Treaty Interpretation and the Vienna Convention on the Law of Treaties: 30 Years on
Queen Mary Studies in Internation...
304 downloads
1055 Views
2MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Treaty Interpretation and the Vienna Convention on the Law of Treaties: 30 Years on
Queen Mary Studies in International Law Edited by
Malgosia Fitzmaurice and Phoebe Okowa
VOLUME 1
Treaty Interpretation and the Vienna Convention on the Law of Treaties: 30 Years on Edited by
Malgosia Fitzmaurice, Olufemi Elias, Panos Merkouris
LEIDEN • BOSTON 2010
This book is printed on acid-free paper. Library of Congress Cataloging-in-Publication-Data Treaty interpretation and the Vienna Convention on the Law of Treaties : 30 years on / edited by Malgosia Fitzmaurice, Olufemi Elias, Panos Merkouris. p. cm. -- (Queen Mary studies in international law ; v. 1) Includes bibliographical references and index. ISBN 978-90-04-18104-5 (hardback : alk. paper) 1. Treaties. 2. Vienna Convention on the Law of Treaties (1969) 3. Treaties--Interpretation and construction. I. Fitzmaurice, M. II. Elias, O. A. III. Merkouris, Panos. KZ1301.I87 2010 341.3’7--dc22 2010003905
ISBN 1877-4822 ISBN 978 90 04 18104 5 Copyright 2010 by Koninklijke Brill NV, Leiden, The Netherlands. Koninklijke Brill NV incorporates the imprints Brill, Hotei Publishing, IDC Publishers, Martinus Nijhoff Publishers and VSP. All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission from the publisher. Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill NV provided that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive, Suite 910, Danvers, MA 01923, USA. Fees are subject to change. printed in the netherlands
CONTENTS List of Contributors ...............................................................................................vii List of Abbreviations ...............................................................................................xi Introduction: Interpretation is a Science, is an Art, is a Science....................... 1 Panos Merkouris PART I ON THE FUNCTION AND CONTENT OF THE CANONS OF TREATY INTERPRETATION Virtuous Interpretation ....................................................................................... 17 Jan Klabbers Treaty, Custom and Time: Interpretation/Application? ....................................39 Philippe Sands and Jeffery Commission Article 31(3) (a) and (b) of the Vienna Convention and the Kasikili/Sedudu Island Case ............................................................................59 Hazel Fox ‘Third Party’ Considerations and ‘Corrective Interpretation’ in the Interpretative Use of Travaux Préparatoires: – Is it Fahrenheit 451 for Preparatory Work? – .......................................................................................75 Panos Merkouris PART II TREATY INTERPRETATION, INTERNATIONAL TRADE AND INVESTMENT LAW The Appellate Body and Treaty Interpretation ...................................................99 Georges Abi-Saab Interpretation in International Trade Law ........................................................111 Brigitte Stern Diversity and Harmonization of Treaty Interpretation in Investment Arbitration ........................................................................................129 Christoph Schreuer
vi
contents
Canons of Treaty Interpretation: Selected Case Studies From the World Trade Organization and the North American Free Trade Agreement .........................................................................................153 Malgosia Fitzmaurice & Panos Merkouris Interpretation of (Allegedly) Self-judging Clauses in Bilateral Investment Treaties ..........................................................................239 Tarcisio Gazzini PART III INTERPRETATION, AND HUMAN RIGHTS Intentionalism and the Interpretation of the ECHR........................................257 George Letsas Conflicting Interpretations of the ICC Statute – Are the Rules of Interpretation of the Vienna Convention Still Relevant? ...........................273 José Doria Interpreting Constitutive Instruments of International Criminal Tribunals: Reflections on the Special Court for Sierra Leone ........................333 Phoebe Okowa Bibliography ..........................................................................................................357 List of Cases ..........................................................................................................373 Index ......................................................................................................................381
LIST OF CONTRIBUTORS Georges Abi-Saab is Emeritus professor of international law, Graduate Institute of International Studies Geneva. He has acted as counsel and advocate for several governments in cases before the ICJ as well as international arbitrations. He has also served twice as Judge ad hoc on the ICJ, as Judge on the Appeals Chamber of the ICTY and ICTR, as a Commissioner of the United Nations Compensation Commission, as a Member of the Appellate Body of the WTO, of the Administrative Tribunal of the IMF and of various international arbitral tribunals (ICSID, ICC, CRCICA, etc.). Professor Abi-Saab is the author of numerous books and articles, including: Les Exceptions Préliminaires dans la Procédure de la Cour Internationale: Etude des Notions Fondamentales de Procédure et des Moyens de leur Mise en Oeuvre (Paris: Pedone, 1967); International Crises and the Role of Law: The United Nations Operation in Congo 1960–1964 (Oxford: OUP, 1978); ‘Wars of National Liberation in the Geneva Conventions and Protocols’, RCADI 165 (1979-IV) and ‘General Course of Public International Law’, RCADI 207 (1987-VII). Jeffery P. Commission, a New York qualified lawyer, is an associate in the London office of Freshfields Bruckhaus Deringer LLP and is a member of the firm’s international arbitration group. Jeffery, trained in both common law and civil law, has particular expertise in matters of international investment arbitration, public international law and international litigation in US courts. Jeffery is an Assistant Editor on Transnational Dispute Management and an Assistant Editor and Case Report Contributor for the Investment Claims site run by Oxford University Press. José Doria – PhD is a former Law Professor in Moscow (Peoples Friendship university of Russia), and a former Legal Advisor with the International Criminal Tribunal for the Former Yugoslavia in The Hague, currently with the Office of the High Commissioner for Human Rights in Geneva. Olufemi Elias MA (Oxon) LLM (Cantab) PhD (London). Dr Elias is currently a Visiting Professor at Queen Mary University of London and the Executive Secretary of the World Bank Administrative Tribunal. Formerly a lecturer in International Law at King’s College London, he has worked as Legal Adviser at the United Nations Compensation Commission in Geneva and as Senior Legal Officer at the Organisation for the Prohibition of Chemical Weapons in The Hague.
viii
list of contributors
Malgosia Fitzmaurice holds a chair of public international law at the Department of Law of Queen Mary, University of London and is the Editor-inChief of the International Community Law Review. She specializes in international environmental law and the law of treaties and has published extensively on these fields. Hazel Fox CMG QC, formerly Director, of the British Institute of International and Comparative Law and General Editor of the ICLQ; member of the Institut de Droit International. Her publications include: International Arbitration: Law and Procedure (co-authored with J. L Simpson) (London: Stevens, 1959), Joint Development of Offshore Oil and Gas: A Model Agreement for States for Joint Development with Explanatory Commentary (Vol. I – London: BIICL,1989 & Vol. II – London: BIICL, 1990), The Law of State Immunity (Oxford: OUP, 2002 & 2nd edn. in 2008). Tarcisio Gazzini has taught at the Universities of Padova, Glasgow and VU Amsterdam. He has published in several areas of international law, including use of force, international trade, economic sanctions, foreign investment, and international organizations. He is an alternate member of the Committee on Non-State Actors recently established by the ILA. Jan Klabbers is professor of International Organizations Law at the University of Helsinki, and Director of the Academy of Finland Centre of Excellence in Global Governance Research. During the academic year 2009–2010 he is one of the inaugural Straus Fellows at New York University Law School. His main publications include The Concept of Treaty in International Law (The Hague: Kluwer Law International, 1996), An Introduction to International Institutional Law (2nd edn., Cambridge: CUP, 2009), and Treaty Conflict and the European Union (Cambridge: CUP, 2008). George Letsas joined the UCL Faculty as Lecturer in Laws in 2006 and was promoted to Reader in 2009. He holds a PhD in Law (2005) from UCL, an MA in Legal and Political Theory with Distinction (2000) from the Department of Political Science at UCL, and an LLB (1999) from the University of Athens. George is currently the Co-Director of the UCL Institute for Human Rights. He is also co-convenor of the Oxford-UCL Colloquium in Legal and Political Philosophy and has also co-chaired the Colloquium in Legal and Social Philosophy since 2007. Panos Merkouris MA (University of Athens), LLM (University of Athens & UCL), PhD (Queen Mary, University of London). Panos Merkouris is the Managing Editor of the International Community Law Review and a scholarship holder from IKY (Greek State Scholarship Foundation) and Queen Mary.
list of contributors
ix
He specializes in law of treaties, international law of the sea, human rights and humanitarian law. His publications include: Research Handbook on International Environmental Law (London: Edward Elgar, 2010 – co-editor with Professor Malgosia Fitzmaurice and Dr. David Ong) and ‘Debating the Ouroboros of Public International Law: The Drafting History of Article 31(3)(c)’, ICLR 9 (2007): 1–31. Phoebe Okowa is Reader in Public International. Law at Queen Mary, University of London. Philippe Sands QC is Professor of Law at University College London and a barrister at Matrix Chambers. Christoph Schreuer is a graduate of the Universities of Vienna, Cambridge and Yale. He has published numerous articles and several books in the field of international law. Since 1992 he has concentrated on international investment law and the main product of this activity is The ICSID Convention: A Commentary (Cambridge: CUP, 2009) He has written expert opinions in many investment cases. He has spent most of his academic career at the Department of International Law of the University of Salzburg, Austria. From 1992 to 2000 he was the Edward B. Burling Professor of International Law and Organization at the Paul H. Nitze School of Advanced International Studies (SAIS) of the Johns Hopkins University in Washington, D.C. From October 2000 to September 2009 he was Professor of International Law at the University of Vienna, Austria. Since December 2008 he is Of Counsel with the law office Wolf Theiss, Vienna. Brigitte Stern is Professor of International Law at the University of Paris I, Panthéon-Sorbonne, and was Professor at the Graduate Institute of International Studies in Geneva, and Director of the CERDIN-Paris I, Centre of Research in International Law. She is also an Expert for International Organizations and Governments – for example, she was a member of the legal team of the Bosnian Government in the Genocide Case against Yugoslavia before the International Court of Justice, and works as a counsel in several arbitrations – as well as an international Arbitrator (Sole Arbitrator, Member or President in ICSID, ICC, UNCITRAL, NAFTA, Energy Charter Treaty cases). She has been for 7 years, until October 2005, the President of the French Commission for the Elimination of Landmines, and is presently a Judge of the United Nations Administrative Tribunal (UNAT). She has published many books, as well as articles in the AFDI, RGDIP, AJIL.
LIST OF ABBREVIATIONS AB AFRC AG AJIL Ann. BIICL BISD BIT BYIL CAFTA-DR FTA CAR CBD CIEL CITES CMC CMLR CTE CUP DRC DSB DSU EC ECHR ECJ ECOMOG ECOWAS ECR ECT ECtHR EHRR EJIL ETS EU FTA FTC GATS
Appellate Body Armed Forces Revolutionary Council Avocat Général American Journal of International Law Annex British Institute of International and Comparative Law Basic Instruments & Selected Documents Bilateral Investment Treaty British Yearbook of International Law Central America-Dominican Republic Free Trade Agreement Central African Republic Convention on Biological Diversity Center for International Environmental Law Convention on International Trade in Endangered Species Center for Marine Conservation Common Market Law Reports Committee on Trade and Environment Cambridge University Press Democratic Republic of Congo Dispute Settlement Body Dispute Settlement Understanding European Community European Convention on Human Rights European Court of Justice ECOWAS Cease-fire Monitoring Group Economic Community of West African States European Court Reports Energy Charter Treaty European Court of Human Rights European Human Rights Reports European Journal of International Law European Treaty Series European Union Free Trade Agreement Free Trade Commission General Agreement on Trade in Services
xii GATT IBFD ICC ICJ Rep. ICJ ICLQ ICLR ICRC ICSID ICTR ICTY ILC ILM ILR Iran-US CTR JWT LNTS MEA MFN MOX MUP NAFTA NATO NGO NILR NYIL OAS OAU OECD OMC OSPAR OUP P5 PCA PCIJ RCADI RGDIP RIAA
list of abbreviations General Agreement on Tariffs and Trade International Bureau of Fiscal Documentation International Criminal Court ICJ Reports International Court of Justice International and Comparative Law Quarterly International Community Law Review International Committee of the Red Cross International Centre for the Settlement of Investment Disputes International Criminal Tribunal for Rwanda International Criminal Tribunal for the former Yugoslavia International Law Commission International Legal Materials International Law Reports Iran – US Claims Tribunal Reports Journal of World Trade League of Nations Treaty Series Multilateral Environmental Agreement Most-Favoured Nation Mixed Oxide Manchester University Press North American Free Trade Agreement North Atlantic Treaty Organization Non-Governmental Organisation Netherlands International Law Review Netherlands Yearbook of International Law Organization of American States Organization of African Unity Organisation for Economic Co-operation and Development Organisation Mondiale du Commerce (i.e. WTO) Convention for the Protection of the Marine Environment of the North East Atlantic Oxford University Press Five Permanent Members of the UN Security Council Permanent Court of Arbitration Permanent Court of International Justice Recueil des Cours de l’ Academie de Droit International Revue Generale de Droit International Public Review of International Arbitral Awards
list of abbreviations RUF SCM SLD SPS Agreement SRSG TED TRC Report TRIPs UK UN UNCITRAL UNITA UNSC UNSCOR UNTS US VCLT WTO WWF YILC
xiii
Revolutionary United Front Subsidies and Countervailing Measures Soft Lumber Division WTO Agreement on the Application of Sanitary and Phytosanitary Measures Special Representative of the Secretary General Turtle Excluder Device Report of the Sierra Leone Truth and Reconciliation Commission Agreement on Trade-Related Aspects of Intellectual Property Rights United Kingdom United Nations UN Commission on International Trade Law National Union for the Total Independence of Angola United Nations Security Council United Nations Security Council Official Records United Nations Treaty Series United States of America Vienna Convention on the Law of Treaties World Trade Organization World Wildlife Fund Yearbook of the International Law Commission
INTRODUCTION: INTERPRETATION IS A SCIENCE, IS AN ART, IS A SCIENCE* Panos Merkouris “ἥxeiV ἀφήxeiV oὐk ἐn tῷ polέmῳ qnήxeiV” which translates to “you will go you will return not in the battle you will perish”
I. Interpretation: Deciphering Oracular Statements One of the most cited quotes on interpretation comes from Lord McNair: “[t]here is no part of the law of treaties which the text-writer approaches with more trepidation than the question of interpretation”.1 However, it is not just the text-writers that approach interpretation with trepidation. From the very start, lives of peoples and entire empires hanged on the balance of interpreting a single statement, an oracle or even a treaty. In 280 BC, Pyrrhus of Epirus decided to attack the Romans. However, before engaging in such an action, he consulted the oracle of Delphi, as was the custom of the times, hoping that Apollo through the oracle would give him a glimpse of what the future held for him. Pythia2 made the oracular statement mentioned in the beginning of this introduction, i.e. “you will go you will return not in the battle you will perish”.3 It was common practice that * the originals and translations of the ancient greek and latin sources were found in the following sites: i) works of Thucydides and Polybius in <www.perseus.tufts.edu/cache/perscoll_ Greco-Roman.html> ii) works of Cicero, Titus Livius, Ennius and Valerius Maximus in <www .thelatinlibrary.com> and iii) works of Polyaenus in <www.attalus.org/translate/polyaenus .html>. The author would also like to thank Ms. Maria Kanellou, PhD candidate in Classics, UCL, for providing her insight into the research and citation of ancient greek and latin sources. 1 Sir Arnold McNair, The Law of Treaties (Oxford: Clarendon Press, 1961), 364. 2 which was the title that was given to the high-priestess of the Oracle at Delphi and through which Apollo was supposed to speak. 3 there is controversy over whether this oracle was actually given to Pyrrhus or to someone else. It is not cited by any ancient greek or latin source, although it is considered to be one of the best examples of delphic ambiguity. However, in some later texts, this oracle appears in Latin, attached to another oracle that is mentioned in the Annales of Ennius and Cicero’s De Divinatione as having been given to Pyrrhus: “Aio te Aeacida Romanos vincere posse”, which translates to “I say that you, son of Aeacus, the Romans will be able to win”; Marcus Tullius Cicero, De Divinatione, 2.56.116; Quintus Ennius, Annales (Fragmenta), 6.174. This oracle is then, in later texts, followed by “ibis redibis nunquam in bello peribis”, which as we have already mentioned translates to: “you will go you will return not in the battle you will perish”; Henry Coppée, Elements of Logic (New York: American Book Company, 1872), 154, <www.archive.org/stream/ elementsoflogicd00copprich/elementsoflogicd00copprich_djvu.txt> (accessed on 1 July 2009). It is also interesting to note that the first part of this oracle – “I say that you, son of Aeacus, the
2
panos merkouris
Pythian oracular statements were very vague and in some cases open to various interpretations, even conflicting ones. This is one of the most notorious of the latter category. Depending on where one would put the comma, you get two completely different interpretations of the same statement. For instance “you will go, you will return, not in the battle you will perish” means that Pyrrhus would return safely. However, should we transpose the comma to the exact next word then the meaning becomes completely different: “you will go, you will return not, in the battle you will perish” is pretty self-explanatory. Pyrrhus would not survive his attack. What makes matters even worse is that at that time Ancient Greek was written in capital letters, with no punctuation so there was no way for Pyrrhus to know which interpretation was the correct one. Eventually, Pyrrhus decided to go for the first interpretation, but it turned out to be the wrong one. After having been beaten by the Romans, he, eventually, died during battle. II. Interpretation of Ancient Treaties/Agreements The realm of interpretation, however, was not the exclusive domain of oracles and seers. The Classical Antiquity 4 is fraught with similar examples of interpretation, or misinterpretation, which in some cases involve ‘treaties’ of the time. It would be beyond the scope of this introduction to list all the examples available,5 however, a few of the most famous incidents will be mentioned to demonstrate how dangerous, or even literally lethal, it could be to allow a text to be open to various interpretations and conversely how beneficial it was to be attuned to the subtleties of words and their possible interpretations; especially at a time when rules and principles of interpretation had yet to be devised and broadly accepted. Pericles, in one instance, had made an agreement, that should his enemies lay down “their iron or steel” their lives would be spared. The terms used and the context within which they were used, clearly indicated that iron or steel referred to arms and Pericles’ enemies, promptly, did so. Still, Pericles issued
Romans will be able to win” – can be interpreted in two ways, as well, depending on what kind of syntax the interpreter uses. The words ‘you’ and ‘Romans’, can be considered subject and object, respectively, or vice versa; giving rise to the following interpretations: i) you will beat the Romans or ii) the Romans will beat you. 4 i.e. the time-period comprising of the civilizations of Ancient Greece and Ancient Rome. 5 for a more detailed analysis of the various incidents involving interpretation in Classical Antiquity see Hugo Grotius, ‘On Interpretation’, in The Law of War and Peace [De Jure Belli ac Pacis], (1625), Book II, Chapter XVI; Emmerich de Vattel, The Law of Nations; or Principles on the Law of Nature: Applied to the Conduct and Affairs of Nations and Sovereigns, (London, 1702), Book II, Ch. XVII; and more recently, David Bederman, Classical Canons: Rhetoric, Classicism and Treaty Interpretation (Aldershot: Ashgate, 2001).
introduction: interpretation is a science, is an art, is a science 3 an order that all those who still had iron clasps on their cloaks be killed as violators of the agreement.6 Polybius also gives an account of an infamous example of treaty interpretation in Book XII of his Histories. When the Locrians first appeared in Sicily they found a great part of the land already being occupied by the Sicels.7 However, the natives, fearing the newcomers admitted them into the country. The Locrians, then, made a sworn agreement with the Sicels that ‘they would be friendly and share the country with them, as long as they stood upon the ground they then stood upon, and kept heads upon their shoulders’. But, while the oaths were being taken, they say that the Locrians put earth inside the soles of their shoes, and heads of garlic concealed on their shoulders, before they swore; and that then they shook the earth out of their shoes, and threw the heads of garlic off their shoulders, and soon afterwards expelled the Sicels from the country.8
Polyaenus in his Stratagems mentions the case of Paches, who when besieging Notium, proposed a conference with Hippias, the general of Pissithnus. However, Pithias, fearing for his life was reluctant to meet with him. Paches, then, made an oath, “that if Hippias would come out and meet him, he would ensure that he was brought back into the city safe and alive”.9 Hippias convinced that his safety was assured through this oath went out to meet him, but Paches left him under guards and immediately took the city by storm. He then ordered Hippias to be brought back into the city, safe and alive, just as had been agreed; and afterwards he ordered him to be executed.10
The next example comes from Ancient Rome. After the Roman general Scipio Asiaticus defeated King Antiochus the Great of Syria in 190 BC at Magnesia ad Sipylum, Rome dispatched Quintus Fabius Labeo in 188 BC to destroy the king’s navy. Eventually, a treaty was agreed upon, which provided that Antiochus would be allowed to retain half his fleet. However, it had never been the intention of Rome to allow for any possibility for Antiochus to retain any power and gradually re-establish himself as a force to be reckoned with. Therefore, Antiochus employed the following interpretation. Instead of taking half of Antiochus’ fleet, he sawed each and every ship of his fleet in half. Thus, argued Labeo, he had acted within their agreement and Antiochus ended up with no ships at all.11
6 see Grotius, supra note 5, Book II, Chapter XVI, para. V; de Vattel, supra note 5, Book II, Ch. XV §233 & Ch.XVII, §280. 7 the Sicels were one of the three main tribes that occupied Sicily prior to the arrival of greek colonists; see Thucydides, The Peloponnesian War, 6.2.1–6.2.6. 8 Polybius, Histories, 12.6. 9 Polyaenus, Stratagems, 3.2. 10 Id.; see also Thucydides, The Peloponnesian War, 3.34.3. 11 Valerius Maximus, Valeri Maximi Factorum et DictorumMemorabilium, 7.3.4.
4
panos merkouris
The final example shows how issues of interpretation of centuries ago, might still strike a familiar chord with the contemporary international lawyer. At the end of the first Punic war, Rome and Carthage signed a treaty, one of the chief provisions of which was that: “The allies of neither of the parties to the treaty shall be attacked by the other”.12 After this treaty, the Romans took the Saguntines as allies. Later on, Hannibal attacked the Saguntines, so the question which arose was the following: If the Carthaginians approved of this action would they be in violation of the treaty? Polybius summarises the main arguments: [the Romans] alleged that the clause … which by their own admission stipulated that ‘the allies of neither party should be attacked by the other’, did not refer to then existing allies only, as the Carthaginians interpreted it; for in that case a clause would have been added, disabling either from making new alliances in addition to those already existing, or excluding allies, taken subsequently to the making of the treaty, from its benefits.13 But since neither of these provisions was made, it was plain that both the then existing allies, and all those taken subsequently on either side, were entitled to reciprocal security. And this was only reasonable. For it was not likely that they would have made a treaty depriving them of the power, when opportunity offered, of taking on such friends or allies as seemed to their interest; nor, again, if they had taken any such under their protection, was it to be supposed that they would allow them to be injured by any persons whatever. But, in fact, the main thing present in the minds of both parties to the treaty was, that they should mutually agree to abstain from attacking each other’s allies, and on no account admit into alliance with themselves the allies of the other: and it was to subsequent allies that this particular clause applied, ‘Neither shall enlist soldiers, or impose contributions on the provinces or allies of the other; and all shall be alike secure of attack from the other side’.14
Essentially, the question which arose was whether the term allies should be interpreted as referring to allies at the time of the conclusion of the treaty or if it should be interpreted in a broader sense so as to include all future allies. This is a question of intertemporality. Thus, intertemporality, an issue of interpretation, which caused so many problems to the drafters of the VCLT and remains up to this day a highly complex issue, was baffling generals, lawyers and politicians as far back as Classical Antiquity.15
12
Polybius, Histories, 3.27; see also Titus Livius (Livy), Ab Urbe Condita, 21.18–21.19. Clement Barksdale, translating and annotating De Jure Belli ac Pacis of Grotius, notes that such a clause was actually added in the Peloponnesian peace accord between the Lacedemonians and the Athenians; Clement Barksdale, trans., The Illustrious Hugo Grotius of the Law of Warre and Peace with Annotations. III Parts and Memorials of the Author’s Life and Death, (London: printed by T. Warren, for William Lee, and are to be sold at his shop at the signe of the Turkshead in Fleet-Street, 1654), Book II, para. LVIII; see also Thucydides, The Peloponnesian War, 5.18.1–5.18.11. 14 Polybius, Histories, 3.29. 15 both Grotius and Vattel, in their respective works, solved this question by reference to the distinction between odious and favourable obligations and treaties and concluded that 13
introduction: interpretation is a science, is an art, is a science 5 All of the aforementioned instances of interpretation16 share some common themes. Interpretation, in those times, focused to an obsessive degree on the letter of the text to the detriment of context, object and purpose, intention and even in some cases the ordinary meaning of the words.17 It is perhaps this abundance of perfidious interpretations that led Cicero to declare in the first book of de Officiis,18 while using examples from roman military history and agreements, that: “Semper autem in fide, quid senseris, non quid dixeris, cogitandum est” which translates into “But in a promise, what you mean, not what you say, is always to be taken into account”!19 III. Interpretation in the 21st Century: Still as Important as Ever… But Why? Interpretation has gone a long way, since those first examples of the Classical Antiquity. Within the framework of the international legal system, principles of interpretation were gradually formulated, based on the pitfalls of past experiences.20 It comes as no surprise that Puffendorf, Vattel and Grotius, in their respective seminal works, elaborated a set of maxims on interpretation by reference to the works of ancient writers and orators, such as Cicero, Thucydides and Polybius, and to examples of treaty interpretation of the past. Their writings introduced in the field of international law notions such as ‘ordinary meaning’, ‘context’, ‘intention’, ‘special meaning’, ‘preparatory work’ and ‘surrounding circumstances’. These maxims were then picked up, tested and applied by the relevant arbitral tribunals leading to an ever increasing corpus of international jurisprudence on the subject of treaty interpretation. This process, culminated in the general acknowledgement of certain customary principles of interpretation, and their eventual reflection21 in Articles 31–33 of the VCLT.22
the term should refer only to the allies existing at the time of the conclusion of the treaty. Grotius, supra note 5, para. XIII; de Vattel, supra note 5, §284. 16 with the possible exception of the last one, dealing with Rome and Carthage. 17 see the incident involving the Locrians. 18 which is a treatise ‘on Moral Duties’ (which is the english title). 19 Marcus Tullius Cicero, De Officiis, 1.40 (emphasis added). 20 see supra Sections I and II. 21 as the international tribunals never cease to tire of repeating. 22 although, it has to be added that the exact wording, let alone the feasibility of their very inclusion, was hotly debated in the discussions of the ILC prior to the VCLT Conferences. On the debates, on the various aspects of treaty interpretation see the articles published in the present volume; also for a general overview see Franciscus A. Engelen, Interpretation of Tax Treaties under International Law: A Study of Articles 31, 32 and 33 of the Vienna Convention on the Law of Treaties and their Application to Tax Treaties (Amsterdam: IBFD, 2004); Olivier Corten, Les Conventions de Vienna sur le Droit des Traités: Commentaire Article par Article (Bruxelles: Bruylant, 2006).
6
panos merkouris
A researcher would be hard pressed to find a case that, in one way or another, did not touch upon the issue of interpretation. A quick browsing through the case-law of the most prominent international courts and tribunals would persuade even the staunchest critics of this claim. The question that unavoidable has to be addressed is what the cause of this situation is. The reason for this pre-eminence of interpretation in the judicial process is the inherent defects of language. Bertrand Russell, a philosopher, logician and mathematician, who, alongisde with Ludwig Witggenstein, was considered as one of the foremost authorities on this subject, surmises eloquently the main problems of language: There are various problems as regards language. First, there is the problem of what actually occurs in our minds when we use language with the intention of meaning something by it; this problem belongs to psychology. Secondly, there is the problem as to what is the relation subsisting between thoughts, words, or sentences, and that which they refer to or mean; this problem belongs to epistemology. Thirdly, there is the problem of using sentences so as to convey truth rather than falsehood; this belongs to the special science dealing with the subject-matter of the sentences in question. Fourthly, there is the question; what relation must one fact (such as a sentence) have to another in order to be capable of being a symbol for that other? This last is a logical question …23
In order for the need for interpretation to vanish, what is required is … accurate Symbolism, i.e. … Symbolism in which a sentence ‘means’ something quite definite. In practice, language is always more or less vague, so that what we assert is never quite precise. Thus, logic has two problems to deal with in regard to Symbolism: (1) the conditions for sense rather than nonsense in combinations of symbols; (2) the conditions for uniqueness of meaning of reference in symbols or combinations of symbols. A logically perfect language has rules of syntax which prevent nonsense, and has single symbols which always have a definite and unique meaning. … [However, no] language is logically perfect, or [do] we believe ourselves capable, here and now, of constructing a logically perfect language, but … the whole function of language is to have meaning, and it only fulfils this function in proportion as it approaches to the ideal language which we postulate.
Since no human language is perfect, the problem is how one can lift the veil of vagueness. It is this role that interpretation fulfils. It is not a perfect tool, it functions on an ad hoc24 basis but it is the best tool we have towards more accurate Symbolism and the disambiguation of the terms used in everyday language. 23 Bertand Russell, ‘Introduction’, in Tractatus Logico-Philosophicus, Ludwig Wittgenstein, (London: Routledge Classics, 2005), at X. 24 interpretations are not carved in stone. They may change in time or due to context, or for a variety of reasons, or even worse may had been erroneous from the start (let us not forget that the human factor is an element in the equation of the judicial interpretative process and as Seneca said: “errare humanum est” [“to err is human”]).
introduction: interpretation is a science, is an art, is a science 7 In international law, this situation is further aggravated by the fact that language is an element of a larger system that of international law; and this latter system is not devoid of complications itself. International law is a “universe of inter-connected islands”,25 a house with many different rooms.26 Each room/ island is a different area of international law, and depending on its area of focus, be it environmental law, trade law etc. it might have developed a terminology that could differ from the generally accepted one. Even further, the proliferation of international courts and tribunals27, which may, or may not, function in a specific area of international law can lead to varieties of approach as to interpretation,28 not to mention, that interpretation in international law is not restricted to treaties.29
25 Joost Pauwelyn, ‘Bridging Fragmentation and Unity: International Law as a Universe of Inter-connected Islands’, Michigan Journal of International Law 25 (2004): 903–16. 26 based on the analogy of the member of the ILC, Ms. Xue Hanqin; ILC Study Group on Fragmentation, Fragmentation of International Law: Difficulties Arising from the Diversification and Expansion of International Law, Report of the Study Group, finalized by Martti Koskenniemi, (13 April 2006) (A/6N.4/L.682), para. 420. 27 Brown Chester, ‘The Proliferation of International Courts and Tribunals: Finding your Way through the Maze’, Melbourne Journal of International Law 3 (2002): 453–75; Thomas Buergnethal, ‘Proliferation of International Courts and tribunals: Is it Good or is it Bad?’, Leiden Journal of International Law 14 (2001): 267–75; Hugh Thirlway, ‘The Proliferation of International Judicial Organs: Institutional and Substantive Issues: The International Court of Justice and other International Courts’, in Proliferation of International Organizations: Legal Issues, Nicolaas M. Blokker and Henricus G. Schermers (eds.) (The Hague: Kluwer Law International), 251–78. 28 see for instance the publications on interpretation by various international courts: Giorgio Gaja, ‘Does the European Court of Human Rights use its Stated Methods of Interpretation?’, in Divenire Sociale e Adeguamento del Diritto: Studi in Onore di Francesco Capotorti, Francesco Capotorti (ed.) (Milano: Giuffré, 1999), 213–27; Donald McRae, ‘Approaches to the Interpretation of Treaties: the Europeran Court of Human Rights and the WTO Appellate Body’, in Human Rights, Democracy and the Rule of Law: Liber Amicorum: Luzius Wildhaber, Luzius Wildhaber and Stephan Breitenmoser (eds.) (Zürich: Dike, 2007), 1407–22; Koen Lenaerts, ‘Interpretation and the Court of Justice: a Basis for Comparative Reflection’, The International Lawyer 41 (2007): 1011–32; and the various installments of Professor Hugh Thirlway in the British Yearbook of International Law entitled the ‘The Law and Procedure of the International Court of Justice’. 29 Interpretation of unilateral acts has been the subject of judicial decisions (see Nuclear Tests case (Australia v. France), Judgment of 20 December 1974, 1974 ICJ Rep. 253) and of discussions within the ILC (see ILC, ‘Guiding Principles Applicable to Unilateral Declarations of States Capable of Creating Legal Obligations with Commentaries’, YILC (2006), Vol.II, Part II); on interpretation of Security Council Resolutions see Alexander Orakhelashvili, ‘The Impact of Peremptory Norms on the Interpretation and Application of United Nations Security Council Resolutions’, EJIL 16 (2005): 59–88 and Efthymios Papastavridis, ‘Interpretation of Security Council Resolutions Under Chapter VII in the Aftermath of the Iraqi Crisis’, ICLQ 56 (2007): 83–118; finally, recently there has been a gradual proliferation in mentions to ‘interpretation of custom’: see the in passim reference at Certain Questions of Mutual Assistance in Criminal Matters (Djibouti v. France), Judgment of 4 June 2008, 2008 ICJ Rep. para. 40; more importantly see In the Consolidated Arbitration under Chapter Eleven of the North American Free Trade Agreement (NAFTA) and the UNCITRAL Arbitration Rules between the Canadian Cattlemen for Fair Trade (CCFT) and the United States of America, Award on Jurisdiction of 28 January 2008, para. 187, <www.naftaclaims.com/Disputes/USA/CCFT/CCFT-USA-Award.pdf> (accessed on
8
panos merkouris IV. Interpretation: Art or Science?
From the above analysis, it comes as no surprise that interpretation is pivotal in the functioning of the international legal system. But this does not help us come any closer to understanding its nature. One of the most famous aphorisms regarding interpretation produced by the ILC is that “the interpretation of documents is to some extent, an art not an exact science”.30 This statement probably reflects the original awkwardness with which the ILC approached the issue of the inclusion of provisions on interpretation in the first drafts of the Articles on the Law of Treaties. Special Rapporteur Sir Humphrey Waldock, at the 726th meeting of the ILC, informed the ILC members that he was working on articles on treaty interpretation but that [t]he subject was a vast and difficult one and he was anxious not to penetrate too deeply into the realm of logic and what might be described as the art of interpretation.31
1 May 2009) (hereinafter CCFT v. US). For a detailed analysis of how the latter case’s interpretation constitutes ‘interpretation of custom’, see Malgosia Fitzmaurice and Panos Merkouris, ‘Canons of Treaty Interpretation: Selected Case Studies from the WTO and the NAFTA’, (published in the present Volume) at Section V.D. 30 ILC, Report of the ILC on the Work of its 18th Session: Draft Articles on the Law of Treaties with Commentaries, (A/6309/Rev.1), reproduced in YILC (1966), Vol. II: 218; On more recent comments on the ‘artistic’ nature of interpretation see: Vladimir D. Degan, L’ Interprétation des Accords en Droit International (The Hague: Martinus Nijhoff, 1963), at 163–4; Charles Rousseau, Droit Itnernational Public,Tôme I: Introduction et Sources (Paris: Dalloz, 1970), at 29; Serge Sur, L’ Interprétation en Droit International Public (Paris: Librairie Générale de Droit et de Jurisprudence, 1974), at 83; Sur, however, makes the statement that “l’interprétation est un art, non un science” (interpretation is an art, not a science), omitting, thus, the characterization ‘exact’, which as we shall see infra is an important element of the aforementioned ILC quote; Martti Koskenniemi, From Apology to Utopia: The Structure of International Legal Argument (Cambridge: CUP, 2005), at 340 et seq.; Jan Klabbers, ‘Interpretation as the Continuation of Politics by Other Means’, Opinio Juris website, (accessed on 1 November 2009); Richard Gardiner, ‘Art and Rules’, Opinio Juris website, (accessed on 1 November 2009); Duncan Hollis, ‘Art and Auto-Interpretation of Treaties’, Opinio Juris website, (accessed on 1 November 2009); Hollis in a very interesting passage makes an artful analogy with musical composition and considers that: Switching to the analogy of musical composition as an art form, there are understandings as to what a fugue or sonata is, and their composition can be seen as being to some extent governed by rules; but within either form very different results can be produced. To the extent that there are rules for treaty interpretation, they constitute a springboard rather than a straightjacket.
However, the common element of the reasoning of most of these authors lies in the fact that their characterization of interpretation as an artistic endeavour is predicated upon its lack of adhering to strict rules and consequent indeterminacy and vagueness, which as well shall examine infra should not ipso facto bar the characterization of interpretation as a science. 31 ILC, 726th Meeting: Law of Treaties, (A/CN.4/167), reproduced in YILC (1964), Vol. I: 20, para. 4 (emphasis added).
introduction: interpretation is a science, is an art, is a science 9 This comment, sparked a huge debate as to whether it was possible or, even desirable, to include rules on interpretation in the draft articles.32 The Chairman very eloquently stated that it “had been said rather too glibly that interpretation was an art; the question was whether there were any rules for practising that art”.33 Waldock agreed to draft some provisions, but maintained that the issue of interpretation could lead the Commission into great difficulties because the approach of jurists to it was so varied. There were two different kinds of rules; general ones, such as the rule that the treaty must be read as a whole, and strictly technical ones. Some rules of a practical nature could be usefully summarized but he would view with apprehension any attempt to delve too deeply into the theoretical issues.34
During the 16th Session of the ILC, at the 765th meeting, the issue of whether it was necessary to include rules on interpretation, was once again raised by Waldock, but the ensuing dialogue, showed that the majority of the ILC members felt that rules on interpretation had to be included,35 a feeling that was affirmed by the comments of the governments to the draft articles. From this point onward, the debate on the inclusion of rules on interpretation, gives its place to debate on the content and nature of those rules. However, the aforementioned and the relevant discussions in the ILC indicate that the main arguments for non-inclusion were that interpretation was not a process that could be regulated, that setting out rules might restrict the judges too much and that if rules of interpretation truly existed and were identifiable, then the practice of international courts would not be so erratic and inconsistent. Be that as it may, and irrespective of whether these arguments had validity or not at the time, the post-VCLT practice seems to have scattered all the above fears. International courts apply the relevant articles on interpretation of the VCLT and make reference to them, even when they do not have to apply the VCLT. There is no shortage of cases, where although the parties are not bound by the VCLT, the international court states that they will apply Articles 31 and 32 of the VCLT, as they enshrine customary international law. Although they do not have to, they still make reference to Articles 31 and 32. Still the question remains. Is the statement of the ILC correct; is interpretation an art, and not a science? If this is true, then this would mean that judicial certainty is just a utopian dream;36 that academics, lawyers and judges alike,
32 see for example the comments made by Verdross, Elias and Paredes in ILC, supra note 31, paras. 15, 24 and 28, respectively. 33 Id., para. 34. 34 Id., para. 36. 35 For arguments on both sides see ILC, 765th – 766th Meetings: Law of Treaties, (A/CN.4/167/ Add.3), reproduced in YILC (1964), Vol. I: 275et seq.; also see ILC, Draft Articles on the Law of Treaties with Commentaries, YILC (1964), Vol.II: 176, at 199–201. 36 if there is no science in legal interpretation, no rules regulating the outcome, then any result would be possible.
10
panos merkouris
are artists and not legal scientists; that judicial decisions, the articles in the present volume are works of art not the products of legal science; most importantly, since interpretation is a key element of legal science, the latter would then be just a fiction, a mirage obstructing the true nature of an ‘art of law’ from shining through. Before we accept, though, our new vocation as artists, let us examine what exactly does “interpretation is an art, not an exact science” mean. If we juxtapose the two elements of the sentence, then it is evident that: i. the term ‘art’ is used as a kind of antithesis to ‘exact science’; and ii. revealing, as well, is the use of the epithet ‘exact’ to characterise ‘science’. From the above two elements we can draw the conclusion that the ILC describes interpretation as an art because, unlike science, where a set of rules will predetermine the exact outcome in any given scenario, interpretation does not seem to follow that deterministic process. It consists of so many unquantifiable elements,37 which render the prediction of an outcome with absolute certainty an impossibility. However, this idea of science as ‘exact’ does not correspond to reality at all. An absolutely deterministic model requires knowledge of the initial conditions with infinite precision. In reality, such infinite precision is impossible. Consequently, even in physics, which is probably the basis for this notion of ‘exact science’ of the ILC there is always a ‘margin of error’; and this ‘margin of error’ is just the first step down the rabbit hole. Chaos theory, quantum physics and Heisenberg’s uncertainty principle,38 are but a few examples, in modern physics, that go towards showing that an ‘exact science’ is a notion of the past. The coup de grâce, however, is delivered in mathematics, by Gödel’s theory of incompleteness, which proved that mathematics, like language, is an incomplete system; that there are statements in mathematics that are undecidable – they can neither be proved nor disproved – and that whether a statement falls within this category or not cannot be known a priori. Thus, with Gödel’s theory, mathematics, which until then was hailed as the perfect/complete system, proved to be just another in a long line of incomplete systems. Based on the above, the ILC cannot chastise interpretation for the shortcomings, i.e. uncertainty, that the ‘exact sciences’ also manifest. Interpretation is regulated by rules,39 which to the degree possible, ensure legal certainty,
37 some of which are non-legal, not the least of which is the human element in the judicial process. 38 according to which it is impossible to measure simultaneously both position and velocity of a microscopic particle with any degree of accuracy or certainty. The more precisely one property is known, the less precisely the other can be known. 39 Articles 31–33 of the VCLT and their customary law counterparts.
introduction: interpretation is a science, is an art, is a science 11 a main goal of legal science. It is no coincidence, that all international courts and tribunals, either explicitly or implicitly follow the process of treaty interpretation enshrined in Articles 31–33 of the VCLT. Not only do these rules satisfy the need for legal certainty, but also they ingeniously use the failings of language to the benefit of the interpretative process. One of the main shortcomings of language as analysed supra is the lack of accurate Symbolism; one word can mean many different things. The drafters of the VCLT ensured that the articles on interpretation would benefit from this, and through the use of terms and phraseology such as ‘any subsequent practice’ of Article 31(3)(b), ‘all relevant rules of international law’ of Article 31(3)(c) ‘supplementary means…including’ and ‘circumstance of [the treaty’s] conclusion’ of Article 32 allowed for a controlled flexibility of the interpretative process, to be able to adapt not only to the particularities of each case, but also to the evolution of the international community and the changes brought about by the lapse of time.40 It is exactly for the aforementioned reasons that the authors of the various chapters of this Volume devoted to interpretation play an important role. We live in interesting times, where the proliferation of international adjudication and of areas of international legal regulation has reached unprecedented heights. In such a diverse and constantly changing environment, interpretation is the only way to keep up, but interpretation is also affected.41 By analysing i) the function and normative content of the Articles on interpretation of the VCLT directly, ii) their application in international trade and investment law and iii) in a human rights context, the authors attempt to produce a clearer image of the rules of interpretation; to define with better precision42 the workings of interpretation, in order for us to understand it, to know it better. Isn’t it serendipitous that the etymological root of ‘science’ is exactly that: to know well43? From the above, it was deduced that interpretation in international law is science, not an exact one, but science, nevertheless. But does this mean
40 see for instance the sudden burst in judicial decisions dealing with Article 31(3)(c) and the principle of systemic integration (e.g., In the Arbitration Regarding the Iron Rhine (‘IJzeren Rijn’) Railway Between The Kingdom of Belgium and the Kingdom of the Netherlands, PCA, Award of the Tribunal of 24 May 2005; EC – Measures Affecting the Approval and Marketing of Biotech Products (29 September 2006), Panel Report, WT/DS291.292.293) or more recent cases, making timid steps in including judicial decisions, within the notion of supplementary means of Article 32 (see CCFT v. US, supra note 29; European Communities – Customs Classification of Frozen Boneless Chicken Cuts (30 May 2005), Panel Report, WT/DS269/R & WT/DS286/R). 41 as analysed in the preceding paragraph. 42 it was mentioned earlier that infinite precision is unattainable for any science. 43 science comes from the latin word ‘scientia’ (‘knowledge’), which originates from the verb ‘scire’ i.e. ‘to know’. The same is also true for the greek word for science, ‘ἐpistήmh’ (which is the root for such english words as: epistemology), originates from the verb ‘ἐpίstamai’, which means ‘to know well’.
12
panos merkouris
ipso facto that interpretation is not an art? Are they so antithetical, that the existence of one negates the other? The author submits that this is not the case at hand. There are numerous examples where science becomes art, or where art uses science to enhance aesthetic value: Fractals,44 are an important section in mathematics, yet their beauty has given rise to fractal art. Most notably, fractals have been identified in Jackson Pollock paintings. M.C. Escher was inspired by mathematics in the creation of several of his works, such as, for instance, the ‘Print Gallery’, ‘Waterfall’ and ‘Up and Down’. Leonardo da Vinci used the Golden Rectangle45 when painting the face of Mona Lisa. That particular geometric shape is considered to be one of the most pleasing and beautiful shapes to look at, which is why many artists have used it in their work. However, the work that probably best epitomises the blend between art and science is the Vitruvian Man; a drawing by Leonardo da Vinci which is both a work of art and of science. The basis for the Vitruvian Man is the correlation of ideal human proportions and geometrical shapes; beauty and geometry; art and science. Consequently, science and art are not mutually exclusive. On the contrary, there are areas of science and art that overlap as shown in the diagram below: A
art
science
B
To produce a single meaning out of a swarm of elements, to juggle various considerations and circumstances, while attributing to each the appropriate gravitas through the application of the principles of interpretation, to end in each case with one single interpretation, applicable to that specific case and at that particular instant, is surely a scientific process, but undeniably, as well, a work of art. So how can we respond to the question of whether interpretation is an art or a science? Interpretation is a science, that is artful, an art that is scientific; 44 a fractal is a rough or fragmented geometric shape that can be split into parts, each of which is a reduced-size copy of the whole. 45 A Golden Rectangle is a rectangle connected to the mathematical constant j (known as ‘golden ratio’ or ‘divine proportion’).
introduction: interpretation is a science, is an art, is a science 13 a science that has characteristics that transform it into art, which art in turn partakes of such scientific elements that make it science, and so on and so forth. Paraphrasing the title of Malvina Halberstam’s article ‘A Treaty is a Treaty is a Treaty’46 and in an attempt to express this indivisible duality of interpretation, to the question: Interpretation is an art or a science? The answer should be: Interpretation is a science is an art is a science.
46 Malvina Halberstam, ‘A Treaty is a Treaty is a Treaty’, Virginia Journal of International Law 33 (1992): 51–68.
Part I
On the Function and Content of the Canons of Treaty Interpretation
VIRTUOUS INTERPRETATION Jan Klabbers I. Introduction Curiouser and curiouser, to paraphrase one of the linguistically most playful stories ever, is the circumstance that over the last decade or so international lawyers have come to be obsessed by rules on treaty interpretation. Treaty interpretation used to be a rather arcane topic, with many resigning to the old adage according to which interpretation was an art, not a science.1 And if it was an art, one could hardly develop general rules on how best to do it, in much the same way that one can hardly develop general rules on how best to listen to Beethoven or Satie, or how best to understand the paintings of Rembrandt or Picasso or understand the novels of Flaubert or Tolstoy. Indeed, another fictional stalwart perhaps put it best: a word means whatever I want it to mean, dixit Humpty Dumpty, rather than what someone else could possibly think I may have had in mind. While on occasion authorities tried to make sense of interpretation, they quickly gave up, usually not reaching much further than the conclusion that the idea of there being a scientific element involved was a charade. In the classic words of Sir Robert Jennings, interpretation is, at best, an art masquerading as a science.2 A glimpse at the leading relevant works tells its own story. During the two decades in which the International Law Commission (ILC) paved the way for the Vienna Convention, its special rapporteurs clearly had low expectations concerning rules on interpretation. Brierly – admittedly not highly interested in codification efforts to begin with3 – never got around to drafting rules on interpretation, and neither did Hersch Lauterpacht.4 Sir Gerald Fitzmaurice was even positively opposed, expecting far more practical results from listing
1 Telling is the motto to the chapter on interpretation in Anthony Aust, Modern Treaty Law and Practice (Cambridge: CUP, 2000), 184: “The interpretation of documents is to some extent an art, not an exact science.” Note also that in this case, there is no reference to the author of the phrase (usually the mottos with which Aust starts his chapters are attributed to a single author), as if the statement is so common that no authorship can be discerned. 2 See Sir Robert Jennings, ‘General Course on Principles of International Law’, RCADI 121 (1967/II): 323–606, at 544. 3 He quit after having produced three reports, and wrote elsewhere that he was not too convinced of the wisdom of codifying international law to begin with. See James L. Brierly, ‘The Codification of International Law’, Michigan Law Review 47 (1948): 2–10. 4 Lauterpacht did however address interpretation in his capacity as Rapporteur for the Institut de Droit International.
18
jan klabbers
a code of principles or maxims that could possibly be relied on by tribunals to justify their decisions, rather than any hard-and-fast rules telling tribunals how to read treaty provisions.5 And even Sir Humphrey Waldock, under whose guidance the ILC’s final draft was steered through the ordeal of two diplomatic conferences, was decidedly lukewarm on rules on interpretation, including them more because he thought this was expected of him than out of genuine expectation that rules on interpretation would be of much use.6 Much the same applies to general textbooks on international law, which would typically contain a chapter on the law of treaties and devote a word or two on interpretation, but not much else, and no overblown expectations either.7 And it is perhaps no coincidence that McNair, in what is still the most impressive comprehensive work on the law of treaties generally, devotes some 150 pages to interpretation but without ever attempting to formulate a general one-size-fits-all rule on treaty interpretation.8 In this light, the current voguish focus on rules of interpretation warrants some explanation. International tribunals increasingly, or so it seems, invoke the rules on treaty interpretation as laid down in articles 31 and 32 of the Vienna Convention on the Law of Treaties.9 This applies to the ICJ which, for instance in Oil Platforms, relied heavily on the rules on interpretation and in particular on what has been called the principle of systematic integration.10 It applies most prominently to the WTO’s dispute settlement body, which time and again invokes the rules on treaty interpretation.11 It applies to arbitral bodies such as the Permanent Court of Arbitration, which in its OSPAR award invoked the rules on interpretation to justify its common-sense conclusion concerning applicable law.12 And it even applies to the European Court of Justice, and its advocates-general, who have come to rely on the Vienna
5 But then, Fitzmaurice would have preferred an expository code on the law of treaties generally, instead of a convention. 6 See Sir Humphrey Waldock, ‘Third Report on the Law of Treaties’, YILC (1964): Vol. II, 5–65, esp. at 54. 7 To illustrate, the 4th edition of Brownlie devotes six of its almost 750 pages to interpretation (and one of these is devoted to reproducing articles 31 and 32 of the Vienna Convention). See Ian Brownlie, Principles of Public International Law (Oxford: Clarendon Press, 1990), 626– 32. Bucking the trend, by the way, much the same applies to the recent 7th edition: still six pages, including a reproduction of articles 31 and 32. See Ian Brownlie, Principles of Public International Law, (Oxford: Clarendon Press, 2008), 630–6. 8 See Lord Arnold McNair, The Law of Treaties (Oxford: Clarendon Press, 1961), 345–489. 9 Note also the wonderful Freudian reference to the ‘Vienna Convention on the Interpretation of Treaties’ in Lori Wallach and Patrick Woodall, WTO: Whose Trade Organization? A Comprehensive Guide to the WTO (New York: New Press, 2004), 235. 10 See also Oil Platforms case (Iran v. United States of America), Merits, Judgment of 6 November 2003, 2003 ICJ Rep. 161, para. 41. 11 See e.g., US – Standards for Reformulated and Conventional Gasoline (29 April 1996), Appellate Body Report, WT/DS2/AB/R, at 16–8. 12 See Dispute Concerning Access to information under article 9 of the OSPAR Convention (Ireland v. United Kingdom), PCA, Final Award of 2 July 2003, paras. 80–6.
virtuous interpretation
19
Convention’s rules on interpretation in cases involving agreements which affect the Community or to which the Community is a party13 – even if all other parties happen to be member states of the same Community.14 Claimants and respondents before the ECJ, moreover, invoke the Vienna Convention’s rules also when it comes to the interpretation of Community instruments, such as regulations.15 Scholars too seem to have discovered interpretation as a topic worthy of attention. It can hardly be a coincidence that recently, two monographs on the process of treaty interpretation have been published in English16 (and it may well be that these are actually the first; I cannot readily think of predecessors17), or that scholarly articles are being produced on how to interpret other international legal instruments, such as decisions of international organizations.18 The International Law Commission too placed much hope on interpretation as a way to combat the fragmentation of international law,19 and most intriguingly perhaps, interpretation is used to present work that is hardly about interpretation as such – not unlike Einstein presenting his theory of relativity as a theory on the interpretation of relativity.20 What spurs this reliance – overreliance perhaps – on rules of interpretation? How can it possibly be explained that the classic authorities never expected too much from rules on interpretation, whereas every self-respecting international lawyer today feels the need to justify their particular way of reading a text with the help of external standards? What does this tell us about the nature of interpretation?
13 See e.g., Case C-199/05, European Community v. Belgium (2006), Opinion of AG StixHackl, ECR: I-10485, e.g., para. 49 (dealing with a provision in the Protocol on privileges and Immunities of the EC). 14 See e.g., Case C-203/07 P, Greece v. Commission (2008), Opinion of AG Mazák, para. 63 (dealing with an agreement between the member states and the Commission relating to the EU’s mission in Nigeria. 15 See e.g., Case C-500/99 P, Conserve Italia v. Commission (2002), ECR: I-867, para. 73. 16 See Richard Gardiner, Treaty Interpretation (Oxford: OUP, 2008) and Ulf Linderfalk, On the Interpretation of Treaties: The Modern International Law as Expressed in the 1969 Vienna Convention on the Law of Treaties (Dordrecht: Springer, 2007). 17 There have been monographs in other languages, but those too have been few and far between: two within the space of a year would seem to be unprecedented. An example of a leading study in French is Vladimir D. Degan, L’Interprétation des Accords en Droit International (The Hague: Martinus Nijhoff, 1963). 18 See Efthymios Papastavridis, ‘Interpretation of Security Council Resolutions under Chapter VII in the Aftermath of the Iraqi Crisis’, ICLQ 56 (2007): 83–118. 19 See Martti Koskenniemi, Fragmentation of International Law: Difficulties arising from the Diversification and Expansion of International Law. Report of the Study Group of the International Law Commission (Helsinki: Erik Castrén Institute, 2007). 20 See e.g., the fine study by George Letsas, A Theory of Interpretation of the European Convention on Human Rights (Oxford: OUP, 2007), which is essentially a theory of human rights rather than a theory of interpretation.
jan klabbers
20
One possible – and highly plausible – answer was suggested by a young David Kennedy almost a quarter of a century ago. Interpretation, so he stated, played out as the functional equivalent of truth, helping to curb power: through interpretation “power seems tamed despite the unavailability of a workable picture of truth.”21 It follows that whoever controls the process of interpretation, therewith controls the truth, or at least the meaning to be given to the text subject to interpretation.22 Interpretation therewith equals power, and being able to decide on how to interpret would be tantamount to dictating the terms of a legal instrument.23 I shall take this as a given, and leave it as it is.24 What I wish to focus on instead is the proposition that interpretation cannot meaningfully be subjected to rules, let alone to legal rules; instead, much depends on the personal characteristics – the virtues, in a term which sounds more outdated than it should – of those who interpret. My main proposition is a plea for virtuous interpretation, as opposed to the sort of virtuoso interpretation that could result from applying rules on interpretation. Section II discusses the difficulties of thinking in terms of rules on interpretation; section III analyzes the question of the ostensible customary nature of the Vienna Convention’s rules on interpretation; in section IV, I will sketch how these rules by their very nature are of little concrete help. Section V will sketch the contours of what I call a virtuous approach to interpretation, while section VI concludes. II. Difficulties of Thinking in Terms of Rules on Interpretation If a picture tells a thousand stories, perhaps a well-chosen anecdote can be quite illuminating as well. So let me start with an anecdote. In the spring of 2008 I participated in a panel discussion at the annual International Studies
21
See David Kennedy, ‘The Turn to Interpretation’, Southern California Law Review 58 (1985): 251–75, at 265. 22 As Koskenniemi puts it succinctly: “The interpretative techniques lawyers use to proceed from a text or a behaviour to its ‘meaning’ create (and do not ‘reflect’) those meanings.” See Martti Koskenniemi, From Apology to Utopia: The Structure of International Legal Argument (Cambridge: CUP, 2005), at 597 (emphasis in original). 23 This is further strengthened by the consideration that interpretation decides not only on how to interpret, but also on what to interpret. For helpful discussion on this point, see Roger Cotterell, The Politics of Jurisprudence: A Critical Introduction to Legal Philosophy (London: Butterworths, 1989), 179–80. 24 In a recent contribution to a leading US law review, it is claimed that when US courts interpret treaties, they should work as if they were interpreting contracts. The Vienna Convention’s rules only make a small cameo appearance when the author notes that his approach “is also consistent with international norms”. This illustrates both the extent to which interpretation is about power (US lawyers can be blissfully ignorant of international niceties), and that international lawyers may well form an interpretive community of their own. See Curtis J. Mahoney, ‘Treaties as Contracts: Textualism, Contract Theory, and the Interpretation of Treaties’, Yale Law Journal 116 (2007): 824–57; the words quoted can be found at 853.
virtuous interpretation
21
Association conference on the topic of interdisciplinary relations, in my case discussing the possibilities and improbabilities of cooperation between international lawyers and those specializing in the study of international relations. As things go, we (the participants in the panel) were later asked to put our thoughts on paper, and true to the polemic nature of the original, I wrote a few pages suggesting that, often enough, calls for interdisciplinary cooperation are exercises of power: one discipline aiming to exercise power, through its own vocabulary and techniques, over another.25 In the process, I claimed that having the same object of study, being occupied with similar things, need not necessarily result in fruitful cooperation, and in order to drive the point home I invented a comparison between a musicologist studying Mozart’s string quartets, and the lady at the ticket counter selling tickets to next week’s performance of one of Mozart’s operas. Both have something to do with Mozart, but it is also immediately obvious that the things they do involving Mozart are radically different and have nothing much to do with one another. So far, so good. Imagine my surprise when one of the referees (because the polemic contribution had to go through a refereeing process) not only did not grasp the essence of what makes a polemic piece (overblown claims, lack of subtlety, lack of evidence, lack of balance) but also was none too happy with my Mozart analogy: this, he felt (I shall presume male authorship), suggested an unwarranted hierarchy and was, moreover, gendered. Leaving polemic style debates aside, the latter two points were, I thought, quite interesting, for at no point had I suggested that being a musicologist is more important than selling tickets (or the other way around, for that matter), and while it was true that I spoke of the lady at the ticket counter, the piece remained silent on the sex of the musicologist.26 Obviously, though, the referee must have thought my fictional musicologist was a man, and must have also presumed that I would automatically think that the profession of musicologist somehow ranks above that of ticket seller, and therewith presumed I would unflinchingly rank man over woman. This then could explain the referee’s point on hierarchy; and this then could explain the point concerning the gendered nature of the example. As noted, though, both interpretations are at least debatable. The piece contained no value statement concerning the respective professions of musicologist and ticket seller. At best, one could claim that since I mentioned the
25 I have made the same point elsewhere as well. See Jan Klabbers, ‘The Relative Autonomy of International Law, or The Forgotten Politics of Interdisciplinarity’, Journal of International Law and International Relations 1 (2004–2005): 35–48. 26 I wrote that the disciplines of international law and international relations
have as much in common as the musicologist studying Mozart’s string quartets, and the lady at the ticket-counter trying to sell tickets to next month’s performance of Die Zauberflöte: both somehow have to do with Mozart, but that’s about it.
22
jan klabbers
musicologist first, I must have felt this to be the more important profession. This is open for rebuttal, however: maybe I ordered them alphabetically (‘musicologist’ comes before ‘ticket seller’, ‘seller’, or ‘salesperson’, after all); maybe I felt that the most important one should be mentioned last. Or maybe I never gave it much thought: sometimes, as they say, a cigar is just a cigar, after all. Likewise, there was, in my draft, no indication of the musicologist’s sex. He could be a he, or she could be a she. He or she could have been transsexual too, for that matter and for all I care. At most, the one feasible conclusion is that in all likelihood my musicologist was human; but this too, I guess, will remain speculation. I simply never gave the matter much thought. Now what does this little anecdote teach us? It teaches us, I think, two related things. First, it teaches us that the author of a text has, in the end, fairly little influence over how his or her text will be interpreted.27 What the author intends will be filtered by the audience. Clearly, this referee jumped to conclusions all of his own making, with no basis in the text that he had been reading. And to the limited extent that his reading had a textual basis, it had crept in unintentionally: I had written of the lady at the ticket counter without giving it much thought, without intending anything beyond illustrating my point. And this relates to the second lesson of the anecdote: regardless of what I had meant to convey, this particular referee had approached my piece with a bag full of background assumptions: his alarm bells responded to the word ‘lady’; his reading was sensitive to possible power differences between people selling tickets and people studying string quartets;28 his reading was sensitive to the undisputable circumstance (I presume, not having checked this) that there are probably more male musicologists than female musicologists. Regardless of what my intention had been, he came to construe my analogy as an example of unwarranted hierarchization and far from gender-neutral.29 And not without justification: I should have realized that whatever I would write would See Jan Klabbers, ‘The Bridge Crack’d: A Critical Look at Interdisciplinary Relations’ International Relations 23 (2009): 119–125. 27 This is something every published author will immediately recognize, I would think. 28 Tantalizingly, of course, it would by no means be impossible that the ticket lady was engaged in a part time profession, paying her way through university training in order to become a musicologist. Surely, for those interested in musicology, selling tickets at the Opera may not be such a bad starting point. This would both strengthen and weaken the element of hierarchy: it would strengthen it by positing a teacher-pupil relationship, but weaken it by suggesting that it is only temporary: at some point in time, the ticket lady will become a musicologist, on a par with the current musicologist, or perhaps even, in the more distant future, his boss. Also not impossible is that the musicologist hailed from a working class background while the lady at the booth would be some sort of Paris Hilton-type socialite: a wealthy heiress entertaining herself by doing volunteer work selling tickets to the opera on occasion. In such a case, clearly any professional hierarchy would be overshadowed by the social hierarchy. But perhaps it is best to let these possibilities slide. 29 It is, incidentally, precisely to take such background assumptions into account that the published version will speak of the ‘gentleman’ selling tickets. See Klabbers, supra note 26.
virtuous interpretation
23
end up filtered by each and every reader; I should have realized that readers make up much their own story, taking some cues from whatever the writer writes but moulding it into a story of their own making, based as much on whatever their background assumptions are as on what I put into the mix. Now, both lessons are fairly trivial; the same things have been pointed out long ago by the likes of political theorist John Pocock30 and literary theorist Stanley Fish.31 Indeed, in particular the latter has done much to enhance our understanding of interpretation, suggesting that interpretation depends not so much on what the authors of a text put into it, but rather on what the reader takes out. That is not to say that readers can do as they please; typically, they are part of an interpretive community, and their understanding will often be based, to quite a large extent, on the sort of assumptions and understandings ‘in force’ in that particular community.32 Thus, to put it plainly, put a human rights text to a human rights lawyer, and she will approach it with the sensitivities of a human rights lawyer, sensitive to possible escape clauses, possible margins of appreciation, and knowing what other authoritative interpreters have held similar texts to mean. Give the same text to a tax lawyer, or a maritime lawyer, however, and chances are that the tax lawyer or the maritime lawyer come to rather different conclusions as to what the text means, not so much because they use different techniques or would necessarily have a different set of political values to endorse, but because they start from different premises, put on different spectacles and have different blind spots.33 III. The Question of the Ostensible Customary Nature of the Vienna Convention’s Rules on Interpretation The above both helps explain the resort to rules on interpretation and helps sketch their limits. It helps explain them in the following way: obviously, if
30
I remember a wonderful quote by Pocock to this effect which I read at some point in the early 1990s and stored on my antique laptop. Alas, it did not survive the implosion of said laptop, and I have been unable to retrieve the quote. 31 See e.g., Stanley Fish, Doing What Comes Naturally: Change, Rhetoric, and the Practice of Theory in Literary and Legal Studies (Oxford: Clarendon Press, 1989). Schauer’s statement “that language has meaning independent of what its initial users (or inscribers) intended to say…” could encompass the same position, although he uses it mostly to describe changes over time, arguing against originalism. See Frederick Schauer, Playing by the Rules: A Philosophical Examination of Rule-based Decision-making in Law and in Life (Oxford: Clarendon Press, 1991), 219. 32 For an application to international law, see Ian Johnstone, ‘Treaty Interpretation: The Authority of Interpretive Communities’, Michigan Journal of International Law 12 (1991): 371–419. 33 More structural still: “By denying knowledge of both subjective and objective meanings, theories of treaty interpretation become a perpetuum mobile which allow challenging each proposed interpretation as ‘subjective’. ” See Koskenniemi, supra note 22, at 337 (emphases in original).
24
jan klabbers
words can mean whatever we want them to mean, then we should have rules on how to give meaning to words. Any order in world politics is only possible, so it would seem, if we have some guidance as to how to read the texts that record our agreements. Without such rules, there can be no guarantees that different parties, different states, different interpretive communities, come to radically difficult conclusions as to what it was again we agreed on. And if that is the case, then order is simply not possible. It so happens such rules are, for purposes of treaty interpretation, laid down in the Vienna Convention on the Law of Treaties, and it so happens, many would claim, that those rules are of customary origin and nature. Even those to whom the Vienna Convention technically does not apply (because they are not parties, or because they are not states, or both) would be bound to respect the rules of the Vienna Convention; and even if at issue are documents other than treaties (resolutions of international organizations, unilateral statements,34 perhaps even judicial decisions), nonetheless the rules of the Vienna Convention ought to be followed. This seems to have become the consensus position: in all matters international, always interpret in accordance with the rules of the Vienna Convention. The Vienna Convention, more in particular its rules on interpretation, therewith come close to being of quasi-constitutional character. That raises two questions: is this burden commensurate with the creation of the Vienna Convention’s rules on interpretation? And can the rules carry such a burden to begin with? To start with the first question: it is sometimes suggested that the rules on treaty interpretation are customary in origin. In other words, so the reasoning goes (or should go, rather), throughout the centuries, states and their agents started to look in similar ways to the texts before them, establishing a general practice, and over time came to accept this general practice as law. While they might, at times, diverge from the rules on interpretation, they would do so by insisting that any divergence would only be temporary, and would be justified by the circumstances at hand, but would not affect the general rule.35 Put like this, it becomes obvious that for any such set of customary rules to arise, two preconditions must be met. First, the behaviour concerned must
34 Note, in this connection, how the ICJ felt compelled to interpret unilateral statements in the Nuclear Tests cases, and while its doing so helped diffuse a potentially explosive dispute, the Court’s reading of those statements has not been universally shared. See e.g., Nuclear Tests case (Australia v. France), Judgment of 20 December 1974, 1974 ICJ Rep. 253. 35 Obviously, members of the ‘invisible college of international lawyers’ (an early incarnation of the idea of interpretive community, and a fairly broad one) will recognize the notion of custom presented here as owing a thing or two to the ICJ’s findings in the Military and Paramilitary Activities in and against Nicaragua (Nicaragua v. United States of America), Judgment of 27 June 1986, 1986 ICJ Rep. 14. The notion is derived from Oscar Schachter, ‘The Invisible College of International Lawyers’, Northwestern University Law Review 72 (1977): 217 et seq.
virtuous interpretation
25
be, in the words of the North Sea Continental Shelf cases, of a “fundamentally norm-creating character”36, which suggests that, as an activity, interpretation can meaningfully be captured in terms of rules to begin with. This now is doubtful: it may very well be (as I have argued elsewhere37) that interpretation is not a rule-governed activity. Indeed, if there is any message to be found in the lament that interpretation is an art, not a science (or an art masquerading as a science), it is precisely this: the activity cannot be captured in rules, at least not in a meaningful way. It is not the case that anyone armed with the rules of interpretation can make sense of, say, the Biological Diversity Convention, or the European Convention on Human Rights. It is, really, not unlike playing the violin or playing tennis: one can give the same guidelines to two different people, and they will reach different standards of sophistication (which, in our case, translates as different levels of understanding of the text before them). To be sure, following the guidebook for playing the violin will teach its users that the violin is a string instrument, not meant to be drummed (except, in an odd sort of way, when playing pizzicato), and should ideally be approached with a bow while its correct positioning is under the chin (rather than, say, on the lap). Likewise, following the guidebook for tennis will probably tell the user that tennis is typically played with balls roughly the size of a fist, softer than those use for golf and bigger than those use for table tennis, but smaller than those used for football; that the general idea is to hit the ball over the net but between the lines, out of the opponent’s reach, and that the scoring rules have fairly little to do with what normally passes for logic in mathematics, but it will not bring the reader on a par with Roger Federer. Second, the idea of there being rules of interpretation would seem to suggest that behind the text, there lies, however deeply hidden perhaps, a single, unitary meaning that can be uncovered (excavated, more like) if only we have the right tools and dig hard and deep enough. This may or may not be plausible when it comes to literary texts. After all, these typically have but a single author, so at least it must be possible – in theory, and always assuming this will be of interest and that interpretation is the search for an author’s intended meaning – to figure out what that single author may have had in mind. It is tempting, of course, to follow a similar logic with respect to law, but doing so is only possible on the idea that law too has a single author whose ideas can meaningfully be excavated. To this end, we tend to think that contracts and treaties are the result of ‘meetings of the mind’. This now is a flawed starting point: treaties, at least, are better seen, in Philip Allott’s glorious
36 See North Sea Continental Shelf cases (Germany v. Denmark and the Netherlands), Judgment of 20 February 1969, 1969 ICJ Rep. 3, para. 72. 37 See Jan Klabbers, ‘On Rationalism in Politics: Interpretation of Treaties and the World Trade Organization’, Nordic Journal of International Law 74 (2005): 405–28.
