, convention, or other practices of the
constituted organs of government.55 As opposed to the formal document, the

50. 1 BRUCE ACKERMAN, WE THE PEOPLE: FOUNDATIONS 35 (1999) ("If we are to do justice
to American realities, we must see that effective power is organized on very different lines, that it has a
very different genealogy from the one set out by our paper Constitution."); see also Weiler, supra note
29, at 174.
51. See KELSEN, supra note 31, at 124-25, 258 (articulating the notion of a formal
constitution); KELSEN, supranote 40, at 9, 127.
52. See, for example, the substantive rights of the ECHR, or the prohibition on the use of force
and the right to self-defense, codified in the Charter of the United Nations. U.N. Charter, arts. 2(4), 5.
53. HANS KELSEN, PURE THEORY OF LAW 222 (Max Knight trans., The Lawbook Exchange
rev. ed. 2002) (1967).
54. KELSEN, supra note 31, at 258.
55. For example, judicial review in the United States is mentioned nowhere in the formal
constitutional document-yet the rule asserted in Marbury that the Court has final say over the validity
of legislation under the constitution would certainly fall into the material constitutional structure of the
United States. See Marbury v. Madison, 5 U.S. (1 Cranch) 137 (1803) (asserting the Court's power of
judicial review and invalidation). Similarly, though passed as normal legislation by a vote of fifty
percent plus one, the Reform Act of 1832 transformed the constitution of the United Kingdom by
radically overhauling the electoral system to expand representative government. See Great Reform Act,
1832, 2 & 3 Will. 4, c. 45 (Eng.) (reforming popular representation in the House of Commons by
eliminating the "rotten boroughs" and significantly expanding the size of the electorate).
2013] Treaty Interpretationand ConstitutionalTransformation 303

material constitution describes the fundamental normative architecture within

which the constituted bodies function.
The material constitution comprehends the full constitution of any
organization. Nothing in it is non-constitutional, and nothing is missing from it
that is constitutional for that given entity. By contrast, the formal constitution is
both incomplete and may include some norms that would not normally be
considered constitutional from the material point of view. However, it bears
noting that the formal document may have material significance, insofar as it
entrenches particularly important norms and provides for authoritative
interpretation or even review. Moreover, and especially important for present
purposes, the interpretation of the formal constitution can have a profound
material effect.5 7 But it is crucial to bear in mind that any organization will
likely entail other un-entrenched norms of the highest constitutional
significance.

C. Juridicalvs. PoliticalPerspective

It is important not to be overly legalistic. For Kelsen, the material

constitution consisted entirely of that highest set of norms articulating how
valid legal norms are generated, applied, interpreted and reviewed throughout a
legal order. This archetypically "juridical" view is interested mainly in
understanding a constitution in terms of its internal hierarchy of norms
organizing how validity is transmitted throughout a legal order. As important as
this point of view is for understanding how a constitution operates, in isolation
it obscures much about the most important facets and effects of its operation.
Constitutions do not belong exclusively to the province of pure legal theory;
they are not just about law, but also about power. In establishing institutions,
rights, and obligations, constitutions create and regulate power, and as such
they also belong to the realm of politics. The study of constitutions should thus
be as multifaceted as its object; it should proceed from two perspectives at
once-the juridical point of view and the perspective of political theory.58
From a political point of view, it is not enough to ask how validity is
transmitted through a legal system. It is just as important to ask how power is
divided up within such a system. Who gets to create, apply, and interpret
norms? Are these powers separated among different bodies? If so, are they
checked? Balanced? How much discretion do these bodies enjoy? And how
much power does the organization wield over the governed? This political
dimension is the central concern of the modem normative ethos of

56. KELSEN, supra note 31, at 125. For example, though formally entrenched, the Eighteenth
Amendment to the U.S. Constitution (on prohibition) had minimal constitutional significance from a
material perspective.
57. Marbury, 5 U.S. (I Cranch) at 173; Kesavananda Bharati v. State of Kerala, A.I.R. 1973
S.C. 1461 (India) (asserting the judicial power to review and invalidate duly enacted constitutional
amendments that violate the "basic structure" of the Constitution).
58. See also Arato, supra note 42, at 638-43 (further elaborating the mutual complimentarity
of the juridical and the political points of view for analyzing, evaluating, and comparing constitutional
forms).
304 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

constitutionalism. It provides the central basis for the comparison and critique
of constitutional forms.
Thus a constitution should be understood as the normative architecture of
an organization-a structure of norms that may or may not entail a formal
constitutional instrument, with both juridical and political dimensions. From
one perspective the constitution represents the fundamental structure of an
organization as a legal order. But at the same time it articulates the structure of
power within that legal order, and provides for the valid exercise of power
through law.

D. Amendment vs. Transformation

To borrow Georg Jellinek's distinction, constitutional change can occur

through two different modes: amendment and transformation. 59 Amendment
should be understood to refer only to change of the formal constitutional
document occurring through formal procedure (e.g., Article V of the U.S.
Constitution or Articles 108 and 109 of the U.N. Charter). It is an inherently
intentional act, which brings about obvious changes in the constitution. Unlike
amendment of the formal constituent instrument, which depends on the willful
use of formal procedures, constitutional transformation can occur through the
less-than-intentional action of the constituted bodies. The issue is not the
intentions of the actors but the consequences of their actions. Intent aside, how
do the practices of constituted bodies contribute to material (if informal)
change in the constitutional order to which they belong?
Constitutional transformation can occur through the organs' exercise of
their delegated powers at all levels of the system (as opposed to only the
constituent level): through the enactment of important legislation, the actions of
the executive power, or interpretation by the courts.60 Moreover it may occur
through a variety of more subtle means, including longstanding usages,
customs, and conventions of the constitution.61 What is important, here, is not
the procedure but the effect of material change in the constitutional order.
Transformation can bring into effect change of all magnitudes, from
minor readjustments to sweeping systemic reorganization.62 The material effect

59. See Jellinek,supra note 1.

60. See id. at 55. For a more recent debate as to the possibility and propriety of informal
constitutional change in context of the United States, see Bruce Ackerman & David Golove, Is NAFTA
Constitutional?, 108 HARv. L. REv. 801 (1995) which examines the emergence of a particular
transformation of the power to bind the nation through intemational treaty under the U.S. Constitution,
through the sustained, interrelated, and occasionally competitive practices of all of three branches of
government. Cf Lawrence Tribe, Taking Text and Structure Seriously: Reflections on Free-Form
Method in ConstitutionalInterpretation,108 HARV. L. REV. 1225-26 (1995) (taking a skeptical view of
the propriety of informal change under the U.S. Constitution, and questioning the validity of the putative
changes to the treaty power identified by Ackerman and Golove).
61. This Article focuses on treaty interpretation as a mechanism for change. Full discussion of
the other informal mechanisms will be confined to another paper, but suffice it to note for the moment
that I mean to refer to constitutional development through non-legal means like conventions in the sense
employed in British constitutional theory since Dicey. See generally DICEY, supra note 45; IvoR
JENNINGs, THE LAW AND THE CONSTITUTION 79-135 (4th ed. 1952).
62. Jellinek, supra note 1, at 54-55; ACKERMAN, supra note 50, at 43.
2013] Treaty Interpretationand ConstitutionalTransformation 305

of a constitutional transformation can be just as profound as any change

achieved through formal amendment. The more difficult issue is to demarcate
what kinds of architectural changes the concept is meant to connote. As noted
above, formal amending procedures may bring into effect changes to the
constitution of both profound and miniscule import from a material
perspective-they are viewed as "constitutional" changes simply by virtue of
their formal incorporation. Constitutional transformation, by contrast,
necessarily refers to a change of some material significance. It will generally
connote two main kinds of change: the reordering of power among the organs
of the organization (e.g. procedural adjustments, or the restructuring of checks
and balances); or changes to the power of the organization as a whole, vis-i-vis
the constituent power.
The focus here will be on material transformation through judicial
interpretation-specifically a constituted court's interpretation of its formal
constitution. It may not be immediately obvious why a court, in interpreting the
formal constitution, contributes to a material transformation. Indeed, it may
seem that the court is not transforming anything, but simply expounding the
text of the formal charter. Here, however, the distinction between juridical and
political perspectives is particularly illuminating.6 From a juridical point of
view, even an expansive or evolutive interpretation might seem to entail no
constitutionalchange. Such an interpretation, by a duly authorized court, would
appear as simply expounding what is already there. But from a political-
theoretical perspective, the image may appear in a substantially different light.
Instead of focusing on the juridical question of the normative validity of the
court's interpretation, the political lens focuses on the effect of the court's
interpretation on the ordering of power throughout the organization to which it
belongs. Where an interpretation entails a new assertion of judicial power, the
reordering of the powers of the organs more generally, or the alteration of the
powers of the organization as a whole, it may be understood as transforming (or
contributing to the transformation of) the constitution of the organization in a
politically significant manner.65 Depending on perspective, an interpretation

63. Some prominent voices in the United States object to the possibility of significant informal
constitutional change in practice, especially in the presence of an express provision on amendment. See
Sanford Levinson, How Many Times Has the United States ConstitutionBeen Amended? (A) <26; (B)
26; (C) 27; (D) >27: Accounting for ConstitutionalChange, in RESPONDING TO IMPERFECTION 13, 18-
19 (Sanford Levinson ed., 1995) (attempting to draw lines between informal change, amendment, and
revision in terms of degree of change); Tribe, supra note 60 (arguing that extent of change matters-
small reinterpretations may be possible, but never the kind of transformation that would change the
"topology" of the text or structure of the Constitution). These objections should be understood in
context: these critics do not deny the general theoretical possibility that informal means may produce
fundamental constitutional change; rather they challenge the propriety of any such process in the
specific context of the U.S. Constitution. It should be understood that conceptually the process connotes
change of all sizes; of course it remains perfectly appropriate to challenge the legality or legitimacy of a
particular transformation on grounds of its magnitude in a particular constitutional legal system.
64. See Arato, supranote 42.
65. See, e.g., Marbury v. Madison, 5 U.S. (I Cranch) 137 (1803); Kesavananda Bharati v.
State of Kerala, A.I.R. 1973 S.C. 1461 (India). Even if these Courts attempted to ground their newly
asserted powers of review in the formal constitution, it would be difficult to deny that these assertions
amounted to monumental transformations of their respective constitutions in a material sense.
306 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

might thus appear as both mere elaboration (juridically) and dramatic

transformation (politically).66
Judicial interpretation can bring about constitutional transformation in
two ways. On the one hand, a given interpretation might bring about a direct,
"first-order" change in the material constitution of the interpreting tribunal.
Such would be the case where a particular interpretation directly alters the
powers of the organization as a whole, or reorders the competences of the
various organs in such a way as to substantially alter the division of powers
within the organization. In other words, the substance of a particular
interpretation may of itself entail the informal and material transformation of
the organization's constitution-for example, an interpretation of the scope of
presidential powers. On the other hand, a judicial body might significantly
transform its constitution in a "second-order" sense, by its approach to
interpretation. Irrespective of the particular substantive interpretations in
question, a court's method of interpretation might gradually bring about the
transformation of the organization to which it belongs.
Formal amendment is surely the most obvious (and uncontroversially
legitimate) mechanism through which the constitutions of international
organizations evolve over time.67 But their material constitutions can change
dramatically through less formal means as well. Of particular importance is the
transformation of their constitutions engendered by the interpretation of their
constituent instruments as internationaltreaties-i.e., the interpretation of an
organization's formal constitution by its constituted judicial body according to
the external law of treaty interpretation. And thus we come back to the
beginning: it is crucial to appreciate the constituent instrument of an
international organization as both a constitution and a treaty. On the one hand,
as in myriad domestic constitutional orders, the judicial organs of international
organizations play an important role in expounding and developing their
material constitutions through the interpretation of their formal constitutions.
On the other hand their interpretive practices do not proceed in a vacuum-they
are expected to and generally do proceed according to the rules of interpretation
under the external international law of treaties.68 The traditional rules of
interpretation exert a real pull. As we shall see, even when judicial organs like
the ICJ and ECtHR deviate from the normal approach to interpretation they
tend to justify their approaches in legalese, by appeal to Articles 31 and 32 of

66. In Jellinek's matchless phrase, "'Be bright and lively in expounding, if you can't expound,
then pound it in .. . ' these words of the poet are the highest maxim for constitutional transformation
through judicial interpretation." Jellinek, supra note 1, at 56 (invoking Goethe, "In Auslegen seid Frisch
und munter, legt ihr 's nicht aus, so leget was unter").
67. To this category we might add amendment and modification according to the default rules
of the VCLT, supra note 8, arts. 39-41, which provide for amendment in the absence of a formal
amendment rule, or under certain conditions the modification of a treaty between some of the parties
only. The rules on amendment and modification of treaties in the VCLT may be understood as
comprising a "default amendment rule" in international law, basically akin to typical constitutional
amendment (and analytically distinct from constitutional transformation).
68. VCLT, supranote 8, arts. 31-33.
2013] Treaty Interpretationand ConstitutionalTransformation 307

the VCLT. But deviate they do, and occasionally with dramatic transformative
effect.

III. SUBSEQUENT PRACTICE AND THE PRIMACY OF CONSENT

It is important not to go too fast. The peculiarities of subsequent practice

in the context of constitutional interpretation need to be thrown into relief, by
contrast to the longstanding and far more common use of the technique in the
interpretation of bilateral and multilateral conventions by third-party courts and
tribunals not engaged in the interpretation of their own constituent instruments
("ad hoc interpretation"). This section begins with a brief excursus elaborating
the contours of subsequent practice as it is traditionally and generally employed
in the ad hoc interpretation of international treaties. It concludes with an
analysis of the interpretive practice of the WTO-AB, as an example of an
international organization adopting a highly restrictive and voluntaristic
approach to subsequent practice, even in the interpretation of its own
constituent instrument.

A. The TraditionalContours of Subsequent Practice

At bottom subsequent practice is a technique of interpretation based on

state consent. The centrality of consent is evident in the modem formulation of
the technique, as codified at VCLT 31(3)(b), which provides that interpreters
should take into account "any subsequent practice in the application of the
treaty which establishes the agreement of the parties regarding its
interpretation. The doctrine envisions giving weight to subsequent practice
in two closely related and mutually consistent ways. On the one hand, the
technique provides a method for uncovering (or construing) the "original"
intentions of the parties. As stated by the Permanent Court of Arbitration in
1912, "the fulfillment of engagements between states, as between individuals,
is the surest commentary on the meaning of those engagements." 70 On the other
hand, under certain conditions, subsequent practice provides a mechanism for
treaty change and adaptation, by giving weight to the parties' evolving
intentions. State consent links these two modes, and renders them
harmonious. 7 The doctrine of subsequent practice thus seeks to resolve two

69. Id., art. 31(3)(b) (emphasis added).

70. Russian Claim for Interest on Indemnities (Russ. v. Turk.), 11 R.I.A.A. 421, 433 (Perm.
Ct. Arb. 1912) ("Considerant que l execution des engagements est, entre Etats comme entre
particuliers,le plus suir commentaire du sens de ces engagements.").
71. See Julian Arato, Subsequent Practiceand Evolutive Interpretation: Techniques of Treaty
Interpretationover Time and Their Diverse Consequences, 9 LAW & PRAC. INT'L CTS. & TRIBUNALS.,
443, 444-45 (2010); Gardiner, supra note 14, at 230-32. As evidenced by the current work plan of the
ILC, the criteria and outer contours of the technique may be in some degree of flux. See Rep. of the Int'l
Law Comm'n, 60th Sess., May 5-June 6, July 7-Aug. 8, 2008, U.N. Doc. A/63/10; GAOR, 63d Sess.,
Supp. No. 10 (2008), Annex A: Treaties over Time [hereinafter Annex A: Treaties over Time]. But what
remains consistent, outside of the context of international organizations, is an appreciation of the
technique's foundations in the common intentions of the parties as objectively evidenced in their
subsequent conduct. See Rahim Moloo, When Actions Speak Louder than Words: The Relevance of
Subsequent Party Conduct to Treaty Interpretation,30 BERKELEY J. INT'L L. (forthcoming 2013).
308 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

questions: What did the parties consent to in the first place? And have they
subsequently agreed to develop and extend the ambit of their common consent
such that the treaty should be viewed as having changed over time?
Whether employed in the search for the parties' original intent, or
whether the treaty should be understood as changing over time in light of the
parties' evolving intentions, subsequent practice is traditionally grounded on
the same basic premise: under international law, states are the masters of their
treaties.72 According to this perennial recitation, a treaty is nothing more than
an agreement between two or more sovereigns. 73 Acting together, the parties
remain free to do with their engagements what they will: whether
interpretation, reinterpretation, the progressive development of vague terms,
outright modification, or even letting the agreement wither away through
desuetude.74
In general, international courts and tribunals tend to be sensitive to the
traditional primacy of consent in relying on subsequent practice. But context is
important. Subsequent practice normally arises as an interpretive criterion in
the context of ad hoc interpretation. 75 In this more typical situation,
international courts and tribunals are charged with interpreting treaties on an ad
hoc basis-either where states call upon a standing tribunal to resolve a
particular dispute over the meaning of their treaty (as in many of the ICJ's
contentious cases), or where the tribunal itself is constituted on an ad hoc basis
to resolve a particular treaty dispute. Ad hoc interpretation may be opposed to
constitutional interpretation in international organizations, where a standing
judicial organ is charged with the interpretation of its own constituent
instrument. 76 At least in the context of ad hoc interpretation, judicial

72. See IAN SINCLAIR, THE VIENNA CONVENTION ON THE LAW OF TREATIES 98 (1984) (citing
LORD MCNAIR, THE LAW OF TREATIES 169 (1961)); VILLIGER, supra note 12, at 46-47 ("Parties are
States for which the treaty is in force ... and they are the masters of the treaty."). This notion of mastery
of the treaty may overly simplify a great many things. See more recently, and more subtly, Roberts,
supra note 18, at 202. But it remains a valuable heuristic for two reasons: first, it is historically
important as the traditional justification for reliance on subsequent practice and indeed for much of the
law of treaties as a whole; second, the expression consistently arises in scholarship and jurisprudence
even today, and thus appears to exert an influence somewhat disproportionate to its merit. Most
strikingly, the word has emerged in the jurisprudence of the German Constitutional Court in its opinion
on the Treaty of Lisbon. See German Constitutional Court in its Decision on the Treaty of Lisbon,
Bundesverfassungsgericht (BVerfG) (Federal Constitutional Court), June 30, 2009, 2 be 2/08, 1 235,
http://www.bverfg.de/entscheidungen/es2009O630_2bve000208en.html (holding that the German
Constitution would prohibit any degree of integration going so far as to transfer "the constituent power
of the Member States as the masters of the Treaties."). Still, the "traditional view" presented here
represents only an ideal type of a certain venerable and influential brand of voluntarism in the
interpretation of treaties. In fairness, not all traditionalists are so traditional.
73. See VILLIGER, supra note 12. Of course not everyone goes so far. As noted above, some
eminent voices believe that certain kinds of treaties should not be so easily linked to the changing will of
the parties. See MERON, supra note 18; SIMMA, supra note 18; and Roberts, supra note 18.
74. See KELSEN, supra note 31, at 119; Marcelo Kohen, Desuetude and Obsolescence of
Treaties, in THE LAW OF TREATIES BEYOND THE VIENNA CONVENTION 350, 352 (Enzo Cannizzaro ed.,
2011).
75. See, e.g., Kasikili/Sedudu Island (Botswana/Namib.), 1999 I.C.J. 1045, ' 74 (13 Dec.). See
also the more common ad hoc arbitral awards, for example Russian Indemnities, II R.I.A.A. at 433; Air
Transport Services Agreement Arbitration (U.S. v. Fr.), 38 I.L.R. 182, 248-55 (1963).
76. See AMERASINGHE, supra note 15, at 49-56; GARDINER, supra note 14, at 245-49.
2013] Treaty Interpretationand Constitutional Transformation 309

interpreters tend to rely on subsequent practice cautiously, in a manner highly

deferential to the consent of the parties. But as we shall see, the picture can be
very different in the context of constitutional interpretation.
In assessing the contours of subsequent practice as it is traditionally (and
typically) deployed, two issues tend to stand out. The first is a matter of
evidence: how can authoritative, subsequent practice be established? The
second question concerns the expansive potential of the doctrine: to what extent
can the subsequent practice of the parties support developing a treaty
provision? The relevance of state consent as a premise undergirding the
technique depends upon the answers to these questions. These two issues will
return as the critical tension points distinguishing the approaches to subsequent
practice within the WTO-AB, ICJ, and ECtHR.
Regarding the first (evidentiary) issue, subsequent practice is usually
considered authentic and relevant to interpretation only insofar as it
demonstrates the parties' common understanding of the treaty's meaning. As
the ILC explains in the Commentaries to its original draft of the VCLT,
subsequent practice requires that all of the parties to a treaty, not just some of
them, act in such a way as to evidence their agreement on the interpretation. 77
Certain preeminent voices have added that the parties' practice must not only
be common, but also concordant and consistent.78 The Commentaries allow that
this need not mean that all of the parties must actively engage in the practice-
it would be sufficient that any non-engaging party simply acquiesce in the
practice of the others.79
The second issue concerns the expansive potential of the doctrine. What
is clear is that subsequent practice can support both interpretation and
reinterpretation. According to some, it can even lead to modification.80
Often the disputes in which subsequent practice proves relevant mark the
first time a treaty provision has been authoritatively interpreted. Here
subsequent practice can provide strong evidence of the parties' original
understanding of their rights and obligations under the agreement. However,
subsequent practice may just as well evidence the parties' intention to
reinterpretthe treaty-in other words to augment or reject a prior interpretation
by positing a new one that may be different or even contrary to its precursor.
The parties are generally free to reinterpret their treaty again and again; the
judicial interpreter should simply examine what their subsequent practice
establishes as to their common understanding about the treaty, even if it entails
a departure from a prior interpretation.

77. ILC DALT, supra note 12, at 222.

78. GARDINER, supra note 14, at 227; SINCLAIR, supra note 72, at 137 (citing Yasseen,
L'interpritationdes traites d'apris la Convention de Vienne sur le droit des traitis, 151 RECUEIL DES
COURS 1, 48. (1978));
79. ILC DALT, supra note 12, at 222; VILLIGER, supra note 12, at 431; see also Arato, supra
note 71, at 460 (noting the perennial problem of identifying such acquiescence).
80. GARDINER, supra note 14, at 243-45; AMERASINGHE, supra note 15, at 54-55.
81. GARDINER, supra note 14 at 225-49.
310 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

Finally, an interpretation or reinterpretation based on the subsequent

practice of the parties may appear, on its face, to go beyond the scope of, or
even contradict, the plain meaning of the treaty text-here interpretation can
shade imperceptibly into modification.82 In the words of the ILC, modification
of treaties by subsequent practice means cases "where the parties by common
consent in fact apply the treaty in a manner which its provisions do not
envisage."83 There remains significant controversy as to whether it is
appropriate to rely on this interpretive doctrine to establish the modification of
a treaty. In finalizing the VCLT, the contracting parties explicitly rejected the
ILC's draft provision supporting treaty modification through subsequent
practice (Draft Article 38).84 Two arguments stood out in their discussions: first
that inclusion of such a provision would cause significant constitutional
problems for those parties whose domestic law requires the ratification of
treaties and treaty amendments; and second, with regard to multilateral
treaties with formal amendment rules, the idea of modification through practice
might sanction an end-run around the ordinary amending procedures.8 The
latter charge loomed especially large: to the extent that Draft Article 38 might
permit a treaty modification that could not have been achieved through formal
amendment procedures, it would seriously undermine the consent-based
foundations of the law of treaties.

82. See, e.g., Air Transport Services Agreement Arbitration (U.S. v. Fr.), 38 I.L.R. 182, 248-
55 (1963); Delimitation of the Border between Eritrea and Ethiopia (Eth./Eri.), 25 R.I.A.A. 83, 1 3.29
(Apr. 13, 2002) (setting out that in determining the weight of subsequent practice in the context of the
boundary delimiting provisions of a border treaty, the "conduct of one Party must be measured against
that of the other" and that "[e]ventually, but not necessarily so, the legal result may be to vary a
boundary established by a treaty").
83. ILC DALT, supra note 12, at 236.
84. Article 38 of the ILC's original Draft Articles on the Law of Treaties explicitly provided
for modification through subsequent practice in a separate provision. Id. art. 38. However this proposed
article was rejected in its entirety-indeed it was the only provision to suffer the fate of wholesale
elimination by the representatives of the parties. United Nations Conference on the Law of Treaties, Ist
Sess., Mar. 26-May 24, 1968, Summary Records of the Plenary Meetings and of the Meetings of the
Committee as a Whole 215, U.N. Doc. A/CONF.39/C.1/SR.38 (Apr. 25, 1968) [hereinafter VCLT
Conference Minutes].
85. Jean-Jacques de Bresson (France) insisted upon a principle of "formal parallelism,"
whereby modifications of a treaty should follow the same domestic procedure as was required to ratify
the original text. "If the manner in which the responsible officials applied the treaty was in itself capable
of leading to modification, that requirement of parallelism could hardly be met." VCLT Conference
Minutes, supra note 84, at 208. See also statement by Masato Fujisake (Japan), noting that "the Japanese
Constitution stipulated that treaties must be concluded with the approval of the Legislature, and the same
rule applied to the amendment of a treaty." Id. at 208.
86. As noted by de Bresson (France), "many international agreements contained specific
provisions on the conditions of their revision: to admit that the parties could derogate from those clauses
merely by their conduct in the application of the treaty would deprive those provisions of all meaning."
Id. at 208.
87. Santiago Martinez-Caro (Spain) best expressed the concern, proclaiming that "article 38
could mean that it was possible and legal to do by tacit agreement what it was impossible and illegal to
do by formal agreement; it could only be regarded as conflicting with the principle pacta sunt
servanda." Id. at 209; see also Tribe, supra note 60 (raising similar objections to the possibility of the
informal amendment of the U.S. Constitution).
2013] Treaty Interpretationand ConstitutionalTransformation 311

And yet the deletion of Draft Article 38 from the final VCLT did not lay
the question to rest, even at the time. Controversial as the prospect may be, at
least a few international courts and tribunals have relied on subsequent practice
to establish treaty modification-both before and after the 1969 conclusion of
the VCLT.89 Finally, it should be borne in mind that the line between
interpretation and modification will always be mercurial. 90
In its traditional form, interpretation on the basis of subsequent practice
relies on the actual existence and nature of the parties' practice. The technique
can support significant treaty change, but only to the extent established by the
parties' common practice in applying the treaty (or their acquiescence in one
another's conduct). Both the possibility and extent of any such change are thus
supposed to remain, more or less, in the hands of the states parties. The idea of
relying upon subsequent practice thus does not undermine the foundational
status of state consent as such. As traditionally understood, the technique is
rather a device for giving voice to the changing intentions of the parties, while
protecting the limits of their consent.

B. Subsequent Practicein the WTO: A TraditionalApproach

The WTO-AB exemplifies the traditional voluntarism of the law of

treaties in its approach to subsequent practice-taking rigorous care to respect
the consent of the parties (and the limits thereof) even in engaging in
constitutional interpretation. When interpreting its complex constituent
instrument, the highest judicial organ of the WTO relies assiduously on
subsequent practice in a way that conforms to the traditional contours of the
technique. As such, the interpretive practices of the WTO-AB provide a useful

88. Several delegations insisted that the provision reflected already extant customary
international law. See, for example, statements by Adolfo Maresca (Italy) ("[A]rticle 38 reflected a legal
fact which had always existed. International law was not a slave to formalism and by reason of its nature
must adapt itself to practical realities"), and Mustafa Kamil Yasseen (Iraq) (stating that Article 38
"reflected positive law"). VCLT Conference Minutes, supra note 84, at 211. Even as the VCLT
committee voted to reject Article 38, special rapporteur Sir Humphrey Waldock of the ILC insisted that
the principle was already enshrined in positive (customary) international law. Id. at 214. Even earlier,
Sir Gerald Fitzmaurice, the previous special rapporteur on the law of treaties, had similarly argued that
subsequent practice was a legitimate means of modification under contemporary international law.
Gerald Fitzmaurice, The Law and Procedure of the International Court of Justice 1951-4: Treaty
Interpretation and Other Treaty Points, 33 BRIT. Y.B. INT'L L. 203, 225 (1957) (stating that just as it is
"the duty of a tribunal 'to interpret treaties, not to revise them', it is equally the duty of a tribunal to
interpret them as revised, and to give effect to any revision arrived at by the parties"-including through
their conduct).
89. See, e.g., Air Transport Services Agreement Arbitration (U.S. v. Fr.), 38 I.L.R. 182, 248-
55 (1963); Delimitation of the Border between Eritrea and Ethiopia (Eth./Eri.), 25 R.I.A.A. 83, 1 3.29,
4.60 (Apr. 13, 2002).
90. In voting against Article 38 at the Vienna Conference, Shabtai Rosenne explained that the
Israeli delegation considered the provision redundant to the future 31(3)(b): "A theoretical distinction
certainly existed between subsequent practice as a means of interpreting a treaty and the modification of
a treaty through subsequent practice in its application; but in practice, the consequences were
substantially the same, so that it did not seem necessary to insert a separate article." VCLT Conference
Minutes, supranote 84, at 213.
91. See Georg Nolte, Subsequent Practiceas a Means ofInterpretation in the Jurisprudenceof
the WTO Appellate Body, in THE LAW OF TREATIEs BEYOND THE VIENNA CONVENTION, supra note 74,
at 138; Feldman, supranote 19.
312 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

control to better gauge the breadth and effects of the more liberal approaches to
subsequent practice of the ICJ and the ECtHR.
The Appellate Body was formally constituted by the relatively recent
Marrakesh Agreement, as part of the modem-day Dispute Settlement Body
(DSB) of the WTO. It must be understood, however, that this seemingly
discrete treaty incorporates a complex web of agreements linking together a
coherent system of trade law reaching back across the past half century. 92
Properly understood, the Body is constituted by this entire inter-locking regime
of treaties, taken as a whole ("the Agreements").93 It has jurisdiction to review
points of law on appeal from the decisions of the lower Panels, concerning the
entire web of Agreements. 94 Its opinions ("Reports") become binding and
authoritative absent an agreement by all of the parties not to adopt it, including
the party winning its appeal ("negative consensus").9 5 As such, the WTO-AB
has essentially binding authority to interpret the Agreements, and enjoys ample
opportunity to do so.
The Agreements may be understood together as the formal constitution of
the WTO system. And indeed they are frequently portrayed in constitutional
terms, often by appeal to their singular importance in the international
economic order.96 But what is most significant, for present purposes, is that the
Agreements establish a set of permanent institutions, charged with various
functions of governance over the parties. Among the constituted bodies, the
WTO-AB is the highest judicial organ of the system. At least formally its task
of authoritatively interpreting the Agreements may be qualified as
constitutional interpretation. Yet, as we shall see, the WTO-AB goes about its
task of interpretation in an especially rigid way, generally eschewing dynamism
in favor of strict deference to member state consent.
The WTO-AB routinely relies on VCLT Article 31 in interpreting the
Agreements.97 In considering subsequent practice as part of its analysis, it
frequently appeals to VCLT 31(3)(b). 98 It has added specificity to the sparse
words of the Vienna rules, specifically invoking Yasseen's formulation. In the

92. As WTO Director-General Lamy notes, "The WTO is a treaty comprising some 500 pages
of text accompanied by more than 2,000 pages of schedules and commitments." Lamy, supranote 47, at
971. Moreover, "50 years' worth of GATT practice and decisions, what is known as the 'GAT
acquis'-havebeen incorporated into what constitutes the new WTO Treaty." Id. at 971-72.
93. See id. at 971-72; Appellate Body Report, Argentinaafeguard Measures on Imports of
Footwear, 81, WT/DS121/AB/R (Dec. 14, 1999) (the WTO Agreements "must afortioribe read as
representing an inseparable package of rights and disciplines which have to be considered in
conjunction").
94. See ISABELLE VAN DAMME, TREATY INTERPRETATION BY THE WTO APPELLATE BODY
293 (2009); Lamy, supra note 47, at 972.
95. Marrakesh Agreement Establishing the World Trade Organization, Annex 2, arts. 6.1,
16.4, 17.4, Apr. 15, 1994, 1869 U.N.T.S. 401 [hereinafter DSU]; see John Jackson, The WTO DSU:
Misunderstandingson the Nature ofLegal Obligation, 91 AM. J. INT'L L. 60 (1997).
96. See Jackson, supra note 26; Jessica Lawrence, Contesting Constitutionalism:
ConstitutionalDiscourse at the WTO, 2 GLOBAL CONSTITUTIONALISM 63 (2013).
97. See, e.g., Appellate Body Report, United States--Standards for Reformulated and
Conventional Gasoline, at 17, WT/DS2/ABIR (Apr. 29, 1996).
98. See, e.g., Appellate Body Report, Japan-Taxes on Alcoholic Beverages, at 10-12,
WT/DS8/AB/R, WT/DS IO/AB/R, WT/DS 1 I/ABIR (Oct. 4, 1996).
2013] Treaty Interpretationand ConstitutionalTransformation 313

view of the WTO-AB, 31(3)(b) permits reliance upon only a "'concordant,

common and consistent' sequence of acts or pronouncements which is
sufficient to establish a discernible pattern implying the agreement of the
parties regarding [the treaty's] interpretation."99 The Body couches its approach
in highly voluntaristic terms. It insists that "'[s]ubsequent practice' in the
application of a treaty may be an important element in treaty interpretation
because it constitutes objective evidence of the understanding of the parties on
the meaning of the treaty."100 And indeed the WTO-AB has consistently
followed through with its rhetoric in practice-by rigorously demarcating the
contours of the doctrine in a manner calculated to reflect only the actual
consent of the parties, and to respect its limits. The caution in the Appellate
Body's approach is manifest in its statements about both the evidentiary criteria
for establishing subsequent practice, and the expansive potential of the doctrine
as applied to the Agreements.
With regard to evidentiary criteria, the WTO-AB has been particularly
clear about whose practice it will look to and what types of conduct it will
accept. The kinds of practice generally at issue reflect the complexity of the
Agreements under the Court's jurisdiction. The issue tends to arise in disputes
relating to specific parties' substantive trade commitments under the
Agreements, for example their bound tariff rates (under the GATT), or their
market commitments for trade in services (under the GATS)-in other words,
technical trade commitments that are negotiated bilaterally or plurilaterally, and
automatically incorporated into the Agreements upon adoption.10' Parties in
dispute before the WTO-AB have relied upon a wide variety of "practices" in
construing one another's trade commitments, including: the practice of the
parties in applying their trade commitments,' 02 as well as their conduct within
the organs and committees of the WTO. 0 3 In all cases, the WTO-AB has
viewed such claims skeptically, with an eye to the primacy of the consent of the
parties.

99. Japan-Alcoholic Beverages II, WT/DS8/AB/R at 12; see also Appellate Body Report,
United States-MeasuresAffecting the Cross-Border Supply of Gambling and Betting Services, 1 192,
WT/DS285/AB/R (Apr. 7, 2005) (stating that to establish authentic subsequent practice, "(i) there must
be a common, consistent and discernable pattern of acts or pronouncements; and (ii) those acts or
pronouncements must imply agreement on the interpretation of the relevant provision").
100. Appellate Body Report, European Communities-Customs Classification of Frozen
Boneless Chicken Cuts, 255, WT/DS269/AB/R, WT/DS286/AB/R (Sept. 12, 2005) (emphasis added)
(quoting 2 Y.B. INT'L L. COMM'N 1, 219, A/CN.4/SER.A/1966/Add. 1).
101. See Appellate Body Report, European Communities-Customs Classification of Certain
Computer Equipment, T 109, WT/DS62/AB/R, WT/DS67/ABIR, WT/DS68/AB/R (June 5, 1998)
("[T]he fact that Members' Schedules are an integral part of the GATT 1994 indicates that, while each
Schedule represents the tariff commitments made by one Member, they represent a common agreement
among all Members."). In WTO parlance, "plurilateral" refers to an agreement between more than two
parties, but not amounting to an agreement among all of the parties.
102. EC-Chicken Cuts, WT/DS269/AB/R at 1 252 (concerning the classification practices of
the European Communities).
103. Japan-AlcoholicBeverages II, WT/DSIO/ABIR at 13 (concerning the adoption of panel
reports by the parties in the DSB); US-Gambling, WT/DS285/AB/R, $ 193 (concerning Scheduling
Guidelines adopted by the Council for Trade in Services).
314 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

First, the WTO-AB only looks to the direct practice of states in applying
the Agreements; it has refused to consider the resolutions or decisions of
various organs and bodies of the WTO as a suitable proxy for the subsequent
practice of the parties.' The Appellate Body has been most firm, in this
regard, in Japan-Alcoholic Beverages II, where it rejected the possibility that
decisions of Panels in previous disputes constituted relevant subsequent
practice, even upon adoption by the parties (either by affirmative consensus,
under GATT 1947, or by "negative consensus" under the WTO).os
Second, even in considering the weight of the parties' conduct in
applying their individual trade commitments under the Agreements, the WTO-
AB will only give weight to practice that evidences a common interpretation of
all of the parties. The WTO-AB has proven unwilling to rely on the conduct of
even a vast majority of the parties in the face of contrary practice by a small
fraction of them. 06
The WTO-AB does not require that absolutely all of the parties
affirmatively engage in the practice. It has noted its willingness, under certain
narrow conditions, to give weight to the practice of some parties coupled with
the acquiescence of the others.107 In EC-Chicken Cuts, the Appellate Body
acknowledged that "not each and every party must have engaged in a particular
08
practice for it to qualify as a 'common' and 'concordant' practice."',
Nevertheless, it demands evidence of active participation in the practice by a
substantial number of parties.
[P]ractice by some, but not all parties is obviously not of the same order as practice
by only one, or very few parties. To our mind, it would be difficult to establish a
'concordant, common and discernible pattern' on the basis of acts or
pronouncements of one, or very few parties to a multilateral treaty, such as the WTO
Agreement.109

Further, it requires that such substantial active participation be

complimented by the discernible acquiescence of the others. The Appellate

104. US-Gambling, WT/DS285/AB/R, } 193 ("We have difficulty accepting Antigua's

position that the 2001 Scheduling Guidelines constitute 'subsequent practice' revealing a common
understanding that Members' specific commitments are to be construed in accordance with W/120 and
the 1993 Scheduling Guidelines. Although the 2001 Guidelines were explicitly adopted by the Council
for Trade in Services . .. they do not constitute evidence of Members' understanding regarding the
interpretation of existing commitments ... [and do not] of themselves, constitute 'subsequent practice'
within the meaning of Article 31(3)(b) of the Vienna Convention"); Japan-Alcoholic Beverages II,
WT/DSIO/AB/R at 13.
105. Japan-Alcoholic Beverages II, WTIDSIO/AB/R at 13-14 ("We do not believe that the
CONTRACTING PARTIES, in deciding to adopt a panel report, intended that their decision would
constitute a definitive interpretation of the relevant provisions of GATT 1947 . . . their character and
their legal status have not been changed by the coming into force of the WTO Agreement .. . [thus we do
not agree] that panel reports adopted by the GATT CONTRACTING PARTIES and the WTO Dispute
Settlement Body constitute subsequent practice in a specific case" (internal quotation marks omitted)).
106. EC-Computer Equipment, WT/DS62/AB/R, % 92-93; id. T 95 ("Inconsistent
classification practice, however, cannot be relevant in interpreting the meaning of a tariff concession.").
107. EC-Chicken Cuts, WT/DS269/AB/R, f 272 ("We do not exclude that, in specific
situations, the 'lack of reaction' or silence by a particular treaty party may, in the light of attendant
circumstances, be understood as acceptance of the practice of other treaty parties.").
108. Id. F259.
109. Id.
2013] Treaty Interpretationand ConstitutionalTransformation 315

Body's discomfort with making assumptions about the intentions of the parties
is manifest. It requires that such acquiescence be evident, and not simply
presumed from silence.
Such situations may occur when a party that has not engaged in a practice has
become or has been made aware of the practice of other parties (for example, by
means of notification or by virtue of participation in a forum where it is discussed),
10
but does not react to it.]
Finally, and perhaps most radically, the WTO-AB will only consider the
conduct of the parties in the interpretation of the Agreements as relevant if it is
sufficiently determinate. In other words, the Appellate Body requires evidence
that the parties not only engaged in a given practice, but that the conduct
reflects their actual understanding of the meaning of the provisions in
question-something like opiniojuris, by analogy to the elements of customary
international law."' The Appellate Body most clearly articulated its
requirement of such determinacy in US-Gambling, where it refused to take
into account certain non-binding guidelines adopted by the Council for Trade in
Services as authentic subsequent practice. In the view of the WTO-AB, the
Guidelines in question were non-binding, and were, moreover, adopted "in the
context of the negotiation of future [trade] commitments." 12 They were
essentially forward-looking113 :
As such, they do not constitute evidence of Members' understanding regarding the
interpretation of existing commitments. Furthermore, as the United States
emphasized before the Panel, in its Decision adopting the 2001 Guidelines, the
Council for Trade in Services explicitly stated that they were to be 'non-binding'
and 'shall not modify any rights or obligations of the Members under the GATS.114
Thus the Appellate Body decided that it could not consider the guidelines as
sufficiently determinate to qualify as "'subsequent practice' revealing a
common understanding."115
The WTO-AB takes a similarly restrained view of the expansive potential
of subsequent practice. While the Appellate Body has indicated its willingness
to consider the subsequent intentions of the parties for purposes of
interpretation, it rejects the possibility that any interpretation may go so far as
to modify the Agreements. 116 Of course the line between interpretation and