26
jan klabbers
phrase, as “disagreement reduced to writing”.38 They are the result of compromise between actors with, often enough, widely diverging ideas, interests and desires. As a result, there is not a single author’s voice to be discerned: a treaty is by definition a compromise between the voices of a variety of authors – otherwise it would not exist. The same applies, mutatis mutandis, to decisions of international organizations and even judicial decisions: one of the reasons why judgments of the Court of Justice of the EU can be so dense is the same need for compromise between judges with widely varying ideas as to what the law says, what the relevant facts are, and how the two combine in the particular case. It is precisely to overcome the need for extreme compromise that some courts allow for separate or dissenting opinions; but the very circumstance of doing so suggest that disagreement is more likely to be expected than unanimity. Indeed, the same point has even been made with respect to domestic legislation which, as the outcome of a legislative process, could more easily be mistaken for having resulted from a single mindset. Thus, Radbruch already pointed out that no single person can claim authorship of laws: laws typically are the result of cooperation between lots of different people, in government, in parliament, and at ministries doing the preparatory work. A law, for Radbruch, is best conceptualized as embodying the will of the state, but the state is not reducible to any individual.39 Likewise, if more radical perhaps, political theorist Bonnie Honig points out that law always has an alien element: it is always made in conjunction with others, who do not quite share our own thoughts.40 Hence, she concludes, at least within democracies (but Radbruch suggests it goes farther than that) law is “the mongrel product of political action”.41 That need not be a bad thing – quite the opposite perhaps – but it does put to rest the idea of law having a single identifiable author. And that idea is even more untenable in international law than in domestic law. This is reflected in the history of rules of interpretation in international law. The classics typically present a laundry list of maxims, but refrain from laying down a single rule. Grotius, for instance, presents a great number of maxims,
38
See Philip Allott, ‘The Concept of International Law’, EJIL 10 (1999): 31–50, at 43. Teubner suggests much the same when discussing contract as “the legal form of a social relationship”. See Gunther Teubner, Law as an Autopoietic System (Oxford: Blackwell, 1993), 115. 39 See Gustav Radbruch, Einführung in die Rechtswissenschaft, (Stuttgart: Koehler Verlag, 1980), 283–4. As a result, Radbruch advocated a teleological approach, claiming that legal interpretation is really about finishing thoughts (‘Zu-Ende-Denken eines Gedachten”); this, however, might be seen again as presupposing some unitary thought – a single author – which only needs to be brought to its logical conclusion. 40 See Bonnie Honig, Democracy and the Foreigner (Princeton, NJ: Princeton University Press, 2001), 29–31. The one exception, it could be postulated, might be a dictatorship, but even here the voices of others cannot with impunity be ignored. See e.g., Lon Fuller, The Morality of Law, (New Haven, CT: Yale University Press, 1969). 41 Honig, supra note 40, at 39.
virtuous interpretation
27
including the best interpretation of carefully delimited sentences (how best to interpret “Carthage shall be free”42), but does not present a general rule, other perhaps than stating that interpretation is a quest for the author’s intent. This would seem to stem first and foremost from his starting point: his entire lengthy discussion of interpretation starts with the proper interpretation of promises, and promises are, after all, unilateral acts, which might make it plausible to presume single authorship. Much the same applies to Vattel: a long list of maxims, to be applied in accordance with the circumstances, but no general rule.43 Much the same applies to later writers as well. Thus, Prudhomme, writing in 1910 and with a particular emphasis on private international law, held that the application of treaties owed much to two conflicting principles (as he called them): the domestic law of the state concerned, and the precise content of the treaty concerned. He was perfectly willing to think of treaty interpretation in terms of theories, hoping to uncover a theory of interpretation, but not in terms of a single rule. Quite the opposite, in fact: he established that in different branches of private international law, the balance between national law and international agreement could fall differently, which if anything would make it impossible to derive at a single rule of interpretation.44 And someone like Sir Ernest Satow, in his monumental treatise on diplomatic law first published in 1917, paid a lot of attention to arranging and classifying the various forms international agreements can take (plus accompanying subtleties as to how seriously they ought to be taken), but none whatsoever to how best to read them – or write them, for that matter.45 All in all, it seems doubtful whether, prior to the Vienna Convention, there was indeed a general rule, let alone one of customary status. If anything, there
42 See Hugo Grotius, On the Law of War and Peace, (Oxford: Clarendon Press, 1925), Book II, Chapter XVI, at 417. This follows the logic of his discussion of legal instruments: he starts with promises, subsequently discusses contracts, and ends with treaties. 43 This calls for nuance. Vattel himself claimed to list a general rule of interpretation: all interpretation was the search for the intentions of the author(s) of the text. It is worth citing him in full:
Since the sole object of the lawful interpretation of a deed ought to be the discovery of the thoughts of the author or authors of that deed – whenever we meet with any obscurity in it, we are to consider what probably were the ideas of those who drew up the deed, and to interpret it accordingly. This is the general rule for all interpretations.
See Emmerich de Vattel, The Law of Nations (New York: AMS Press, reprint of 1863 edition), 247 (emphasis in original). To the extent that this is a general rule to begin with, it is a rule as to the purpose of interpretation, not on how to interpret (as he also realized: witness the phrase “and to interpret it accordingly”). Indeed, his ‘rule’ was preceded by a number of general maxims and followed by a fairly large number of more detailed instructions, taking up some 30 pages. 44 See Auguste Prudhomme, La Loi Territoriale et les Traités Diplomatiques devant les Juridictions des États Contractants (Paris: LGDJ, 1910). 45 See Sir Ernest Satow, A Guide to Diplomatic Practice (London: Longmans, 1917).
28
jan klabbers
was a bit of debate involving two different, but ever-so related things. First, there was debate on whether there would be a rule on interpretation to begin with (never mind what the rule would look like). Charles Cheney Hyde, for one, opined that the process of interpretation “is not fettered by many prohibitive rules”46, and quoted with much gusto a similar opinion by Westlake, who had written that the quest to find out the real intention of the parties “is not to be shackled by any rule of interpretation”.47 Short and sweet, Hyde concluded, “the formation of rules of interpretation can hardly serve a useful purpose”.48 Indeed, later through out the 20th century, the main debate turned around the evidential question: which documents can be admitted before a court or other decision-making body to shed light on the intention of the parties, and which should be excluded?49 While some favoured the full panoply of materials possible50, others presented more modest visions. Thus, Fachiri may have been among the first to advocate a restrictive approach to allowing extraneous evidence, on the theory that what mattered was the manifest intentions of the parties.51 Interestingly, he attributed the less restrictive approach to ‘continental jurists’, and juxtaposed these to ‘common law jurists’, in particular English lawyers. This would seem to suggest (hold the thought) that there is a connection between rules of interpretation and legal tradition, which in turn may represent a connection between interpretation and power: the eventual effect of a treaty depends, in part at least, on how it is interpreted. Hence, it becomes of relevance to control the process of interpretation, and in doing so, ‘our’ method is better (for us, at least), than ‘theirs’. There is a second debate, however implicit, which manifests itself throughout the twentieth century in particular. For Grotius, as noted, interpretation was the search for the intention of the promisor;52 and from this, he came to see interpretation of treaties, without being very explicit, as the search for the intention of the drafters. In this, he was followed by many generations: interpretation was, for a long time, conceptualized as the search for what the drafters had in mind. 46 See Charles Cheney Hyde, ‘Concerning the Interpretation of Treaties’, AJIL 3 (1909): 46–61, at 47. 47 Id. (citing Westlake). 48 Id., at 54. 49 See generally Jan Klabbers, ‘International Legal Histories: The Declining Importance of Travaux Préparatoires in Treaty Interpretation?’, NILR 50 (2003): 267–88. 50 Lauterpacht was willing to allow a lot in this respect, possibly to counter an overly rigid textualism in some of his contemporaries. For him, the treaty could only acquire meaning in light of its drafting history; the ordinary meaning of the text could give rise to little more than a presumption. See Sir Hersch Lauterpacht, ‘De l’Interprétation des Traités: Rapport’, Annuaire de L’Institut de Droit International 43 (1950/II): 366–460. 51 See Alexander P. Fachiri, ‘Interpretation of Treaties’, AJIL 23 (1929): 745–52. 52 “The measure of correct interpretation is the inference of intent from the most probable indications. These indications are of two kinds, words and implications; and these are considered either separately or together.” Grotius, supra note 42, at 409.
virtuous interpretation
29
Yet, over time, this came to be accompanied by two competing conceptualizations. Or, perhaps it is more accurate to state that the old idea split up into three distinct branches, according to which interpretation of a treaty could come to stand for three different things. In addition to finding out what the intentions of the drafters were when they agreed on a text, interpretation could also serve to elucidate the meaning of the text itself (thus, in relative isolation from what the drafters may have had in mind), or to elucidate what the drafters aimed to accomplish (thus, in relative extrapolation from what the drafters may have had in mind).53 Those three possibilities then having given rise, over time, to three competing approaches to interpretation: the textual approach aims at unveiling the meaning of the text; the historical approach (‘originalism’) aims to retrieve the intentions of the drafters, and the teleological school aspires first and foremost to do justice to the goals underlying the treaty. It would seem fair to say that proponents of all three can be found, both in the literature and in the relevant decisions of courts, before the conclusion of the Vienna Convention. As a result, it becomes difficult to argue that any of them stood out and came to be codified as pre-existing customary law in the Vienna Convention. Sorel sums it up nicely when he states that the ILC, drafting what would become articles 31 and 32 of the Vienna Convention, was faced with “des indices sérieux, mais pas toujours concordants, qui lui permettent d’élaborer une condification de l’interprétation sans la certitude que la base est fermement coutumière.”54 There is another, more fundamental, argument militating against too easily concluding that any rule on interpretation could establish customary law. If rules on interpretation be proper rules of law, it would follow that their transgressions could be classified as wrongful acts, in much the same way as returning a refugee to his homeland to face persecution would constitute a wrongful act, or committing torture would constitute a wrongful act.55
53 This division is borrowed from Sir Humphrey Waldock, supra note 6, at 53; it is also followed by Sir Ian Sinclair, The Vienna Convention on the Law of Treaties (Manchester: MUP, 1984), 114–5. Other authors come to different classifications, with arguably the most subtle being that proposed by Bos. Bos distinguishes five distinct approaches to interpretation, and makes further distinctions according to the intellectual approach taken (logical or comparative) and according to the effect of interpretation (restrictive or extensive). See Maarten Bos, A Methodology of International Law (Oxford: North Holland, 1984), 139. 54 See Jean-Marc Sorel, ‘Article 31 – Convention de 1969’, in Les Conventions de Vienne sur le Droit des Traités: Commentaire Article par Article, Olivier Corten & Puierre Klein (eds.) (Brussels: Bruylant, 2006), 1285–338, at 1301. A similar conclusion appears to be reached by Mark E. Villiger, Customary International Law and Treaties (Dordrecht: Martinus Nijhoff, 1985), 327–56, who continues, however, by suggesting that post-Vienna Convention practice suggests the emergence of a customary rule, but one which deviates from the general rule laid down in the Vienna Convention. 55 See likewise Shabtai Rosenne, ‘Conceptualism as a Guide to Treaty Interpretation’, reprinted in Shabtai Rosenne, An International Law Miscellany (Dordrecht: Martinus Nijhoff, 1993), 441– 55. Admittedly, this point owes much to a particular jurisprudential conception of
30
jan klabbers
Now suppose a police chief in, say, LaVerkin, Utah, blissfully unaware of any relevant case-law, is called upon to apply the Vienna Convention on Consular Relations which, as is well-known, grants foreign suspects a right to consular assistance. Suppose now our police chief interprets the relevant provision as stating that foreign suspects shall under no circumstances be granted consular assistance. Obviously, this interpretation of the text would be silly; obviously, it would result in a violation of the Convention attributable to the US; obviously, a decent case could be made to accuse the police chief of incompetence, obstruction, illiteracy, or downright lunacy. But it would be most unorthodox to construe this as not only a violation of the Consular Convention but also, on top of this, as a violation of the rule on interpretation. If one were to sue the police chief, it would be for violating the Consular Convention; if one were to sue the US before the ICJ, it would be for violating the Consular Convention, not for violating any interpretation rule.56 By the same token, while it would decidedly be unorthodox to interpret article 3 of the European Convention on Human Rights as allowing for torture as long it takes place only on Wednesdays,57 such would not be construed, in any meaningful way, as a violation of any interpretation rule. Instead, it would simply, and plausibly, be construed as a violation of the European Convention itself.58 The conclusion thus presents itself that no rule on interpretation could likely be of customary law nature;59 interpretation rules simply are of a different quality than ordinary norms of behaviour. Instead of guiding behaviour, rules on interpretation are methodological devices, suggesting to judges and international law in which rules are by definition rules of behaviour. If it be accepted that international law also knows other rules (secondary rules, in Hart’s terminology), then the point changes its character: it then becomes whether methodological rules (such as a rule on how best to read treaties) can ever be of customary nature: does it make sense, to involve a facile analogy, that there be a legal obligation, if one is in possession of a hammer and nails, to use the hammer for hammering in nails, and not use a rock for the same purpose? Or is this not so much a legal obligation but something of a different nature? The natural thing to do, no doubt, would be to use the hammer instead of the rock, but could one claim that it is legally wrong to use the rock? The reference, by the way, is to Herbert L.A. Hart, The Concept of Law (Oxford: Clarendon Press, 1961). 56 This is borne out by the German memorial submitted in the LaGrand case before the ICJ: it contains a detailed list of all alleged violations of the Vienna Convention on Consular Relations by the US, but at no point suggests that the rule on interpretation was violated. See <www.icj-cij.org/docket/files/104/8552.pdf> (accessed on 30 September 2008). 57 Article 3 provides, in fairly absolute terms: “No one shall be subjected to torture or to inhuman or degrading treatment or punishment”. 58 Moreover, as Rosenne has astutely observed, many allegations of breach of treaty can be recast as differences on the proper interpretation of the terms. See Shabtai Rosenne, Breach of Treaty (Cambridge: Grotius, 1985), 121. Letsas, supra note 20, makes much the same point when discussing what he refers to as ‘good faith violations’. 59 The WTO Dispute Settlement Understanding refers to interpretation of the WTO “in accordance with customary rules of interpretation of public international law”. While these words themselves are not as lucid as they could have been, this is usually taken as a reference to articles 31 and 32 of the Vienna convention. Yet, if what I am suggesting is correct, a nasty philosophical problem arises: if it is implausible to think of interpretation in terms of (legal) rules,
virtuous interpretation
31
other decision-makers how relevant legal instruments may be read. As such, they are immensely valuable; but they do not fit the regular pattern of rules. One cannot, in any meaningful way, violate them. IV. How the Vienna Convention Rules on Interpretation Are by their Very Nature of Little Concrete Help As the above illustrates, it is important to have some rules on interpretation. Those rules may not be of customary law nature (perhaps not even be of a legal nature if by that we mean that they ought to be followed lest we be subjected to legal sanctions), but are of great importance when it comes to guaranteeing the unity of international law – indeed, when it comes to guaranteeing the very relevance of international law. Without some guidelines on how to read treaties, things could easily derail.60 Still, the Mozart-anecdote mentioned above plus the conclusions it gave rise to also suggest that there are limits to what interpretation – and having rules on interpretation – can achieve: whatever the good that rules on interpretation can do, they cannot prevent that people come to a text with widely diverging background assumptions and knowledge packages; they cannot prevent that people approach a text from very different backgrounds – and this does not even factor in the will to power. At best, groups of people who have undergone similar training, who have been exposed to similar earlier experiences, and who have acquired similar sensibilities, may come to form an interpretive community: typically, people working in the field of water law will have some basis for reaching common understandings of, say, a convention on aquifers; the rest of us, not working in water law, probably do not even know what aquifers are. Thus put, a problem automatically emerges: if it is the case that interpretation only makes sense to certain audiences (fellow members of the interpretive community at hand) but not to others, does this not imply that international law is bound to be shattered after all? On such a reading, ‘fragmentation’ is still too benign a term, for the limits of interpretation would suggest that communication between discreet interpretive communities is excluded, at any rate as a practical matter. That may not be much of a problem when it concerns literary texts (and it is worth remembering that the idea of interpretive communities arose in the context of literary studies): it may be that not much of value is how then to interpret this reference in the WTO Dispute Settlement Understanding? To hold it to be devoid of content is not very satisfactory, and would presuppose (equally difficult to accept as a starting point) that the drafters had no idea what they were doing. Yet, to accept it as a distinct possibility (“a rule is a rule when enough people say it is”) also does not strike as ideal. 60 See generally also William Twining and David Miers, How to Do Things with Rules (London: Butterworths, 1991).
32
jan klabbers
lost when readers of Sartre cannot communicate with readers of Tolstoy.61 But in matters of international law this does become problematic: if environmental experts are fundamentally incapable of communicating with trade experts, or if the human rights community cannot talk to the security community, then something is amiss.62 Perhaps the most sophisticated theoretical account and response stem from social systems theory, in its several manifestations. Its auctor intellectualis, Niklas Luhmann, pioneered the notion of ‘structural couplings’ to discuss the way a system such as law interacts with its environment, but mostly he seems to have been thinking of interaction between broad social systems (law and politics, economics and society, law and economics, et cetera) rather than the interactions between subsystems – interpretive communities – inter se. The most he claimed about the internal workings of such broad systems was the suggestion that somehow, all specific legal branches must be read against a general background. Discussing the structural coupling between law and economy in ancient times, he fleetingly suggests (and I quote him somewhat out of context) that clinical isolation is not likely: “Property cannot be distinguished from family relations, and contracts, if one can call them that, are embedded in general obligations within a framework of balanced reciprocal relations.”63 Likewise, Teubner’s notion of law as an autopoietic system conceives law as a fairly coherent system juxtaposed to other social systems, rather than as a system comprising clashing subsystems.64 In a later work, however, Teubner and his co-author Fischer-Lescano sharpen the focus considerably, and posit the notion of ‘Vernetzung’ as a possible cure for the shattering of international law into a multitude of subregimes. ‘Vernetzung’, while not directly translatable, would boil down to the creation of networks between the various sub-regimes and sub-disciplines in order to manage conflicts between them without, perhaps, actually resolving those conflicts. The greatest achievement of ‘Vernetzung’, they write, is not that it helps dissolve intra-legal conflict, but rather that it einen unauflöslichen Widerspruch, der sich in der Normenkollisionen manifestiert, in ein tragbares Gegeneinander von verschiedenen Ebenen und
61 For argument’s sake, at least. I would suspect, however, that the proposition cannot be sustained at any rate: wouldn’t one at least expect Sartre readers to be able to communicate with those who read de Beauvoir or Camus? 62 And it is not implausible to discern distinct interpretive communities within recognized fields: marine pollution and climate change will by many considered as parts of international environmental law, but their respective practitioners may find communication with one another not all that easy. 63 See Niklas Luhmann, Law as a Social System (Oxford: OUP, 2004), 386. While the point on property may have been lost to history, the point on contract still seems appropriate, in all its generality. 64 See Teubner, supra note 38.
virtuous interpretation
33
Subsystemen, von Netzwerkknoten, Knotenrelationen und Gesamtvernetzung übersetzt.65
It is here then that rules on interpretation would seem to come in: the connecting points between the various sub-regimes can be formed by having uniform rules on interpretation. If all of us use the same interpretive devices, whether we are interpreting a convention on aquifers or rules on free trade, whether we are trying to make sense of a Security Council regulation or, instead, aim to make sense of the right to own property, then somehow the various subsystems can be seen to communicate with each other; they no longer exist, to paraphrase the classic dictum of the WTO’s Appellate Body, in ‘clinical isolation’ of each other.66 Or do they? There are several problems with this optimistic vision. One is, that strictly speaking, the mere circumstance that all interpreters use the same interpretive rule can never mean more than that they actually use the same interpretive device. The unity this seems to create is a mere simulacrum of unity, for it will still mean that trade rules will be given a trade-friendly interpretation, and human rights rules will be given a human rights interpretation. The interpretive community cannot escape from itself merely by using the same interpretive rules also used by others: it will still be self-referential; it will still bite itself in the tail.67 To posit a facile, but apt, analogy: to insist that everyone uses the same interpretive device is akin to suggesting that everyone should ride a bike, but without there being any agreement as to where all these people should cycle to or which side of the road they should use. There may be unity in the method of transportation, but not in the destination.68 This alone suggests that the insistence by all and sundry that articles 31 and 32 VCLT are to be applied, is based on an erroneous assumption, for if it is the case that if seasoned members of an interpretive community can come to radically different interpretations of one and the same text and can do so by both relying on the rules on interpretation, then those rules themselves contribute little to any elucidation.69 65
Andreas Fischer-Lescano and Gunther Teubner, Regime-Kollisionen: Zur Fragmentierung des globalen Rechts (Frankfurt am Main: Suhrkamp, 2006), 60. This calls for a (very liberal) translation, perhaps: the great virtue of ‘Vernetzung’ is that it translates an unsolvable contradiction which manifests itself in a collision of norms, into a manageable opposition of different levels and subsystems, of network nodes, nodal relations, and common embeddedness. 66
See US - Reformulated and Conventional Gasoline, supra note 11, at 17. The theme is played with in Panos Merkouris, ‘Debating the Ouroboros of International Law: The Drafting History of Article 31(3)(c)’, International Community Law Review 9 (2007): 1–31. 68 And to complete the analogy: if everyone is just biking without a sense of direction, then collisions will be well-nigh inevitable. 69 Already the European Court of Human Rights in Golder, possibly the first post-Vienna Convention judicial decision investigating the rules on interpretation in some depth, notes that 67
34
jan klabbers
Add to this the circumstance that the rule(s) of articles 31 and 32 are rather open-ended themselves – embodying, as they do, a compromise between various approaches which itself goes back to a compromise concerning the various distinct activities that treaty interpretation signifies – and it will be obvious that not too much ought to be expected from articles 31 and 32 as such. While it goes too far to suggest that ‘anything goes’ under these provisions, still, ‘quite a bit goes’ would be a fairly accurate synopsis. The most sophisticated observers have noted as much, of course, and have come to advocate a more modest variation. Instead of merely suggesting that everyone jump on their bikes, they also indicate at least some lanes which may have been cleared for bike use, and a general direction in which to cycle. For this group, article 31(3)(c) of the Vienna Convention should be singled out and regarded as the embodiment of a principle of systemic integration.70 Article 31(3)(c) specifies that when interpreting a treaty the interpreter should take into account “any relevant rules of international law applicable in the relations between the parties”. Apart from the interpretive problems this itself engenders (does it mean parties to that other treaty, or parties to the dispute at hand?71), at best and on its own it can only suggest that the sub-regimes be interpreted against the background of general international law – and it is in precisely this sense that the rule was relied on by the ICJ in Oil Platforms.72 But the rule cannot go further and prescribe communication between the various sub-regimes. This is most obviously visible when those subsystems themselves claim no longer to form part of international law: the EC is often considered a case in point. But this also applies in less extreme cases: as the WTO Biotech example makes clear,73 article 31(3)(c) justifies the building of a fence around a sub-system. As long as the interpreter is keen on protecting or safeguarding his or her own system, the principle of systemic integration is bound to remain toothless.
it could use those rules at the behest of both the United Kingdom and the then-extant European Commission of Human Rights (which, for all practical purposes, argued on behalf of the applicant). In other words: both parties thought their diametrically opposed conclusions could follow from one and the same rule of interpretation. See Golder v. United Kingdom, EHRR 1 (1975): 574, para. 29. Or maybe the Court just felt compelled to refer to articles 31 and 32 to take the wind out of the sails of the dissenting opinion discussing interpretative devices by Judge Sir Gerald Fitzmaurice. 70 The underlying ambitions are explored in Jan Klabbers, ‘Reluctant Grundnormen: Articles 31(3)(c) and 42 of the Vienna Convention on the Law of Treaties and the Fragmentation of International Law’, in Time, History and International Law, Matthew Craven, Malgosia Fitzmaurice and Maria Vogiatzi (eds.) (Leiden: Martinus Nijhoff, 2007), 141–61. 71 A WTO panel famously reached the former conclusion, thus unhelpfully holding that relevant rules of international law can only be taken into account if those rules apply to all WTO members. See EC – Measures Affecting the Approval and Marketing of Biotech Products (29 September 2006), Panel Report, WT/DS291.292.293. 72 See supra note 10. 73 See supra note 71.
virtuous interpretation
35
V. A Virtuous Approach to Interpretation If this is correct, then it would seem to follow that unity cannot be created by the mechanistic reliance on written rules alone, no matter how clever or wellwritten the rules in question may be. The main reason is, of course, that at the end of the day, rules or no rules, interpretation is still a human activity, depending on the efforts of human beings, their intellectual capacities, their sensibilities and, perhaps most of all, their sense of virtue. This is, to a lawyer, largely barren territory: indeed, to the extent that law aims to catch future behaviour in rules, its very aim is to make sure that social relations can unfold without having to rely on the peoples’ virtues. While it may be doubted whether Madison was right when he suggested that no government was necessary if only men were angels74 (some things need regulation regardless of individual virtue: if only to figure out which side of the road to cycle on), nonetheless a decent case can be made that large chunks of law are necessary because we cannot trust our virtues to do the work: if only as a fall-back, we may need to stipulate that contracts are binding75; we cannot be sure that others keep their word (or, more darkly, that we ourselves keep our word when confronted with tempting offers, as Ulysses realized all too well). In short, lawyers tend to focus on rules and principles, on setting external standards for behaviour. Yet, experience suggests that external standards alone cannot do the trick. Even the most established rule tends to get circumvented when there seems a decent justification for doing so; and even the hierarchically most superior rules get circumvented when the end seems to justify the means. And those are fairly ideal situations: in messy everyday life, it may often be that there is some uncertainty on the scope of the rule, or its field of application, or the precise behaviour it demands. In those circumstances, an appeal to external standards alone cannot suffice, for it is precisely this external standard that is in doubt. In a thoughtful recent book, constitutional theorist Jeff Powell set out a number of virtues that, he feels, ought to inform (and may well, as a practical matter, actually inform) judges when they make their decisions. To his mind, judging is not a matter of simply applying rules to facts or, more sophisticated (albeit slightly so) simply applying personal politics to cases. Instead, judging is construed as a moral activity, tapping into a “judge’s individual obligations
74 See James Madison, ‘The Federalist No. 51’, in The Federalist Papers, Alexander Hamilton, James Madison and John Jay, (New York: Bantam Books, 1982), 261–65. 75 Practice actually bears this out: as Macauley observed a long time ago, businessmen use contract law as a safety-net rather than as the sole guide for their actions. See Stewart Macauley, ‘Non-contractual Relations in Business: A Preliminary Study’, American Sociological Review 28 (1963): 55–67.
36
jan klabbers
as a moral actor.”76 A judge (or other interpreter – there is no particular need to stick to the setting in which Powell writes) needs to make choices between often equally plausible or possible alternatives, and those choices, Powell insists, are moral choices.77 Powell identified a number of virtues. Interpreters should be humble about their own roles, and not be conceited enough to think that their interpretation is the only possible one: the interpretation to end all interpretations. Interpreters should pay heed to past decisions and interpretations: they cannot just overturn them because they happen to disagree (Powell speaks of acquiescence). Interpreters should display integrity about reaching their conclusions, and should show candour in making transparent the reasons for their judgment.78 Much the same can be said about acts of interpretation generally. While judges, as guardians of legality (or, in Powell’s setting, guardians of the US Constitution) may have to be especially virtuous, Powell’s list of requirements may come in handy with respect to everyday interpretation of treaties and other international legal instruments by courts and tribunals as well as policymakers and civil servants. Its main attraction is that it recognizes the art and politics of interpretation (indeed, he specifies that one of the normative bases of his approach is a recognition of the primacy of the political79) and does not try to hide these behind a thin, rapidly peeling, veneer of technical craftsmanship. This will not solve all disagreement, obviously, and neither is it meant to. The main idea is to recognize what others have referred to as the ‘morality of disagreement’, ever so pressing in a pluralist world of some 200 states and more than six billion people. As a result, at best one can hope for political debate, open and without obstacles between participants on an equal footing. This calls for rules on how to conduct the debate, of course, but rules of interpretation cannot play that role, for their main function is to stifle today’s debate in the name of yesterday’s (seeming) agreement. VI. Concluding Remarks If it is true that the way texts are read owes as much to the audience as it does to the author, then it follows that creating rules on how to read those texts alone will not be of much help: one would also need to have some guidelines on who gets to read those texts. Bluntly put, what matters is not so much how
76 H. Jefferson Powell, Constitutional Conscience: The Moral Dimension of Judicial Decision (Chicago, IL: University of Chicago Press, 2008), 3. 77 In a sense, he picks up where Burton leaves off: see Steven J. Burton, Judging in Good Faith (Cambridge: CUP, 1992). 78 Powell, supra note 76, esp. at 107. 79 Id., at 110.
virtuous interpretation
37
to interpret, but who has the power to do so, as generations of lawyers, advocating the creation of more and more international courts and tribunals, have intuitively realized. Yet, curiously enough, with the creation of all those courts and tribunals, it would seem that the call for rules on interpretation has not diminished: quite the contrary, it has increased. This would seem to suggest a painful black hole somewhere, a black hole that cannot be filled by creating yet more rules and yet more courts and tribunals: indeed, creating yet more rules and yet more tribunals would only result in more of the same problem. The reason for this is that such remedies focus, so the speak, on the wrong end of the equation: the psychology of government, as Sir Ivor Jennings observed in the midst of a world war, is as important as the precise rules at issue.80 Any rule, any treaty, can be misread; it is for this reason that the pacta sunt servanda rule not merely stipulates that treaties are binding, but also that they shall be performed in good faith. And needless to say, this is also why the notion of good faith is included in the general rule on interpretation as laid down in article 31. It is perhaps time to re-discover this particular element of the rule: the reference to the virtues – the good faith – of the interpreter. This suggests that individuals should approach the interpretation of texts with humility, acquiescence, integrity and candour, as Powell proposes, rather than with the mechanistic sense of applying a rule as if it were mathematics. What matters is not virtuoso technique, but virtuous reading. This is nothing new, of course: virtue ethics as such is often traced back to Aristotle,81 and more recent authorities too come close enough to advocating something along these lines.82 But the fact that there is nothing new about insisting on the integrity of the interpreter makes doing so all the more relevant.
80 See Ivor W. Jennings, The Law and the Constitution, (London: University of London Press, 1943), at xxxi. 81 A useful collection of articles is Roger Crisp and Michael Slote, Virtue Ethics (Oxford: OUP, 1997), whereas some of the classic texts are usefully assembled in Jennifer Welchman, The Practice of Virtue: Classic and Contemporary Readings in Virtue Ethics (Indianapolis: Hacket, 2006). 82 See e.g., Koskenniemi’s use of Kant’s moral politician in Martti Koskenniemi, ‘Constitutionalism as Mindset: Reflections on Kantian Themes in International Law and Globalization’, Theoretical Inquiries in Law 8 (2007): 9–36. See also in a similar vein arguing that the end should not always be allowed to justify the means, Jan Klabbers, ‘The Commodification of International Law’, Select Proceedings of the European Society of International Law 1 (2006): 341–58.
TREATY, CUSTOM AND TIME: INTERPRETATION/APPLICATION?* Philippe Sands and Jeffery Commission “Yet in speaking of application it was necessary to speak of interpretation.”1 “He did not think it was possible to draw a distinction between the interpretation of a treaty and its application, as the Special Rapporteur had attempted to do: for once a treaty had been correctly interpreted, i[t] had to be applied according to that interpretation. ”2 “… it was hard to draw the dividing line between interpretation and application.”3
I. Introduction Article 31(3)(c) of the 1969 Vienna Convention on the Law of Treaties directs that in interpreting a treaty “there shall be taken into account, together with the context … any relevant rules of international law applicable in the relations between the parties”. Six years ago I wrote an article entitled ‘Treaty, Custom and the Cross-fertilization of International Law’, which tried to identify and assess international judicial and arbitral practice on this issue.4 The article was based on a reasonably thorough trawl of the case law and commentary, and concluded that Article 31(3)(c) appeared “to have been expressly relied upon only very occasionally in judicial practice”, and had “attracted little academic comment”.5 Put simply, there appeared to be “a general reluctance to refer to Article 31(3)(c)”.6 The ICJ had not looked at the issue, aside from Judge Weeramantry’s * This Chapter was based on the presentation made by the author during the ‘First Annual Conference on International Law: Interpretation under the VCLT’ organized by Queen Mary and Eversheds in 2006, and is current as of that date. 1 Comments of Mr. Yasseen at the 729th ILC Meeting on the Third Report on the Law of Treaties by Sir Humphrey Waldock, on the proposed Article 56 (The inter-temporal law); Comments of Mustafa Kamil Yasseen (A/CN.4.SR.729), reproduced in YILC (1964): Vol.I at 37, para. 35. 2 Comments of Mr. Verdross at the 728th ILC Meeting on the Third Report on the Law of Treaties by Sir Humphrey Waldock, on the proposed Article 56 (The inter-temporal law); Comments of Alfred Verdross (A/CN.4.SR.728), reproduced in YILC (1964): Vol. I at 33, para. 6. 3 Comments of Mr. Jimenez de Arechaga at the 728th ILC Meeting on the Third Report on the Law of Treaties by Sir Humphrey Waldock, on the proposed Article 56 (The inter-temporal law); Comments of Jimenez de Arechaga (A/CN.4.SR.728), reproduced in YILC (1964): Vol.I, at 34, para. 11. 4 Philippe Sands, ‘Treaty, Custom and the Cross-Fertilization of International Law’, Yale Human Rights and Development Law Journal 1 (1999): 85. 5 Id., at 95. 6 Id.
40
philippe sands and jeffery commission
observations in his separate opinion in the Gabcikovo-Nagymoros case, that the ability of Article 31(3)(c) to address relationships between treaty and custom “scarcely covers this aspect with the degree of clarity requisite to so important a matter”.7 It was not yet at the forefront of the ILC’s agenda. In concluding that research effort, I proposed that Article 31(3)(c) be focused on as “an available tool”, with the recognition that [s]hort of the International Law Commission providing greater clarity on its utilization—an unlikely prospect in the short term—it will be for judicial and quasi-judicial bodies to take the lead.8
No doubt for unrelated reasons, greater attention has been paid to the provision in recent years. International courts and tribunals have begun to focus on its requirements, occasionally invoking Article 31(3)(c);9 the ILC has referred to it as part of its study on the fragmentation of international law;10 and scholars and practitioners have also discussed it.11 A significant development – and perhaps even a tipping point – was the ICJ’s judgment on the merits in the Oil Platforms case.12 As noted by one commentator, the ICJ’s judgment in Oil Platforms “has shone a searchlight onto one of the most neglected corners of the interpretation section of the Vienna Convention, namely Article 31(3)(c)”.13 The searchlight has identified a number of difficulties. In particular, despite the apparent quasi-unanimity on the principal legal issues before the Court in Oil Platforms (14–2, and 15–1, respectively), the eleven separate opinions, declarations and dissents reflect a crucial difference of approach with respect to the relationship between interpretation and application under Article 31(3) (c). This is not surprising, since the scope of Article 31(3)(c) has long 7 Gabcikovo-Nagymaros Project (Hungary v. Slovakia), Judgment of 25 September 1997,1997 ICJ Rep. 7 at 114. 8 Sands, supra note 4, at 105. 9 The Iran-United States Claims Tribunal, the European Court of Human Rights, Arbitral Tribunals established pursuant to multilateral agreements, and the WTO Appellate Body. 10 ILC, Report of the International Law Commission (58th Session, 1 May-9 June and 3 July-11 August 2006) (A/61/10): 413–6 (hereinafter ILC Report); ILC Study Group on Fragmentation, Fragmentation of International Law: Difficulties Arising from the Diversification and Expansion of International Law, Report of the Study Group, finalized by Martti Koskenniemi, (13 April 2006) (A/6N.4/L.682): 206–44, at paras. 410–80 (hereinafter ILC Study Group Final Report); ILC Study Group on Fragmentation, Fragmentation of International Law: Difficulties Arising from the Diversification and Expansion of International Law, (29 July 2005) (A/CN.4/L.676) (hereinafter ILC Study Group Preliminary Report). 11 Campbell McLachlan, ‘The Principle of Systemic Integration and Article 31(3)(c) of the Vienna Convention’, ICLQ 54 (2005): 279; Anja Lindroos and Michael Mehling, ‘Dispelling the Chimera of ‘Self-Contained Regimes’: International Law and the WTO’, EJIL 16 (2005): 857; Alan Boyle, ‘Further Development of the Law of the Sea Convention: Mechanisms for Change’, ICLQ 54 (2005): 563. 12 Oil Platforms case (Iran v. United States of America), Merits, Judgment of 6 November 2003, 2003 ICJ Rep. 161, reprinted in ILM 42 (2003): 1334. 13 McLachlan, supra note 11, at 279.
treaty, custom and time: interpretation/application?
41
focused minds on the question of where interpretation ends and application begins. The concerns of Messrs. Jimenez de Arechaga, Verdross, and Yasseen, amongst others, that are set out at the beginning of this lecture date back to the ILC debates of more than forty years ago on a related draft article dealing with the issue of inter-temporal law, ultimately omitted from the convention.14 The related draft article, article 56 (‘The inter-temporal law’), last appeared in the travaux prepararoires to the Vienna Convention in Sir Humphrey Waldock’s Third Report on the law of treaties in 1964. It specifically provided that “the application of a treaty shall be governed by the rules of international law in force at the time when the treaty is applied”.15 While the differences voiced more than forty years ago were initially in respect of draft article 56, the issues went on to provoke controversy in discussions of Article 31(3)(c), and are equally pertinent today.16 In short, at what point does the interpretation of a treaty by reference to other rules of international law become the application of those other rules of international law?17 It is an important question which permits of no easy answer. This paper considers the decision rendered in the Oil Platforms case, and explores how it and how 14
The text of what is now Article 31(3)(c) began as Article 70(1)(b) in Waldock’s Third Report on the law of treaties in 1964: the terms of the treaty shall be interpreted in good faith in accordance with the natural and ordinary meaning to be given to each term -… [and] (b) in the context of the rules of international law in force at the time of the conclusion of the treaty.
While the idea that treaties should be interpreted in the context of the rules of international law was ‘taken for granted’, [a]ll the discussion and controversy in the Commission was addressed to the second part of the provision – namely the suggestion that that normative environment should be constructed on the basis of the law in force at the moment of the conclusion of the treaty;
ILC Study Group Final Report, supra note 10, at 216–7, paras. 430–1. 15 Draft Article 56, The inter-temporal law, provided as follows: 1. A treaty is to be interpreted in the light of the law in force at the time when the treaty was drawn up. 2. Subject to paragraph 1, the application of a treaty shall be governed by the rules of international law in force at the time when the treaty is applied.
Sir Humphrey Waldock, Third Report on the Law of Treaties, (A/CN.4/167 & Add.1–3), reproduced in YILC (1964): Vol. II, 5, at 8. 16 Id. 17 See Lord Arnold McNair, The Law of Treaties (Oxford: Clarendon Press, 1961), at 365; The words ‘interpret’, ‘interpretation’ are often used loosely as if they included ‘apply, application’. Strictly speaking, when the meaning of the treaty is clear, it is ‘applied’, not ‘interpreted’. Interpretation is a secondary process which only comes into play when it is impossible to make sense of the plain terms of the treaty, or when they are susceptible of different meanings.
As observed recently by one commentator, [t]here is a crucial distinction between interpretation and application, which both the Court [in Oil Platforms] and certain individual judges seemed to blur or, at least, appeared to do so within the language of their reasoning.
Duncan French, ‘Treaty Interpretation and the Incorporation of Extraneous Legal Rules’, ICLQ 55 (2006): 281, at 290.
42
philippe sands and jeffery commission
Article 31(3)(c) have been received by courts and tribunals in the time since the judgment. II. The Oil Platforms Case A. Factual Background The case originated from the November 1992 application by the Islamic Republic of Iran to the International Court of Justice instituting proceedings against the Government of the United States in respect of a dispute aris[ing] out of the attack [on] and destruction of three offshore oil production complexes, owned and operated for commercial purposes by the National Iranian Oil Company, by several warships of the United States Navy on 19 October 1987 and 18 April 1988, respectively.
Iran contended that these actions constituted a ‘fundamental breach’ of various provisions of the 1955 Treaty of Amity, Economic Relations and Consular Rights between the United States and Iran, as well as of international law.18 The Application invoked, as a basis for the Court’s jurisdiction, Article XXI(2) of the 1955 Treaty.19 The United States contended that the actions of its military did not breach the Treaty, and proceeded to institute a counter-claim alleging that Iran’s military actions in the Persian Gulf amounted to a breach of the Treaty. In its Judgment of 12 December 1996, the Court rejected the preliminary objections of the US according to which the 1955 Treaty did not provide any basis for the jurisdiction of the Court and found that it had jurisdiction, on the basis of article XXI(2), of the 1955 Treaty, to entertain the claims made by Iran under Article X(l), of that Treaty.20
18 Specifically, Iran claimed that the attacks breached Articles I (providing for “firm and enduring friendship between the parties”) and X (providing for “freedom of commerce and navigation”) of the 1955 Treaty, and international law. During Iran’s oral submissions, the arguments concerning purported violations of international law were dropped. 19 Article XXI, paragraph 2, of the 1955 Treaty provides:
Any dispute between the High Contracting Parties as to the interpretation or application of the present Treaty, not satisfactorily adjusted by diplomacy, shall be submitted to the International Court of Justice, unless the High Contracting Parties agree to settlement by some other pacific means. 20 Article X, paragraph 1, of the 1955 Treaty provides: “Between the territories of the two High Contracting Parties there shall be freedom of commerce and navigation”. Article XX(l)(d), upon which the US asserted that its actions were necessary to protect its national security, provides:
The present treaty shall not preclude the application of measures: …(d) necessary to fulfill the obligations of a High Contracting Party for the maintenance or restoration of international peace and security, or necessary to protect its essential security interests.
treaty, custom and time: interpretation/application?
43
B. The Judgment on the Merits The Court was called upon to interpret two provisions of 1955 Treaty of Amity, Economic Relations and Consular Rights between the United States and Iran. Specifically, the Court was requested to determine whether actions by Iran which were alleged to imperil neutral commercial shipping in the Iran-Iraq war, and later destruction by the US Navy of three Iranian oil platforms in the Persian Gulf were breaches of the Treaty. The Judgment of the Court is in two parts. First, the Court, by fourteen votes to two (Judges Al-Khasawneh and Elaraby dissenting): (i) found that the actions of the USA against Iranian oil platforms on 19 October 1987 and 18 April 1988 could not be justified as measures necessary to protect the essential security interests of the USA under Article XX(l)(d) of the 1955 Treaty of Amity, Economic Relations and Consular Rights between the USA and Iran, as interpreted in the light of international law on the use of force; (ii) found further that the Court could not uphold Iran’s submission that those actions constitute a breach of the USA’s obligations under Article X(l), of that treaty, regarding freedom of commerce between the territories of the parties, and that, accordingly, Iran’s claim for reparation could not be upheld. Second, the Court, by fifteen votes to one (Judge Simma dissenting), found that the counter-claim of the USA concerning the breach of the obligations of Iran under Article X(l) of the 1955 Treaty, regarding freedom of commerce and navigation between the territories of the parties, could not be upheld; and accordingly, that the counter-claim of the USA for reparation could not be upheld. In addressing the first question – whether the US attacks could be justified under Article XX(l)(d) – the Court stated that interpretation of the 1955 Treaty must take into account ‘any relevant rules of international law applicable in the relations between the parties’ (Article 31(3)(c) ). Specifically, the Court stated that: The Court cannot accept that Article XX, paragraph 1(d) of the 1955 Treaty was intended to operate wholly independent of the relevant rules of international law on the use of force, so as to be capable of being successfully invoked, even in the limited context of a claim for breach of the Treaty, in relation to an unlawful use of force. The application of the relevant rules of international law relating to this question thus forms an integral part of the task of interpretation entrusted to the Court by Article XXI, paragraph 2, of the 1955 Treaty.21
Despite the quasi-unanimity with respect to decisions on the legal claims before the Court (14–2, and 15–1, respectively), a divergence of views on the
21
Oil Platforms case, supra note 12, at para. 41.
44
philippe sands and jeffery commission
question of the appropriate approach to interpretation manifested itself in two declarations, seven separate opinions, and two dissenting opinions. In brief, [t]he criticism of the Court’s reasoning was intense and the distinction between separate and dissenting opinions, while technically existing, was not reflected in the style or content of these individual opinions.22
However, of the eleven individual views,23 only four expressly refer to Article 31(3)(c), namely the views of Judges Higgins, Buergenthal, Simma and Elaraby. Each merits separate consideration. Judge Higgins, who had previously examined Article 31(3)(c), took issue with the Court’s approach.24 While acknowledging that it is “commonplace that treaties are to be interpreted by reference to the rules enunciated in Article 31 of the Vienna Convention on the Law of Treaties”, she maintained that the Court had read this provision as incorporating the totality of the substantive international law the use of force.25 In doing so, she rightly observed that approach ignored the fact “that Article 31, paragraph 3, requires ‘the context’ to be taken into account: and ‘the context’ is clearly that of an economic and commercial treaty”.26 She added: What is envisaged by Article 31, paragraph 3(c), is that a provision that requires interpretation in Article XX, paragraph 1(d), will be illuminated by recalling what type of a treaty this is and any other ‘relevant rules’ governing Iran-United States relations. It is not a provision that on the face of it envisages incorporating the entire substance of international law on a topic not mentioned in the clause – at least not without more explanation than the Court provides.27
Judge Higgins considered that the Court had not interpreted Article XX(l)(d) by reference to the rules on treaty interpretation, but rather had “invoked the concept of treaty interpretation to displace the applicable law”. In so doing “[i]t has replaced the terms of Article XX, paragraph 1(d), with those of
22 Jorg Kammerhofer, ‘Oil’s Well that Ends Well? Critical Comments on the Merits Judgment in the Oil Platforms Case’, Leiden Journal of International Law 17 (2004): 695, at 716. 23 Of the 16 Judges, 11 appended either a separate opinion or declaration—Vice-President Ranjeva and Judge Koroma appended declarations to the Judgment of the Court; Judges Higgins, Parra-Aranguren, Kooijmans, Buergenthal, Owada, and Simma appended separate opinions to the Judgment of the Court; Judges Al-Khasawneh and Elaraby appended dissenting opinions to the Judgment of the Court; and Judge ad hoc Rigaux appended a separate opinion to the Judgment of the Court. 24 Rosalyn Higgins, ‘Time and Law: International Perspectives on an Old Problem’, ICLQ 46 (1997): 501, at 519;
The Vienna Convention on the Law of treaties as it finally emerged itself contains no general rule on the inter temporal question. The only hint one gets is in Article 31, which provides simply that treaties are to be interpreted in accordance with any relevant rules of law. 25 26 27
Oil Platforms case, supra note 12, para. 45. Id., para. 46. Id.
treaty, custom and time: interpretation/application?
45
international law on the use of force and all sight of the text of Article XX, paragraph 1(d) is lost”.28 Judge Buergenthal was equally critical of the Court’s approach, focusing on the jurisdictional restraints on the Court’s freedom of treaty interpretation, given the consensual nature of the Court’s jurisdiction.29 Put simply, he maintained that the substantive rules of international law, conventional or customary, could not be brought into the litigation through the back door by invoking Article 31, paragraph 3(c), of the Vienna Convention on the Law of Treaties in the absence of specific jurisdiction conferred by the Parties on the Court to rule on them.30
He added: the problem with the Court’s reliance on this provision of the Vienna Convention is that, while the rule is sound and undisputed in principle as far as treaty interpretation is concerned, it cannot have the effect of allowing the Court to take account, as it does here, of those ‘relevant rules of international law applicable between the parties’, which the parties to the dispute have not submitted to the Court’s jurisdiction under the dispute resolution clause of the 1955 Treaty. That is, the principles of customary international law and whatever other treaties the parties to a dispute before the Court may have concluded do not by virtue of Article 31, paragraph 3(c) become subject to the Court’s jurisdiction.31
Judge Simma refers to Article 31 (3)(c) in a footnote to his opinion, agreeing that the Court accepts, and rightly so, the principle according to which the provisions of any treaty have to interpreted and applied in the light of the treaty law applicable between the parties as well as of the rules of general international law ‘surrounding’ the treaty.32
Judge Simma found it ‘regrettable’ that the Court did not restate and reconfirm the fundamental principles of the law of the United Nations as well as customary international law on the use of force.33 However, Judge Simma acknowledged that since the Court’s jurisdiction “is limited to the bases furnished by the 1955 Treaty”, “it would not have been possible for the Court to go as far as stating in the dispositif of its Judgment”.34 While dissenting on the first question before the Court, Judge Elaraby nevertheless agreed with the Court’s approach on Article 31(3)(c), stating that the Court “rightly observed” in paragraph 41 that: 28 29 30 31 32 33 34
Id., para. 49. Id., para. 28. Id. Id., para. 22. Id., para 9. Id., para 6. Id.
46
philippe sands and jeffery commission Moreover, under the general rules of treaty interpretation, as reflected in the 1969 Vienna Convention on the Law of Treaties, interpretation must take into account ‘any relevant rules of international law applicable in the relations between the parties’ (article 31, paragraph 3(c)). The Court cannot accept that Article XX, paragraph 1(d) of the 1955 Treaty was intended to operate wholly independent of the relevant rules of international law on the use of force, so as to be capable of being successfully invoked, even in the limited context of a claim for breach of the Treaty, in relation to an unlawful use of force. The application of the relevant rules of international law relating to this question thus forms an integral part of the task of interpretation entrusted to the Court by Article XXI, paragraph 2, of the 1955 Treaty.35
While the remaining judges did not expressly invoke or cite Article 31(3)(c) in their respective separate opinions, a number of them offered comments about the use of general international law for purposes of interpreting the 1955 Treaty. Judge Kooijmans stated that the Court had no choice but to rely “on the body of general international law” as “[g]eneral international law is therefore indispensable as a standard of interpretation of the provisions of the 1955 Treaty”.36 Yet, Judge Kooijmans criticized the court’s approach in considering the actions of the United States “exclusively in the light of the right of selfdefense and returns only at a rather late stage to the terms of Article XX, paragraph 1(d)”.37 Judge Koroma agreed that the Court held, “rightly” in his view, “that the application of general international law on the question forms part of the interpretation process”.38 In contrast, Judge Parra-Aranguren held that the Court did not have jurisdiction to examine the defenses advanced by the United States on the basis of Article XX, paragraph 1(d), to justify its hypothetical breach of Article X, paragraph 1, of the 1955 Treaty.39
Judge Owada observed that the Court is “entitled” to get into an examination of the scope and the relevance of the rules of general international law relating to the use of force, but only to the extent that such examination, ancillary to the examination of Article XX, paragraph 1(d), is found to be necessary for clarifying the interpretation and application of Article XX, paragraph 1(d).40
Judge Al-Khasawneh opined that he did not feel that the ultra petita rule was infringed nor that the concept of lex specialis (assuming the 1955 Treaty was one) would operate to exclude the operation of rules of international law that have a peremptory character.41 35 36 37 38 39 40 41
Id., para. 3.2. Id., paras. 48–9. Id., para. 24. Id., at 1378. Id., para. 14. Id., para. 14. Id., para. 9.
treaty, custom and time: interpretation/application?
47
C. The Legacy of The Oil Platforms Case Much has been written on the Oil Platforms case42 The decision has been described as: “[a] disappointment”,43 “quite unusual in its logic and structure”,44 “a brave, but flawed and ultimately dangerous move by the Court”,45 “[a] fascinating and confusing mix of formal consensus and substantive dissensus”,46 and “a roller coaster of an opinion”.47 It has itself been the subject of numerous symposia, and been analyzed and critiqued by counsel for both parties to the case and by practitioners and scholars. The importance of the Court’s reference to Article 31(3)(c) in Oil Platforms, and the fundamental differences in interpretive approach it generated, go well beyond the facts and disposition of the dispute itself. The heart of the discord between the judges in Oil Platforms seems to turn on competing conceptions as to the nature of the international legal order and the function of the international judiciary. It may also reflect the “differences in legal training and educational background of the judges”.48 The United States “argued that paragraph 1(d) has an autonomous field of operation, totally distinct from general international law”, going so far as to argue “that the Treaty of Amity was a ‘self-contained regime’ ”.49 Yet, as noted by one of Iran’s counsel, “[t]he Treaty of Amity is not a self-contained regime; it is a normal bilateral treaty governed by international law”.50 While not specifically invoking Article 31(3)(c), Iran referred the court to a passage from the decision of the Iran-United States Claims Tribunal in Amoco International Finance case: As a lex specialis in the relations between the two countries, the Treaty [of Amity] supersedes the lex generalis, namely customary international law. This does not
42
Oil Platforms case, supra note 12. Andrew Garwood-Gowers, ‘Did the ICJ Miss the Boat on the Law on the Use of Force?’, Melbourne Journal of International Law 5 (2004): 241, at 254. 44 David H. Small, ‘The Oil Platforms Case: Jurisdiction Through The – Closed-Eye of the Needle’, The Law and Practice of International Courts and Tribunals 3 (2004): 112, at 113. 45 James A.Green, ‘The Oil Platforms Case: An Error in Judgment?’, Journal of Conflict and Security Law 9 (2004): 357, at 386. 46 Veijo Heiskanen, ‘Oil Platforms: Lessons of Dissensus’, Nordic Journal of International Law 74 (2005): 179. 47 Harvey Rishikof, ‘When Naked Came the Doctrine of ‘Self-Defense’: What is the Proper Role of the International Court of Justice in Use of Force Cases?’, Yale Journal of International Law 29 (2004): 331. 48 Heiskanen, supra note 46, at 11, para. 27; 43
Those hailing from the common law tradition seem more inclined to define the Court’s function in terms of adjudication, where those with civil law and other similar more policy-oriented backgrounds tend to define it in terms of dispute settlement. 49 Oil Platforms case, Compte Rendu 2003/16 (Monday, 3 March 2003), at 12, para. 6 (citing Oil Platforms case, Compte Rendu 2003/12 (Wednesday, 26 February 2003), at 18, para. 17.20). 50 Id.
48
philippe sands and jeffery commission mean, however, that the latter is irrelevant in the instant Case. On the contrary, the rules of customary law may be useful in order to fill in possible lacunae of the Treaty, to ascertain the meaning of undefined terms in its text of, more generally, to aid interpretation and implementation of its provisions.51
To achieve the conclusion it wanted the Court had to invoke a standard independent from the 1955 Treaty, for which it relied on Article 31(3)(c). Whether the Court did so to “stress its relevance in the 21st century”,52 to “send a message to the world’s remaining superpower as to the limits on its military powers”,53 or to confirm “that the rules, principles and categories of international judicial competence—based on, or inspired by, the consensual principle—are not carved in stone”,54 the impact of the decision is significant in suggesting a desire for unity and coherence in the international legal order. As discussion at the ILC has subsequently put it, Article 31(3)(c) embodies a “principle of systemic integration”, a guideline according to which treaties should be interpreted against the background of all the rules and principles of international law—in other words, international law understood as a system.55
Moreover, the fact that “[t]his is the first time that the Court has expressly used this rule of interpretation”, indicates that it “must be assessed as an action in favour of the unity of the international legal system”.56 III. The Operationalization of 31(3)(c) A. WTO Appellate Body57 Article 31(3)(c) has previously been resorted to in certain Appellate Body proceedings, prior to the Oil Platforms case, including in the Appellate Body Reports in United States-Import Prohibition of Certain Shrimp and Shrimp Products, and in European Communities-Measures Concerning Meat and Meat Products (Hormones).58 This trend continues in both the arguments of parties
51 Amoco International Finance Corporation v. Iran (1987) Iran-US CTR 15: 189, at 222, para. 112. Judge Kooijmans referred to this passage in his separate opinion, at para 23. 52 Andreas Laursen, ‘The Judgment by the International Court of Justice in the Oil Platforms Case’, Nordic Journal of International Law 73 (2004): 160. 53 Pieter H.F. Bekker, Oil Platforms (Iran v. United States), AJIL 98 (2004): 550, at 557. 54 Alexander Orakhelashvili, Oil Platforms (Islamic Republic of Iran v. United States), Merits, Judgment of 6 November 2003’, ICLQ 53 (2004): 753, at 761. 55 ILC Study Group Preliminary Report, supra note 10, at 11, para. 27. 56 Natalia Ochoa-Ruiz and Esther Salamanca-Aguado, ‘Exploring the Limits of International Law relating to the Use of Force in Self-defence’, EJIL 16 (2005): 499, at 509. 57 See Lindroos and Mehling, supra note 11; ILC Study Group Final Report, supra note 10 at 223–8, paras. 443–50. 58 US – Import Prohibition of Certain Shrimp and Shrimp Products (6 November 1998), Appellate Body Report, WT/DS58/AB/R; EC – Measures Concerning Meat and Meat Products
treaty, custom and time: interpretation/application?
49
and the reports of panels and of the Appellate Body, despite the absence of any direct reference to the ICJ’s decision. 1. European Communities – Customs Classifications of Frozen Boneless Chicken Cuts59 In a September 2005 report, the Appellate Body discussed Article 31(3)(c). The European Communities, as appellant, and Brazil and Thailand, as other appellants, each appealed certain issues of law and legal interpretations developed in the Panel Reports. The Panel was initially established to consider complaints by Brazil and Thailand concerning certain measures of the European Communities, pertaining to the classification of frozen boneless salted chicken cuts for tariff treatment. In the course of determining what constituted context for interpreting the term “salted” in heading 02.10 of the EC Schedule, the Panel received submissions from Brazil, Thailand and the European Communities: (a) Brazil argued that that the Harmonized System constituted ‘context’ within the meaning of Article 31(2)(b) of the Vienna Convention; (b) Thailand maintained that the Harmonized System qualified as context under Article 31(1) or under Article 31(3)(c); and the (c) European Communities held the view that that Harmonized System was a ‘relevant rules of international law’, within the meaning of Article 31(3)(c). The Panel did not definitively decide which was position was correct, but rather decided to treat the Harmonized System “as if it qualifies as ‘context’ under Article 31(2)”60 Before the Appellate Body, the parties argued as follows: 194. … in response to questioning at the oral hearing, all participants, and the United States, as third participant, agreed that the Harmonized System was relevant for interpretation of the terms of the EC Schedule. The participants took the view that the Harmonized System was context for the purposes of interpreting the terms of the EC Schedule or a ‘relevant rule of international law applicable in the relations between the parties’ that should be ‘taken into account, together with the context’, pursuant to Article 31(3)(c) of the Vienna Convention. The United States, as third participant, was of the view that the Harmonized System
(Hormones) (13 February 1998), Appellate Body Report, WT/DS26/AB/R. Article 31(3)(c) is also referenced in passing by Brazil in the Appellate Body’s 1997 Report in Brazil – Measures Affecting Desiccated Coconut (21 February 1997), Appellate Body Report, AB-1996-4 and by the EC in the Appellate Body’s 1998 Report in EC – Customs Classification of Certain Computer Equipment (5 June 1998) Appellate Body Report, AB-1998-2. 59 EC – Customs Classifications of Frozen Boneless Chicken Cuts (12 September 2005), Appellate Body Report, WTO/DS269/AB/R, WTO/DS286/AB/R, AB-2005-5; EC – Customs Classifications of Frozen Boneless Chicken Cuts (30 May 2005), Report of the Panel, WTO/ DS269/R, WTO/DS286/R. 60 EC – Customs Classifications of Frozen Boneless Chicken Cuts, Appellate Body Report, supra note 59, para. 194.
50
philippe sands and jeffery commission was a ‘supplementary means of interpretation’, within the meaning of Article 32 of the Vienna Convention.61
The Appellate Body concluded that there was broad consensus among the GATT Contracting Parties to use the Harmonized System as the basis for their WTO schedules, and such this agreement constituted ‘context’ under Article 31(2)(a) of the Vienna Convention for the purpose of interpreting the WTO agreements. In a footnote the Court stated that in view of this determination, we do not find it necessary to determine whether the Harmonized system could constitute a ‘relevant rule of international law’, within the meaning of Article 3l (3)(c) of the Vienna Convention.62
2. Mexico—Tax Measures on Soft Drinks and Other Beverages63 The dispute concerned certain tax measures imposed by Mexico on soft drinks and other beverages that use any sweetener other than cane sugar. The tax measures included: (a) a 20 per cent tax on the transfer or, as applicable, the importation of soft drinks and other beverages that use any sweetener other than cane sugar (‘soft drink tax’); (b) a 20 per cent tax on specific services (commission, mediation, agency, representation, brokerage, consignment and distribution), when provided for the purpose of transferring products such as soft drinks and other beverages that use any sweetener other than cane sugar (‘distribution tax’); and (c) a number of requirements imposed on taxpayers subject to the ‘soft drink tax’ and to the ‘distribution tax’ (‘bookkeeping requirements’). In proceedings before the Panel, Mexico argued that pursuant to Article 31(3)(c), in interpreting the WTO ‘covered agreements’, the Panel must take into account ‘any relevant rules of international law applicable in the relations between the parties’.64 Specifically, Mexico argued that “[c]learly, the NAFTA sets out such rules for the relations between Mexico and the United States which are relevant to the present dispute”, and that “the Panel must consider the NAFTA in this dispute”.65 The Panel found that: And, therefore, Mexico asserts that the Panel must consider the NAFTA in this dispute. Article 31(3) of the Vienna Convention, however, pertains to the interpretation of the terms of a treaty, and provides that ‘relevant rules of international
61
Id. Id., at fn. 384 63 Mexico – Tax Measures on Soft Drinks and Other Beverages (6 March 2006), Appellate Body Report, WT/DS308/AB/R, AB-2005-10; Mexico – Tax Measures on Soft Drinks and Other Beveragesl (7 October 2005), Report of the Panel, WT/DS308/R. 64 Mexico – Tax Measures on Soft Drinks and Other Beverages, Report of the Panel, supra note 63, para. 4.402. 65 Id. 62
treaty, custom and time: interpretation/application?
51
law applicable in the relations between the parties’ shall be taken into account along with the context of the treaty’s terms. Mexico has not identified any terms of the WTO Agreement for which it might be using the NAFTA or ‘general principles of international law’ as relevant context for interpretation of the meaning of the WTO Agreement’s terms. Mexico[‘s] reference to Article 31(3) does not change the fact that interpretation and application of the NAFTA are outside the Panel’s terms of reference.66
On appeal, the Appellate Body upheld the Panel’s Report, but did not offer any view on Mexico’s argument premised on Article 31(3)(c). B. European Court of Human Rights67 The European Court of Human Rights continues to refer to Article 31(3)(c) since the ICJ’s judgment in the Oil Platforms case.68 1. Case of Mamatkulov and Askarov v. Turkey69 In a judgment of the Grand Chamber rendered on 4 February 2005, the Court again relied upon Article 31(3)(c). The case concerned the applications of two Uzbek nationals against the Republic of Turkey first lodged with the Court under Article 34 in March 1999. The applications concerned the applicants’ extradition to the Republic of Uzbekistan. In a February 2003 Chamber judgment, the Chamber held unanimously that: (a) there had been no violation of Article 3; (b) that Article 6 was not applicable to the extradition proceedings in Turkey, and that no separate issue arose concerning the applicant’s complaint under Article 6 of the Convention; (c) that there had been a violation of Article 34 of the Convention; and (d) the finding of a violation constituted in itself sufficient just satisfaction for the non-pecuniary damage sustained by the applicants. In the Grand Chamber’s discussion of the alleged violation of Article 34 of the Convention, the Court reiterated that the Convention must be interpreted
66
Id., para. 4.4483. See Mark E. Villiger, ‘Articles 31 and 32 of the Vienna Convention on the Law of Treaties in the Case-Law of the European Court of Human Rights’, in Internationale Gemeinschaft und Menschenrechte: Festschrift für Georg Rezz zum 70 Geburtstag am 21 Januar 2005, Jürgen Bröhmer and Georg Rezz (eds.) (Köln: Heymann, 2005), 317. In addition to the Grand Chamber judgments in the Mamatkulov and Askarov v. Turkey, EHRR 41 (2005): 494, and Bosphorus Hava Yollari Turizm Ve Ticaret Anonim Sirketi v. Ireland, EHRR 42 (2006): 1, Article 31(3)(c) has also been referenced briefly in several ordinary chamber decisions and judgments, including: (i) Manoilescu and Dobrescu v. Romania and Russia (2005), Application no. 60861/00; (ii) Bianchi v. Switzerland (2006), Application no. 7548/04; and (iii) Nikolaus and Jurgen Treska v. Albania and Italy (2006), Application no. 26937/04. 68 Golder v. United Kingdom EHRR 1 (1975): 574 and Loizidou v. Turkey, EHRR 23 (1997): 513; ILC Study Group Final Report, supra note 10, at 219–21, paras. 435–8. 69 Mamatkulov and Askarov v. Turkey, EHRR 41 (2005): 494. 67
52
philippe sands and jeffery commission
in the light of the rules set out in Article 31(3)(c). In particular, in its discussion of powers to issue interim measures, the Grand Chamber stated that: The Court must determine the responsibility of the States in accordance with the principles of international law governing this sphere, while taking into account the special nature of the Convention as an instrument of human-rights protection (Golder v. the United Kingdom, judgment of 21 February 1975, Series A no. 18, § 29). Thus, the Convention must be interpreted so far as possible consistently with the other principles of international law of which it forms a part (Al-Adsani v. the United Kingdom [GC], no. 35763/97, § 60, ECHR 2001-XI).70
2. Case of Bosphorus Hava Yollari Turizm Ve Ticaret Anonim Sirketi v. Ireland 71 The case originated in an application against Ireland lodged with the European Commission of Human Rights under former Article 25 of the Convention by a Turkish company, Bosphorus Hava Yollari Turizm, on 25 March 1997, alleging that “the impounding of its leased aircraft [from the national airline of the former Yugoslavia] by the respondent State breached its rights under Article 1 of Protocol No.1”.72 In a decision rendered June 30, 2005, the Court discussed article 31(3)(c), and the Convention in the context of Ireland’s compliance with its legal obligations flowing from EC law, in particular Article 8 of EC Regulation 990/93: 150. The Court considers it evident from its finding at paragraphs 145–148 immediately above, that the general interest pursued by the impugned action was compliance with legal obligations flowing from the Irish State’s membership of the EC. It is, moreover, a legitimate interest of considerable weight. The Convention has to be interpreted in the light of any relevant rules and principles of international law applicable in relations between the Contracting Parties (Article 31 § 3(c) of the Vienna Convention on the Law of Treaties of 23 May 1969 and Al-Adsani v. the United Kingdom [GC], no. 35763/97, § 55, ECHR 2001-XI), which principles include that of pacta sunt servanda.73
70
Id., para. 111. Bosphorus Hava Yollari Turizm Ve Ticaret Anonim Sirketi v. Ireland, supra note 67. 72 Id., paras. 1–3. 73 Dispute Concerning Access to information under article 9 of the OSPAR Convention (Ireland v. United Kingdom), PCA, Final Award of 2 July 2003; the Mox Plant Case (Ireland v. United Kingdom), PCA, Request for Provisional Measures Order of 3 December 2001; Pope & Talbot Inc. and the Government of Canada, Award on the Merits of Phase 2 of 10 April 2001, (accessed on 1 May 2009); Pope & Talbot Inc. and the Government of Canada, Award in Respect of Damages of 31 May 2002; Bosphorus Hava Yollari Turizm Ve Ticaret Anonim Sirketi v. Ireland, supra note 67, para. 150. 71
treaty, custom and time: interpretation/application?