110. Id.T272.
111. In US-Gambling, the WTO-AB emphasized that the technique requires a demonstration
that the practice in question actually reflects the parties' view on interpretation: to qualify as practice,
the relevant "acts or pronouncements must imply agreement on the interpretation of the relevant
provision." US-Gambling, WT/DS285/AB/R at 1 192; see also Japan-Alcoholic Beverages II,
WT/DSIO/AB/R, at 13.
112. US-Gambling,WTIDS285/ABIR, 1 193.
113. See Nolte, supranote 91, at 138.
I14. US-Gambling,WT/DS285/AB/R, 193.
115. Id.
116. See Appellate Body Report, European Communities-Regimefor the Importation, Sale,
and Distribution of Bananas, Second Recourse to Article 21.5 of the DSU by Ecuador, iJ 391-93,
WT/DS27AB/RW2/ECU, WT/DS27/AB/RW/USA (Nov. 26, 2008) (explaining, while specifically
considering interpretation on the basis of subsequent agreement (VCLT 31(3)(a)), that modification
through interpretation would be impermissible).
316 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

modification is hazy. What is important is that the Appellate Body draws a line
at all, and that it does so out of fidelity to the text as the best and highest
evidence of the intentions of the parties.
In sum, the WTO-AB will only consider the conduct of the parties where
two conditions are met: first, if there is evidence that a substantial number of
parties have actively engaged in the practice, no states have acted in a directly
contrary fashion, and the acquiescence of the others can be demonstrated (as
opposed to simply presumed on the basis of their silence); and second, if there
is some determinate evidence that the practice actually represents an agreement
of the parties regarding interpretation. And even where such conditions are met,
the Appellate Body will only rely on such practice to interpret the
Agreements-never to modify their provisions.
From the constitutional perspective, this glimpse into the WTO-AB's
interpretive practice yields a static and restrained picture. As the foregoing
evidences in microcosm, the WTO-AB makes sincere efforts to respect the
extent of and limits to the consent of the parties, as evident in its outsized
reliance on textual analysis in interpretation over and above other canons.1 It
thus rigorously restricts its use of subsequent practice so as not to take liberties
in holding parties to interpretations to which they might not have consented.'
In other words, while the WTO-AB may be said to be engaging in
"constitutional interpretation" as a formal matter, its approach must be
characterized as restrained, voluntaristic, and highly textualist; one would be
pressed to point to any major idiosyncrasies in its approach to subsequent
practice as compared to the typical case of ad hoc dispute resolution. The
WTO-AB treats the parties with genuine deference, as veritable masters of the
treaties. The same cannot so easily be said of either the ICJ or the ECtHR.
IV. CONSTITUTIONAL INTERPRETATION IN THE UNITED NATIONS: "THE
PRACTICE OF THE ORGANIZATION" AS SUBSEQUENT PRACTICE IN THE ICJ

As compared to the WTO, the United Nations is a highly dynamic

constituted organization. Its formal constitution, the U.N. Charter, establishes
six official organs,119 among which the ICJ represents the "principal judicial

117. See Douglas Irwin & Joseph Weiler, Measures Affecting the Cross-Border Supply of
Gambling and Betting Services (DS 285), 7 WORLD TRADE REv. 71, 89 (2008) (branding the interpretive
approach of the WTO-AB as "textual fetishism").
118. Its restrictive attitude toward interpretation is not limited to 31(3)(b). For example, Jos6
Alvarez notes the WTO-AB's similar caution in interpreting the Agreements in light of "other rules of
international law applicable in the relations between the parties[.]" Jos6 Alvarez, The FactorsDriving
and Constraining the Incorporation of International Law in WTO Adjudication, in THE WTO:
GOVERNANCE, DISPUTE SETTLEMENT, AND DEVELOPING COUNTRIES 611 (Merit Janow ed., 2008)
(quoting VCLT 31(3)(c)). Alvarez suggests that the WTO-AB's aversion to any appearance of law-
making can be partially explained by the express terms of Article 3.2 of the DSU, which states that
"1recommendations and rulings of the DSB cannot add to or diminish the rights and obligations provided
in the covered agreements," and in part by a more general perception of their own "tenuous legitimacy."
Id. at 616.
119. U.N. Charter art. 7 (including in the full list the U.N. General Assembly (UNGA), the U.N.
Security Council (UNSC), the Economic and Social Council (ECOSOC), the Trusteeship Council, and
the Secretariat). See generally ALF Ross, CONSTITUTION OF THE UNITED NATIONS: ANALYSIS OF
2013] Treaty Interpretationand ConstitutionalTransformation 317

organ of the United Nations."l 20 The Charter formally delegates certain powers
to each of the constituted bodies, and provides a general framework for their
interaction. 121 It should be borne in mind throughout that the formal terms of
the Charter are extremely difficult to change, in light of its rigid provisions on
amendment.122
The Charter leaves the ICJ in a relatively weak position. First, the Charter
confers upon the Court only two thin bases of jurisdiction. It may only exercise
jurisdiction over states in "contentious cases" with their express consent, and
only its judgments in these kinds of cases are technically binding.123 Its only
other basis of jurisdiction is its power to render non-binding advisory opinions
at the request of a competent institution.124 Nevertheless, as a material matter
the advisory opinions of the ICJ carry significant weight, especially insofar as
they express the opinion of the Court on an interpretation of the Charter.125
Arguably, the non-binding form actually frees the Court to adopt more
adventurous legal positions.126

STRUCTURE AND FUNCTION (1950) (providing an early analysis of the United Nations in constitutional
terms).
120. U.N. Charter art. 92.
121. However, in finalizing the Charter at the 1945 San Francisco Conference, the parties
indicated their understanding that each principal organ of the United Nations would have some leeway
to interpret its own competence for itself. 11 DOCUMENTS OF THE UNITED NATIONS CONFERENCE ON
INTERNATIONAL ORGANIZATION, SAN FRANCISCO 37 (1945); see SIMON CHESTERMAN, THOMAS
FRANCK & DAVID MALONE, LAW AND PRACTICE OF THE UNITED NATIONS: DOCUMENTS AND
COMMENTARY 8 (2008).
122. U.N. Charter arts. 108-09 (requiring a two-thirds super-majority vote, plus the ratifications
of two-thirds of the members. Also required is the assent of each of the five permanent members of the
UNSC); see HANS KELSEN, THE LAW OF THE UNITED NATIONS: A CRITICAL ANALYSIS OF ITS
FUNDAMENTAL PROBLEMS 818-19 (2000) (confirming that the facially different amendment provisions
in Articles 108 and 109 amount, numerically, to the same thing). By comparison, the amendment rule of
the WTO is, for most matters, significantly more flexible. A handful of fundamental articles can only be
amended by unanimity, including core substantive provisions of the GATT, General Agreement on
Trade in Services (GATS), and Agreement on Trade-Related Aspects of Intellectual Property Rights
(TRIPS), the rules of decision-making and interpretation (Article IX), and the amendment rule itself
(Article X). The rest may be amended under a variety of procedures requiring a qualified majority-
thereby altering only the rights and obligations of those parties who have adopted the amendment. See
WTO Agreement, art. X.
123. Statute of the International Court of Justice art. 35, June 26, 1945, 59 Stat. 1055, 33
U.N.T.S. 993 [hereinafter I.C.J. Statute]. Thus, as against the stronger jurisdictional bases of the WTO-
AB and especially the ECtHR, the ICJ enjoys comparatively few opportunities to interpret the Charter.
Moreover, it lacks any formal power to interpret the Charter with universally binding effect. See
ALVAREZ, supra note 2, at 68.
124. U.N. Charter art. 96(a) (empowering the UNGA and UNSC to request an "advisory
opinion on any legal question"); id. art. 96(b) (adding that "[o]ther organs of the United Nations and
specialized agencies, which may at any time be so authorized by the General Assembly, may also
request advisory opinions of the Court on legal questions arising within the scope of their activities").
125. MOHAMMED SAMEH M. AMR, THE ROLE OF THE INTERNATIONAL COURT OF JUSTICE AS
THE PRINCIPLE JUDICIAL ORGAN OF THE UNITED NATIONS 115 (2003).
126. See Jos6 Alvarez, The New Dispute Settlers: (Half) Truths and Consequences, 38 TEX.
INT'L L.J. 405, 423, 431 (2003) (suggesting that the very softness of the advisory-jurisdiction form may
afford international courts like the ICJ greater leeway to make general pronouncements of law, with
greater legal effect, than is possible in the context of binding international dispute settlement). By
contrast and perhaps counterintuitively, as Alvarez notes elsewhere, the WTO-AB may be "more not
less reticent to make broad pronouncements of law" than the ICJ, despite enjoying an apparently
stronger basis of jurisdiction and formally binding adjudicative authority; precisely because of these
apparent strengths the Body may be all the more wary that "the perception of judicial activism
318 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

The ICJ frequently relies on subsequent practice when appraised with the
interpretation of treaties.127 Usually it hews quite closely to the traditional
contours of the technique. However, it has developed a peculiar approach to
subsequent practice when engaged in the interpretation of the U.N. Charter.
The Court has developed its sui generis approach gradually, over the course of
three advisory opinions spanning forty years: Certain Expenses, Namibia, and
Wall. In each case, the Court has progressively broadened the contours of its
interpretive approach. Taken together, these three cases represent a significant
material transformation of the constitution of the United Nations as a whole.

A. From "Subsequent Practice"to the "Practiceof the Organization"

Everything begins with Certain Expenses (1962). At issue there was the
interpretation of the General Assembly's (UNGA) budgetary competence. The
Charter delimits the organ's budgetary power at Article 17, which provides in
relevant part: "(1) The General Assembly shall consider and approve the
budget of the organization; (2) The expenses of the Organization shall be borne
by the members as apportioned by the General Assembly."1 28
The Court was asked to render an advisory opinion as to whether certain
expenditures authorized by the UNGA (specifically, extraordinary expenses to
cover the costs of U.N. operations in the Congo (ONUC) and the U.N.
Emergency Force in the Middle East (UNEF) 129) fell within its competence
under Article 17, or were rather ultra vires. In relevant part, the Court
addressed several arguments intended to cabin the General Assembly's
budgetary authority under Article 17 as implicitly limited to "regular" or
"administrative" expenses. 130 The ICJ ultimately found that the budgetary
expenditures were well within the General Assembly's competences. For
present purposes, the ICJ's very plausible substantive interpretation of the
Charter as including such expenditures is not particularly noteworthy. What is
extraordinary is the Court's method of interpretation. 131
The Court articulated its approach in very traditional terms. It noted that
when it interprets the Charter, it follows "the principles and rules applicable in
general to the interpretation of treaties" for the simple enough reason that "the
Charter is a multilateral treaty, albeit a treaty having certain special
characteristics."l32 But its next move was critical: the Court then indicated that
it would consider, in this regard, "the manner in which the organs concerned

encourage WTO Members to adopt reforms that would weaken the WTO dispute settlement system."
Alvarez, supra note 118, at 628.
127. See Kasikili/Sedudu Island (Botswana/Namib.), 1999 I.C.J. 1045 (13 Dec.); Temple of
Preah Vihear (Cambodia v. Thai.), 1962 I.C.J. 6, 34 (June 15).
128. U.N. Charter art. 17.
129. Certain Expenses of the United Nations, Advisory Opinion, 1962 1.C.J. 151, 156 (July 20).
130. Id. at 159, 165.
131. The gravity of the Court's interpretive approach was already clear at the time to a handful
of the judges writing separate opinions, and above all Sir Percy Spender. Id. at 189-90 (separate opinion
of Judge Spender). The full potential of the Court's nascent canon would only become apparent in
subsequent opinions.
132. Id. at 157.
2013] Treaty Interpretationand ConstitutionalTransformation 319

'have consistently interpreted the text' in their practice."1 33 Everyone agreed at

the time that the general law of treaties had long included a principle
authorizing the consideration of subsequent practice like that eventually
codified at VCLT 31(3)(b).1 34 Yet subsequent practice had always referred to
the subsequent conduct of the states parties. There seems to have been
absolutely no precedent for the notion that the practice of the organs of an
international organization might act as a proxy for subsequent state conduct.135
And this position engendered no small amount of controversy among the
judges in the event-certain of whom decried the seismic potential of the
Court's novel interpretive shift.' 36
Though muddying the novelty of its position, the Court was abundantly
clear in articulating its approach to interpreting the Charter: Certain Expenses
stands for the proposition that the consistent practice of the organs of an
international organization is an authentic and appropriate guide to the meaning
of that organization's constituent instrument.137 The Court relied upon its
nascent doctrine alongside a barebones textual analysis to dismiss the attempts
to limit the scope of the General Assembly's budgetary power. First, the Court
stated that Article 17 cannot be interpreted as excluding peace and security
from the budgetary competence of the UNGA in light of that organ's
longstanding conduct. The Court noted that "[i]t is a consistent practice of the
General Assembly to include in the annual budget resolutions, provision for
expenses relating to the maintenance of international peace and security." 38 On
this basis, it concluded that "there is no justification for reading into the text of
Article 17, paragraph 1, any limiting or qualifying word before the word
'budget. "'l139
Finally, the Court turned from its general interpretation of Article 17 of
the Charter to the more specific question of whether the particular impugned
expenses fell within the UNGA's broad mandate. Here again, the Court gave
great weight (indeed seemingly decisive weight) to the UNGA's own practice.
With regard to funding UNEF operations, the Court concluded that, "from year
to year, the expenses of UNEF have been treated by the General Assembly as

133. Id.
134. Id. at 189-90 (separate opinion of Judge Spender).
135. The Court claimed that its approach was nothing new and cited several "precedents,"
including Competence of the General Assembly for the Admission of a State to the United Nations,
Advisory Opinion, 1950 I.C.J. 4, 8-9 (Mar. 3). The Court did note there that the organs at issue had
"consistently interpreted the text" in a way that conformed to its interpretation of the plain text.
However, the Court in no way relied upon such practice as anything like an authentic criterion of
interpretation. The Court explicitly decided upon an interpretation based on the plain meaning of the
text, and only then indicated in passing that such meaning was home out by the practice of the relevant
organs. As noted by Judge Spender in his separate opinion to Certain Expenses, the Court could not
claim any precedent in 1962 for the notion that it may actually ground its interpretation of the Charter on
organ practice. Certain Expenses of the United Nations, 1962 I.C.J. 151, 194-95 (July 20) (separate
opinion of Judge Spender).
136. Id. at 189-90 (separate opinion of Judge Spender); id. at 230-31 (dissenting opinion of
President Winiarski).
137. Certain Expenses, 1962 I.C.J. at 157-65.
138. Id. at 160.
139. Id. at 161.
320 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

expenses of the Organization within the meaning of Article 17, paragraph 2, of

the Charter."l 40 Turning to ONUC, the Court reasoned that the "General
Assembly has twice decided that even though certain expenses are
'extraordinary' and 'essentially different' from those under the 'regular
budget', they are none the less 'expenses of the organization"' for purposes of
Article 17(2).141 Most surprisingly of all, several of the Resolutions relied upon
were not unanimous, and indeed entailed substantial negative votes by diverse
and representative states, including even the USSR-a Permanent Member of
the Security Council.142 The Court considered it sufficient that the Resolutions
in question were "adopted by the requisite two-thirds majority" necessary for
authorizing expenditures.143
To reiterate, it is not the particular interpretation of the UNGA's
budgetary authority that is important here-the substance of the Court's
interpretation was on its face eminently plausible. What gives pause is the
Court's subtle assertion of a canon of interpretation based on the subsequent
practice of the organs of the Organization. And, indeed, the innovative nature
of this assertion was not lost on everyone at the time: while the opinion of the
Court blithely assumes its method as a given,'" several of the judges
underscore its novelty in their separate opinions. 145 CertainExpenses represents
a highly significant extension of the traditional state-centric canon based on the
subsequent practice of the states parties in applying a treaty to encompass
"organ practice."l46 For the purpose of interpreting the formal constitution of an
international organization, the practice of the constituted bodies (organs) would
henceforth join the practice of the constituent power (states parties) in the
pantheon of authoritative evidentiary criteria. Most critically, though left
unsaid, the Court considered that the resolutions of an organ like the General
Assembly can be afforded interpretive weight even if adopted in the face of
dissenting votes-making it difficult, in other words, to argue that such "organ

140. Id. at 175.

141. Id. at 178-79.
142. Great weight was placed, for example, on Resolution 1089 (XI), which was adopted by a
vote of sixty-two to eight, with seven abstentions. Id. at 174 (citing U.N.G.A, Res. 1089 (XI)
A/RES/1089 (XI) (Dec. 21, 1956)); see also Christopher Peters, Subsequent Practice and Established
Practice of InternationalOrganizations:Two Sides of the Same Coin?, 3 GOETTINGEN J. INT'L L. 617,
640 (2011). The Court also gave weight to Resolution 1151 (XII), adopted with fifty-one votes in favor,
eleven against, and nineteen abstentions (including a negative vote by the USSR). Official voting
records for the foregoing Resolutions are available at the website of the General Assembly. United
Nations General Assembly-Voting Records (last visited Apr. 17, 2013), http://www.un.org/en/ga
/documents/voting.asp.
143. Certain Expenses of the United Nations, 1962 I.C.J. 151, 174 (July 20).
144. Indeed, the Court relied on previous cases that provide little support for its method. See,
e.g., Competence of the General Assembly for the Admission of a State to the United Nations, Advisory
Opinion, 1950 I.C.J. 4, 8-9 (March 3). The ICJ may be best understood here as indulging in a typical
case of what Ackerman and Golove have called the judicial "myth of continuity." Ackerman & Golove,
supra note 60, at 890-91 (explaining the tendency of U.S. Courts to insist that there is nothing new
under the sun, even while contributing to the thorough transformation of the Constitution).
145. See Certain Expenses, 1962 I.C.J. at 189-90 (separate opinion of Judge Spender); id. at
230-31 (dissenting opinion of President Winiarski).
146. See also ALVAREZ, supra note 2, at 87-92.
2013] Treaty Interpretationand Constitutional Transformation 321

practice" is relevant simply by virtue of reflecting the agreement of the parties

on the interpretation of the Charter.
Nine years later, in 1971, the Court clarified its approach in a second
advisory opinion in the Namibia case. The Court there confirmed that it viewed
organ practice as connected to and grounded in the venerable doctrine of
interpretation on the basis of the subsequent practice of the parties. Moreover, it
clarified the expansive scope of its technique as a doctrine capable of
establishing an interpretation at clear odds with the plain text.
In relevant part, Namibia interprets the Charter on the basis of the
consistent practice of the UNSC. In Resolution 284 (1970) the Council
requested the Court's opinion on the legality of South Africa's continued
presence in Namibia (originating under the Mandate system of the League of
Nations).147 The crucial moment comes at the outset of the advisory opinion, in
the Court's assessment of its jurisdiction. South Africa objected, inter alia, that
the Council's request for an advisory opinion suffered from fatal procedural
flaws of sufficient gravity as to render Resolution 284 ultra vires, null and
void-thereby removing the grounds of the ICJ's advisory jurisdiction.148
South Africa's main complaint alleged fatal voting irregularities in the
adoption of the request for an advisory opinion.149 The Charter articulates the
Council's voting procedures at Article 27(3), providing that decisions of the
UNSC on non-procedural matters "shall be made by an affirmative vote of nine 50
members including the concurring votes of the permanent members."1
Resolution 284 was adopted by twelve members of the Security Council with
three abstentions; however, South Africa insisted that the Resolution was ultra
vires nevertheless, because two of the abstainers were permanent members-
Great Britain and the Soviet Union.' 5 1 The case thus turned on the
interpretation of Article 27(3): is "concurring vote" limited to a vote in the
affirmative, or is concurrence broad enough to include abstentions? South
Africa alleged that the phrase "including the concurring votes" of the P5
requires that each permanent member explicitly vote in the affirmative for a
decision of the Council to be legally valid.152 The Court rejected that
interpretation, reading the provision to require only that the permanent
members do not vote in the negative-thereby crystallizing the "veto" as we
know it today.153
Again, what is crucial for present purposes is the Court's methodology.
The ICJ expended no words assessing the "ordinary meaning" of Article 27(3),

147. Legal Consequences for States of the Continued Presence of South Africa in Namibia
(South West Africa) Notwithstanding Security Council Resolution 276 (1970), Advisory Opinion, 1971
I.C.J. 16, T I (June 21).
148. Id. 20.
149. Id. 21.
150. U.N. Charter art. 27(3) (emphasis added).
151. Namibia, 1971 I.C.J. 21.
152. Id. 21.
153. Id. 22.
322 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

let alone its context or object and purpose. 154 It relied exclusively on the
Council's practice to establish its interpretation. The critical passage reads:
[T]he proceedings of the Security Council extending over a long period supply
abundant evidence that presidential rulings and the positions taken by members of
the Council, in particular its permanent members, have consistently and uniformly
interpreted the practice of voluntary abstention by a permanent member as not
constituting a bar to the adoption of resolutions .... This procedure followed by the
Security Council, which has continued unchanged after the amendment in 1965 of
Article 27 of the Charter, has been generally accepted by Members of the United
Nations and evidences a general practice of that Organization. 5 5

Herein lie several important insights into the Court's use of organ practice in
interpreting the Charter, each of which needs to be excavated.
First, this passage elucidates the connection between organ practice and
the subsequent practice of the states parties to the Charter. Rather than simply
emphasizing the practice of the Council as such, the Court indicates that the
organ's practice is probative because it represents the subsequent practice of
the members of the Council, and "in particular its permanent members." 56
Moreover, the Court insists on the importance that such practice has been
"generally accepted by the Members of the United Nations" as a whole.157 The
Court indicates that the probative value of organ practice for interpretation will
depend either upon whether the practice is accepted by the membership as a
whole, or whether the members have at a minimum acquiesced. ss The Court
thus indicates that it relies on "organ practice," coupled with the members'
acquiescence, as a kind of proxy for the subsequent practice of the states
parties. 59 It subsumes this coupling under the broader concept of a "general
practice of the Organization."' 60
Namibia further extends the expansive potential of the Court's
interpretive approach. The case differs critically from Certain Expenses in that

154. See VCLT, supra note 8, art. 31(1). Indeed, the Court made no mention, in this regard, of
the Vienna Convention on the Law of Treaties of 1969.
155. Namibia, 1971 I.C.J.Ij 22.
156. Id.
157. Id.
158. Recall that under the traditional view of subsequent practice, the subsequent conduct of
one or more parties will be considered authentic subsequent practice even in the absence of similar
conduct by all parties, so long as those not engaging in the practice acquiesce in the others' conduct.
159. Recall, however, that in Certain Expenses the Court relied on certain GA Resolutions
adopted over the dissenting votes of some Member States-a practice to which the Court would return
in the 2004 Wall opinion. Legal Consequences of the Construction of a Wall in the Occupied Palestinian
Territory, Advisory Opinion, 2004 I.C.J. 136 (July 9). To the extent that the Court considers the practice
of the organization relevant only insofar as it reflects the understanding of the Membership as a whole, it
seems to thus consider the simple fact of negative votes to be insufficiently determinate to defeat the
possibility that a Resolution reflects an interpretation of the Charter. The view is not entirely untenable:
a negative vote could have a variety of explanations, and it is not outrageous for the Court to require a
more explicit protest to defeat the presumption that a UNGA Resolution carries interpretive weight.
160. Namibia, 1971 I.C.J. T, 22. By encapsulating the entirety (organ practice plus Member
acquiescence) within the concept of a "general practice of the Organization," the Court reintroduces
some obscurity into its interpretive approach. Without further illuminating the meaning of the
expression, Namibia simply indicates that the "general practice of the Organization" is the plenary
criterion for interpretation-merely evidenced by the practice of its organs and the acquiescence of its
members. What remains clear is those elements will be, together, sufficient to establish an interpretation.
2013] Treaty Interpretationand ConstitutionalTransformation 323

the resulting interpretation is not particularly easy to square with the text. The
interpretation in Certain Expenses is by any measure a plausible reading of
Article 17(2)-it simply construes "expenses of the organization" as including
"extraordinary" expenses as opposed to merely those required to meet normal
administrative budgetary needs. 161 The Court's interpretation in Namibia, by
contrast, entails greater linguistic contortion. Article 27(3), again, provides that
decisions of the UNSC "shall be made by an affirmative vote of nine members
including the concurring votes of the permanent members."l62 The requirement
of their concurring votes is framed in very active language, and strongly seems
to require the affirmative votes of the P5. Yet the Court inverts the plain
language, requiring only that none of the P5 actively vote against the proposal.
In the Court's reading, "By abstaining, a member does not signify its objection
to the approval of what is being proposed," and does not thereby defeat the
measure for lack of concurrence.163 Namibia thus confirms that the Court's
technique of interpretation can establish interpretations that strain the text,
arguably amounting to a modification.
In one brief paragraph, Namibia expounds two aspects of the doctrine of
"organ practice" in dramatic fashion: it elucidates the evidentiary criteria of the
technique as organ practice coupled with member acquiescence; and it extends
the scope of the technique to include interpretations at odds with the plain text,
arguably even amounting to a modification.165 The Court thereby establishes a
rebuttable presumption that the organs can establish an interpretation of the
Charter through their practice, even one difficult to square with the text, with
the burden on the member states to protest or reject the interpretive relevance
of any organ practice that might establish an interpretation with which they
disagree.166

161. Certain Expenses of the United Nations, Advisory Opinion, 1962 I.C.J. 151, 178-79
(July 20).
162. U.N. Charter art. 27(3) (emphasis added).
163. Namibia, 1971 I.C.J. 22.
164. Writing in dissent on other grounds, Judge Fitzmaurice stated that "the practice of an
organization, or of a particular organ of it, can modify the manner of exercise of one of its functions (as
for instance in the case of the veto in the Security Council which is not deemed to be involved by a mere
abstention)." Id. 94 (dissenting opinion of Judge Fitzmaurice) (continuing to note that "such practice
cannot, in principle, modify or add to the function itself'). It may be noted in passing that this is the
same Sir Gerald Fitzmaurice who had previously sat as ILC Special Rapporteur on the Law of Treaties
and there insisted on the possibility that subsequent practice can establish treaty modification.
GARDINER, supra note 14, at 245 (stating, of Namibia, that "this seems an example of an interpretation
close to accepting amendment by practice, particularly given that the preparatory work of the Charter
suggests an intent contrary to the adopted practice on the part of the eventual permanent members of the
Security Council"). This position runs afoul of the harsh counsel of Judge Spender in the earlier case of
CertainExpenses, refusing to agree "that a common practice pursued by an organ of the United Nations,
though ultra vires and in point of fact having the result of amending the Charter, may nonetheless be
effective as a criterion of interpretation." Certain Expenses, 1962 I.C.J. at 189-90 (separate opinion of
Judge Spender).
165. Namibia, 1971 I.C.J. 122.
166. As established by Certain Expenses, the mere fact that an organ's recommendation was
adopted over negative votes does not vitiate its interpretive value as the conduct of that organ. Certain
Expenses, 1962 I.C.J. at 174. It would thus appear that, in the view of the ICJ, protest must be more
explicit and determinate to rebut the presumption in favor of assigning weight to the recommendations
or decisions of the political bodies.
324 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

The Wall opinion represents the final and most recent case in our trio. It
mirrors Namibia in that it concerns a challenge to the legal validity of the
request for an advisory opinion under the terms of the Charter-this time issued
by the General Assembly. In its Tenth Emergency Session in December 2003,
the UNGA resolved to request an advisory opinion from the ICJ concerning the
legality of Israel's construction of a wall in the Occupied Palestinian
Territories, including in and around East Jerusalem.167 Israel alleged, however,
that the General Assembly's request exceeded its competence under the
Charter. Though normally the UNGA would be competent to issue such a
request, Israel contended that its competences were curtailed by Article 12(1)
of the Charter-which limits the General Assembly's sphere of action when the
Security Council is apprised of a situation. The issue was thus whether, in
requesting the Court's advisory opinion, the UNGA impermissibly encroached
upon the UNSC's competence.
Article 12, paragraph I of the Charter provides: "[W]hile the Security
Council is exercising in respect of any dispute or situation the functions
assigned to it in the present Charter, the General Assembly shall not make any
recommendation with regard to that dispute or situation unless the Security
Council so requests."l68 In light of Article 12(1), and "given the active
engagement of the Security council with the situation in the Middle East,
including the Palestinian Question," Israel contended that "the General
Assembly acted ultra vires under the Charter" in requesting an advisory
opinion on the construction of the wall in the Occupied Palestinian Territory-
rendering the request null and void, and stripping the Court of jurisdiction. 169In
deciding the case, the Court confirmed that it would examine the significance
of Article 12(1) by70
"having regard to the relevant texts and the practice of the
United Nations."'
As in Certain Expenses and Namibia, the Court mainly looked to the
practice of the relevant organs, this time including both the UNGA and the
UNSC; unlike those two preceding cases, the Court found itself confronted
with a glaring inconsistency in the organs' practice over time.
As regards the practice of the United Nations, both the General Assembly and the
Security Council initially interpreted and applied Article 12 to the effect that the
Assembly could not make a recommendation on a question concerning the
maintenance of international peace and security while the matter remained on the
Council's agenda. 7'
And yet, of course things can change-not least in the realm of legal
interpretation. In view of the more recent "practice of the United Nations,"
including that of both the UNGA and the UNSC, the Court found that "this

167. See Legal Consequences of the Construction of a Wall in the Occupied Palestinian
Territory, Advisory Opinion, 2004 I.C.J. 136, 1l (July 9).
168. U.N. Charter art. 12(l).
169. Wall, 2004 I.C.J. T24.
170. Id. T 25.
171. Id.27.
2013] Treaty Interpretationand ConstitutionalTransformation 325

interpretation of Article 12 has evolved subsequently."l 72 The Court noted that

"the General Assembly deemed itself entitled in 1961 to adopt
recommendations in the matter of the Congo. . . and in 1963 in respect of the
Portuguese colonies," in both cases while the situation "still appeared on the
Council's agenda, without the Council having adopted any recent resolution
concerning them."l 73 Once again, the Court relied mostly on Resolutions
adopted in the face of substantial negative votes, including by a member of the
P5.174 Even more probative for the Court was the fact that "there has been an
increasing tendency over time for the General Assembly and the Security
Council to deal in parallel with the same matter concerning the maintenance of
international peace and security."'175 Finally, the Court took into consideration
the opinion of the Legal Counsel of the United Nations, who confirmed in
response to a question posed by Peru "that the Assembly interpreted the words
'is exercising the functions' in Article 7612 of the Charter as meaning 'is
exercising the functions at this moment."'
In other words, the Court held that the "practice of the United Nations"
evidences a dramatic reinterpretation of the Charter. Although Article 12 may
once have been understood as preempting the General Assembly's request for
an opinion in a case like this, it can no longer be understood as so limiting the
organ's competence today. 77
Wall pushes the ball in three ways. One concerns the evidentiary criteria
required to establish a "practice of the organization" relevant to the
interpretation of the Charter. The other two implicate the expansive potential of
interpretation on the basis of organizational practice.
First, the Court confirms and extends its position on the question of
whose subsequent practice is relevant for interpreting the Charter. Echoing its
brief remark in Namibia, the Court states that the relevant criterion is the
"practice of the United Nations."' 78 As in both Namibia and Certain Expenses,
the Court in Wall looks primarily to the practice of the relevant organs-the

172. Id.
173. Id.
174. Id. (citing G.A. Res. 1600 (XV), U.N. Doc. A/RES/1600 (Apr. 15, 1961) (adopted with 60
votes in favor, 16 votes against, and 23 abstentions, including a dissenting vote by the USSR); G.A. Res.
1913 (XVIII), U.N. Doc. A/RES/1913 (Dec. 13, 1963) (adopted by 91 affirmative votes over 2 negative
votes)). Official voting records for the foregoing Resolutions are available at the website of the General
Assembly. Voting Records, UNITED NATIONS, http://www.un.orglen/ga/documents/voting.asp (last
visited Apr. 17, 2013).
175. Wall, 2004 I.C.J. 27. The Court also cites the matters involving Cyprus, South Africa,
Angola, Southern Rhodesia and more recently Bosnia and Herzegovina and Somalia. The Court adds
that "[i]t is often the case that, while the Security Council has tended to focus on the aspects of such
matters related to international peace and security, the General Assembly has taken a broader view,
considering also their humanitarian, social and economic aspects." Id.
176. Id. (quoting U.N. GAOR 23rd Sess., 1637th mtg. 9, U.N. Doc. A/C.3/SR.1637, (Dec. 12,
1968)).
177. Id.128.
178. Id. $T 27-28; see also Legal Consequences for States of the Continued Presence of South
Africa in Namibia (South West Africa) Notwithstanding Security Council Resolution 276 (1970),
Advisory Opinion, 1971 I.C.J. 16, 1 22 (June 21) (using the more general phrase "practice of th[e]
[o]rganization," to be established by the general practice of the organs of the organization coupled with
the acquiescence of the Member States).
326 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

UNSC and the UNGA, both of whose competences are at stake in the
interpretation of Article 12. But here the Court goes further in extending its
examination beyond the practice of only the organs of the United Nations,
attaching weight to statements by the Legal Counsel of the United Nations.179
The Legal Counsel is indeed a constituted body, and may be many things
more-but the office is by no means a formal organ under the terms of the
Charter.'so And, unlike the UNSC and UNGA, the office cannot claim to speak
on behalf of even one state. Thus the Court seems to indicate that
organizational practice might not need to reflect state practice even indirectly.
The Court's consideration of the views of the U.N. Legal Counsel may
seem like only a marginal extension. Yet closer scrutiny reveals its singular
symbolism. This apparently small move is important because it undermines the
link between organizational practice and the affirmative "subsequent practice"
of any actual states. Even if only in this small way, the opinion seems to
indicate that acquiescence may be enough to maintain the bond between the
"practice of the organization" and the intentions of its members. Henceforth,
the Court indicates, the presumption shall be that the organization as a whole
determines the meaning of the Charter; insofar as they disagree with the
practice of the organization, the burden falls upon the states parties to express
their disapproval.
Second, the Court confirms its position in Namibia concerning the
expansive potential of the interpretive technique by again adopting an
interpretation that contorts the Charter's text. To reiterate, Article 12 states that
"while the Security Council is exercising in respect of any dispute or situation
the functions assigned to it in the present Charter, the General Assembly shall
not make any recommendation with regard to that dispute or situation unless
the Security Council so requests." 8 The provision seems categorical, on its
face. In Wall, however, the Court interprets the provision as less-than-
categorical in three ways. First, it reads Article 12(1) to allow the UNGA to
adopt recommendations regarding a situation on the Council's docket about
which the latter has not adopted any recent resolutions.182 Second, it reads the
phrase "is exercising the functions" as "is exercising the functions at this
moment."183 Finally, it indicates that in some cases the UNSC and UNGA may
act in concert consistently with Article 12(1) by focusing on different aspects of
a situation-the Council might focus on international peace and security, while
the General Assembly might take a broader view, considering "humanitarian,
social and economic aspects" of the situation. Taken together these three
exceptions chafe against the categorical "shall not" language of Article 12(1),

179. Wall, 2004 I.C.J. T 27.

180. See U.N. Charter art. 7 (establishing only six formal organs: the UNGA, UNSC,
ECOSOC, Trusteeship Council, ICJ, and the Secretariat).
181. Id. art. 12(l).
182. Wall, 2004 I.C.J. T 27.
183. Id. (emphasis added).
184. Id.
2013] Treaty Interpretationand ConstitutionalTransformation 327

arguably amounting to a modification of the provision.185 In any event, they

confirm the Court's view that organizational practice can support the very
significant development of the Charter's text.
Finally, the Court makes clear that practice may support a
reinterpretation, even establishing a new interpretation that is both entirely
inconsistent with the prior practice of the organization and difficult to square
with the text. Not only did the Court significantly contort Article 12(1) on the
basis of the practice of the organization, but it did so in spite of a long history
of materially countervailing practice. What mattered was not that the relevant
organs had always engaged in a certain practice, but that once they started
doing so they continued to do so consistently. Though the earlier practice may
have supported an earlier interpretation (one, we might add, more consistent
with the text), the Court relied on a later phase of practice to find that the
1 86
correct interpretation of Article 12 "has evolved subsequently."
Thus Wall clarifies and develops the expansive potential of the ICJ's
interpretive doctrine based on the "practice of the organization," as well as its
evidentiary criteria. Wall makes clear that organizational practice may ground
not only an interpretation at significant odds with the text, but also successive
interpretations that chafe against both the text and the prior consistent practice
of the organization. In other words, the practice of the organization can
establish interpretation, reinterpretation, and arguably even successive
modifications. As to the establishment of the authentic practice of the
organization, the Court suggests that it is not only the practice of U.N. organs
(composed of representatives of states) that carry weight, but also the practice
of non-representative bodies like the office of the Legal Counsel of the United
Nations. Carried to its full conclusions, the interpretive rule asserted in Wall
seems to unravel all bonds between the practice of the organization and the
practice of the states parties to the Charter save one: the gossamer thread
requiring at least tacit consent through acquiescence (meaning here the absence
of outright and direct protest).

B. A TransformativeJurisprudenceandIts PoliticalEffects

Big things can occur through small steps.' 87 Through three advisory
opinions spanning several generations, the ICJ has contributed to a sizeable
transformation of the material constitution of the United Nations, resulting in
an organization more autonomous, assertive, and powerful over its constituent
member states. It did not accomplish this feat alone, but rather in conjunction
with the more active organs (the UNGA and the UNSC). Still, at the heart of

185. See Jessica Liang, Modifying the UN Charter Through Subsequent Practice:Prospectsfor
the Charter'sRevitalisation, 81 NORDIC J. INT'L L. 1, 10 (2012) (suggesting that the interpretation in
Wall represents a "procedural modification").
186. Wall, 2004 1.C.J. 27.
187. See Ackerman & Golove, supra note 60, at 873 (using the term "piecemeal
precedentialism" to describe the contribution of the U.S. courts to the transformation of the U.S.
Constitution, not through a single judgment but through the combination of a series of disparate opinions
acting in conjunction with the movement of the political branches of government).
328 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

this subtle evolution lies the Court's increasingly expansive assertion that the
Charter may be legitimately interpreted, reinterpreted, and even modified on
the basis of the practice of the organization. Before assessing the constitutional
implications of the ICJ's interpretive practice, it is worth pausing for a moment
to take stock-by comparing its reliance on organizational practice to the
traditional contours of the subsequent practice rule as it is usually employed in
ad hoc dispute settlement.
As traditionally employed, subsequent practice has relatively strict
evidentiary criteria: the rule only grants interpretive weight to the conduct of
the states partiesto a treaty; and only if all of the parties engage in the conduct,
or at least materially acquiesce in one or another's conduct in applying the
treaty.'s The traditional rule is also generally understood as having a relatively
limited scope. While everyone agrees that subsequent practice can support
interpretation and reinterpretation, it is controversial whether the doctrine
extends to modification under the general law of treaties.189 Even those who
contend that the subsequent practice of the parties can support modification
tend to indicate that such expansive change would demand a more exacting
evidentiary standard.190 The rationale behind both the strictness of the
evidentiary criteria, and the discomfort surrounding the possibility of informal
modification, is the sanctity of state consent in the law of treaties.
As it stands today, the ICJ's understanding of the role of practice in the
interpretation of its constituent instrument contrasts sharply with the traditional
rule. In interpreting the Charter, the Court has loosened the evidentiary criteria
for establishing authentic practice, and it has relied on the technique to
establish quite substantial treaty change over time. First, in lieu of relying on
the direct practice of the parties, the Court treats the "practice of the
organization" as a proxy for state conduct-including the practice of organs
like the UNSC and UNGA and other non-representative bodies like the office
of legal counsel.19 The Court has proven willing to give this proxy
presumptive interpretive weight, indicating that it will consider such practice
authoritative so long as the parties appear to have acquiesced.192 At least in the
case of the practice of the General Assembly, mere evidence that a number of
states voted against a particular Resolution has proven insufficient to defeat this
presumption, even where a substantial and diverse group votes against,

188. See GARDINER, supra note 14, at 109-12. See also the view of the WTO-AB, that conduct
only counts as subsequent practice where it is "concordant, common, and consistent" among the parties.
Appellate Body Report, Japan-Taxes on Alcoholic Beverages, at 13, WT/DS8/AB/R, WT/DSIO/AB/R,
WT/DSI1 /AB/R (Oct. 4, 1996).
189. See supra text accompanying note 84.
190. See Legal Consequences for States of the Continued Presence of South Africa in Namibia
(South West Africa) Notwithstanding Security Council Resolution 276 (1970), Advisory Opinion, 1971
I.C.J. 16, ' 94 (June 21) (dissenting opinion of Judge Fitzmaurice); AMERASINGHE, supra note 15, at
463 (suggesting that especially clear and determinate evidence of the intentions of the parties would be
necessary in such cases); Peters, supra note 142, at 632.
191. Wall, 2004 I.C.J.1127.
192. Namibia, 1971 I.C.J. T 22.
2013] Treaty Interpretationand Constitutional Transformation 329

including a member of the P5. In other words, the Court seems to treat the
practice of the organization as authoritative so long as no party explicitly
protests the relevance of a particular conduct for interpretation. Second, as
regards the expansive potential of the technique, the Court has relied upon the
consistent practice of the organization to establish: an interpretation of the
Charter; a reinterpretation of the Charter at odds with previous consistent
practices; and even interpretations and reinterpretations at significant odds with
the text, arguably amounting to the informal modification of the Charter's
express terms. 194
From the perspective of the law of treaties, the ICJ's gradual
jurisprudence on subsequent practice represents a thorough broadening of a
patently malleable rule of interpretation. Although the breadth of its doctrine
may be remarkable, there is nothing particularly surprising about Courts
stretching techniques of interpretation to suit their ends. From the point of view
of constitutional theory, however, and in conjunction with the active practice of
the UNGA and the UNSC, it would appear that the Court's interpretive
practices have contributed to the dramatic transformation of the constitution of
the United Nations as a whole.
Recall that judicial interpretation may lead to constitutional
transformation in two ways. On the one hand, a particular interpretation may
bring about an immediate first-order change, where the judicial body directly
interprets (and changes) the normative architecture of the organization-as, for
example, where it interprets provisions of its constituent instrument delineating
the competences of the organs. On the other hand, a judicial body may bring
about a second-order transformation by asserting a particular method of
interpretation over time, insofar as its methodological approach itself alters or
develops the material constitution of the organization. The series of advisory
opinions outlined above appears to qualify under both types. However the
second, here, proves far more significant than the first.
Each of the three advisory opinions arguably entails a degree of first order
constitutional change.' 95 In CertainExpenses, the Court developed the material
budgetary authority of the UNGA, though admittedly in a manner that fits quite
comfortably within the formal text.' 96 Namibia, by contrast, entailed very little
in the way of a material transformation but rather confirmed the long-standing
practice of the UNSC regarding voting-procedures; the interpretation does,
however, appear to strain the Charter's formal text.' 97 Finally, Wall entailed the
clearest first-order transformation. The Court acknowledged that both the plain
text and the consistent practice of the UNGA and UNSC had long ago