53
The Court ultimately found that the impoundment of the aircraft did not give rise to a violation of Article 1 of Protocol No. 1 to the Convention. C. Arbitral Tribunals Article 31(3)(c) has also been resorted to by a number of arbitral tribunals established pursuant to multilateral agreements, including in arbitrations in the Mox Plant Case, and in Pope and Talbot Inc. v. Canada.74 Since the ICJ’s judgment in Oil Platforms Article 31(3)(c) has been referred to on several occasions. 1. Case Concerning the Auditing of Accounts between the Netherlands and France in the Application of the 1991 Additional Protocol to the 1976 Convention on the Protection of the Rhine against Pollution by Chlorides75 In October 1999, the Netherlands requested that a dispute with France concerning the interpretation and application of the 1991 Additional Protocol to the 1976 Convention on the Protection of the Rhine against Pollution by Chlorides be submitted to arbitration, as provided for by Article 13 of the Convention. The 1991 Additional Protocol sought to improve the water quality of the Rhine and to regulate the reduction of chloride levels in the river, and provided that France shall take measures to reduce chloride levels in the Rhine, including the temporary storage of chlorides in its territory, to be funded jointly by the parties to the Protocol. The arbitral tribunal was constituted in February 2000, and the dispute concerned the financing of the operations undertaken by France under the Protocol. In particular, a determination of the amount that France should pay back to the Netherlands to complete the final auditing of accounts envisaged by paragraph 4.2.1 of Appendix III of the Protocol. In an award rendered March 12, 2004, an Arbitral Tribunal (Krzysztof Skubiszewski, Pieter Kooijmans and Gilbert Guillaume) acknowledged the need to consider any relevant rules of international law pursuant to Article 31(3)(c). In particular, the Netherlands argued that the ‘polluter pays principle’ constituted an applicable rule of international law within the ambit of Article 31(3)(c). While remarking that the principle was included in certain
74 ILC Study Group Final Report, supra note 10, at 235, para. 467. Article 31(3)(c) has also been referenced in passing in one Iran-United States Claims Tribunal award since the ICJ’s Oil Platforms decision, in Iran v. The United States of America, Case No. Bl (Counterclaim), Interlocutory Award of 9 September 2004, Award No. ITL 83-B1-FT. 75 Affaire Concernant L’Apurement des Comptes entre le Royaume des Pays-Bas et la Republique Francaise en Application du Protocole du Septembre 1991 Additionnel à la Convention Relative à la Protection du Rhin contre la Pollution par les Chlorures du 3 Decembre 1976, Sentence Arbitrale du 12 Mars 2004; ILC Study Group Final Report, supra note 10, at 211–2, para. 421.
54
philippe sands and jeffery commission
bilateral and multilateral treaties, the Tribunal found that it did not constitute a principle of general international law.76 2. The Iron Rhine Arbitration77 The dispute between Belgium and the Netherlands concerned the ‘Iron Rhine’ (‘IJzeren Rijn’, as it is known in Dutch), “a railway linking port of Antwerp, Belgium to the Rhine basin in Germany, via the Netherlands provinces of Noord-Brabant and Limburg”.78 The origins of the Iron Rhine are in the negotiations surrounding the separation of Beligum from the Netherlands in the 1830’s, in particular, the 1839 Treaty of Separation between Belgium and the Netherlands. The Arbitration Agreement posed a number of specific questions for the Arbitral Tribunal, including but not limited to: (1) to what extent is Dutch legislation and the decision-making power based thereon in respect of the use, restoration, and modernization of railway lines on Dutch territory applicable, in the same way, to the historical route of the Iron Rhine on Dutch territory; (2) to what extent does Belgium have the right to perform or commission work with a view to the use, restoration, adaptation and modernization of the historical route of the Iron Rhine on Dutch territory; and (3) to what extent any cost items and financial risks associated with the use, restoration, adaptation and modernization of the historical route of the Iron Rhine on Dutch territory be borne by Belgium or by the Netherlands.79 In an award dated May 24, 2005, an Arbitral Tribunal (Rosalyn Higgins, Guy Schrans, Bruno Simma, Alfred H.A. Soons, and Peter Tomka) relied upon Article 31(3)(c) in interpreting certain phrases contained in Article XII of the 1839 Treaty of Separation: 58. It is to be recalled that Article 31, paragraph 3, subparagraph (c) of the Vienna Convention on the Law of Treaties makes reference to ‘any relevant rules of international law applicable in the relations between the parties.’ For this reason – as well as for reasons relating to its own jurisdiction – the Tribunal has examined any provisions of European law that might be considered of possible relevance in this case (see Chapter III below). Provisions of general international 76
The Tribunal stated the following with respect to Article 31(3)(c): 102. En s’ acquittant de sa tache, le Tribunal doit aussi tenir compte ‘de toute regle pertinente de droit international applicable dans les relations entre les Parties’ (article 31, paragraphe 3, littera c) de la Convention de Vienne. Le Tribunal note que les Pays-Bas, a l’appui de leur demande, ont fait reference au principe de ‘pollueur payeur’. 103. Le Tribunal observe que ce principe figure dans certains instruments internationaux, tant bilateraux que multilateraux, et se situe a des niveeaux d’effectivite variables. Sans nier son importance en droit conventionnel, le Tribunal ne pense pas que ce principe fasse parties du droit international general.
77 In the Arbitration Regarding the Iron Rhine (‘IJzeren Rijn’) Railway Between The Kingdom of Belgium and the Kingdom of the Netherlands, PCA, Award of the Tribunal of 24 May 2005 (hereinafter Iron Rhine Arbitration). 78 Id., at 7, para. 16. 79 Id., at 3–4, para. 3.
treaty, custom and time: interpretation/application?
55
law are also applicable to the relations between the Parties, and thus should be taken into account in interpreting Article XII of the 1839 Treaty of Separation and Article IV of the Iron Rhine Treaty. Further, international environmental law has relevance to the relations between the parties.80
3. Saluka Investments BV v. the Czech Republic 81 On March 17, 2006, an UNCITRAL arbitral tribunal in Saluka Investments BV (The Netherlands) v. The Czech Republic rendered a partial award citing and relying upon VCLT article 31(3)(c) and the ICJ’s decision in Oil Platforms. The arbitration arose out of the Czech Government’s privatization of one of the major Czech banks, known as IPB, by selling the State’s shareholding to a company within the Nomura Group of companies, a major Japanese merchant banking and financial services group of companies. The IPB shares were later transferred to another Nomura subsidiary, Saluka Investments BV, constituted under the laws of the Netherlands. Saluka claimed, in particular, that the Czech Republic acted in relation to Saluka and its investment in a manner inconsistent with the Czech Republic’s obligations under the [1991] bilateral investment treaty between the Netherlands and the Czech Republic
claiming that is was “deprived of its investment contrary to Article 5 of that treaty, and that contrary to Article 3, its investment was not treated fairly and equitably”.82 The Arbitral Tribunal (Arthur Watts, Yves Fortier, and Peter Behrens), found that it had jurisdiction over the dispute and that while the Czech Republic had not acted in breach of Article 5 of the treaty, it had acted in breach of Article 3 of the Treaty, and reserved the questions as to appropriate redress for the second phase of the arbitration. In its discussion of Saluka’s claims under Article 5 of the treaty, the tribunal held that Article 5 imports into the Treaty the customary international law notion that a deprivation can be justified if it results from the exercise of regulatory actions aimed at the maintenance of public order.83
In support of this finding, the Tribunal referred to Article 31(3)(c) and the ICJ’s Oil Platforms decision, stating that: In interpreting a treaty, account has to be taken of ‘any relevant rules of international law applicable in the relations between the parties’ – a requirement which
80
Id., at 28, para 58. In the Matter of an Arbitration Under the UNCITRAL Rules 1976, Saluka Investments BV (The Netherlands) v. the Czech Republic, Partial Award of 17 March 2006 (hereinafter Saluka Investments). 82 Id., at 10, para. 26. 83 Id., at 52, para. 254. 81
56
philippe sands and jeffery commission the International Court of Justice (‘ICJ’) has held includes relevant rules of general customary international law.84
IV. The Work of the ILC In 2002, as part of its study on the fragmentation of international law, and the concomitant difficulties arising from the diversification and expansion of international law, the ILC embarked on a highly laudable and significant undertaking. Included amongst a number of study areas was a study on the interpretation of treaties in the light of ‘any relevant rules of international law applicable in the relations between the parties’ (article 31(3)(c) of the Vienna Convention on the Law of Treaties), in the context of general developments in international law and concerns of the international community.85
As the substantive outcome of its work, the ILC prepared both: [(1)] a relatively large analytical study on the question of fragmentation, composed on the basis of the individual outlines and studies submitted by individual members of the Study Group during 2003–2005 and discussed in the Group;
and (2) “a single collective document containing a set of conclusions emerging from the studies and the discussions in the Study Group”.86 Put simply, the idea behind the ILC’s work on Article 31(3)(c) has been to illustrate by examples, drawn from the practice of international courts and tribunals, techniques available for lawyers as they approach problems that appear to involve conflicts between rules or rule-systems.87
Article 31(3)(c) was not meant to be seen as “a panacea in reducing fragmentation”, but rather a call “upon lawyers to interpret treaties so as to ensure consistency with their normative environment”; “the integration into the process of legal reasoning – including reasoning by courts and tribunals – of a sense of meaningfulness”.88 On 17 July 2006, the ILC Study Group completed its 256 page analytical study and adopted a report containing 42 conclusions. Of the 42 conclusions, seven relate specifically to Article 31(3)(c), and concern “systemic integration”, “interpretation as integration in the system”, “application of systemic integration”, “application of custom and general principles of law”, “application of
84
Id. ILC Study Group Preliminary Report, supra note 10, at 3, para. 4. 86 ILC Report, supra note 10 at 401, para. 235. 87 ILC Study Group Final Report, supra note 10, at 245, para. 485. 88 ILC Study Group Preliminary Report, supra note 10, at 12, para. 28; ILC Study Group Final Report, supra note 10, at 211, para. 419. 85
treaty, custom and time: interpretation/application?
57
other treaty rules”, “inter-temporality”, and “open or evolving concepts”.89 As observed during the debates at the ILC, “the systemic nature of international has received clearest formal expression in that article [Article 31(3)(c)]”, an Article that “operates like a ‘master key’ to the house of international law”.90 The conclusions regarding interpretation and application in the context of Article 3 l(3)(c) and the objective of systemic integration are worth setting out in full: (17) Systemic integration. Article 31(3)(c) VCLT provides one means within the framework of the VCLT, through which relationships of interpretation […] may be applied. It requires the interpreter of a treaty to take into account ‘any relevant rules of international law applicable in relations between the parties’. The article gives expression to the objective of ‘systemic integration’ according to which, whatever their subject matter, treaties are a creation of the international legal system and their operation is predicated upon that fact. (18) Interpretation as integration in the system. Systemic integration governs all treaty interpretation, the other relevant aspects of which are set out in the other paragraphs of Article 31–32 VCLT. These paragraphs describe a process of legal reasoning, in which particular elements will have greater or less relevance depending upon the nature of the treaty provisions in the context of interpretation. In many cases, the issue of interpretation will be capable of resolution with the framework of the treaty itself. Article 31(3)(c) deals with the case where materials sources external to the treaty are relevant in its interpretation. These may include other treaties, customary rules or general principles of law. (19) Application of systemic integration. Where a treaty functions in the context of other agreements, the objective of systemic integration will apply as a presumption 89 ILC Report, supra note 10, at 413–6, paras. 17–23. The twentieth through twenty-third conclusions of the ILC Report regarding Article 31(3)(c) provide as follows:
20. Application of custom and general principles of law. Customary international law and general principles of law are of particular relevance to the interpretation of a treaty under article 31(3)(c) especially where: (a) the treaty rule is unclear or open-textured; (b) the terms used in the treaty have a recognized meaning in customary international law or under general principles; (c) the treaty is silent on the applicable law and it is necessary for the interpreter, applying the presumption in conclusion (19)(a) above, to look for rules developed in another part of international law to resolve the point. 21. Application of other treaty rules. Article 3 l(3)(c) also requires the interpreter to consider other treaty-based rules so as to arrive at a consistent meaning. Such other rules are of particular relevance where parties to the treaty under interpretation are also parties to the other treaty, where the treaty rule has passed into or expresses customary international law or where they provide evidence of the common understanding of the parties as to the object and purpose of the treaty under interpretation or as to the meaning of a particular term. 22. Inter-temporality. International law is a dynamic system. A treaty may convey whether in applying article 3 l(3)(c) the interpreter should refer only to rules of international law in force at the time of conclusion of the treaty or may also take into account subsequent changes in the law. Moreover, the meaning of a treaty provision may also be affected by subsequent developments, especially where there are subsequent developments in customary law and general principles of law. 23. Open or evolving concepts. Rules of international law subsequent to the treaty to be interpreted may be taken into account especially where the concepts used in the treaty are open or evolving. This is the case, in particular, where: (a) the concept is one which implies taking into account subsequent technical, economic or legal developments; (b) the concept sets up an obligation for further progressive development for the parties; or (c) the concept as a very general nature or is expressed in such general terms that it must take into account changing circumstances. 90
ILC Study Group Final Report, supra note 10, at 211, para. 420.
58
philippe sands and jeffery commission with both positive and negative aspects: (a) [t]he parties are taken to refer to customary international law and general principles of law for all questions which the treaty does not itself resolve in express terms; (b) [i]n entering treaty obligations, the parties do not intend to act inconsistently with generally recognized principles of international law. Of course, if any other result is indicated by ordinary methods of treaty interpretation that should be given effect, unless the relevant principles were part of jus cogens.91
V. Conclusions The ILC Study Group has acknowledged the difficulty in distinguishing between ‘interpretation’ and ‘application’. As the Study Group noted, although a tribunal may only have jurisdiction in regard to a particular instrument, it must always interpret and apply that instrument in its relationship to its normative environment – that is to say ‘other’ international law
but [t]his is not to say that it would in practice be easy – or even possible – to distinguish these aspects from each other.92 This is a conceptual point, as the Study Group indicated. It “refers to the way any right or obligation is doublesided – a creation of a treaty that is ‘applicable’ and in substance determined through ‘interpretation’ ”.93 That double-sided aspect is not about to disappear.
91
ILC Report, supra note 10, at 413–4, paras. 17–19. ILC Study Group Final Report, supra note 10, at 212–3, at para. 423, fn. 580 93 Id. The ILC considered the report of the Study Group at its fifty-eighth session, held in Geneva from 1 May to 9 June 2006, and from 3 July to 11 August 2006. At its 2902nd meeting, the Commission decided to take note of the conclusions of the Study Group and at its 2911th meeting, commended them to the attention of the General Assembly. 92
ARTICLE 31(3) (A) AND (B) OF THE VIENNA CONVENTION AND THE KASIKILI/SEDUDU ISLAND CASE* Hazel Fox I. Introduction On 13 December 1999 the International Court of Justice gave its judgment in the Kasikil/Sedudu Island (Botswana/Namibia) case.1 The decision settled satisfactorily, and Botswana and Namibia accepted and applied the judgment relating to a boundary dispute. Why then did four judges dissent? Why do two other judges add qualifying remarks and certain jurists criticise the judgment? In trying to answer I shall only direct my remarks to the Court’s treatment of ‘subsequent practice’ as a method of interpretation of Art III of 1890 Anglo German Treaty as to the meaning of ‘the centre of the main channel’ of the Chobe River. The dispute related to the location of the boundary where that island divides the Chobe, a tributary of the Zambezi River, between to the north the State of Namibia, formerly the German colony of SW Africa, and to the south the State of Botswana, formerly the British protectorate of Bechuanaland. The ICJ was requested in the Special Agreement to interpret the Anglo-German Treaty of 1890 which drew the boundary between the German and British possessions and provided by Article III that the boundary should follow ‘the centre of the main channel’ of the Chobe to its junction with the Zambezi. The words ‘the main channel of ’ were added shortly before the treaty was signed at the specific instance of Lord Salisbury, the Foreign Secretary and Prime Minister of Great Britain at the time. The International Court interpreted and applied Article III awarding the island to Botswana but declaring, – and this was accepted by Botswana, – that the tourist boats of both Botswana and Namibia should have free access to both channels round the island. The judgment of the International Court was accepted by both parties and there would seem to be no further cause for difference between the two States in that location. The accurate delimitation of boundaries and the allocation of
* This Chapter was based on the presentation made by the author during the ‘First Annual Conference on International Law: Interpretation under the VCLT’ organized by Queen Mary and Eversheds in 2006, and is current as of that date. 1 Case Concerning Kasikili/Sedudu Island (Botswana v. Namibia), Judgment of 13 December 1999, 1999 ICJ Rep. 1045 (hereinafter Kasikili/Sedudu Judgment). In the interest of transparency it is noted that the author acted as one of the counsel for Botswana in this case.
60
hazel fox
shared water resources remain a matter of constant concern to governments in Southern Africa and is one where the applicable principles of international law identified by the Court are likely to prove of great assistance.2 The third paragraph of the dispositif of the Judgment that in the two channels around the Island the nationals of and vessels flying the flags of Botswana and Namibia should enjoy equal national treatment provides a valuable application of the principle of equitable utilization of shared water resources.3 The ICJ decision is also likely to be a useful precedent to solve other potential river boundary disputes relating to the Zambezi and the Chobe.4 Nonetheless, the Court’s assessment of the relevance of the ‘subsequent practice of the States’ in solving the parties’ dispute has proved controversial, it challenges the general principle as stated by Judge Huber that “a juridical fact must be appreciated in the light of the law contemporary with it and not of the law in force at the time when a dispute arises or falls to be settled”.5 On this count it provoked debate among certain of the judges themselves and gave rise to criticism from jurists.6 It will be recalled that article 31 of the 1969 Vienna Convention on Treaties (VCLT), after stating the general rule of interpretation in paragraph l, and defining the context to which reference may be made in the second paragraph, deals with subsequent practice in paragraph 3, which provides that there shall be taken into account, together with the context a) any subsequent agreement between the parties regarding the interpretation of the treaty or its application of its provisions and b) any subsequent practice in the application of the treaty which establishes the agreement of the parties regarding its interpretation.
2 The Court’s identification of the thalweg or median line between the banks as a method of delimitation of a boundary in a water course (Kasikili/Sedudu Judgment supra note 1, para. 24) was applied in the Case concerning the Frontier Dispute (Benin/Niger), Judgment of 12 July 2005, 2005 ICJ Rep. 90, para. 144. 3 See Kasikili/Sedudu Judgment supra note 1, Separate Opinion of Judge Kooijmans. 4 Peter Ashton, ‘South African Water Conflicts: are they inevitable or preventable?’ in Water for Peace in the Middles East and South Africa, Green Cross International (ed.) (Geneva: Green Cross International, 2000), 94–8. He identifies two islands on the Zambezi and five islands on the Chobe where the boundary is uncertain and describes the Kasikili/Sedudu case as
an excellent example of a water based conflict that reached a high level of tension preventing resolution of the problem by disputing parties, thus requiring a third party (the ICJ) to arbitrate the dispute. 5 Island of Palmas (or Miangras) case (United States of America v. the Netherlands) (1928), RIAA 2: 829, at 839. 6 Charalambos Apostolidis, ‘L’Affaire d’ Ile de Kasikili/Sedudu (Botswana/Namibie): l’Arret de CIJ du 13 Decembre 1999’,45 Annuaire Francaise de Droit International 45 (1999) : 434; Paul Tavernier ‘Observations sur le Droit Intertemporel dans l’Affaire de l’Ile de Kasikili/Sedudu (Boswana/Namibie);Cour Internationale de Justice, Arret du 13 decembre 1999’, RGDIP 104 (2000) : 429; Malcolm Shaw, ‘Case concerning Kasikili/Sedudu Island (Botswana/Namibia)’, ICLQ 49 (2000) : 964; Donald Greig, Intertemporality and the Law of Treaties (London: BIICL, 2001), 114–8.
article 31(3) (a) and (b) of the vienna convention
61
The relevance of this provision to the boundary dispute between Botswana and Namibia raised three issues. 1) Had Article 31 (3) subparagraphs (a) and (b) any application to the Kasikili/ Sedudu case given that Botswana and Namibia were not parties to the Vienna Convention on Treaties? 2) If the rules of interpretation set out in Article 31 (3) are to be applied in the case as customary law, is the customary law applicable to a treaty concluded in 1890 to be assumed to be the same as the modern rules of interpretation as set out in Article 31? and 3) If such an assumption is made, is any agreement or practice of the parties subsequent to 1890 at any time up to the determination of the meaning of the treaty by the International Court admissible as evidence? or is such subsequent agreement or practice confined to a period immediately following the conclusion of the 1890 Treaty? Shaw identifies these last two questions as the temporal factor and the crtitical date.7 II. Interpretation of a Treaty by Reference to Subsequent Practice Distinguished from Other Methods In assessing its relevance to the determination of the boundary concerning Kasikili/Sedudu Island VCLT Article 31(3) is to be distinguished from other distinct operations which may be resorted to where a query as to a meaning of words in a treaty arises. The amendment or revision of the treaty is the most far-reaching of such operations since it may change the legal effect of the treaty provisions, it may result in the replacement of the original treaty by a new treaty satisfying all the procedural requirements of part IV of the Vienna Convention relating to the amendment of treaties. Closely allied to such amendment but possibly more in conformity with the parties’ original imperfectly expressed intent, will be an authoritative interpretation of the meaning of the disputed words; according to Schwarzenberger “the right of giving authoritative interpretation of a legal rule belongs solely to the person or body who has the power to modify or suppress it”. Such a body may be the international court or tribunal entrusted by the parties to make a binding decision on the matter but more usually will be the parties themselves. Such an authoritative interpretation may modify the terms of the treaty, but in a manner not strictly in accordance with any amendment procedures provided by the treaty or by the constitutional requirements of the States party to it. In Schwarzenberger’s words inferences made from subsequent behaviour of the contracting parties may only be taken into account for the purpose of 7
Shaw, supra note 6, at 968.
62
hazel fox
“strengthening its conclusion as the common intention of the parties or the meaning of the treaty at the time of it conclusion”.8 An example of such a modification which could not be reached by mere interpretation of the original treaty text is the modification of the phrase “concurring votes of permanent members” of the Security Council in article 27 of the UN Charter so as to permit the abstention of a permanent member with regard to a resolution of the Security Council without being deemed to have exercised the veto against it.9 Thirdly as envisaged in the third paragraph of VCLT Article 31, and as the method under consideration in the Kasikili/Sedudu case, there may take place the elucidation of the text of the treaty by reference to the subsequent agreement or practice of the parties establishing an agreement as to the interpretation of the treaty. This method accepts the ordinary meaning of the words of the treaty as predominant; the words are not challenged as ambiguous, obscure or leading to an absurd result, as in Article 32, but elucidation of that ordinary meaning is made by reference to a subsequent agreement of the parties. Such subsequent agreement was described by the International Law Commission in its commentary on the draft Vienna Convention as representing “an authentic interpretation by the parties which must be read into the treaty for the purposes of interpretation”; and in the same commentary the subsequent practice of the parties was stated to be of obvious importance “for it constitutes objective evidence of the understanding as to the meaning of the parties”.10 As Myres S McDougall states “[t]he effect of such an exercise is not to modify the treaty but ‘to confirm or supplement the parties’ original expectations”.11 A fourth somewhat unorthodox type of operation relating to the ascertainment of the meaning of words in a treaty appears to have been adopted by the International Court in its judgment in Kasikili/Sedudu, and is further discussed below. In that case the Court made use of “factual findings.. at different periods, not disputed at the time” which “whilst not constituting subsequent 8 Georg Schwarzenberger International Law, Vol. I: International Law as Applied by International Courts and Tribunals (London: Stevens, 1957), 531–2. 9 See Legal Consequences for States of the Continued Presence of South Africa in Namibia (South West Africa) notwithstanding Security Council Res. 276 (1970), Advisory Opinion of 21 June 1971, 1971 ICJ Rep. 16, at 22, recognising this practice; Karol Wolfke, ‘Treaties and Custom: Aspects of Interrelation’ in Essays in the Law of Treaties: a Collection of Essays in honour of Bert Vierdag, Jan Klabbers and Rene Lefeber (eds.), 31,\at 34. Also The Question of Jaworzina, Advisory Opinion (1923), PCIJ (Ser. B), No. 8, for a decision of the Conference of Ambassadors as to a division of territory which was in conformity with, but not specifically so identified by, a previous agreement between the Czech and Polish Foreign Ministers. 10 YBILC (1966): Vol II, at 221, paras. 14.and 15, cited in the Kasakili/Sedudu Judgment, supra note 1, para. 48. Judge Koiijmans in his separate opinion considered that where a claim of acquisitive prescription is raised in respect of the interpretation of a treaty it must be assessed as a question of subsequent practice; he did not consider it possible to “evaluate a claim to prescriptive title positively and at the same time to evaluate a claim concerning subsequent practice negatively”, para 19. 11 Myres S McDougall, Harold Lasswell, and James C. Miller, The Interpretation of International Agreements and World Public Order (New Haven: New Haven Press, 1994), 132–4.
article 31(3) (a) and (b) of the vienna convention
63
practice” support conclusions of the court as to the ordinary meaning of the treaty. III. The Scope of Interpretation Pursuant to VCLT Article 31(3)(a) and (b) As regards the third method of interpretation as provided in VCLT Article 31(3)(a) and (b), it is to be noted that the term ‘the interpretation of a treaty’ is used by reference to any subsequent agreement of the parties whereas ‘the application of a treaty’ is used in respect of subsequent practice of the parties. The two-fold exercise required by the use of these terms is well explained by Haraszti in the opening words of his book Some Fundamental Problems in the Law of Treaties 1971 where he states “A legal rule manifesting itself in whatever form cannot be applied unless its content has been elucidated”. In his view the two operations of interpretation and application, though closely related, can be segregated from each other inasmuch as interpretation has “the elucidation of the text as to its meaning as its objective” whilst application “implies the specifying the consequences devolving on the contracting parties” [and in certain exceptional cases also to third States in the given situation].12 Article 31(3) (a) speaks first of “any subsequent agreement between the parties regarding the interpretation of the treaty”. The word ‘agreement’ is less rigorous than a ratified treaty. It would seem that some arrangement less formally complete than another treaty achieved by amendment is contemplated, and hence article 31.3 subpara. a) extends to an informal agreement recorded in the minutes of a meeting or a press release, provided it constitutes “concordant practice” in Haraszti’s words, or the “genuine shared expectations of the parties” in McDougall’s phrase. To read it to mean an ‘understanding’, as Judge Weeramantry who dissented in the Kasikili case asserted, may go too far; he would treat the “silence and inaction of one party in the face of regular use of the Island by the Masubia tribe” as coming within the ‘subsequent agreement of the parties’. In his dissenting opinion Judge Weeramantry states that “the word ‘agreement’ in subpara 3(a) of the Treaty Convention can be read in the sense of ‘understanding’, and can therefore cover silence and inaction as well”. He relies on Sir Humphrey Waldock’s statement as Rapporteur to the ILC that “the assent of a party to a transaction may be inferred from its reaction or absence of reaction to the practice”.13 But this was written to support the choice
12 Gyorgy Haraszti, Some Fundamental Problems in the Law of Treaties (Budapest: Akadémiai Kiadó, 1973), 18. 13 Kasikili/Sedudu Judgment, supra note 1, Dissenting Opinion of Judge Weeramantry, paras. 23–4.
64
hazel fox
of the word ‘understanding’ and ignores the deletion of that word for ‘agreement’ in the final version of the text of the convention.14 An additional argument in support of some expression, however informal, of actual agreement between the parties, rather than the acquiescence of one treaty party to the course of action of the other, would be if the VCLT, by first listing ‘agreement’ and then ‘practice’ as admissible evidence in the interpretation of the treaty, could be said to envisage some order of hierarchy in Article 31.3.(a) and (b). Certainly the Court in its judgment by the use of the word ‘a fortiori’ in paragraph 63 appears to rank ‘subsequent agreement’ above ‘subsequent practice’ when it states that it concludes that the events demonstrate the absence of agreement between South Africa and Bechuanaland with regard to the location of the boundary around Kasikili/Sedudu Island and the status of the Island. Those events can, therefore constitute “subsequent practice in the application of the treaty [of 1890] which establishes the agreement of the parties regarding its interpretation” (1969 Vienna Convention on the Law of Treaties, Art. 31, para. 3 (b)). A fortiori, they cannot have given rise to an “agreement regarding the interpretation of the treaty or the application of its provisions” (Id., Art. 31, para. 3 (b)).15 Yet the International Law Commission which drafted the Vienna Convention was adamant that Art 31 contained no hierarchy of methods, though it did so on the basis that the weaker form of ‘understanding’ would be adopted for Art 31,(3) (a). It is possible to argue, therefore, that ‘agreement’ being the version which was adopted in the final text introduces a hierarchy, since it appears to insist on an element of common intent of the parties and hence distinguishes it from mere practice of the parties. The ILC’s use of ‘understanding’ on the other hand would refute any ranking since the term seems merely to stress the subjective inference to be drawn from the practice envisaged in (b) or acts of active or passive acquiescence covered by a).16 IV. The Application of Article 31 (3) Subparagraphs (a) and (b) to the Kasikili/Sedudu Case given that Botswana and Namibia were not Parties to the Vienna Convention on Treaties? The applicability of Article 31 of the Vienna Convention to the territorial dispute between Namibia and Botswana is not straight forward. How could an interpretative method set out in a convention adopted in 1969 have any relevance to a Treaty entered into eighty years previously? Why was it possible to apply methods of interpretation laid down by a non-retroactive treaty of 1969
14 UN Conference on the Law of Treaties, First Session (26 March – 24 May 1968), at 442, para.29. 15 Kasikili/Sedudu Judgment, supra note 1, para. 63. 16 Apostolidis, supra note 6, at 448.
article 31(3) (a) and (b) of the vienna convention
65
to a treaty of 1890? Neither Botswana or Namibia were parties to the VCLT but “both of them consider that Article 31 of the VCLT is applicable in as much as it reflects customary international law.”17 From this admission the Court considered that it had authority to apply to the determination of the issues the rules as to subsequent practice of the parties as set out in the Convention. The Court cited Article 4 of the Convention as providing that “it applies only to treaties … concluded by States after the entry into force of the… Convention with regard to such States”.18 The International Court solved the temporal difficulty by declaring that it had previously in Territorial Disputes (Libya/Chad)19 and Oil Platforms (Iran/USA) Preliminary Objections20 held that customary law found expression in Article 31 and accordingly declared that Article 4 of the Convention might be disregarded by reason of the parties’ acknowledgement of the customary force of Article 31. Judge Oda in his separate opinion considered that the judgment placed “excessive reliance upon the Vienna Convention on the Law of Treaties for the purpose of the Court’s interpretation of the 1890 Anglo-German Treaty”. In his view the rules of interpretation provided by the VCLT only applied to treaties concluded between States; since Botswana and Namibia were neither parties to the Anglo-German Treaty nor the VCLT it was not a case for the application of the VCLT’s interpretation rules.21 The succession of Namibia and Botswana to the treaty rights of Germany and Great Britain was complicated by the intervening status of SW Africa as a British, and later a South African, mandate territory prior to its independence as the State of Namibia; but inquiry into this intricate question was made unnecessary by the parties’ direction in Article 1 of the Special Agreement referring the dispute to the Court that it be determined by the Court on the basis of the Anglo German Treaty of 1890. The two States, Botswana and Namibia, thereby implicitly acknowledged that they were the successor States to Great Britain and Germany respectively. Judge Oda’s criticism, however, is more telling as to the application of the VCLT as a rule of customary international law. His objection has two aspects, the first being did the parties’ acceptance of the applicability of a customary 17
Kasikili/Sedudu Judgment, supra note 1, para. 18. Tavernier comments that the Court’s omission of the words in Article 4 “Without prejudice to any rules set forth in the present Convention to which treaties would be subject under international law independently of the Convention”, implies that the VCLT’s formulation in Article 31 of the customary rules of interpretation only applies to States which are parties to the Convention. “The better view is that Article 4 is only a ‘saving clause’ and that as long as an independent customary rule can be identified any non- retroactivity is of no effect’ ”; Tavernier, supra note 6, at 431. Mark Villiger, Customary International Law and Treaties (The Hague: Kluwer Law International, 1997), 252, para. 395. 19 Territorial Dispute (Libya v. Chad), Judgment of 3 February 1994, 1994 ICJ Rep. 4, at 21. 20 Oil Platforms (Iran v. United States), Preliminary Objections, Judgment of 12 December 1996, 1996 ICJ Rep. 812, para. 23. 21 Kasikili/Sedudu Judgment, supra note 1, Separate Opinion of Judge Oda, para. 4. See Shaw, supra note 6, at 968. 18
66
hazel fox
rule suffice to give it legal force? The International Court disposed of this question when in the Nicaragua Case it examined the effect of the prohibition against the use of force and permissibility of self defence as rules of customary international law, independent of the UN Charter provisions which had been expressly excluded by the parties. It stated that the concurrence of the views of the parties does not ... dispense the Court from having itself to ascertain what are the rules of customary international law are applicable. The mere fact that States declare their recognition of certain rules is not sufficient for the Court to consider these as being part of customary international law and as applicable as such to those States.
The Court in that case gave as its reason the direction in article 38 of its Statute to the Court to apply international custom “as evidence of a general practice accepted as law” and hence deduced that in ascertaining the existence of custom it “may not disregard the essential role of general practice in the formation of customary law.” It concluded that the shared view of the parties as to the content of what they regard as the rule is not enough. The Court must satisfy itself that the existence of the rule in the opinio juris of states is confirmed by practice.22
Whilst the International Court of Justice will, therefore, generally, in addition to the parties’ acceptance, require proof of state practice and opinion juris before applying as international custom a rule, it would not seem necessary to do so for the application of the rules relating to interpretation as set out in VCLT Article 31. No trawl through the general practice of states may be required in a particular case because the travaux preparatoires of Article 31 make it plain that its content as rules of customary law was well established at the time of their incorporation into the Vienna Convention. Such a conclusion would seem supported by the Beagle Channel arbitration where the tribunal stated that the invocation of Article 31 may be justified on the basis, not just of the parties’ wishes, but also of the fact that it enshrined “the traditional canons of treaty interpretation”.23 V. Is the Customary Law Applicable to a Treaty Concluded in 1890 to Be Assumed to Be the Same as the Modern Rules of Interpretation as Set Out in Article 31? Yet, it does not necessarily follow that an international custom recognised at the date of the adoption of the Vienna Convention in 1969 was in existence in 22 Military and Paramilitary Activities in and against Nicaragua (Nicaragua v. United States of America), Judgment of 27 June 1986, 1986 ICJ Rep. 14, at 97–8, para.184. 23 Beagle Channel Arbitration (Argentina v. Chile) (1977), ILR 52: 93, paras. 124 and 127.
article 31(3) (a) and (b) of the vienna convention
67
1890 at the conclusion of the Anglo-German Treaty. The second aspect of Judge Oda’s objection relates to whether the parties’ acceptance that the case be decided ‘on the basis of the 1890 Anglo-German Treaty’ restricted the customary law to such rules of interpretation of treaties as existed at the time of the conclusion of the treaty in 1890. If as the Court accepted customary law is to be applied, may it assume that such rules are identical to the modern customary rules of interpretation as set out in article 31.(3) (a) and (b)? This second question raises the issue of intertemporality. As Tavernier writes: La référence à la régie coutumière ne supprime toutes les difficultés en matière de non-rétroactivité et de droit intertemporel. En effet, il faudrait encore déterminer à partir de quelle date la régie coutumière existait, ce qui peut s’ avérer extrêmment difficile, voire impossible…s’il s’ agit d’un traité ancien comme le traité de 1890 entre l’Allemagne et le Royaume-Uni.24
Clearly any practice of the parties relied upon must take place after the treaty is in force but is the period of time open-ended and unlimited? or is the meaning of what was agreed restricted to the sense given it in the immediate practice of the parties implementing the treaty? Tavernier suggests one solution to this problem might be to treat the provisions in Article 31 more as a direction rather than rules and hence as unnecessary to apply strict rules of temporality to their application. He reminds one of a comment made by Reuter: There has been much discussion about the method of interpretation; the most difficult issue being the distinction between what can be the subject of specific rules of interpretation and what in a concrete case should be left to the interpreter’s art.25
Certainly the Court in the Kasikili/Sedudu case did not seem to recognise any difficulty relating to time in its application of customary rules of interpretation, baldly stating: In order to illuminate the meaning of words agreed upon in 1890, there is nothing that prevents the Court from taking into account the present-day scientific state of knowledge as reflected in the documentary material submitted to it by the Parties.26
and cited the Laguna del Desierto case in support.27 Judge Higgins made a declaration supporting the Court’s finding that the northern channel is the main channel within the meaning of Article III of the 1890 Treaty but rather than 24
Tavernier, supra note 6, at 433. Text as in Paul Reuter, Introduction to the Law of Treaties (London: Kegan Paul International, 1995, trans. by José Mico and Peter Haggernacher), 95, para. l40. 26 Kasikili/Sedudu Judgment, supra note 1, para. 20. 27 Dispute concerning the course of the frontier between B.P. 62 and Mount Fitzroy (Argentine v. Chile), Arbitral Award of 21 October 1994, ILR 113: 76, para. 157 (also known as the Laguna del Desierto case). 25
68
hazel fox
treating the Court’s finding as a determination of the ordinary meaning of the words, would seem to classify ‘the main channel’ as a generic term: such a generic term, as stated in respect of the term ‘territorial status’ in the Aegean Continental Shelf case gives rise to “a presumption that its meaning is intended to follow the evolution of law and to correspond with the meaning attached to the expression by the law in force at any given time”.28 Judge Higgins considered that the Court might use contemporary knowledge and scientific data to assist in the task of identification, through the legal technique of evaluating evidence, of the ‘main channel,’ “a somewhat general term decided upon by the Parties in 1890” and of its application to “a geographic and hydrological situation much better understood today”. But she warned “that we must never lose sight of the fact that we are seeking to give flesh to the intention of the Parties’ expression in generalized terms in 1890. We must trace a thread back to the point of departure…our task is to decide what general idea the Parties had in mind, and then make reality of that general idea through the use of contemporary knowledge”.29
A variation of this approach is also to be found in the dissenting opinion of Judge Weeramantry who maintained that we are not interpreting or applying a legal concept, in which case intertemporal principles might, in certain cases attract the meaning that the concept bears at the time of interpretation. Rather, we are here examining a question of fact as to which two channels was considered by the Parties at that time to be the main channel.
However, as we shall see, by stressing a ‘principle of contemporaneity’, the judge restricted the admissibility of factual evidence to a restricted period after the conclusion of the treaty.30 The Court itself by reference to a 1887 Resolution of the Institut de Droit International in which the word ‘channel’ was used as a synonym for the term thalweg, noted that
28 Aegean Continental Shelf case (Greece v. Turkey), Jurisdiction of the Court, Judgment of 19 December 1978, 1978 ICJ Rep. 3, at 32, para. 77. See a useful summary by Judge Guillaume in his declaration paras 16–9, in the Dispute regarding Navigational and Related Rights (Costa Rica v. Nicaragua), Judgment of 13 July 2009 of the contemporary (renvoi fixe) and evolutive (renvoi mobile) methods of treaty interpretation; the issue before the ICJ in that case being whether free navigation of the San Juan River “por los objectos de commercio” extended to commercial carriage of persons, and in particular tourists, as well as goods; available at: <www.icj-cij.org/ docket/files/133/15330.pdf> (accessed on 1 November 2009). 29 Kasikili/Sedudu Judgment, supra note 1, Declaration of Judge Higgins, para. 4. 30 Kasikili/Sedudu Judgment, supra note 1, Dissenting Opinion of Judge Weeramantry, para.29. The judges also interpreted the Anglo-German Treaty to be an agreement relating to spheres of interest, not a boundary treaty and hence an agreement dealing with spatial zones, while boundary treaties involve points or lines that have no breadth. “In consequence he concluded that there was greater flexibility in the definition of the boundary in question”; Kasikili/ Sedudu Judgment, supra note 1, para. 98.
article 31(3) (a) and (b) of the vienna convention
69
at the time of the conclusion of the 1890 Treaty, it may be that the terms ‘centre of the [main] channel’ and ‘Thalweg des Hauptlaufes’ were used interchangeably (para 25).
Apart from this reference to state practice in 1890, no attempt was made by the Court to establish what the customary rules as to subsequent practice of parties with regard to the interpretation of treaties were in 1890. Subsequent writers who have looked into this point have somewhat grudgingly concluded that the position in 1890 was probably not much different from that set out in the convention. Greig tentatively suggests, on the basis of the Chamizal arbitration,31 and by reference to construction of documents in municipal law that the rules of interpretation might have been stricter and that there could be no departure from the strict meaning without ambiguity; in consequence the supplementary means, now set out in VCLT Article 32 were necessary.32 He, himself considered that the Treaty itself provided a solution, – one which neither the parties nor the Court raised: in his view any ambiguity arising as to the application of the 1890 Anglo German Treaty by reason of the occasional use of the island for cultivation by the Masubia tribe gave rise to the obligation of Great Britain under article VI of the 1890 Treaty to rectify the frontier. VI. The Period after the Conclusion of the Treaty during which Subsequent Practice as to the Meaning of the Treaty is Admissible Assuming the Court is satisfied as to the existence of customary rules of interpretation, it is also necessary to determine whether there is any restriction on the period in which evidence as to subsequent practice may be admitted. All four dissenting judges, Weeramantry, Para-Rangurem, Fleischauer and Rezek, were clearly of the view that the period for admission of a subsequent agreement to interpret the 1890 treaty should be confined to the immediate decades after its conclusion. Judge Para-Aranguren in his dissenting opinion stated that the creation of the mandate by the League of Nations in 1920 rendered South Africa incapable of disposing by its own actions of the territory of South Africa, and consequently concluded that 1914 [when SW Africa was occupied by British forces during the first world war] is the latest date to be taken into account for the determination of the subsequent practice of the parties, Germany and Great Britain, in regard to the interpretation of Article III of the Anglo-German Agreement.
Applying this finding to the facts of the case, Judge Para-Arangurem concluded that 31 32
Text in AJIL 5 (1911): 782. Greig, supra note 6, 114–8.
70
hazel fox the Masubia of the Eastern Caprivi were the only tribesmen who occupied Kasikili/Sedudu Island, at least until 1914; that their occupation was peaceful and public (para 88)…. that the [Masubia] chiefs were agents of the [German] colonial administration and their acts represent the subsequent practice of the parties for the interpretation of the 1890 Anglo-German Agreement.33
Myres S MacDougall and his co-authors who considered this matter considered that preference should be given to evidence immediately following the treaty in ascertaining the true intention of the parties with regard to words in the treaty but that “a course of action [pursued] by those who must live with and under the provisions of the instrument” should also be admissible as evidence of subsequent agreement in accordance with VCLT Article 31(3)(b).34 VII. The Court’s Application of a Novel Method of Interpretation In the Kasikili/Sedudu case the Court was unable to find any subsequent agreement of the parties as to the precise location of the boundary in the vicinity of the Island nor was it able to conclude that the evidence of the subsequent practice of Namibia and Botswana or their predecessors in title provided support for any agreement as to how Article 3 of the Anglo-German Treaty 1890 should be applied where the Chobe river divided at the island. But rather than resorting to the supplementary means of interpretation in Article 32 of the Vienna Convention, a resort which would have required a ruling that the meaning of Article 3 of the 1890 Treaty was ambiguous or obscure or led to a result manifestly absurd or unreasonable, the Court appears to have found a fourth more unstructured method of interpretation. It supported its conclusion as to the ordinary meaning of the words in Article III by use of the parties’ own ‘factual findings’ as to that meaning, though these the Court held neither amounted to a subsequent agreement nor subsequent practice as to such an agreement. This new method of treaty interpretation was arrived at as follows: In paragraph 79 of its judgment the Court rejected the solution of the issue in the case by the methods of interpretation authorised by Article 31 (3) (a) and (b). The Court in paragraphs 21 to 41 had reviewed the linguistic, hydrological and other criteria and concluded that in accordance with the ordinary meaning of the terms the northern channel must be regarded as the main channel. It then, after an extensive review of the practice adduced by the parties subsequent to the treaty, in paragraph 79 concluded that that practice of the parties did not result in any agreement regarding the interpretation or application
33 Kasikili/Sedudu Judgment, supra note 1, Dissenting Opinion of Judge Para-Arangurem, paras.43 and 87. 34 McDougall, Lasswell. and Miller, supra note 11, at 144.
article 31(3) (a) and (b) of the vienna convention
71
of the 1890 treaty nor any practice of the parties in its application establishing the agreement of the parties regarding its interpretation. Yet it went on to say; However the Court is bound to note that on at least three occasions, at different periods – in 1912, in 1948 and in 1985-surveys carried out on the ground identified the channel of the Chobe to the north and west as the main channel of the river around Kasikili/Sedudu Island…The factual findings on these occasions were not, as such, disputed at the time. The court finds that these facts, while not constituting subsequent practice by the parties in the interpretation of the 1890 Treaty, nevertheless support the conclusions which it has reached by interpreting Article III, para. 2 of the 1890 Treaty in accordance with the ordinary meaning to be given its terms (see para. 41 above).35
As to the particular evidence to which it referred, the Court was of the view, and it is difficult to think it was wrong, that, despite three separate occasions when the question of the correct location of the boundary in relation to the disputed island was addressed by the relevant governmental authorities, they never reached an ‘agreement’ to bring this practice within the meaning of either subpara. a) or b) of VCLT Article 31(3). Capt. Eason’s report of 1912 after visiting the river that “the North should be claimed as the main channel” appears in the words of the Court’s judgment, “never to have been made known to Germany and to have remained at all times an internal document of the British authorities”.36 The Trollope/Redman/Dickinson proposals of 1947– 1951 to clarify the situation failed to be adopted or result in any agreement by the higher authorities either in Bechuanaland or South Africa.37 Finally as regards the finding of the Joint Survey of 1985 that ‘the main channel now passes Kasikili/Sedudu island to the west and the north of it’ the Court declared that “it cannot conclude…that in 1984–1985 South Africa and Botswana had agreed on anything more than the dispatch of a joint team of experts.”38 In paragraph 79 of its judgment, accordingly, the Court concluded that the subsequent practice of the parties did not result in any agreement regarding the interpretation of the 1890 treaty or the application of its provisions nor in any practice as to its application which established the agreement of the parties regarding its interpretation. Thus, it would seem the International Court has evolved a new method of treaty interpretation by employing ‘factual findings’ of parties on separate occasions over time as to the meaning of the contested words, factual findings which were not disputed, to support the court’s conclusions as to ordinary meaning of words in Article III of 1890 Treaty.
35 36 37 38
Kasikili/Sedudu Judgment, supra note 1, para. 80. Id., para. 55. Id., para. 63. Id., para. 68.
72
hazel fox
Not all the Judges supported the adoption of this new method. Four judges dissented being of the view that the Court’s enquiry should have been confined to the years immediately succeeding the conclusion of the treaty, or not later that 1914 which would have resulted in the identification of the southern channel as the main channel; Judge Fleischauer considered that insufficient attention had been given by the Court to the object and purpose of the treaty and its context. Judge Higgins in her declaration found there to be no ordinary meaning of the main channel; she was content to review the whole period up to the time of the judgment to ascertain the meaning of the generic term ‘the main stream’. For her the aerial photography and satellite imagery which showed “year round, taking one season with another the north to be the broader and more important channel” was the decisive element which led her to agree with the Court’s decision that the northern channel was the main channel.39 She further placed no reliance on the facts said to be found by Eason, Trollope and Redman, nor did she think it useful to accept as ‘facts’ findings of the Joint Team of Experts. In his declaration Judge Koroma appears to accept the Court’s findings as ‘the shared intentions’ of the parties. This new method of interpretation by the Court has attracted a wider criticism. Shaw regards the Court’s method as controversial, particularly as the techniques used by the persons concerned at those times were not clearly explained and may not have matched the methodological rigour of some of the scientific data submitted by the parties.
He makes a telling criticism: It is perhaps problematic for the Court to accept ‘facts’, the legal consequences of which were not accepted by the relevant parties, as supportive of legal conclusions reached by the Court.40
As he further points out, the Court seemed to uphold an opposite view as regards the map evidence: although unlike the surveys on the ground “the parties largely ignored the maps”41 as relevant to the location of the boundary, there were ‘factual findings’ that the early maps and the 1933 Bechuanaland map and the 1949 South Africa map showed the boundary in the southern channel, it appears that, on the basis of the maps available at the time, the boundary had until then [1947] been supposed to be located in the southern channel of the Chobe.42
39 Judge Rezek who dissented considered the greater ‘visibility’ of the northern channel by reference to aerial photography could have no relevance for either the colonial powers who drew the 1890 boundary, their local agents or the indigenous communities in the immediate decades after the treaty since it was not capable of perception at that time; Id., para. 4. 40 Shaw supra note 6, at 975. 41 Kasikili/Sedudu Judgment, supra note 1, para. 85. 42 Id., para. 62.
article 31(3) (a) and (b) of the vienna convention
73
Yet the Court rejected the evidence relating to the maps because it was unsupported by either agreement or practice of the parties in conformity with the requirements of VCLT Article 31(3)(a) or (b). Greig is more ready to accept the Court’s method but suggests it produces further problems. In commenting on the use made by the court of the ‘factual findings’ of the parties, he questions whether if the threshold of an agreement or an understanding is not reached, can the conduct simply be ignored? if there is no duty to respond, or if there is an adverse reaction to the conduct, it should not be the case that that the conduct (and any reaction to it) can simply be disregarded. It is the circumstances viewed as a whole which establish a gradation in the value of such evidence, rather than determine an artificial cut-off point where the practice ceases to be of relevance because it fails to give rise to an agreement or understanding as to the meaning of the treaty in issue.
Greig, therefore, favours the inclusion of all relevant evidence as an aid to the Court in its interpretation of a treaty, and would not narrow the Court’s enquiry so as to exclude evidence which relates to the parties’ understanding as to any ambiguity or obscurity in the terms of the treaty. However, in his view, such an interpretation of article 31 would seem to amount as far as subsequent conduct is concerned to the total disregard of Art 32. It is at this point in the Court’s treatment of Article 31 that Greig questions whether such a broad construction of the scope of article 31 can possibly be treated as a customary rule of treaty interpretation at the time or shortly after the 1890 Treaty came into force. He thus queries its applicability to the determination of the meaning of Article 3 of the 1890 Treaty. If the Court’s admission of factual findings is a method of treaty interpretation, however novel, authorised by the Vienna Convention even though it falls short of a subsequent agreement of the parties as to the meaning of the treaty, this would, in Greig’s view, establish that the effect of the VCLT is not to codify pre-existing law but to produce a shift in the law to render it inappropriate for application to treaties of earlier time.43 VIII. Conclusion As stated at the beginning, the International Court of Justice’s judgment in the Kasikili/Sedudu case successfully brought to an end a boundary in dispute between the parties. It achieved this by introducing a method by which reliance was placed on subsequent practice of the parties which did not, as required by the rules of interpretation set out in VCLT Article 31(3)(a) and (b), result in an agreement of the parties as to the meaning of the disputed
43
Greig, supra note 6, at 122.
74
hazel fox
words in the treaty. Further, in following this method, it made little attempt to enquire whether the customary rules of interpretation at the time of the conclusion of the treaty were the same or different from those set out in Article 31 nor, if such customary rules permitted reference to the subsequent practice of the parties, whether such practice was admissible solely within a restricted period after the conclusion of the treaty. Such a judgment, in which scant attention is given to the period of time in which the facts and the applicable law are to be determined and reliance placed on selective facts drawn from the whole period of the dispute to establish the meaning of words in a treaty, invites the conclusion suggested by Reuter that the task of the Court in interpreting a treaty is more an art than the application of strict rules.
‘THIRD PARTY’ CONSIDERATIONS AND ‘CORRECTIVE INTERPRETATION’ IN THE INTERPRETATIVE USE OF TRAVAUX PRÉPARATOIRES: – IS IT FAHRENHEIT 451 FOR PREPARATORY WORK? – Panos Merkouris I. Prolegomenon Fahrenheit 451 is the temperature at which paper burns. Ray Bradbury in the same-titled book envisaged a future where all books were burnt, as they were considered not as instruments of intellectual growth and enlightenment but as bearers of confusion, deceit and decay. This article examines whether Article 32 of the Vienna Convention on the Law of Treaties, which refers to supplementary means of interpretation, and more specifically the recourse to travaux préparatoires is facing a similar future. Are we witnessing a simmering fire that will eventually consume the travaux préparatoires as an element of treaty interpretation and reduce them to mere embers or are they still an incontestable element of the interpretative process? The importance or not of recourse to travaux préparatoires has been, since the first evolutionary steps of international law, the target of intense debate. It has been argued that travaux préparatoires focus the interpretative process on the intention of the negotiators, whereas the decisive element is the intention of the ratifiers;1 that it is misleading,2 too ample,3 a tabula in nauflagrio for all
1 Alexander P. Fachiri, ‘Interpretation of Treaties’, AJIL 23 (1929): 745 et seq.; However, Sir Hersch Lauterpacht does away with this argument by demonstrating that the international tribunals have invariably focused on the intention of the negotiators who are the main actors in the creation of the treaty and not the ratifiers, see Sir Hersch Lauterpacht, ‘Les Travaux Préparatoires et l’Intérpretation des Traités”, RCADI 48-II (1934): 709, at 795–6. 2 de Luna had serious doubts about the importance of travaux préparatoires, in the sense that the most important provisions are usually discussed and negotiated in informal discussions (one form of which is corridor-diplomacy). Consequently, the true reasons behind the adoption of a certain formula will not be found in the records of the meetings and are, thus, not “a full guide to the will of the parties”; Comments of Antonio de Luna (A/CN.4/SR.766), reproduced in YILC (1964): Vol. I, at 285, para. 36; Bartos on the other hand turned his criticism on the fact that the travaux préparatoires were of dubious value since States usually made conflicting declarations or a term was construed in several different ways, making it, usually, impossible to figure out which construction had prevailed; Comments of Milan Bartos (A/CN.4/SR.766), reproduced in YILC (1964): Vol. I, at 287, para. 57; see also comments of Sir Ian Sinclair, in Vienna Conference on the Law of Treaties, Meetings of the Committee as a Whole, 33rd Meeting, para. 8. 3 In the sense that preparatory work especially of multilateral treaties can be extremely voluminous and the task of sifting through the relevant material is, if not insurmountable, at least
76
panos merkouris
the provisions that a State might not like;4 that recourse to travaux préparatoires is sometimes impossible, due to the fact that records are not always published or made available to the public5 or that, in any event, a great number of treaties were concluded when a great number of States had not yet come into existence and thus such a historical interpretation would not be just or equitable for them;6 that the recourse to travaux préparatoires can lead to different interpretations between different tribunals,7 that referring to the travaux préparatoires, essentially, has an ossifying effect on the development of the relevant notions, incorporated in the treaty or provision in question;8 or finally, that if the importance of travaux préparatoires is over-stressed, then “it would be as difficult to agree on what the records would contain as it would be to agree on the text of the treaty itself ”.9 These are but a few of the main arguments against the importance of the travaux préparatoires in the interpretative process of international courts and tribunals. Accordingly, each of these arguments against the importance of preparatory work can be paired off with a counter-argument in favour of travaux préparatoires. An in-depth presentation of all such arguments, as well as all the accompanying problematic areas of the travaux préparatoires would exceed the scope and purpose of this article. Instead, ‘third party’ considerations and highly odious; Lauterpacht, supra note 1, at 798 et seq.; However, Lauterpacht, himself, recognises the advantages, as to legal certainty, that such a completed survey can offer. 4 A tabula in nauflagrio means a wooden plank in the midst of a shipwreck and indicates that it would give the States an arsenal effective enough to circumvent any undesirable provision; Comments of Sir Erick Beckett, ‘De l’ Interpretation des Traités’, Annuaire de l’ Institut de Droit International 43 (1950): Tome I, at 438–40, also cited in the comments of Jimenez de Arechaga, in Vienna Conference on the Law of Treaties, Meetings of the Committee as a Whole, 31st Meeting, para. 64; similarly see comments of Jose Maria Ruda, in Vienna Conference on the Law of Treaties, Meetings of the Committee as a Whole, 33rd Meeting, para. 25. 5 Jan Klabbers, ‘International Legal Histories: The Declining Importance of Travaux Préparatoires in Treaty Interpretation?’, NILR 50 (2003): 267, at 280. 6 Id.;, this argument, in essence, reflects the problem of whether a Court or Tribunal can have recourse to travaux préparatoires of a treaty in which one of the Parties of the dispute in question had not participated. The most famous case on this issue is the River Oder case and the subtleties of this problem will be analysed infra at Part II. Travaux préparatoires with respect to Third Parties. 7 National or international, see Quincy Wright, ‘The Interpretation of Multilateral Treaties’ AJIL 23 (1929): 94, at 101; this is perhaps on of the first expressions of the fear of fragmentation. This argument, however, is more of a criticism with respect to the independence and impartiality of the judges rather than an inherent defect of the travaux préparatoires themselves, see Lauterpacht, supra note 1, at 796. 8 Comments of Antonio de Luna (A/CN.4/SR.871), reproduced in YILC (1966): Vol. I, Part II, at 194, para. 16, citing the Diversion of Water from the Meuse Case; Case of Diversion of Water from the Meuse, 1937 PCIJ (Ser. A/B), No.70, at 21; see also Klabbers, supra note 5, at 281. By referring to the travaux préparatoires the interpreter restricts himself to the arguments and scenarios envisaged by the drafters, which may long have been overrun by the current developments. However, there is another side to this argument; only through the study of the travaux préparatoires can the interpreter benefit from the wide range of possibilities and problems envisaged and discussed in the negotiating circles of the treaty; Lauterpacht, supra note 1, at 796–7. 9 Klabbers, supra note 5, at 281.
interpretative use of travaux préparatoires
77
the notion of ‘corrective interpretation’ were selected from the wide array of debatable issues with respect to preparatory work. This selection was not arbitrary but dictated by methodological considerations. Both these elements constitute extremities of the notion of travaux préparatoires. The question of whether preparatory work to which certain parties to a dispute have not participated, i.e to which they are for all purposes ‘third parties’, touches upon the outer limits of the travaux préparatoires from the standpoint of the subject of a treaty. On the other hand, the debate over the existence or not of a ‘corrective’ function of the preparatory work alongside the more classical ‘confirmatory’ and ‘determinative’ functions is, once again, an outer limit but this time from the viewpoint of the functional span of preparatory work. An analysis based on these two axes, can shed some light on the true limits of the recourse to travaux préparatoires and by implication answer the question, what the future holds for this interpretative tool. II. Travaux Préparatoires with Respect to Third Parties One of the most notorious issues with respect to the travaux préparatoires relates to the principle pacta tertiis nec nocent nec prosunt. According to this principle things that have been agreed between other parties do not bind a State, without its consent.10 Based on the consent-based structure of the international community this principle would seem to be self-evident. The problems, however, begin when one tries to transpose, mutatis mutandis, this principle to cover a wide diversity of instruments and documents. In the present case, the pacta tertiis principle would prima facie raise questions as to the admissibility of recourse to travaux préparatoires in cases where one of the parties to the dispute had not taken part in the negotiations of the treaty in dispute. The main argument on which this approach is based is that a great number of the existing treaties were concluded when a great number of States had not even come into existence. To insist on a significant role for historical interpretation is to deny a voice to roughly three quarters of today’s states, simply for not existing independently at the time of the drafting of a great number of treaties.11
A similar argument, but from the perspective of negotiator vs. acceding States was put forward within the ambit of the ILC.12
10 this principle has also found its place among the articles of the VCLT, see Section IV, Articles 34–8 of the VCLT. 11 Klabbers, supra note 5, at 280. 12 de Luna, supra note 8, at 194.