193. Certain Expenses of United Nations, Advisory Opinion, 1962 I.C.J. 151, 174 (July 20);
Wall, 2004 I.C.J. T 27; see supra text accompanying notes 142 and 174.
194. Wall, 2004 I.C.J. 11 27-28.
195. As noted above these opinions are non-binding, but the advisory opinions of the "principal
judicial organ of the United Nations" are highly authoritative to the extent that they entail the
interpretation of the Charter. U.N. Charter, art. 92; see also AMR, supra note 125, at 115.
196. See supra text accompanying note 139. The case represents at most a development of the
material constitution of the United Nations that sits comfortably with the formal Charter.
197. See supra text accompanying note 164.
330 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

established an authentic interpretation of the Charter to the effect that the

former could not make any recommendations concerning a situation currently
on the docket of the latter. However the Court considered that recent and
apparently consistent trends in the practices of both organs established a
reinterpretation of the Charter permitting the Assembly to make such
recommendations where the Council had failed to act, or was not actively
considering the situation "at that moment." Wall thus entailed a clear material
shift in the ordering of competences between the two bodies. Moreover, this
shift is difficult to square with the Charter's formal text.
The importance of each of these first-order developments should not be
dismissed. Taken together, however, the series of advisory opinions reveals a
much more imposing second-order constitutional transformation. Through the
gradual articulation of its principle of interpretation on the basis of the "practice
of the organization," the Court has materially transformed the constitution of
the United Nations in such a way as to render this constituted organization
significantly more autonomous and indeed more powerful vis-i-vis the
constituent power, namely the states parties to the Charter.
The single most important fact about the Court's approach to
interpretation, from the perspective of constitutional theory, is that its reliance
on the practice of the organization frays the threads of consent tying the United
Nations to its member states. The Court presumes that the practice of the
organs and other constituted bodies of the organization may establish an
interpretation of the Charter, except in the remote situation where the relevance
of such practice to interpretation is specifically protested within the
membership-and simply voting against a resolution apparently does not count
(at least not in the UNGA). As the Court has twice held, interpretation on the
basis of such practice may contort the Charter's formal text.' 99 As a result, the
Court has developed a mechanism for affecting informal constitutional change
within the United Nations, capable of skirting the stringent requirements of the
formal amendment rule.200 In other words, through these advisory opinions the
ICJ has sanctioned, and partially developed, an informal mode of constitutional
change based on the usages, customs and practices of the various constituted
bodies. The Court's approach empowers the constituted bodies to develop the
material constitution of the United Nations without the express assent of the
parties. Indeed it leaves the constituent member states with only the narrowest

198. Certain Expenses, 1962 I.C.J. at 174; Namibia, 1971 I.C.J. T 22.
199. Namibia, 1971 I.C.J. T 22; Wall, 2004 I.C.J. I.27-28.
200. The Court's principle of interpretation on the basis of the autonomous practice of the
organization, when coupled with the often expansive practice of the more active political organs, would
appear to vindicate Judge Lauterpacht's portrayal of the U.N. Charter as a living constitution. Voting
Procedure on Questions Relating to Reports and Petitions Concerning the Territory of South-West
Africa, Advisory Opinion, 1955 I.C.J. 67, 106 (June 7) (separate opinion of Judge Lauterpacht) (stating
that the interpretation of the Charter must take into account "its operation in actual practice and in the
light of the revealed tendencies in the life of the Organization").
2013] Treaty Interpretationand ConstitutionalTransformation 331

forms of recourse to prevent such change (through explicit and determinate

201
protest) or to counteract it (by recourse to the rigid formal amendment rule).
Ultimately, the full constitutional significance of the ICJ's approach to
interpretation lies in the fact that it empowers the other organs of the United
Nations significantly more than it empowers itself. Unlike other tribunals, the
World Court is institutionally incapable of playing the role of gatekeeper. It
cannot monitor the practice of the various organs and decide when such
practice is sufficiently consistent as to establish an interpretation or
reinterpretation of the Charter. The ICJ simply lacks the jurisdictional
capabilities: it has no compulsory jurisdiction, nor may it render advisory
opinions of its own accord. Moreover, its contentious cases are only binding on
the parties to the dispute and its advisory opinions are not formally binding on
anyone.202 But its blessing of constitutional change and development through
practice has long resonated throughout the organization. By far the most
dramatic examples of transformation through practice have not been litigated
before the ICJ, but can be easily justified under its interpretive approach203: for
example the UNSC's creation of criminal tribunals to try international crimes
in Rwanda and the Former Yugoslavia,204 as well as its more recent legislative
experiments in promulgating general norms of indefinite duration to combat
terrorism205 and nuclear proliferation.206 Indeed, as Alvarez rightly notes, "the
principle which accords weight to institutional practice is a powerful tool for
the expansion of power at the expense of the wishes of even a majority of the
membership."207

201. It is true that a kernel of this idea may be found in the original travaux of the Charter,
embedded in a well-known statement adopted by the San Francisco Conference. The drafters there
acknowledged that the malleable terms of the Charter may from time to time have to be interpreted by
the organs of the organization themselves. However "it is to be understood," the statement insists
that if an interpretation made by any organ of the Organization ... is not generally
acceptable it will be without binding force. In such circumstances, or in cases where it is
desired to establish an authoritative interpretation as a precedent for the future, it may be
necessary to embody the interpretation in an amendment to the Charter.
ALVAREZ, supra note 2, at 79. Alvarez rightly characterizes this statement on interpretation as "neither
total abdication to national sovereignty nor idealized resort to supranational authority." Id. This kernel
seems, however, a far cry from the thoroughly supranational principle of reliance on the "practice of the
organization" in the form actually worked out in the jurisprudence of the ICJ-whereby the Charter may
be interpreted, reinterpreted, and arguably modified on the basis of even contested organ practice (and,
apparently, the practice of other non-organic bodies). The extent to which the essence of the ICJ's
approach is contained in the statement by the drafters in San Francisco is perhaps debatable. In any case
the unfolding of this kernel over sixty years of practice must be understood as a profound constitutional
movement.
202. U.N. Charter art. 96; see Alvarez, supranote 126, at 423, 431.
203. See generally Arato, supra note 42 (examining, in this regard, the UNSC's assertion of
general legislative powers in the areas of international criminal law, combatting terrorism, and nuclear
non-proliferation).
204. S.C. Res. 955, U.N. Doc. S/Res/955 (Nov. 8, 1994) (establishing the ICTR); S.C. Res.
827, U.N. Doc. S/Res/827 (May 25, 1993) (establishing the ICTY).
205. S.C. Res. 1267, U.N. Doc. S/Res/1267 (Oct. 15, 1999); S.C. Res. 1373, U.N. Doc.
S/Res/1373 (Sept. 28, 2001).
206. S.C. Res. 1540, U.N. Doc. S/Res/1540 (Apr. 28, 2004).
207. ALVAREZ, supra note 2, at 92.
332 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

From the juridical perspective of the law of treaties and treaty

interpretation, the ICJ takes a very broad approach to the doctrine of
subsequent practice. But its breadth can be (and has been) bracketed as the
province of a special regime: the interpretive community of the United
Nations.208 From a political-theoretical perspective, however, it appears that the
broad interpretive rule has contributed to the material transformation of the
constitution of the organization. The material significance of the Court's
approach to interpretation lies in the fact that it sanctifies a second track of
constitutional change, outside of the formal amendment rule. The informal
track is, in the end, more readily available to the other organs than to the Court
itself. But what is important is that the "principal judicial organ of the United
Nations" has sanctioned the more active organs' past and future reliance on this
informal mode of change. True, the approach most directly empowers the other
organs, but at a higher level of abstraction it significantly contributes to the
autonomy of the organization as a whole. Under the color of the law of treaties,
the Court has illuminated a path by which the organization can develop and
expand the scope of its competences, through the sustained practice of its more
capable organs. This informal mechanism circumvents the rigid formal
amendment rule and enables organizational development without direct
recourse to the constituent member states. Taken together, and through no
small interpretive alchemy, the organs of the United Nations have contributed
to the gradual transformation of the Organization, enhancing the autonomy and
power of the constituted bodies over and above the constituent power.

V. THE TRANSFORMATION OF THE EUROPEAN COURT OF HUMAN RIGHTS:

"EUROPEAN CONSENSUS" AS SUBSEQUENT PRACTICE

Of the Courts considered in this Article, the European Court of Human

Rights is the most dynamic. Like the WTO-AB and the ICJ, the ECtHR is a
standing judicial organ of an international organization (the CoE) that engages
in a form of constitutional interpretation-through the sustained interpretation
of its constituent instrument (the ECHR). Its mandate includes the
interpretation of fundamental human rights as codified in the Convention, as
well as the mechanisms for their enforcement (including its own judicial
competences). And alone among the bodies considered here, the ECtHR has
expressly described its task as a constitutional enterprise.2o Like the ICJ-
though in a different way and with important differences in result-the
ECtHR's approach to constitutional interpretation has proven transformative in
effect.
However the organizational position of the ECtHR differs from that of
either the WTO-AB or the ICJ in three crucial respects, relating to its peculiar

208. See, e.g., AMERASINGHE, supra note 15, at 49-55; GARDINER, supra note 14, at 246-49.
209. Loizidou v. Turkey (preliminary objections) [GC], App. No. 15318/89, Eur. Ct. H.R. 1 75
(Mar 23, 1995) (describing the Court's task as interpreting a "constitutional instrument of European
public order").
2013] Treaty Interpretationand ConstitutionalTransformation 333

formal constitutional arrangement as an organ of the CoE; its bases of

jurisdiction; and its ad hoc procedures for formal amendment.
First, though usually studied in its own right, the ECtHR is formally a
judicial organ of the Council of Europe, alongside the supreme legislative
Committee of Ministers and a deliberative Parliamentary Assembly (PACE),
and amidst a wide range of specialized bodies and committees. This complex
organization was not created all at once but rather by separate successive
treaties. The CoE was originally founded by the 1949 Treaty of London,
establishing the Committee of Ministers and the Assembly. Only in the
following year did the members adopt the ECHR, codifying a bill of
fundamental human rights and establishing two new organs of the CoE to
enforce its provisions (the ECtHR and the now-defunct European Commission
on Human Rights). The ECHR is nevertheless understood as a fundamental and
integral instrument of the CoE, and its ratification is a condition of membership
in the organization. From a constitutional perspective, the ECHR is thus best
understood as both the constituent instrument of the ECtHR and an integral
aspect of the larger formal constitution of the CoE.
Second, the ECtHR enjoys a far stronger jurisdictional foothold than
either the ICJ or the WTO. Under the formal terms of the ECHR, the ECtHR
has compulsory jurisdiction over all of its members. 210 Moreover, it may hear
cases between the states parties, as well as complaints by individuals against
the member states.211 In both types of case, the ECtHR enjoys express authority
to interpret the Convention with binding effect on all parties.212 The possibility
of direct individual complaints is of critical importance, because it affords the
ECtHR an exponentially greater caseload than most international courts and
213
tribunals. The caseload causes problems, but for present purposes the point is
that it creates an enormous number of opportunities for the ECtHR to consider
and reconsider its interpretations of the Convention.
Third, in contrast to the WTO Agreements and the U.N. Charter, there is
no express procedure for amending the ECHR. Rather, amendment occurs
through an ad hoc procedure, whereby the Committee of Ministers of the CoE
drafts a proposal for an additional protocol to be submitted to the parties for

210. ECHR, supra note 23, arts. 32-34, 46-48.

211. Id. arts. 33, 34 (using the terms "Inter-State Cases" and "Individual Applications").
212. Id. arts. 32-46. In addition the Court has the capacity to render Advisory Opinions upon
request by two-thirds majority of the Committee of Ministers. Id. arts. 46-48 (The Court may refuse any
such request at its discretion). Although this procedure is now getting more political attention it has to
date only been used twice, regarding relatively minor issues concerning the election of new judges.
Advisory Opinion on Certain Legal Questions Concerning the Lists of Candidates Submitted with a
View to the Election of Judges to the European Court of Human Rights (Advisory Opinion No. 1), Eur.
Ct. H.R. (Feb. 12, 2008); Advisory Opinion on Certain Legal Questions Concerning the Lists of
Candidates Submitted with a View to the Election of Judges to the European Court of Human Rights
(Advisory Opinion No. 2), Eur. Ct. H.R. (Jan. 22, 2010).
213. As a result, the ECtHR suffers from an enormous backlog of cases, implicating all aspects
of the Convention. In 2009, 57,200 applications were allocated to judges, and the backlog reached
119,300. Fact Sheet: Protocol 14, The Reform of the European Court of Human Rights, COUNCIL OF
EUROPE (May 25, 2010), http://www.echr.coe.int/NR/rdonlyres/5721 1BCC-C88A-43C6-B540-AF0642
E81D2C/0/CPProtocole
14EN.pdf.
334 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

ratification. The procedure can take two different forms: an "amending

protocol" (requiring unanimous consent and ratification, and binding on all
parties) or an "optional protocol" (requiring a certain number of ratifications to
come into force, and binding only upon the ratifying states). 214 The parties have
relied on optional protocols to expand the substantive rights of the ECHR,2 15
while resorting to amending protocols to alter the institutional machinery of the
Convention (with one exception).21 6 On the one hand, this ad hoc procedure has
proven effective: the ECHR has been amended no less than fourteen times,
more than once with comprehensive and systematic effect. 1 On the other
hand, the pressure to adopt the unanimity-based procedure for amending the
organizational aspects of the Convention makes it difficult for the parties to
challenge the ECtHR's understanding of its competence through formal means.
Through its sustained interpretive practice the ECtHR has developed a
second, informal track of constitutional change within the Convention system.
Like the ICJ, the ECtHR has transformed its constitution through reliance on
subsequent practice in the interpretation of its constituent instrument, the
ECHR. In some ways it is even more explicit than the World Court. It expressly
grounds its approach in VCLT 31(3)(b), and indeed focuses much more directly
on the actual practice of the states parties.218 Yet just as explicitly, it has proven
willing to rely on the subsequent practice of a majority of the parties, even in

214. The approach is in many ways more flexible than the express amendment rule of the U.N.
Charter, and similar in spirit to the multifaceted amendment provisions of the WTO Agreement.
215. See, e.g., Protocol 7 to the Convention for the Protection of Human Rights and
Fundamental Freedoms, Nov. 22, 1984, Europ. T.S. No. 117 (entered into force Nov. 1, 1988)
(restricting the death penalty).
216. See, e.g., Protocol No. II to the Convention for the Protection of Human Rights and
Fundamental Freedoms, Restructuring the Control Machinery Established Thereby, May 11, 1994,
Europ. T.S. No. 155 (entered into force Nov. 1, 1998) (abolishing the European Commission on Human
Rights, allowing individuals to apply directly to the Court, and giving compulsory jurisdiction to the
Court over all disputes arising out of the ECHR); Protocol No. 14 to the Convention for the Protection
of Human Rights and Fundamental Freedoms, Amending the Control System of the Convention, May
13, 2004, Europ. T.S. No. 1194 (entered into force June 1, 2010) (establishing a "single-judge"
procedure to increase efficiency, whereby initial admissibility decisions may be made by one judge,
rather than by committee of three judges, as well as providing for enhanced enforcement mechanisms in
connection with the CoE-such as issuing interpretations of past judgments and/or ruling on a
Respondent Party's compliance with adverse judgments. But see Protocol l4bis to the Convention for
the Protection of Human Rights and Fundamental Freedoms, May 27, 2009, Europ. T.S. No. 204
(entered into force Oct. 1, 2009) (adopted as an optional protocol in order to afford the Court access to
the single judge procedure for cases against the ratifiers only, in order to address its backlog of cases
while waiting for the final ratification of Protocol 14).
217. See, e.g., Protocol 11, supra note 216; Protocol 14, supra note 216.
218. The ECtHR does, in fact, give significant weight to reports and resolutions of other organs
and bodies of the CoE; but it considers such soft law to be relevant to its interpretive task not as
subsequent practice, but as "relevant rules of international law applicable in the relations between the
parties" via VCLT, supra note 8, art. 31(3)(c). Although this aspect of the Court's interpretive approach
falls outside the scope of the present article, I have argued elsewhere that it has had a similarly
transformative effect. Julian Arato, Constitutional Transformation in the ECtHR: Strasbourg's
Expansive Recourse to External Rules of InternationalLaw, 37 BROOK. J. INT'L L. 349 (2012). Under
the rubric of VCLT, supra note 8, art. 31(3)(c), the Court has proven willing to interpret the Convention
on the basis of non-binding sources emanating from the CoE and other international organizations, as
well as treaty provisions neither signed nor ratified by even a majority of the States parties (including
even the Respondent in the particular case). See, e.g., Demir v. Turkey [GC], App. No. 34503/97, Eur.
Ct. H.R. (Nov. I1, 2008).
2013] Treaty Interpretationand ConstitutionalTransformation 335

plain view of contrary practice by the minority. Moreover, the ECtHR is much
more articulate than the ICJ about the expansive potential of its interpretive
doctrine: it has plainly stated that it regards the subsequent practice of the
parties as capable of establishing an interpretation, reinterpretation, and
modification of the Convention. Despite its nod to the terms of 31(3)(b), the
Court has developed a broad, sui generis approach to subsequent practice. And
as opposed to the ICJ's approach to organizational practice, the ECtHR's
approach is significantly more Court-empowering.
Before going any further, and in view of the ECtHR's enormous and
varied case-law, it is worth clarifying what this section is and is not meant to
do. First, the following is not an argument about how the ECtHR generally
interprets the ECHR or how it tends to do so. I am here intentionally drawing
out a selection of disparate cases with the intention of demonstrating a
particular assertion of judicial power, even if the ECtHR only makes full use of
that power relatively infrequently. Second, and in a similar vein, this section is
not meant to suggest that the ECtHR usually takes an expansive approach to the
interpretation of the rights in its charge. While most of the cases considered
here involve the expansion of rights, the full picture of the ECtHR's
jurisprudence is considerably more checkered.219 The following selection of
cases is rather meant to draw a very particular constellation, reflecting the
ECtHR's capacious understanding of its competences under the law of treaty
interpretation. Specifically, I want to demonstrate its willingness-some of the
time-to expand and even modify the ECHR on the basis of the subsequent
practice of a majority of the parties, despite clear evidence of a non-consenting
minority. Moreover it does so openly, as a matter of doctrinal principle under
the rubric of "European consensus."
Irrespective of whether the ECtHR makes full use of this self-generated
power in all, or even most cases, the assertion itself has had a profound and
material constitutional meaning. The sustained assertion of the doctrine of
European consensus represents a significant transformation of the relationship
between this constituted supranational Court and the constituent power-the
states parties to the ECHR.

A. "European Consensus": The Weight of Majorities,Minorities, and

Trends

The ECtHR expressly relies upon the Vienna rules in interpreting all
aspects of the ECHR, including both the substantive rights of the Convention
and its organizational provisions concerning the Court's competences,
jurisdiction, and procedure. As the Court acknowledged in Golder, "it should
be guided by Articles 31 to 33 of the Vienna Convention of 23 May 1969 on

219. Though the ECtHR has significantly advanced the cause of human rights in Europe, it is
often criticized for affording the member states too much leeway under the Convention. See, e.g., Oren
Gross & Fionnuala Ni Aolain, From Discretion to Scrutiny: Revisiting the Application of the Margin of
Appreciation Doctrine in the Context of Article 15 of the European Convention of Human Rights, 23
HUM. RTS. Q. 625 (2001).
336 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

the Law of Treaties," which "enunciate in essence generally accepted principles

of international law."220 The Court frequently relies upon the practice of the
parties as an authentic and wholly sufficient criterion of interpretation and has
explicitly grounded its approach on VCLT article 31(3)(b) on several
occasions. 22 1
However, the Court usually relies on the practice of the parties in a
somewhat eccentric context: the determination of the margin of appreciation of
the member states to decide individually the extent to which they will enforce
or limit the enforcement of the Convention rights at the national level. At the
heart of this much-criticized doctrine is an idea of deference to the parties and
respect for their varied legal traditions. The Convention is a supranational
instrument codifying open-textured human rights, enforceable directly by the
nationals of its member states; therefore the Court must, in its view, afford a
degree of respect to the different ways in which the parties balance rights
protection with other legitimate priorities. The Court will, however, always
enforce a certain minimum level of protection, a level it determines as a matter
of treaty interpretation in light of the Vienna rules-including inter alia, by
reference to the subsequent practice of the parties. Under the auspices of VCLT
article 31(3)(b), the Court examines the parties' domestic laws and de facto
administrative practices, 222 as well as their inter-state conduct,223 to determine
whether or not there is a consensus among them as to the appropriate minimum
level.224 It further appeals to such practices in order to determine whether it
should expand the appropriate minimum over time, shrinking the margin of
appreciation with regard to the right in question.
Within this narrow context, however, the ECtHR adopts an
extraordinarily broad construction of VCLT article 31(3)(b) itself, notable in
two important respects. First, the singular feature of the Court's approach is
that it does not require the common practice of all of the parties, even when it
is aware of the contradictory practice of a minority of parties. So long as the
practice of the parties evidences the agreement of most of them on a given
interpretation, the Court will consider the interpretation authentic and
established. 225 In the Court's words, it looks to practice in order to establish a
"European consensus"-but it really means "near-consensus" or even an

220. Golder v. United Kingdom, App. No. 4451/70, Eur. Ct. H.R. (Feb. 21, 1975) (plenary).
221. Bankovid v. Belgium (dec.) [GC], App. No. 52207/99, Eur. Ct. H.R. 11 56, 62 (Dec. 12,
2001); Loizidou v. Turkey (preliminary objections) [GC], App. No. 15318/89, Eur. Ct. H.R. T 75 (Mar.
23, 1995); Cruz Varas v. Sweden, App. No. 15576/89, Eur. Ct. H.R. 1100 (Mar. 20, 1991) (plenary).
222. See, e.g., Soering v. United Kingdom, App. No. 14038/88, Eur. Ct. H.R. (July 7, 1989)
(plenary).
223. See Al-Saadoon v. United Kingdom, App. No. 61498/08, Eur. Ct. H.R. (Mar. 2, 2010);
Ocalan v. Turkey [GC], App. No. 46221/99, Eur. Ct. H.R. (May 12, 2005).
224. Note, of course, that the Court relies on all other provisions of the Vienna Rules in setting
this level, and will not necessarily set a low level simply because no such consensus can be found. For
example, even in the absence of a "European Consensus," the Court may nevertheless deny the margin
in light of other "relevant rules of international law" as per VCLT, supra note 8, art. 31(3)(c). See Demir
v. Turkey [GC], App. No. 34503/97, Eur. Ct. H.R. (Nov. 11, 2008); Arato, supra note 218, at 349.
225. See Kanstantsin Dzehtsiarou, European Consensus and the Evolutive Interpretation of the
European Convention on Human Rights, 12 GER. L.J. 1730, 1733 (2011).
2013] Treaty Interpretationand ConstitutionalTransformation 337

evolving trend toward consensus.226 Second, the ECtHR openly considers

interpretation on the basis of Article 31(3)(b) to be capable of establishing quite
expansive treaty change over time. The Court explicitly holds that
interpretation on the basis of the subsequent practice of the parties can establish
an amendment of the Convention. These two areas of breadth compound one
another. In the most spectacular cases, the Court has expressly grounded treaty
modification on European consensus even in the face of explicitly contrary
practice by a handful of the parties. 227

1. Establishing Consensus: The Subsequent Practiceof (Most of)

the Parties

The ECtHR has expressly identified the doctrine of European consensus

with interpretation on the basis of the subsequent practice of the parties to the
ECHR within the meaning of VCLT article 31(3)(b). 228 In relying on such
consensus to justify a developmental interpretation of the Convention as a
"living" instrument, the ECtHR's stated purpose is to enhance the protection of
human rights in lockstep with tendencies among the parties to buttress and
expand rights protection. In the Court's words, a failure to maintain a dynamic
approach in view of such tendencies would "risk rendering [the Convention] a
bar to reform or improvement." 229
But the doctrine of European consensus departs significantly from the
notion of subsequent practice as it is usually understood. As one commentator
has recently noted, by "consensus" the ECtHR means "a general agreement
among the majority of Member states of the Council of Europe about certain
rules and principles identified through comparative research of national and
international law and practice." 230 The critical word in this observation is
"majority" as opposed to true "consensus": the Court has frequently proven
willing to interpret the ECHR on the basis of the consistent practice of most of
the parties, even in view of the contrary practice of the respondent state or other
parties not represented in a given case. In this vein, the Court takes the notion
of interpretation on the basis of practice extraordinarily far. Even if some states

226. Indeed, the Court has proven willing to rely on what it takes to be an "emerging
consensus" when coupled with other considerations. See, e.g., Demir, App. No. 34503/97, Eur. Ct. H.R.
85 ("The consensus emerging from specialised international instruments and from the practice of
Contracting States may constitute a relevant consideration for the Court when it interprets the provisions
of the Convention in specific cases."); Christine Goodwin v. United Kingdom [GC], App. No. 28957/95,
Eur. Ct. H.R. 84-85 (July 11, 2002) (noting a weak but "emerging consensus" among the parties, but
attaching "less importance to the lack of evidence of a common European approach to the resolution of
the legal and practical problems posed, than to the clear and uncontested evidence of a continuing
international trend in favour not only of increased social acceptance of transsexuals but of legal
recognition of the new sexual identity of post-operative transsexuals"). For further discussion, see infra
Subsection V.A.I.
227. See also Al-Saadoon, App. No. 61498/08, Eur. Ct. H.R.; (calan [GC], App. No.
46221/99, Eur. Ct. H.R.; infra Subsection V.A.2.
228. Ocalan [GC], App. No. 46221/99, Eur. Ct. H.R. 163; Loizidou v. Turkey (preliminary
objections) [GC], App. No. 15318/89, Eur. Ct. H.R. 1 37 (Mar. 23, 1995).
229. Christine Goodwin, App. No. 28957/95, Eur. Ct. H.R. 1 74 (citing Stafford v. United
Kingdom [GC], App. No. 46295/99, Eur. Ct. H.R. 1 68 (May 28, 2002)).
230. Dzehtsiarou, supranote 225, at 1733.
338 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

actually protest the interpretation or engage in the opposite practice, the

affirmative practice of a large majority of parties will suffice to establish an
expansive interpretation.23 1
The urtext of European consensus is the perennially cited judgment in
Tyrer v. United Kingdom.232 The case concerned the corporal punishment of
juveniles as provided by law on the Isle of Man, a territory of the United
Kingdom. The question before the Court was whether the specific form of
judicial corporal punishment at issue (known as "birching") constituted
degrading punishment and was thus in contravention of the proscriptive terms
of Article 3 of the ECHR.233 Birching had long been administered on the Isle of
Man, and indeed the Attorney General for the Isle attempted to rest his case on
the argument "that the judicial corporal punishment at issue in this case was not
in breach of the Convention since it did not outrage public opinion in the
Island."234
The Court famously noted that it was not-and could not-be impressed
by such arguments about consistent usages within just one member state. First,
the Court held that the Convention was not stuck in time, and the mere fact that
some state activities may have been permissible in one generation does not
guarantee that the ECHR will not prohibit the same activity in the future. The
Court insisted that it must "recall that the Convention is a living instrument
which . . . must be interpreted in the light of present-day conditions."235 In
words that have echoed throughout the ECtHR's long jurisprudence, the Court
proclaimed that it "cannot but be influenced by the developments and
commonly accepted standards in the penal policy of the member states of the
Council of Europe in this field." 23 6 Even though the practice was long and
overtly considered permissible within this semi-autonomous territory of the
United Kingdom, the Court determined that the Convention had evolved to
proscribe such behavior in light of the clear and consistent practice of the other
member states.
In other words, the Court held that it would look to the practice of the
member states to determine the meaning of the Convention. Where such
practice consistently and overwhelmingly develops in a certain direction, the
Court may consider the Convention as having evolved. Such evolution is
possible even given the clear and consistent contrary practice of a respondent
state in a given case.
Since Tyrer, European consensus has played a significant role in the
Court's development of the Convention. One example is particularly
illuminating: the Court's 1995 judgment on Preliminary Objections in Loizidou

231. Loizidou [GC], App. No. 15318/89, Eur. Ct. H.R. T 79.
232. App. No. 5856/72, Eur. Ct. H.R. (Apr. 25, 1978).
233. Id. T 28; see also ECHR, supra note 23, art. 3 ("No one shall be subjected to torture or to
inhuman or degrading treatment or punishment.").
234. Tyrer, App. No. 5856/72, Eur. Ct. H.R. 31.
235. Id.
236. Id.
2013] Treaty Interpretationand ConstitutionalTransformation 339

v. Turkey.237 The case is notable, first, because the Court explicitly grounds its
reliance on European consensus in Article 31(3)(b) of the VCLT. Indeed the
Court construes Article 31(3)(b) as condoning its approach to interpretation on
the basis of the practice of a large majority of parties, even in the face of
contrary practice by some of them. Second, the Court makes especially clear
that it is willing to ignore not only the contrary practice of a respondent state in
a particular case, but even similar practice among a handful of member states
not party to the proceedings. As Loizidou establishes, in the Court's broad
conception of Article 31(3)(b) it is wholly appropriate to interpret and
reinterpret the Convention on the basis of the common and consistent practice
of most of the states irrespective of the disagreement of the outliers.
Loizidou was, at bottom, a property claim. The applicant, a Cypriot
national, claimed title to several plots of land in Northern Cyprus; she alleged
that she had been prevented from peacefully enjoying her property by the
Turkish military since Turkey's invasion and occupation of Northern Cyprus.238
The pertinent issue, for present purposes, revolves around Turkey's
dispute of the Court's jurisdiction. Turkey claimed that the Court lacked
jurisdiction over Loizidou's claim, inter alia, because it lacked territorial
jurisdiction. In acceding to the ECHR, Turkey deposited two declarations
defining its understanding of its commitments under Articles 25 and 46 of the
Convention (defining the competences of the now-defunct European
Commission on Human Rights and the ECtHR, respectively).239 Turkey
claimed that the Court lacked jurisdiction ratione loci because "in accordance
with their declarations under Articles 25 and 46 ... they had not accepted
either the competence of the Commission or the Court to examine acts and
events outside their metropolitan territory." 240 It was not seriously disputed that
the claims at issue arose outside of Turkey's metropolitan territory (Turkey did
not, after all, claim sovereignty over Northern Cyprus); the case rather turned
on the legal significance of Turkey's interpretive declarations attempting to
limit its commitments under the Convention.241
Turkey strongly contested the notion that there was any consistency in the
parties' practice sufficient to establish an authentic subsequent practice

237. Loizidou v. Turkey (preliminary objections) [GC], App. No. 15318/89, Eur. Ct. H.R. 71
(Mar. 23, 1995).
238. Id. 1 10-ll.
239. Turkey's initial declaration in acceding to Article 25 stated that "(i) the recognition of the
right of petition extends only to allegations concerning acts or omissions of public authorities in Turkey
performed within the boundaries of the territory to which the Constitution of the Republic of Turkey is
applicable." Id. 1 15. Turkey renewed its position in this regard in 1990 and 1993. Id. 25-26. In
addition, Turkey opted in to Article 46 in 1990, declaring then that it would hereby recognize
as compulsory ipso facto and without special agreement the jurisdiction of the European
Court of Human Rights in all matters concerning the interpretation and application of the
Convention which relate to the exercise of jurisdiction within the meaning of Article I of
the Convention, performed within the boundaries of the national territory of the Republic of
Turkey.
Id. T 27.
240. Id. 155.
241. Id.T 65-89.
340 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

demonstrating their agreement that the jurisdictional provisions of the ECHR

were not susceptible to reservation. True, Turkey noted that its interpretive
declaration gave rise to a handful of statements opposing its limitation of the
jurisdictional clauses.242 But in spite of these statements, Turkey contended that
it had not "been established that there is a practice reflecting an agreement
among all ContractingPartiesconcerning the attachment of conditions to these
instruments of acceptance." 243 It strongly emphasized that its declarations were
consistent with the similar practice of other parties, including Cyprus itselfa4
and the United Kingdom,245 both of which claimed limitations to the Court's
jurisdiction. The United Kingdom's declaration in particular included very
similar restrictions on the territorial scope of the Court's jurisdiction, excluding
the state's colonial holdings. 246
The Court firmly rejected Turkey's analysis of the evidentiary criteria for
authentic subsequent practice. The Court as usual began its analysis with a nod
to Tyrer, indicating that even the procedural and jurisdictional provisions of the
Convention "cannot be interpreted solely in accordance with the intentions of
their authors as expressed more than forty years ago." 247 However the Court

242. Id.167.
243. Id. (emphasis added).
244. In opting in to Article 25 in 1988, Cyprus declared
that the competence of the Commission by virtue of Article 25 . . . of the Convention is not
to extend to petitions concerning acts or omissions alleged to involve breaches of the
Convention or its Protocols . . . if the acts or omissions relate to measures taken by the
Government of the Republic of Cyprus to meet the needs resulting from the situation created
by the continuing invasion and military occupation of part of the territory of the Republic of
Cyprus by Turkey.
Id. 30 (quoting Declaration from the Minister for Foreign Affairs of Cyprus, Convention for the
Protection of Human Rights and Fundamental Freedoms, Sept. 2, 1988, available at http://conventions
.coe.int/Treaty/Commun/ListeDeclarations.asp?PO=CYP&NT=005&MA=999&CV=0&NA=Ex-25&CN
=999&VL=1&CM=5&CL=ENG). This is more precisely a restriction of jurisdiction ratione materiae,
but it has an implicit territorial scope.
245. The United Kingdom submitted a declaration when opting into Article 25 in 1966, which it
has successively renewed as required under the provision, providing that
the Government of the United Kingdom of Great Britain and Northern Ireland recognise, in
respect of the United Kingdom of Great Britain and Northern Ireland only and not, pending
further notification, in respect of any other territory for the international relations of which
the Government of the United Kingdom are responsible . .. the competence of the European
Commission of Human Rights to receive petitions . .. [and] the jurisdiction of the European
Court of Human Rights as compulsory.
Id. T 33 (quoting Declarations Contained in Two Letters from the Permanent Representative of the
United Kingdom, Convention for the Protection of Human Rights and Fundamental Freedoms, Jan. 14,
1966, available at http://conventions.coe.int/Treaty/Commun/ListeDeclarations.asp?PO=UK&NT=005
&MA=999&CV=0&NA=Ex-25&CN=999&VL=1&CM=5&CL=ENG). The declaration underscores
this point, adding that the United Kingdom's acceptance of Article 25
does not extend to petitions in relation to anything done or occurring in any territory in
respect of which the competence of the European Commission of Human Rights to receive
petitions has not been recognised by the Government of the United Kingdom or to petitions
in relation to anything done or occurring in the United Kingdom in respect of such a
territory or of matters arising there.
Id.
246. Id. T67.
247. Id. T 71 (adding that its approach under Tyrer v. United Kingdom, App. No. 5856/72, Eur.
Ct. H.R. (Apr. 25, 1978), "is not confined to the substantive provisions of the Convention, but also
2013] Treaty Interpretationand ConstitutionalTransformation 341

acknowledged that interpretation is not a free for all-it must proceed

according to the terms of VCLT article 31. The Court stated that in determining
"whether Contracting Parties may impose restrictions on their acceptance of the
competence of the Commission and Court, . . . [i]t shall also take into account,
together with the context, 'any subsequent practice in the application of the
treaty which establishes the agreement of the parties regarding its
interpretation."'248 On the basis of a highly revealing analysis of the subsequent
practice of the parties, the Court rejected the possibility of unilaterally
restricting the territorial scope of the jurisdiction of either the Commission or
the Court under Articles 25 and 46.
The Court emphasized that "[s]ince the entry into force of the Convention
until the present day, almost all of the thirty Parties to the Convention, apart
from the respondent Government, have accepted the competence of the
Commission and Court to examine complaints without restrictions ratione
loci."249 The Court explicitly considered the exceptions raised by Turkey.
Noting that the Cypriot declaration was eventually withdrawn, the Court turned
to and dismissed the much more relevant declaration by the United Kingdom
regarding Article 25.250 The Court acknowledged that the declaration looks
quite a bit like contrary practice. On its face "the restriction was formulated by
the United Kingdom ... in order to exclude the competence of the Commission
to examine petitions concerning its non-metropolitan territories."251 However, it
notes that this evidence of contrary practice was ultimately not dispositive. In
the critical passage, the Court declares that it
is not called upon to interpret the exact scope of this declaration which has been
invoked by the respondent Government as an example of a territorial restriction.
Whatever its meaning, this declaration and that of Cyprus do not disturb the
evidence of a practice denoting practically universal agreement amongst
Contracting Parties that Articles 25 and 46 ... of the Convention do not permit
territorial or substantive restrictions. 252
In other words, the Court makes it clear that the practice of the United
Kingdom is not, on its own, dispositive whether or not it amounts to a contrary
practice confirming the respondent's argument that there exists no common and
concordant practice evidencing the agreement among all the parties. The
"practically universal" practice of the parties suffices to ground an
interpretation of the competences of the Commission and Court under Articles
25 and 46 in spite of the expressly contrary practice of a small handful of
states. 253

applies to those provisions, such as Articles 25 and 46 (art. 25, art. 46), which govern the operation of
the Convention's enforcement machinery").
248. Id. 73.
249. Id. 79 (emphasis added).
250. Id. 79-80.
251. Id. 80.
252. Id. (emphasis added).
253. The Court thus considers "the restrictions ratione loci attached to Turkey's Article 25 and
Article 46 (art. 25, art. 46) declarations . .. invalid." Id. 89. Indeed, the Court ultimately went so far as
to rule that "the impugned restrictions can be separated from the remainder of the text leaving intact the
342 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

Thus the first major point of breadth in the Court's approach to

subsequent practice goes to evidentiary criteria. The ECtHR has relied on
European consensus to develop a wide variety of substantive Convention
rights, 254 as well as to develop provisions relating to its competences,
jurisdiction and procedure.255 Often the Court simply relies on trends in
256
practice. As Loizidou makes clear, the Court will not be dissuaded from
relying on the practice of most of the parties to establish an interpretation
binding on all of them, even in the face of clear and consistent contrary practice
by the respondent state257 and/or a handful of others.258
It must be re-emphasized that the Court does not by any means tend to
overturn the margin through reliance on consensus or near-consensus among
the parties; it is common for the Court to come to precisely the other result, 259
occasionally in surprising ways. 260 Critically important, however, is that the

acceptance of the optional clauses" and thus "contain valid acceptances of the competence of the
Commission and Court." Id. 97-98.
254. For example, the Court has relied upon European Consensus to develop Article 3 in
Ocalan v. Turkey [GC], App. No. 46221/99, Eur. Ct. H.R. 1 164 (May 12, 2005) (regarding the death
penalty); and Selmouni v. France [GC], App. No. 25803/94, Eur. Ct. H.R. 101 (July 28, 1999)
(regarding the definition of torture); to develop the Article 6 right to a fair trial in Airey v. Ireland, App.
No. 6289/73, Eur. Ct. H.R. 126 (Oct. 9, 1979) (regarding the right to free legal aid in civil cases); and to
develop the Article 8 right to a private life in Christine Goodwin v. United Kingdom [GC], App. No.
28957/95, Eur. Ct. H.R. 74 (July 11, 2002) (regarding the entitlement of post-operation transsexuals to
legal recognition of their gender reassignment). For further discussion, see Dzehtsiarou, supra note 225,
at 1731-32.
255. Loizidou v. Turkey (preliminary objections) [GC], App. No. 15318/89, Eur. Ct. H.R. 71
(Mar. 23, 1995).
256. See, e.g., Christine Goodwin [GC], App. No. 28957/95, Eur. Ct. H.R. 84-85; Loizidou
(preliminary objections) [GC], App. No. 15318/89, Eur. Ct. H.R. T 75-79.
257. See, e.g., Demir v. Turkey [GC], App. No. 34503/97, Eur. Ct. H.R. J 52, 85 (Nov. 11,
2008) (interpreting the freedom of association in Article 11 as applying to public workers on the basis of
the subsequent practice of the "vast majority" of member states, despite Turkey's clear contrary
practice); Tyrer v. United Kingdom, App. No. 5856/72, Eur. Ct. H.R. (Apr. 25, 1978).
258. See, e.g., Al-Saadoon v. United Kingdom, App. No. 61498/08, Eur. Ct. H.R. 120 (Mar.
2, 2010); (Scalan [GC], App. No. 46221/99, Eur. Ct. H.R. 1 163; Loizidou (preliminary objections)
[GC], App. No. 15318/89, Eur. Ct. H.R. 179.
259. See, e.g., StObing v. Germany, App. No. 43547/08, Eur. Ct. H.R. T 61 (Apr. 12, 2012)
(finding "no consensus between the member States as to whether the consensual commitment of sexual
acts between adult siblings should be criminally sanctioned"); Sahin v. Turkey [GC], App. No.
44774/98, Eur. Ct. H.R. T 109 (Nov. 10, 2005) (noting that "[i]t is not possible to discern throughout
Europe a uniform conception of the significance of religion in society," and as a result "the role of the
national decision-making body must be given special importance . . . when it comes to regulating the
wearing of religious symbols in educational institutions, especially . . . in view of the diversity of the
approaches taken by national authorities on the issue").
260. For example, in considering Ireland's ban on abortion, after finding a sufficiently large
consensus against the practice, the Court nevertheless resorted to the margin in light of the particularly
fraught moral and political considerations surrounding the question. A, B & C v. Ireland [GC], App. No.
25579/05, Eur. Ct. H.R. 235-37 (Dec. 16 2010) ("[T]he Court considers that there is indeed a
consensus amongst a substantial majority of the Contracting States of the Council of Europe towards
allowing abortion on broader grounds than accorded under Irish law .... However, the Court does not
consider that this consensus decisively narrows the broad margin of appreciation of the State."); see also
Sahin, App. No. 44774/98, Eur. Ct. H.R. T 3 (Tulkens, J., dissenting) (refusing to accept the "the
margin-of-appreciation approach" on the basis of the "diversity of practice between the States on the
issue of regulating the wearing of religious symbols in educational institutions and, thus, the lack of a
European consensus in this sphere"). Judge Tulkens objected that the majority took an unjustifiable view
of the comparative law, for "in none of the member States has the ban on wearing religious symbols
2013] Treaty Interpretationand ConstitutionalTransformation 343

Court does not clearly demarcate between cases where consensus is or is not
sufficient to overcome the margin in a principled way. By avoiding abstract
statements about the kind, amount, or degree of practice necessary to establish
a European consensus, the Court maintains a gatekeeping role for determining
the scope and evolution of Convention rights.