78
panos merkouris [I]n the case of general multilateral conventions, many of the parties will or may have joined by subsequent accession, not having taken any part in the original framing of the text, and must be assumed to have joined not so much on the basis of what the actual framers ‘intended’ as on the basis of what the text itself says and apparently means.13
Finally, along approximately the same lines, one could argue that using travaux préparatoires in such cases flies against the face not only of the principle of pacta tertiis nec nocent nec prosunt but of equity and good faith as well. The most notorious case, which has lent support to this line of argumentation, has been the Commission of the River Oder case. In that case the PCIJ considered as inadmissible and did not take into consideration the minutes of the Commission of the Peace Conference in Paris.14 The reasoning behind this was that three of the parties to the case brought before it had not participated in the travaux préparatoires of the Treaty of Versailles, hence, according to the Court, “the record of this work can not be used to determine, in so far as they are concerned, the import of the Treaty”.15 Similar considerations were reiterated in the Young Loan Abritration16 and acknowledged by several academics.17 At this point two facts need to be pointed out in order to accord to these two cases their proper meaning and weight. Firstly, it is true that the in the River Oder case the PCIJ barred the recourse to travaux préparatoires because States-parties to the dispute had not participated in the negotiations. However, this approach is in reality a more restrictive approach than the more expansive interpretation that had been suggested by the UK in the Advisory Opinion on Nationality Decrees in Tunis and Morocco. In that case the UK had suggested that the travaux préparatoires should not be admissible vis-à-vis all the acceding States because otherwise these States might find themselves bound by an interpretation based on documents and negotiations of which they had no knowledge at the time of their accession.18 The Court in that case did not deal with that issue; however, the PCIJ with the River Oder case judgment seems to
13 Sir Gerald Fitzmaurice, ‘The Law and Procedure of the International Court of Justice, 1951–4: Treaty Interpretation and other Treaty Points’, BYIL 33 (1957): 203, at 265. 14 this Commission prepared the articles of the Treaty of Versailles which established the Commission of the River Oder. 15 Territorial Jurisdiction of the International Commission of the River Oder, 1929 PCIJ (Ser. A), No.23, at 42. 16 Young Loan Arbitration, (UK, US, France, Belgium and Switzerland v. Federal republic of Germany), Arbitral Award of 16 May 1980, ILR 59 (1980): 494, at 544–7. 17 Sir Ian Sinclair, The Vienna Convention and the Law of Treaties, (Manchester: MUP, 1984), 1424; Arnold D. McNair, The Law of Treaties, (Oxford: Clarendon Press, 1961), 420–1; the issue was also raised by the Yugoslav Government in their Comments to the draft Articles of the VCLT. 18 Advisory Opinion Concerning Nationality Decrees issued in Tunis and Morocco, Sir Ernest Pollock, Pleadings, 1923 PCIJ (Ser. C), No.2, at 197.
interpretative use of travaux préparatoires
79
have restricted this bar on recourse to travaux préparatoires only to cases where the acceding States are Parties to a dispute and not as a general rule. Secondly, in the Young Loan Arbitration the situation is somewhat different than in the River Oder case. In the latter three States had not participated in the negotiations; in the former case, however, Germany had participated albeit in the final stage of the negotiations. The Tribunal held, however, that because of this late entry into the process of negotiations and, mostly, due to the atypical structure and work method of the Conference it could not take into consideration the procès-verbaux of the negotiations.19 It seems that the Tribunal made a distinction between, on the one hand, States that enter the negotiations at the last minute but make no use of their chance to study the travaux préparatoires, in which case the travaux préparatoires are opposable against them, and, on the other hand, States that had not participated in the totality of the negotiations and also had no knowledge of certain documents,20 in which case the relevant travaux préparatoires cannot be invoked against them.21 It seems that the Young Loan Arbitration, in essence, set down a minimum threshold for barring travaux préparatoires, in the form of the actual possibility of accessing the relevant documentation. As shown above the jurisprudence invoked to support non recourse to travaux préparatoires with respect to third States is not so clear – cut as would at first glance seem. From a theoretical and systemic point of view the inevitable counter-argument is that such an approach would lead to an insurmountable fragmentation/molecularization of the treaties. It is contrary to the principle of legal certainty to have two distinct sets of rules; one applying to original negotiating states and a different applicable to acceding States.22 Furthermore, it has been seriously contested whether this jurisprudence reflects “actual practice”.23 This point calls for further elaboration. According to Shabtai Rosenne, States which acceded to multilateral conventions had no hesitation in using the travaux préparatoires,24 a finding which is given further support in the more recent jurisprudence.25 Not only the acceding parties but the ICJ itself 19
Young Loan Arbitration, supra note 16, at 1200–1. an ignorance which is not attributable to an omission on their part. 21 Eric Canal-Forgues, ‘Remarques sur le Recours aux Travaux Préparatoires dans le Contentieux International’, RGDIP 97 (1993) : 901, at 906. 22 Sinclair, supra note 17, at 144. 23 ILC, The International Law Commission Commentary on the draft Vienna Convention, YILC (1966): Vo. II, at 223; Sir Humphrey Waldock, Third Report on the Law of Treaties, (A/ CN.4/167 & Add.1–3), reproduced in YILC (1964): Vol. II, 5, at 58–9; Sir Robert Jennings & Arthur Watts (eds.), Oppenheim’s International Law (London: Longman, 1992): Vol. I, at 1278; Sinclair, supra note 17, at 144. 24 Comments by Shabtai Rosenne, (A/CN.4/SR.872), reproduced in YILC (1966): Vol. I, Part II, at 200, para. 34. 25 US – Imposition of Countervailing Duties on Imports of Fresh and Chilled Atlantic Salmon from Norway (28 April 1994), Panel Report, SCM/153, para.51; Import, Distribution and Sale of 20
80
panos merkouris
had in its own jurisprudence relied heavily on the travaux préparatoires relevant to Articles 36(5) and 37 of its Statute, as revised at the San Francisco Conference, despite the fact that one or both of the parties to the dispute had not been present at the Conference.26 Arbitral Tribunals have also resorted to an examination of the travaux préparatoires in similar circumstances.27 Alongside these cases another group of arbitral decisions needs to be mentioned. In these decisions the relevant tribunals used documents to which parties to the dispute had not participated, but the modus of this recourse is quite unique. In the Mines Aroa Arbitration, for instance, the arbitrator in order to interpret certain provisions of the 1903 Protocol that Venezuela had signed and which established a British-Venezuelan Reparations Commission, examined not only the travaux préparatoires of that Protocol but also the travaux préparatoires of an in pari materia agreement between Venezuela and Germany.28 It is true that in all honesty, this latter cannot be considered as travaux préparatoires, nevertheless such interpretative practice has been often resorted to and can offer valuable insight as to the documents available in the interpretative process. A similar technique was used in the Kummerow,29 Betancourt30 and Sambaggio31 Arbitrations. From the previous analysis it is evident that from the early stages of the discussion on the travaux préparatoires it had been realized that the solution adopted in the PCIJ case was extremely problematic. To not allow for recourse to travaux préparatoires when one of the Parties to a dispute was an acceding State was tantamount to throwing the baby out with the bathwater. On the other hand to ignore the need for a qualification of admissible travaux préparatoires with respect to acceding States would be an inequitable response contrary to the principle of good faith. As Hersch Lauterpacht very eloquently put it, a rigid rule “is not necessarily a wise rule”32. What need to be balanced are Alcoholic Drinks by Canadian Provincial Marketing Agencies (22 March 1988), Panel Report, L/6304–35S/37, para. 3.58. 26 see for instance the Case of the Aerial Incident of 27 July 1955; In that case the ICJ held that the meaning of Article 36§5 of its Statute could be determined by recourse to the travaux préparatoires of the San Francisco Conference regardless of whether the States-parties to a dispute were originals signatories or not; Aerial Incident of 27 July 1955, (Israel v. Bulgaria), Judgment of 26 May 1959, 1959 ICJ Rep. 127, at 140 et seq.; see also the Joint Dissenting Opinion by Judges Sir Hersch Lauterpacht, Wellington Koo and Sir Percy Spender, at 179–180. 27 Italy v. Federal Republic of Germany Arbitration, ILR 29 (1959): 448, at 460 et seq.; Greece (on behalf of Apostolidis) v. Federal Republic of Germany Arbitration, ILR 34 (1960): 242–5; Aerial Incident of 27 July 1955, supra note 26, at 127. 28 Mines Aroa Arbitration, (UK v. Venezuela) (1903), reprinted in John Bassett Moore, History and Digest of the International Arbitrations to which the United States has been a Party, together with Appendices Containing the Treaties Relating to such Arbitations, and Historical and Legal Notes, (hereinafter Moore Arbitrations): Vol. V., at 498. 29 Kummerow Arbitration (1892), Moore Arbitrations : Vol. V., at 526. 30 Betancourt Arbitration, Moore Arbitration : Vol. V, at 939. 31 Sambaggio Arbitration, Moore Arbitration : Vol. V, at 666. 32 Lauterpacht, supra note 1, p. 807.
interpretative use of travaux préparatoires
81
considerations of the knowledge that the acceding State may have of the travaux préparatoires,33 the number of other signatories, and the accessibility of the relevant documents. The status of a State as acceding or not should not be used as a pretext for not allowing recourse to the travaux préparatoires in order to identify the intention of the drafters.34 Almost identical opinions were expressed within the ILC as well, where it was stated that it was more prudent to avoid adopting a general rule dealing with this issue. It should be decided on a case by case analysis.35 For this reason certain criteria were proposed; essentially that the travaux préparatoires have been published and that they are generally accessible.36 It is important to notice that the original proposal by Lord McNair envisaged that these criteria were to apply cumulatively so as to provide greater safeguards that the third Parties were actually aware of and duly informed on the relevant documents. There is quite an extensive jurisprudence37 supporting the application of these criteria. For the scope of this research what needs to be pointed
33 and in most cases a State will have made an extensive research on the negotiating documents. 34 Lauterpacht, supra note 1, at 807. 35 Comments by Shabtai Rosenne, (A/CN.4/SR.872), reproduced in YILC (1966): Vol. I, Part II, at 201, para. 35; However, Rosenne made the clarification that this was only in the case of published and available travaux préparatoires. It would be seriously contestable, on the other hand, whether confidential material could be invoked against third parties. This supposition seems to derive from the principle of pacta tertiis nec nocent nec prosunt, as enshrined in the relevant Articles of the VCLT. 36 Lord Arnold McNair, ‘Comments on the Report of Lauterpacht’, Annuaire de l’ Institut de Droit International 43 (1950): Tome I, at 450–2; these criteria have also been adopted by Professors Stone, Scharzenberger and Sir Gerald Fitzmaurice, see Brijesh N. Mehrisch, “Travaux préparatoires as an Element in the Interpretation of Treaties”, Indian Journal of International Law 11 (1971): 39, at 43; in the case of the WTO the criterion of accessibility seems to apply even after accession but prior to the dispute coming before a Court; the reason for this is that in such an event the State would have the chance to either leave the WTO or at least make a declaration clarifying its position; Michael Lennard, ‘Navigating by the Stars: Interpreting the WTO Agreements’, Journal of International Economic Law 5 (2002): 17, at 49. More recently both the WTO Panel and the Appellate Body in EC – Chicken Classification reaffirmed the aforementioned criteria:
… [when] parties have deemed notice of a particular event, act or instrument through publication, they may be considered to have had constructive knowledge and that such knowledge suffices for the purposes of Article 32 of the Vienna Convention (emphasis added)
European Communities – Customs Classification of Frozen Boneless Chicken Cuts (30 May 2005), Panel Report, WT/DS269/R & WT/DS286/R, para. 7.346. [297] As far as an act or instrument originating from an individual party may be considered to be a circumstance under Article 32 for ascertaining the parties’ intentions, we consider that the fact that this act or instrument was officially published, and has been publicly available so that any interested party could have acquired knowledge of it, appears to be enough. Of course, proof of actual knowledge will increase the degree of relevance of a circumstance for interpretation. (emphasis added)
European Communities – Customs Classification of Frozen Boneless Chicken Cuts (27 September 2005), Appellate Body Report, WT/DS269/AB/R & WT/DS286/AB/R, para. 297. 37 especially within the ambit of the WTO Dispute Panels and Appellate Body, see infra following paragraphs and footnotes for the relevant case-law.
82
panos merkouris
out is that these criteria are not devoid of problems. They pose themselves a series of issues, for instance how accessible need these documents be. According to the US – Combed Cotton Safeguards case, the test is one of due diligence. As long as the Party in question “has exercised due diligence in complying with its obligations of investigation, evaluation and explanation” then it cannot be held responsible or held bound for negotiations that it could not have known.38,39 Another interesting point is that of the complementary or hierarchical interrelationship between the requirement of publication and that of accessibility. Lord McNair’s proposal opted for a cumulative approach.40 The courts, however, seem to rely more on the formal criterion of publication rather than that of genuine accessibility.41 This seems to be a residual effect of the more classical state-centered and bilateral-oriented structure of the international community of the former decades. However, with the shift towards multilateralism and the rise of the individual as an active player on the international scene, accessibility will most likely become more and more the rule rather than the exception.42 The examination of the current state of practice with respect to the use of travaux préparatoires in connection to third parties has shown that this element of treaty interpretation is far from feeble and static. The axiom of pacta tertiis nec nocent nec prosunt seems to allow for certain exceptions when it comes to treaty interpretation and travaux préparatoires. Of course such leniency is not without its conditions. Safeguards exist in order to ensure legal certainty;43 these can be either of a formal nature, i.e. publication of the relevant documents or of a more practical nature i.e. the genuine possibility of accessing the documents in question. Irrespective of these criteria, the
38
US – Combed Cotton Yarn from Pakistan (5 November 2001), Appellate Body Report, WT/ DS192/R, para. 79. 39 a similar approach seems to have been taken in the Young Loan Arbitration, albeit not expressly stated, see supra analysis on Young Loan Arbitration. 40 See supra note 36. 41 in the US – Lamb Safeguards Case the Panel focused solely on the fact that the documents referred to were numbered but did not address the issue of accessibility of these documents; US – Safeguard Measures on Import of Fresh, Chilled or Frozen Lamb Meat from New Zealand and Australia (US – Lamb) (16 May 2001), Panel Report, WT/DS177/R & WT/DS178/R, para. 7.110 et seq.; in the US – Shrimp Case the Appelate Body relied on the travaux préparatoires of the International Trade Organization and the Havana Charter; US – Import Prohibition of Certain Shrimp and Shrimp Products (US – Shrimp) (6 November 1998), Appellate Body Report, WT/DS58/AB/R, paras.152–7; The problem with this is that although these documents are numbered documents, nevertheless their accessibility is severely restricted; only certain governments, researchers in Geneva and a few libraries would hold copies of these documents and in most cases in microfiche; see Lennard, supra note 36, at 51. 42 one example of this is that the travaux préparatoires of the Founding Treaties of the EC, have only recently been made accessible to the public in the framework of the effort towards Good Governance, on ‘Good Governance’ see Commission of the European Communities, European Governance – A White Paper (25 July 2001), (COM(2001)428 final). 43 to the degree, of course, that this is possible when we are dealing with interpretation.
interpretative use of travaux préparatoires
83
conclusion remains the same. The theory and practice of international law allows for a great flexibility in the use of travaux préparatoires even when it comes to third parties. This has as a direct result the insertion and application of a large corpus of documents to a larger number of subjects during the judicial interpretative process. III. Travaux Préparatoires as a Means of ‘Corrective’ Interpretation The examination of the outer limit from the standpoint of subjects, led us to the conclusion that the recourse to travaux préparatoires, at least to that respect, is alive and well. We will now take an introverted look into what is exactly the function of the recourse to preparatory work. Article 32 of the VCLT states that [r]ecourse may be had to supplementary means of interpretation, including the preparatory work of the treaty … in order to confirm the meaning resulting from the application of article 31, or to determine the meaning when the interpretation according to article 31: a) leaves the meaning ambiguous or obscure; or b) leads to a result which is manifestly absurd or unreasonable.44
A simple reading of Article 32 would seem at first glance to allow recourse to travaux préparatoires for reasons of confirmation or for determination but the latter only in case of ambiguity/obscurity or if the interpretation according to article 31 led to a manifestly absurd or unreasonable result. No mention whatsoever is made of any other possibility. During the discussion on the Articles on the Law of Treaties at the ILC, no reference was made to the possibility of the travaux préparatoires having a further function.45 It was only at the very end of the negotiations during the Vienna Conference on the Law of Treaties that the Portuguese delegate, Mr. Crucho de Almeida, made a passing reference to the scenario of conflicting interpretations between the travaux préparatoires and the ‘clear’ meaning of the text.46 He envisaged the possibility of the travaux préparatoires leading
44 Article 31 of the 1969 Vienna Convention on the Law of Treaties (1969), UNTS 1155 (1980): 331; ILM 8 (1969): 679 (emphasis added). 45 the main debate revolved around whether travaux préparatoires was an element of treaty interpretation that should be put in parity with the other elements incorporated in Article 31. 46 Mr. Crucho de Almeida said:
in its Advisory Opinion on the interpretation of the Convention of 1919 concerning employment of women during the night, the Permanent Court of International Justice had found that the preparatory work confirmed the conclusion reached on a study of the text of the Convention. In that particular instance, the judges had been fortunate because the two elements of interpretation had yielded the same results. There had been other decisions in international case law when the natural meaning of the text had coincided with the historical meaning. But a rule could not be based on coincidences,
84
panos merkouris
to a different interpretation than the one that an original Article 31-based interpretation might suggest. Facing this problem, Mr. de Almeida, felt that the principle of good faith was the golden rule and, therefore, the travaux préparatoires should prevail. However, his awkwardness was evident, since immediately afterwards he stated that he felt that such a solution would destroy the hierarchy between the two articles. Judge Schwebel in his seminal article on this issue of ‘corrective’ interpretation through the use of travaux préparatoires, argued in favour of this function of Article 32 of the VCLT.47 The theoretical basis for this is the principle of good faith and Vattel’s axiom that “[we ought to reject that] which renders the act null and without effect”.48 According to Schwebel’s logical construction if Article 32 is not allowed such a function then its role would be marginalized to the extent of merely confirming the interpretation arrived at through Article 31 of the VCLT. According to him “clauses may not be interpreted to be surplusage”49 and therefore interpretative logic dictates that a ‘corrective’ role should be recognized for Article 32. However seductive as this construction may be it, nevertheless, seems to have an inherent logical error. At this point it is helpful to remember a quote by Bertrand Russell There are various problems as regards language … Fourthly, there is the question; what relation must one fact (such as a sentence) have to another in order to be capable of being a symbol for that other? This last is a logical question …50
It is this last problem that Russell points to, that will be our guide in trying to identify whether the ‘corrective’ function is truly reflective of the role that the travaux préparatoires play in the interpretative process. The issue is merely a logical question, if there is a gap in logic then the initial statement has some
[(emphasis added)] and that was precisely the case with article 28. What would happen if, though the text of a treaty was apparently clear, in seeking confirmation in the preparatory work and other surrounding circumstances a divergent meaning came to light? It was impossible to be sure in advance that those circumstances would confirm the textual meaning of the treaty. If the emphasis were placed on good faith it would appear that in such a case those circumstances should be taken into consideration, although they did not lead to the confirmation of the meaning resulting from the application of article 27. But that would destroy the hierarchy between articles 27 and 28.
Crucho de Almeida, Vienna Conference on the Law of Treaties, Meetings of the Committee of the Whole, 33rd Meeting, para. 56 47 Stephen M. Schwebel, ‘May Preparatory Work be Used to Correct rather than Confirm the ‘Clear’ Meaning of a Treaty Provision?’, in Theory of International Law at the Threshold of the 21st Century, Jerzy Makarczyk (ed.) (The Hague: Kluwer Law International, 1996), 541–7. 48 Emmerich de Vattel, The Law of Nations; or the Principles of the Law of Nature: Applied to the Conduct and Affairs of Nations and Sovereigns (London:G.G.J & J. Robinson, 1792), Book II, Ch. XVII, at 234, para. 283. 49 Schwebel, supra note 47, at 545. 50 Bertand Russell, ‘Introduction’, in Tractatus Logico-Philosophicus, Ludwig Wittgenstein, (London: Routledge Classics, 2005), at X.
interpretative use of travaux préparatoires
85
fault. Bearing in mind this, we return to the examination of the logical validity of the theoretical construction of the ‘corrective’ function of the preparatory work. Article 32 is a two-pronged article. It allows use of travaux préparatoires not only for confirmation purposes but also for determinative purposes. Consequently, lack of recognition of the ability to ‘correct’ would by no means render Article 32 useless,51 as the second prong, i.e. that of determining ambiguous or obscure meanings of terms, would remain intact. This conclusion is further reinforced by the fact52 that the very idea of travaux préparatoires ‘correcting’ the ‘clear’ meaning is a sophism at best.53 In order, however, to prove this point we need to examine whether despite the lack of a theoretical basis, this idea is supported by international jurisprudence. The solitary statement made by Mr. de Almeida accompanied by the fact that Schwebel in his article makes only one passing reference to a singular international case54 begs the question whether de Almeida was the first person to identify this issue. The reality could not be further from the truth. English and German courts, in their domestic jurisprudence had already come across this scenario.55 The international jurisprudence, which is more pertinent to the subject, had already at the time of the VCLT negotiations several examples to furnish; a number which was further augmented in the years to come and through the proliferation of international courts and tribunals. The cases referring to the issue of ‘corrective’ interpretation although each having its own unique characteristics and intricacies may, for reasons of facilitating their study, be categorized in the following groups: A. Cases Where the Issue Was an Unintentional Mistake Incorporated in the Text of the Treaty/Document In the Island of Timor Case the tribunal referred to prior negotiations in order to determine certain boundaries between the Dutch and Portuguese possessions in the Island of Timor. The tribunal scrutinized the travaux préparatoires of the international acts of 1902 and 1904 and came to the conclusion that the difficulties they had in making sense of the 1904 convention sprung from a “falsa demonstratio”.56 The negotiators had mixed up and given the wrong
51
as Judge Schwebel suggested. as will be demonstrated in the following analysis. 53 a crucial point to which we shall return after examining the relevant jurisprudence. 54 Schwebel, supra note 47, at 541, citing his own Dissenting Opinion in the Case of Maritime Delimitation and Territorial Questions between Qatar and Bahrain (Qatar v. Bahrain), Jurisdiction and Admissibility, Judgment of 15 February 1995, 1995 ICJ Rep. 6, at 27 et seq. 55 albeit, resolving it in favour of a textual interpretation, for the British stance see Lauterpacht, supra note 1, at 723; for the German courts’ stance, Id., at 732, citing Decisions of the Reichsgericht, (Civil Cases), Vol. LXXIV, at 81. 56 Lauterpacht, supra note 1, at 756. 52
86
panos merkouris
name to the river in question.57 Consequently the Court decided to interpret the relevant provisions in accordance with the true intention of the Parties and not the erroneous text58; in short it corrected the text. On the other hand, a strict textual interpretation is mirrored in AG Tesauro’s Opinion in the Commission v. Spain case. The issue here was that a legislative text adopted by the Council of Ministers differed from that proposed by the Commission. A subsequent examination of the travaux préparatoires revealed that this was an unintentional error during the linguistic revision done by the Commission’s legal service. For reasons of legal certainty AG Tesauro supported that the European Court of Justice should base its decision on the text published in the Official Journal irrespective of what the travaux préparatoires had brought to light.59 Setting aside the fact that the Court did not eventually deal with this issue, it has to be stressed that this was only an Opinion of an Avocat Géneral that cannot preempt the hypothetical stance of the Court or subtract from the importance and finality of other decided cases that deal with this issue.60 This group of cases is by its nature restricted. Errors entering unintentionally the text of a treaty are more the exception than the rule. Nevertheless a further examination of the other groups may lead us to a more coherent picture of the true value with respect to the travaux préparatoires. B. The Travaux Préparatoires Are Not Referred To but the Possibility of ‘Correction’ is Acknowledged In the case of the Rumanian Minister of War v. The Turkish Government the Tribunal held that a treaty, unlike a law, is a synallagmatic convention and that the travaux préparatoires in order to be useful in the interpretative process, … must establish the common intention of the contracting Parties in order to give to a certain clause of the Treaty a meaning … that is not derived directly from its terms61 (emphasis added)
57 58
Oé-Sunan instead of Kabun or Lèos. the Tribunal held that there is no reason to dwell on the mistake of name made … by the negotiators … when they gave to Kabun or Lèos the name of Oè Sunan, and that there is on the contrary reason to accept that it is this very Kabun or Lèos that the Parties intended to consider as properly serving as a frontier from point A north [and that] [f]rom what precedes [i.e. the examination of the travaux préparatoires], there evolves, in other words, the conviction that the will of the contracting Parties ought to be interpreted in the sense that… the frontier follows the thalweg of the river Kabun or Lèos in a northerly direction as far as the source of this last water-course wrongly called Oè Sunan (unofficial English Translation)
Boundaries in the Island of Timor Case, (Netherlands v. Portugal) (1914), RIAA XI: at 502. 59 Case C-277/93, Commission v. Spain (1994), Opinion of AG Tesauro, ECR (1995): I-5515; CMLR 1 (1995): 866, paras. 8–9. 60 see infra analysis of the Badische Erfrischungs-Getränke case and the Stauder case. 61 (author’s translation), the official text goes as follows: “doivent établir l’intention commune des Parties Contractantes de donner à une certaine clause du Traité une signification … qui ne
interpretative use of travaux préparatoires
87
and that “it must be clearly and incontestably proven that the contracting parties intended to say something other than what they said”.62 Although the Tribunal concluded that the rule was that there is a presumption that the Parties have correctly translated their intention within the terms of the text,63 the previous paragraphs show that this presumption is rebuttable. It is true that the threshold that this judgment sets is high, but it is mandated by legal certainty and the ‘extrinsic’ nature of the travaux préparatoires as an element of interpretation. As long as the travaux préparatoires can offer hard evidence of a contrary to the text intention, then the Tribunal would follow such an interpretation, correcting the meaning of the text.64 This takes us to the next group where not only the possibility of ‘correction’ is acknowledged but the travaux préparatoires themselves are examined.65 C. Reference to Travaux Préparatoires and Acceptance of Possibility of ‘Correction’ In the North Atlantic Fisheries Case the Court left open the possibility of an interpretation other than that offered by the text itself as long as there is “evident proof to the contrary”.66 The landmark Lotus Case contains a rather interesting passage with reference to travaux préparatoires. The examination by the Court of the relevant preparatory work led it to the conclusion that the body of travaux préparatoires did not contain anything “calculated to overrule the construction indicated by the actual terms”.67 Once again although in the present case the Court merely resorted to the travaux préparatoires as a means of confirmation, this dictum seems to allow a contrario for an overruling of the interpretation promoted by the ‘actual terms’ if sufficient evidence to the contrary was furnished by the travaux préparatoires. découlent pas directement de ses termes”, Rumanian Minister of War v. The Turkish Government, reprinted in Recueil des Décisions des Tribunaux Arbitraux Mixtes, institúes par les Traités de Paix, Gilbert Gidel (ed.) (Paris, Recueil Sirey, 1928): Vol. VII, at 996 (hereinafter Recueil des Décisions des Tribunaux Arbitraux Mixtes). 62 (author’s translation), the official text goes as follows: “…, il faut qu’il soit nettement et incontestablement prouvé que les Parties Contractantes ont voulu dire autre chose que ce qu’elles ont dit”, Id. 63 Id. 64 to a similar conclusion arrived the Harvard Research, ‘Article 19: Interpretation of Treaties’, AJIL 29 (1935): 961. 65 albeit in some cases merely for confirmation purposes. 66 North Atlantic Coast Fisheries Case (Great Britain v. United Staets of America) (1910), reprinted in James Scott (ed.), Hague Court Reports (New York: OUP, 1916), 186; the same was reiterated in the Advisory Opinion on the Polish Postal Service in Danzig, 1925 PCIJ (Ser. B), No. 11, 6 et seq.; also of relevance is the David. J. Adams case, where the American and British Claims Arbitration Tribunal based on the travaux préparatoires stated that “the literal meaning of an isolated clause is often shown not to be the meaning really understood as intended” (author’s translation), Annual Digest (1919–1922), Case No. 243, cited in Lauterpacht, supra note 1, at 789. 67 The Case of the S.S. Lotus, (France v. Turkey), 1927 PCIJ (Ser. A), No.10, 17.
88
panos merkouris
An almost identical wording was used by the PCIJ in the Advisory Opinion concerning the competence of the European Commission of the Danube.68 What is striking in this latter case is that the Court in the same decision advocates in favour of two diametrically opposed solutions. Initially the Court pays lipservice to the fact that “preparatory work should not be used for the purpose of changing the plain meaning of the text”.69 This would seem to shut the door completely to any possibility of corrective interpretation through the travaux préparatoires. However, after the relevant examination the Court repeats almost verbatim the aforementioned dictum of the Lotus case.70 This seemingly schizophrenic approach may seem prima facie as impossible to lead to any logical and coherent compromise. One way of resolving this would be simply to accept this contradiction as an error of the Court and that, consequently one of the two statements is without effect. However, Vattel’s axiom that “[we ought to reject that] which renders the act null and without effect”71 and which applies mutatis mutandis to the interpretation of judicial awards, bars any such solution. The only remaining possibility is to examine these statements to see if their contradiction is merely fictional. Indeed, that is the case. The two dicta mentioned use different words. The first uses the term “plain [meaning]” whereas the second uses the word “actual [terms]”. By applying Article 31 of the VCLT the ordinary meaning of these two words is not identical. “Plain” means clear,72 whereas “actual” means real, existing in fact.73 Consequently the first dictum is nothing more than a reformulation of Vattel’s axiom, which the international courts repeat quite often as a mantra, i.e. in claris non fit interpretatio. Thus, the second dictum, refers to all the cases where the first dictum does not apply, i.e. when there is a need of confirmation or when there is ambiguity or obscurity.74 Consequently, not only are these
68 in this case the Court stated that “the records of the preparation of the Definitive Statute do not in the opinion of the Court, furnish anything calculated to overrule the construction indicated by the actual terms of Article 6”, Advisory Opinion No.14 relative to the Competence of the European Commission of the Danube, 1927 PCIJ (Series B), No.14, 28. 69 Advisory Opinion No.14 relative to the Competence of the European Commission of the Danube, supra note 68, at 31. 70 Id., similarly see PCIJ (Ser. A), No. 20–1, at 30; (Ser. A/B), No. 47, at 249–53; In a similar context as to the interpretative value of recourse to travaux préparatoires is the Dissenting Opinion of Judge Anzilotti in the Case concerning the Employment of Women during the Night. In that case, Judge Anzilotti stated that “it would be necessary to refer to the preparatory work, which in such case, would be adduced not to extend or limit the scope of a text clear in itself, but to verify the existence of an intention not necessarily emerging from the text but likewise not necessarily excluded by that text”, PCIJ (Ser. A/B), No.50, at 388; Despite the importance, however, that Judge Anzilotti reserves for the travaux préparatoires his dictum seems to be more in favour of a determinative function of the travaux préparatoires rather than a corrective one. 71 de Vattel, supra note 48, at 234, para. 283. 72 Albert S. Hornsby, Oxford Advanced Learner’s Dictionary (Oxford: OUP, 1995), 880–1. 73 Id., at 13. 74 the issue, to what extent Vattel’s axiom is an obscurantist tautology and whether the mere fact of seeking interpretation shows that a text is not clear will be analysed infra.
interpretative use of travaux préparatoires
89
two dicta not in collision, but the decision itself acknowledges the possibility of an interpretation emanating from the travaux préparatoires weighing more than the interpretation based on a simple reading of the text. Not all cases, however, acknowledge this possibility. In Aschberg Hopwood and Crew, Ltd. v. Quaritch the issue was whether the term “cancelled” in Article 310 of the Treaty of Versailles was intended to mean the same as “dissolved”.75 Although this contention seemed to be corroborated by the travaux préparatoires76 the Tribunal found that the text was clear enough and no recourse to travaux préparatoires was necessary.77 What is important is, however, that the Tribunal acknowledged that [i]t may be that in the correspondence between the High Contracting Parties as to this Article, some mistake has crept in, but here the Treaty has a clear meaning and that meaning must prevail.78
Essentially, the Tribunal set aside the different meaning of the travaux préparatoires by referring to a possible, fictional mistake, in order to reinforce the validity of the text and the solution it adopted; i.e. that in any event, the text was so clear that it should prevail.79 D. Application of the Travaux Préparatoires for ‘Corrective’ Purposes The jurisprudence of the Mixed Arbitral Tibunals created by the Peace Treaties furnishes us with quite a few examples of a ‘corrective’ or ‘quasi-corrective’ interpretation. In the case Polyxene Plessa v. the Turkish Government, despite there being a specific exception in Article 58 of the treaty of Lausanne, stating that the Contracting Parties, with the exception of Greece, renounce any claim for reparations for war damages, the Court was of a different opinion. It held that the text did not reproduce, with sufficient precision, the true intention of the parties80 and that the exception in question was not meant to regulate legal rights and interests but merely to take into consideration certain considerations of a sentimental nature.81 Similar solutions were adopted in other cases as well.82
75
a term used in Article 299 of the said Treaty. Yi-Ting Chang, The Interpretation of Treaties by Judicial Tribunals (New York: Columbia University Press, 1933), at 49–51 77 Aschberg Hopwood and Crew, Ltd. v. Quaritch, reprinted in Recueil des Décisions des Tribunaux Arbitraux Mixtes V (1926): 332. 78 Id. 79 according to Chang, however, this case is more of the exception than the rule, Chang, supra note 76, at 51; the importance of this case is further diminished by the Tribunal’s resort to a fictitious mistake. 80 as clearly demonstrated in the documents of the travaux préparatoires. 81 Polyxene Plessa v. the Turkish Government, Recueil des Décisions desTribunaux Arbitraux Mixtes VIII (1929): 224 et seq. 82 Eastern Bank Ltd. v. the Turkish Government, Recueil des Décisions desTribunaux Arbitraux Mixtes VIII (1929): 188 et seq.; Antippa v. Germany, Recueil des Décisions desTribunaux 76
90
panos merkouris
Although arguments have been made, within the ambit of the ICJ, in favour of the travaux préparatoires being acknowledged such a function,83 it has not been merely the arbitral tribunals, or the ICJ and the PCIJ who have had the monopoly on this issue. The European Court has also found itself between a rock and a hard place when called upon to qualify one form of interpretation over another. The importance of these cases is further highlighted if one considers the fact that the European Court of Justice has been up to very recently quite reserved in its employment of the examination of travaux préparatoires for interpretative reasons.84 Two are the most pertinent cases. In the Badische Erfrischungs-Getränke case the Court held that the text adopted by the Council varied from the one proposed by the Commission and this, it felt, showed that the Council had an intention to deviate from the proposed text.85 What is relevant, though, is that the Court did not dismiss the interpretation based on the travaux préparatoires, ab initio, but weighed both interpretations and simply held that the intention shown in the text itself was more substantive in the case at hand. That the result of this process may differ on a case by case basis is shown by the fact that in the Stauder case the Court opted for the interpretation arising out of the travaux préparatoires and ‘corrected’ the meaning of the provisions in question.86
Arbitraux Mixtes VII (1928); In this last case the contradiction between the travaux préparatoires and the text itself is even more evident. The Tribunal interpreted the term “acts commited” by Germany as incorporating all acts, regardless of whether they were committed in violation of international law or in accordance with it. This clash is further evidenced by the fact that it flies in the face of a considerable amount of jurisprudence stating clearly that the Treaty of Versailles as to that point referred only to acts committed in violation of international law; in more detail see Lauterpacht, supra note 1, at 760–1. 83 see Case Concerning the Maritime Delimitation and Territorial Questions between Qatar and Bahrain (Qatar v. Bahrain), Jurisdiction and Admissibility, Dissenting Opinion of Judge Schwebel, 1995 ICJ Rep. 27, at 27–39; Judge Schwebel based on this Opinion would later on publish the most known Article on this issue of ‘corrective’ interpretation; Schwebel, supra note 47, at 541–7; similarly see also in the aforementioned case the Dissenting Opinions of Judge Koroma, at 69–73 and of Judge Valticos, at 75–6. 84 see Anna Bredimas, Methods of Interpretation and Community Law (Oxford: NorthHolland Publishing Co., 1978), 57–65; Hans Kutscher, ‘Methods of Interpretation as seen by a Judge at the Court of Justice’, in Reports of a Judicial and Academic Conference held in Luxembour in 27–28 September 1976 (Luxembourg: Court of Justice of the European Communities, 1976), 1, at 1–5 & 22; this is probably partly due to the fact that the travaux préparatoires of the founding treaties were up till very recently not available to the public. However, the European Court of Justice seems to be using more and more the travaux préparatoires in its more recent jurisprudence especially with respect to the interpretation of provisions of secondary Community measures; more analytically see Søren Schønberg & Karin Frick, ‘Finishing, Refining, Polishing: on the Use of Travaux Préparatoires as an Aid to the Interpretation of Community Legislation’, European Law Review 28 (2003): 149–71. 85 Case C-17/96, Badische Erfrischungs-Getränke (1997), ECR I-4617; CMLR 1 (1998): 341. 86 Case 29/69, Stauder (1969), ECR 419, CMLR 19 (1970): 112.
interpretative use of travaux préparatoires
91
E. Cases Where the Issue Was Not ‘Correction’ but Nullification This final set of cases can possible raise some doubts as to its correlation with ‘corrective’ interpretation. Nevertheless two arguments can be raised regarding the examination of such cases. Firstly, nullification is merely the extreme of ‘correction’. ‘Correction’ is in essence a generic term that can refer equally to ‘correction’ of a specific provision on the one hand, as well as ‘correction’ of the treaty as a whole on the other. It is the latter notion of ‘correction’ that the act of nullification of a specific provision can correspond to. If a specific provision is completely erroneous with respect to the ‘meaning’ of the rest of the treaty then one way of correction is naturally its deletion. Secondly, a more practical argument can be made that regardless of whether these cases fall per se within the ambit of ‘corrective’ interpretation their examination is, nevertheless, not only beneficial but essential since it yields useful information regarding the limits of the role of the travaux préparatoires in the interpretative process. The Cayuga Indians Claims case is unique in the sense that one party to the dispute, the US, suggested not a corrective interpretation but that a certain provision should be completely discarded because the travaux préparatoires showed that the article in question was “only a ‘nominal’ provision, not intended to have any definite application”.87 The Tribunal’s rejection of this argument cannot be construed as proof of the lack of a ‘corrective’ function of the travaux préparatoires. The foundation for the Tribunal’s decision was that in essence the US requested not an interpretation but a nullification of a provision and for this the Tribunal fell back on one of Vattel’s principles; i.e. that “a clause must be interpreted as to give it a meaning rather than so as to deprive it of meaning”.88 Along similar lines, within the ECJ the opinion has been expressed that travaux préparatoires are “only an aid to interpretation and may not defeat the application of a binding rule of law”.89 Again here, however, the issue is not one of which form of conflicting interpretations should prevail, but whether an interpretation presupposing the actual or in essence nullification of a provision of Community legislation can be allowed.
87 Cayuga Indians Claims Case, American and British Claims Arbitration Tribunal, Award of 22 January 1926, reprinted in AJIL 20 (1926): 587. 88 Id.; this is based on Vattel’s principle, “[we ought to reject that] which renders the act null and without effect”, de Vattel, supra note 48, at 234, para.283; It s not strange that the Tribunal used this principle since Vattel’s principles have influenced the interpretative process for centuries. The most notorious example being the in claris non fit interpretatio principle. 89 Opinion of AG Fennelly in Case C-80-96, Quelle Schickedanz (1998), ECR I-123; CMLR 1 (1998): 373, paras. 26–7.
92
panos merkouris F. Summation
In the preceding analysis we examined the theoretical basis of the proposition that travaux préparatoires could be acknowledged a third function, beyond those recognized in Article 32 of the VCLT; in addition to ‘confirm’ and ‘define’ the meaning of a provision under interpretation, it was examined whether the travaux préparatoires have also a ‘corrective’ role when they seem to support an interpretation that is contrary to the ‘clear’ meaning of the terms. The analysis showed that from a doctrinal point of view this theory seemed to be lacking in concrete arguments. It seemed to be more in the sphere of Sollen rather than Sein. Nevertheless, in order to have a more comprehensive understanding of the subject the relevant jurisprudence was analysed. Bearing in mind that the majority of cases brought before international courts and tribunals touch at some point or another upon the issue of interpretation it would seem, at first glance, bizarre that relatively so little jurisprudence exists on the issue of ‘corrective interpretation’. However, this is quite expected. In most cases the Court utilizes the travaux préparatoires for confirmatory purposes, and in the majority of the remaining to define an ambiguous provision. The vast majority of the jurisprudence confines itself within the limits of Article 32, and only in a handful of cases the clash between travaux préparatoires and text is raised and in most of these this clash is hinted at through obiter dicta. Judges will usually avoid making such a clear reference to a disputed topic if there is another more conciliatory way out of that bind. Consequently, despite their shortcomings in numbers, the aforementioned cases can provide the necessary basis to draw conclusions regarding the attribution or not of a ‘corrective function’ to the travaux préparatoires. Initially, it would seem that the relevant jurisprudence contradicts our original conclusions regarding the theoretical validity of the additional role of the travaux préparatoires. The cases mentioned above seem to allow for a ‘corrective function’. However, were we to halt our inquiry at this point we would have committed a grave error in logic.90 The previous conclusion is predicated upon the existence of such a function. This is by no means a given. Only through the process of elimination can we be certain that such a conclusion is valid. If none of the already existing roles of the travaux préparatoires91 do not suffice to explain the previous jurisprudence, then and only then can we accept that the travaux préparatoires can ‘correct’ the ‘clear’ meaning of the text. In doing so we need to revisit the theory of the ‘corrective function’ of the travaux préparatoires. This theory has as a starting point that Vattel’s axiom
90 91
falling prey to Russell’s fourth question/problem about language. as enshrined in Article 32 of the VCLT.
interpretative use of travaux préparatoires
93
in claris non fit interpretatio is in essence an “obscurantist tautology”.92 If a clear meaning exists then there would be no reason to go into the interpretative process, let alone correct the ‘clear’ meaning. This is, precisely, the pitfall of the whole theory. If we follow this theory to its logical conclusion, three are the possible scenarios that arise: i. the travaux préparatoires are used merely to confirm the meaning already arrived at through Article 31; ii. the text is ambiguous and the travaux préparatoires are used to resolve this ambiguity or iii. the text is ‘clear’ but the travaux préparatoires offer a different interpretation. However, this third possibility is not valid. As was mentioned earlier the entire theoretical construct of ‘corrective interpretation’ is predicated upon denying Vattel’s axiom. Nevertheless, in order to talk about a clash between the text and the travaux préparatoires, Schwebel’s theory requires the existence of a clear meaning (see previous paragraph); it requires the validity of Vattel’s axiom. But it is precisely this axiom which has already been discarded, at the very start of the ‘corective interpretation’ hypothesis. Essentially, the ‘corrective function’ of Article 32 is an interpretative Catch-22. For its existence it requires what it has already rejected. If Vattel’s axiom is an “obscurantist tautology” then ‘corrective interpretation’ as a separate and new function of Article 32 is an obscurantist self-negation. But what then of all the jurisprudence that we analysed? Once again, first impressions may be deceiving. What remains to be seen is whether the aforementioned jurisprudence may yield different results examined under the light of these considerations. If one is left only with the functions of ‘determination’ and ‘confirmation’, only the former could be of any assistance. Article 32 of the VCLT requires ambiguity or obscurity. Neither the text nor the travaux préparatoires of the VCLT itself delineate93 the level of ambiguity required. It was left up to the judges to decide when to consider the text ambiguous enough to resort to Article 32. Even the least amount of ambiguity could suffice.94 The fact that a provision enters the interpretative process means ipso facto that there is ambiguity. It is then the judge’s task to decide whether this ambiguity
92 Comments by Myres S. McDougal, of the United States Delegation to the Committee of the Whole of the Vienna Conference on the Law of Treaties, in Vienna Conference on the Law of Treaties, Meetings of the Committee of the Whole, 31st Meeting, at 167, para. 38. 93 and this was probably done intentionally bearing in mind that for the same reason they restrained from defining what exactly is travaux préparatoires, what are the other “supplementary means” mentioned in Article 32 etc. 94 as is evidenced by the extensive jurisprudence of the ECJ, in more detail see Schønberg & Frick, supra note 84.
94
panos merkouris
is tantamount to zero or whether, taking into consideration all the relevant circumstances, that it is sufficient for the activation of Article 32. Bearing these considerations in mind, the jurisprudence analysed supra, can be seen in a different light; not as a contradiction between the interpretation of the travaux préparatoires and that of the text, but as a clarification, a determination of ambiguous terms. All the cases where the ‘corrective function’ allegedly occurred or was hinted at can be understood as merely an application of the ‘determinative’ function.95 What takes place is not correction as a separate function of Article 32, but merely a clearing-up of the ambiguity or absurdity arrived at through Article 31, and a determination of the meaning of a provision. This is exactly the point; it is not a separate function, but it is already integrated in the ‘determinative’ function. Within Article 32 correcting the meaning of a provision is merely just one extreme manifestation of determination, not a separate function. IV. Conclusion The hero in Fahrenheit 451 eventually ends up with a group of people, who are essentially ‘living books’. Each has memorized a book in order to salvage it from destruction. One of them tries to make a point to the hero96 that “[you’re not important. You’re not anything. Some day the load we’re carrying with us may help someone”.97 Similarly, it can be said for the travaux préparatoires that they are not important as such, without any context of reference. However, this does not mean that they are useless and, thus, disposable. On the contrary, they become of the utmost importance during the interpretative process for the reason mentioned in the quote above; “[s]ome day the load [they]’re carrying with [them] may help someone”.98 And that someone is the judge. The previous analysis showed that the current practice allows for quite a bit of flexibility and expansiveness as to the use of the preparatory work with respect to third parties. This enhances the importance and field of application of the travaux préparatoires. This holds true even for the second area of investigation; ‘corrective’ interpretation. The fact, that the analysis of the previous sections showed the fallacies of this theory does not subtract from the importance of the travaux préparatoires or diminish its function. Although the
95 and even in the cases where the Court did not opt for the solution suggested by the travaux préparatoires that also falls within the established structure of Article 32 of the VCLT and the competence of the judge to weigh the relevance and evidentiary importance of the travaux préparatoires. 96 who has also memorised parts of books. 97 Ray Bradbury, Fahrenheit 451 (New York, NY: Harper Collins Publishers, 2004), 171. 98 Id.
interpretative use of travaux préparatoires
95
‘corrective’ function was not added to the existing ones of Article 32 i.e ‘confirmatory’ and ‘determinative’, this does not mean ipso facto that there is less impact of the use of preparatory work. Sometimes ‘less is more’ and this is such a case. All the cases that were supposedly supporting this theory of a ‘corrective’ function, were shown to easily fall within the scope of one of the two existing functions; either the ‘determinative’ or the ‘confirmatory’ one. Why then do we need to invent a new function, when the existing ones more than amply compensate? Even more so, when the invented function suffers from logical errors and inconsistencies. It is not the lack of such a function that subtracts from the interpretative importance of the travaux préparatoires, it is the invention and support of expansive functions which may weaken the foundation of Article 32 of the VCLT, especially when there is absolutely no lacuna to be filled. Consequently, our rejection of the ‘corrective’ function, as a separate function of Article 32, is in essence an affirmation of the still great importance of the role of the travaux préparatoires in interpretation. In response then to the question posed in the title of this article; travaux préparatoires are far from facing a ‘Fahrenheit 451’ future, they are similar to the final group of people in the book; the living, or if not living, at least the existing memory of what transpired. Fragmented, perhaps, inconsistent, true as well, but memory nonetheless. And if one insists on a definitive answer on what is the role of the preparatory work in the interpretative process then the answer was and remains the same: And when they ask us what we’re doing, you can say, We’re remembering! 99
99
Id. (emphasis added).
Part II
Treaty Interpretation, International Trade and Investment Law
THE APPELLATE BODY AND TREATY INTERPRETATION* Georges Abi-Saab I. Introduction The participants in this final roundtable—Judge Gilbert Guillaume of the International Court of Justice, Judge Paolo Mengozzi of the European Court of Justice, Court of First Instance, Judge Allan Rosas of the European Court of Justice, Judge Tullio Treves of the International Tribunal for the Law of the Sea, and myself—are invited to partake in a Brechtian play, to exercise “distanciation”: each actor, while playing his role, sheds a critical look at the persona he is incarnating, as if he were evaluating its behaviour and commenting on its predicament from the outside. The actors here are international judges, and they are invited to reflect, at arms length, on the role they are playing on their diverse benches with respect to treaty interpretation. However, in order for the roundtable to yield a meaningful comparison and conclusion, these reflections have to follow parallel lines. In consequence, the participants have been asked, each from the perspective of his bench, to address the following questions: 1. To what extent do they rely explicitly on the customary rules of interpretation of public international law, and particularly by referring to their codified version in articles 31 and 32 of the Vienna Convention on the Law of Treaties (Vienna Convention)?; and to what extent do they rely on special rules of interpretation, whether in addition to, or in substitution for, the above-mentioned general rules? 2. How do they consider that their respective institutions would be characterized if their record were to be situated, in terms of judicial policy, along the scale going from strict constructionism to teleological interpretation? 3. To what extent have their respective fora gone in their use of interpretative methods, beyond the interpreted instruments, by calling, for example, on paragraph 3(c) of article 31 of the Vienna Convention (‘any relevant rules of international law applicable in relations between the parties’), and more particularly through reference to general international law? * This article is reprinted from: Georges Abi-Saab ‘The Appellate Body and Treaty Interpretation’, in The WTO at Ten: The Contribution of the Dispute Settlement System, Giorgio Sacerdoti et al. (eds.) (Cambridge: CUP, 2006), 453–64, with the kind permission of the author and the publisher. It was originally presented in a round-table composed of judges of several international tribunals, analyzing the approach of their respective benches to treaty interpretation.
100
georges abi-saab II. The Specifities of the Appellate Body
Before attempting to answer these questions from the perspective of the Appellate Body of the WTO, a preliminary issue has to be addressed that does not arise in relation to the fora of the other participants; an issue epitomized in the name or designation of the institution as a ‘body’ rather than court or tribunal. It is the issue or the question that is thus raised concerning the legal nature or characterization of the Appellate Body and of its belonging to the ‘judicial universe’, which necessarily determines the modalities of its activities, including interpretation. In his excellent essay in this volume, Peter van den Bossche, rightly describes the establishment of the Appellate Body as an ‘afterthought’.1 Not that the founding fathers did not know what they were doing in Marrakech. They obviously wanted to establish a double-deck procedure of dispute settlement. But all indications converge in pointing out that they had not thought out all the ramifications of what they were doing. It seems, according to the recollections of some of the negotiators of the DSU, that the Appellate Body was initially conceived as an exceptional recourse to harness the odd ‘rogue panel’, not – as it turned out to be – as a genuinely self-standing organ, not standing idly, but functioning in continuum to review over half of all the released panel reports. The ambiguities in the language of the DSU and in some of the Appellate Body’s institutional features, reflect the heritage of the GATT, with the insistence that it was a mere agreement and not an organization, that it had no autonomous organs (that is, organs with autonomous, though delegated, powers), with all decisions taken by the ‘contracting parties’, acting collectively, whence the necessity of consensus; a heritage very much alive in the WTO, with the formula: ‘this is a member-driven organization’ serving as a mantra. These ambiguities also reflect the hesitancy of some members, and not the least of them, to surrender explicitly any final decision, be it to a judicial body. Thus, apart from the denomination, panels and Appellate Body issue ‘reports’, not judgments or awards. These reports, though making ‘findings’ and ‘determinations’, do not draw the legal consequences in the form of binding decisions addressed to the parties (res judicata), but rather deliver ‘recommendations’ addressed to the Dispute Settlement Body (DSB) – the plenary political organ – which is the one that adopts them formally; and it is its decision, and not that of the panel or the Appellate Body, that finally binds the parties by the legal findings (‘recommendations and rulings’) of the report and their ensuing legal consequences. 1 Peter van der Bossche, ‘From Afterthought to Centerpiece: the WTO Appellate Body and its Rise to Prominence in the World Trading System’, in The WTO at Ten: The Contribution of the Dispute Settlement System, Giorgio Sacerdoti et al. (eds.) (Cambridge: CUP, 2006), 289.
the appellate body and treaty interpretation
101
It is true that this adoption of the report by the Dispute Settlement Body (DSB) has become a mere ritual or formality in the WTO. This is because the decision-making procedure for the adoption of the reports has passed from the purely consensual system under the GATT – requiring a ‘positive consensus’ of the ‘Contracting Parties’, including the parties to the dispute (and thus giving each of them, the same as the other contracting parties, a right to veto such adoption) – to the automatic or mandatory system of adoption under the WTO, unless the same all-inclusive consensus is achieved to set the report aside (commonly referred to as ‘negative’ or ‘reverse consensus’).2 In this respect, it suffices to ponder the functions of the dispute settlement system, as laid down in the DSU, to discover that the Appellate Body is entrusted with a purely judicial function. Indeed, according to paragraph 2 of Article 3 of the DSU: a. the dispute settlement system aims at providing “security and predictability to the multilateral trading system”; b. this is to be achieved not by any means, but by the legal means of “preserv[ing] the rights and obligations of Members under the covered agreements” (without “adding to”, or “diminishing” these rights and obligations); c. and this, through ‘clarifying’ (i.e. interpreting) the provisions of the covered agreements “in accordance with the customary rules of interpretation of public international law”. Moreover, within that system, the Appellate Body’s jurisdiction is limited to reviewing “issues of law…and legal interpretations” in panel reports (Art. 17.6 of the DSU). How can any entity fulfill these functions and abide by these injunctions – which express, albeit clumsily, some of the major parameters of the judicial function – except by acting as a judicial organ? Still, the ambiguities in the language and the institutional procedures, as well as the common reference to the Appellate Body as a ‘quasi-judicial’ organ, combine to maintain, be it subconsciously, the question about its legal nature, and have cast it from the outset into the pirandellian role of “an organ in search of identity”. However, once established, institutions evolve according to their innerdynamics; international experience reveals that an entity, however ambiguous or lacunary its institutional make-up, once it perceives itself as entrusted with
2 It should be noted, however, that “[t]his adoption procedure is without prejudice to the right of Members to express their views on an Appellate Body report” (Art. 17/14 of the DSU); a right they readily exercise, usually during the session of the DSB in which the report is declared adopted.
102
georges abi-saab
the exercise of the judicial function, evolves according to a legal genetic code towards greater judicialisation. Thus, in the manner of a self-fulfilling prophecy, the Appellate Body has, from the outset consciously and systematically affirmed and consolidated its judicial character both in its modalities of functioning and its processes of reasoning. Particularly, throughout its decade of existence, the Appellate Body’s approach to interpretation has been a constant affirmation by the Appellate Body of itself as a judicial organ. The fact that the Appellate Body conducts itself as a judicial organ, regardless of the denomination, the ‘quasi’ qualification and the rest – i.e. the fact that the ambiguities raised by these elements do not affect the “judicial process” itself- does not mean, however, that they have no influence at all on the Appellate Body, and, more particularly, on its “judicial policy”, as shall be seen below under part IV. III. The Appellate Body’s Methodology of Interpretation Turning to the methodology of interpretation followed by the Appellate Body (and to the extent to which it relies on, and explicitly refers to, the customary rules of interpretation of public international law and their codified versions in Articles 31 and 32 of the Vienna Convention), it should be recalled that the original GATT 1947 says nothing about interpretation. Indeed, the GATT 1947 did not regulate in any detail the process of dispute settlement, limiting itself merely to establishing the principle of a right to consultation in Article XXII, and of the jurisdiction of the ‘Contracting Parties’ ultimately to authorize the suspension of concessions in cases of “nullification or impairment” of benefits in Article XXIII. The system evolved, with the emergence of panels that progressively patterned themselves after the ad hoc arbitration model, without reaching its threshold, as the adoption of their reports remained subject to the positive consensus of the Contracting Parties, including the parties to the dispute. And no specific methodology of interpretation emerged from the jurisprudence of these panels, though it was dominated in general by a search for the initial intention of the Contracting Parties. By contrast, the DSU establishes an elaborate system of dispute settlement and, as was mentioned above, provides as one of its functions “to clarify the existing provisions of [the covered] agreements in accordance with customary rules of interpretation of public international law”(Article 3.2 of the DSU)… In its first report, the Appellate Body identified what it described as “a fundamental rule of treaty interpretation [whose] most authoritative and succinct expression” figures in Article 31(1) of the Vienna Convention, which provides: A Treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose.
the appellate body and treaty interpretation
103
Characterizing it as the “general rule of interpretation” (the title of Article 31 in the Vienna Convention), the Appellate Body specifies that this rule “has attained the status of a rule of customary or general international law”, and that [a]s such, it forms part of the ‘customary rules of interpretation of public international law’ which the Appellate Body has been directed, by Article 3.2 of the DSU, to apply.3
In its second report, referring to its above-quoted statement about the legal status of Article 31 of the Vienna Convention, the Appellate Body added: There can be no doubt that Article 32 of the Vienna Convention, dealing with the role of supplementary means of interpretation, has also attained the same status [of a rule of customary or general international law].4
Thus, from the outset, the Appellate Body adopted a clear stand and a comprehensible (albeit obvious) methodology on interpretation. This was particularly important, as interpretation lies at the heart and constitutes the bulk of the work of the Appellate Body, given the limitation of its mandate to reviewing “issues of law … and legal interpretations” (Article 17.5 of DSU). Giving clear directives on interpretation was made even more necessary by the very wide coverage and the extreme degree of detail, specificity, and complexity (not to mention the loose drafting) of the agreements, which render the process of interpretation more hazardous and laborious. It was also very useful to give guidance to the members of the larger legal community who work with the agreements – whether within the legal systems of Member governments or as their representatives and lawyers on the international level – as well as to the members of WTO panels and secretariat; these being largely commercial or trade lawyers who are not necessarily conversant with the arcane subtleties of international law. Having been clearly laid down, how have these guidelines concerning the methodology of interpretation been applied or followed in practice? The novelty of the methodology for the legal community that was called upon to apply it was reflected in the laborious and mechanistic handling of Articles 31 and 32 of the Vienna Convention, manifest particularly through two remarkable symptoms from which even Appellate Body reports were not always immune. In the first place, not only were these two articles reproduced or cited at every breath, whether warranted or unwarranted; but, more seriously, they were treated as if they were substantive provisions of the covered agreements
3 US – Standards for Reformulated and Conventional Gasoline (29 April 1996), Appellate Body Report, WT/DS2/AB/R, DSR 1996: I, 3, at 16–7. 4 Japan – Taxes on Alcoholic Beverages (4 October 1996), Appellate Body Report, WT/DS8/ AB/R, WT/DS10/AB/R; WT/DS11/AB/R/Section D, DSR 1996: I, 97, at 104.
104
georges abi-saab
that are directly applicable as such, and themselves interpreted literally, in the manner of GATT and WTO panels. This approach is totally oblivious of the legal nature of principles of interpretation as ‘secondary rules’ (or ‘rules on rules’), that is, rules governing the lifecycle of the ‘substantive’ or ‘primary rules’ laying down the prescribed conduct of the legal subjects; and this apart from the perennial controversy over the true nature of these principles as legal rules or mere logical rules of reasoning. In any case, the DSU refers to the “customary rules of interpretation of public international law” and not to Articles 31 and 32 of the Vienna Convention; articles that purport to codify these customary rules, that is, to provide a written representation or a snapshot of them at a certain moment. But the representation may not be completely faithful, or may not reflect certain aspects of the customary rules that may also continue to evolve beyond the moment the snapshot was taken. This is why in its most recent reports, the Appellate Body uses a more guarded language in this respect, such as “the principles codified in Articles 31 and 32 of the Vienna Convention”;5 or “subsequent practice … within the meaning of the customary rule of interpretation codified in Article 31(3)(b) of the Vienna Convention”.6 The other remarkable symptom in the handling of the principles of interpretation through Articles 31 and 32 of the Vienna Convention is the way the process of interpretation is perceived and followed: as a rigid sequence of autonomous or discrete steps, each of which has to be explicitly addressed and ‘exhausted’, before moving on to the next one.7 There is obviously a certain logic in the sequence laid down in “the general rule of interpretation” of Article 31(1), and in the progression from Article 31 to Article 32. The interpreter has to start with the hard core of the operation, which is the text to be interpreted (that is, the words and language of the provision); before moving from the text, if need be (that is, if the text is not clear enough), but still by way of internal inference (‘internal’ to the instrument bearing the interpreted text), to context, which consists of the structure and the other provisions of the instrument (and related instruments), to the ‘object and purpose’ of this instrument, if the object and purpose can be fathomed from its provisions, including its preamble where they are frequently expressly stated.
5 United States – Countervailing Duties on Certain Corrosion-Resistant Carbon Steel Flat Products from Germany (28 November 2002), Appellate Body Report, WT/DS213/AB/R, para. 61. 6 Chile – Price Band System and Safeguard Measures Relating to Certain Agricultural Products (23 September 2002), Appellate Body Report, WT/DS207/AB/R, para. 213. 7 See for recent examples, the Appellate Body Reports in US – Gambling and EC – Chicken Cuts.
the appellate body and treaty interpretation
105
Only if internal inference fails to clarify the point at issue, would the interpreter resort to ‘external inference’ (that is, look ‘outside’ the interpreted instrument), whether the ‘subsequent practice’ of the parties (that Article 31(2)(b) of the Vienna Convention treats as an informal or tacit agreement by conduct of the parties to the treaty on its interpretation); or lastly, the preparatory works and circumstances surrounding the conclusion of the treaty (that Article 32 describes as ‘supplementary means of interpretation’). However logical this progression may be as an analytical and pedagogic device, interpretation remains one integrated operation, which uses several tools simultaneously to shed light from different angles on the interpreted text; tools that should not be seen as watertight compartments or as a series of separate sub-operations but, rather, as connected (even overlapping) and mutually reinforcing parts of a whole; of a continuum or a continuous and multifaceted process that cannot be reduced to a mechanical operation and which partakes as much of art (the art of judgement) as of science (the science of law). Be that as it may, the Appellate Body has abundantly referred to the different principles of interpretation in Articles 31 and 32 of the Vienna Convention, as well as to the principle of effectiveness or ‘effet utile’ that can be implied from these provisions.8 However, the degree of emphasis and the frequency of reference vary greatly among them, as a function of the ‘judicial policy’ of the Appellate Body, a subject to be discussed later on. A final remark on the principles of interpretation is warranted. It is noteworthy that, despite the oft claimed specificity of international economic law, and more particularly of international trade law, and even more so of WTO law, a thorough look at the jurisprudence (case-law) of the Appellate Body does not reveal any mention of, or reference to, one or more rules of interpretation specific to this particular field that would come to complement or substitute for the above-mentioned general rules. But, of course, the specificity of the subject-matter may affect the relative weight attributed to the various principles of interpretation, which is a question of ‘judicial policy’ as mentioned above, and to which we now turn. IV. The Appellate Body’s ‘Judicial Policy’ of Interpretation The reference to the customary rules of interpretation of public international law and their identification as those codified in Articles 31 and 32 of the Vienna Convention is but a first step that does not tell the whole story. These rules are a toolkit used by all international fora; but they are used in different 8 For a good example of these references, see WTO Appellate Body Secretariat, WTO Appellate Body Repertory of Reports and Awards 1995–2004 (Cambridge: CUP, 2005), 179–89.
106
georges abi-saab
ways, yielding varied results. This is because, as anyone who has had the opportunity to sit on more than one of these fora knows, although they are all supposed to undertake the same type of activity, that is, to exercise the judicial function, of which interpretation occupies a substantial part, each of them constitutes a legal universe quite different from the others. The difference can be explained not only by the disparity in the structure, mandate, and composition, but also, and perhaps mainly, in the ‘judicial policy’ of each forum, which is also a function of its environment. The Permanent Court of International Justice (PCIJ) was unique in that it was the first and only standing international judicial organ on the universal level with jurisdiction to apply general international law. As such, it could act as the organ of the international legal system at large, and thus managed to develop in full autonomy—together with its successor the International Court of Justice (ICJ)—a clear concept of the international judicial function with its components and limits; a concept that served as a model for all the other international tribunals that came later on. But beyond this hard-core concept, each judicial organ develops its own judicial policy to adapt the exercise of its judicial activity to its environment. And the environment weighs more heavily on the formulation of the judicial policy when it is a close and closed environment, as in the case of the Appellate Body, where the organ is part of a larger institutional setup in charge of the ‘management’ of a specialized field of legal regulation. This is in addition to the special historical factor in the present case of the GATT heritage (described above under part II), which still conditions the mindset of the WTO Membership. These factors combine to explain the judicial policy of the Appellate Body on interpretation, which appears, at first glance, as belonging to the strict constructionist school that interprets texts literally and narrowly. Indeed, there is a great emphasis on words, even individual words, greater than in any other international forum, which verges on obsession (recalling to mind the description coined by Professor René-Jean Dupuy of “obsédé textuel”). This is reflected by the abundant references to dictionaries. Indeed, the Oxford Shorter Dictionary is perhaps the most quoted title in the Appellate Body reports; in any case, more than any covered agreement. It is worth noting, however, that the Appellate Body has recently shown some reserve about the use of dictionaries to identify the ‘natural meaning’ of words.9 Another manifestation of the strict constructionist and literal approach is the tendency to stick the reasoning very closely to, and keep it in constant contact with, the words (the interpreter proceeding in a manner reminiscent
9
United States – Measures Affecting the Cross-Border Supply of Gambling and Betting Services (4 July 2005), Appellate Body Report, WT/DS285/AB/R, para. 166.
the appellate body and treaty interpretation
107
of the definition of ‘sedentary species’, in the United Nations (UN) Convention on the Law of the Sea, as being “unable to move except in constant physical contact with the seabed or the subsoil”).10 Faced with this obsession with words, with constructions such as “the second word in the last part of the third sentence in paragraph 2 of article …”, the man from Mars may ask: “Where is the reasoning in all that?” Indeed, the reasoning seems to disappear in a maze of words. But, in fact, it is there, though camouflaged in this seemingly wild word- (rather than goose-) chase. The question remains: If the reasoning is there, is it not better to shed the camouflage? This judicial policy of strict constructionism is followed basically in interpreting the covered agreements, but much less, however, in interpreting the DSU, for reasons to be discussed later on. Beyond the text, resort to context in interpretation, though frequent, can be rather guarded, particularly when it implies going from one covered agreement to another. Object and purpose is much less used or referred to, at least expressly. Of course, it does not sit well with strict constructionism. The object or finality is ‘telos’ in Greek, and object and purpose lead to teleological interpretation, which is the opposite of strict constructionism. In practice, however, much of the reasoning in interpretation is informed by the object and purpose, either consciously or subconsciously, where they can be identified, even though they may not figure explicitly as such in the analysis. Indeed, they are frequently disguised in the search for ‘effet utile’, or even the initial common intention of the Contracting Parties. V. WTO Law and General International Law Finally, given what has just been said, can there still be any place for the use of interpretative methods that go beyond the interpreted instrument, particularly paragraph 3(c) of Article 31 of the Vienna Convention, which brings to the fore the question of referring to, or applying, general international law? The answer to this question has to be given against the background of the heritage of the GATT (described above under part I), with its surviving mantra that “this is a Member-driven organization” and that Members control everything; which generates the impression among them that the agreements are self-sufficient, constituting a hermetic or ‘self-contained regime’; an impression reinforced by the injunction of Article 3.2 of the DSU to the dispute settlement system not to add to or diminish the rights and obligations of the parties under the covered agreements.
10
Article 77.4 of the UN Convention on the Law of the Sea.
108
georges abi-saab
However, as we have already seen, the DSU itself directs the Appellate Body to interpret the covered agreements “in accordance with customary rules of interpretation of public international law” (Article 3.2 of the DSU). And the Appellate Body, in its very first report discussed above, after citing this provision of the DSU added: “That direction reflects a measure of recognition that the General Agreement is not to be read in clinical isolation from public international law”.11 Does this “measure of recognition” extend beyond the rules of interpretation? This has been questioned by some. But the rules of interpretation themselves include Article 31(3)(c), which refers to “any relevant rules of international law applicable in relations between the parties”. These a fortiori include the rules of general international law which are applicable to all members of the international community. But this provision does not provide the criteria (or the circumstances) warranting such reference. The debate within the Membership is about ‘filling gaps’. It all depends, however, on the gaps we are talking about. In this respect, it suffices to make three remarks by way of conclusion. First, the DSU provides a bare outline of the dispute settlement system. But, for the Appellate Body to exercise its judicial activity, it needs detailed rules of procedure and evidence, which are hardly provided therein. The Appellate Body had to formulate these rules, particularly the rules of evidence, from scratch, by reference to the general principles of law and of international law. And nobody protested against that. Secondly, the covered agreements are treaties providing substantive legal regulation of world trade. But as treaties, their lifecycle is governed by the law of treaties which is composed of ‘secondary rules’ not reproduced in their provisions, but in the Vienna Convention. Thus, whenever an issue of the law of treaties as such arises in relation to the covered agreements (and unless it is expressly addressed in the relevant agreement), the Appellate Body has as a matter of course referred to the Vienna Convention; again, without raising any controversy. Thirdly, the controversy may arise when the reference to general international law concerns the substantive regulation (that is, the primary rules) of the covered agreements. But here, again, there is room for distinction. In the first place, Article XXII of the GATT defers to the compulsory decisions of the Security Council under Chapter VII of the UN Charter. This is but one example of mandatory international law, the wider category of which are the peremptory or jus cogens rules of general international law, such as the prohibition of individual use of force and of genocide.
11
US – Gasoline, Appelate Body Report, supra note 3, at 16–7.
the appellate body and treaty interpretation
109
Beyond that, while not technically jus cogens, there are certain general principles of international law, or of law tout court, without which it is impossible to imagine how any legal system can function—in other words, principles inherent in the concept of legal system itself—such as the principles of good faith and proportionality. Still, the reference to these two principles by the Appellate Body has raised some controversy, which leaves open the question of the limits of reference to general international law for the time being.
INTERPRETATION IN INTERNATIONAL TRADE LAW Brigitte Stern I. A Theoretical Approach of the Problem of Interpretation A. The Scope of Interpretation In principle, judges apply existing law. If we retain this premise, it would appear that judges do not play any creative role, that case-law is not a source of law. This is all the more so in international law, where the principle of the ‘relative effect’ of judicial decisions applies. This principle, laid down in Article 59 of the ICJ Statute, is a consequence of State sovereignty.1 Nevertheless, in order to apply a rule, it must first be stated and then interpreted. The issue is thus whether international judges hold any creative power at the two stages of statement and interpretation of the rules. I believe that an affirmative response may be given in both cases. 1. The Margin of Interpretation in the Statement of Customary International Law We all know that custom plays an important role in international law: a customary rule, born of practice, is an unwritten rule. As such, in some cases, its existence only becomes certain once it has been stated by a judge. It is well known that the existence of a customary rule implies the presence of a substantive element and a subjective element of opinio juris. In case of a dispute between States concerning the existence of a customary rule, the judge must verify the presence of these two elements before stating the rule. As regards the substantive element, the judge does not have much room to maneuver: either a certain practice exists or it does not. That is not the case, however, for opinio juris, a purely subjective element, the existence of which is not always easy to prove. Kelsen believed that the presence of this element gives the judge a great deal of room to maneuver and that, in reality, it simply serves to mask the judge’s role in the creation of law. More recently, one author went so far as to claim that a customary rule is often no more than a “legal hypothesis confirmed by a judicial decision”.2 1 ICJ Statute, Art. 59: “The decision of the Court has no binding force except between the parties and in respect of that particular case”. 2 The complete text reads in French:
En effet, l’induction conduit à une hypothèse, à la différence de la déduction, qui contraint à une conclusion. Cette hypothèse, lorsqu’elle est retenue par le juge, prend force juridique. Ainsi, du point de
112
brigitte stern
I consider that the importance of custom in international law enhances the role played by judges in the creation of the law. Yet, I also believe that the judge’s creative role exists in every legal system, since every rule must be interpreted before it can be applied. 2. The Scope of Interpretation of the Meaning of a Conventional Rule of International Law In reality, in my opinion, every interpretation involves an element of creation. Yet, as will be seen, not all lawyers agree on this point. (a) The Different Types of Interpretation When applying a legal rule, the meaning or signification of the rule must first be ascertained. There are differing ways of applying a rule: – either by laying down a subsidiary rule of application; – or by making a concrete act of individual application; – or by resolving a conflict of interpretation concerning the application of the rule. In all these cases, applying the rule requires interpretation, i.e. a determination of its meaning, of its true content. However, it is possible to distinguish two very different forms of interpretation depending on the body called upon to determine the meaning: – what may be called ‘authentic interpretation’, when the interpretation is effectuated by bodies that are empowered to apply the law; and – what may be called ‘non-authentic interpretation’, by private individuals who attempt to apply a law in order to avoid a sanction, or by jurists, law professors, etc. The question, then, is what constitutes an authentic or non-authentic interpretation of international law. In international law, those subject to the law are often the same as the ones that are empowered to apply the law. Thus, contrary to what happens in national law, interpretation by the subjects of international law, which are at the same time subjects and authors of the law, is an authentic interpretation. As already stated, the application of a legal rule always involves interpretation. However, the scope of such interpretation may vary. vue de la logique, la coutume est une hypothèse juridictionnelle, un raisonnement inductif, consacré par l’autorité de la chose jugée.