2. Expansive Potential:Interpretation,Reinterpretation,and
Modification

The second peculiarity of the doctrine of European consensus as a flavor

of subsequent practice concerns the extent of its potential to establish treaty
change over time. Recall that under general international law the expansive
potential of subsequent practice is hotly contested. A few authorities, including
some arbitral awards, have considered clear subsequent practice to be capable
of establishing a modification. But such cases are few and far between, and
generally concern bilateral agreements. Even while straining the text in both
Namibia and Wall, the ICJ has balked at explicitly referring to its
interpretations of the Charter as "modifications."261 The ECtHR, by contrast,
expressly considers subsequent practice capable of establishing a modification
of the ECHR, even in plain view of contrary practices among a minority of the
parties. The Court's reliance on the technique to interpret, reinterpret, and
successively modify the terms of the ECHR is particularly vivid in a trilogy of
opinions on the legality of capital punishment under the Convention: Soering v.
262 .. 263
United Kingdom, Ocalan v. Turkey, and Al-Saadoon v. United
Kingdom.264
The relevant question in each of the three cases concerned the legality of
the death penalty under the ECHR. Each of the cases turned on the
interpretation of Articles 2 and 3. The latter provides in categorical terms that
"no one shall be subjected to torture or to inhuman or degrading treatment or
punishment."265 Article 2 specifically protects the right to life, however its
terms explicitly leave room for capital punishment by law. Article 2(1)
provides that "everyone's right to life shall be protected by law. No one shall
be deprived of his life intentionally save in the execution of a sentence of a
courtfollowing his conviction of a crimefor which this penalty is provided by
law."266 In each case a claim was raised-at first on the applicant's behalf, 267

extended to university education, which is intended for young adults, who are less amenable to
pressure." Id.
261. Legal Consequences of the Construction of a Wall in the Occupied Palestinian Territory,
Advisory Opinion, 2004 I.C.J. 136, 27-28 (July 9); Legal Consequences for States of the Continued
Presence of South Africa in Namibia (South West Africa) Notwithstanding Security Council Resolution
276 (1970), Advisory Opinion, 1971 I.C.J. 16, T 22 (June 21).
262. Soering v. United Kingdom, App. No. 14038/88, Eur. Ct. H.R. (July 7, 1989) (plenary).
263. Ocalan v. Turkey [GC], App. No. 46221/99, Eur. Ct. H.R. (May 12, 2005).
264. Al-Saadoon v. United Kingdom, App. No. 61498/08, Eur. Ct. H.R. (Mar. 2, 2010).
265. ECHR, supranote 23, art. 3.
266. Id. art. 2(1) (emphasis added).
267. Although Soering did not himself suggest that the death penalty per se violated Article 3,
Amnesty International intervened on his behalf, submitting that "the evolving standards in Western
344 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

and in the latter two instances by the applicant himself268-that the death
penalty had become so offensive to European mores as to violate Article 3,
despite the express permissive terms of Article 2(1).
Over the course of these three cases, the Court determined that the
permissive language of Article 2(1) had been modified by the subsequent
practice of the member states so as to no longer provide an exception to the
right to life for capital punishment of any kind, thereby clearing the way for an
evolutive interpretation of Article 3 prohibiting the death penalty on the basis
of the subsequent practice of the parties. 269
Soering v. United Kingdom concerned a German national who was
detained in the United Kingdom pending extradition to the United States to face
double-homicide charges in Virginia. 270 In relevant part, the Court considered
an intervention on Soering's behalf by Amnesty International, concerning the
inconsistency of the death penalty with Article 3. Amnesty submitted that "the
evolving standards in Western Europe regarding the existence and use of the
death penalty required that capital punishment should now be considered as an
inhuman and degrading punishment within the meaning of Article 3."271
The Court noted up front that "[c]apital punishment is permitted under
certain conditions by Article 2(1) of the Convention."272 At the same time, the
Court cited Tyrer to the effect that it "cannot but be influenced by the
developments and commonly accepted standards in the penal policy of the
member states of the Council of Europe." 273 The Court noted sweeping changes
in the legal practices of the parties since the ECHR had come into force.
De facto the death penalty no longer exists in time of peace in the Contracting
States to the Convention. In the few Contracting States which retain the death
penalty in law for some peacetime offences, death sentences, if ever imposed, are
nowadays not carried out. This virtual consensus ... is reflected in Protocol No. 6
to the Convention, which provides for the abolition of the death penalty in time of
peace. Protocol No. 6 was opened for signature in 1983, which in the practice of the
Council of Europe indicates the absence of objection on the part of any of the
Member States of the Organisation; it came into force in March 1985 and to date
has been ratified by thirteen Contracting States to the Convention, not however
274
including the United Kingdom.
The obvious question was thus how to square text and practice: on the
one hand Article 2(1) provides a clear and rigid window for states to impose
capital punishment by law; on the other hand, clear and consistent practice had

Europe regarding the existence and use of the death penalty required that the death penalty should now
be considered as an inhuman and degrading punishment within the meaning of Article 3." Soering, App.
No. 14038/88, Eur. Ct. H.R.I 10 1 (plenary).
268. Al-Saadoon, App. No. 61498/08, Eur. Ct. H.R. 1 102; Ocalan [GC], App. No. 46221/99,
T 157.
269. Al-Saadoon, App. No. 61498/08, Eur. Ct. H.R. 1120.
270. Soering, App. No. 14038/88, Eur. Ct. H.R. IT 11-13.
271. Id. 101.
272. Id.
273. Id. T 102 (citing Tyrer v. United Kingdom, App. No. 5856/72, Eur. Ct. H.R. 1 31 (Apr. 25,
1978)).
274. Id. (internal quotation marks omitted).
2013] Treaty Interpretationand ConstitutionalTransformation 345

emerged, indicating that the membership was coming to consider the death
penalty to be inconsistent with Article 3.275 The Court explained that the
question of whether "these marked changes have the effect of bringing the
death penalty per se within the prohibition of ill-treatment under Article 3 must
be determined on the principles governing the interpretation of the
Convention."276 In no uncertain terms, the Court explained that its approach to
interpretation might sometimes privilege practice over even the clearest textual
provision:
Subsequent practice in national penal policy, in the form of a generalised abolition
of capital punishment, could be taken as establishing the agreement of the
Contracting States to abrogate the exception provided for under Article 2 §
1 ... and hence to remove a textual limit on the scope for evolutive interpretation of
Article 3.277
At this stage the Court contented itself to register its view in obiter dictum. The
Court emphasized the fact that the parties had opted to rely on the normal
278
provisions for amending the ECHR by formal protocol2. In view of the
promulgation of Protocol No. 6, and its several signatories to date, the Court
considered that "the intention of the Contracting Parties as recently as 1983 was
to adopt the normal method of amendment of the text in order to introduce a
,,279
new obligation to abolish capital punishment in time of peace. Further, the
Protocol reflects the parties' choice to pursue change "by an optional
instrument allowing each State to choose the moment when to undertake such
an engagement."280 As a result the Court held that at this time Article 3 could
not be interpreted as generally prohibiting the death penalty. 281
Sixteen years later, in Ocalan v. Turkey, a new applicant followed up on
the Court's dicta in Soering, arguing that the parties had, by their practice
through the ensuing years, "abrogated the exception set out in the second
sentence of Article 2 § 1 of the Convention and that the death penalty
constituted inhuman and degrading treatment within the meaning of Article
3Y'282
The Court recalled that in Soering it had found the states' evident
283
intention to rely on the normal amendment rule to be dispositive. It noted,

275. Id. T 103.

276. Id. 102.
277. Id. 103 (emphasis added).
278. Id. 102 (citing ECHR, supra note 23, Protocol No. 6).
279. Soering, App. No. 14038/88, 103.
280. Id.
281. Id. The Court ultimately ruled that Mr. Soering's extradition would violate Article 3 by
subjecting the applicant to the extreme emotional hardship attendant to capital punishment in the United
States, and would for that reason be nevertheless impermissible. In the Court's view,
having regard to the very long period of time spent on death row in such extreme conditions,
with the ever present and mounting anguish of awaiting execution of the death
penalty ... the applicant's extradition to the United States would expose him to a real risk
of treatment going beyond the threshold set by Article 3.
Id.T 111.
282. Ocalan v. Turkey [GC], App. No. 46221/99, Eur. Ct. H.R. $ 157 (2005).
283. Id. T 163.
346 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

however, that "the legal position as regards the death penalty has undergone a
considerable evolution since Soering was decided." 28 It emphasized that the
"de facto abolition" noted in Soering in respect of twenty-two Contracting
States in 1989 had "developed into a de jure abolition in forty-three of the
forty-four Contracting States and a moratorium in the remaining State that has
not yet abolished the penalty, namely Russia." 285 Furthermore, all of the parties
had now signed Protocol No. 6, and forty-one had already ratified it.286 In light
of what it took to be such overwhelming practice, the Court decided it did not
need to await ratification by the three remaining States. It proceeded to
conclude that the death penalty exception in Article 2(1) had been adequately
modified, explaining that "[a]gainst such a consistent background, it can be
said that capital punishment in peacetime has come to be regarded as an
unacceptable ... form of punishment that is no longer permissible under
Article 2."287
Thus the Court no longer considered itself bound to respect the parties'
choice to rely on formal amendment. In light of the overwhelming tendency of
their practice, (to which the optional Protocol obviously contributed), the Court
was willing to jump ahead and find Article 2(1) modified so as to no longer
permit the death penalty in peacetime.
Nevertheless, the case concerned the death penalty in time of war. The
Court found that the parties' practice was significantly less imposing with
regard to capital punishment in wartime and remained insufficient to abrogate
the permissive clause of Article 2(1) in that circumstance. The Court
emphasized, as it had in Soering, that "by opening for signature Protocol No.
13 concerning the abolition of the death penalty in all circumstances, the
Contracting States have chosen the traditional method of amendment of the text
of the Convention in pursuit of their policy of abolition."288 By the date of its
judgment, three states had yet to sign Protocol No. 13, and sixteen had yet to
ratify it. The Court noted that "this final step towards complete abolition of the
death penalty .. . can be seen as confirmation of the abolitionist trend in the
practice of the Contracting States." 289 However it left its analysis there, stating
only obliquely that:

284. Id.
285. Id. (quoting Ocalan v. Turkey, App. No. 46221/99, Eur. Ct. H.R. T 195 (Mar. 3, 2003)).
286. Id.
287. Id.
288. Id. 1164
289. Id. It is worth noting, here, that the Court gives some weight to a very different kind of
practice in determining whether a European consensus has emerged as to the meaning of the ECHR. Not
only does the Court look at what the parties actually do or do not prohibit in their domestic law, but also
to their undertakings on the supranational level (e.g., in considering the numbers of signatures and
ratifications of the optional Protocol No. 13). It would seem as though the Court considers these
supranational commitments at least relevant to the determination of whether a sufficient level of "near-
consensus" has emerged to justify developing the Convention. It might be argued that this gives a
signature more legal significance (as "practice") than it would normally have under the law of treaties.
See VCLT, supra note 8, art. 18 (obliging a signatory only "to refrain from acts which would defeat the
object and purpose of [the] treaty" where the agreement requires ratification to become binding).
2013] Treaty Interpretationand ConstitutionalTransformation 347

For the time being, the fact that there is still a large number of States who have yet
to sign or ratify Protocol No. 13 may prevent the Court from finding that it is the
established practice of the Contracting States to regard the implementation of the
death penalty as inhuman and degrading treatment contrary to Article 3 of the
Convention, since no derogation may be made from that provision, even in times of
war. 290
Resting once more on dicta, the Court found it was "not necessary ... to reach
any firm conclusion on these points" since the respondent would have violated
Article 2 in any event by imposing the death penalty at the culmination of an
unfair trial. 291
Five years later, and twenty-six years after Soering, the Court finally
vitiated the permissive clause of Article 2(1) in Al-Saadoon & Mufdhi v. United
Kingdom. The Court noted that "Ocalan did not exclude that Article 2 had
already been amended so as to remove the exception permitting the death
penalty." 292 On the basis of intervening practice, it emphasized that "the
position has evolved" once again.
According to the Court, the subsequent practice of the parties over the
span of the Convention's long history overwhelmingly pointed in one direction.
All forty-seven member states had, by 2010, signed Protocol No. 6, and all
except for Russia had ratified it.294 In the intervening five years since Ocalan,
ratifications of Protocol 13 had jumped to 42 member states; it had been signed
but not ratified by a further three (Armenia, Latvia and Poland). Indeed, it was
signed by the respondent, "the United Kingdom, on 3 May, 2002, ratified on 10
October, 2003 and entered into force in respect of that state on 1 February,
2004."295 The Court somewhat dismissively acknowledged evidence of
contrary practice, observing that "Azerbaijan and Russia are alone in not
having signed the Protocol." 296 In spite of the fact that the parties had evidently
continued to rely on the formal amendment rule-given so much weight in
Soering and even in Ocalan-the Court considered the trend as having finally
crystalized on the basis of sufficiently overwhelming practice.
These figures, together with consistent State practice in observing the moratorium
on capital punishment, are strongly indicative that Article 2 has been amended so as
to prohibit the death penalty in all circumstances. Against this background the Court
does not consider that the wording of the second sentence of Article 2 § 1 continues
to act as a bar to its interpreting the words 'inhuman or degrading treatment or
297
punishment' in Article 3 as including the death penalty.
On the basis of practice subsequent to Ocalan, the Court finally interpreted
Article 2 as being thoroughly amended so as to no longer permit the death

290. Ocalan [GC], App. No. 46221/99, Eur. Ct. H.R. 165.
291. Id.
292. AI-Saadoon v. United Kingdom, App. No. 61498/08, Eur. Ct. H.R. T 120 (Mar. 2, 2010).
293 Id.
294. Id. T116.
295. Id.T117.
296. Id.
297. Id. T 120.
348 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

penalty under any circumstances-in peacetime or in times of war.298

Henceforth, the Court thus held, capital punishment would be prohibited in all
contexts under an evolutive interpretation of Article 3.
Taken together Soering, Ocalan, and AI-Saadoon illuminate several
aspects of the Court's approach to interpreting the Convention on the basis of
the subsequent practice of the member states. First, and most obviously, the
cases stand for the proposition that the Court will rely on the subsequent
practice of the parties to interpret, reinterpret, and even successively modify the
ECHR. Second, the Court has twice indicated its hesitation to modify the
Convention on the basis of informal practice while the parties appear to have
opted to amend the Convention through formal procedures-especially by
optional protocol, binding only upon those who opt-in.299 However, the Court
has also twice decided to stop waiting for the relevant Protocols to be signed or
ratified unanimously, choosing instead to rely on the states' conduct in signing
and ratifying the Protocols alongside their national legislative and executive
practices as further evidence of a practice of sufficient weight as to establish an
amendment of the Convention. 300
Finally, the Court confirmed throughout the death penalty cases that it
would rely on mere near-consensus to modify the Convention even in the face
of contrary practice, as it does in cases of interpretation where modification is
not at issue (like Tyrer and Loizidou). The Court does rely on quite a high
degree of practice in icalan and Al-Saadoon. But it leaves completely open the
question of how much practice is necessary to establish a European consensus
sufficient to modify the plain text of the Convention where the parties had not
opted to employ formal amendment procedures.

B. "European Consensus " as a TransformationofJudicialPower

In terms of form and content, the ECtHR's approach to subsequent

practice differs markedly from that of the ICJ. The World Court relies on organ
practice as a tenuous proxy for the practice of the parties, and has in many ways
proven insensitive to the difficulty of establishing state consent to its
interpretations. By contrast, the ECtHR looks directly at the practice of the
parties, but has frequently expressed its willingness to rely on majority practice
even in the face of the clear and consistent contrary practice of a handful of
outlier parties-thus even more willfully ignoring the traditional notion that
authentic subsequent practice should establish the agreement of all of the

298. Id.
299. Soering, App. No. 14038/88, Eur. Ct. H.R. 103 (refusing to interpret Article 2(1) as
amended on the basis of practice in light of the parties' choice to amend the provision through Protocol
No. 6); Ocalan v. Turkey [GC], App. No. 46221/99, Eur. Ct. H.R. 164 (May 12, 2005) (refusing to
interpret Article 2(1) as prohibiting the death penalty in all circumstances, in light of the parties' attempt
to amend the Convention via the optional Protocol No. 13).
300. Ocalan [GC], App. No. 46221/99, Eur. Ct. H.R. T 165 (finding the Convention as having
been modified to prohibit the death penalty in peacetime in spite of the non-unanimous adherence to
Protocol No. 6); AI-Saadoon, App. No. 61498/08, Eur. Ct. H.R. T 120 (finding the Convention as having
been modified to bar capital punishment in all circumstances in spite of the non-unanimous adherence to
Protocol No. 13).
2013] Treaty Interpretationand Constitutional Transformation 349

parties. Further, while the ICJ has proven willing to contort the text of the
Charter in relying on organ practice, the ECtHR has gone further in explicitly
relying on European consensus to establish the modification of the clear text of
the Convention-again even in plain view of the contrary practice of a handful
of member states. Yet in spite of their differences, a critical conceptual
similarity remains: like the World Court, the ECtHR's approach to subsequent
practice has materially and dramatically transformed its constitutional position,
rendering it significantly more autonomous from the membership. Even more
clearly than the ICJ's interpretive principle, the ECtHR's assertion of judicial
competence strains the bonds of consent between the organization and its
constituent states parties-the nominal "masters of the treaty."
Recall that treaty interpretation can contribute to a constitutional
transformation in two ways: directly, through particular substantive
interpretations; or in a second-order sense, connected to the overall method of
interpretation.
Some of the Court's substantive interpretations fit into the first-order type
of transformation in that they involve the direct expansion or reinterpretation of
an organ's competences or procedures as expressed or implied by its
constituent instrument.30 1 Loizidou fits into this category, insofar as the Court
interpreted the provisions in Articles 25 and 46 as not amenable to reservations
concerning the territorial scope of the Court's jurisdiction (as well as that of the
Commission).302 The Court construed those provisions as empowering it to
hold a state party's interpretive declaration on territorial scope to be invalid,
and further as entitling it to sever that invalid reservation from the state's
general accession to the jurisdiction of the Court-thus preserving its
jurisdiction over the respondent.303 However, most of the ECtHR's
jurisprudence concerns the scope of the substantive rights in the ECHR, and not
the more organizational features of the Convention.m While these
interpretations are often expansive, and in many cases explicitly evolutive, such
rights-expansion does not necessarily entail a reordering of the organization's
constitution. 305 Rights expansion (and limitation) is indeed a normal and
necessary part of the function of all courts charged with rights-adjudication. 306

301. Recall the analogous case of the ICJ's interpretation of the competences of the UNGA in
Wall. See supra Part IV.
302. Loizidou v. Turkey (preliminary objections) [GC], App. No. 15318/89, Eur. Ct. H.R. 171
(Mar. 23, 1995).
303. Id. T 97.
304. But see Mamatkulov v. Turkey [GC], App. Nos. 46827/99, 46951/99, Eur. Ct. H.R. TT
111, 123-24 (2005).
305. Arguably the interpretation of the contours and limits of substantive rights could amount to
a transformation where such rights are expanded to such an extent as to create altogether new rights.
306. See, e.g., J.H.H. WEILER, Fundamental Rights and Fundamental Boundaries: On the
Conflict of Standards and Values in the Protection of Human Rights in the EuropeanLegal Space, in
THE CONSTITUTION OF EUROPE 102 (1999); George Letsas, Strasbourg'sInterpretiveEthic: Lessons for
the InternationalLawyer, 21 EUR. J. INT'L L. 509 (2010); Lucas Lixinski, Treaty Interpretationby the
Inter-American Court of Human Rights: Expansionism at the Service of the Unity of InternationalLaw,
21 EuR. J. INT'L L. 585 (2010).
350 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

However the Court's method of interpretation has had a much more

dramatic constitutional effect in the second-order sense. Implicit in the
ECtHR's reliance on subsequent practice is an assertion of a significant judicial
power-one that has transformed the relationship between the Court and the
member states by diluting the parties' mastery over the treaty.
Through its interpretive practice, the Court has come to assert the
competence to literally modify the Convention on the basis of the consistent
practice of a mere majority of states parties, even in the face of open, active,
and longstanding contrary practice of a minority. This assertion of judicial
competence under 31(3)(b) (qua European consensus) seriously erodes the
bonds of consent tethering the Court to each member state individually. By
broadly construing its mandate under VCLT 31(3)(b) (and despite that
provision's voluntaristic pedigree), the Court has successfully transformed its
constitutional position vis-A-vis the parties-empowering itself to hold them to
an evolving Convention capable of substantial modification even without their
unanimous consent. This is not to say that the Court has rendered itself a
completely autonomous lawmaker; its assertion still depends on the practice of
some of the parties, and usually a substantially qualified majority. But the fact
that it is willing to ignore the dissenting minority's evident lack of consent in
establishing a modification of the Convention through interpretation represents
a significant step. Moreover, the Court has been careful not to make a numbers
game out of the doctrine of European consensus-it has never spelled out an
amount or percentage of parties that would suffice to establish a particular
interpretation, reinterpretation, or modification. It has carefully left the matter
to its own discretion, thus enabling it to determine on a case-by-case basis
whether or not a sufficient degree of practice has crystallized into a new
interpretation binding on any holdouts.
The constitutional meaning of the ECtHR's interpretive practice might be
distinguished from that of the ICJ on one level: while the latter mostly
empowers other organs of the organization, the former is more clearly court-
empowering. As noted above, the ICJ lacks compulsory jurisdiction over states,
and its advisory opinions lack binding effect. The full constitutional meaning of
its approach to interpretation is only manifest in the practice of the more
powerful organs like the UNGA and UNSC, acting under the aegis of the ICJ's
interpretive blessing.307 By contrast the ECtHR's assertion represents a
transformation of its own judicial power vis-A-vis the parties. The Court may
not always or even generally rely on the full extent of this competence, on the
basis of a variety of doctrinal considerations or political calculations; what is
remarkable and transformative is that the Court asserts the competence to
develop and modify the Convention on the basis of contested majoritarian
practice at all, while maintaining its discretion to decide when and when not to
do so. And unlike many other international courts and tribunals, the Court has
no shortage of opportunities to reevaluate its interpretations of the Convention
over time.