Max Gounelle, La Motivation Des Actes Juridiques En Droit International Public: Contribution à Une Théorie de l’ Acte Juridique en Droit International Public (Paris: Pedone, 1979), 75–6.
interpretation in international trade law
113
(b) The Different Degrees of Interpretation Various types of indeterminacy in the concrete application of a norm may be distinguished. a. First, there is what Kelsen called the ‘relative indeterminacy’ in the application of the law, which could more clearly be called “minimal indeterminacy”. In other words, a norm can never regulate every detail of the inferior norm or its concrete application. There is always an irreducible minimum of indeterminate elements which are, thus, subject to interpretation. b. In some cases, there may be ‘intended indeterminacy’, aiming to leave the body applying the rule a certain degree of freedom: e.g. a criminal law providing for alternative penalties of prison or a fine. Also, many international treaties, or some of their clauses, remain voluntarily vague, in the absence of agreement between the parties. Difficulties are thus postponed until the time when the treaty will have to be applied, i.e. interpreted. c. Finally, there are the cases that generally raise the most complex questions of interpretation, where there is ‘unintended indeterminacy’. The three most frequent examples are: – an unclear linguistic meaning; – a linguistic meaning in contradiction with the real will of the author of the norm; and – a contradiction between two norms, or two aspects of the same norm. Having listed the varying degrees of indeterminacy requiring interpretation of the applicable norm, it is necessary to present successively: – the traditional theory of interpretation; and – Kelsen’s theory of interpretation. B. The Traditional Theory of Interpretation According to the traditional, most widespread theory, the aim of interpretation is to clarify the meaning of an applicable norm, through the use of a certain number of methods of cognition. In other words, interpretation is an intellectual operation which should lead to the determination of the correct – and the only correct – meaning of the norm, out of a number of possible meanings. The most frequently used methods are: – – – –
the search for the presumed intention of the author(s) of the norm; a strict literal analysis of the text; use of a contrario reasoning; a reasoning by analogy, etc.
114
brigitte stern
Kelsen rejected the use of such methods, which he considered to be purely arbitrary. He did not believe that the arbitrary application of one or the other of these intellectual processes could be used as a basis, in positive law, for what might be termed the exact solution, or the necessary solution. According to him: It is a matter of complete indifference whether one neglects the text in order to stick to the legislator’s presumed will, or strictly observes the text in order to avoid concerning oneself with the legislator’s (usually problematic) will.3
In the same way, in any given case, recourse to a contrario reasoning or reasoning by analogy will lead to opposing results; yet there is no legal reason to prefer one method over the other. C. Kelsen’s Theory of Interpretation For Kelsen, every norm defines a general framework within which there are a number of possible applications. Thus, in his opinion, there is not a single possible application of each norm, or sole exact interpretation, but rather numerous possible interpretations. In this context, it is important to distinguish the two types of authentic and non-authentic interpretation, mentioned earlier. 1. Non-Authentic Interpretation: an Act of Cognition Non-authentic interpretation, through an act of cognition, implies the determination of all the possible meanings, the establishment of all the different possible significations within the framework of the norm. As long as a particular meaning comes within the given framework, it is correct, possible. Thus, for example, where there is linguistic indeterminacy in a text, a judicial decision applying any one of the possible linguistic meanings would be correct: That a judicial decision is based on a statute means in truth simply that the decision stays within the frame the statute represents, means simply that the decision is one of the individual norms possible within the frame of the general norm, not that it is the only norm possible.4
Thus, after the ‘cognitive’ stage, i.e. after the non-authentic interpretation, another step must be taken in order to allow a choice between the possible meanings exposed by the non-authentic interpretation.
3 Hans Kelsen, Introduction to the Problems of Legal Theory (Pure Theory of Law) (Oxford: Clarendon Press, 1992), 81. 4 Id., at 80.
interpretation in international trade law
115
2. Authentic Interpretation: an Act of Will Authentic interpretation, i.e. interpretation made by a body authorised to apply the law, is not the result of a purely intellectual action based on legal theory. Rather, it is an act of will, belonging to the sphere of legal policy. Kelsen thus denounced an illusion resulting from the belief that when an entity authorised to apply the law interprets a norm it conducts a series of neutral, abstract operations, based on the application of certain logical methods, or means of cognition, in order to determine the correct Interpretation, with a capital I. A choice must be made, of course, but it is not a legally compelling choice and all the possible interpretations are equally valid. Kelsen thus shows clearly that through such action the judges are led to make choices, judges are creators of law: When a legal body applies the law, the interpretation of the applicable law through a cognitive act combines with an act of will through which the body applying the law makes a choice between the possibilities revealed by cognitive interpretation.5
It must be noted that, in some cases, those who propose non-authentic interpretations do not present all of the possibilities, choosing amongst them, for example with a view to influencing the bodies applying the law. One thinks of lawyers and doctrinal writers. Yet whenever legal theoreticians present their interpretation as being the only exact meaning, rather than the preferable meaning for non-legal reasons (justice, social equilibrium, etc.), they create an illusion. According to Kelsen, “to act in this way is falsely to present as scientific truth that which is simply a political value judgment”.6 On the contrary, the jurist’s role is to point out all of the possible interpretations of unclear norms, in order to force the authors of the norms to make them clearer; that is the only way to increase legal security. As a concluding remark, it can be mentioned that naturally if the judge can create law in a single concrete case, he also plays a creative role in a more general way, through the setting of precedents. Of course, in international law, one does not think of binding precedents, but it cannot be denied that if the ICJ or the Appellate Body of WTO says something, it has an important value, even if not binding stricto sensu for future cases. II. A Concrete Approach: the Shrimp/Turtle Case It is important to point out at the outset that the rules of interpretation laid down in the Vienna Convention on the Law of Treaties are the indispensable
5 6
Id., at 460. Id., at 463.
116
brigitte stern
benchmark and are cited in this case by the Appellant, the Appellees and the Panel, as well as by the Appellate Body. We know that Article 3:2 of the Dispute Settlement Understanding (DSU) refers to the ‘customary rules of interpretation of public international law’.7 The Appellate Body, like the Panel before it, found these rules in the provisions of Article 31 of the Vienna Convention. In fact, the Appellate Body gave a veritable ‘master class’ on the rules of interpretation, reproaching the Panel with not having followed all of the steps required by the applicable rules of interpretation of public international law. It is best to quote this passage in extenso: As we have emphasised numerous times, these rules call for an examination of the ordinary meaning of the words of a treaty, read in their context, and in the light of the object and purpose of the treaty involved. A treaty interpreter must begin with, and focus upon, the text of the particular provision to be interpreted. It is in the words constituting that provision, read in their context, that the object and purpose of the states parties to the treaty must first be sought. Where the meaning imparted by the text itself is equivocal or inconclusive, or where confirmation of the correctness of the reading of the text itself is desired, light from the object and purpose of the treaty as a whole may usefully be sought.8
It is interesting to note the distinction – rarely made, it seems – between the object and purpose derived from a provision of the treaty and the object and purpose of the treaty as a whole. This case poses once again, in acute form, the problem of reconciling trade liberalisation with national environmental protection policies. It is well known that the United States adopted legislation imposing an import ban on shrimps from countries that harvest them using methods which could harm certain species of marine turtle and that the implementation of this legislation was challenged by a number of shrimp-exporting countries in Asia on the grounds that it infringed GATT/WTO rules. A. The Limits on the Interpretative Competence of the Panel in the Interpretation of Article XX A number of countries – India, Malaysia, Pakistan and Thailand – requested the establishment of a panel to examine the United States measures, which they claimed were inconsistent with GATT/WTO rules, in particular Articles I, XI, XIII, and could not be justified under the Article XX exceptions. As for the United States, it considered that the measures it had adopted were justified by the Article XX’s exceptions. The Panel’s Report was circulated on 15 May 1998.
7 US – Import Prohibition of Certain Shrimp and Shrimp Products (12 October 1998), Appellate Body Report, WT/DS58/AB/R, para. 114. 8 Id.
interpretation in international trade law
117
1. The Panel’s Interpretation Having found that there had indeed been a violation of article XI:1, the Panel held that there was no need for it to examine the allegations concerning articles I:1 and XIII:1, but it considered that it was required to verify whether this was a case where there existed an exception authorising non-compliance with a GATT/WTO rule, as the United States claimed. The Panel’s reasoning was broadly as follows. It interpreted the introductory text of article XX, taken into consideration before any further examination, as authorising only measures which did nothing to jeopardise the multilateral trading system; indeed, according to the Panel, any type of measure which, if generalised, would threaten the multilateral trading system is prohibited, even though in itself it does not constitute such a threat: … we are of the opinion that the introductory text of article XX, interpreted within its context and in the light of the object and purpose of GATT and of the WTO Agreement, only allows Members to derogate from GATT provisions so long as, in doing so, they do not undermine the WTO multilateral trading system … We are of the view that a type of measure adopted by a Member which, on its own, may appear to have a relatively minor impact on the multilateral trading system, may nonetheless raise a serious threat to that system if similar measures are adopted by the same or other Members … In our view, if an interpretation of the introductory text of article XX were to be followed which would allow a Member to adopt measures conditioning access to its market for a given product upon the adoption by the exporting Members of certain policies, including conservation policies, GATT 1994 and the WTO Agreement could no longer serve as a multilateral framework for trade among Members as security and predictability of trade relations under those agreements would be threatened … this would rapidly lead to the end of the WTO multilateral trading system.9
The Panel drew therefore the following conclusion: “In the light of the findings above, we conclude that the import ban on shrimp and shrimp products as applied by the United States on the basis of Section 609 of Public Law 101–162 is not consistent with article XI:1 of GATT 1994 and cannot be justified under article XX of GATT 1994.”10
Thus, the Panel found against the United States measures on two concurrent grounds: – on the one hand, because it constituted a quantitative restriction on trade prohibited by Article XI:1 of the GATT 1994.
9 US – Import Prohibition of Certain Shrimp and Shrimp Products (15 May 1998), Panel Report, WT/DS58/R, paras. 7.44 and 7.45; cited in US – Import Prohibition of Certain Shrimp and Shrimp Products, Appellate Body Report, supra note 7, para. 112. 10 US – Import Prohibition of Certain Shrimp and Shrimp Products, Panel Report, supra note 9, para. 8.1.
118
brigitte stern
– on the other hand, because it did not meet the requirements for falling within the definition of the trade-restrictive measures authorised in exceptional cases by Article XX, inasmuch as it constituted “unjustifiable discrimination between countries where the same conditions prevail”, unjustifiable in that it posed a threat to the multilateral trading system. 2. The ‘Lesson in Interpretation’ from the Appellate Body The Appellate Body did not agree at all with the approach followed by the Panel. As far as the interpretation of Article XX was concerned, the Appellate Body gave what could be called ‘a lesson in interpretation’ to the Panel, in insisting on the logical sequence in the analysis of Article XX. According to the Appellate Body, the Panel did not examine the ordinary meaning of the words of Article XX. This article deals with the manner in which the measure is applied and the design of the measure itself, which are addressed by the introductory text of Article XX and its various paragraphs, respectively, to be examined separately. The Panel examined only the general design of the measure in the light of the introductory text of Article XX, instead of examining, as it should have done, first the general design of the measure with reference to Article XX(g) and then its application with reference to the introductory text. Thus, the Panel misinterpreted Article XX, on the one hand, because it failed to examine what the Appellate Body calls “the immediate context of the chapeau”,11 that is to say, paragraphs (a) to (j), and, on the other, because it failed to look into the object and purpose of the introductory text of Article XX. The Appellate Body does not deny that maintaining the multilateral trading system is a fundamental premise underlying the WTO Agreement. At the same time, however, according to the Appellate Body, it is not a right or an obligation, nor is it an interpretative rule. The Panel’s misinterpretation was a result of its not following the essential procedural steps for the analysis of Article XX. Thus, it is necessary first to examine the acceptability of the measure criticised with respect to the various paragraphs of Article XX, before verifying whether or not it is being applied in a manner such as to constitute abuse. The Appellate Body therefore held that by examining the introductory text first and constructing a new a priori test, the Panel had made an error in legal interpretation that would render any domestic measure that derogated from GATT obligations incapable of justification, even if it implemented an important and legitimate policy covered by one of the paragraphs of Article XX. Clearly, the purpose of this article is to authorise certain measures which would not be authorised in its absence, precisely because they threaten the multilateral trading system. 11
Id., para. 116. Original emphasis.
interpretation in international trade law
119
The Appellate Body thus insisted that in its reasoning the Panel should have proceeded in two successive non-interchangeable steps: analysis of the measure itself in light of the paragraphs of Article XX; analysis of the non-abusive application of the measure in light of the introductory text of Article XX. B. The Extensive Powers of Interpretation Used by the Appellate Body 1. An Evolutionary Interpretation12 of ‘Exhaustible Natural Resources’ To use an evolutionary interpretation means that the interpretation has to be based on contemporary concepts and not on those prevalent at the time the provision was drafted. Such an interpretation was utilised by the Appellate Body for the interpretation of both ‘natural resources’ and ‘exhaustible resources’. When it included living organisms among natural resources, the Appellate Body observed that it is too late in the day to suppose that Article XX(g) of the GATT 1994 may be read as referring only to the conservation of exhaustible mineral or other nonliving natural resources.13
Similarly, the Appellate Body adopted an evolutionary interpretation of the scope of the term ‘exhaustible natural resources’: The words of Article XX(g), ‘exhaustible natural resources’, were actually crafted more than 50 years ago. They must be read by a treaty interpreter in the light of contemporary concerns of the community of nations about the protection and conservation of the environment.14
According to the Appellate Body, “the generic term ‘natural resources’ in Article XX(g) is not ‘static’ in its content or reference but is rather ‘by definition, evolutionary’ ”.15 In order to give an idea of these contemporary concerns, the Appellate Body went on to make a long inventory of all the multilateral treaties on the protection of the environment. This approach is an application of the strong statement made by the Appellate Body in its first Report on Gasoline, to the effect that WTO law should not be considered and applied ‘in clinical isolation of international law’.
12 On this notion of evolutive interpretation, see Brigitte Bollecker, ‘L’ Avis Consultatif du 21 Juin 1971 dans l’Affaire de la Namibie (Sud-Ouest Africain)’, Annuaire Francais de Droit International 17 (1971): 271, at 288–94. 13 US – Import Prohibition of Certain Shrimp and Shrimp Products, Appellate Body Report, supra note 7, para. 131. Emphasis added. 14 Id., para. 129. 15 Id., para. 130.
120
brigitte stern
This notion of concepts that are, by definition, evolutionary was first introduced, as we know, by the International Court of Justice (ICJ) in the Namibia case,16 decided in 1971. In so doing, it departed from the traditional doctrine still being espoused by the ICJ in 1966 when it held that in order to determine the rights and duties, past and present, of the parties under the Mandate … the Court must place itself in the period when the Mandate system was instituted.17
The question thus to be raised was whether Section 609 was concerned with the conservation of ‘exhaustible natural resources’. The complainants claimed that living resources, being renewable, were not exhaustible and that this term could therefore only be applied to mineral or non-living resources. However, the Appellate Body rejected this narrow interpretation, since modern biological sciences teach us that some species are susceptible of extinction.18 As sustainable development is one of the objectives of the WTO Agreement, it follows that the concept ‘exhaustible natural resources’ is, by definition, evolutionary, that is to say must take into account the resources threatened with exhaustion at the time of the interpretation. From there it is easy to proceed to the conclusion that sea turtles constitute exhaustible natural resources since they are listed in Appendix I of the CITES19 Convention.20 2. A Teleological Interpretation of the Admission of Amicus Curiae Briefs In Shrimp-Turtle, two environmental NGOs21 made submissions at the panel stage. A little later, the Panel received another submission from a third environmental NGO.22 At the Appellate Body stage, there were more environmental NGOs wanting to participate: the United States appended to its Appellant’s
16 Legal Consequences for States of the continued presence of South Africa in Namibia (South West Africa), Advisory Opinion of 21 June 1971, 1971 ICJ Rep. 16, at 31. Opinion cited in US – Import Prohibition of Certain Shrimp and Shrimp Products, Appellate Body Report, supra note 7, fn 109. 17 South West African cases (Ethiopia v. South Africa and Liberia v. South Africa), Judgments of 18 July 1966, 1966 ICJ Rep. 6, at 23. See also Rights of Nationals of the United States of America in Morocco (France v. United States of America), Judgment of 27 August 1952, 1952 ICJ Rep. 167, at 189, from which it follows that in order to establish the meaning of a legal concept in an historical context, the way in which that concept was originally understood in the context must be taken into account. 18 As early as 1982, a GATT panel, in United States – Prohibition of Imports of Tuna and Tuna Products from Canada (22 February 1982), Panel Report L/5198 – 29S/91, had endorsed the agreement between the two parties to treat tuna as an exhaustible natural resource. 19 1973 Convention on the International Trade in Endangered Species of Fauna and Flora (CITES), UNTS 993: 243. 20 Emphasis added. 21 The Center for Marine Conservation (CMC) and the Center for International Environmental Law (CIEL). 22 The World Wildlife Fund (WWF).
interpretation in international trade law
121
submission, three ‘independent briefs’23 submitted by three groups of NGOs,24 to which all the Appellees objected, both before the Panel and before the Appellate Body. The Panel refused to accept the amicus curiae briefs submitted by NGOs, considering that it was not competent to do so. More precisely, the Panel had declared that accepting non-requested information from non-governmental sources would be “incompatible with the provisions of the DSU as currently applied”.25 The question was again raised at the Appellate Body level. To the various arguments already put forward against the submission of such exhibits to the Panel, there was now added one more, namely that taking them into consideration would be inconsistent with the very essence of the appeal procedure which, according to Article 17.6 of the DSU, is limited to “issues of law covered in the panel report”. The NGO documents were said to contain a number of factual assertions which, according to the Appellees, made them inadmissible in the appeal stage. Therefore the Appellate Body had to decide whether amicus curiae briefs should be accepted before the Panel and before the Appellate Body. (a) The Treatment of the NGOs Submissions before the Panel With regard to the Panel’s treatment of the NGOs briefs, the Appellate Body was to criticise the fundamental legal position taken by the Panel, while conceding that it had the authority to treat the NGOs documents in the way it actually did. The Appellate Body rejected the Panel’s textual analysis: authority to seek information is not properly equated with a prohibition on accepting information which has been submitted without having been requested by the panel. A panel has the discretionary authority either to accept and consider or to reject information and advice submitted to it, whether requested by a panel or not. The fact that a panel may motu proprio have initiated the request for 23
In respect of these exhibits, the United States stated the following: Encouraging the use of TEDs in order to promote sea turtle conservation is a matter of great importance to a number of nongovernmental environmental organisations. Three groups of these organisations – each with specialised expertise in conservation of sea turtles and other endangered species – have prepared submissions reflecting their respective independent views with respect to the use of TEDs and other issues. The United States is submitting these materials to the Appellate Body for its information attached hereto as US Appellant Exhibits 1–3.
United States Appellant’s submission, para. 2, footnote 1. 24 1. Earth Island Institute; Humane Society of the United States; and Sierra Club; 2. Center for International Environmental Law (CIEL); Centre for Marine Conservation; Environmental Foundation Ltd; Mangrove Action Project; Philippine Ecological Network; Red Nacional de Accion Ecologica; and Sobrevivencia; and 3. Worldwide Fund for Nature and Foundation for International Environmental Law and Development. 25 US – Import Prohibition of Certain Shrimp and Shrimp Products, Panel Report, supra note 9, para. 7.8.; cited in US – Import Prohibition of Certain Shrimp and Shrimp Products, Appellate Body Report, supra note 7, para. 100.
122
brigitte stern information does not, by itself, bind the panel to accept and consider the information which is actually submitted.26
In particular, the Appellate Body found the Panel’s reading of its ‘right to seek information’ under Article 13 of the DSU too literal or, in its own words, “unnecessarily formal and technical”.27 The Appellate Body considered that the words “seek … from any individual or body which it deems appropriate” implied the right of the Panel to choose the source of information and decide whether or not to request information, as well as the authority to accept a requested document, to reject a requested document, to accept a nonrequested document, to reject a non-requested document, to dispose of a requested document as it thinks fit and to dispose of a non-requested document as it thinks fit. In so far as the Panel allowed the United States to attach the NGOs briefs to its own submission, the Appellate Body considered that this was a legitimate means that the Panel could use to obtain certain information from NGOs. The conclusion is therefore clear: We find, and so hold, that the Panel erred in its legal interpretation that accepting non-requested information from non-governmental sources is incompatible with the provisions of the DSU. At the same time, we consider that the Panel acted within the scope of its authority under Articles 12 and 13 of the DSU in allowing any party to the dispute to attach the briefs by non-govermental organisations, or any portion thereof, to its own submissions.28
On what legal basis rested the Panel’s authority, according to the Appellate Body? First of all, with regard to the authority granted to panels under Article 13 of the DSU, there can be no denying that it relies upon an extremely flexible interpretation of the act of ‘seeking’. To convince oneself of this, it is sufficient to compare the wording of Article 13 with, for example, that of Article 61.3 of the Rules of the European Court of Human Rights: Article 61.3: The President of the Chamber may, in the interests of the proper administration of justice, invite or grant leave to any Contracting State which is not a party to the proceedings, or any person concerned who is not the applicant, to submit written comments or, in exceptional cases, to take part in a hearing. Requests for leave for this purpose must be duly reasoned …29
It is therefore debatable whether the verb ‘to seek’ can be interpreted by turning what might be described as an ‘active seeking’ into a ‘passive seeking’, as the Appellate Body did, without assigning oneself a judicial law-forming role, which explains why there was no shortage of government criticism. 26 US – Import Prohibition of Certain Shrimp and Shrimp Products, Appellate Body Report, supra note 7, para. 108. Original emphasis. 27 Id., para. 107. 28 Id., para. 110. 29 <www.echr.int/Fr/fdoc/rulesofcourt.htlm> (accessed on 1 May 2009). Emphasis added.
interpretation in international trade law
123
(b) The Treatment of the NGOs Submissions before the Appellate Body In a ruling on this preliminary procedural matter issued on 11 August 1998, the Appellate Body accepted for consideration the information provided by the NGOs. In its Report, it only partially explained how this information would be taken into account. In fact, the Appellate Body had already questioned the United States about the exact linkage between its submission and those of the NGOs. On 13 August 1998, the United States replied as follows: The three submissions prepared by non-governmental organisations reflect the independent views of those organisations … The United States agrees with the legal arguments in the submissions of the non-governmental organisations to the extent those arguments concur with the U.S. arguments set out in our main submission.30
The Appellate Body then admitted the NGO briefs (i) in so far as they formed an integral part of the Appellant’s submission, that is to say in so far as the Appellant took them into account, and (ii) in so far as they concurred with the arguments in the main submission. On what legal basis then rested the authority of the Appellate Body? It must be conceded that, with regard to the authority of the Appellate Body, no clear answer was given in the Shrimp/Turtle case. Later, in United States-Carbon Steel, it was assumed by the Appellate Body to rest on the basis of Article 17 of the DSU, and perhaps subsidiarily on Article 16.1 of the Working Procedures for Appellate Review, but such legal interpretation is far from watertight. The recourse to Article 17.9 of the DSU in United States-Carbon Steel merely indicates that the Appellate Body has wide discretion to adopt procedural rules, but the use of Article 17.9 depends on the observance of a consultation and notification procedure, which quite clearly was not followed in this particular case. Nor does it seem that Article 16.1 of the Working Procedures for Appellate Review, if it had been formally invoked, would have been able to provide a firm foundation for a new general procedural rule such as the Appellate Body claimed to have laid in United States-Carbon Steel. In fact, this article is a sort of safeguard clause that enables the Appellate Body to give a ruling in unprecedented procedural situations; however, it is explicitly intended for use in specific and exceptional circumstances and not for laying down general procedural rules, since it stipulates that “a division may adopt an appropriate procedure for the purposes of that appeal only”. Indeed, the Appellate Body acknowledged this in the Asbestos case, by establishing procedural rules for the admission of amicus curiae briefs applicable to that case only.
30 US – Import Prohibition of Certain Shrimp and Shrimp Products, Appellate Body Report, supra note 7, para. 86.
124
brigitte stern
There can be no denying that any jurisdictional body has implicit powers that enable it to organise the proceedings taking place before it in the best possible way. But it is one thing to say that such procedural powers exist and quite another completely to upset the balance of an interstate dispute settlement system by introducing a procedural innovation with far reaching consequences, which, it seems to me, is what happened in this case. (c) The Political Implications It is the entire philosophy of international trade dispute settlement that is at stake. A comparative analysis shows that amicus curiae briefs are not admitted in purely intergovernmental proceedings, whereas they are accepted in cases in which private persons are exclusively or partially involved. This broad acceptance of amicus curiae briefs is opening up access to the WTO dispute settlement system, which was intended to be intergovernmental, to private actors. As Hervé Ascencio has said, “the amicus curiae is the procedural breach in the dike through which individuals, companies and associations can pour when party status is reserved for states.”31 This throwing open of the WTO dispute settlement mechanism initiated by the rulings of the Appellate Body confirms the ‘retournement du monde’,32 about which there is so much talk. Whatever its advantages and disadvantages, it would seem difficult for the WTO to go into reverse on this question of the admission of amicus curiae briefs, particularly as it is obvious that the position taken by the Appellate Body is tending to spill over into other international disputes.33 Whatever the legal criticisms that might be leveled against the Appellate Body’s decision, it has been adopted by Members within the context of the Dispute Settlement Body (DSB). According to Arthur Appleton, the ruling in Shrimp-Turtle shows “the Appellate Body’s determination to maintain a degree of public support for the WTO among what is frequently termed ‘civil society’ ”.34 31 Hervé Ascensio, ‘L’ Amicus Curiae devant les Juridictions Internationales’, RGDIP 105 (2001): 897, at 900. 32 See in particular, Bertrand Badie and Marie-Claude Smouts, Le Retournement du Monde. Sociologie de la Scène Internationale (Paris: Presses de la Fondation Nationale de Sciences Politiques, 1995). 33 See in particular, the NAFTA arbitration in an international investment case involving a government and an investor, In the Matter of Arbitration under Chapter 11 of the NAFTA and the UNCITRAL Rules, between Methanex Corporation and the United States of America, Decision of Tribunal on the Petition from Third Persons to intervene Amici Curiae of 15 January 2001, <www.naftalaw.org/methanex/Methanex-Amicus.Decision.pdf> (accessed on 1 May 2009). See also another NAFTA arbitration along the same lines, United Parcel Service of America Inc. v. Canada, Decision of 17 October 2001. 34 Arthur Appleton, ‘Shrimp/Turtle: Untangling the Nets’, Journal of International Economic Law 2 (1999): 477, at 494.
interpretation in international trade law
125
The same author has drawn attention to the ‘perverse’ effect of reverse consensus, which results in any ruling of the Appellate Body being necessarily endorsed by Members, even if there is a large majority which considers that ruling to be legally unsound. Thus, although automaticity prevents losing parties from blocking the adoption of appellate Reports, it also severely restricts the possibility that Members can limit the effects of the Appellate Body rule making. Because of automaticity, amicus submissions are now here to stay.35
In reality, from a political point of view, considering the strength of international feeling and the WTO’s desire to establish its credibility in the eyes of civil society, it would seem difficult directly to oppose the movement initiated by the Appellate Body. The author just quoted has also drawn attention to the ‘political’ relevance of the Appellate Body’s decision, not so much from the standpoint of the Members of the WTO, many of whom, to say the least, have reservations about it, but from the standpoint of the WTO itself, as an institution with ambitions to be at the centre of the international trading and, indeed, economic system: From a political perspective this decision satisfies a public relations objective— reducing the distrust of the WTO which exists in the environmental community. This task is accomplished without seriously compromising the trade rights guaranteed under the WTO Agreement.36
It is beyond doubt that if in the future, by a decision of the Members of the WTO, NGOs were to be refused access to the dispute settlement system via amicus curiae briefs – which would mean a setback for civil society relative to the present situation – the credibility of the system would be seriously impaired; on the other hand, continuing to accept such briefs would reinforce, within that civil society, the feeling that economic justice is being done. 3. A Politically Balanced Interpretation of the Interaction between Trade and Environment One of the main contributions of this case seems to be the idea that environmental protection is a real consideration within the context of the WTO Agreement. As one author has pointed out, “[i]n each of the Appellate Body’s three reversals of panel findings, it appeared to respond to charges that the
35 Arthur Appleton, ‘Amicus Curiae Submissions in the Carbon Steel Case: Another Rabbit from the Appellate Body’s Hat?’, Journal of International Economic Law 3 (2000): 691, at 699. 36 Appleton, supra note 34, at 477. See also: Amicus briefs from nongovernmental organisations could further enhance the legitimacy, and acceptance, of the WTO dispute settlement process; Thomas Cottier, ‘The WTO and Environmental Law: Three Points for Discussion’, in Trade and Environment: Bridging the Gap, Agatha Fijolkowski and James Cameron (eds.) (London: Cameron May, 1998), 56, at 59.
126
brigitte stern
WTO’s dispute settlement process is trade-biased.”37 After finding that the US measure failed to meet WTO requirements, the two Reports took good care not to leave an anti-environmental impression. However, there is a clear difference in outlook between the Panel’s approach and that of the Appellate Body. The Panel seems to have kept environmental objectives subordinate to the liberalisation of trade: While the WTO Preamble confirms that environmental considerations are important for the interpretation of the WTO Agreement, the central focus of that agreement remains the promotion of economic development through trade; and the provisions of GATT are essentially turned toward liberalisation of access to markets on a non-discriminatory basis.38
However, in a slight softening of its position it also noted that our findings regarding article XX do not imply that recourse to unilateral measures is always excluded, particularly after serious attempts have been made to negotiate; nor do they imply that, in any given case, they would be permitted.39
The Appellate Body seems to have endeavoured to bring the trade-off between liberalisation and protection of the environment more neatly into balance and warned against an interpretation whose net effect was unfavourable to the environment. To prevent any such construction, the Appellate Body gave its own interpretation, making quite clear what could not be said about the Report adopted: In reaching these conclusions, we wish to underscore what we have not decided in this appeal. We have not decided that the protection and preservation of the environment is of no significance to the Members of the WTO. Clearly, it is. We have not decided that the sovereign nations that are Members of the WTO cannot adopt effective measures to protect endangered species, such as sea turtles. Clearly, they can and should. And we have not decided that sovereign states should not act together bilaterally, plurilaterally or multilaterally, either within the WTO or in other international fora, to protect endangered species or to otherwise protect the environment. Clearly, they should and do.40
The Appellate Body sought to define what it called “the relationship between trade and the environment”.41 It considered this relationship to be indisputable and in need of no further proof than the intentions expressed by the
37 Gregory Schaffer, ‘United States – Import Prohibition of Certain Shrimp Products’, AJIL 93 (1992): 508. 38 US – Import Prohibition of Certain Shrimp and Shrimp Products, Panel Report, supra note 9, para. 7.42. Emphasis added. 39 Id., para. 7.61. 40 US – Import Prohibition of Certain Shrimp and Shrimp Products, Appellate Body Report, supra note 7, para. 185. 41 Id., para. 154.
interpretation in international trade law
127
Ministers meeting at Marrakesh when they took the important decision to establish a Committee on Trade and Environment (CTE): Considering that there should not be, nor need be, any policy contradiction between upholding and safeguarding an open, non-discriminatory and equitable multilateral trading system on the one hand, and acting for the protection of the environment, and the promotion of sustainable development on the other …42
The Article XX exceptions should not be interpreted narrowly but rather in such a way as to strike a balance between trade concerns and the other concerns taken into account in the various paragraphs of Article XX. It is important to note that the Appellate Body sees its task as being the comprehensive regulation of world trade, with all the parameters – and not just the strict rules of trade – being taken into account. This rejection of a narrow interpretation of the exceptions is a departure from the traditional GATT approach as described by the Panel in the Tuna I case: … previous panels had established that Article XX is a limited and conditional exception from obligations under other provisions of the General Agreement, and not a positive rule establishing obligations in itself. Therefore, the practice of panels has been to interpret Article XX narrowly.43
Although the Appellate Body still regards Article XX(g) as a limited and conditional exception to the basic obligations of the GATT, it is now, under the rules of the WTO, seeking to establish a new balance: Turning then to the introductory text of Article XX, we consider that it embodies the recognition on the part of WTO Members of the need to maintain a balance of rights and obligations between the right of a Member to invoke one or another of the exceptions of Article XX, specified in paragraphs (a) to (j), on the one hand, and the substantive rights of the other Members under the GATT 1994, on the other hand.44
Although the choice of the words rights/substantive rights may suggest the maintenance of a certain hierarchy of objectives, we are far from the narrow interpretation of the exceptions proposed by the GATT panels, which has been under attack since the Appellate Body’s very first ruling. The
42 Preamble to the Decision on trade and the environment, cited in US – Import Prohibition of Certain Shrimp and Shrimp Products, Appellate Body Report, supra note 7, para. 154. 43 US – Restrictions on Imports of Tuna (16 August 1991), Panel Report, ILM 30 (1991): 1594, at 1619, para. 5.22. 44 US – Import Prohibition of Certain Shrimp and Shrimp Products, Appellate Body Report, supra note 7, para. 156:
Exercise by one Member of its right to invoke an exception, such as Article XX(g), if abused or misused, will, to that extent, erode or render naught the substantive treaty rights in, for example, Article XI:1, of other Members.
For another formulation of the same idea, see para. 158.
128
brigitte stern
environmental exception should not be interpreted narrowly as permitting derogation from the principles that govern trade but as an objective to be achieved in the same way as the liberalisation of trade itself. According to Pierre Monnier, “the Appellate Body has displayed great sensitivity to the problems of the environment”.45 Undoubtedly, that is the main contribution made by this decision. This was true before the Article 21.5 Report, and is even more true after, as this Report considered that, thanks to a few aesthetic modifications, the American regulation did conform with Article XX. Environment then seems to have won. But at the same time, the Report accepted a very broad power of regulation of world activities by States, that can be used by any State that has the power to impose its views. This has been called in the context of the protection of the environment ‘green imperialism’ by Professor Petersman. So the open question is whether this green imperialism should be rejected because it is ‘imperialism’ and therefore disregards the sovereignty of all the States but the one imposing its power, or should it be welcomed because it is ‘green’, in the name of the welcomed goal of the protection of the environment? I leave this question of political interpretation open …
45 ‘L’ environnement dans la jurisprudence de 1’OMC’, Les notes Bleues de Bercy 186 (1–15 July 2000): 2.
DIVERSITY AND HARMONIZATION OF TREATY INTERPRETATION IN INVESTMENT ARBITRATION* Christoph Schreuer Investment arbitration typically involves a variety of treaties. Most frequently it is based on bilateral investment treaties (BITs). Alternatively, multilateral regional treaties such as the North American Free Trade Agreement (NAFTA) or the Energy Charter Treaty (ECT) are the basis of consent to arbitration. In addition, multilateral treaties such as the Convention on the Settlement of Investment Disputes between States and the Nationals of Other States (ICSID Convention) are frequently interpreted and applied. Investment arbitration takes place before ad hoc tribunals. Their composition varies from case to case. This makes it considerably more difficult to develop a consistent case law than in a permanent judicial institution such as the International Court of Justice (ICJ), The European Court of Human Rights (ECtHR) or The Court of Justice of the European Communities. This article looks at the methodology of tribunals in interpreting treaties, at their willingness to follow previous decisions on like questions and at possible methods to achieve consistency and harmonization.1 I. The Methods of Treaty Interpretation Adopted by Tribunals A. Interpretation in Accordance with the Vienna Convention on the Law of Treaties Tribunals almost invariably start by invoking Article 31 of the Vienna Convention on the Law of Treaties (VCLT) when interpreting treaties. There
* This Chapter was based on the presentation made by the author during the ‘First Annual Conference on International Law: Interpretation under the VCLT’ organized by Queen Mary and Eversheds in 2006, and is current as of that date. 1 Most of the decisions cited in this article may be found at one of the following websites: or www.investmentclaims.com/ (accessed on 1 May 2009).
130
christoph schreuer
are numerous decisions to this effect.2 For instance the Tribunal in Siemens v. Argentina3 said: 80. Both parties have based their arguments on the interpretation of the Treaty in accordance with Article 31(1) of the Vienna Convention. This Article provides that a treaty be “interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose.” The Tribunal will adhere to these rules of interpretation in considering the disputed provisions of the Treaty. …4
In referring to the rules of interpretation contained in the VCLT, tribunals sometimes point out that these rules reflect customary international law.5 Thus the Tribunal in Tokios Tokeles v. Ukraine6 said: 27. As have other tribunals, we interpret the ICSID Convention and the Treaty between the Contracting Parties according to the rules set forth in the Vienna Convention on the Law of Treaties, much of which reflects customary international law. Article 31 of the Vienna Convention provides that “[a] treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in light of its object and purpose.7
At times, tribunals will also refer to the supplementary means of interpretation contained in Article 32 of the VCLT. In the words of the Tribunal in Noble Ventures v. Romania:8 50. … reference has to be made to Arts. 31 et seq. of the Vienna Convention on the Law of Treaties which reflect the customary international law concerning treaty interpretation. Accordingly, treaties have to be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of the object and purpose of the Treaty, while 2 See also AAPL v. Sri Lanka, Award, 27 June 1990, paras. 38–42; MTD v. Chile, Award, 25 May 2004, para. 112; Enron v. Argentina, Decision on Jurisdiction (Ancillary Claim), 2 August 2004, para. 32; Salini v. Jordan, Decision on Jurisdiction, 29 November 2004, para. 75; Plama v. Bulgaria, Decision on Jurisdiction, 8 February 2005, paras. 117, 147–165; Sempra Energy Intl. v. Argentina, Decision on Jurisdiction, 11 May 2005, para. 141; Camuzzi v. Argentina, Decision on Jurisdiction, 11 May 2005, para. 133; Methanex v. United States, Award, 3 August 2005, Part II, Chapter B, paras. 15–23, Part IV, Chapter B, para. 29; Eureko v. Poland, Partial Award, 19 August 2005, para. 247; Aguas del Tunari, S.A. v. Bolivia, Decision on Jurisdiction, 21 October 2005, paras. 88–93, 226, 230, 239; National Grid v. Argentine Republic, Decision on Jurisdiction, 20 June 2006, paras. 51, 80; Canfor v. United States, Tembec v. United States, Terminal Forest Products v. United States, Order of the Consolidation Tribunal, 7 September 2005, paras. 59, 86, 95, 113; Bogdanov v. Moldova, Award, 22 September 2005, section 4.2.4; Saluka v. Czech Republic, Partial Award, 17 March 2006, para. 296 et seq.; Suez v. Argentina, Decision on Jurisdiction, 16 May 2006, ICSID Case No. ARB/03/17, para. 54; Azurix v. Argentina, Award, 14 July 2006, paras. 307, 360, 391. 3 Siemens v. Argentina, Decision on Jurisdiction, 3 August 2004. 4 para. 80. 5 National Grid v. Argentine Republic, Decision on Jurisdiction, 20 June 2006, para. 51; Canfor v. United States, Tembec v. United States, Terminal Forest Products v. United States, Order of the Consolidation Tribunal, 7 September 2005, para. 59. 6 Tokios Tokeles v. Ukraine, Decision on Jurisdiction, 29 April 2004. 7 Para. 27, footnotes omitted. See also paras. 46, 75 and 85. 8 Noble Ventures v. Romania, Award, 12 October 2005.
diversity and harmonization of treaty interpretation
131
recourse may be had to supplementary means of interpretation, including the preparatory work and the circumstances of its conclusion, only in order to confirm the meaning resulting from the application of the aforementioned methods of interpretation. Reference should also be made to the principle of effectiveness (effet utile), which, too plays an important role in interpreting treaties.9
B. Interpretation in Accordance with the Treaty’s Object and Purpose Among the principles contained in Article 31 VCLT an interpretation that looks at the treaty’s object and purpose is particularly popular.10 In the context of BITs this often leads to an interpretation that is favourable to investors. For instance, the Tribunal in Noble Ventures v. Romania11 said: 52. The object and purpose rule also supports such an interpretation. While it is not permissible, as is too often done regarding BITs, to interpret clauses exclusively in favour of investors, here such an interpretation is justified. Considering, as pointed out above, that any other interpretation would deprive Art. II (2)(c) [an umbrella clause] of practical content, reference has necessarily to be made to the principle of effectiveness, also applied by other Tribunals in interpreting BIT provisions (see SGS v. Philippines, para. 116 and Salini v. Jordan, para. 95). …12
On the other hand the Tribunal in Amco v. Indonesia13 in interpreting the ICSID Convention pointed out that investment protection was also in the longer term interest of host States: “…to protect investments is to protect the general interest of development and of developing countries”.14 The most frequent way to find a treaty’s object and purpose was to look at the preamble.15 The Tribunal in Siemens v. Argentina16 said in this respect: 81. The Tribunal considers that the Treaty has to be interpreted neither liberally nor restrictively, as neither of these adverbs is part of Article 31(1) of the Vienna Convention. The Tribunal shall be guided by the purpose of the Treaty as expressed in its title and preamble. It is a treaty “to protect” and “to promote” investments. The preamble provides that the parties have agreed to the provisions
9
para. 50. See also Lauder v. Czech Republic, Award, 3 September 2001, para. 292; MTD v. Chile, Award, 25 May 2004, paras. 104, 105; CMS v. Argentina, Award, 12 May 2005, para. 274; Aguas del Tunari, S.A. v. Bolivia, Decision on Jurisdiction, 21 October 2005, paras. 153, 240–1, 247; Continental Casualty Co. v. Argentina, Decision on Jurisdiction, 22 February 2006, para. 80; Azurix v. Argentina, Award, 14 July 2006, para. 360. 11 Noble Ventures v. Romania, Award, 12 October 2005. 12 para. 52. 13 Amco v. Indonesia, Decision on Jurisdiction, 25 September 1983, ICSID Reports 1: 389. 14 At para. 23. See also Award, 20 November 1984, ICSID Reports 1: 413, at para. 249. 15 SGS v. Philippines, Decision on Jurisdiction, 29 January 2004, para. 116; Occidental Exploration and Production Co. v. Ecuador, Award, 1 July 2004, para. 183; CMS v. Argentina, Award, 12 May 2005, para. 274; Bogdanov v. Moldova, Award, 22 September 2005, section 4.2.4; Aguas del Tunari v. Bolivia, Decision on Jurisdiction, 21 October 2005, para. 241; Saluka v. Czech Republic, Partial Award, 17 March 2006, para. 299; Azurix v. Argentina, Award, 14 July 2006, para. 360. 16 Siemens v. Argentina, Decision on Jurisdiction, 3 August 2004. 10
132
christoph schreuer of the Treaty for the purpose of creating favorable conditions for the investments of nationals or companies of one of the two States in the territory of the other State. Both parties recognize that the promotion and protection of these investments by a treaty may stimulate private economic initiative and increase the wellbeing of the peoples of both countries. The intention of the parties is clear. It is to create favorable conditions for investments and to stimulate private initiative.17
One tribunal warned against over-extending the method of looking at object and purpose. The Tribunal in Plama v. Bulgaria18 said: 193. … the Tribunal is mindful of Sir Ian Sinclair’s warning of the “risk that the placing of undue emphasis on the ‘object and purpose’ of a treaty will encourage teleological methods of interpretation [which], in some of its more extreme forms, will even deny the relevance of the intentions of the parties.19
C. Restrictive or Effective Interpretation The discussion about the propriety of a restrictive or effective interpretation is by no means new. The idea that treaty provisions that constitute derogations from sovereignty call for a restrictive interpretation is usually rejected nowadays.20 Object and purpose oriented methods of interpretation typically lead to results that favour effectiveness. Investment tribunals have grappled with these conflicting theories in a number of cases. The correct interpretation of “umbrella clauses” is particularly prominent in these cases.21 The Tribunal in SGS v. Pakistan22 clearly favoured a restrictive approach. It said: 171. … The appropriate interpretive approach [to the umbrella clause] is the prudential one summed up in the literature as in dubio pars mitior est sequenda, or more tersely, in dubio mitius23
The Tribunal in Noble Ventures v. Romania24 seemed to proceed from a similar presumption: 55. Thus, an umbrella clause, when included in a bilateral investment treaty, introduces an exception to the general separation of States obligations under
17
para. 81. Footnote omitted. Plama v. Bulgaria, Decision on Jurisdiction, 8 February 2005, para. 193. 19 para. 193. Footnote omitted. 20 See Christoph Schreuer, ‘The Interpretation of Treaties by Domestic Courts’, BYIL 45 (1971): 255, at 283–301. 21 Umbrella clauses put undertakings by the host State, especially in investment contracts, under the treaties’ protective umbrella. These clauses while common in BITs may also be contained in other treaties for the protection of investments. The Energy Charter Treaty in Article 10(1), last sentence also contains an umbrella clause: “Each Contracting Party shall observe any obligations it has entered into with an Investor or an Investment of an Investor of any other Contracting Party”. Generally see: Anthony C. Sinclair, ‘The Origin of the Umbrella Clause in the International Law of Investment Protection’, Arbitration International 20 (2004): 411. 22 SGS v. Pakistan, Decision on Jurisdiction, 6 August 2003. 23 para. 171. 24 Noble Ventures v. Romania, Award, 12 October 2005. 18
diversity and harmonization of treaty interpretation
133
municipal and under international law. In consequence, as with any other exception to established general rules of law, the identification of a provision as an “umbrella clause” can as a consequence proceed only from a strict, if not indeed restrictive, interpretation of its terms and, more generally, in accordance with the well known customary rules codified under Article 31 of the Vienna Convention of the Law of Treaties (1969).25
This quote is curious in more than one respect. The Tribunal seemed to link its preference for a restrictive interpretation to the VCLT which, however, is silent on the point. Even more surprisingly, the Tribunal then proceeded to an application of the umbrella clause that gives it full effect.26 Other tribunals clearly rejected a restrictive interpretation.27 For instance, the Tribunal in Aguas del Tunari, S.A. v. Bolivia28 said: 91. … the Vienna Convention represents a move away from the canons of interpretation previously common in treaty interpretation and which erroneously persist in various international law decisions today. For example, the Vienna Convention does not mention the canon that treaties are to be construed narrowly, a canon that presumes States can not have intended to restrict their range of action.29
Some tribunals clearly embraced a method of interpretation that gives full effect to the provision in question. The Tribunal in SGS v. Philippines30 said: 116. The object and purpose of the BIT supports an effective interpretation of Article X(2) [umbrella clause]. The BIT is a treaty for the promotion and reciprocal protection of investments. According to the preamble it is intended “to create and maintain favourable conditions for investments by investors of one Contracting Party in the territory of the other”. It is legitimate to resolve uncertainties in its interpretation so as to favour the protection of covered investments.31
The Tribunal in Eureko v. Poland32 also subscribed to an effective interpretation: 248. … The context of Article 3.5 [the umbrella clause] is a Treaty whose object and purpose is “the encouragement and reciprocal protection of investment”, a treaty which contains specific provisions designed to accomplish that end, of which Article 3.5 is one. It is a cardinal rule of the interpretation of treaties that each and every operative clause of a treaty is to be interpreted as meaningful
25
para. 55. Paras. 46–62. See at para 61: “… the Parties had as their aim to equate contractual obligations governed by municipal law to international treaty obligations as established in the BIT”. 27 Loewen v. United States, Decision on Jurisdiction, 9 January 2001, para. 51; Methanex v. United States, Decision on Jurisdiction, 7 August 2002, paras. 103–5; Suez v. Argentina, Decision on Jurisdiction, 16 May 2006, para. 59, 64; Suez v. Argentina, AWG Group v. Argentina, Decision on Jurisdiction, 3 August 2006, paras. 60, 61, 66. 28 Aguas del Tunari, S.A. v. Bolivia, Decision on Jurisdiction, 21 October 2005. 29 para. 91. Footnote omitted. 30 SGS v. Philippines, Decision on Jurisdiction, 29 January 2004. 31 para. 116. 32 Eureko v. Poland, Partial Award, 19 August 2005. 26
134
christoph schreuer rather than meaningless. It is equally well established in the jurisprudence of international law, particularly that of the Permanent Court of International Justice and the International Court of Justice, that treaties, and hence their clauses, are to be interpreted so as to render them effective rather than ineffective.33
Some tribunals, probably wisely, distance themselves from rules purporting to prescribe a restrictive or an extensive interpretation.34 The Tribunal in Mondev v. United States35 said: 43. In the Tribunal’s view, there is no principle either of extensive or restrictive interpretation of jurisdictional provisions in treaties. In the end the question is what the relevant provisions mean, interpreted in accordance with the applicable rules of interpretation of treaties. These are set out in Articles 31–33 of the Vienna Convention on the Law of Treaties, which for this purpose can be taken to reflect the position under customary international law.36
D. Special Rules of Interpretation 1. Expressio Unius est Exclusio Alterius The proposition that the expression of one thing is the exclusion of another is not really a rule of law but a purported rule of logic. The limited value of this kind of logic is demonstrated by the conflicting ways in which this “rule” has been received by tribunals. The Tribunal in Waste Management v. Mexico37 seemed to embrace it when it said: 85. Where a treaty spells out in detail and with precision the requirements for maintaining a claim, there is no room for implying into the treaty additional requirements, whether based on alleged requirements of general international law in the field of diplomatic protection or otherwise.38
Similarly, the Tribunal in National Grid v. Argentina39 found that the noninclusion of dispute settlement in a list of exceptions to the application of an MFN clause was an indication that it should be seen as included. The Tribunal said: 82. The Tribunal observes that the MFN clause does not expressly refer to dispute resolution or for that matter to any other standard of treatment provided for specifically in the Treaty. On the other hand, dispute resolution is not included among the exceptions to the application of the clause. As a matter of 33
para. 248. El Paso v. Argentina, Decision on Jurisdiction, 27 April 2006, paras. 68–70; Pan American Energy v. Argentina, Decision on Preliminary Objections, 27 July 2006, paras. 97–99; Azurix v. Argentina, Award, 14 July 2006, para. 307. 35 Mondev v. United States of America, Award, 11 October 2002. 36 para. 43. Footnote omitted. 37 Waste Management v. United Mexican States, Award, 30 April 2004. 38 para. 85. 39 National Grid v. Argentine Republic, Decision on Jurisdiction, 20 June 2006. 34
diversity and harmonization of treaty interpretation
135
interpretation, specific mention of an item excludes others: expressio unius est exclusio alterius.40
By contrast, other tribunals seemed to reject this principle. The Tribunal in Enron v. Argentina41 said: 46. The fact that a treaty may have provided expressly for certain rights of shareholders does not mean that a treaty not so providing has meant to exclude such rights if this can be reasonably inferred from the provisions of such treaty. Each instrument must be interpreted autonomously in the light of its own context and in the light of its interconnections with international law.42
The expressio unius principle also did not find favour with the Tribunal in Siemens v. Argentina.43 The Tribunal said: 140. … If the Treaty should be interpreted as alleged by Argentina – by excluding from its application every specific situation that has not been included –, we would be bound to reach the conclusion that, in cases of discrimination, arbitrary measures, or treatment short of the just and equitable standard, there would not be a right to compensation under the Treaty – an unlikely intended result by the Contracting Parties given the Treaty’s purpose. If a matter is dealt with in a provision of the Treaty and not specifically mentioned under other provisions, it does not necessarily follow that the other provisions should be considered to exclude the matter especially covered.44
The problem with the expressio unius principle is not so much a lack of consistency of the tribunals but its limited usefulness. Whether the mention of one item or a list of items in a provision really excludes the relevance of other items depends very much on the particular circumstances and cannot be answered in a generalized way. Similarly, the question whether a provision in one treaty may be taken as proof that another treaty that lacks such a provision was meant to exclude the effects of the provision is difficult to answer in a generalized way with the tools of abstract logic. 2. Interpretation in the Light of Other Treaties The large number of BITs, often containing similar or identical provisions, lends itself to a comparative approach. Especially the BITs of the host State but also of the investor’s home State with third countries often lead to extensive comparisons and inferences.45 The similarities and differences in the treaties offer infinite possibilities to try and draw conclusions.
40
para. 82. Footnote omitted. Enron v. Argentina, Decision on Jurisdiction, 14 January 2004. 42 para. 46. 43 Siemens v. Argentina, Decision on Jurisdiction, 3 August 2004. 44 para. 140. 45 For an extensive comparative analysis of this kind see Aguas del Tunari, S.A. v. Bolivia, Decision on Jurisdiction, 21 October 2005, paras. 289–314. 41
136
christoph schreuer
At the simplest level, it seems plausible that identical or very similar wording in different treaties has the same meaning unless a different meaning can be gathered from the circumstances. The Tribunal in Enron v. Argentina46 said to this effect: 47. … Indeed, the interpretation of a bilateral treaty between two parties in connection with the text of another treaty between different parties will normally be the same, unless the parties express a different intention in accordance with international law. A similar logic is found in Article 31 of the Vienna Convention in so far as subsequent agreement or practice between the parties to the same treaty is taken into account regarding the interpretation of the treaty. There is no evidence in this case that the intention of the parties to the Argentina-United States Bilateral Treaty might be different from that expressed in other investment treaties invoked.47
But even this seemingly simple principle has narrow limits. Taken out of its specific context a seemingly similar provision can assume an entirely different meaning. This is illustrated by the interpretation of national treatment clauses such as Article 1102 of the NAFTA or similar provisions in BITs. Tribunals have refused to simply adopt the practice of GATT/WTO for the interpretation of national treatment clauses in investment treaties.48 The Tribunal in Methanex v. United States49 said in this context: As to the third general principle, the term is not to be examined in isolation or in abstracto, but in the context of the treaty and in the light of its object and purpose. One result of this third general principle, being relevant to Methanex’s first argument on GATT jurisprudence and Article 1102 NAFTA, is that, as noted by the International Tribunal for the Law of the Sea in The MOX Plant case (as also applied in The OSPAR case): “the application of international law rules on interpretation of treaties to identical or similar provisions of different treaties may not yield the same results, having regard, to, inter alia, differences in the respective contexts, objects and purposes, subsequent practice of parties and travaux preparatoires.”50
3. The Significance of Model Treaties for Interpretation BITs are typically based on model treaties. Many States have model BITs which serve as a starting point for negotiations with a prospective treaty partner. To what extent the final product will resemble a country’s model will depend on the relative weight of the negotiating partners, on whether the other country
46
Enron v. Argentina, Decision on Jurisdiction, 14 January 2004. para. 47. In the same sense: Sempra Energy Intl. v. Argentina, Decision on Jurisdiction, 11 May 2005, at para. 144. 48 See Occidental Exploration and Production Co. v. Ecuador, Award, 1 July 2004, at paras. 174–6. But see for a contrary approach: Pope and Talbot v. Canada, Award on the Merits, 10 April 2001, at paras. 45–63, 68–9, ICSID Reports 7: 102. 49 Methanex Corp. v. United States, Award, 3 August 2005. 50 Part II, Chapter B, para. 16. See also paras. 25–37. 47
diversity and harmonization of treaty interpretation
137
also has a model BIT that it wants to promote and on the general circumstances of the negotiations. In Siemens v. Argentina,51 Argentina had argued that the fact that a particular treaty provision departed from Germany’s model BIT and was hence specifically negotiated should be given special weight in its interpretation. The Tribunal rejected this contention. It said: 106. The Respondent has stressed the fact that the dispute settlement clause departs from the standard bilateral investment treaty of Germany in order to support its argument that this was a clause specially negotiated and hence which should be differentiated from the rest. The acceptance of a clause from a model text does not invest this clause with either more or less legal force than other clauses which may had [sic] been more difficult to negotiate. The end result of the negotiations is an agreed text and the legal significance of each clause is not affected by how arduous was the negotiating path to arrive there. The Tribunal feels bound, in its interpretation of the Treaty, by the expressed intention of the parties to promote investments and create conditions favorable to them. The Tribunal finds that when the intention of the parties has been clearly expressed, it is not in its power to second-guess their intentions by attributing special meaning to phrases based on whether they were or were not part of a model draft.52
In National Grid v. Argentina53 the United Kingdom had included a clarifying phrase concerning the reach of the MFN clause in its model treaty. But that model BIT was subsequent to the BIT that was applicable in the particular case which did not contain the clarifying phrase. The Tribunal found that the opening phrase “[f]or the avoidance of doubt” indicated that the clarifying phrase reflected the intention of the United Kingdom even with respect to earlier BITs. The Tribunal said: The implication in the wording of this additional paragraph is that, all along, this was the UK’s understanding of the meaning of the MFN clause in previously concluded investment treaties.54
E. The Use of Travaux Preparatoires According to Article 32 of the VCLT, the materials reflecting the preparatory work to a treaty only figure as supplementary means of interpretation. They are to be used only to confirm a meaning resulting from the primary means of interpretation contained in Article 31 or to determine the meaning if the primary means leave the meaning ambiguous or obscure or lead to a result that is manifestly absurd or unreasonable. In practice, resort to travaux preparatoires seems to be determined less by their position among the canons of interpretation than by their availability. 51 52 53 54
Siemens v. Argentina, Decision on Jurisdiction, 3 August 2004. para. 106. National Grid v. Argentine Republic, Decision on Jurisdiction, 20 June 2006. para. 85. Footnote omitted.
138
christoph schreuer
The drafting history of the ICSID Convention is documented in detail, readily available and easily accessible through an analytical index.55 As a consequence ICSID tribunals frequently resort to it. By contrast, the negotiating history of BITs is typically not documented. Therefore tribunals do not have the possibility to rely on travaux preparatoires even if they are minded to do so. In Aguas del Tunari, S.A. v. Bolivia56 the Tribunal requested the parties to submit any available evidence concerning the BITs interpretation and practice.57 It summarized the unsatisfactory outcome of this attempt in the following terms: 274. This sparse negotiating history thus offers little additional insight into the meaning of the aspects of the BIT at issue, neither particularly confirming nor contradicting the Tribunal’s interpretation.58
The position with NAFTA occupies a middle ground. For a number of years the documents illustrating the negotiating history were unavailable to the public. This had led to complaints about an inequality of arms between a respondent State which had access to these materials and a claimant investor who did not. In July 2004 the NAFTA Free Trade Commission announced the release of the negotiating history of Chapter Eleven of NAFTA dealing with investment.59 Tribunals have referred to aspects of NAFTA’s negotiating history.60 The Tribunal in Methanex v. United States61 not only denied a request for documentary disclosure of the travaux to a number of NAFTA Articles.62 It also stressed the limited relevance of the negotiating history under Article 32 of the VCLT. The Tribunal said: 22. Article 32: With respect to Article 32 of the Vienna Convention, Methanex has sought disclosure from the USA of the negotiating history of Articles 1101, 1102, 1105 and 2101 NAFTA in order to resolve the issues of their interpretation, as considered further below in Chapter II H of this Award. For present purposes, it is sufficient to note that, pursuant to Article 32, recourse may be had to supplementary means of interpretation only in the limited circumstances there specified. Other than that, the approach of the Vienna Convention is that the text of the treaty is deemed to be the authentic expression of the intentions of the
55 ICSID, History of the ICSID Convention. Documents Concerning the Origin and the Formulation of the Convention on the Settlement of Investment Disputes between States and Nationals of Other States (Washington, DC: ICSID, 1968, reprinted in 2001). 56 Aguas del Tunari, S.A. v. Bolivia, Decision on Jurisdiction, 21 October 2005. 57 para. 268. 58 para. 274. 59 The documents are published at <www.naftaclaims.com/commission.htm> (accessed on 1 May 2009) It is unclear whether the available documentation covers all existing documents. 60 See e.g., Grand River Enterprises Six Nations Ltd. v. United States, Decision on Jurisdiction, 20 July 2006, para. 35. 61 Methanex Corp. v. United States, Award, 3 August 2005. 62 Part II, Chapter H, paras. 1–26.
diversity and harmonization of treaty interpretation
139
parties; and its elucidation, rather than wide-ranging searches for the supposed intentions of the parties, is the proper object of interpretation.63
II. The Authority of Previous Decisions A. The Treatment of ‘Precedents’ Reliance on past decisions is a fundamental feature of any orderly decision process. Drawing on the experience of past decisions plays an important role in securing the necessary uniformity and stability of the law. The need for a coherent case law is evident. It strengthens the predictability of decisions and enhances their authority. In actual fact, tribunals in investment disputes, including ICSID tribunals, rely on previous decisions of other tribunals whenever they can. At the same time, it is also well-established that the doctrine of precedent, in the sense known in the common law, does not apply in international adjudication.64 In other words, tribunals in investment arbitrations are not bound by previous decisions of other tribunals. Each tribunal is constituted ad hoc for the particular case. Therefore, ICSID cannot be expected to act like an international court such as the ICJ or ECtHR. The first part of Article 53(1) of the ICSID Convention states: “The award shall be binding on the parties…”. This may be read as excluding the applicability of the principle of binding precedent to successive ICSID cases.65 Nothing in the Convention’s travaux preparatories suggests that the doctrine of stare decisis should be applied to ICSID arbitration. Tribunals have pointed out repeatedly that they are not bound by previous ICSID cases. In the annulment proceedings in Amco v. Indonesia66 the ad hoc Committee stated: 44. Neither the decisions of the International Court of Justice in the case of the Award of the King of Spain nor the Decision of the Klockner ad hoc Committee are binding on this ad hoc Committee. The absence, however, of a rule of stare decisis in the ICSID arbitration system does not prevent this ad hoc Committee from sharing the interpretation given to Article 52(1)(e) by the Klockner ad hoc Committee. This interpretation is well founded in the context of the Convention and in harmony with applicable international jurisprudence. Therefore this ad hoc Committee does not feel compelled to distinguish strictly between the ratio decidendi and obiter dicta in the Klockner ad hoc Committee decision.67 63
Part II, Chapter B, para. 22. Footnote omitted. See Mohamed Shahabuddeen, Precedent in the World Court (Cambridge: CUP, 1996). 65 Art. 59 of the Statute of the International Court of Justice is more specific on this point by saying: “The decision of the Court has no binding force except between the parties and in respect of that particular case”. 66 Amco v. Indonesia, Decision on Annulment, 16 May 1986, ICSID Reports 1: 521. 67 para. 44; see also Amco v. Indonesia, Decision on Jurisdiction, 25 September 1983, ICSID Reports 1: 395. 64
140
christoph schreuer
Similarly, in LETCO v. Liberia,68 the Tribunal, before quoting authority from other ICSID tribunals, said: “Though the Tribunal is not bound by the precedents established by other ICSID Tribunals, it is nonetheless instructive to consider their interpretations”.69 Tribunals operating under the NAFTA in the framework of the ICSID Additional Facility have reached the same result. For instance, the Tribunal in Feldman v. Mexico70 said: … this Tribunal has also sought guidance in the decisions of several earlier NAFTA Chapter 11 Tribunals that have interpreted Article 1110. The Tribunal realizes that under NAFTA Article 1136(1), “An award made by a Tribunal shall have no binding force except between the disputing parties and in respect of the particular case”, and that each determination under Article 1110 is necessarily fact-specific. However, in view of the fact that both of the parties in this proceeding have extensively cited and relied upon some of the earlier decisions, the Tribunal believes it appropriate to discuss briefly relevant aspects of earlier decisions …71
Other tribunals have followed the same line in finding that they are not bound by previous decisions but will take due account of them.72 The question of the authority of previous decisions became the subjectmatter of heated debate in some of the cases against Argentina. Despite numerous decisions to the contrary, Argentina tenaciously kept raising the same jurisdictional objections over and over again. In the Decision on Jurisdiction in the resubmitted Vivendi case,73 one of Argentina’s objections concerned the question of whether the participation of foreign shareholders in a domestically incorporated company constituted an ‘investment’. The Tribunal not only rejected the Argentinean objection but added an appendix to its decision in which it listed previous decisions that had rejected that same argument. The Tribunal said:74 94. Finally, numerous arbitral tribunals have rejected this very same jurisdictional objection as shown by the 18 cases referred to in Appendix 1 to this Decision. In each of those eighteen cases the tribunals upheld the right of shareholders to pursue such claims. In 11 cases, the Argentine Republic was respondent 68
LETCO v. Liberia, Award, 31 March 1986, ICSID Reports 2: 346. At p. 352. 70 Feldman v. Mexico, Award, 16 December 2002, ICSID Reports 7: 341. 71 para. 107. 72 EnCana v. Ecuador, Award, 3 February 2006, para. 189; El Paso v. Argentina, Decision on Jurisdiction, 27 April 2006, para. 39; Suez v. Argentina, Decision on Jurisdiction, 16 May 2006, paras. 26, 31, 60–65; Jan de Nul & Dredging International v. Egypt, Decision on Jurisdiction, 16 June 2006, paras. 63, 64; Azurix v. Argentina, Award, 14 July 2006, para. 391; Pan American Energy v. Argentina, Decision on Preliminary Objections, 27 July 2006, para. 42; Grand River Enterprises Six Nations Ltd. v. United States, Decision on Jurisdiction, 20 July 2006, para. 36. 73 Compania de Aguas del Aconquija, S. A. & Vivendi Universal (formerly Compagnie Generale des Eaux) v. Argentine Republic, Decision on Jurisdiction, 14 November 2005. 74 para. 94. 69
diversity and harmonization of treaty interpretation
141
and asserted, and lost, this same objection. In the last one of these cases, the Tribunal observed that the very objection which Argentina raises in this case has been made numerous times, never, so far as the Tribunal has been aware, with success.75
Despite the tedium of repeated identical objections, tribunals have treated Argentina’s right to reintroduce arguments that had failed in other cases and its insistence on the irrelevance of previous decisions with respect and caution. In Enron v. Argentina76 the Tribunal said: 40. The Tribunal is of course mindful that decisions of ICSID or other arbitral tribunals are not a primary source of rules. The citations of and references to those decisions respond to the fact that the Tribunal in examining the claim and arguments of this case under international law, believes that in essence the conclusions and reasons of those decisions are correct.77
In a subsequent decision in the same case the Tribunal was even more specific. It said: The Tribunal agrees with the view expressed by the Argentine Republic in the hearing on jurisdiction held in respect of this dispute, to the effect that the decisions of ICSID tribunals are not binding precedents and that every case must be examined in the light of its own circumstances.78
By far the most extensive discussion of the value of previous decisions as ‘precedents’ can be found in AES Corp. v. Argentina.79 In that case the Claimant had pointed out that all of Argentina’s objections to jurisdiction had been raised repeatedly in similar terms in other cases and that these same objections had been rejected consistently by other tribunals.80 In response, Argentina insisted on the specificity of each treaty involved and said: Repeating decisions taken in other cases, without making the factual and legal distinctions, may constitute an excess of power and may affect the integrity of the international system for the protection of investments.81
The Tribunal agreed with Argentina and stated that the provisions of Article 25 of the ICSID Convention, together with fundamental principles of public international law, dictate that: … each decision or award delivered by an ICSID Tribunal is only binding on the parties to the dispute settled by this decision or award.82 There is so far no 75 Gas Natural SDG, S.A. v. The Argentine Republic, ICSID Case No. ARB/03/10, Decision of the Tribunal on Preliminary Question on Jurisdiction, 17 June 2005. [footnote original]. 76 Enron v. Argentina, Decision on Jurisdiction, 14 January 2004. 77 para. 40. 78 Enron v. Argentina, Decision on Jurisdiction (Ancillary Claim), 2 August 2004, para. 25. 79 AES Corp. v. Argentina, Decision on Jurisdiction, 26 April 2005, paras. 17–33. 80 paras. 17, 18. 81 Quoted at para. 22. 82 Article 53 of the ICSID Convention. [footnote original].
142
christoph schreuer rule of precedent in general international law; nor is there any within the specific ICSID system for the settlement of disputes between one State party to the Convention and the National of another State Party. This was in particular illustrated by diverging positions respectively taken by two ICSID tribunals on issues dealing with the interpretation of arguably similar language in two different BITs.83
The AES Tribunal pointed out that each BIT has its own identity and that striking similarities in the wording of many BITs often dissimulate real differences.84 The Tribunal drew the following conclusion: From the above derive at least two consequences: the first is that the findings of law made by one ICSID tribunal in one case in consideration, among others, of the terms of a determined BIT, are not necessarily relevant for other ICSID tribunals, which were constituted for other cases; the second is that, although Argentina had already submitted similar objections to the jurisdiction of other tribunals prior to those raised in the present case before this Tribunal, Argentina has a valid and legitimate right to raise the objections it has chosen for opposing the jurisdiction of this Tribunal.85
At the same time the Tribunal rejected the “excessive assertion” that “absolutely no consideration might be given to other decisions on jurisdiction or awards delivered by other tribunals in similar cases”.86 In case of a high level of similarity or identity of underlying legal questions the Tribunal did not feel barred as a matter of principle from considering the position taken by other tribunals.87 This led the Tribunal to the following compromise solution: An identity of the basis of jurisdiction of these tribunals, even when it meets with very similar if not even identical facts at the origin of the disputes, does not suffice to apply systematically to the present case positions or solutions already adopted in these cases. Each tribunal remains sovereign and may retain, as it is confirmed by ICSID practice, a different solution for resolving the same problem; but decisions on jurisdiction dealing with the same or very similar issues may at least indicate some lines of reasoning of real interest; this Tribunal may consider them in order to compare its own position with those already adopted by its predecessors and, if it shares the views already expressed by one or more of these tribunals on a specific point of law, it is free to adopt the same solution.88
83 SGS Societe Generale de Surveillance S.A. v. Islamic Republic of Pakistan, ICSID Case N° ARB/01/13 and SGS Societe Generate de Surveillance S.A v. Republic of the Philippines, ICSID Case N° ARB/02/6. [footnote original]. 84 paras. 24–5. 85 para. 26. 86 para. 27. 87 para. 28. 88 para. 30.
diversity and harmonization of treaty interpretation
143
Having made these broad statements on the limited value of “precedents”, the Tribunal actually proceeded to examine and rely on previous decisions by other tribunals.89 The decision in Gas Natural v. Argentina90 demonstrates the Tribunal’s cautious attitude towards ‘precedents’ not only in its wording but also in the decision’s structure. The Tribunal first gave its Decision on Jurisdiction without reference to previous cases.91 After having reached a result it stated: 36. The Tribunal wishes to emphasize that it has rendered its decision independently, without considering itself bound by any other judgments or arbitral awards. Having reached its conclusions, however, the Tribunal thought it useful to compare its conclusion with the conclusions reached in other recent arbitrations conducted pursuant to the ICSID Arbitration Rules and arising out of claims under contemporary bilateral investment treaties. We summarize a few of these decisions here, and confirm that we have not found or been referred to any decisions or awards reaching a contrary conclusion.92
Only after having made that statement does the Tribunal examine a number of previous decisions which it finds to be in line with its own conclusions.93 Its conclusion is as follows: 52. In sum, the Tribunal is satisfied that its analyses and decisions, independently arrived at, are consistent with the conclusions of other arbitral tribunals faced with similar issues. It does not follow that the ultimate decisions of this Tribunal on the merits will be wholly consistent with those of other arbitral tribunals, because different claims have been based on different treaties and different factual situations.94
Whether a decision that relies preponderantly or exclusively on previous decisions might be subject to annulment for that reason may be subject to doubt. No decision on annulment in ICSID proceedings has ever annulled an award because it rested its reasoning on precedents. But an application for annulment that alleges an excess of powers or a failure to state reasons because the tribunal has simply relied on earlier decisions without making an independent decision or developing its own reasons is entirely possible. From the perspective of tribunals it seems wiser not to expose themselves to this charge.
89 paras. 51–9, 70, 73, 86, 89, 95–7. See also Bayindir v. Pakistan, Decision on Jurisdiction, 14 November 2005, para. 76: “The Tribunal agrees that it is not bound by earlier decisions, but will certainly carefully consider such decisions whenever appropriate”. [referring to AES Corp. v. Argentina]. 90 Gas Natural SDG, S.A. v. Argentina, Decision on Jurisdiction, 17 June 2005. 91 paras. 20–35. 92 para 36. 93 paras. 37–51. 94 para. 52.
144
christoph schreuer B. Inconsistent Decisions
In some cases tribunals did not follow earlier decisions but adopt different solutions. At times they simply adopted a different solution without distancing themselves from the earlier decision. At other times they referred to the earlier decision and pointed out that they were unconvinced by what another tribunal had said and that, therefore, their decision departed from the one adopted earlier. A clear example of a rejection of an earlier decision occurred in SGS v. Philippines.95 The Tribunal discussed the earlier decision in SGS v. Pakistan96 and voiced its disagreement with some of the answers given there. The SGS v. Philippines Tribunal said: As will become clear, the present Tribunal does not in all respects agree with the conclusions reached by the SGS v. Pakistan Tribunal on issues of the interpretation of arguably similar language in the Swiss-Philippines BIT. This raises a question whether, nonetheless, the present Tribunal should defer to the answers given by the SGS v. Pakistan Tribunal. The ICSID Convention provides only that awards rendered under it are “binding on the parties” (Article 53(1) ), a provision which might be regarded as directed to the res judicata effect of awards rather than their impact as precedents in later cases. In the Tribunal’s view, although different tribunals constituted under the ICSID system should in general seek to act consistently with each other, in the end it must be for each tribunal to exercise its competence in accordance with the applicable law, which will by definition be different for each BIT and each Respondent State. Moreover there is no doctrine of precedent in international law, if by precedent is meant a rule of the binding effect of a single decision. There is no hierarchy of international tribunals, and even if there were, there is no good reason for allowing the first tribunal in time to resolve issues for all later tribunals. It must be initially for the control mechanisms provided for under the BIT and the ICSID Convention, and in the longer term for the development of a common legal opinion or jurisprudence constante, to resolve the difficult legal questions discussed by the SGS v. Pakistan Tribunal and also in the present decision.97
The consistency of decisions has become a prominent issue in investment arbitration.98 There have been a number of instances in which tribunals sitting in different cases have come to conflicting conclusions on identical questions. Four examples may suffice to illustrate this point:
95
SGS v. Philippines, Decision on Jurisdiction, 29 January 2004. SGS v. Pakistan, Decision on Jurisdiction, 6 August 2003, ICSID Reports 8: 383. 97 SGS v. Philippines, at para. 97. Footnotes omitted. 98 In addition to conflicting answers to similar questions in different cases there is the occasional problem of conflicting outcomes of parallel proceedings concerning the same dispute. See especially Ronald S. Lauder v. The Czech Republic, Final Award, 3 September 2001, ICSID Reports 9: 66 and CME v. The Czech Republic, Partial Award, 13 September 2001, ICSID Reports 9: 121. 96
diversity and harmonization of treaty interpretation
145
1. Many BITs in their consent clauses contain phrases referring to “all disputes concerning investments” or “any legal dispute concerning an investment”. Tribunals have given conflicting meanings to these.99 2. Umbrella clauses have received divergent interpretations in the practice of tribunals.100 3. A common condition in BITs for the institution of arbitration proceedings is the observance of so-called waiting periods. This means that an amicable settlement must have been attempted through consultations or negotiations for a certain period of time. The reaction of tribunals to these provisions has not been uniform. In some cases the tribunals found that noncompliance with the waiting periods did not affect their jurisdiction.101 In other cases they reached the opposite conclusion.102 4. Most BITs and some other treaties for the protection of investment contain most favoured nation (MFN) clauses. This has led to the question of whether the effect of MFN clauses extends to the provisions on dispute settlement in these treaties. Tribunals have given conflicting decisions on this point.103 III. Consistency and Harmonization Fortunately the problem of inconsistency is not pervasive. Most tribunals carefully examine earlier decisions and accept these as authority most of the time. But sometimes they disagree with them and make their disagreement
99 Salini Costruttori SpA et Italstrade SpA c/Royaume du Maroc, Decision on Jurisdiction, 23 July 2001, Journal de Droit International (2002): 196, ICSID Reports 6: 400, para. 61; Compania de Aguas del Aconquija, S. A. & Vivendi Universal (formerly Compagnie Generale des Eaux) v. Argentine Republic, Decision on Annulment, 3 July 2002, ICSID Reports 6: 340, para. 55; SGS v. Pakistan, Decision on Jurisdiction, 6 August 2003, ICSID Reports 8: 383, para. 55; SGS v. Philippines, Decision on Jurisdiction, 29 January 2004, ICSID Reports 8: 518, paras. 131–5. 100 SGS v. Pakistan, Decision on Jurisdiction, 6 August 2003, ICSID Reports 8: 383 at paras. 163–73; SGS v. Philipines, Decision on Jurisdiction, 29 January 2004, ICSID Reports 8: 518, paras. 125, 128; Joy Mining v. Egypt, Award, 6 August 2004, para. 81; CMS v. Argentina, Award, 12 May 2005, paras. 296–303; Eureko v. Poland, Partial Award, 19 August 2005, paras. 244–60; Noble Ventures v. Romania, Award, 12 October 2005, paras. 42–62; El Paso v. Argentina, Decision on Jurisdiction, 27 April 2006, paras. 66–86; Pan American Energy v. Argentina, Decision on Preliminary Objections, 27 July 2006, paras. 92–115. 101 Ethyl Corp. v. Canada, Decision on Jurisdiction, 24 June 1998, ICSID Reports 7: 12 at paras. 76–88 and in Ronald S. Lauder v. The Czech Republic, Final Award, 3 September 2001; Wena Hotels v. Egypt, Decision on Jurisdiction, 29 June 1999, ICSID Reports 6: 74, 87; SGS v. Pakistan, Decision on Jurisdiction, 6 August 2003, ICSID Reports 8: 383, para. 184. 102 A. Goetz v. Burundi, Award, 10 February 1999, ICSID Reports 6: 5, at paras. 90–3; Enron Corp. and Ponderosa Assets, L.P. v. Argentina, Decision on Jurisdiction, 14 January 2004, para. 88. 103 Maffezini v. Spain, Decision on Jurisdiction, 25 January 2000, ICSID Reports 5: 396, paras. 38–64; Siemens v. Argentina, Decision on Jurisdiction, 3 August 2004, paras. 32–110; Salini v. Jordan, Decision on Jurisdiction, 29 November 2004, paras. 115, 119; Plama v. Bulgaria, Decision on Jurisdiction, 8 February 2005, paras. 216–26; Gas Natural v. Argentina, Decision on Jurisdiction, 17 June 2005, paras. 24–31, 41–9.
146
christoph schreuer
known. In addition, the growing number of simultaneous cases makes it increasingly likely that tribunals may reach conflicting results without realizing it. Therefore, the problem of conflicting awards is a reality and has led to a discussion on how to address the problem. A. Interpretative Statements by States Occasionally the two States parties to the BIT may issue a joint statement without binding force on a question of interpretation pending before a tribunal. In CME v. Czech Republic104 the BIT between the Czech Republic and the Netherlands provided for “consultations” with a view to resolving any issue of interpretation and application of the Treaty. Pursuant to this procedure, the Netherlands and the Czech Republic issued ‘Agreed Minutes’ containing a ‘common position’ on the BIT’s interpretation, after the Tribunal had issued a partial award.105 The Tribunal took this joint statement into account as supporting its view.106 Unilateral assertions of the disputing State party, on the meaning of a treaty provision, made in the process of ongoing proceedings are of limited value. Such a statement will be perceived as self-serving since it is probably determined by the desire to influence the tribunal’s decision in favour of the State offering the interpretation. Alternatively, the non-disputing State party to a BIT may give a unilateral statement on the interpretation of the treaty. Such a statement may or may not confirm the position of the disputing State party to the treaty. In Aguas del Tunari v. Bolivia,107 the Claimant had submitted statements made by Ministries of the Government of the Netherlands to the Parliament of the Netherlands.108 In addition, the Tribunal wrote to the Legal Adviser to the Foreign Ministry of the Netherlands enquiring about certain aspects of the BIT’s interpretation.109 The Tribunal found the information thus obtained not helpful.110 It said: … the Tribunal can find no subsequent practice … which establishes an agreement of the parties regarding the interpretation of the BIT. In addition, the response from the Netherlands provides no additional information of the type suggested by Article 31 of the Vienna Convention on the Law of Treaties as being possibly relevant and upon which a general interpretative position might be based.111
In one case the government of the Claimant’s nationality took the unusual step of writing to ICSID to voice its disapproval of an interpretation given by an
104 105 106 107 108 109 110 111
CME v. The Czech Republic, Final Award, 14 March 2003, ICSID Reports 9: 264. paras. 87–93. paras. 437, 504. Aguas del Tunari v. Bolivia, Decision on Jurisdiction, 21 October 2005. paras. 249–57. paras. 47, 258–9. paras. 260–3. para. 262.
diversity and harmonization of treaty interpretation
147
ICSID tribunal. In SGS v. Pakistan112 the Swiss Government in a letter to ICSID’s Deputy Secretary-General stated that the Swiss authorities were wondering why the Tribunal had not found it necessary to enquire about their view of the meaning of the provision in the Pakistan-Switzerland BIT in spite of the fact that the Tribunal attributed considerable importance to the intent of the Contracting Parties in drafting it. The Swiss authorities were alarmed by the interpretation given by the Tribunal to the provision. The letter added that the interpretation ran counter to the intention of Switzerland when concluding the Treaty and was neither supported by the meaning of similar articles in BITs concluded by other countries nor by academic comments.113 It is obvious from these examples of practice that occasional views expressed by States parties to treaties on the meaning of particular provisions are not a viable method to achieve uniformity of interpretation. B. Institutionalized Mechanisms 1. Official Interpretations Plans to create institutionalized mechanisms to achieve uniform interpretations have yielded limited results so far. The NAFTA has a mechanism whereby the Free Trade Commission (FTC), a body composed of representatives of the three States parties, can adopt binding interpretations of the treaty.114 The FTC has made use of this method in July 2001 in interpreting the concepts of ‘fair and equitable treatment’ and ‘full protection and security’ under Article 1105 of the NAFTA.115 NAFTA tribunals have accepted this interpretation as binding.116 The Tribunal in Methanex v. United States117 said:
112
SGS v. Pakistan, Decision on Jurisdiction, 6 August 2003. See Stanimir A. Alexandrov, ‘Breaches of Contract and Breaches of Treaty’, The Journal of World Investment & Trade 5 (2004): 555, at 570–1; Emmanuel Gaillard, ‘Investment Treaty Arbitration and Jurisdiction Over Contract Claims -the SGS Cases Considered’, in International Investment Law and Arbitration, Todd Weiler (ed.) (London: Cameron May, 2005), 325, at 341–2. 114 NAFTA Article 2001(1): The Parties hereby establish the Free Trade Commission, comprising cabinet-level representatives of the Parties or their designees. NAFTA Article 1131(2): An interpretation by the Commission of a provision of this Agreement shall be binding on a Tribunal established under this Section. 115 FTC Note of Interpretation of 31 July 2001. 116 See Mondev International Ltd. v. United States of America, Award, 11 October 2002, ICSID Reports 6: 192, paras. 100 et seq.; United Parcel Service of America, Inc. v. Canada, Award, 22 November 2002, ICSID Reports 7: 288, para. 97; Adf Group, Inc. v. United States of America, Award, 9 January 2003, ICSID Reports 6: 470, paras. 175–8; Loewen Group, Inc. and Raymond L. Loewen v. United States of America, Award, 26 June 2003, ICSID Reports 7: 442, paras. 124–8; Waste Management, Inc. v. United Mexican States, Award, 30 April 2004, paras. 90–1. See also United Mexican States v. Metalclad Corp., Judgment, Supreme Court of British Columbia, 2 May 2001, ICSID Reports 5: 236, paras. 61–5. 117 Methanex v. United States, Award, 3 August 2005. 113
148
christoph schreuer With respect to Article 1105, the existing interpretation is contained in the FTC’s Interpretation of 31st July 2001. Leaving to one side the impact of Article 1131(2) NAFTA, the FTC’s interpretation must also be considered in the light of Article 31(3)(a) of the Vienna Convention as it constitutes a subsequent agreement between the NAFTA Parties on the interpretation of Article 1105 NAFTA.118
BITs do not normally have institutional mechanism to obtain authentic interpretations of their meaning. But the United States Model BIT of 2004 provides for a mechanism that is similar to the one in the NAFTA: Article 30(3): A joint decision of the Parties, each acting through its representative designated for purposes of this Article, declaring their interpretation of a provision of this Treaty shall be binding on a tribunal, and any decision or award issued by a tribunal must be consistent with that joint decision.
This method is efficient, but has a serious drawback. States will strive to issue official interpretations to influence proceedings to which they are parties. As the example of the July 2001 interpretation of the FTC under NAFTA demonstrates, the home States of disputing investors are less interested in interpretations favourable to their nationals in pending disputes than in interpretations that favour State respondents generally. It is obvious that a mechanism whereby a party to a dispute is able to influence the outcome of judicial proceedings, by issuing official interpretation to the detriment of the other party, is incompatible with principles of a fair procedure and is hence undesirable. 2. Appeals Procedures Another perceived solution is the creation of an appeals facility that would open the possibility to review decisions thereby increasing the chances of a consistent case law. A number of US treaties foresee this possibility in the form of an appellate body or similar mechanism.119 The United States Model BIT of 2004 contains the following provision in an Annex: Annex D Possibility of a Bilateral Appellate Mechanism: Within three years after the date of entry into force of this Treaty, the Parties shall consider whether to establish a bilateral appellate body or similar mechanism to review awards rendered under Article 34 in arbitrations commenced after they establish the appellate body or similar mechanism.
This idea of a bilateral appeals mechanism has found entry into the US BIT with Uruguay.120 It is doubtful whether appellate bodies established under different bilateral treaties would lead to a coherent case law. Separate mechanisms for different 118
Part II, Chapter, at para. 23. Generally see Barton Legum, ‘The Introduction of an Appellate Mechanism: the U.S. Trade Act of 2002’, in Annulment of ICSID Awards, Emmanuel Gaillard and Yas Banifatemi (eds.) (New York, NY: Juris Publishing, 2004), 289. 120 See 2004 Uruguay-US BIT, ILM 44 (2005): 268, 296, Annex E. 119
diversity and harmonization of treaty interpretation
149
treaties, even if used regularly, will only have a limited impact. The issues in question, such as the proper interpretation of umbrella clauses or the reach of MFN clauses, recur in the context of many treaties. A harmonizing effect will be achieved only if the institutional mechanism applies to all or at least many treaties. A similar idea though directed at a multilateral appeals mechanism is reflected in the CAFTA-DR Free Trade Agreement (FTA)121 as well as in US FTAs with Singapore122 and Chile.123 To this effect the FTA with Chile provides: 10. If a separate multilateral agreement enters into force as between the Parties that establishes an appellate body for purposes of reviewing awards rendered by tribunals constituted pursuant to international trade or investment agreements to hear investment disputes, the Parties shall strive to reach an agreement that would have such appellate body review awards rendered under Article 10.25 in arbitrations commenced after the appellate body’s establishment.124
ICSID at one point floated a draft that foresaw the creation of an appeals facility at ICSID. A Discussion Paper of October 2004125 pointed to the danger of inconsistencies, and offered the prospect of a single appeal mechanism as an alternative to multiple mechanisms. An annex to the paper presented possible features of an ICSID Appeals Facility. Submission to the Appeals Facility would have to take place by treaty. It would be available for ICSID as well as for non-ICSID awards rendered in investor-State arbitrations. An appeal would be heard by an appeal tribunal consisting of three members and selected for each case from a panel of 15 persons. The appeal would have to be based on an error of law or fact or one of the grounds for annulment set out in Article 52 of the ICSID Convention. The appeal tribunal would have the power to uphold modify or reverse the award concerned. In a subsequent Working Paper of May 2005126 the idea of an appellate mechanism was dropped, at least for the time being, since “it would be premature to attempt to establish such an ICSID mechanism at this stage”. The project of a multilateral appeals mechanism at ICSID would have created certain problems. Article 53 of the ICSID Convention says that an award shall not be subject to any remedy except those provided for in the Convention.
121 Central America-Dominican Republic Free Trade Agreement, 5 August 2004, Article 10.20(10). 122 Singapore-US FTA, 1 January 2004, Article 15.19(10). 123 Chile-US FTA, 1 January 2004. 124 Article 10.19(10). 125 ICSID Secretariat, Possible Improvements of the Framework for ICSID Arbitration (Discussion Paper of 2 October 2004) <www.worldbank.org/icsid/improve-arb.pdf> (accesssed on 1 December 2006). 126 ICSID Secretariat, Suggested Changes to the ICSID Rules and Regulations (Working Paper of the ICSID Secretariat, 12 May 2005) (accessed on 1 December 2006).
150
christoph schreuer
In addition, the appeals mechanism would have excluded the application of the annulment procedure under Article 52 of the ICSID Convention. Any attempt to amend the ICSID Convention would be far too complex to be realistic. The idea was that bilateral treaties providing for appellate mechanisms would operate in partial derogation of the ICSID Convention. While this avenue may be feasible under the law of treaties, it is questionable whether an organ created by a multilateral convention is the right body to take the initiative towards the derogation of that convention by way of bilateral or regional treaties. An appeal before three arbitrators selected from a panel of fifteen would not have guaranteed any uniformity of decisions. The different composition of appeals tribunals would not offer a guarantee against inconsistency. Another problem would be the avoidance of national setting aside procedures for non-ICSID awards. The legal bases for these setting aside procedures are national arbitration laws. It would be necessary to ensure that, where an appeal is available under the international procedure, there would be no possibility to turn to a national court for the setting aside of the award. Equally, it would be important to shield the decisions of the appeal tribunal from review by national courts. Any other solution would lead to an undesirable situation of competing and possibly conflicting appeals procedures. 3. Preliminary Rulings An appeals facility is not necessarily the best mechanism to achieve coherence and consistency in the interpretation of investment treaties. Appeal presupposes a decision that will be attacked for some alleged flaw in order to be repaired. Rather than try and fix the damage after the fact through an appeal, it is more economical and effective to address it preventively before it even occurs. A method to secure coherence and consistency that has been remarkably successful is to allow for preliminary rulings while the original proceedings are still pending.127 Under such a system a tribunal would suspend proceedings and request a ruling on a question of law from a body established for that purpose. This procedure has been very successfully used in the framework of the European Community to secure the uniform application of European Community Law by domestic courts. It is contained in Article 234 (formerly
127 The idea has been put forward before: see Gabrielle Kaufmann-Kohler, ‘Annulment of ICSID Awards in Contract and Treaty Arbitrations: Are there Differences?’, in Annulment of ICSID Awards, Emmanuel Gaillard and Yas Banifatemi (eds.) (New York, NY: Juris Publishing, 2004), 189. See also Gabrielle Kaufmann-Kohler, ‘In search of Transparency and Consistency: ICSID Reform Proposal’, Transnational Dispute Management 2(5) (2005): 8.
diversity and harmonization of treaty interpretation
151
177) of the Treaty establishing the European Community128 and provides for preliminary rulings by the European Court upon the request of domestic courts.129 Adapted to investment arbitration this method could provide for an interim procedure whenever a tribunal is faced with a fundamental issue of investment treaty application, a situation where it wants to depart from a decision by a previous tribunal or where there are conflicting previous decisions. In such a situation the tribunal might be required to suspend proceedings and request a ruling from the central decision maker. Once that ruling has been forthcoming the original tribunal would continue its proceedings. This method has turned out to be very successful to ward off inconsistency and fragmentation. A number of details would have to be worked out. One is under what circumstances a tribunal would request a preliminary ruling and whether it would be under an obligation to do so. Another would be whether these rulings would bind the tribunal or would merely constitute recommendations. Not least, the composition of a body charged with giving preliminary rulings would need to be discussed. Preliminary rulings would leave Article 53 of the ICSID Convention untouched. They would not affect the principles of expediency and finality, two of the chief assets of arbitration. And they would help to prevent the development of inconsistencies rather than create a costly and time consuming repair mechanism.
128
Treaty Establishing the European Community, Article 234: The Court of Justice shall have jurisdiction to give preliminary rulings concerning: (a) the interpretation of this Treaty; (b) the validity and interpretation of acts of the institutions of the Community and of the ECB; the interpretation of the statutes of bodies established by an act of the Council, where those statutes so provide. Where such a question is raised before any court or tribunal of a Member State, that court or tribunal may, if it considers that a decision on the question is necessary to enable it to give judgment, request the Court of Justice to give a ruling thereon. Where any such question is raised in a case pending before a court or tribunal of a Member State against whose decisions there is no judicial remedy under national law, that court or tribunal shall bring the matter before the Court of Justice.
129 From among the numerous writings on preliminary rulings under the EC Treaty see e.g., Mads Andenas (ed.), Article 177 References to the European Court - Policy and Practice (London: Butterworths, 1994); David Anderson, References to the European Court (London: Sweet and Maxwell, 1995); Alan Dashwood and Angus Johnston, (eds.), The Future of the Judicial System of the European Union (Oxford: Hart Publications, 2001); Gráinne de Búrca, G. and Joseph H. H. Weiler, (eds.), The European Court of Justice (Oxford: OUP, 2001).
CANONS OF TREATY INTERPRETATION: SELECTED CASE STUDIES FROM THE WORLD TRADE ORGANIZATION AND THE NORTH AMERICAN FREE TRADE AGREEMENT Malgosia Fitzmaurice & Panos Merkouris* I. Introductory Remarks Treaty interpretation is a notoriously difficult subject. As Lord McNair observed “[t]here is no part of the law of treaties which the text-writer approaches with more trepidation than the question of interpretation.”1 The main purpose of this article is to analyse certain new developments in treaty interpretation as contained in the jurisprudence of such organisations as the World Trade Organisation (WTO) and the North American Free Trade Agreement (NAFTA), organisations that have not been traditionally linked to the application, development or crystallisation of the rules of general (classical) international law.2 The jurisprudence of the WTO and NAFTA relating to treaty interpretation will be assessed to determine whether it complies with the canons of interpretation set out in the 1969 Vienna Convention on the Law of Treaties (1969 VCLT)3 and their crystallisation through the judgments of the World Court. The study will not limit itself to particular aspects of the canons of interpretation (as included in the 1969 VCLT) but will examine them in their entirety. It may be observed that the 1986 Vienna Convention on the Law of Treaties between States and International Organisations or between International Organisations (1986 VCLT), that is not yet in force, follows the same canons of interpretation as the 1969 VCLT.4 There is no doubt that the canons of interpretation as enshrined in the 1969 VCLT form part and parcel of customary international law. This has been
* The present article is an updated and more extensive version of the same-titled article that appeared in ARIEL 10 (2005): 41–94. 1 Lord Arnold McNair, The Law of Treaties (Oxford: Clarendon Press, 1961), 364. 2 On these issues see Jonathan I. Charney, ‘Is International Law Threatened by Multiple International Tribunals?’ RCADI 271 (1998): 139–55. 3 1969 Vienna Convention on the Law of Treaties, ILM 8 (1969): 679; Entry into force 27 January 1980. 4 1986 Vienna Convention on the Law of Treaties between States and International Organisations or between International Organisations, ILM 25 (1986): 543; Anthony Aust, Modern Treaty Law and Practice (Cambridge: CUP, 2000), 7–8.
154
malgosia fitzmaurice & panos merkouris
confirmed in several Judgments of the International Court of Justice (ICJ), the latest being the Kasikili/ Sedudu Island case.5 The Court observed: [a]s regards interpretation of that Treaty [1890 Anglo-German Treaty], the Court notes that neither Botswana nor Namibia are parties to the Vienna Convention on the Law of Treaties of 23 May 1969, but that both of them consider that Article 32 of the Vienna Convention is applicable inasmuch as it reflects customary international law. The Court itself had already had occasions in the past to hold that customary international law found expression in Article 31 of the Vienna Convention…6
II. The 1969 VCLT Canons of Interpretation-General Considerations The purpose of interpreting a treaty is to establish the meaning of the text that the parties intended it to have “in relation to circumstances with reference to which the question of interpretation has arisen.”7 The approach adopted by the International Law Commission (ILC) to the interpretation of treaties was based on the jurisprudence of the World Court. The Commission stated in its Commentary during its work on the codification of the law of treaties that “the jurisprudence of the International Court contains many pronouncements from which it is permissible to conclude that the textual approach to treaty interpretation is regarded as established by law”.8 On the basis of the Court’s practice, Fitzmaurice drew up the following principles of treaty interpretation: I. Principle of actuality or textuality, means that treaties are to be interpreted as they stand, on the basis of their actual words; II. Principle of the natural and ordinary meaning, means that, subject to the principle of contemporaneity, where applicable, particular words and phrases are to be given their normal, natural, and unstrained meaning in the context in which they occur. This principle can only be replaced by direct evidence that the terms used are to be understood in a different manner than the natural and ordinary one and if interpretation would lead to an unreasonable or absurd result. Principle III, integration, is of fundamental importance since
5 Case Concerning Kasikili/Sedudu Island (Botswana v. Namibia), Judgment of 13 December 1999, 1999 ICJ Rep. 1045; Territorial Dispute (Libya v. Chad), Judgment of 3 February 1994, 1994 ICJ Rep. 4, at para. 41; Oil Platforms (Iran v. United States), Preliminary Objections, Judgment of 12 December 1996, 1996 ICJ Rep. 812, at para. 23. 6 Kasikili/Sedudu Island case, supra note 5, at para. 18. In the following Judgments, the Court stated that Art. 31 of the 1969 VCLT reflects international customary law: Libya/Chad case, supra note 5, at 21; Oil Platforms case, supra note 5, 812, at para. 23. 7 Lassa F. L. Oppenheim, Robert Y. Jennings, and Arthur D. Watts (eds.), Oppenheim’s International Law 9th ed. (London: Longman, 1996), Vol. 1. Parts 2 to 4, 1266. 8 ILC, ‘Commentary’, YILC (1966): Vol. II, at 220.
canons of treaty interpretation
155
it indicates that treaties are to be interpreted as a whole, and particular parts, chapters or sections also as a whole. Therefore, no part of a treaty can be interpreted out of context. Principle IV, of effectiveness (ut res magis valeat quam pereat) indicates that treaties are to be interpreted with reference to their declared or apparent objects and purposes; and particular provisions are to be interpreted as to give the fullest effect consistent with the normal sense of the words and with other parts of the text, and in such a way that a reason and meaning can be attributed to every part of the text. Principle V, of subsequent practice permits recourse to the subsequent practice of parties in relation to the treaty. It may be even desirable to do so in order to obtain the best and most reliable evidence derived from how the treaty was interpreted in practice as to what the correct interpretation is. Principle VI, of contemporaneity, means that the terms of a treaty must be interpreted according to the meaning which they possessed, or which has been attributed to them, in the light of linguistic usage current at the time when the treaty was originally concluded.9 With certain simplifications, it may be said that there are three main schools of interpretation: one reflecting subjective or the ‘intention of parties’ approach; a second reflecting objective or the ‘textual’ approach and a third reflecting teleological or the ‘object and purpose’ approach. These schools of interpretation are not “mutually exclusive”10 and the 1969 VCLT is meant to reflect both these approaches. The principle of effectiveness is the most controversial and difficult and in fact covers more than one principle. As Thirlway explained, the first element of this principle is the rule that all provisions of a treaty, or other instrument, must be viewed as intended to have significance and to be necessary to contain the intended meaning, i.e., the interpretation is suspect if it does not convey some meaning. The second element is the rule that the instrument as a whole and each one of its provisions must be taken as intended to achieve some end. Therefore an interpretation that would make the text ineffective to achieve its object is similarly suspect. The second element is similar to the ‘object and purpose’ criterion and as such is “to be employed with discretion”11 since one of the most complicated issues in treaty interpretation is the use of the ‘object and purpose of a treaty’ criterion. It is the second element of the principle of effectiveness that is reflected in the maxim ut res magis valeat quam pereat. Article 31 of the Vienna Convention, stipulates that a treaty should be interpreted in “the light of its
9 Sir Gerald Fitzmaurice, ‘Treaty Interpretation and Other Treaty Points, 1951–1954’, BYIL 33 (1957): 203–93. 10 Sir Ian Sinclair, The Vienna Convention on the Law of Treaties (Manchester: MUP, 1984), 115. 11 Hugh Thirlway, ‘The Law and Procedure of the International Court of Justice’, BYIL 69 (1992): 44.
156
malgosia fitzmaurice & panos merkouris
object and purpose.” However, there are many controversies surrounding the term ‘object and purpose’ which is considered as vague and ill defined, something that makes its use as a tool of treaty interpretation unreliable.12 It would appear that the rule under the second element of the principle of effectiveness is that which is included in the 1969 VCLT. The most difficult and controversial is of course the reconciliation of the objective and the subjective approach in treaty interpretation, a task that is almost impossible. Koskenniemi observes that according to the subjective approach, treaties bind because they express consent. According to the objective approach, they bind due to the requirements of considerations of teleology, utility, reciprocity, good faith or justice. The doctrine of ‘normal’ meaning fails to deal with the fact that already the ascertainment of the ‘normal’ requires interpretation and that the very emergence of the dispute conclusively proves this…
The same author notes that the principal goal of interpretation (to give an expression to the subjective party intentions) cannot be achieved, since it is impossible to ascertain real, subjective party intent.13 According to the ILC, “the starting point of interpretation is the elucidation of the meaning of the text, not an investigation ab initio into the intention of the parties”.14 Therefore, the Commission was of the view that there is a presumption that the text of the treaty must be a real expression of the intentions of the parties. It appears that the preferred method of interpretation of the ICJ was the reliance on the text of a treaty (see below). The basic underlying principle of treaty interpretation contained in the 1969 VCLT is that treaties will be interpreted in good faith, which is the embodiment of the principle pacta sunt servanda (contained in Article 26 of the 1969 VCLT). The ‘rule’ (in singular) of interpretation as enshrined in Article 31 is a procedure consisting of three elements: the text, the context and the object and purpose of a treaty.15 Doctrine, however, is divided to some extent whether the rule of interpretation as contained in the 1969 VCLT is the reflection of a pure contextual approach or is a combination of various approaches. Jennings and Watts in the ninth edition of Oppenheim’s International Law support the exclusively textual approach. However the view of the majority, supported by the text of Article 31, is that the wording of the general rule of interpretation as contained in the 1969 VCLT is phrased in such a manner as to “give some
12 See e.g., Isabelle Buffard and Karl Zemanek, ‘The ‘Object and Purpose’ of a Treaty: An Enigma?’, ARIEL 3 (1998): 311–43; Jan Klabbers, ‘Some Problems Regarding the Object and Purpose of Treaties’, Finnish Yearbook of International Law 7 (1997): 138–61. 13 Martti Koskenniemi, From Apology to Utopia: The Structure of International Legal Argument (Cambridge: CUP, 1989), 291–94, 298–9. 14 ILC, ‘Commentary’, YILC (1966): Vol. II, at 220. 15 Aust, supra note 4, at 187.
canons of treaty interpretation
157
comfort to supporters of both intentions and the object and purpose schools”.16 Although the World Court was a staunch supporter of the textual interpretation, doubts as to the principles on which Article 31 was based remain as one of the unresolved issues of treaty interpretation. The phrase “context of a treaty” means any instrument that is of relevance to the conclusion of a treaty, as well as the preamble to the treaty and its annexes. As it has been observed, paragraphs 1, 2, and 3 of Article 31 of the 1969 VCLT do not create any hierarchical order but embody a logical progression, nothing more than that.17 The Court has consistently adhered to the textual interpretation considering it by far the most important.18 In the Libya/Chad case, the Court stated that [a] treaty must be interpreted in good faith, in accordance with the ordinary meaning to be given to its terms in their context and in the light of its object and purpose. Interpretation must be based above all upon the text of a treaty. As a supplementary measure recourse may be had to means of interpretation such as the preparatory work of the treaty.19
The determination of the ordinary meaning of a term is done in the context of the treaty and in the light of its object and purpose.20 Although the Vienna Convention is very clear as to the use of supplementary means of interpretation including travaux preparatiores, the debate relating to the use and importance of preparatory work has never ceased and is an example of the difficulties that treaty interpretation poses. The use of these means is justified either as a confirmation of the meaning of the terms of the treaty or in cases where the application of the rule of interpretation enshrined in Article 31 leaves the meaning ambiguous and obscure, or leads to a result which is manifestly absurd or unreasonable.21 However, in 16
Donald W. Greig, ‘Oppenheim Revisited: An Australian Perspective’, Australian Yearbook of International Law 14 (1992): 227, at 269–70. 17 Aust, supra note 4, at 187. 18 Maritime Delimitation and Territorial Question Between Qatar and Bahrain (Qatar v. Bahrain), Jurisdiction and Admissibility, Judgment of 15 February 1995, 1995 ICJ Rep. 4, at 14 et seq. 19 Libya/Chad case, supra note 5, at para. 41; the same principle was confirmed in the Kasikili/ Sedudu Island case, supra note 5 at para. 18. 20 E.g., On the Interpretation of the Convention of 1919 Concerning Employment of Women During Night, 1932 PCIJ (Ser. A/B) No. 50. The terms of Art. 3 of this Convention caused doubt whether or not it was meant to apply to certain categories of women other than manual workers. The Court said as follows: “|t]he wording of Art. 3, considered by itself, gives rise to no difficulty; it is general in its terms and free from ambiguity or obscurity.” 21 United Nations, United Nations Conference on the Law of Treaties, Official Records (1968): 178: The controversial character of preparatory work was evident at the Vienna Conference on the Vienna Convention. The US delegate was in favour of the inclusion of the preparatory work equally with the text in determination of the intention of the parties. The opposite view was represented by the UK delegate who found that the preparatory work is often confusing and bear no relation to “the ultimate text of the treaty; unequal, because not all delegates spoke on
158
malgosia fitzmaurice & panos merkouris
practice, the use of supplementary means of interpretation gave rise to many different, often contradictory views. A similar approach was adopted by the Court in Kasikili/Sedudu Island case, in which it referred to preparatory work in confirmation of the underlying reasoning concerning the objective of Article III of the 1890 Treaty.22 The same approach was adopted by the WTO Panel in the Shrimp/Turtle case (see below).23 The use of travaux preparatoires was a subject of a considerable controversy in the 1995 Qatar v. Bahrain case. The Court decided that in this case use of preparatory work, on account of its fragmentary nature, could be only resorted to with caution. The Judgment of the Court was the subject, however, of critical analysis by Judge Schwebel.24 The “circumstances of the conclusion of a treaty”, according to the ILC include both the contemporary circumstances and the historical context in which the treaty was concluded.25 The World Court has on occasion referred to the circumstances of the conclusion of the treaty.26 In the Qatar v. Bahrain case, the Court, did not find that the circumstances of the signing of the Doha Minutes provided any conclusive supplementary means for interpreting the text.27 III. The World Trade Organisation (WTO) and Interpretation of Treaties A. General Issues The relationship between the WTO and general international law has become a point of particular interest after the Uruguay Round when the organization was established.28
particular issue; and partial because it excluded the informal meetings between heads of delegations at which final compromises were reached and which were often the most significant feature of any negotiation”; The French delegate expressed the similar view, at 176. 22 Kasikili/Sedudu Island case, supra note 5, at paras. 45–6: the Court said as follows: “[t]he travaux preparatoires of the Treaty concerning south-west Africa and the Caprivi Strip in particular support this reasoning”. 23 Shrimp/Turtle case, infra note 61. 24 Dissenting Opinion of Judge Schwebel, in the Qatar v. Bahrain case, supra note 18, at 25 et seq. Sir H. Lauterpacht was a proponent of recourse to preparatory work in order to establish the intention of the parties. See Sir Hersch Lauterpacht, ‘Restrictive Interpretation and the Principle of Effectiveness in International Treaties’, BYIL 26 (1949): 48, at 73; see also Lord McNair, supra note 1, at 366. 25 ILC, ‘Commentary’, YILC (1966): Vol. II, at 52. 26 Anglo-Iranian Oil Company case (United Kingdom v. Iran), Preliminary Objections. Judgment of 22 July 1952, 1952 ICJ Rep. 105. 27 Qatar v. Bahrain case, supra note 18, at para. 42. 28 Agreement Establishing the World Trade Organisation in Final Act Embodying the Results of the Uruguay Round of Multilateral Treaty Negotiations, ILM 33 (1994): 1125.
canons of treaty interpretation
159
Unlike the 1947 General Agreement on Trade and Tariffs (GATT), the WTO represents a coherent legal system consisting of several Agreements.29 The WTO system, also established an extensive dispute settlement system. Under the Dispute Settlement Understanding (DSU), a Dispute Settlement Body (DSB) consists of Panels, and an Appellate Body.30 The obvious difference between the GATT panels and the WTO dispute settlement bodies is that the WTO is specifically required to invoke the rules of interpretation of treaties in relation to WTO agreements. According to Article 3.2 of the DSU, all WTO agreements (not only the GATT 1994) are to be interpreted in accordance with the customary rules of interpretation of public international law. Under the GATT 1947, the rules of interpretation of treaties as codified in the VCLT were also applied. However, unlike in the WTO, their application was not institutionalised.31 There are many members of the WTO that are not parties to the 1969 VCLT. However, the Appellate Body of the WTO, declared, as the ICJ did, that the 1969 VCLT represents the codification of the rules of international customary law, and therefore is binding on all parties to the WTO following the jurisprudence of the ICJ.32 B. Selected Case-Studies This section of the study will explore, on the basis of the selected case-law, the principles of treaty interpretation adopted by the WTO. This section will be divided into two parts: the first will analyse the application of the rules of interpretation of the 1969 VCLT (generally); the second will investigate to what extent, if at all, the principles underlying the general structure and the 29 Texts are reproduced in ILM 33 (1994): 1114 et seq.; Agreement establishing the WTO, at 1144; General Agreement on Tariffs and Trade (GATT 1994), at 1154; Several Undertakings on various GATT provisions, at 1156–1166; General Agreement of Trade and Services (GATS), at 1168; Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPs), at 1197; Dispute Settlement Understanding (DSU), at 1226. Numerous Annexed Agreements to the multilateral Agreements in Trade in Goods, such as the Agreement on Agriculture and the Agreement on Subsidies are available at the WTO website: <www.wto.org> (accessed 1 May 2009). 30 Ann. 2 to the 1994 WTO Agreement, ILM 33 (1994): 1226. 31 See Jan Klabbers, ‘Jurisprudence in International Trade Law; Art. XX of GATT’, JWT 26 (1992): 86; The principles of interpretation of treaties were included in the following cases: US – Restrictions on Import of Sugar (22 June 1989), Panel Report, Supp. BISD 36S/331 (for the application of the principle of ordinary meaning as enshrined in Art. 31 (1) of the 1969 VCLT); in the EEC-Regulation on Imports of Parts and Components (16 May 1990), Panel Report BISD 37S/132, (Art. XX para. d of the GATT 1947 was interpreted in accordance with Art. 31 of the 1969 VCLT). 32 Japan – Taxes on Alcoholic Beverages (4 October 1996), Appellate Body Report, WT/DS8/ AB/R, WT/DS10/AB/R; WT/DS11/AB/R/Section D, at 10; US – Standards for Reformulated and Conventional Gasoline (29 April 1996), Appellate Body Report, WT/DS2/AB/R, at 16–7. The case-law of the WTO is available at <www.vvto.org/english/tratop_e/dispu._e/.staply_ e.doc> (accessed 1 May 2009).
160
malgosia fitzmaurice & panos merkouris
legal character of the WTO influence the method of interpretation of treaties adopted by the WTO (specifically). Additionally, the question arises whether specific rules of interpretation, if any such rules exist, are a consequence of the WTO as a potential self-contained regime. In the view of the present authors, the most representative cases concerning the issue of interpretation of treaties are the Reformulated Gasoline, the Shrimp/Turtle33 and the Biotech cases, while other cases such as the US – Continued Zeroing present issues of extreme theoretical value. 1. Reformulated Gasoline Research in this case will be based on the jurisprudence of the Appellate Body (AB) of the WTO. The analysis of the Panel will be included insofar as it was subject to AB’s critical observations. Both Venezuela and Brazil filed a complaint against the United States concerning the effects of rules prescribed under the US Clean Air Act for foreign exported gasoline. The US relied on Article XX of the GATT 1994, as relating to measures aiming at conservation of natural resources pursuant to Article XX (g).34 The case was subject both to Panel35 and AB proceedings.36 According to the AB the Panel’s Report was based on an incorrect application of the rule of interpretation as enshrined in the 1969 VCLT.37 The AB made also a most important statement, the view of the present authors, by expressing: “[t]hat the General Agreement is not to be read in clinical isolation from public international law”.38 It does not, however, mean that the WTO Agreement forms part of general international law. The principle that they should not be read in isolation from public international law does not automatically entail the incorporation of these agreements into general international law in its entirety. What it means is that WTO Agreements follow certain principles of public international law. This however does not exclude the possibility that the WTO Agreements constitute a self-contained regime.
33
Shrimp/Turtle case, infra note 61. Art. XX 1994 GATT: “General Exceptions: Subject to the requirements that such measures are not applied in a manner which would constitute a means of arbitrary or unjustifiable discrimination between countries where the same conditions prevail, or disguised restriction on international trade, nothing in this Agreement shall be construed to prevent the adoption or enforcement by any contracting party of measures: […] (g) relating to the conservation of exhaustible natural resources if such measures are made effective in conjunction with restrictions on domestic production or consumption.” 35 US – Standards for Reformulated and Conventional Gasoline (29 January 1996), Panel Report, WT/DS2/R. 36 Reformulated Gasoline case, AB Report, supra note 32. 37 Id., at 16–7. 38 Id. 34
canons of treaty interpretation
161
First of all, the Panel took the view that clean air was a “natural resource” that could be “depleted”. Along these lines the Panel concluded that a policy to reduce the depletion of clean air was a policy to conserve an exhaustible natural resource within the ambit of Article XX paragraph (g). However, immediately after, the Panel concluded that “the less favourable baseline establishment methods” were not primarily aimed, at the conservation of exhaustible natural resources and thus fell outside the justification provided for by Article XX paragraph (g). The Panel in its interpretation of the phrase ‘relating to’, fell upon the previous case law of the WTO, i.e., the 1987 Herring Salmon case.39 The Panel did not attempt to clarify whether the formulation ‘direct connection’ was being used as a synonym for ‘primarily aimed at’, or whether a new additional element, on top of ‘primarily aimed at’ was demanded. The AB criticised the Panel’s reasoning for posing a question whether the ‘less favourable treatment’ of imported gasoline was ‘primarily aimed at’ the conservation of natural resources, rather than whether the ‘measure’, (the baseline establishment rules), were ‘primarily aimed at’ the conservation of clean air. According to the customary rule of interpretation as codified in Article 31 of the 1969 VCLT, there had to be a finding that the measure provided ‘less favourable treatment’ under Article III paragraph 4, before the Panel examined the ‘General Exceptions’ contained in Article XX. The AB continued by saying that the chapeau of Article XX made it clear that it is the ‘measures’ that have to be examined under Article XX paragraph (g) and not the legal finding of ‘less favourable treatment’.40 The Panel concluded that the baseline establishment rules were not sufficiently proved by the US to be ‘necessary’ under Article XX paragraph (b), since alternative measures either consistent or inconsistent with the General Agreement were reasonably available to the US for achieving the aim of protecting human, animal and plant life.41 According to the AB, the Panel erred again in its interpretation since it appeared to have applied the ‘necessary’ test not only in the examination of the baseline establishment rules under Article XX paragraph (b), but in the course of applying Article XX paragraph (g) as well.42 The main criticism of the Panel’s Report by the AB related to the Panel’s lack of consideration for the basic rule of treaty interpretation, i.e., it did not give proper meaning to the words of the treaty, in this case the GATT, in their context and in the light of the treaty’s object and purpose. 39 Canada – Measures Affecting Exports of Unprocessed Herring and Salmon (22 March 1988), Panel Report, BISD 35S/98, para. 4.4, cited in the Reformulated Gasoline case, Panel Report, supra note 35, para. 6.37. 40 Id. 41 Reformulated Gasoline case, Panel Report, supra note 35, paras. 6.25–6.31; Reformulated Gasoline case, AB Report, supra note 32, at 16. 42 Reformulated Gasoline case, AB Report, supra note 32, at 16.
162
malgosia fitzmaurice & panos merkouris
The AB followed the rule of interpretation included in the 1969 VCLT. It looked at the provision in question (“relating to the conservation of exhaustible natural resources”), in its context and in such a manner as to give effect to the purposes and objects of the 1994 GATT. The AB interpreted the relevant Articles of the GATT 1994: Article XX paragraph (g), Articles I, III and IX as linked together.43 The AB said as follows: [t]he context of Article XX (g) includes the provisions of the rest of the General Agreement, including in particular Articles I, III and IX; conversely, the content of Articles I and III and IX includes Article IX. Accordingly, the phrase ‘relating to the conservation of exhaustible natural resources’ may not be read so expansively as to seriously subvert the purpose and object of Article III: 4. Nor may Article III: 4 be given so broad a reach as effectively to emasculate Article XX (g) and the policies and interests it embodies. The relationship between the affirmative commitments set out in, e.g., Articles I, III and IX, and the policies and interests embodied in the ‘General Exceptions’ listed in Article XX, can be given meaning within the framework of the General Agreement and its object and purpose by a treaty interpreter only on a case-by-case basis, by careful scrutiny of the factual and legal context of given dispute, without disregarding words actually used by the WTO Members themselves to express their intent and purpose.44
In the view of the present authors this is a crucial pronouncement of the AB as to interpretation: i. It interprets the relationship between the relevant provisions in the light of the ‘object and purpose’ of a treaty (the General Agreement) in general, as well as in light of the ‘object and purpose’ of the Articles in question (the principle of effectiveness, see below); ii. it sets a principle, that the rule of interpretation of the ‘object and purpose’ of a treaty can only be accorded meaning on a case-by-case basis. Therefore, there is no single universal formula readily applicable to all instances; iii. from the previous point follows that each and every case will have to be examined within the factual and legal context of the dispute at hand; and iv. each and every case will have to be examined with due consideration accorded to the words actually used by WTO Members themselves to express their intent and purpose. It appears that by adding the formulation “…without disregarding words actually used by the WTO Members themselves to express their intent and purpose”,45 the AB embodied the rule of interpretation contained in Article 31 paragraph (4) VCLT: “A special meaning shall be given to a term if it is established that the parties so intended”. 43 44 45
According to the principle of integration as laid out by Fitzmaurice, supra note 9. Reformulated Gasoline case, AB Report, supra note 32, at 17–8. Id., at 18.
canons of treaty interpretation
163
The AB, as observed above, followed scrupulously the principle of integration (one of the most fundamental principles) as evidenced by the Herring and Salmon case.46 It is against the background of the findings in the Herring and Salmon case that the AB analysed the requirement of ‘primarily aimed at,’ and the legality of the imposition of the baseline and concluded that this requirement was fulfilled. The AB reasoned in a very persuasive manner that the baseline establishment rules, taken as a whole, i.e., baselines for domestic refiners, along with provisions relating to baselines for blenders and importers of gasoline, have to be connected to the ‘non-degradation’ requirements set out in the Gasoline Rule.47 Again, in the implementation of the principle of integration, these provisions cannot be understood if analysed strictly as they stand by themselves, i.e., in separation from the other sections of the Gasoline Rule that constitute the context in which these provisions have to be assessed.48 According to the AB, the Panel made yet another serious error in the application of the rule of interpretation. This failure was in fact the subject of the appeal by the US. It concerned the fact that the Panel did not deal with the issue whether the baseline establishment rules “are made effective in conjunction with restrictions on domestic production or consumption”, since it came to the conclusion that those rules had not even satisfied the preceding requirement, of ‘relating to’ in a sense of being ‘primarily aimed at’ the conservation of clean air. It must be mentioned that there was a disagreement as to the interpretation of the second clause of Article XX paragraph (g). The US submitted that the second clause of Article XX (g) requires that the burdens entailed by the regulation of the level of pollutants in the air emitted in the course of combustion of gasoline, must be imposed solely on, or in respect to, imported gasoline. Venezuela and Brazil, put forward a different argument, based on other panel reports such as in the Herring and Salmon case, that a measure deemed as “made effective in conjunction with restrictions on domestic production or consumption”. a measure must be ‘primarily aimed at’ making effective certain restrictions on domestic production or consumption. These states also claimed that the US failed to prove the imposition of restrictions on domestic production or consumption since clean air was not an exhaustible natural resource within the meaning of Article XX paragraph (g).49 An interesting argument put forward by Venezuela was that to be properly regarded as ‘primarily aimed at’ the conservation of natural resources, the
46 47 48 49
Herring and Salmon case, Panel Report, supra note 39, para. 4.6m. Reformulated Gasoline case, AB Report, supra note 32, at 19. Id. Id.
malgosia fitzmaurice & panos merkouris
164
baseline establishment rules must not only “reflect a conservation purpose” but they must also be shown to have “some positive conservation effect”.50 In order to interpret the above question, the AB once again gave full recognition to the rule of interpretation of the 1969 VCLT. Having applied this rule, the AB came to the conclusion that the meaning of ‘made effective’, when used in connection with a measure, which is a governmental act or a regulation, may be observed to refer to such a measure as being ‘operative,’ meaning of being ‘in force,’ or as having ‘come into effect.’51 It really meant that the clause “if such measures are made effective in conjunction with restrictions on domestic product or consumption” imposes a requirement that the measures in question impose restrictions both on imported and domestic gasoline, in the name of “even-handedness”.52 The AB made an interesting observation: [t]here is, of course, no textual basis for requiring identical treatment of domestic and imported products. Indeed, where there is identity of treatment-constituting real, not merely formal, equality of treatment-it is difficult to see how inconsistency with Article III: 4 would have arisen in the first place.53
However, if no restrictions are imposed on domestic products, but limitations are imposed only on imported ones, the measure cannot be viewed as aimed at conservation but as a discriminatory one against imported goods. This was not the case. The AB, established that the clause: “if made effective in conjunction with restrictions on domestic production or consumption”, was not intended to establish an “empirical effects test” for the availability of the Article XX paragraph (g) exception. The difficulties noted by the AB concerned the determination of causation, and the impossibility of the authoritative statement of the effects of a given measure, after a short period of time.54 The AB also came to the conclusion that, in that particular case, that measure was not designed as a conservation regulation, to start with, therefore was not ‘primarily aimed at’ conservation of natural resources at all. The first step in this complicated interpretation process led the AB to conclude-that baseline establishment rules of the Gasoline Rule fell within the terms of Article XX paragraph (g). The second step was to determine whether those rules also met the requirements of the chapeau of Article XX. Conditions that justify the imposition of the protection of Article XX must be in conformity with its opening clauses. It may be stated that the chapeau addresses not the question of what constitutes a measure but rather the manner in which this measure is imposed. In order to address this issue the AB took into consideration preparatory work
50 51 52 53 54
Id., at 20. Id. Id. Id. Id., at 21.
canons of treaty interpretation
165
(travaux preparatoires). The purpose and object of the chapeau of Article XX was to prevent the abuse of exceptions of what was later to become Article XXI. Therefore, with the assistance of preparatory work, the AB was able to draw the conclusion that the measure in order to be in conformity with Article XX “must be applied reasonably, with due regard both to the legal duties of the party claiming the exception and the legal rights of the parties concerned”.55 The AB followed the rule laid down in Article 32 of the 1969 VCLT (supplementary means of interpretation).56 Although not stated in an explicit manner in the AB’s Report, in the view of the present authors, preparatory work was applied in order to clear the possible obscurity. The AB further conducted interpretation according to the fundamental rule that interpretation must give effect to all terms of the treaty. This means that the interpreter is not free to adopt a reading that would result in reducing whole clauses of a treaty to redundancy or inutility. Applying this principle, the Panel, came to the conclusion that “[t]he provisions of the chapeau cannot logically refer to the standard [s] by which a violation of a substantive rule has been determined to have occurred”.57 The AB rightly observed that the formulation of the chapeau is rather ambiguous. The guidance as to the overall interpretation should be viewed from the point of view of the “general object and purpose of avoiding abuse or illegitimate use of the exceptions to substantive rules available under Article XX”.58 The AB observed that any discrimination for importers would have been avoided if the US had imposed either the same statutory baselines for foreign and domestic gasoline or individual baselines for foreign as well as domestic refiners.59 Finally, the AB established that the US had committed two omissions: a) it had not explored adequate means, including in particular cooperation with both Governments of Venezuela and Brazil; b) it had not taken into consideration the costs for foreign refiners resulting from the imposition of statutory baseline. Detailed interpretation of Article XX, as a whole allowed the AB to conclude that the baseline establishment rules in the gasoline rule constituted an “unjustifiable discrimination” and a “disguised restriction on international trade”.60 Thus, the AB decided that baseline establishment rules, although within the terms of Article XX were not justified and not entitled to the protection under Article XX in its entirety. The AB reached the following conclusions when interpreting the provisions of the GATT 1994 and of the Panel Report: 55 56 57 58 59 60
Id., at 22. Art. 32 VCLT. Reformulated Gasoline case, AB Report, supra note 32, at 22. Id., at 24–5 Id., at 25. Id., at 28.
166
malgosia fitzmaurice & panos merkouris
i. the Panel erred in law as to its conclusion that the baseline establishment rules contained in Part 80 of the Title 40 of the Code of the Federal Regulations did not fall within the terms of Article XX paragraph (g) of the General Agreement; ii. the Panel made an error by not deciding whether the baseline establishment rules contained in Part 80 of Title 40 of the Code of federal Regulations failed to meet the conditions of the chapeau of Article XX of the GATT 1994; iii. the baseline establishment rules contained in Part 80 of the Code of Federal Regulations failed to meet the conditions of the chapeau of Article XX of the GATT 1994 and therefore were not justified under the General Agreement, since they undermined the purpose and object of the introductory clauses of Article XX, which is generally the prevention of abuse of the exceptions of Article XX: iv. the chapeau, by its express terms, does not question the measure or its specific content as such, but rather the manner in which the measure is applied; and finally, v. the AB cautioned against, the application of Article XX exceptions so as to “frustrate or defeat” legal obligations of the holder of rights under the GATT 1994. The foregoing preliminary conclusion may be drawn from the AB Report. The AB followed the provisions of the 1969 VCLT on interpretation; therefore it reached a substantively different result from the Panel, which had not followed these rules. First, it investigated the content of Article XX paragraph (g) against the Gasoline Rule and against the findings of the Panel and then proceeded to interpret the whole of Article XX, in particular its chapeau in conjunction with the purpose and aim of the entire 1994 GATT. The AB took into account all elements of Article 31 of the 1969 VCLT. Therefore, it must be stated that the AB, unlike the Panel, remained firmly rooted in general international law. 2. The Shrimp/Turtle Case From the point of view of substantive issues regarding WTO law, a slightly different solution was adopted in the Shrimp/Turtle case.61 It related to the relationship between multilateral environmental agreements (MEAs) and the 1947 GATT, since this issue arose before in the Tuna/Dolphin cases.62 The
61
US – Import Prohibition of Certain Shrimp and Shrimp Products (15 May 1998), Panel Report, WT/DS58/R; US – Import Prohibition of Certain Shrimp and Shrimp Products (12 October 1998), Appellate Body report, WT/DS58/AB/R. 62 US – Restrictions on Imports of Tuna, (16 June 1994, unadopted), Panel Report, DS/29R, at 842.
canons of treaty interpretation
167
Shrimp/Turtle case is more interesting, however, from the point of view of treaty interpretation. The dispute concerned trade restrictions imposed by the United States on shrimp not caught by harvesting methods using the turtle excluder device (TED). Similarly, like in the Reformulated Gasoline case, the legal problem derived from, inter alia, the interpretation of Article XX of the 1994 GATT, in particular its chapeau and paragraph (g) and in the alternative paragraph (b), in other words from the issue of interpretation of the notion of ‘unjustifiable discrimination.’ On the issue of the interpretation of Section 609 of the US Law in the light of Article XX, the Panel said as follows: … [w]e are of the opinion that the chapeau of Article XX, interpreted within its context and in the light of the object and purpose of GATT and of the WTO Agreement, only allows Members to derogate from GATT multilateral provisions so long as, in doing so, they do not undermine the WTO multilateral trading system, thus also abusing the exceptions contained in Article XX. Such undermining and abuse would occur when a Member jeopardises the operation of the WTO Agreement in such a way that guaranteed access and non-discriminatory treatment within multilateral framework would no longer be possible. […] We consequently find that when considering a measure under Article XX, we must determine not only whether the measure on its own undermines the WTO multilateral trading system, but also whether such type of measure, if it were adopted by other Members, would threaten the security and predictability of the multilateral trading system.63
The Panel continued ..it appears to us that, in light of the context of the term ‘unjustifiable’ and the object and purpose of the WTO Agreement, the measure at issue constitutes unjustifiable discrimination between countries where the same conditions prevail and thus is not within the scope of measures permitted under Article XX.64
Finally, the Panel concluded; “[w]e therefore find that the US measure at issue is not within the scope of measures permitted under the chapeau of Article XX”.65 (a) Issues Relevant to the US Appeal The main issues raised by the US in its appeal were as follows: i. erroneous interpretation of Article XX, the interpretation of which was not based on ordinary meaning and context;
63 64 65
Shrimp/Turtle case, Panel Report, supra note 61, para. 7. 44. Id., para. 7.49. Id., para. 7.62.
168
malgosia fitzmaurice & panos merkouris
ii. unfounded interpretation of the chapeau of Article XX as requiring the Panel to determine whether a measure constitutes a ‘threat to multilateral trading system’. That interpretation would add an entirely new obligation to existing ones under Article XX; iii. the US also submitted that the Panel erred on yet another element of the rule of interpretation because it failed to take into account the context of the term ‘unjustifiable discrimination’; and finally, iv. the Panel was incorrect in applying the object and purpose of the WTO Agreement in interpreting Article XX.
(b) The Arguments of Joint Appellees The Appellees put forward the following arguments: i. they stressed that the illegality of the unilateral measures imposed by the US constituted an abuse of Article XX; ii. they challenged the US interpretation of the chapeau term ‘unjustifiable’ in the light of the object and purpose of the Article XX of the 1994 GATT and the object and purpose of the WTO agreement; iii. they supported the findings of the Panel, as identifying the purpose and object of the chapeau as the prevention of the abuse of Article XX exceptions, and associating the prevention of such abuse with the preservation of the multilateral trading system. As it is often in treaty interpretation, the notion of the object and purpose of the treaty was the subject of diverse interpretation in this case. In its submissions, Malaysia, inter alia, maintained that the Appellant misconceived the Panel’s findings in that: the Panel did not in any way allude to the supremacy of trade concerns over nontrade, and did not fail to recognise that most treaties have no single, undiluted object and purpose but a variety of different objects and purposes.66
Another issue that in a similar fashion caused divergence of views was the relationship of the measures adopted by the US in Section 609 of the US Law with the provisions of the chapeau of Article XX. Australia submitted that the Panel failed to interpret the terms of the chapeau of Article XX “in accordance with customary rules of interpretation of public international law, in particular with its ordinary meaning and in context”.67 The same country stated that the Panel made an error in that it examined whether Section 609 met the requirements of the chapeau prior to analysing 66 67
Shrimp/Turtle case, AB Report, supra note 61, para. 47. Id., para. 55.
canons of treaty interpretation
169
whether it met the requirements of any of the paragraphs of Article XX. According to Australia, the incorrect interpretation by the Panel of the term ‘unjustifiable discrimination’ resulted from the erroneous interpretation of the ‘object and purpose’ of the WTO Agreement in construing the GATT 1994. In general, it may be said that the interpretative approach adopted by the Appellate Body in the Reformulated Gasoline case, i.e. that the legality of each and every measure has to be assessed first in the light of the relevant paragraph of Article XX (in most cases (g) ) and secondly in the light of the chapeau of Article XX, met with the agreement of most of the parties, such as the EU.68 The EU also stressed the importance of the test of the purpose and object. It focused its analysis of the interpretation of Article XX on the presumption that the general aim of this Article should not be construed in such a manner so that trade concerns always prevail over non-trade ones, such as environmental and health concerns. The EU was also of the view that each and every case should be judged on its own merits. The general test as applied by the Panel, i.e. whether a measure threatens the security and predictability of the multilateral trading system and elevates trade concerns above all others, is inconsistent with the object and purpose of the WTO Agreement. In cases where a WTO member, on its own, takes a ‘reasonable’ measure with the aim of protecting and preserving a particular global environmental resource, such a measure would have to be consistent with general international law. A member first of all has to prove that such a measure is ‘reasonable’, i.e. no more restrictive than required to protect shared environmental resources and relates directly to an environmental objective. (c) The Appellate Body Analysis of the Panel’s Findings First of all the AB observed that the Panel, when applying the “customary international rules of interpretation of public international law”, did not follow all the steps required by Article 3.2 of the DSU. The Panel erred in that it did not examine expressly the ordinary meaning of the words of Article XX. The Panel did not analyse specifically how the application of Section 609 constitutes “a means of arbitrary or unjustifiable discrimination between countries where the same conditions prevail, or a disguised restriction on international trade”. The Panel focused instead on the design of the measure itself. The design of the measure however should be analysed against the paragraphs of Article XX. The Panel failed to analyse the immediate context of the chapeau, i.e., paragraphs (a) to (j) of Article XX. Further, the Panel erred as to the object and purpose test, i.e. it did not look into the object and purpose of the chapeau of Article XX, but into the whole
68
Id., para. 69.