307. See supra Section IV.B.

2013] Treaty Interpretationand Constitutional Transformation 351

At the highest level of abstraction, the interpretive approaches of the ICJ

and ECtHR have had common constitutional effects. Both have asserted
extraordinarily broad approaches to interpretation on the basis of subsequent
practice, capable of establishing thoroughgoing changes to their respective
constituent instruments on the basis of (at best) tenuous evidence of the consent
of their respective parties. In so doing, both have opened their respective
constitutional systems to a second track of informal change through a
combination of practice and interpretation, parallel to, but more flexible than,
the normal procedures for formal amendment. Both Courts have thus
contributed, in different ways and to different degrees, to the transformation of
their constitutions. As a result of their interpretive judicial practices, the ECtHR
(as a judicial organ) and the U.N. organization as a whole have grown more
autonomous, and more capable of binding the states that created them.
From the juridical perspective of the law of treaties, the ECtHR's
approach to interpretation appears to be broad, but not unimaginably broad-
just as in the case of the ICJ's reliance on organ practice. In the scholarship, the
breadth of the ECtHR's approach to practice tends to be cabined as the
idiosyncratic approach of a special regime for the protection of human rights.308
From the perspective of political theory, however, it appears that the technique
of interpretation has played an important role in transforming the ECtHR as a
constituted supranational judicial body with significant governance functions
on the international plane. Through its interpretive practice, the Court has
successfully accrued to itself the ability to develop and even modify the
ECHR-a multilateral, consent-based treaty-on the basis of the contested
practice of a majority of the parties. Article 31(3)(b) was and remains the
crucible for this exemplar of judicial alchemy.

VI. CONCLUSION: LEGITIMATE AUTHORITY, ACCOUNTABILITY, AND THE

AUTONOMY OF INFORMAL TRANSFORMATION

By bringing together the juridical and the political, the constitutional

frame sheds light upon the meaning of interpretive practices within
international organizations-meaning that remains obscure from the doctrinal
perspective of the law of treaties alone. Viewed in purely juridical terms, the
ICJ and the ECtHR have developed broad approaches to the interpretation of
their constituent instruments. They stretch the ambit of VCLT Article 31(3)(b)
in interpreting the U.N. Charter and the ECHR. Still, from the perspective of
the law of treaties, their expansive approaches can be explained away as two
sets of special interpretive rules characteristic of two special treaty regimes.309
But from the political point of view, their interpretive practices appear much
more vibrant and meaningful. It is of the highest political significance that
these Courts have developed such broad techniques in the interpretation of their

308. See, e.g., MAGDALENA FOROWICZ, THE RECEPTION OF INTERNATIONAL LAW IN THE
EUROPEAN COURTOF HUMAN RIGHTs 372 (2010); Letsas, supra note 306.
309. See GARDINER, supra note 14, at 247.
352 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

own constituent instruments.3o In so doing, these constituted judicial bodies

have transformed the composition and capacities of their respective
organizations. What matters most, at this altitude, is that these Courts'
interpretive practices have enhanced the material power and autonomy of the
organizations to which they belong vis-i-vis the states parties. What appears as
expansive treaty interpretation from one point of view thus appears as dramatic
constitutional transformation from another.
As a normative matter, informal constitutional change is not necessarily a
bad thing. Sometimes transformation brings happy results, and in some
contexts it may even be necessary. The possibility of transformation through
interpretation should not be rejected out of hand, nor should the practices of the
ICJ and the ECtHR be subjected to bludgeons like the label of judicial activism.
Rather, I want to suggest that informal changes must primarily be assessed on a
case-by-case basis, in light of the particularities of the very different
constitutional orders within which they occur.
Constitutional interpretation in the WTO cannot be measured by the same
metric as would be appropriate for the United Nations, nor can either be
evaluated in the same way as the interpretive practices of the ECtHR. The
functions of each treaty organization must be taken into account. For example,
the informal expansion of mechanisms for the regional protection of
individuals' human rights may be desirable in an international order populated
by states. But the informal empowerment of organs charged with the
maintenance of international peace and security may be more troubling-as in
the case of the UNSC's arrogation of the power to order states to freeze the
assets of specific individuals suspected of financing international terrorism. 311
And of course the desirability of informal change through interpretation may
again be very different in the context of the WTO-AB, in light of its function of
stewardship over a web of technical rules for the liberalization of trade. All
constitutional transformation entails the reordering of power within a system of
governance, either among the constituted bodies or as against the governed; the
merits of any such reconfiguration can only be appropriately judged in relation
to the specific function and powers of the particular organization in question.
Further militating in favor of a case-by-case approach is the fact that, in
the context of international organizations, constitutional transformation can
also affect relations of power among the parties-and in innumerable context-
driven ways. The ICJ's ascription of normative weight to organizational
practice greatly empowers those states capable of steering the political bodies,
for example the P5 in the case of the UNSC, or those states in a position of
influence within the UNGA. Likewise, the ECtHR's willingness to rely on

310. Indeed, when charged with the interpretation of unrelated treaties the ICJ has actually
adopted a rather strict approach to subsequent practice. See Kasikili/Sedudu Island (Botswana/Namib.),
1999 I.C.J. 1045, T 74 (Dec. 13) (requiring determinacy in the practice of the parties as evidencing an
interpretation of the Treaty; a condition which the I.C.J. did not view as necessary in relying on "organ
practice" in Certain Expenses).
311. See, e.g., Thomas Franck, The "Powers of Appreciation ": Who Is the Ultimate Guardian
of UN Legality?, 86 AM. J. INT'L L. 519, 520 (1992).
2013] Treaty Interpretationand ConstitutionalTransformation 353

majority practice may enhance the power of some blocs against others,
although the effect seems weak insofar as the Court remains the ultimate
gatekeeper as to whether a new consensus has emerged. This is not to say that
the transformations described above ultimately boil down to reconfigurations of
power among the member states. To the contrary, the reordering of power
among particular members occurs alongside (and indeed through) the
empowerment of the larger organization over the membership as a whole.
Rather the point is that the varied ways in which informal change shifts power
among their member states further complicates any general assessment of the
phenomenon in the context of international organizations.
At the same time, informal tracks of constitutional change are inherently
likely to give rise to certain general problems that deserve special emphasis.
Whatever the desirability of particular changes in particular organizations, it is
important to bear in mind certain risks. Two problems in particular share a
special affinity with the informal transformation of public international
organizations. The first is a problem of legitimate authority, stemming from the
autonomy of transformation through constitutional interpretation. The second is
a problem of accountability, arising out of the informality of constitutional
change through interpretation in the face of formal procedures of amendment.
Both problems are all the more serious in light of the exponential growth in the
delegation of public power to transnational institutions in recent years. By way
of conclusion, the following paragraphs briefly address each problem in turn.
The fundamental tenet of the law of treaties is, of course, the consent of
the parties. Thus we see the perennial recital that the parties always remain the
masters of their treaties,312 not least in the context of treaty interpretation over
time.3 This principle grounds the legitimacy of any treaty-based organization.
And yet the ICJ's and ECtHR's reliance on broad modalities of subsequent
practice belie the parties' mastery, and indeed fly in the face of the traditional
doctrine of "state consent above all." The law of treaties does envision the
expansive interpretation of treaties over time, but not absent some basis in
consent. In the context of a treaty-based organization, the erosion of consent
through dynamic constitutional interpretation can give rise to a serious problem
of legitimation. The glosses on subsequent practice by the ICJ (practice of the
organization) and the ECtHR (European consensus) may differ in their
contours, but they similarly strain the bonds of state consent from which these
bodies derive their treaty-based authority in the first place.
This consent-based problem of legitimate authority is not merely
speculative; it has arisen at various milestones in our story of transformation.
The problem was raised most eloquently by Judge Percy Spender of the ICJ at
the very beginning. Writing separately in Certain Expenses, he decried the

312. See the opinion of the German Constitutional Court in its Decision on the Treaty of
Lisbon, Bundesverfassungsgericht (BVerfG) (Federal Constitutional Court), June 30, 2009, 2 be 2/08,
235, http://www.bverfg.de/entscheidungen/es20096302bveOO0208en.html
313. See VILLIGER, supra note 12, at 46-47.
354 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

conversion of (contested) organ practice to "subsequent practice" as alchemical

heresy:
The Charter establishes an Organization. The Organization must function through
its constituted organs. The functions and authorities of those organs are set out in
the Charter. However the Charter is otherwise described the essential fact is that it
is a multilateral treaty. It cannot be altered at the will of the majority of the Member
States, no matter how often that will is expressed or asserted against a protesting a
minority and no matter how large be the majority of Member States which assert its
314
will in this manner or how small the minority.

In Spender's view, the Court's reliance on non-unanimous UNGA resolutions

in interpreting the Charter is of itself an impermissible transgression of the
voluntaristic spirit of the law of treaties. Judge Fitzmaurice continued the
charge at the next stage in his dissent in Namibia, insisting that the possibility
of modification only compounds the problem of consent underlying the ICJ's
interpretive jurisprudence.315 Similar worries resound in the separate opinions
316
and dissents of disappointed judges of the ECtHR, with echoes of their
protestations rippling throughout the high courts of the CoE member states. 317
International organizations like the United Nations and the ECtHR are
critical in the porous world of the twenty-first century. It is admittedly
important that such governance institutions enjoy a degree of flexibility in
exercising their powers. But taken far enough, the casual treatment of consent
may weaken the authority of these crucial organizations. In some cases the risk

314. Certain Expenses of the United Nations, 1962 I.C.J. 151, 196 (July 20) (separate opinion
of Judge Spender); see also id. at 189-90 ("Nor can I agree . . .that a common practice pursued by an
organ of the United Nations, though ultra vires and in point of fact having the result of amending the
Charter, may nonetheless be effective as a criterion of interpretation.").
315. Legal Consequences for States of the Continued Presence of South Africa in Namibia
(South West Africa) Notwithstanding Security Council Resolution 276 (1970), Advisory Opinion, 1971
I.C.J. 16, 220 (June 21) (Fitzmaurice, J., dissenting) ("[W]hereas the practice of an organization, or of a
particular organ of it, can modify the manner of exercise of one of its functions (as for instance in the
case of the veto in the Security Council which is not deemed to have been involved by a mere
abstention), such practice cannot, in principle, modify or add to the function itself.").
316. Indeed, the very same Sir Gerald Fitzmaurice, writing as a judge of the ECtHR after his
tenure on the ICJ, reserved scathing words for the Court's evolutive method of interpreting the ECHR in
dissent to the judgment in Golder. Golder v. United Kingdom, App. No. 4451/70, Eur. Ct. H.R. 37
n.21 (Feb. 21, 1975) (Fitzmaurice, J., dissenting) ("It is one thing for a national constitution to allow part
of its legislative processes to be effected by means of judge-made 'case law': quite another for this
method to be imposed ab extra on States parties to an international convention supposed to be based on
agreement."); see also Loizidou v. Turkey (preliminary objections) [GC], App. No. 15318/89, Eur. Ct.
H.R. (Mar. 23, 1995) (Golctikki and Pettiti, JJ., dissenting) (decrying the Court's willingness to
expansively develop the Convention and its own power to sever the territorial restrictions attached to
Turkey's accession to the jurisdiction of the ECtHR on the basis of a supposed "uniform and consistent
practice" evidenced by the statements of just five States).
317. In a recent lecture at the ECtHR entitled "What are the Limits to the Evolutive
Interpretation of the Convention," Baroness Hale, Judge of the U.K. Supreme Court, chastised the
ECtHR for its dynamic approach to interpretation on the basis of practice. Noting the Convention's
appellation as a "living document," she contends that it is difficult to determine the "natural limits to the
growth of the living tree." In the end, she notes, "the standard most often appealed to in the Court's
jurisprudence is the common European understanding. But sometimes ... this is judged by the standards
to be found in the domestic legislation of the member States; and at other times .. . it is judged by
evolving European attitudes and beliefs." Sometimes, finally, "it seems to get some way ahead of both."
Baroness Hale, Dialogue Between Judges: What are the Limits to the Evolutive Interpretation of the
Convention?, EUR. CT. H.R. 18 (2011), http://www.echr.coe.int/NR/rdonlyresD901069F-76AO-401 F-B
F48-248FC8OE728A/OIDIALOGUE_201 _EN.pdf.
2013] Treaty Interpretationand Constitutional Transformation 355

may be worth taking.318 But the more the bonds of consent are strained through
change on the informal track, the more likely are such legitimation problems to
arise.
In addition to the problem of consent, constitutional transformation can
give rise to serious issues of accountability. The formal amendment of an
international organization occurs through the willful act of the states parties, in
accordance with clear procedures. The constituent power remains in control of
the development of its constitution. By contrast, informal tracks of
constitutional change take the development of the organization out of the hands
of the parties, and thereby supplant a form of international law making. In the
case of the ECtHR, such competence accrues to the Court itself as gatekeeper;
in the case of the United Nations, the ICJ's rule empowers the political organs
to develop their competences for themselves without the need for judicial
imprimatur. But in both instances the constituted bodies gain a sizeable degree
of control over their constitutions, with only remote accountability to the
constituent states parties.
The rigidity of amendment rules in most international organizations only
amplifies the problem. The formal means of amending the U.N. Charter and the
ECHR are difficult to operationalize. Some, dissatisfied with the sluggishness
of these organizations' amendment procedures, openly advocate the use of
broad interpretative methods to get around such formalism in the interest of
peace, justice, and other moral imperatives. As Dicey observed long ago,
rigidity in amendment rules can incentivize informal tracks of constitutional
change. 320 But where such bodies do resort to informal means of change, the
very same rigidity of the formal amendment procedures inhibits the parties'
capacity to correct the actions of the constituted bodies.321 In other words,
informal change achieves flexibility at the expense of oversight.
So what is to be done? In the first place, it must be reiterated that, in some
cases, informal constitutional change can have highly desirable results despite
its attendant complications. The assessment of constitutional transformation in
international organizations must take place on a case-by-case basis, in light of
each particular organization's specific functions. Moreover, abuses cannot be

318. As Andrew Guzman has recently argued, "An excessive commitment to consent can
cripple efforts to use international law as a tool to help solve the world's largest problems." Andrew
Guzman, Against Consent, 52 VA. J. INT'L L. 747, 753 (2012).
319. See Liang, supra note 185 (advocating modification through interpretation to facilitate the
adaptation of the U.N. organization to changing external conditions); Letsas, supra note 281 (advocating
the continued expansive interpretation of the ECtHR on the basis of truly moral concerns, and even
elevating the Strasbourg Court's interpretive ethic as a "lesson to international lawyers"); see also SATO,
supra note 14.
320. DICEY, supra note 45, at 469-70.
321. The UNSC presents a striking illustration of this point. Because each of the five Permanent
Members of the Council enjoy a veto over any amendment of the U.N. Charter, the amending power is
virtually unavailable as a means of overriding the bodies' self-arrogation of power in the realm of
international peace and security. See COHEN, supra note 32, at 282-83. More generally, the more rigid
the rules of amendment the more likely are the constituted bodies to resort to informal avenues, and the
more difficult it becomes to correct abuses. As Dicey explains, "a constitution may be undermined by
the passing of laws which, without nominally changing its provisions, violates its principles." DICEY,
supranote 45, at 475.
356 THE YALE JOURNAL OF INTERNATIONAL LAW [Vol. 38: 289

simply washed away through sweeping positive gestures. What is crucial is to

understand such processes, and to curtail excesses that risk undermining the
authority of international institutions. To this end, two approaches to reform
appear promising-one arising out of the law of treaties, the other relating to
institutional design.
International law still operates on a fundamentally consensual basis,
despite ongoing trends toward something more communitarian.322 Some of the
most promising approaches to securing greater legitimacy and accountability in
international organizations are thus grounded in positive international law. One
approach entails the further development of the law of treaties, specifically with
regard to lacunae in the codified rules of treaty interpretation, so as to better
guide interpretation in international institutions. 323 It is especially telling that
the ILC has itself begun to reassess the proper criteria and scope of subsequent
practice. Georg Nolte, as Chairman of the Study Group on Treaties over Time,
opened the project with a clarion call:
Treaties are not just dry parchments. They are instruments for providing stability to
their parties and to fulfill the purposes which they embody. They can therefore
change over time, must adapt to new situations, [and] evolve according to the social
needs of the international community.... It is suggested that the Commission
revisit the law of treaties as far as the evolution of treaties over time is concerned.
Problems arise frequently in this context. As certain important multilateral treaties
324
reach a certain age, they are even more likely to arise in the future.
The outcome of such projects remains to be seen, but the project of clarifying
and potentially limiting the appropriate contours of interpretation over time on
a multilateral level is plausible and laudable.
The second avenue of reform goes beyond doctrine, attacking the
problem of informal change in international organizations from a more political
sensibility. Though rarely easy to accomplish, formal institutional design
remains a viable path toward mitigating the legitimation and accountability
problems of informal change, and toward reducing the perceived need to resort
to informality in the first place. In conceiving and constructing international
organizations, statesmen and international lawyers should draw from
constitutional theory and the normative ethos of modern constitutionalism. As
in the domestic state, though of course not in the same way, in the context of
powerful international organizations the value of tools like the internal

322. See Simma, supra note 18, at 233; Villalpando, supra note 18.
323. In this regard some scholars have more explicitly suggested reappraising the rules of
interpretation in the context of international organizations, so as to better foster legitimacy. Some, for
example, suggest that legitimacy can be at least partially secured if the evidentiary requirements for
establishing modification through subsequent practice are understood as being higher than those for
establishing a "mere interpretation" fitting more comfortably with the text. AMERASINGHE, supra note
15, at 463; see also Peters, supra note 142, at 632.
324. Annex A: Treaties over Time, supra note 71, T 1, 6. The ILC has since recast and
formalized its study of treaties over time, appointing Georg Nolte as Special Rapporteur for the newly
constituted topic Subsequent Agreements and Subsequent Practice in Relation to the Interpretation of
Treaties. Rep. of the Int'l L. Comm'n, 64th Sess., May 7-June 1, July 2-Aug. 3 2012, T 227, U.N. Doc.
A/67/10, GAOR, 67th Sess., Supp. No. 10 (2012).
2013] Treaty Interpretationand ConstitutionalTransformation 357

separation of powers and checks and balances cannot be overstated.325 And

perhaps more flexible amendment procedures may mitigate the perceived need
to resort to informal modes of change. Structural reforms of this kind can range
from the minute to the grand-from the plausible to the unfathomable. And it is
important to remember that in the context of institutional design, small steps
can have big effects.
But while both clarifying the rules of interpretation and institutional
design are important, it should be obvious from the foregoing that neither will
be a panacea. It would be futile and undesirable to try to fully constrain the
discretion of judicial interpreters by systematically trying to predetermine the
contours of the various doctrines of interpretation.326 And it is similarly
unrealistic to expect international judicial organs to completely restrain
themselves out of respect for more flexible amendment procedures. Indeed the
ECtHR has made some of its most expansive forays in parallel to the use of
formal amendment procedures by the parties-as in Ocalan and Al-Saadoon
where the Court took note of the parties' contemporaneous attempts at
amending the Charter through formal protocol, and simply crossed the Rubicon
on the laggards' behalf.
It is not clear that we should expect (or seek) an easy solution to the
problems attending informal constitutional change. What is needed today are
new principles of international authority and accountability-so as to better
understand legal and political change in international organizations, to
anticipate these processes and their attendant problems, and to subject them to
informed critique. The venerable ground of constitutional theory and the related
ethos of modem constitutionalism provide a fertile place to start.

325. See Arato, supranote 42.

326. Indeed, the ILC envisions its project in a much more subtle light. See Annex A: Treaties
over Time, supra note 71.