170
malgosia fitzmaurice & panos merkouris
object and purpose of the GATT 1994 and the WTO Agreement, “which object and purpose is described in an overly broad manner”.69 This interpretative error resulted in the Panel’s very broad formulation that the measures in question, which according to the Panel undermined the WTO multilateral trading system, were against the scope of permissible measures under the chapeau of Article XX. The AB made a very important observation as to the interpretative value of the duty to maintain (not to undermine) the general multilateral trading system. It referred to its findings in the United States-Gasoline case in which it stated that it is important to underscore that the object and purpose of the introductory clauses of Article XX is generally the prevention of abuse of the exceptions of Article XX. However, the Panel ignored in its interpretation the manner in which the measure in question had been applied as to constitute abuse or misuse of a given kind of exception and instead applied this general interpretative directive mechanically. The AB observed that it is almost impossible to identify abuse or misuse of specific provisions of Article XX upon interpreting the chapeau of Article XX prior to the examination of the specific exceptions threatened with abuse. Furthermore, the standards established by the chapeau are very broad in scope, therefore when applied in a particular case the content of these standards will vary, as the kind of measure under analysis varies. For example, the AB said that the standard of “arbitrary discrimination” may be “different for a measure that purports to be necessary to protect public morals than for one relating to the products of prison labour”.70 In order to reverse the Panel’s findings, the AB proceeded with the twotiered analysis of the imposition of the measure. The US invoked primarily Article XX paragraph (g), and alternatively (b). The AB began with the analysis of the compatibility (or incompatibility) of the measure under paragraph (g), with the interpretation of the notion of ‘exhaustible natural resources’. This was not fully explored by the Panel, who adopted the incorrect sequence of interpretation: the ‘chapeau-down’ approach. It. must also be taken into consideration that the wording of Article XX was drafted more that 50 years ago; and that at present the aim of sustainable development is supported by the WTO Agreement as evidenced by its Preamble.71 The AB, made yet another fundamental statement: it said that from the perspective embodied in the Preamble, the generic notion of ‘natural resources’ in Article XX paragraph (g)
69
Id., para. 116. Shrimp/Turtle case, AB Report, supra note 61, para. 120. 71 “[t]he parties to this Agreement, Recognizing that their relations in the field of trade and economic endeavor […] seeking both to protect and preserve the environment and to enhance the means for doing so in a manner consistent with their respective needs and different levels of economic development…” 70
canons of treaty interpretation
171
is not static in this context or reference but rather “by definition, evolutionary”.72 The AB, in accordance with the principle of effectiveness, found that measures to conserve exhaustible natural resources, whether living or nonliving, may fall within Article XX paragraph (g). Further, the AB found that sea turtles as species are a living exhaustible resource of a migratory type. Next, the AB, had to establish whether the second part of Article XX paragraph (g): ‘relating to’, was applicable in this case. The AB stated that it was proven that there was a ‘relationship between the general structure and design of the measure at hand, Section 609, and the policy goal it purports to serve that is the conservation of sea turtles.’73 The next interpretative task was to analyse “if such measures are made in conjunction with restrictions on domestic production or consumption”. Having examined the domestic legislation, the AB decided that this requirement was also fulfilled. Further, the AB proceeded to analyse the general provisions of the chapeau of Article XX. Contrary, to what was pleaded by the US, the AB came to the conclusion that the policy goal of a measure cannot serve as a rationale or justification under the standards of the chapeau XX. According to the principles of interpretation, in order for the policy measure to be justified under the chapeau of Article XX, it has to be first justified under Article XX paragraph (g). However, it does not follow that a measure that is justified under Article XX (g), fulfils automatically the requirements of the chapeau. Next step in the interpretation was to examine the ordinary meaning of the words of the chapeau. The interpretation of these provisions indicates that there are three standards in the chapeau: first, arbitrary discrimination between countries where the same conditions prevail; second, unjustifiable discrimination between countries where the same conditions prevail; and third, a disguised restriction on international trade. Therefore, for the measure to be applied in a manner that would constitute “arbitrary or unjustifiable discrimination between countries where the same conditions prevail”, all these elements must exist. It is obvious from the above analysis that the standards contained in the chapeau of Article XX are different from those embodied in Article XX paragraph (g) and Article XI: I of the GATT 1994. 72 The AB adopted the approach taken by the ICJ in Legal Consequences for States of the Continued Presence of South Africa in Namibia (South West Africa) notwithstanding Security Council Res. 276 (1970), Advisory Opinion of 21 June 1971, 1971 ICJ Rep. 16, at 31, that since the concepts embodied in a treaty are evolutionary, their interpretation has to confirm with the subsequent development of law. This approach was confirmed by the ICJ in the GabcikovoNagymaros Project (Hungary v. Slovakia), Judgment of 25 September 1997, 1997 ICJ Rep. 7, at para. 112, in which the Court said that “Consequently, the Treaty is not static, and is open to adapt to emerging norms of international law”. See also Gabrielle Marceau, ‘A Call for Coherence in International Law. Praises for the Prohibition Against ‘Clinical Isolation’ in WTO Dispute Settlement’, JWT 33 (1999): 120–3. This author calls this approach ‘evolutive approach.’ 73 Shrimp/Turtle case, AB Report, supra note 61, para. 137.
172
malgosia fitzmaurice & panos merkouris
The AB stressed further that the preamble of the WTO Agreement clearly indicates that world resources should be used in accordance with the objective of sustainable development. Therefore, interpretation of the agreements annexed to the WTO Agreement (including the GATT 1994) must be influenced by this aim. The factors relevant to the interpretation of the chapeau of Article XX, according to the AB, are therefore the following: the ordinary meaning of the chapeau of Article XX, in light of its object and purpose and as part of the context of the chapeau, must take into consideration, the specific language of the preamble to the WTO Agreement.74 As to the chapeau of Article XX, according to the AB, it embodies the recognition on the part of WTO members of a requirement to maintain the balance of rights and obligations between the right of a Member to invoke one of the exceptions enumerated in Article XX, on one hand, and the substantive rights of other Members under the GATT 1994, on the other hand. Interpretation of the language of the chapeau makes it quite clear that all exceptions of Article XX are limited and conditional exceptions of the substantive obligations contained in other provisions of the GATT 1994. Thus “the ultimate availability of the exception is subject to the compliance by the invoking Member with the requirement of the chapeau”.75 This interpretation of the chapeau was confirmed by the preparatory work (the negotiating history) to Article XX of the GATT 1947. The formulation of the chapeau of Article XX expresses the principle of good faith, which is a general principle of law and at the same time a general principle of international law. As the AB observed one of the applications of this principle is the doctrine of abus de droit. The AB observed that this principle prohibits the abusive exercise of states’ rights, which must be carried out reasonably. Therefore, the exercise of rights by a Member that results in abuse amounts to treaty violation. The AB in its interpretation not only interpreted the language of the chapeau, but sought additional guidance, as appropriate, from the ‘general principles of international law’,76 according to Article 31 paragraph 3 (c) of the 1969 VCLT. In the view of the present writers, one of the most interesting features, of the interpretation techniques applied by the AB in this case is that while interpreting the treaty it took into account, as a guiding principle, not. only the actual text, but also the substance of rights and duties of Members under the 1994 GATT. The AB stated explicitly that the task of interpreting and applying the chapeau is fundamentally one of locating and marking the line of equilibrium 74 75 76
Shrimp/Turtle case, AB Report, supra note 61, para. 156. Id., para. 157. Id., para. 159.
canons of treaty interpretation
173
between the right of a Member to invoke an exception under Article XX and the rights of other Members stemming from substantive provisions of the GATT 1994. Further, the AB observed that the location of the line of equilibrium, as enshrined in the chapeau of Article XX, is not fixed and static. This line moves according to the variations of the kind and the shape of the measures at stake and according to the differences in facts of each and every case. It was in the light of the above interpretation that the AB analysed the measure adopted by the US. Thus, although the measure itself, as was proved, fell within the ambit of Article XX paragraph (g), it might nonetheless constitute “a means of arbitrary or unjustifiable discrimination between countries where the same conditions prevail”, or “a disguised restriction on international trade”. The task before the AB was therefore to determine whether the application of a measure constituted an abuse or misuse of the provisional justification made available under Article XX paragraph (g). The AB looked both at the substantive and procedural aspects of the measure. First, the AB scrutinised whether Section 609 had been applied in a fashion amounting to “unjustifiable discrimination between the countries where the same conditions prevail”. The application of the measure by the US was flawed in that it was intended to have a coercive effect on specific policy decisions made by foreign governments. The AB considered this policy measure irreconcilable with the declared policy objective of conservation of sea turtles and viewed this as an attempt to influence WTO members to adopt the same comprehensive regulatory regime as that applied in the US, without any inquiry into the appropriateness of the regulatory programme for the conditions in those exporting countries. Most important, in the view of the present writers, was the observation made by the AB that the US failed to engage in serious across-the-border negotiations with the objective of concluding bilateral or multilateral treaties for the protection and conservation of sea turtles, before enforcing the import prohibitions against the shrimp exports of those other Members.77
This was also a finding of the Panel.78 The effect of this was, according to the AB. discriminatory and unjustifiable. The failure to enter into meaningful negotiations may be seen as unilateralism in the application of Section 609. Other important evidence of ‘unjustifiable discrimination’ was a different period of time allowed for different countries to implement the TEDs. Next, the AB examined whether Section 609 had been applied in a manner constituting “arbitrary discrimination between countries where the same conditions prevail”. 77 78
Id., para. 166. Shrimp/Turtle case, Panel Report, supra note 61, para. 7.56.
174
malgosia fitzmaurice & panos merkouris
Due to the difference in treatment of various Members as regards the grant of certifications, the AB found that the United States measure is applied in a manner which amounts to a means not just of ‘unjustifiable discrimination’, but also of ‘arbitrary discrimination’, contrary to the requirements of the chapeau of Article XX.79
In conclusion of this legally very elaborate interpretation of Article XX, the AB decided that although the measure of the US served the environmental objective and was recognised as legitimate under Article XX paragraph (g) of the GATT 1994, the application of this measure by the US was effected in a manner which constituted arbitrary and unjustifiable discrimination between Members of the WTO. Therefore, the measure was not in conformity with the chapeau of Article XX. The Members of the WTO are entitled to adopt their own policies aimed at protecting of the environment, as long as in doing this they fulfill their obligations and respect the rights of other Members deriving from the WTO Agreement. In general, as Cameron and Gray emphasise, “[t]he link between treaty interpretation and requirements stated in Article 3.2 of the DSU is now entrenched in WTO law”.80 Failure of the WTO Panel to observe the canons of treaty interpretation results in overturned rulings as evidenced, inter alia, in the above cases. Therefore, the correct process of interpretation consists of the following steps: 1. Analysis of the text of the provision; 2. Recourse to the test of the object and purpose of the treaty, in cases where the meaning of the text is unclear, or does not lead to conclusive results. The Panel’s interpretation in the Shrimp/Turtle case was subject to severe criticism by the AB due to its incorrect sequence of interpretative steps taken: the determination of the object and purpose of the chapeau first, prior to the examination of the relevant provisions of Article XX, therefore rendering the provision absurd. The Panel erred also by examining only the design of the US measure prohibiting shrimp imports and not its application. Marceau makes an interesting observation as to the scope of the applied principles of interpretation by the WTO. She submits that when interpreting the WTO Agreement panels and ABs are obliged to take into account canons of interpretation included in other sources of international law: custom and general principles of international law, not only the 1969 VCLT.81 79
Shrimp/Turtle case, AB Report, supra note 61, para. 184. James Cameron and Kevin R. Gray, ‘Principles of International Law in the WTO Dispute Settlement Body’, ICLQ 50 (2001): 256. 81 Marceau, supra note 72, at 116. 80
canons of treaty interpretation
175
Two further cases within the WTO system, expound on the issue of ‘object and purpose’ and its correct application. In US – Stainless Steel the Panel held [w]e note that the term ‘object and purpose’ in Article 31(1) is preceded by ‘its’ whereas the term ‘context’ is preceded by ‘their’. Thus, we consider that Article 31(1) refers to the object and purpose of the treaty as a whole, rather then specific provisions thereof. Had drafters intended to refer to the object and purpose of specific provisions, they would have used ‘their’, not ‘its’, before ‘object and purpose’, as in the case of the word ‘context’.82
but felt it necessary to include in a footnote that, despite the fact that Article 31 referes to the object and purpose of a treaty as a whole, this did “not preclude the interpreter from taking into account the object and purpose of specific treaty provisions, where warranted”.83 This had already been established in the EC – Chicken Classification by the Appellate Body, which stated we do not believe that Article 31(1) excludes taking into account the object and purpose of particular treaty terms, if doing so assists the interpreter in determining the treaty’s object and purpose on the whole. We do not see why it would be necessary to divorce a treaty’s object and purpose from the object and purpose of specific treaty provisions, or vice versa. To the extent that one can speak of the ‘object and purpose of a treaty provision’, it will be informed by, and will be in consonance with, the object and purpose of the entire treaty of which it is but a component.84
In summary, the procedure, which these two cases outline, is that the judges, first and foremost, have to examine the object and purpose of a treaty as a whole. The objects and purposes of specific provisions can also be taken into account as long as they assist the judges in the interpretative process. However, WTO law cannot be interpreted only by reference to the ‘object and purpose’ of certain provisions in isolation from or to the exclusion of ‘the object and purpose’ of the entire treaty.85 3. The Biotech Case86 This case is interesting as it concerns Article 31(3)(c) of the 1969 VCLT, which was called “one of the most neglected corners of the interpretation section of 82 United States – Final Anti-Dumping Measures of on Stainless Steel from Mexico (20 December 2007), Panel Report, WT/DS334/R, para. 6.16; similarly, see European Communities – Customs Classification of Frozen Boneless Chicken Cuts (27 September 2005), Appellate Body Report, WT/DS269/AB/R & WT/DS286/AB/R, para. 238; and Sinclair, supra note 10, 130–5. 83 Id., para. 6.16, FN 6. 84 EC – Chicken Classification, Appellate Body Report, supra note 82, para. 238. 85 Id., para. 239. 86 European Communities-Measures Affecting the Approval and Marketing of Biotech Products, Interim Reports of the Panel, 7 February 2006, WT/DS291/INTERIM: WT/DS292/INTERIM; WT/DS293/INTERIM. The Report was secret, but was leaked and made public by the Friends of Earth Europe, which provided the leaked report. Available at <www.trade-environment.org/ page/theme/tewto/biotechcase.htm> (accessed 1 May 2009); see also CIEL, ‘EC-Biotech: Overview
176
malgosia fitzmaurice & panos merkouris
the Vienna Convention”.87 This rule of interpretation gained importance due to the work of the International Law Commission (the ‘ILC’) Study Group on Fragmentation of International Law: Difficulties Arising from the Diversification and expansion of International Law.88 Briefly speaking the subject matter of the case concerned the challenge to three types of measures adopted by the EC and EC member States: an alleged moratorium on approval of biotech products; product specific measures related to the approval of biotech products; and several EC members’ measures relating to the import and/or marketing of specific biotech products.89 The Panel made several observations in this case as regards the understanding of the content of Article 31(3)(c). First of all the Panel defined what it meant by the scope of the ‘rules of international law’ are referred to in this Article. The Panel reached the conclusion that customary law, international treaties, as well as general principles of international law (following the jurisprudence of the Appellate Body US-Shrimp case), should be taken into consideration. Therefore
and Analysis of the Panel’s Interim Report’ (2006), <www.ciel.org/Publications/EC_Biotech_ Mar06.pdf> (accessed on 1 May 2009). 87 Campbell McLachlan, ‘The Principle of Systemic Integration and Art. 31(3)(c) of the Vienna Convention’, ICLQ 54 (2005): 279; Art. 31(3)(c): “There shall be taken into account, together with the context: … any relevant rules of international law applicable in the relations between the parties”; See also Duncan French, ‘Treaty Interpretation and the Incorporation of Extraneous Legal Rules’, ICLQ 55 (2006): 281–314; Panos Merkouris, ‘Debating the Ouroboros of International Law: The Drafting History of Article 31(3)(c)’, International Community Law Review 9 (2007): 1–31; Joost Pauwelyn, ‘Bridging Fragmentation and Unity: International Law as a Universe of Inter-Connected Islands’, Michigan Journal of International Law 25 (2004): 903 et seq.; and the Comment on this essay by J. Meltzer, at 917 et seq. and Pauwelyn’s reply to Meltzer’s comments, at 924–7; see also Joost Pauwelyn, Conflict of Norms in Public International Law (Cambridge: CUP, 2003), 203. Pauwelyn’s general view is that the Appellate Body should consider other rules of international law in the interpretation of WTO agreements when the law reflects the ‘common intentions’ of the parties to the WTO. He argues that it does not mean that all the parties to the WTO treaty must formally and explicitly agreed, one after the other to the new non-WTO rule; nor even that this rule must otherwise legally bind all WTO members; but rather, that this new rule can be said to be at least implicitly accepted or tolerated by all WTO members, in the sense that it can reasonably be considered to express the common intentions or understanding of all members as to the meaning of the WTO term concerned.
Joost Pauwelyn, ‘The Role of Public International Law in the WTO: How Far Can We Go?’ AJIL 95 (2001): 575–6; finally, see Isabelle van Damme, Treaty Interpretation by the WTO Appellate Body (Oxford: OUP, 2009), at 355–76; van Damme argues that Article 31(3)(c) is of limited value and its importance should not be exaggerated. 88 ILC Study Group on Fragmentation, Fragmentation of International Law: Difficulties Arising from the Diversification and Expansion of International Law – Conclusions of the Study Group (18 July 2006) (A/CN.4/L702); ILC Study Group on Fragmentation, Report of the Study Group, (2 May 2006) (A/CN.4/L682/Add.l); ILC Study Group on Fragmentation, Fragmentation of International Law: Difficulties Arising from the Diversification and Expansion of International Law, Report of the Study Group, finalized by Martti Koskenniemi, (13 April 2006) (A/6N.4/L.682). 89 See CIEL, supra note 86, at 9.
canons of treaty interpretation
177
we would agree that if the precautionary principle is a general principle of international law, it could be considered a ‘rule of international law’ within the meaning of Article 31(3)(c).90
The term ‘parties’ as understood by the Panel lead it to the following conclusion: the rules of international law to be taken into account in interpreting the WTO agreements at issue in this dispute are those which are applicable in the relations between the WTO members.91
This understanding of the applicable rules by the Panel differed from that of the EU. The EU suggested that the Panel must interpret the WTO agreements at issue in this case in the light of other rules of international law “even if these rules are not binding on all parties to the dispute”.92 However, the panel was of the view that such an interpretation would violate States’ sovereignty as it would mean that a treaty to which a State is a party might be affected by the interpretation of another treaty to which it is not a party.93 The Panel also stated that requiring that a treaty be interpreted in the light of other rules of international law which bind the States parties to the treaty ensures or enhances the consistency of the rales of international law applicable to these States and thus contributes to avoiding conflicts between the relevant rules.94
The Panel took into consideration the following treaties: the Convention on Biological Diversity (CBD) and the Biosafety Protocol. It also analysed the relevance of the precautionary principle. As to the CBD, the Panel took the view that since the United States is not a party to it, it is not applicable in the relations between the US and all other WTO members and therefore the Panel rejected the argument of the EU that it should be required to take account of this Convention in interpreting the multilateral WTO agreements relevant in this case.95 The same argument was put forward in relation to the Biosafety Protocol. As Argentina, Canada and the US did not ratify it, it cannot be taken into account for the purposes of Article 31(3)(c) interpretation.96 The Panel also rejected the argument that by reason of the mere signing of a treaty (according to Article 18 of the 1969 VCLT) a state must refrain from acts which would defeat object and purpose its. It argued that ‘the object and purpose of a treaty’ cannot be reasonably considered to constitute a “rale of international law”.97 90 91 92 93 94 95 96 97
Supra note 86, para, 7.67. Id., para. 7.68. Id., para. 7.7.71. Id., para. 7.71. Id., para. 7.69. Id., para. 7.74. Id., para. 7.75. Id., para. 7.74 (and note 200).
178
malgosia fitzmaurice & panos merkouris
The application of the precautionary principle led to a heated debate as to its character. The EU strongly supported the view that it already has become a “fully-fledged” general principle of international law;98 whilst the US strongly advocated the view that precaution has not become a rule of international law and therefore a fortiori could not become a rule of customary international law.99 The Panel relied on the AB Report in the EC-Hormones decision, which was inconclusive as to the legal status of the precautionary principle. The Panel therefore stated as follows: [s]ince the legal status of the precautionary principle remains unsettled, like the Appellate Body before us, we consider that prudence suggests that we not attempt to resolve this complex issue, particularly if it is not necessary to do so.100
The Panel’s interpretation of Article 31(3)(c) was criticised in the following terms: [while it] may not be manifestly wrong, it does not contribute to building channels of dialogue in an increasingly fragmented international legal system. The Panel’s apparent attempt to avoid conflicts between relevant rules of international law lead it to conclude that the Vienna Convention did not establish a legal obligation for interpreting bodies to take into account treaties that were not ratified by all parties to the treaty being interpreted. However, this conclusion stands at odds with the responsibility of an interpreting body to take into account those treaties, especially when they address issues of global concern where the interests of the international community are involved.101
98
Id., para. 7.78. Id., para. 7.82. 100 Id., para. 7.7 note 82, para. 7.89. It has to be added that the Panel also considered in interpreting WTO agreements, rules of international law that are not applicable in the relations between the WTO members and therefore are not covered by the category of rules under Art. 31(3)(c). The Panel referred to the EC argument in the US-Shrimp case, the Appellate Body interpreted the WTO rules by reference to treaties not binding on all parties to the proceedings, including the CBD. The Panel reached the conclusion that it could consider such rules if it deemed such rules to informative, which proved to be unnecessary. The Panel relied on Art. 31(1) of the 1969 VCLT, on the rule that a treaty must be interpreted in accordance with the ordinary meaning to be given to the words in their context and in light of its object and purpose. It decided that in this case, it was not necessary to rely on the particular provisions of the CBD and the Biosafety Protocol invoked by the EC in interpreting the WTO agreements at issue in this case. See paras. 7.91–7.93 and 7.95. 101 See CIEL, supra note 86, at 49. Pauwelyn says as follows: 99
[n]otwithstanding the limited jurisdiction of the WTO panels (WTO claims only), it has become standard practice for WTO panel and the Appellate Body to use non-WTO law when interpreting the meaning of the terms in the WTO agreement. Such interpretation can, for example, lead to broader GATT exceptions, as in US-Shrimp, where the Appellate Body interpreted the words ‘exhaustible natural resources’ in GATT Art. XX (g) with reference to certain environmental treaties. It may also narrow the scope of GATT rules or exceptions, as shown in the Oil Platforms case, where a treaty provision similar to GATT Art. XX (b) (iii) on essential security interests was interpreted restrictively with reference to rules of general international law prohibiting the use of force.
See Pauwelyn, supra note 87, at 913–4.
canons of treaty interpretation
179
The problem, which arose in the above case was foreseen by McLachlan in his excellent article. He identified the problem as encountered by the WTO Panel as follows: …it is inherently unlikely that there will be a precise congruence in the identity of the parties to two treaties. If complete identity of the parties were required before the other patty could be regarded as being ‘applicable in the relations between the parties’, it would have the ironic effect that the more membership of a particular treaty such as the WTO Covered Agreements expanded, the more treaties would be cut off from the rest of international law.102
The author of the article puts forward four possible solutions, out of which the solution adopted by the Panel in this case falls into the category of “clear but very narrow standard”.103 Article 31(3)(c) represents a more general principle of treaty interpretation: so-called systemic integration within the international law system,104 according to which, whatever their subject matter, “treaties are a creation of the international legal system and their operation is predicated upon this fact”.105 The task of this principle (which was not included by Fitzmaurice in his rules of treaty interpretation, see above), is to find a workable solution as regards the fragmentation of international law.106 According to the ILC: systemic integration governs all treaty interpretation, the other relevant aspects of which are set out in the other paragraphs of Articles 31–32 VCLT… Article 31(3)(c) deals with the case where material sources external to the treaty are relevant in its interpretation. These may include other treaties, customary rules or general principles of law.107
IV. Other Points of Treaty Interpretation in the WTO A. The Principle of Effectiveness The principle of effectiveness (ut res magis valeat quam pereat) was one of the principles applied by the GATT 1947. The ILC during its codification of the 102
McLachlan, supra note. 87, at 314. Id. 104 McLachlan, supra note 87, at 280: see also the ILC Report of 18 July 2006, supra note 88, on the same subject, at 13–7. See also Oil Platforms case (Iran v. United States of America), Merits, Judgment of 6 November 2003, 2003 ICJ Rep. 161, para 41. 105 The ILC Report of 18 July 2006, supra note 88, on the same subject, at 14. 106 McLachlan, supra note 87, at 318. 107 the ILC Report of 18 July 2006, supra note 88, at 14–5. The ILC stated that 103
where a treaty functions in the context of other agreements, the objective of systemic integration will apply as a presumption with both positive and negative effects: (a) the parties are taken to refer to customary international law and general principles of law for all questions which treaty does not itself resolve in express terms; (b) in entering into treaty obligations, the parties do not intend to act inconsistently with generally recognized principles of international law.
180
malgosia fitzmaurice & panos merkouris
law of treaties recognised this principle as useful in the event when a treaty is open to two interpretations and one of them deprives the treaty of any effect, then good faith and the object and purpose of the treaty require that the effective interpretation should be applied.108 However, this principle of interpretation is one of the most controversial and difficult to implement. First of all, some writers suggest that there may be two separate rules under the label ‘principle of effectiveness’. The first is the rule that all provisions of the treaty or other instrument must be supposed to have been intended to have significance and to be indispensable to convey the intended meaning; consequently, an interpretation that does not render the required effects is not correct. The second is the rule that the instrument as a whole, and each of its provisions, must be taken as intended to achieve some result; and that an interpretation which would disable the text to achieve this objective is similarly incorrect.109 The second rule of effectiveness is similar to the ‘object and purpose’ criterion and therefore is akin to the teleological school of interpretation, based on the notion of the ‘object and purpose’ of the treaty. It has to be applied with a degree of caution, since [t]he teleological approach, in some of its more extreme forms, will even deny the relevance of the intention of the parties; it is in effect based on the concept that, whatever the intentions of the parties may have been, the convention as framed has a certain object and purpose, and the task of the interpreter is to ascertain this object and purpose rather than interpret the treaty as to give effect to it.110
The view was also expressed that …in so far as the rule of effectiveness is identical with the principle of good faith it has good justification of its own and cannot be regarded as a technical or artificial rule of construction. But it is for that very reason that it must not deliberately be allowed to assume an existence independent of intention, express or legitimately implied, of the parties.111
The jurisprudence of the WTO has applied the principle of effectiveness without specific differentiation. The use of this principle in, for example, JapanTaxes on Alcoholic Beverages case, indicates that the AB applied this principle as a rule of interpretation in the first of the above meanings. It stated as follows: “[a] fundamental tenet of treaty interpretation flowing from the general rule of interpretation set out in Article 31 is the principle of
108 ILC, Report of the ILC on the Work of its 18th Session: Draft Articles on the Law of Treaties with Commentaries, (A/6309/Rev.1), reproduced in YILC (1966), Vol. II: 218, at 219. 109 Thirlway, supra note 11, at 44. 110 Sir Ian Sinclair, supra note 10, at 131. 111 Lauterpacht, supra note 24, at 74.
canons of treaty interpretation
181
effectiveness (ut res magis valeat quam pereat)”. In United States-Standards for Reformulated and Conventional Gasoline, we noted that [o]ne of the corollaries of the ‘general rule of interpretation’ in the Vienna Convention is that interpretation must give meaning and effect to all terms of the treaty. An interpreter is not free to adopt a reading that would result in reducing the whole clauses or paragraphs of a treaty to redundancy or inutility.
The above analysis of the jurisprudence of the WTO indicates a frequent application of the interpretative rule of effectiveness in the meaning of the object and purpose (usually the object and purpose of the 1994 GATT). In this meaning it was most clearly applied in the above-mentioned Japan-Tax case. In relation to the interpretation of Article III:1 of the GATT 1994, the AB said as follows: [t]he terms of Article III must be given their ordinary meaning—in their context and in the light of their overall object and purpose of the WTO Agreement. Thus, the words actually used in the Article provide the basis for an interpretation that must give meaning and effect to all its terms. […] Any other reading of Article III would have the effect of rendering the words of Article III: 1 meaningless, thereby violating the fundamental principle of effectiveness in treaty interpretation.112
Cameron and Gray expressed the opinion that the preferred interpretative approach of the DSB within the WTO system of the principle of effectiveness is one that analyses provisions within the context of the whole WTO treaty system. Such an application “challenges the notion of lex specialis, where each agreement could operate in isolation from each other”.113 The present authors agree with the first part of this statement; the second part however, concerning lex specialis, may be disputed. The WTO constitutes a whole, unitary nexus of interrelated obligations, represented by agreements within the network of one legal system. Therefore, particular specific agreements of the WTO cannot be lex specialis in relation to other agreements of the same system. The whole of the WTO system may be considered to be lex specialis in relation to the general system of international law, but its particular components are not lex specialis towards each other. Finally, before bringing this Section of our analysis to a closure we have to examine a most recent use of the principle of effectiveness which occurred in US – Continued Zeroing. The importance of this case cannot be stressed enough because it attempted to clarify the meaning of an extremely
112 113
Japan – Taxes on Alcoholic Beverages Appellate Body Report, supra note 32, Section G. Cameron and Gray, supra note 80, at 257.
182
malgosia fitzmaurice & panos merkouris
problematic provision, that of Article 17.6(ii) of the 1994 Anti-Dumping Agreement which states the following: the panel shall interpret the relevant provisions of the Agreement in accordance with customary rules of interpretation of public international law. Where the panel finds that a relevant provision of the Agreement admits of more than one permissible interpretation, the panel shall find the authorities’ measure to be in conformity with the Agreement if it rests upon one of those permissible interpretations.114
The relationship between this provision and that of DSU Article 3.2 has always been a problematic one. In this case the AB attempted to elucidate how this standard should be applied and in doing so reiterated some basic concepts of the interpretative process, as enshrined in Articles 31 and 32 of the VCLT. The principles of interpretation that are set out in Articles 31 and 32 are to be followed in a holistic fashion. The interpretative exercise is engaged so as to yield an interpretation that is harmonious and coherent and fits comfortably in the treaty as a whole so as to render the treaty provision legally effective. A word or term may have more than one meaning or shade of meaning, but the identification of such meanings in isolation only commences the process of interpretation, it does not conclude it. Nor do multiple meanings of a word or term automatically constitute ‘permissible’ interpretations within the meaning of Article 17.6(ii). Instead, a treaty interpreter is required to have recourse to context and object and purpose to elucidate the relevant meaning of the word or term. This logical progression provides a framework for proper interpretative analysis. At the same time, it should be kept in mind that treaty interpretation is an integrated operation, where interpretative rules or principles must be understood and applied as connected and mutually reinforcing components of a holistic exercise.115
and further added that the proper interpretation of the second sentence of Article 17.6(ii) must itself be consistent with the rules and principles set out in the Vienna Convention. This means that it cannot be interpreted in a way that would render it redundant, or that derogates from the customary rules of interpretation of public international law. However, the second sentence allows for the possibility that the application of the rules of the Vienna Convention may give rise to an interpretative range and, if it does, an interpretation falling within that range is permissible and must be given effect by holding the measure to be in conformity with the covered agreement. The function of the second sentence is thus to give effect to the interpretative range rather than to require the interpreter to pursue further the interpretative exercise to the point where only one interpretation within that range may prevail.116
114 Article 17(6)ii. of the 1994 Agreement on Implementation of Article VI of the General Agreement on Tariffs and Trade, <www.wto.org/english/docs_e/legal_e/19-adp.pdf> (accessed on 1 May 2009). 115 United States – Continued Existence and Application of Zeroing Methodology (19 February 2009), Appellate Body Report, WT/DS350/AB/R, para. 268 (emphasis added). 116 Id., para. 272 (emphasis added).
canons of treaty interpretation
183
The aforementioned passages of the AB in the US – Continued Zeroing case are important for two reasons. First, the AB reiterated that the elements of treaty interpretation within Articles 31 and 32 are not hierarchically stuctured. They are all integral and essential parts of the interpretative process. As the AB very succinctly phrased it, they are “connected and mutually reinforcing components of a holistic exercise”.117 Secondly, and perhaps more importantly, the AB attempted to address the correct interpretation of Article 17.6(ii). It is guided in this endeavour by the principle of effectiveness (ut res magis valeat quam pereat). Based on this principle, it logically follows, that the second part of Article 17.6(ii) must have a raison d’ être; it cannot be devoid of any meaning. The first part, which refers to interpretation in “accordance with customary rules of interpretation” cannot be considered as rendering useless – inutile – the second part. In order to cope with this problem the AB had already in previous decisions declared that application of the rules of treaty interpretation in Articles 31 and 32 of the Vienna Convention could give rise to, at least, two interpretations of some provisions of the Anti-Dumping Agreement, which, under that Convention, would both be ‘permissible interpretations’ 118
The AB in US – Continued Zeroing expanded on this, further, by stating that if Article 17.6(ii) is to be interpreted effectively, then this would mean that there is a possibility that the application of the rules of the VCLT may lead to an interpretative range. By ‘interpretative range’ what the AB probably meant was a variety of interpretations that are not identical, but do not conflict either.119 These interpretations differ from each other either subtly or substantially, depending on the scale of the ‘interpretative range’ that the tribunal will find. Perhaps, this construction might be easier to comprehend if we try to visualise it. If we consider the entire span of possible interpretations of a specific provision, as a circle, then each interpretation corresponds to a different positioning of the radius. The sector of the circle created by these two radii would then be the interpretative range120 (see diagram).
117
Id., para. 268. United States – Anti-Dumping Measures on Certain Hot-Rolled Steel Products from Japan (23 August 2001), Appellate Body Report, WT/DS184/AB/R, para. 59 (hereinafter US – HotRolled Steel). 119 on why conflicting interpretations are not covered by this ‘interpretative range’ see analysis infra. 120 If three or more interpretations are permissible, then the section would be defined by the two interpretations/radii that are farther apart, i.e that create the greatest angle and the largest sector (the reason for this is that all the other interpretations would fall within the section defined by these two interpretations). 118
184
malgosia fitzmaurice & panos merkouris A
Interpretative Range
B
0
If any interpretation falls within that range (in the diagram it is the sector A0B) it is permissible and, thus, under Article 17.6(ii) should be considered as being in conformity with the 1994 Anti-Dumping Agreement.121 Consequently, according to the AB’s viewpoint Article 17.6(ii) functions in the following way. First, the customary rules of interpretation are applied. Second, this process might end up with two or more interpretations being ‘permissible’.122 In that case, the second part of Article 17.6(ii) becomes operational and all measures that are in conformity with these ‘permissible’ interpretations are legal and considered in conformity with the Anti-Dumping Agreement.123 In this way, the AB felt that it had given effect to the entire provision in accordance with the customary rules of interpretation. However, this whole theoretical model is based on the premise that the application of the customary
121
US – Continued Zeroing, supra note 115, para. 272; previous cases had also dealt with the issue of the meaning of Article 17.6(ii), but only briefly and none made mention of this notion of ‘interpretative range’, although all acknowledged the possibility of more than one ‘permissible’ interpretations. US – Continued Zeroing. See US – Hot-Rolled Steel, Appellate Body Report, supra note 118, para. 60; European Communities – Anti-Dumping Duties on Imports of CottonType Bed Linen from India (12 March 2001), Appellate Body Report, WT/DS141/AB/R, paras. 63–5; Thailand – Anti-Dumping Duties on Angles, Shapes and Sections of Iron or Non-Alloy Steel and H-Beams from Poland (5 April 2001), Appellate Body Report, WT/DS122/AB/R, para. 127. 122 i.e as falling within the aforementioned ‘interpretative range’. 123 US – Continued Zeroing, supra note 115, paras. 270–1; Generally on Article 17.6(ii) and the fact that no panel or the Appellate Body has ever made actual use of the notion of ‘permissible interpretations’ see James P. Durling and Matthew R. Nicely, Understanding the WTO AntiDumping Agreement: Negotiating History and Subsequent Interpretation (London: Cameron May, 2002), at 580–5; James P. Durling, ‘Deference, but only when Due: WTO Review of AntiDumping Measures’, Journal of International Economic Law 6 (2003): 123, at 142–3; Richard Cunningham and Troy Cribb, ‘Dispute Settlement through the Lens of ‘Free Flow of Trade’: A Review of WTO Dispute Settlement of US Anti-Dumping and Countervailing Duty Measures’, Journal of International Economic Law 6 (2003): 155, at 161–4; Daniel Tarullo, ‘Paved with Good Intentions: The Dynamic Effects of WTO Review of Anti-Dumping Action’, World Trade Review 2 (2003): 373, at 377–9; Holger Spamann, ‘Standard of Review for World Trade Organization Panels in Trade Remedy Cases: A Critical Analysis’, Journal of World Trade 38 (2004): 509,
canons of treaty interpretation
185
rules on interpretation can have as an end result more than one permissible interpretations; and therein lies the problem. Articles 31 and 32 of the VCLT, and their customary law counterparts, have been designed to function as a complete system. That means that for any dispute the application of the customary rules on interpretation would always lead to a result, a singular interpretation of the provision in question. If this was not the case, if Articles 31 and 32 allowed for more than one simultaneously valid interpretations within one case, then international judicial settlement would abound with non liquet decisions, as will be analysed infra. The US after the adoption of the AB Report stated that the VCLT itself allowed for a multiplicity of ‘permissible’ interpretations, which was evidenced by the fact that even “Article 32 made clear that the application of the rules of interpretation in Article 31 could lead to a meaning that was ‘ambiguous’ ”.124 However, that is incorrect, or to be more in spirit with the issue since we are discussing permissible interpretations, this is not a ‘permissible’ interpretation of the rules on interpretation. It is true that Article 31 might lead to an ambiguous result, but that is exactly why Article 32 was inserted. To either confirm the result arrived to via Article 31 or determine the meaning when there is ambiguity. Its function is both ‘confirmatory’ and ‘determinative’.125 Article 32 is what completes the system of interpretation, ensures that the judge can always arrive at one interpretative result and closes the door to any non liquet scenarios. It is no accident that for that reason exactly, the term ‘supplementary’ was selected for Article 32, because ‘to supplement’ means ‘to complete’. Since then, Article 17.6.(ii) talks about application of the customary rules of interpretation, this means both Article 31 and 32, thus the US argument, which is based on an isolated application of Article 31 fails. The combined application of Article 31 and 32 would always produce a singular result. In order to reinforce this conclusion, let us apply the argumentum ad absurdum logical tool. Let us consider as a given the multiplicity of ‘permissible’ interpretations through the application of Articles 31 and 32. In that case, if we apply Article 17.6.(ii)126 the interpretative process could be represented as follows: at 511; Edwin Vermulst, The WTO Anti-Dumping Agreement: A Commentary (Oxford: OUP, 2005), at 228–32; see also John Greenwald, ‘WTO Dispute Settlement: An Exercise in Trade Law Legislation’, Journal of International Economic Law 6 (2003): 113, at 117 who laments that due this lack of use: “in the hands of the panels and the Appellate Body, Article 17.6 has quickly become a dead letter”. 124 Dispute Settlement Body, ‘Minutes of Meeting held on 19 February 2009’, WT/DS/B/M/ 265, para.78 <docsonline.wto.org/GEN_highLightParent.asp?qu=%28%40meta%5FSymbol+ WT%FCDSB%FCM%FC%2A+and+%40meta%5FMeet%5FDate+19%2F02%2F2009%29& doc=D%3A%2FDDFDOCUMENTS%2FT%2FWT%2FDSB%2FM265%2EDOC%2EHTM&cu rdoc=3&popTitle=WT%2FDSB%2FM%2F265> (accessed on 1 July 2009). 125 In more detail see Panos Merkouris, ‘Third – Party Considerations and ‘Corrective Interpretation’ in the Interpretative Use of Travaux Préparatoires: Is it Fahrenheit 451 for Preparatory Work?’, published in the present Volume. 126 in the way that the AB effectively interpreted it in US – Continued Zeroing.
186
malgosia fitzmaurice & panos merkouris multiple permissible interpretations Ar.31,32 + 17.6(ii)(phrase 2) = Judgment on measure
However, what happens if we transport the scenario of multiple ‘permissible’ interpretations to a tribunal that has only recourse to Articles 31 and 32? How will that Tribunal reach a judgment? multiple permissible interpretations Ar.31,32 + ??? = Judgment on measure If we accept the possibility of multiple ‘permissible’ interpretations then where is the equivalent of Article 17.6(ii) in this process? How will the judge be able to select between the multiple ‘permissible’ interpretations? The argument that 17.6(ii) does not actually help the judge select between the many ‘permissible’ interpretations is not valid, since 17.6(ii) does not select because it accepts them all. But this is not the case, with most treaties and disputes where the international court has to reach a singular interpretation. Equally, no argument can be made in resorting to other principles of law, such as equity, because these, even more so since they are customary law, are already included within Article 31 and 32.127 It is evident, then, than another element is needed to complete the interpretative process and allow the judge to reach a judicial conviction. But no such element exists128 and, thus, we have reached an absurdum. Ergo, the interpretative system as described by Articles 31 and 32 is a complete system, which does not allow for its function to result in multiple ‘permissible’ interpretations. A further ‘nail in the coffin’ of the hypothesis of arriving at a number of ‘permissible’ interpretations through the application of Articles 31 and 32 is that, usually, in a dispute the two parties will claim conflicting interpretations of a certain provision. No middle ground exists, in this case, for the so called ‘interpretative range’. Either the interpretation of party A will be allowed or that of party B.129 Both cannot be considered ‘permissible’. This was regarded as self-evident by the AB in US – Continued Zeroing which stated:
127 the most appropriate for this would be Article 31(3)(c). On equity as an element of interpretation within Article 31 see Anastasios Gourgourinis, ‘Delineating the Normativity of Equity in International Law’, ICLR 11 (2009): 327–47. 128 this situation is, further, aggravated by the fact that this hypothetical element should also have customary law status, since in many cases Articles 31 and 32 are not applied as such but as custom. If that is the case, then we should be able to easily spot the practice and opinio juris of this element. 129 Otherwise, acceptance of both conflicting interpretations would essentially be a non liquet decision. There is also a third possibility, that the court adopts its own interpretation, but in
canons of treaty interpretation
187
We further note that the rules and principles of the Vienna Convention cannot contemplate interpretations with mutually contradictory results. Instead, the enterprise of interpretation is intended to ascertain the proper meaning of a provision; one that fits harmoniously with the terms, context, and object and purpose of the treaty. The purpose of such an exercise is therefore to narrow the range of interpretations, not to generate conflicting, competing interpretations. Interpretative tools cannot be applied selectively or in isolation from one another. It would be a subversion of the interpretative disciplines of the Vienna Convention if application of those disciplines yielded contradiction instead of coherence and harmony among, and effect to, all relevant treaty provisions.130
Since one interpretation from each party to a dispute, would by definition be in conflict with each other, the only other possibility remaining, would be from one party to put to the court two or more interpretations of the same provision, all of which would give legal effect to their hypothetical actions. But even this scenario cannot lead us to a ‘multiple permissible interpretations’ outcome. The court would examine each interpretation in sequential order. The moment it found one interpretation that it considered permissible, it would halt the interpretative process. Since already one basis for the legality of a certain action has been established, judicial economy dictates that there is no reason to look further for any other additional basis, as it would serve no purpose.131 From the above analysis, the conclusion is that the ‘effective’ interpretation of Article 17.6(ii) by the AB and the acknowledgement of the application of Article 31 and 32 leading to multiple ‘permissible’ interpretations – or as it was termed in the US – Continued Zeroing, leading to an ‘interpretative range’ – raises more problems than it solves. Essentially by applying the principle of effectiveness in such a way as it did, the AB in its effort to ensure the effectiveness of Article 17.6(ii) threatened that of a much larger system, that of Articles 31 and 32. This does not mean that Article 17.6(ii) should be rendered useless; only that the interpretation of Article 17.6(ii) should be approached differently. It is
order to reach at that point, for reasons of judicial economy, it would have to have already rejected the interpretations of both parties. So in this case, once again, we have ended up with only one interpretation. 130 US – Continued Zeroing, supra note 115, para. 273; similarly a member of the Appellate Body in his Concurring Opinion had the following to say: The range of meanings that may constitute a permissible interpretation does not encompass meanings of such wide variability, and even contradiction, so as to accommodate the two rival interpretations. One must prevail. The Appellate Body has decided the matter. At a point in every debate, there comes a time when it is more important for the system of dispute resolution to have a definitive outcome, than further to pick over the entrails of battles past. With respect to zeroing, that time has come.
US – Continued Zeroing, supra note 115, para. 312. 131 The judgments of the ECtHR are characteristic in this example, as very often when a certain violation has been found, it cursorily disposes of other similar claims, based on the same facts by stating that it “considers it unnecessary to examine this complaint”.
188
malgosia fitzmaurice & panos merkouris
helpful to recall the reason why the drafters inserted Article 17.6(ii) in the first place. At the end of the Uruguay Round negotiations, Article 17.6(ii) was key to the acceptance of the other provisions of the Anti-Dumping Agreement. The existence of such a provision confirmed that Members were aware that the text would pose particular interpretive challenges, at least in part because it was drafted to cover varying and complex anti-dumping systems around the world and longstanding differences concerning methodology. The negotiators, therefore, had indicated that it would have been a legal error not to respect a permissible interpretation of the Anti-Dumping Agreement.132
The goal is, thus, to give meaning to the second part of Article 17.6(ii) which states: [w]here the panel finds that a relevant provision of the Agreement admits of more than one permissible interpretation, the panel shall find the authorities’ measure to be in conformity with the Agreement if it rests upon one of those permissible interpretations.
without, at the same time, threatening the integrity of the system of interpretation as delineated by Articles 31 and 32, which is mentioned in the first part of Article 17.6(ii) and goes as follows: “the panel shall interpret the relevant provisions of the Agreement in accordance with customary rules of interpretation of public international law”. If we are to achieve this, three are the possible interpretations: 1. The ‘non – rigorous’ approach – This approach, focuses on the first part of Article 17.6(ii). The idea is that the principles of customary international law are not to be applied at the same level of intensity as regularly required by Article 31 and 32. They are to be applied less rigorously, thus, leaving open the option of multiple ‘permissible’ interpretations arising, without hurting the effectiveness of Articles 31 and 32 per se. There are a number of problems with this interpretation. First, it doesn’t seem to be based on the text, as the first part is pretty straightforward. Secondly, it presupposes that the interpretative process of Articles 31 and 32 consists of repeating the same steps, using the same elements, over and over again, but every time with an added level of scrutiny/intensity, until at last the judge remains with one interpretation. This, however, portrays interpretation as a linear process, which is far from the reality.133 There are no different levels of scrutiny/ intensity of the various elements of treaty interpretation. Finally, this approach would be open to extreme abuse, since there is no precedent in practice or any guidelines in law, judicial decisions or legal texts. A situation could arise, where the level of intensity used under Article 17.6(ii) was so 132 133
Dispute Settlement Body, supra note 124, para. 79. as we have already seen.
canons of treaty interpretation
189
low as to be non-existent, thus allowing for a flood of ‘permissible’ interpretations. In such a scenario, it would be the entirety of Article 17.6(ii) that would have been rendered ineffective, due to our interpretation. 2. The ‘negative’ approach – Again, this approach has as its starting point the first part of Article 17.6(ii). The customary rules of interpretation are usually applied by searching what is the correct interpretation. Article 17.6(ii) in order to have meaning as a whole, would presuppose a reversal of this methodology. Instead of looking for what is the correct interpretation the judges should focus on excluding what is not. This way, if conclusive evidence exists against a specific interpretation it would be excluded. All the rest would be impermissible. This approach, as well, is not devoid of problems. First of all, as with the first approach, this is also not based on the text of the Article. Furthermore, it is not at all certain that the result would be anything different from the normal interpretative process. If, as we have already demonstrated supra, the outcome of Article 31 and 32 application is reaching a singular interpretation, then it would not matter whether you try to find the correct interpretation directly or through a process of elimination. If anything, this latter process might be more time-consuming and onerous for the judges. 3. The ‘generic’ approach – This final approach differs from the previous two in the sense that it accepts and applies the first part of Article 17.6(ii) as it stands. However, this will lead us to only one interpretation. How, then, can this be reconciled with the second part of Article 17.6(ii) which talks about “more than one permissible interpretation”? International law is no stranger to the notion of ‘generic’ terms.134 It is common understanding that some notions or even legal provisions are ‘generic’ in the sense that they can be understood in various ways either nunc or ad futurum. The court by applying the first part of Article 17.6(ii) has reached one interpretation of a certain provision. However, by acknowledging its generic term, it recognises the possibility of other permissible interpretations, now or in the future, but which can be only adopted in a different case. In the present case, only one interpretation is permissible.135 However, this acknowledgment of the ‘generic’ nature of certain provisions is more than enough to give effect to the second part of Article 17.6(ii). The relevant article states: “[w]here the panel finds that a relevant provision … admits of more than one permissible interpretation”. Nowhere, does it say that the court has to identify the 134 see Aegean Continental Shelf case (Greece v. Turkey), Jurisdiction of the Court, Judgment of 19 December 1978, 1978 ICJ Rep. 3; also Rosalyn Higgins, ‘Some Observations on the InterTemporal Rule in International Law’, in Theory of International Law at the Threshold of the 21st Century, Jerzy Makarczyk and Krzysztof Skubiszewski (eds.) (The Hague: Kluwer Law International, 1996), 173–181; Donald Greig, Intertemporality and the Law of Treaties (London: BIICL, 2001). 135 see supra analysis on how only two interpretations are brought in fornt of a court in each dispute, out of which only one can prevail.
190
malgosia fitzmaurice & panos merkouris
exact content of each permissible interpretation. It just has to acknowledge that a certain provision can be permissibly interpreted in various ways. The acknowledgment of the ‘generic’ nature of a provision does exactly that. The remaining part of Article 17.6(ii) also poses no issues since it is the logical culmination of the interpretative process. Since an interpretation has been arrived at, the measure is deemed to be effective. Under this light, we can accept the ‘interpretative range’ notion of the AB. This approach seems also to best reflect and serve the considerations and fears of the drafters of the 1994 Anti-Dumping Agreement. They were aware that the text would pose particular interpretative challenges, due to the fact that it covered varying and complex anti-dumping systems and differences in methodology and they wanted to insert a provision that would both protect them and remind the court of the special nature of the provisions of the Anti-Dumping Agreement. An objection could be raised, then, as to the usefulness of the second part of Article 17.6(ii). If it is simply a re-statement of the obvious and applicable, is it not, in essence, a provision devoid of any effectiveness/ meaning? But in that case, the same could be argued for the first part of that Article, or in fact for any such clause inserted in any agreement. Since the principles of interpretation are customary law, they would ipso facto apply to any dispute dealing with interpretation, so there is no apparent need for their inclusion in a treaty clause. Need is not to be equated with effectiveness. The fact that it is not necessary to restate the obvious does not ipso facto render the provisions that do so meaningless and ineffective. On the contrary, if they state the obvious, then they definitely have meaning; even more so in the case of Article 17.6(ii), which apart from stating the obvious actually establishes the way that the Anti-Dumping Agreement is to be viewed and interpreted by the judges, as an evolving, ‘living’ instrument. Although, our analysis started from the principle of effectiveness, in order finally to apply this principle correctly, we once again, had to resort to all the elements of treaty interpretation; text, context, object and purpose, intention as well as travaux préparatoires; which proves even further, the interconnectedness and complimentarity of all these elements. B. The Principle of Non-Retroactivity The jurisprudence of the WTO has also dealt with the principle of nonretroactivity as codified in Article 28 of the VCLT136 in a number of cases, 136
Art. 28 of the 1969 VCLT: Non-retroactivity of treaties: unless a different intention appears from the treaty or is otherwise established, its provisions do not bind a party in relation to any act or fact which took place or any
canons of treaty interpretation
191
most notably in the Desiccated Coconut case137 and in the EC Bananas case.’138 In the first of the above cases, Philippines brought a case against Brazil under the GATT 1994 for a countervailing measure that was introduced after the 1994 GATT entered into force. The arguments of the parties were as follows. Brazil argued that customary international law did not permit the retroactive application of treaty obligations. Relying on Article 28 of the 1969 VCLT, Brazil was of the view that the relevant act at issue in this dispute was the initiation and subsequent conduct of a countervailing duty investigation, which began on 21 June 1994, before the date of the entry into force of the WTO Agreement and GATT 1994 on 1 January 1995. Brazil thus contended that GATT 1994 could not bind it with respect to this investigation.139 The Philippines represented the view that reliance on Article 28 of the 1969 VCLT was erroneous because the act in question in this dispute was the imposition of the countervailing measure on 18 August 1995, that was after the GATT 1994 had entered into force for the Parties. The only act undertaken by Brazil prior to the entry into force of the WTO Agreements was the initiation of the investigation, but all other significant acts, such as the imposition of provisional measures, occurred after the entry into force of the WTO Agreements. In addition, the Philippines submitted that general principles of international law, as codified in the 1969 VCLT required that even pre-existing measures should be reviewed in the light of new obligations deriving from the new agreement. From this follows that even if Brazil’s investigation had been concluded and the countervailing measure had been imposed before the entry into force of the WTO Agreement, Brazil’s continued implementation of such a measure would have to be assessed in the light of its obligations under the GATT 1994. The Philippines also claimed that in the US-Reformulated Gasoline case (see above), the AB had examined the US rule making process that took place prior to the effective date of the WTO Agreements. This, in the view of the Philippines, demonstrated that the panel could consider events prior to the entry into force of the WTO Agreements in evaluating the consistency of a measure with these Agreements.140
other situation which ceased to exist before the date of the entry into force of the treaty with respect to that party. 137 Brazil – Measures Affecting Desiccated Coconut (17 October 1996), Panel Report, WT/ DS22/R; Brazil – Measures Affecting Desiccated Coconut (21 February 1997), Appellate Body Report, WT/DS22/AB/R. 138 EC – Regime for the Importation, Sale And Distribution of Bananas (22 May 1997), Panel Report, WT/DS27/R/HND; EC – Regime for the Importation, Sale And Distribution of Bananas (9 September 1997), Appellate Body Report, WT/DS27/AB/R. 139 Brazil – Measures Affecting Desiccated Coconut, Panel Report, supra note 137. The main arguments of the Parties can be found at Part V, para. 23. 140 Id., paras. 26–7.
192
malgosia fitzmaurice & panos merkouris
This argument was rejected by Brazil who contended that the USReformulated Gasoline case did not support the Philippines’ position. It was the view of Brazil that Venezuela and Brazil challenged the continued maintenance of discriminatory standards by the US after the entry into force of the WTO and not the rule making process that had led to the imposition of the those standards.141 The Parties to the Desiccated Coconut case also analysed the Panel’s decision in the US-Countervailing Duties on Non-Rubber Footwear from Brazil (Non-Rubber Footwear case).142 According to the Philippines, the Panel in this case recognised that general principles of international law required that preexisting measures must be reviewed in light of new obligations imposed by a new agreement. Accordingly, the Philippines postulated that the same argumentation should apply to the Brazilian countervailing measures as governed by the GATT 1994.143 Brazil on the other hand was of the view that the Panel’s decision in the Non-Rubber Footwear case did not support the argument of the Philippines, since in Brazil’s view the Panel concluded that any obligation under the SCM Agreement would be met by conducting a review of its conformity with the new obligations at the request of an interested party. The Philippines did not request such a review. Moreover, Brazil argued that in the Non-Rubber Footwear case it was stated that Article 28 of the 1969 VCLT prevented the application of the new treaty to pre-existing acts, i.e., the actual investigation, and only permitted its application to the continuing implementation. Therefore, since that situation had continued after the entry into force of the WTO, the collection of duties was subject to the WTO Agreement and the Philippines could seek a review to ensure that the continued collection was in accordance with WTO rules.144 The Parties also referred to the Judgment of the Permanent Court of International Justice (PCIJ) in the Phosphates in Morocco case,145 regarding the interpretation of the concept of retroactive application addressed in Article 28 of the 1969 VCLT. This case dealt with the question whether the ‘situation’ (the regime creating the cartel that led to the monopolisation of phosphate trade) was subject to the 1931 Agreement, under which the complaint was brought, or whether it occurred prior to that Agreement. The Court found that because the act underlying the dispute arose in 1920, even though the monopolisation continued after 1931, the dispute could not be brought under the 1931 Agreement. Brazil contended that this decision established the principle codified in Article 28 of the 1969 VCLT and that, similarly to the Phosphates 141
Id., para. 28. US-Countervailing Duties on Non-Rubber Footwear from Brazil (3 June 1995), Panel Report, SCM/94, paras. 4.5 and 4.10. 143 Desiccated Coconut case, Panel Report, supra note 137, para. 29. 144 Id., paras. 30–1. 145 Phosphates in Morocco case (Italy v. France), 1938 PCIJ (Ser. A/B) No 74, at 23. 142
canons of treaty interpretation
193
in Morocco case, the WTO Members only intended to be subject to WTO obligations after the date of entry into force, indicated in the WTO Agreement, rather than having it enter into force as a soon as a sufficient number of states had ratified it.146 According to the Philippines the interpretation of the PCIJ was different. It believed that the PCIJ case had only made a pronouncement that it had no jurisdiction over disputes involving alleged international law violations that originated from definitive acts that took place before the Parties ratified the instruments through which they subjected themselves to the compulsory jurisdiction of the Court.147 The AB observed that Article 28 of the 1969 VCLT contains a general principle of international law concerning the non-retroactivity of treaties. The AB was quite clear that absent of a contrary intention, a treaty cannot apply to acts or facts which took place, or situations which ceased to exist, before the date of its entry into force.148 The Panel and the AB decision was based on the interpretation of Article 32.3 of the SCM Agreement that reads as follows: …the provisions of this Agreement shall apply to investigations, and reviews of existing measures, initiated pursuant to applications which have been made on or after the entry into force for a Member of the WTO Agreement.
This Article was according to the AB an express statement of intention. The relationship between the SCM Agreement and Article VI of the GATT 1994 is set out in Article 10 and 32.1 of the SCM Agreement. Both the Panel and the Appellate Body found the inapplicability of the GATT provisions on this case as a ground for action. The investigation of the Philippine subsidy included both procedural and substantive aspects continuous up to the imposition of any countervailing duty and therefore subject to the previous subsidy regime. The AB also stated that it took into account the integrated nature of the WTO Agreement in combination with the specific language of Articles 10 and 32. l. of the SCM Agreement which are not separable from the rights and obligations of the GATT 1994 or the WTO Agreement taken as whole.149 In the EC-Bananas case before the AB the subject matter was a complaint both by the EC and Ecuador, Guatemala, Honduras, Mexico and the United States (the ‘Complaining Parties’) from certain issues of law and legal interpretations in the Panel Reports.150 The Panel was established to consider a complaint by the Complaining Parties against the EC concerning the regime 146 147 148 149 150
Desiccated Coconut case, Panel Report, supra note 137, para. 32. Id., para. 33. Desiccated Coconut case, AB Report, supra note 137, Part D. Id. EC – Bananas case, supra note 138.
194
malgosia fitzmaurice & panos merkouris
for the import, sale and distribution of bananas established by Council Resolution (‘Regulation 404/93’) and subsequent EC related legislation. The legal issue was the application of the General Agreement on Trade in Services treaty, (GATS), which entered into force after the introduction of a trade measure. The EC argued that taking into account that the GATS entered into force on 1 January 1994, only factual situations regarding the EC banana import regime arising after the entry into force of the GATS and not before, should be considered under GATS Articles II and XVII. The EC claimed that the Panel erred in its interpretation of what constitutes a ‘situation’ within the meaning of general international law as codified in Article 28 of the 1969 VCLT.151 The Panel stated that the scope of our legal examination includes only actions which the EC took or continued to take, or measures that remained in force after entry into force of the GATS. Likewise consistency or inconsistency with requirements of Articles II and XVII of GATS would be made with respect to the period after the entry into force of the GATS.
The Panel added in a footnote that the EC measures at issue-may be considered as continuing measures, which in some cases were enacted before the entry into force of the GATS did not cease to exist after that date” (the opposite situation envisaged in Article 28).152
The Complaining Parties argued in respect to Article 28 of the 1969 VCLT that the Panel was correct to characterise the measure at issue as a continuing measure that was in some cases enacted before the entry into force of the GATS. It was argued that such a measure was excluded by the ILC from the scope of Article 28 of the 1969 VCLT. The EC raised the question whether the Panel erred in giving retroactive effect to Articles II and XVII of the GATS, in violation of the principle of Article 28 of the 1969 VCLT.153 The EC further submitted that the continuing situation at issue was not the continued existence of Regulation 404/93 and other related regulations, but the alleged discrimination against foreign service suppliers. The EC was of the view that de facto discrimination is a fact at a particular point in time and does not necessarily continue for as long as a law remains in force. The EC maintained that there was no ground for the assumption that the factual data relating to 1992, even if correct, continued to exist after the entry into force of the GATS. Therefore, the EC contended that the non-retroactivity principle did apply in that case and as a result invalidated the Panel’s conclusion of inconsistency of the EC import licensing regime with Articles II and XVII of the GATS.154 151 152 153 154
EC – Bananas case, AB Report, supra note 138, para. 52. EC – Bananas case, Panel Report, supra note 138, para. 7.331. Id., para. 235. Id., para. 236.
canons of treaty interpretation
195
The AB, however, upheld the findings of the Panel and stated that [i]t is, however, evident from the terms of its finding that the Panel concluded, as a matter of fact, that de facto discrimination did continue to exist after entry into force of the GATS. Thus the actual finding is beyond review by the Appellate Body. Thus, we do not reverse or modify the panel’s conclusion…155
C. Related Issues The jurisprudence of the WTO also provides examples of other points of interpretation that are not strictly speaking within the canons of interpretation of the 1969 VCLT.156 Therefore, they fall outside the scope of this study. One such example is the principle in dubio mitius or of restrictive interpretation, whereby limitations on state sovereignty are not to be presumed. This principle was enunciated in the Wimbledon case.157 However, this principle was treated as rather obsolete by Lord McNair.158 However, it may be noted that that the opposite view is also expressed in the doctrine of international law, among others for example by Jennings and Watts.159 This principle was, however, resurrected by the WTO, both by the Panel and the AB, in the EC-Measures Concerning Meat and Meat Products (Hormones) case (Hormones case).160 The issue at question was the interpretation of the Agreement on Sanitary and Phytosanitary Measures (SPS Agreement). The Panel and the AB analysed Articles 3.1 and 3.2 of the SPS Agreement. The Panel ruled that the measures adopted by the EC must be ‘based on’ international standards under Article 3.1 of the SPS Agreement, meaning that they must ‘conform to’ international standards as required under Article 3.2. The AB rejected the Panel’s interpretation of Article 3.1 of the SPS Agreement that measures ‘based on’ meant that members must ‘conform to’ the standards, guidelines and recommendations of the Codex Alimentarius. Therefore, the measure might be ‘based on’ the standard, without however fully incorporating all of its elements. The AB also observed that both notions (‘based on’ and ‘conform to’) are used in the SPS Agreement in different articles and in different paragraphs of the same article, therefore indicating the possibility of various choices and reflecting an intention to convey a different meaning.161
155
Id., para. 237. Donald W. Greig, International Law (London: Butterworth, 1976), 480–1. 157 The Wimbledon case (France, Italy, Japan and the UK v. Germany), 1923 PCIJ (Ser. A) No. 1, 15. 158 Lord Arnold McNair, ‘Treaties and Sovereignty’, in Symbolae Verzijl: presentees au Professeur J. H. W. Verzijl a l’ occasion de son LXXieme anniversaire, Jan H. W. Verzijl and Frederik M. Van Asberck (eds.) (The Hague: Nijhoff, 1958), 222. 159 Jennings and Watts, supra note 7, at 1278. 160 EC – Measures Concerning Meat and Meat Products (Hormones case) (16 January 1998), Appellate Body Report, WT/DS48/AB/R. 161 Id., para. 164. 156
196
malgosia fitzmaurice & panos merkouris
The AB, stated that [w]e cannot lightly assume that sovereign states intended to impose upon themselves the more onerous, rather than the less burdensome, obligation by mandating conformity or compliance with such standards, guidelines and recommendations.162
V. The North American Free Trade Agreement The North American Free Trade Agreement (NAFTA) was signed in 1993 and entered into force on 1 January 1994 between the United States, Mexico and Canada. This Agreement established a free trade area.163 Its purpose was to promote trade and investment among the parties.164 The question that will be dealt with in this essay is whether, and to what extent, the canons of interpretation as enshrined in the 1969 VCLT have been followed in the jurisprudence of the NAFTA. It should be noted however that a detailed description of the settlement of disputes procedure will not be included as it exceeds the framework of this study. Briefly, the crucial provisions for dispute settlement are found in Chapter Twenty of NAFTA. Chapter Twenty includes provisions relating to the avoidance and settlement of all disputes as regards the interpretation or application of the Agreement, with the exception of matters covered in Chapter Eleven (Investment), Chapter Fourteen (Financial Services), and Chapter Nineteen (Antidumping and Countervailing Duty Final determinations). The rule is that when general disputes are not resolved through consultation within a specified period of time, the matter may be referred at the request of either party to a non-binding panel, set up under Article 2008. Under this Article, any party may request the establishment of an Arbitral Panel. A panel under Chapter Twenty is established similarly to one under dispute settlement provisions of the WTO. The NAFTA is not isolated from general international law. Article 102 (2) titled ‘Objectives’, clearly states that [t]he Parties shall interpret and apply the provisions of this Agreement in the light of its objectives set out in paragraph 1 and in accordance with applicable rules of international law.
Of general importance is also Article 103 in which the parties affirm existing rights and obligations with respect to each other under the GATT and other agreements to which they are parties. However, in the event of discrepancy
162 163 164
Id., para. 165. Art. 101 NAFTA. Art. 102 NAFTA.
canons of treaty interpretation
197
between the NAFTA and the GATT 1994 Agreements, the NAFTA prevails, except as otherwise provided in the agreement. A different provision however is applicable in relation to conflicts with environmental and conservation agreements.165 A. The Matter of Cross Border Trucking Services Case There are numerous examples of the rules relating to interpretation adopted in the course of dispute resolution in the NAFTA, under different chapters. One such example is the decision adopted by the Arbitral Panel pursuant to Chapter Twenty in the Matter of Cross-Border Trucking Services case.166 In this case the Panel had the task of interpreting provisions on land transportation in the NAFTA area concerning Reservations for existing Measures and Liberalisation Commitments (Section VII), Services (Section VIII) and Investment (Section IX). In order to interpret the case at hand, the Panel confirmed that the mandatory standard for the interpretation of detailed provisions of the NAFTA was included in Article 102 (2). The Panel, in order to interpret particular provisions, analysed them in the context of the objectives of the NAFTA, as specified in Article 102 (1). These objectives develop the principal purpose of the NAFTA as enshrined in its Preamble, i.e., inter alia, to “create an expanded and secure market for the goods and services produced in their territories”. The Panel followed the examples of international tribunals and courts and determined the principal objective of the treaty by resorting to the Preamble.167 Therefore, it concluded that the main objective of the NAFTA was the liberalisation of trade, something already stressed by the Panel in the Agricultural Products arbitration.168 The Panel said in this case as follows: [a]s a free trade agreement, NAFTA has the specific objective of eliminating barriers to trade among the three contracting Parties. […] Any interpretation adopted by the Panel must, therefore, promote rather than inhibit NAFTA’s objectives. Exceptions to obligations of trade liberalisation must perforce be viewed with caution.169
The Panel followed the well-accepted principle established in the jurisprudence of the GATT and the WTO that exceptions to general obligations are to
165
Art. 104 NAFTA. NAFTA Arbitral Panel Established Pursuant to Chapter Twenty – In the Matter of CrossBorder Trucking Services (6 February 2001), Final Panel Report, Sec. File No. USA-MEX98-20080-01. 167 Id., para. 219 and accompanying note 233. 168 NAFTA Arbitral Panel Established Pursuant to Art. 2008 – In the Matter of Tariffs Applied by Canada to Certain United States Origin Agricultural Products (2 December 1996), Final Panel Report, Sec. File ODA 95-2008-01, para. 122. 169 Trucking Services case, Panel Report, supra note 166, para. 219. 166
198
malgosia fitzmaurice & panos merkouris
be construed narrowly.170 The Panel also observed that the Parties intended to “preserve their flexibility to safeguard the public welfare”.171 In the Trucking Services case the Panel made a very important observation that “[i]n identifying the rules of interpretation of international law referred to in Article 102 (2), the Panel need go no further than the 1969 Vienna Convention on the Law of Treaties”.172 Both Parties decided that the 1969 VCLT was a proper tool to use for interpretation. The Panel specifically emphasised that the guiding provision for interpretation was included in Article 31(1) of the 1969 VCLT. Therefore, the Panel continued, in addition to the ordinary meaning of the terms, interpretation must take into consideration the context, object and purpose of the treaty. The context includes, in addition to the text of the treaty, its preamble and annexes, and any agreement relating to the treaty. The Panel further added, following the provisions of Article 31, that, if necessary any subsequent practice and any relevant rules of international law, applicable in the relations between the parties, should be taken into account. The Panel stated that if these criteria prove to be insufficient, recourse should be allowed to supplementary means of interpretation in accordance with Article 32.173 The Panel correctly concluded that only if the ordinary meaning of the words, as established through the study and analysis of the context, appear to contradict the purpose and object of the treaty, other means of interpretation should be allowed. The Panel stated that in the case at hand, it deemed recourse to Article 32 of the 1969 VCLT unnecessary.174 The Panel made yet another interesting statement, namely, that Article 31 must be applied, as all other Articles of the 1969 VCLT, in conjunction with Article 26 (pacta sunt servanda), of the same convention.175 A novel and interesting point was made by the Panel as to the invocation by both Parties of their national legislation on land transportation. The Panel referred to Article 27 of the 1969 VCLT that states that “[a] party may not invoke the provision of its internal law as justification for its failure to perform a treaty”. Therefore, on the grounds of this provision, the Panel declined to examine national laws and decided to examine only international law. Thus,
170 Id. and accompanying note 234. The Panel gave the examples of the following cases: US – Restrictions on Imports of Tuna (22 February 1982), Panel Report, L/5198-29S91; Reformulated Gasoline case, AB Report, supra note 32; Shrimp case, supra note 61; Thailand – Restrictions on Importation of and Internal Taxes on Cigarettes (7 November 1990), Panel Report, DS10/R-37S/200, at 87. 171 Id., para. 219. 172 Id., para. 220. 173 Id., para. 221. 174 Id., para. 222. 175 Art. 26 of the 1969 VCLT: “[e]very treaty in force is binding upon the parties to it and must be performed by them in good faith”: Trucking Services case, Panel Report, supra note 166, para. 223.
canons of treaty interpretation
199
national laws either of Mexico or the US should not be utilised for the interpretation of the NAFTA.176 Substantively, the interpretative dispute concerned the problem of reservations for existing measures and liberalisation commitments (Annex I), in particular the question of reservations for a phase-out of national treatment. The Panel commenced its interpretative analysis by examining the interpretative note177 (Note). The relevant part of the Note178 clearly indicates two main points of interpretation: i. when interpreting liberalisation commitments regarding Phase-Out elements in Annex I, the elements of reservation must be considered in the light of the relevant provisions of the Chapters against which the reservation is adopted; and ii. the Phase-Out element of a reservation are to prevail over all other elements of the reservation. Another important interpretative point concerning the ‘measures’ in Annex I was also clarified by the Note. The measures concern the laws, regulations or other measures, as qualified, where indicated by the ‘Description’ element, for which reservation is adopted. The Note, explicitly adopted a hierarchy of rules for the interpretation of the agreed reservations. It stated in paragraph 3 (b) that if the Measures were qualified by a liberalisation commitment from the Description element, the Measures element as so qualified should prevail over all other elements.179 The Panel also observed that, in the light of the Note, the text of the phaseout elements in Annex I, relating to both the liberalisation and cross-border services, was unambiguous if based on the ordinary meaning of the words.180 Generally, as regards reservations, the Panel observed that the reservations under the heading Land Transportation included in Annex I were contrary to the principal objective of the NAFTA as set out in its Preamble and against achieving concrete objectives as specified in Article 102 (2). The Panel presumed that such reservations were intended as a necessary structural element
176
Id., at para. 224. The interpretative Note is one of the most important aspects of Chapter Eleven of the NAFTA. Interpretation is done by the Free Trade Commission and its purpose is to submit a jointly agreed interpretation, in order to retain a certain degree of control over the interpretation of the Agreement, in the process of adjudication of cases by the ad hoc Tribunals. Art. 1136 of the NAFTA is not construed on the basis of stare decisis principle. However, previous Decisions, of Tribunals are taken informally into consideration during the decision-making process. 178 Trucking Services case, Panel Report, supra note 166, para. 229. 179 Id., paras. 231–5. 180 Id., para. 236. 177
200
malgosia fitzmaurice & panos merkouris
that was essential to the establishment of a Free Trade Area, the ultimate goal of the NAFTA.181 In view of the above, the Panel recalled a legal principle expressed in the maxim exceptio est stricissimae applicationis, meaning that reservations to treaties are to be construed restrictively. The Panel further relied on the Wimbledon case182 and the Hormones case,183 in which cases it was decided that a treaty provision must take precedence over a general rule of international law. Accordingly, the Panel decided that the phase – out provisions concerning reservations must be given full legal force over all other elements of Annex I. If the Parties wished to create any mode of subsequent acceptance or condition to the liberalisation commitments agreed on in the phase-out elements of Annex I, they would have or could have used a different wording. It was the view of the Panel that the phase-out provisions of Annex I must prevail over all other elements of the same Annex. Therefore, the US had failed to demonstrate the existence of any valid legal grounds for their noncompliance with the NAFTA liberalisation commitments regarding Land Transportation Services and Investment in Annex I. In view of the above, the phase-out obligations of the US under Annex I as regards cross-border trucking services and investment had priority unless there was some other, different, provision under the NAFTA that could supersede these obligations.184 The Panel next proceeded to examine, analyse and interpret whether the US was in breach of Article 1202 (national treatment for cross-border services) and 1203 (most-favourednation treatment for cross-border services) of the NAFTA, by failing to lift its moratorium on the processing of applications by Mexican owned trucking firms for authority to operate in US border states. In view of the expiry on 17 December 1995 of the US reservation to Annex I, in connection to cross-border trucking, the maintenance of the moratorium could be justified either under the language of Articles 1202 and 1203, or on the basis of some-other provisions of the NAFTA, i.e. Chapter Nine (standards) or Article 2101 (general exceptions). Neither the US nor Mexico claimed that Annex I itself contained an exception that would justify US actions and the US did not rely on Chapter Nine as a defence. Both parties relied on their interpretation of Articles 1202, 1203 and 2101.185 The arguments of the parties may be summarised as follows: the US argued that Mexico’s truck transportation regulatory system did not maintain the same rigorous standards as the systems in the US and Canada. For that reason the ‘in like circumstances’ language of Article 2102 meant that “service providers [in Mexico] may be treated differently in order to address the legitimate 181 182 183 184 185
Id., para. 237. Wimbledon case, supra note 157. EC – Hormones case, AB Report, supra note 160, para. 253. Trucking Services case, Panel Report, supra note 166, paras. 238–40. Id., para. 241.
canons of treaty interpretation
201
regulatory objective”. On the other hand, the trucking regulatory system of Canada, which maintained the same standards as that of the US. was not in contravention of the most-favoured-nation treatment under Article 1203 and Canadian trucking firms were treated as “in like circumstances in relation to the US trucking firms unlike the Mexican trucking firms”. Furthermore, the US also invoked Article 2101 general exceptions to NAFTA obligations on grounds of health, safety and consumer protection.186 Mexico submitted a very different interpretation of Articles 1202 and 1203. It did not rely on the argument of the equivalence of the Mexican and the US regulatory trucking system, but it pleaded that Mexican trucking firms were entitled to the same rights as US carriers under the law of the United States, that is “consideration on their individual merits and a full opportunity to contest the denial of operating authority”. According to Mexico, any other position is in contravention of Articles 1202 and 1204 of the NAFTA. Mexico, relied on travaux preparatories to the NAFTA. It appeared that governments understood that motor carriers would have to comply fully with the standards of the country in which they were providing their services.187 Canada (which participated in the proceedings on the basis of Article 2013) submitted essentially the same interpretation as Mexico.188 The pertinent part of Article 1202 provides as follows: “1. [e]ach Party shall accord to service providers of another Party treatment no less favourable that it accords, in like circumstances, to its own service providers”. Article 1203 states as follows: [e]ach party shall accord to service providers of another Party treatment no less favourable than it accords, in like circumstances, to service providers of any other Party or of a non-Party.
The Panel took into consideration that under US law, operating authority applications received from US and Canadian owned and domiciled carriers are treated on an individual basis. Consequently, the blanket refusal of the US to review the application for operating authority from Mexican trucking service providers on an individual basis gave rise to suggestions of discrepancy with the national treatment obligation and most-favoured-nation treatment (since Canadian carriers were treated on an individual basis). The Panel proceeded then to interpret the phrase ‘in like circumstances’ in Articles 1202 and 1203. In order to do that it sought guidance in other agreements that used similar language and had recourse to travaux preparatoires. The Panel observed that the Parties were in agreement that the use of the
186
Id., para. 242. Id., para. 243. The second defense of the US i.e., the exceptions on the grounds of health, safety and consumer protection were similarly not accepted by Mexico. However, it did not lead to issues of interpretation, therefore it will not be dealt with in this essay. 188 Id., para. 244. 187
202
malgosia fitzmaurice & panos merkouris
phrase ‘in like circumstances’ was intended to have the meaning ‘like services and service providers,’ as suggested by Canada and Mexico during NAFTA negotiations. The US in its Memorial, asserted that the phrase ‘in like circumstances’ was not substantively different from the phrase ‘in like situations,’ as used in bilateral investment treaties (not disputed by Mexico).189 Most importantly, no Party asserted that the use of the phrase ‘in like circumstances’ in NAFTA Chapter Twelve was intended to have a different meaning from this in the United States-Canada Free Trade Agreement (FTA). For example, Mexico in its Memorial observed that the “immediate source” of the ‘in like circumstances’ language in Articles 1202 and 1203 of the NAFTA was the FTA. The Parties also agreed that the language of the FTA regarding national treatment gave support to the interpretation of the phrase used in the NAFTA to permit differentiated treatment, where appropriate, to meet legitimate regulatory objectives. The Parties did not dispute that the interpretation that allowed different national treatment might be appropriate and consistent with the Party’s national obligations.190 The Panel relied on Article 1402 of the FTA as the most instructive provision in the interpretation of the national treatment requirement. Having analysed the above Article, the Panel observed that, for the purposes of interpretation, this Article was more elaborate as to the provision of a more detailed and specific limitation on one Party’s right to depart from its national treatment than that included in the brief text of Article 1202. The Panel also observed that similar national treatment obligations had already been interpreted in one case before the GATT, where it was permitted to impose certain requirements concerning imports that were different from those imposed on domestic products.191 Identical treatment is not necessarily required as regards intellectual property violations concerning imported goods compared to domestically produced goods. However, the Panel observed that in the Section 337 case, it was recognised that formally identical requirements for imports may in fact provide less favourable treatment in specific circumstances.192 The Panel interpreted the relevant provisions in the light of the general objectives of the NAFTA which are contained in Article 102 paragraph 1 (a): “to eliminate barriers to trade in, and facilitate the cross-border movement of, goods and services between the territories of the Parties”. These general objectives are further elaborated in more specific principles of the NAFTA, including national treatment. The Panel was also mindful that provisions of the Agreement required an interpretation in light of the objectives and applicable 189
Id., paras. 248–9. Id., para. 249. 191 US Section 337 of the Tariff Act of 1930 (7 November 1989), Panel Report, L/6439-36S/345, para. 5.11, mentioned in the Trucking Services case, Panel Report, supra note 166, para. 251. 192 Id., para. 251. 190
canons of treaty interpretation
203
rules of international law. Finally, the Panel reminded itself that the same term was used in the FTA. Taking into account all these elements, the Panel was of the view that the proper interpretation of Article 1202 required that differential treatment should not exceed the necessary requirements such as safety and that such treatment should be equivalent to the treatment accorded to domestic service providers. The Panel was also of the view that a broad interpretation of the phrase ‘in like circumstances’ would defeat the purpose of Articles 1202 and 1203. In light of the above reasoning, the Panel reached the conclusion that the position adopted by the US, namely that the phrase ‘in like circumstances’ permitted continuation of the moratorium on accepting applications for operating authority in the US from Mexican owned and domiciled carriers, was “an overly-broad reading of the clause”.193 As to the US suggestion that Article 2101 allowed the US to refuse to accept applications from Mexican tracking service providers due to safety concerns, the Panel expressed the view that the ‘in like circumstances’ language, as an exception, should be interpreted narrowly, and that this was also applicable to Article 2101. To further substantiate this interpretation, the Panel relied on the GATT/WTO history and the FTA language. The Panel observed that, although no explicit language limitations on the scope of the ‘in like circumstances’ expression was contained in Chapter Twelve, the general exception in Article 2101 (2), invoked by the US, was similar to the language of Article XX of the GATT and to the FTA proviso limiting the exceptions to national treatment to situations where “the difference in treatment is no greater than necessary for…health and safety consumer protection reasons”.194 The Panel noted that in view of the above discussion, under Article 2101, safety measures such as the US moratorium on accepting applications for the operating authority from Mexican trucking service providers might be justified only to the extent that they were “necessary to secure compliance” with rules and regulations that were otherwise consistent with the NAFTA. This interpretation was also in accordance with the GATT/ WTO jurisprudence, e.g. the Reformulated Gasoline and the Shrimp/Turtle cases, discussed above. The close interpretative analysis of the above cases convinced the Panel that, in order for the US to justify the imposition of its moratorium, it must first secure compliance with laws or regulations that do not discriminate between the Parties; and the moratorium must not be arbitrary, or introduce unjustifiable discrimination, or disguised restrictions on trade. Furthermore, the US failed to demonstrate that there were no alternative means of achieving US safety goals that were more consistent with the NAFTA moratorium.195
193 194 195
Id., paras. 258–9. Trucking Services case, Panel Report, supra note 166, para. 260. Id., paras. 269–70.
204
malgosia fitzmaurice & panos merkouris
The final interpretative considerations were given to the Preamble of the NAFTA that states that the Parties “resolve to…preserve their flexibility to safeguard public welfare”. The Panel decided that the US could not rely on this provision as an independent ground for failing to comply with its obligations under various provisions included in the NAFTA text and Annex I. The Panel relied on Article 31 of the 1969 VCLT which considers the Preamble as a part of the ‘context’ in which a treaty is interpreted. However, there was no suggestion that the preambular language was intended to overrule the textual obligations. The style of the Preamble that uses language such ‘resolve’ (rather than ‘agree to’, ‘shall’ or ‘must’), indicates that the Preamble is aspirational in character. The Panel also observed that the Parties have “resolved to…create an expanded and secure market for the goods and services produced in their territories…”. This phrase corresponds with the obligation placed upon the US by Articles 1202 and 1203. Taking into account all the above, the Panel came to the conclusion that there is no legally sufficient basis for interpreting ‘in like circumstances’ as permitting a blanket moratorium on all Mexican trucking firms. Nor is the departure from national treatment and most-favoured-nation treatment under these Articles justified under 2101.196
The issue before the Panel with regard to investment – i.e., the determination whether the failure by the US government to take appropriate regulatory measures to eliminate the moratorium on Mexican investments in companies providing international transportation by land constituted a breach of Articles 1102, 1103 and 1104 of the NAFTA – was decided by the Panel on the basis of the substantive analysis of the relevant provisions of the NAFTA and the WTO, not on the basis of the interpretation of the provisions of the Agreement. Therefore, it will not be dealt with in this essay. B. The Tariffs Applied by Canada to Certain US–Origin Agricultural Products Case It may be said that the classical case from the point of view of the interpretative analysis of the text is the arbitral award in the matter of the Tariffs Applied by Canada to Certain US-Origin Agricultural Products case.197 The substantive issue in this case was whether the customs duties imposed by Canada on certain US origin agricultural products, following ‘tariffication’ in accordance with the agreements reached as a result of the Uruguay Round, were in breach of the relevant provisions of the NAFTA. The US argued that these duties were in breach of Article 302 (1) and (2) of the NAFTA. Canada, on the other hand, asserted that the obligation to ‘tariffy’, arising from the Uruguay Round 196 197
Id., paras. 277–278. US – Agricultural Products case, Panel Report, supra note 168.
canons of treaty interpretation
205
agreements (in particular the WTO Agreement on Agriculture), had been incorporated into the NAFTA. In Canada’s view this was done by virtue of Article 710 of the FTA that was incorporated in the NAFTA by Annex 702.1. According to Canada, this provision took precedence over the obligation contained in the Article 302 of the NAFTA not to ‘increase’ or ‘adopt’ any customs duty.198 It is worth mentioning that in order to untangle the complex relationship between the FTA, the NAFTA and the GATT and the agreements of the WTO – in particular the WTO Agreement on Agriculture – the parties to the dispute directed the Panel in its interpretation to the actual wording used, to the object and purpose of the agreements, to the preparatory work and to the practice of the Parties, both contemporaneous and subsequent. The Panel observed that the Parties were not in agreement either as to the relevance of or the weight to be given to these considerations.199 The Panel’s starting point of interpretation of the NAFTA was Article 102 (2). The parties agreed that the applicable rules of international law included in Articles 31 and 32 of the 1969 VCLT were generally accepted as reflecting customary international law. Therefore, according to the rule of interpretation included therein, the Panel started its interpretation by the identification of the plain and ordinary meaning of the words used. In order to do so, the Panel, took into consideration the meaning actually to be attributed to words and phrases looking at the text as a whole, examining the context in which the words appear and considering them in the light of the object and purpose of the treaty. The Panel also stressed the importance of subsequent agreements and subsequent practice as factors to be taken into account in an interpretation. The Panel did not eliminate the possibility of recourse, when necessary, to supplementary means of interpretation, such as preparatory work, according to Article 32 of the 1969 VCLT.200 It should also be noted that the Panel included in its interpretative task the background of the case, i.e. the trade liberalisation agreements, and concluded that the agreements under consideration should be interpreted against these background agreements. The objective of the NAFTA is to eliminate barriers to trade among the Parties. The principles and rules through which the objectives of the NAFTA are elaborated can be found in Article 102 (1): national treatment, the most favoured-nation treatment and transparency. Therefore, the interpretation of the NAFTA provisions by the Panel should aim at promoting rather that impeding the NAFTA objectives and should be based on the premise that exceptions to obligations of trade liberalisation must be approached with caution.201
198 199 200 201
Id., paras. 22–48. Id., para. 117. Id., paras. 118–21. Id., para. 122.
206
malgosia fitzmaurice & panos merkouris
The Panel noted the difficulties regarding the interpretation of these agreements, which stem from the fact that the NAFTA incorporates obligations from numerous other agreements such as the FTA and the GATT. The Panel observed that the terminology used in the drafting of various provisions both within and across these agreements was quite inconsistent and lacking in uniformity. As a result of these inconsistencies, the Panel faced the twofold task first of determining the meaning from the presence of certain words and secondly of “divining meaning from the absence of particular words”.202 The Panel was also mindful of the concerns of Mexico, that the NAFTA be interpreted as an agreement on its own terms and that this interpretation not be governed by the interpretation of the FTA, to which Mexico was not a Party.203 One of the central issues was the question of the temporal application of Article 710 of the FTA, an issue that is not often encountered. Canada argued that the effect of Article 710 of the FTA was to incorporate rights and obligations under the GATT, and under all related agreements, even those subsequent to the entry into force of the NAFTA, into the NAFTA. From this Article follows, according to Canada, the binding force of the obligations under the WTO Agreement on Agriculture. The United States, on the other hand, asserted that Article 710 of the FTA only referred to rights and obligations under the GATT, or under other agreements negotiated under the GATT, existing at the time that the FTA entered into force or, at the most, existing at the time that the NAFTA entered into force.204 Having taken account the incorporation of Article 710 of the FTA into the NAFTA, the Panel was of the view that the date of entry into force was the critical date. However, the Panel decided that the most convenient approach to this issue was first to look at the meaning of Article 710 in the context of the FTA and then to determine whether the meaning was changed when this Article was incorporated into the NAFTA.205 The Panel then proceeded to analyse the meaning of Article 710. The point at issue was the interpretation of the word ‘retain’, the use of which in the view of the US was conclusive. The Panel did not accept the restrictive interpretation offered by the US, since one can retain rights that exist at present, and one can also retain rights under a regime that develops and evolves into the future. Such an example can be found in Article 1608 (2) of the FTA under which “[e]ach Party and investors of each Party retain their respective rights and obligations under customary international law…”. According to the Panel, this evolutionary aspect contained a reference to customary international law. Therefore, the wording of Article 710 of the FTA 202 203 204 205
Id., para. 123. Id., para. 124. Id., paras. 132. Id., para. 133.
canons of treaty interpretation
207
prima facie might be interpreted as having two meanings: either relating only to rights or obligations under the GATT in existence at the date of the incorporation of Article 710 of the FTA or relating both to existing rights and obligations and those that come into existence through subsequently negotiated agreements. The Panel decided to look elsewhere for guidance as to which of the two meanings reflected more closely the intention of the Parties.206 The Panel set about this task in accordance with Article 31 of the 1969 VCLT. The Panel had no doubt that the Parties had other means for limiting the operation of FTA Article 710 to existing rights and obligations if that was their intention. The word ‘existing’, which according to FTA Article 201 meant “in effect at the time of the entry into force of this Agreement”, could have been introduced. ‘Existing’ was used in Article 104 of the FTA, which is a general affirmation of rights and obligations under bilateral and multilateral agreements to which both States are parties. In the light of the above, the use of the word ‘existing’ would have made it clear that the ‘rights and obligations’’ mentioned above were only in existence at the time that the agreement entered into force.207 The Panel researched other similar contexts and noticed that the Parties were able to express their intention to preserve only existing GATT rights and obligations.208 However, analysis of the language of FTA Article 710 clearly showed that such an explicit formulation did not exist. Such an omission indicates that future rights and obligations were not excluded. This reasoning was supplemented by the wording of Article 1801 (2) of the FTA. This Article provides that disputes arising under both this Agreement and the General Agreement on Tariffs and Trade, and agreements negotiated thereunder (GATT) may be settled in either forum, according to the rales of that forum…
The Panel inferred from this formulation that it seemed unlikely that the Parties agreed in this Article that the option of selecting between the GATT and the FTA dispute settlement applied only to disputes arising under the GATT and negotiated agreements, or to the rules of either forum, only as they existed on January 1, 1989; and that no such right would have been relevant in respect of any evolution under the GATT or to any agreements negotiated after that time. The Panel therefore concluded that the terms of FTA Article 710, considered in their context and in light of the object and purpose of the FTA as required by the Vienna Convention, are forwardlooking.209
206 207 208 209
Id., paras. 134–5. Id., para. 136. E.g., Art. 501, 1003 FTA. Id., at para. 145.
208
malgosia fitzmaurice & panos merkouris
The next step in the interpretation was to consider whether the forwardlooking character of FTA Article 710 was changed when it was incorporated into the NAFTA. In order to answer this question, the US regarded the wording of paragraph 4 of NAFTA Annex 702.1 of particular importance.210 This paragraph made no reference to agreements negotiated under the GATT. According to the US, this omission clearly indicated that the parties did not wish to bring WTO Agreements into the NAFTA.211 The Panel did not find the interpretation submitted by the US persuasive. According to the Panel, the inclusion of the NAFTA Annex 702.1 (4) could only be viewed as providing greater certainty. The role of this Annex, as interpreted by the Panel, was to clarify FTA Article 710 by explicitly mentioning the GATT PPA exemptions and the GATT waivers that were not referred to in the text of FTA Article 710 itself. The Panel interpreted the language of Annex 702.1 (4) as inclusive, not exclusive. The Panel also analysed the general circumstances of the conclusion of the NAFTA and did not find anything that would assist in the interpretation of the FTA Article 710 as a NAFTA provision. Therefore, there was no reason to reword this article in order to bring it closer to the terminology of the NAFTA since it retained the same prospective effect it had in the FTA.212 The next question dealt with by the Panel was whether other considerations confirmed or contradicted the interpretation of FTA Article 710 submitted by the Parties. The Panel thus considered whether the consequences that would result from these interpretations assisted the Panel’s analysis whether or not Article 710 was ‘forward-looking’. The US argued that it could not consent to obligations that their content could not be anticipated in advance, whilst Canada characterised an interpretation that would freeze in 1989 a set of obligations that was about to become obsolete as “inherently absurd” and added that the stand adopted by the US placed in conflict both the NAFTA and the WTO Agreement on Agriculture, which should be compatible.213 The Panel was of the view that if the FTA incorporated into the NAFTA rights and obligations under the GATT as of the date the NAFTA came into effect, then Article XI of the GATT that permits certain agricultural quotas as well as GATT PPA exemptions and GATT waivers, would still remain in force
210
Para. 4 NAFTA Ann. 702.1 provides as follows: The Parties understand that Art. 710 of the Canada-United States Free Trade Agreement incorporates the GATT rights and obligations of Canada and the United States with respect to agricultural, food, beverage and certain related goods, including exemptions by virtue of para. (1) (b) of the Protocol of Provisional Application of the GATT and waivers granted under Art. XXV of the GATT.
211 212 213
US-Agricultural Products case, supra note 168, paras. 146–7. Id., paras. 149–50. Id., para. 155–6.
canons of treaty interpretation
209
under the NAFTA as between the US and Canada. That would make rights and obligations of these States different from those in force between other WTO Members. The Panel found it difficult to sustain the frozen-in-time theory, since nothing in the record indicated an intention by either party to preserve agricultural quotas under the NAFTA notwithstanding the outcome of the Uruguay Round. The Parties were mindful that if the Round was successfully completed it would result in new agreements. For that reason the interpretation of FTA Article 710 (as incorporated into the NAFTA), in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose, is that the Parties intended their rights and obligations under both the FTA and the NAFTA not to be limited to those rights and obligations under agreements existing at the time the FTA and then the NAFTA were negotiated.214 This conclusion of the Panel led to the next step in interpretation which had to determine what “particular rights and obligations with respect to agricultural, food, beverage and certain related goods” arising from the Uruguay Round agreements were in fact brought into the NAFTA by virtue of Article 710 of the FTA. The interpretative task of the Panel was to analyse the text of Article 4.2 of the WTO Agreement on Agriculture, i.e., whether this Article imposed the duty to tariffy. This Article reads as follows: [m]embers shall not maintain, resort to, or revert to any measures of the kind which has been required to be converted into ordinary customs duties, except as otherwise provided for in Article 5 and Annex 5.
The Panel stated that on its face value this Article did not create the obligation to tarrify. However, the formulation “which have been required to be converted into ordinary customs duties” indicated the separate existence from Article 4.2 of a provision containing an antecedent obligation to tariffy.215 The Panel decided that, “[s]ince as a matter of general principle some meaning must be attributed to the words in Article 4.2”, it had to look beyond the text of the provision and take into consideration its context, any subsequent agreements or practice and, if necessary rely on supplementary means of interpretation, such as travaux preparatoires of the WTO Agreement on Agriculture and the circumstances of its conclusion generally, according to Articles 31 and 32 of the 1969 VCLT.216 The Panel commenced its task by looking into the negotiations constituting the Uruguay Round. Its objective in relation to agricultural trade was to achieve greater liberalisation of agricultural trade by improving market access
214 215 216
Id., paras. 165–6. Id., para. 171. Id., para. 172.
210
malgosia fitzmaurice & panos merkouris
through, inter alia, the reduction of import barriers. The system of achieving this, as proposed by the US in 1988, was the conversion of non-tariff barriers to ‘tariff equivalents’ (or so-called ‘tariffication’), which was based on the principle of the elimination of non-tariff barriers and the establishment of tariffrate quotas in their place. All relevant documents produced in the course of solving these problems used mandatory language, e.g. “[t]ariff equivalents shall be established…”. This language was not, however, included in Article 4.2 of the WTO Agreement on Agriculture. According to Article 32 of the 1969 VCLT, the relevant documents – i.e., the Dunkell Draft, the Modalities Document and the documents on which they were based – form part of the travaux prepararoires and circumstances of the conclusion of the WTO Agreement on Agriculture. The Panel justified recourse to preparatory work by the obscurity of the content of Article 4.2. The Modalities Document was considered, for the purposes of interpretation, as part of the context of the WTO Agreement on Agriculture, pursuant to Article 31 (2) of the 1969 VCLT. According to Article 31 (3) (b) of the 1969 VCLT the practice of states in tariffying their non-tariff barriers when becoming WTO members could be regarded as “subsequent practice in the application of the treaty which establishes the agreement of the parties regarding its interpretation”. In general, the Panel identified a quid pro quo which involved the removal of non-tariff barriers for the establishment and the application of tariff equivalents. The Panel read the wording of Article 4.2 as nothing more than an imprecise method of referring to a nexus of rights and obligations by which non-tariff barriers were replaced with tariffs, which constituted an equivalent protection through the establishment of over-quota tariffs. This was expressly acknowledged by the US in its Statement of Administrative Action on the implementation of the Uruguay Round Agreements into US law. On the entry into force of the WTO Agreement on 1 January 1995, tariff schedules annexed to the Marrakesh Protocol became schedules to GATT 1994. This package was accepted by both the US and Canada when they became parties to the WTO. Therefore, in the Panel’s view there was no ground for going behind these agreements and questioning the tariffs included in the tariff schedule of Canada. The establishment by the Parties to the WTO Agreement on Agriculture of tariffs to replace nontariff barriers (the tariffication) gave rise to more specific rights and obligations.217 The next step in the interpretation of the Panel was to specify the extent to which these ‘rights and obligations’ were brought into the NAFTA by Article 710 of the FTA. According to the Panel, FTA Article 710 was not limited to non-tariff barriers, since Chapter Seven of the FTA was not restricted exclusively to non-tariff barriers. Similarly, Chapter Seven of the NAFTA is also not restricted to such
217
Id., paras. 171–90.
canons of treaty interpretation
211
a limitation. This interpretation follows from Section A of the NAFTA that applies to “measures adopted or maintained by a Party relating to agricultural trade”.218 Article 201 (1) explains what is understood under a ‘measure.’ It includes “any law, regulation, procedure, requirement or practice”. The interpretation of the wording clearly indicates that it. is not limited to non-tariff barriers and that it can be applied to tariffs. The Panel was mindful that the interpretation of the content of FTA Article 710 had to be undertaken in such a manner as to ensure that it could not be used as a ground for defeating the objectives of the NAFTA as a free trade agreement. The Panel inferred from the particular character of the NAFTA as a free trade agreement, that it provided special rules applicable to trade between its parties that may be different from those applicable multilaterally. From this follows that Article 710 of the FTA could not have been intended to provide for the entire incorporation of GATT rights and obligations relating to agricultural products. The Panel thus adhered to the US views that FTA Article 710 could not be interpreted as a simple substitution of the WTO tariff schedule for the NAFTA tariff schedule.219 A further important result of the Panel’s interpretation was the statement that Article 710 did not incorporate the entire wholesale provisions of the GATT and subsequently of the WTO rights and obligations. It permitted only for the incorporation of those tariffs that resulted from tariffication. In respect of products formerly subject to quotas, the result was that in-quota tariffs on agricultural products applying between Canada and the US were those established under the NAFTA, and tariffs on over-quota imports were those established in the WTO tariff schedule. WTO in-quota tariffs did not enter the NAFTA and they were not part of the tariffication package (the quid quo quo). 1220 The Panel also interpreted the relationship between NAFTA Chapters Three and Seven, i.e. Articles 302 (1) and 701 (2). According to the Panel, the establishment of over-quota tariffs on import products at rates higher than the NAFTA rates for in-quota imports of such products resulted in inconsistency between FTA Article 710 and NAFTA Article 302 (1). In these circumstances, the Panel decided that NAFTA Article 701 (2) was applicable which stipulated that if there was an inconsistency between Section A of Chapter Seven and another provision of the NAFTA, then, according to Article 701 (2), Section A should prevail to the extent of the inconsistency. Article 710 prevailed to the extent of the inconsistency. Therefore FTA Article 710 should prevail.221
218 219 220 221
NAFTA Art. 701 (1). US-Agricultural Products case, supra, note 168, paras. 191–6. Id., para. 201. Id., paras. 202–7.
212
malgosia fitzmaurice & panos merkouris
The general conclusion of the Panel was that Article 710 of the FTA had the effect of bringing into the NAFTA the replacement regime for agricultural non-tariff barriers, established under the WTO as part of the Uruguay Round. This amounted to the obligation not to introduce or maintain such non-tariff barriers and in the right to introduce tariffs that were a result of tariffication (as included in their tariff schedules) to over-quota imports of agricultural products, together with the obligation to reduce these tariffs and ensure minimum volumes of imports. These rights were not subject to Article 302 (1) of the NAFTA.222 C. The Pope and Talbot Inc. and the Government of Canada Case Of relevance to the above issues of interpretation were two Awards in the Pope and Talbot case issued by the Arbitral Tribunal under Chapter Eleven of the NAFTA.223 The point at issue was the interpretation of Article 1105 paragraph 1 of the NAFTA Agreement which states as follows: [e]ach Party shall accord to investments of investors of another Party treatment in accordance with international law, including fair and equitable treatment and full protection and security.
The Arbitral Tribunal elaborated on the interpretation of this provision in the Award on the Merits Phase 2.224 These awards are important – if not groundbreaking – from the following points of view: first the interpretation of the minimum standard of treatment of aliens; secondly the role of the Note by the Free Trade Commission in relation to the interpretation submitted by the Tribunal; thirdly the relationship between the amendment and the interpretation of a treaty. The subject matter of this case was alleged multiple breaches of the NAFTA that resulted from the temporary settlement by Canada of a persistent dispute with the US concerning the alleged subsidization of softwood lumber. The settlement resulted in the introduction by Canada of an export quota system imposed in order to limit quantities of lumber imported to the US. The Claimant alleged breaches of several provisions of the NAFTA that were
222
Id., para. 208. Pope & Talbot Inc. and the Government of Canada, Interim Award of 26 June 2000; Pope & Talbot Inc. and the Government of Canada, Award on the Merits of Phase 2 of 10 April 2001, <www.dfait-maeci.gc.ca/tna-nac/documents/Award_Merits-e.pdf> (accessed on 1 May 2009); Pope & Talbot Inc. and the Government of Canada, Award in Respect of Damages of 31 May 2002. See also the following cases and articles: S. D. Myers Inc. and the Government of Canada, Award of 12 November 2002; ‘Metalclad Judicial Review Proceeding. Court’s Reasons for Judgment issued on 2 May 2001’, British Columbia Law Review 89 (2001): 359; J. Christopher Thomas, ‘Reflections on Article 1105 of NAFTA: History, State Practice and the Influence of Investors’, ICSID Review -FJ1L 21 (2002): 21–101. 224 Pope and Talbot case, Merits Award, supra note 223, interpretation of Art. 1005, at paras. 105–18 and application of Art. 1005, at paras. 120–85. 223
canons of treaty interpretation
213
dismissed (e.g. Articles 1106 and 1102 as well as allegations that the export control amounted to expropriation). Of interest to this study is however the claim that Canada had breached its obligations under Article 1105 of the NAFTA. The Tribunal held that Canada breached the Agreement on this ground, in relation to what was called ‘Verification Review Episode’.225 The investor and Canada agreed that the provision of Article 1105 established a minimum standard of treatment that applied apart from the treatment a NAFTA Party might grant to its own or to other countries’ investors and investments. The disagreement arose as to the content of that minimum standard.226 In brief, Canada submitted that Article 1105 contained the standard of treatment of aliens that was enshrined in customary international law, as this standard was elaborated in the Neer case, according to which breaches must be ‘egregious’ in order to trigger off state responsibility.227 The Investor asserted that the “international law requirements” of Article 1105 included: 1) All sources of international law as enshrined in Article 38 of the Statute of the International Court of Justice; 2) The concept of good faith (including the principle pacta sunt servanda); 3) The World Bank Guidelines on foreign direct investment; 4) The NAFTA Parties’ other treaty obligations; and 5) The body of domestic law of each NAFTA Party that addresses the exercise of domestic regulatory authority.228 The Tribunal did not agree with the Canadian interpretation and said that Canada was addressing the content of the requirements of international law rather than the other conditions of Article 1105; namely “fair and equitable treatment and full protection and security”.229 The Tribunal presented two possible ways in which this provision might be interpreted: one possible interpretation was that the text of Article 1105 suggested that these elements were already included in international law (the view expressed by both Parties).230 A second possible interpretation was that the ‘fairness elements’, i.e. fair and equitable treatment and full protection and security, were additive to the requirements of international law. From this interpretation follows that investors under the NAFTA Agreement were entitled to international minimum standards in addition to the elements of fairness.231 The Tribunal acknowledged that the language of Article 1105 appeared to suggest otherwise, since it stated that fairness elements were included in international law,232 but
225
On this issue see Pope and Talbot case, Merits Award, supra note 223, at paras. 156–81. Id., para. 106. 227 Id., para. 108; Neer Case (US v. United Mexican States), Decision of 15 October 1926, RIAA 4: 60, at 61–2. 228 Id., para. 107. 229 Id., para. 109. 230 Id. 231 Id., para. 110. 232 Id. 226
214
malgosia fitzmaurice & panos merkouris
remarked that an interpretation that plainly departed from the ordinary meaning of the language of Article 1105 was based on the different language used in bilateral investment treaties (BITs) concluded between the US and other states, from which Article 1005 evolved.233 The Tribunal interpreted the language employed in these agreements as “expressly adopting the additive character of the fairness elements”.234 The Tribunal further departed from the 1969 VCLT canons of interpretation, (the interpretation based on a textual analysis), and focused on the intention of the parties that went beyond the text of the Article. To this end, it expressed the view that the US did not furnish evidence that Parties to the NAFTA intended to reject the additive character of the BITs and that the US relied only on the language of Article 1105 in order to prove the intention of the Parties.235 In order to prove that the elements of fairness were additive to the requirements of Article 1105, the Tribunal took into account treaties concluded between the US and other Parties. This interpretative approach did not comply with Article 31 of the VCLT that includes for the purpose of the interpretation of a treaty any agreements or instruments concluded between parties to a treaty, not with other states outside the treaty framework. The Tribunal, having examined treaties concluded with states other than NAFTA Parties, came to the conclusion that it was unlikely that NAFTA Parties intended to limit the scope of Article 1105 as between themselves when they had granted broader rights to States outside the NAFTA.236 The facts however were different: …the history of North American investment relations actually is the opposite. In view of the periodic investment irritants with their more powerful neighbour, until NAFTA neither Canada nor Mexico had previously shown any enthusiasm for negotiating treaty-based investment protections and they approached investment negotiations with caution. … Article 1105…comported with the United States’ understanding that fair and equitable treatment and full protection and security referred to customary international law.237
The assumed lesser treatment of NAFTA investors as compared to entitlements granted to other States’ investors prompted the Tribunal to conclude that this interpretation lead to “a patently absurd result”.238 Therefore, the Tribunal relied on Article 32 of the 1969 VCLT. As was observed:
233
E.g., the US 1987 Model BIT adopts in Art. II.2 the following formulations: [i]nvestment shall at all times be accorded fair and equitable treatment, shall enjoy full protection and security and shall in no case be accorded treatment less than required by international law.
234 235 236 237 238
Pope and Talbot case, Merits Award, supra note 223, para. 111. Id., para. 114. Id., para. 115. Thomas, supra note 223, at 78–9. Pope and Talbot case, Merits Award, supra note 223, para. 118.
canons of treaty interpretation
215
…the ‘absurdity’ that the Tribunal thought it had established resulted from an incorrect application of the Vienna Convention Article 32. That Article is applied only when the ‘interpretation according to Article 31 (which in this case is applied to Article 1105) leads ‘to a result which manifestly absurd or unreasonable’ Here, the absurdity was found not from a good faith interpretation of the ordinary meaning of the terms of Article 1105 in their context and in light of the NAFTA’s objective and purpose, but rather from referring to treaties between a NAFTA Party and other states.239
The interpretative techniques used by the Tribunal resulted in a decision on minimum international standards in which additive elements exceeded the requirements of international law, “without any threshold limitations that the conduct complained of be ‘egregious,’ ‘outrageous,’ or ‘shocking,’ or otherwise extraordinary”.240 The Parties to the NAFTA found the Award of the Arbitral Panel difficult to accept. Therefore, they had recourse to Article 1131 (see above) that provides for the opportunity to exercise control over the interpretation of the NAFTA Agreement Chapter Eleven, by jointly agreeing on interpretation. The gist of this Article is that Parties may agree in their capacity as Members of the Free Trade Commission that an interpretation of the Agreement needs to be provided, and that this interpretation will be binding on the Tribunal. The Note on Interpretation included the following: B. Minimum Standard of Treatment in Accordance with International Law 1. Article 1105 (1) prescribes the customary international minimum standard of treatment of aliens as the minimum standard of treatment of investors of another Party. 2. The concept of ‘fair and equitable treatment’ and ‘full protection and security’ do not require treatment in addition to or beyond which is required by customary international law minimum standard of treatment of aliens.241
The Tribunal in connection to the Note, asked for certain clarifications from the Parties, most importantly on the issue of the retroactive effect of the Commission’s ruling on the decision made previously by the Tribunal. The principal issues for the determination by the Tribunal were as follows: 1) whether the interpretation put forward by the Commission was a valid exercise of the Commission’s power and as such binding on the Tribunal; 2) if answered in the affirmative what would the effect of the interpretation be in relation to awards already made by the Tribunal (the retroactivity issue); and 3) the nature and content of customary international law in the context of
239
Thomas, supra note 223, at 79–80. Pope and Talbot case, Merits Award, supra note 223, para. 118. See the critical remarks on the interpretation applied in this case by Honourable Mr. Justice Tysoe, in the Metalclad Judicial Review case, British Columbia Law Review 89 (2001). 241 Pope and Talbot case, Damages Award, supra note 223, at 5–6. 240
216
malgosia fitzmaurice & panos merkouris
Article 1105 and its application to the facts of the case.242 An additional question was whether Canada, against which the ruling on Article 1105 was made, could be part of the revised interpretative process under the light of the principle that no-one shall be judge in his own cause and in order to assure due process before an impartial Tribunal.243 As to the retroactivity issue, the Tribunal stated that the phrase ‘shall be binding’ in Article 1103 paragraph 2 should be understood as mandatory rather than prospective. Viewed as such “it is incumbent on the Tribunal to assess the impact of the Interpretation upon its prior findings with respect to Article 1105”.244 The Note on Interpretation assessed fairness elements as subsumed in customary international law rather than as additive to it. The Tribunal in its interpretation rejected the static concept of customary law (therefore it did not agree with the interpretation presented by Canada, i.e. that principles of customary law were frozen in the amber of time at the Neer decision). This approach would amount to the award of damages only in cases of conduct found to be egregious or a failure to meet internationally required standards.245 The Tribunal made a very interesting observation that it did not adhere to a static but to an evolutionary concept of customary international law.246 This approach influenced its interpretation and was in line with the interpretative techniques of the WTO’s AB. The Tribunal stated that there had been considerable evolution in the field of customary international law since the 1920s which had evolved through state practice. The Tribunal further observed that international agreements constitute State practice and contribute to the creation of customary law!247 Further, the Tribunal noted that since the 1920s, the range of actions subject to international concern acquired a broader dimension and exceeded the scope of international delinquencies, considered in the Neer case to include the concept of fair and equitable treatment. This development was acknowledged by the OECD in its Draft Convention on the Protection of Foreign Property that stated that this concept was already customary in BITs then in force. The Tribunal in particular emphasised two points included in the commentary to the draft: the first point
242
Id., at 9. Id., at 8. 244 Id., para. 51. 245 Id., para. 57. 246 Id., para. 58. See also Sir Robert Jennings’, expert opinion in the Methnex v. United States that the Neer case was concerned not with fair and equitable treatment but whether the State concerned had committed an ‘international delinquency.’ In the Matter of Arbitration under Chapter 11 of the NAFTA and the UNCITRAL Rules, between Methanex Corporation and the United States of America, Decison. of Tribunal on the Petition from Third Persons to intervene Amici Curiae of 15 January 2001. 247 Id., para. 59. The Tribunal did not comment on the role of the opinio iuris in the formation of customary law. 243
canons of treaty interpretation
217
was that fair and equitable treatment required treatment at least as good as that accorded by a State to its own nationals; the second point was that the above mentioned concept was embodied in customary international law. The tribunal further explained that many of BITs required state conduct to be evaluated under the fairness elements, apart from the standards of customary law. Even those BITs that did not provide that these elements were owned independently of the requirements of international customary law, added the fair and equitable treatment protection of these rights that previously were protected by international customary law. Therefore, BITs enlarged the scope of protection beyond international delinquencies.248 The Tribunal, however, did not address the question of the content of Article 1105 as including ‘additive’ standards and stated only that, according to the most favoured nation clause (MFN) contained in Article 1103, standards of BITs could be derived. The Tribunal did not find it necessary to make such a pronouncement.249 The Tribunal therefore applied the strict standards of Canada embodied in the Interpretation Note and used the Canadian term ‘egregious’ behaviour. It found that Canada, even having applied this limited standard, violated Article 1105. According to the Tribunal the actions of Canada’s Soft Lumber Division (SLD) “would shock and outrage every reasonable citizen of Canada, they did shock and outraged the Tribunal”.250 For these reasons, the Tribunal concluded that the conduct of the SLD in the Verification Episode breached the fair and equitable treatment requirement of Article 1105, even having applied the strict and limited standards of Canada’s formulation.251 As to the assessment of the Interpretation Note by the Commission, namely whether it was an amendment or an interpretation, the Tribunal refrained from pronouncing on this subject and said only, in very cautious terms that: …were the Tribunal required to make a determination whether the Commission’s action is an interpretation or an amendment, it would choose the latter. However, for the reasons discussed below, this determination is not required. Accordingly, the Tribunal has proceeded on the basis that the Commission’s action was an ‘interpretation’.252
D. The Canadian Cattlemen for Fair Trade (CCFT) v. The United States of America Case The claimants were all Canadian nationals engaged in beef and cattle business and argued that since 20 May 2003 the United States had maintained
248 249 250 251 252
Id., paras. 60 and 61. Id., paras. 9 and 66. Id., para. 68. Id., para. 69. Id., para. 47.
218
malgosia fitzmaurice & panos merkouris
prohibitions and restrictions on Canadian-origin livestock and beef products, most notably an absolute prohibition on the shipment of live cattle from Canada under the authority of the Animal Health Protection Act. This ban was enforced, according to the claims of the CCFT in spite of abundant evidence that Canadian-origin livestock and beef products were safe and posed minimal risks of Bovine Spongiform Encephalopathy (BSE) to the United States. They maintained, thus, that the United States is obligated under NAFTA Article 1102(1) to accord national treatment to Canadian investors with respect to their investments in Canada, that the ban breached this obligation, and that by reason of this alleged breach they incurred losses when the profitability and value of their cattle-related investments in Canada decreased.253
On the other hand the US argued, first, that the entirety of the measures adopted were completely within their ambit of discretion, especially after the discovery of BSE in a cow in Alberta, Canada; secondly, that the CCFT was not eligible to have their claims resolved under Chapter Eleven because their investments were in Canada and, thus, the US had no “obligation to provide national treatment to those investments, nor the obligation to arbitrate claims relating to them”.254 The Parties to the dispute agreed on a bifurcated procedure to the effect that the tribunal instead of dealing with the entirety of the case, i.e. jurisdictional issues and merits, all in one judgment, it would pass judgment only on a ‘Preliminary Issue’ which was defined as: Does this Tribunal have jurisdiction to consider claims under NAFTA Article 1116 for an alleged breach of NAFTA Article 1102(1) where all of the Claimants’ investments at issue are located in the Canadian portion of the North American Free Trade Area and the Claimants do not seek to make, are not making and have not made any investments in the territory of the United States of America?
What this meant, was that the entirety of the case, essentially, revolved around one single issue of interpretation; that of the meaning of Article 1102(1) and, more precisely, whether the notion of ‘investment’ was part and parcel of that of ‘investor’ or whether Chapter Eleven could be interpreted in such a way, that an investor could be isolated from his investment.255 It comes as no surprise that 253 In the Consolidated Arbitration under Chapter Eleven of the North American Free Trade Agreement (NAFTA) and the UNCITRAL Arbitration Rules between the Canadian Cattlemen for Fair Trade (CCFT) and the United States of America, Award on Jurisdiction of 28 January 2008, para. 3 (hereinafter CCFT v. US case), <www.naftaclaims.com/Disputes/USA/CCFT/CCFTUSA-Award.pdf> (accessed on 1 May 2009). 254 Id. 255 It must be reiterated that the main crux of this case lied in the fact that the CCFT consisted of investors of a Party to NAFTA (Canada) that had not invested, and did not intend to invest, in another NAFTA Party (i.e. the US), but had invested only in the territory of their home State.
canons of treaty interpretation
219
this was also reflected in the text of the judgment, which is in its overwhelming majority dedicated to an analysis of the interpretative process and application of the principles of interpretation to the legal texts relevant to the case. Building upon the aforementioned, it is no coincidence that this interpretation-driven stance of the court is mirrored in the structuring of the judgment; whereas in most cases after a brief presentation of the facts and the claims from claimants and respondents, the court will then proceed to the analysis of each and every one of the claims raised, in the CCFT v. US case the Tribunal constructed its judgment in sections, each of which does not correspond to a different claim but to a different element of treaty interpretation. Here are, for instance, some of the titles of the relevant sections: “I.II. Ordinary Meaning in Light of the Context and Object and Purpose of the NAFTA and of its Negotiating History (Article 31(1) and 32 VCLT)”, “I.III. Subsequent Agreement and Practice (Article 31 (3) VCLT)” and most notably “I. VI. Decisions in Other Cases as Supplementary Means of Interpretation (Article 32 VCLT)”.256 However, the importance of CCFT v. US does not lie in the structure of its judgment. A number of the Tribunal’s dicta are also of high academic and practical value. Let us examine the title used in Section I.II of the judgment i.e. “Ordinary Meaning in Light of the Context and Object and Purpose of the NAFTA and of its Negotiating History (Article 31(1) and 32 VCLT)”. The structure of this Section, and the corresponding analysis would seem to follow verbatim the order both internal (text, context, object and purpose) and external (Article 31 followed by Article 32) of the relevant articles on treaty interpretation of the VCLT. Nevertheless, a closer look will show that this is just a mirage. In that section when analysing the notion of context, the Tribunal introduces also the element of “negotiating history”,257 and to top that the object and purpose is analysed after the ‘negotiating history/context’, while another part of Article 32, ‘other supplementary means’, is analysed at a later section. The way the Tribunal proceeded, although seemingly not following the structure of Articles 31 and 32, actually captures the very essence of the judicial interpretative process and is more in tune with the true spirit of the principles of interpretation enshrined in Articles 31 and 32 of the VCLT. Text, context, object and purpose, preparatory work are not in a hierarchical relationship with each other, based on the order in which they are mentioned in the relevant articles of the VCLT. As the drafters of the VCLT themselves, made clear, this was a logical/temporal sequence258 not a hierarchical one.259
256
regarding the innovation and importance of this last section see infra. see CCFT v. US, supra note 253, Section I.II.4.3.d.iii, para. 155 et seq. 258 and one not carved in stone for that matter. 259 Sir Humphrey Waldock, the Special Rapporteur, responding to Professor Ago’s claim that the seperation of Articles 31 and 32 (at that stage of drafting they were Articles 69 and 70, 257
220
malgosia fitzmaurice & panos merkouris
It was a suggestion of a process that seemed logical, but in no way was it meant to bind the judges hand and foot. Consequently, in the present case, this seemingly chaotic intermingling of the various elements of treaty interpretation is not in contradiction with the way that Articles 31 and 32 should be applied. It is actually in conformity. It is not a violation but a validation of Articles 31 and 32 of the VCLT. Shifting now our analysis to the individual elements, the Tribunal made a very interesting pronouncement on the issue of object and purpose of Chapter Eleven. First, it established that Chapter Eleven is not a stand-alone chapter, but part of a larger, much more complex treaty, the various parts of which are in a constant state of interdependence and amphidromous influence. 164. NAFTA Chapter Eleven cannot be reduced simply to an ordinary bilateral investment treaty… it functions as part of a larger agreement that requires analysis of how the different parts of the treaty interrelate in a range of circumstances. 165. Chapter Eleven itself reflects this interdependence thoroughly. Beginning with its gateway, Article 1101, which carves out measures covered by Chapter 14 (Financial Services), Chapter Eleven is replete with provisions showing its interconnectedness to other parts of the NAFTA. Article 1112(1) makes Chapter Eleven in its entirely subordinate to the provisions of other NAFTA chapters where there are inconsistencies – thereby establishing the supremacy of other dispute resolution mechanisms in the event of conflict.15 Article 1115 contains a savings clause that insures that the rights and obligations of the NAFTA States Parties under the dispute settlement provisions of Chapter 20 of the treaty are preserved, subject to the provisions of Article 1138. The latter article also clarifies the relationship between Chapter Eleven and national-security based actions under Article 2102. A number of other provisions of Chapter Eleven, including Articles 1116 and 1117, discussed earlier, import provisions from outside the
respectively), made it clear that that those two phases of interpretation succeeded each other (see YBILC (1966) Vol. I Part II: 205, para. 23), made the following enlightening statement: “the process of interpretation was essentially a simultaneous one, though logic might dictate a certain order of thought” (emphasis added); ILC, 873rd Meeting: Law of Treaties, (A/CN.4/186 and Addenda; A/CN.4/L.107, L.115), reproduced in YILC (1966), Vol. I Part II: 206, para. 36. The ILC, also, in its Commentary made some revealing comments as well as to the relationship of the various elements of treaty interpretation: Those observations appeared to indicate a possible fear that the successive paragraphs of article 27 might be taken as laying down a hierarchical order for the application of the various elements of interpretation in the article. … [T]he application of the means of interpretation in the article would be a single combined operation. All the various elements, as they were present in any given case, would be thrown into the crucible, and their interaction would give the legally relevant interpretation.
and most importantly that the Commission] considered that the article, when read as a whole, cannot properly be regarded as laying down a legal hierarchy of norms for the interpretation of treaties. The elements of interpretation in the article have in the nature of things to be arranged in some order. But it was considerations of logic, not any obligatory legal hierarchy, which guided the Commission in arriving at the arrangement proposed in the article. (emphasis added)
ILC, Report of the ILC on the Work of its 18th Session: Draft Articles on the Law of Treaties with Commentaries, (A/6309/Rev.1), reproduced in YILC (1966), Vol. II: 218, paras. 8–9.
canons of treaty interpretation
221
Chapter. And provisions outside of Chapter Eleven – for example, the provisions of Article 2103 – affect an investor’s exercise of the rights granted by Chapter Eleven.260
Having established that, the Tribunal then analysed the phenomenon of the parallel existence of several objects and purposes, i.e. object and purpose of a treaty and object and purpose of a specific provision (or chapter) within that treaty and how one can identify these. 166. Thus, it is clear to this Tribunal that NAFTA Chapter Eleven cannot be viewed in isolation but must be considered in light of its larger context – the NAFTA as a whole. However, that does not mean that Chapter Eleven itself must bear the whole weight of the diverse purposes set out in Article 102. Those purposes, it is clear, apply to the treaty in its complex entirety, and some are wholly irrelevant to Chapter Eleven. As one Chapter Eleven Tribunal, the ADF Tribunal, has stated, the NAFTA’s overall purpose clause is akin to a lex generalis, while the particular chapter is a lex specialis… The corollary is that particular segments of the treaty may reflect a much more limited set of purposes than the overall purposes clause sets forth. As Chapter Eleven lacks its own object and purpose clause, as a stand-alone investment treaty would likely have, its meaning must be discerned principally from textual analysis of the lex specialis and other relevant context.261
Essentially, the Tribunal resolved the issue of different objects and purposes by invoking the tool of lex specialis and lex generalis. According to the judges in the CCFT v. US each provision of a larger treaty does not and cannot reflect the totality of the purposes of the treaty. Similarly, Chapter Eleven of NAFTA does not share all the purposes of NAFTA as stated in Article 102. This, of course, does not mean that there could be a conflict between the purposes of NAFTA and Chapter Eleven and that is exactly why the Tribunal suggested that the relationship existing between these two sets of objects and purposes is that of lex generalis and lex specialis.262 The objects and purposes of NAFTA, thus, offer a backdrop for the identification and/or clarification of the corresponding purposes of each Chapter.263 Since Chapter Eleven does not have an ‘object and purpose’ clause, devoted specifically to it, its ‘object and purpose’ will be identified through a textual analysis of the Chapter, its context and of course with the help of the already existing clause of objects of NAFTA, the lex generalis, incorporated in Article 102. Here, again, the interconnectedness of the various elements of treaty interpretation is evident. Since the object and
260
CCFT v. US, supra note 253, paras. 164–5. Id., para. 166. 262 the Tribunal finds support for this in: In the Matter of an Arbitration under Chapter Eleven of the North American Free Trade Agreement between ADF Group Inc. and the United States of America, Award of 9 January 2003, ICSID Case No. ARB(AF)/00/1, para. 147, (hereinafter ADF Group v. US), <www.state.gov/documents/organization/16586.pdf> (accessed on 1 May 2009). 263 for the stance adopted in the WTO Panels and AB, see supra Section III. 261
222
malgosia fitzmaurice & panos merkouris
purpose is not clearly stated, we resort to all the other elements in order to identify it. A further point that was raised in the proceedings was whether there was a subsequent agreement between the parties or subsequent practice regarding the interpretation of Article 1101(1)(a).264 The relevant provisions of the VCLT are Articles 31(3)(a) and (b), which state: 181. Article 31(3) There shall be taken into account, together with the context: (a) any subsequent agreement between the parties regarding the interpretation of the treaty or the application of its provisions; (b) any subsequent practice in the application of the treaty which establishes the agreement of the parties regarding its interpretation…. (emphasis added)
The US argued that all three NAFTA Parties had confirmed its interpretation of Chapter Eleven’s purpose, i.e. to protect investors and investments with respect to another NAFTA Party’s territory. As for practice, the US mentioned the US Statement of Administrative Action which stated that Chapter Eleven “applies where such firms or nationals make or seek to make investments in another NAFTA country”;265 the United States’ Congressional Budget Office’s analysis of the NAFTA; the Canadian Statement on Implementation of the NAFTA; Canada’s own statement in the S.D. Myers arbitration, where it reiterated that “Chapter Eleven applies only to investors that have, or are seeking to make, investments in the territory of the disputing Party”;266 and Mexico’s similar statement in the Bayview arbitration that Chapter Eleven does not intend “to protect […] the property of one state[’s nationals] in that same state”.267 The Claimants on their part argued that no such subsequent agreement existed, all the more so since the Parties never formally agreed upon the interpretation of the NAFTA through the mechanism NAFTA itself provides, i.e. the Free Trade Commission pursuant to Article 2001(2).268 As for the subsequent practice, the CCFT argued that the US statements were not sufficient, whereas the Canadian and Mexican statements in the S.D. Myers and Bayview arbitration, respectively, had no bearing upon the present case, as the factual background of those cases was completely different.269 The Tribunal, on its part, first laid down the theoretical foundations of Article 31(3)(a) and (b) of the VCLT. Citing Sir Ian Sinclair it stated: [t]he value and significance of subsequent practice will naturally depend on the extent to which it is concordant, common, and consistent. A practice is a sequence
264 The text of the Article goes as follows: “Article 1101(1)(a): Scope – 1. This Chapter applies to measures adopted or maintained by a Party relating to: (a) investors of another Party;” 265 CCFT v. US, supra note 253, para. 174. 266 Id., para. 175. 267 Id., para. 176. 268 Id., para. 179. 269 Id., para. 180.
canons of treaty interpretation
223
of facts or acts and cannot in general be established by one isolated fact or act or even by several individual applications.270
and that: reference may be made to “subsequent practice that clearly establishes the understanding of all the parties regarding its interpretation”271 … The Tribunal is, furthermore, of the view that, as the rule states, any ‘subsequent agreement’ or ‘subsequent practice’ of the kind contemplated by the rule ‘shall be taken into account, together with the context’ of the treaty terms subject to interpretation.272
Essentially, what is required of subsequent practice, to be relevant for the purposes of Article 31(3)(b) VCLT is to satisfy, to a certain extent, three criteria. It has to be: – concordant – common, and – consistent But before turning to the issue of subsequent practice, the Tribunal, first, had to deal with whether a subsequent agreement existed. This top – bottom approach makes sense and is actually common practice. If a common set of facts is used to claim both ‘subsequent agreement’ and ‘subsequent practice’ the relevant tribunal will make a judgment on the former, which has much stricter requirements, and only if these are not satisfied turn to ‘subsequent practice’ and see if the less severe criteria mentioned above are met.273 In the present case, the Tribunal held that although the instances brought forth by the Respondent were “certainly suggestive of something approaching an agreement…[it did] not rise to the level of a ‘subsequent agreement’ ”,274 and unlike The ADF Group v. US case, here no document or fact was deemed as revealing a common intention of the parties reaching the level of ‘subsequent agreement’.275 An interesting point of the judgment was the fact that the Tribunal opted not to equate the term ‘agreement’ of Article 31(3)(a) with ‘understanding’. In the Kasikili/Sedudu Island case, Judge Weeramantry, deviating from the position adopted by the majority of the Court, argued that “the
270 Id., para. 182, citing. Sir Ian Sinclair, The Vienna Convention on the Law of Treaties (Manchester: MUP, 1984), 137. 271 Sir Ian Brownlie, Principles of Public International Law (Oxford: OUP, 1998), 635. 272 CCFT v. US, supra note 253, para. 183. 273 that is, being concordant, common and consistent. For a more detailed analysis of whether there is or not a hierachical relationship between Article 31(3)(a) and (b) see Lady Hazel Fox, ‘Article 31(3) (a) and (b) of the Vienna Convention and the Kasikili/Sedudu Island Case’: Section III, published in the present Volume. 274 CCFT v. US, supra note 253, para. 187. 275 In more detail, the Tribunal in The ADF Group v. US case had to deal with a document, the ‘FTC Interpretation of 31 July 2001’. This document, was accepted by all parties and was
224
malgosia fitzmaurice & panos merkouris
word ‘agreement’ in subpara. 3(a) of the Treaty Convention can be read in the sense of ‘understanding’, and can therefore cover silence and inaction as well”.276 However, this equation was based on certain statements made by Sir Humphrey Waldock during the travaux préparatoires of the VCLT, which were aimed at the term ‘understanding’ and not the term ‘agreement’ which was later adopted.277 The Tribunal in CCFT v. US, correctly, refrained from setting ‘agreement’ and ‘understanding’ on the same level when it held that: the Tribunal notes the absence of any Article 1128 submission by Canada before this Tribunal. This cannot be seen as evidence of Canadian support for the Claimants’ position on this issue, but it also cannot be seen as evidence of Canadian opposition. The Tribunal concludes that there is no ‘subsequent agreement’ on this issue within the meaning of Article 31(3)(a) of the Vienna Convention.278
Having ruled on the non-existence of a ‘subsequent agreement’ the Tribunal then, foreseeably, turned to the question of whether the instances put forward could amount to ‘subsequent practice’ under Article 31(3)(b). 188. The question remains: is there ‘subsequent practice’ that establishes the agreement of the NAFTA Parties on this issue within the meaning of Article 31(3)(b)? The Tribunal concludes that there is. Although there is, to the Tribunal, insufficient evidence on the record to demonstrate a ‘subsequent agreement between the parties regarding the interpretation of the treaty or the application of its provisions’, the available evidence cited by the Respondent demonstrates to us that there is nevertheless a ‘subsequent practice’ in the application of the treaty which establishes the agreement of the parties regarding its applications…. 189. On the record before this Tribunal … there is evidence of a sequence of facts and acts that amounts to a practice that is concordant, common, and consistent…. [which] confirms the Tribunal’s interpretation of the ordinary meaning of Article 1101(1)(a) of the NAFTA, as set out above.279
considered by the Tribunal to reflect the true intentions of the Parties as to the interpretation of certain provisions of NAFTA and, thus, “[n]o more authentic and authoritative source of instruction on what the Parties intended to convey in a particular provision of NAFTA, is possible”; ADF Group v. US, supra note 262, para. 177. However, the CCFT v. US case was different in the sense that no such document existed and that is probably the reason why the US tried to tie the aforementioned quote from the ADF Group v. US case, not so much with ‘subsequent agreement’ but more with ‘subsequent practice’; see CCFT v. US, Transcripts of Hearing of 9 October 2007, p. 47, <www.naftaclaims.com/Disputes/USA/CCFT/CCFT-USA-Transcript% 20Day-01.pdf> (accessed on 1 May 2009), and CCFT v. US, supra note 253, para. 177. 276 Case Concerning Kasikili/Sedudu Island (Botswana v. Namibia), Judgment of 13 December 1999, Dissenting Opinion of Judge Weeramantry, 1999 ICJ Rep., para. 23. (hereinafter Kasikili/ Sedudu Judgment) 277 Thirlway’s statement was the following: “the assent of a party to a transaction may be inferred from its reaction or absence of reaction to the practice”, Kasikili/Sedudu Judgment supra note 5, Dissenting Opinion of Judge Weeramantry, para. 24; in more detail on this issue see analysis by Lady Hazel Fox, ‘Article 31(3) (a) and (b) of the Vienna Convention and the Kasikili/ Sedudu Island Case’: Section III, published in the present Volume. 278 CCFT v. US, supra note 253, para. 187 (emphasis added). 279 Id., paras. 188–9. (emphasis added).
canons of treaty interpretation
225
Consequently, the court remained true to the application of the three criteria to the facts produced by the parties, i.e being concordant, common and consistent, in order to substantiate the existence of the ‘subsequent practice’ of Article 31(3)(b). Despite the theoretical value of all the aforementioned passages of the CCFT v. US judgment, it is actually Section I.VI. which steals the limelight. The issue that that Section deals with can be seen from the very start. The title of Section I.VI. goes as follows: ‘Decisions in Other Cases as Supplementary Means of Interpretation (Article 32 VCLT)’. The Tribunal in CCFT v. US is one of the very few courts that have expressly acknowledged judicial decisions as other supplementary means. It is, also, revealing to go through the reasoning of the court on this matter extensively: 50. On the other hand, Article 32 VCLT permits, as supplementary means of interpretation, not only preparatory work and circumstances of conclusion of the treaty, but indicates by the word ‘including’ that, beyond these two means expressly mentioned, other supplementary means may be applied. Article 38 [paragraph 1.d.] of the Statute of the International Court of Justice provides that judicial decisions are applicable for the interpretation of public international law as ‘subsidiary means’. Therefore, they must be understood to be also supplementary means of interpretation in the sense of Article 32 VCLT. 51. That being so, it is not obviously clear how far arbitral decisions are of relevance to the Tribunal’s task. It is at all events plain that the decisions of other tribunals are not binding on this Tribunal, and the Tribunal refers in this connection to paragraphs 73–76 of the Decision on Jurisdiction in Bayindir Insaat Turizm Ticaret v. Islamic Republic of Pakistan of November 14, 2005 (ICSID Case No. ARB/03/29). This does not, however, preclude the Tribunal from considering other arbitral decisions and the arguments of the Parties based upon them, to the extent that it may find that they throw useful light on the issues that arise for decision in this case. Such an examination will be conducted by the Tribunal later in this Award, after the Tribunal has considered the Parties’ contentions and arguments regarding the various issues argued and relevant to the interpretation of the NAFTA provisions at stake.280
Article 32 of the VCLT deals with the so-called supplementary means of interpretation. However, the list of supplementary means, mentioned in Article 32 – travaux préparatoires and ‘circumstances of conclusion’ of a treaty – is clearly not a numerus clausus. The use of the term ‘including’ in the text of Article 32 is pretty straightforward. It was the clear intention of the drafters of the VCLT to allow the judge certain flexibility in the interpretative process. As can be seen from the travaux préparatoires of the VCLT, there was originally a huge debate as to whether rules of interpretation existed, let alone should be included in the VCLT.281 Some argued in favour of the existence of ‘principles’ 280
CCFT v. US, supra note 253, paras. 50–1 (emphasis in original). – ILC, 726th Meeting: Law of Treaties, (A/CN.4/167), reproduced in YILC (1964), Vol.I: 20, para. 4, 15, 24, 28; ILC, 765th – 766th Meetings: Law of Treaties, (A/CN.4/167/Add.3), reproduced 281
226
malgosia fitzmaurice & panos merkouris
of interpretation, not ‘rules’, while others feared that putting any such rules down in black-letter law, might make them to rigid, and eventually hinder the work of the judges.282 In this climate, one can understand the effort that the drafters put in ensuring a level of flexibility for the rules of interpretation they would end up with. Article 32 is but one more example of that effort. What is, nevertheless, strange is that the term ‘including’ although not an object of controversy, was not an object of discussion either. The entirety of debate focused on the notion of travaux préparatoires and its various complexities.283 This, on the one hand, gives the judges a relative ‘clean slate’ to work with, on the other hand, though, it deprives the judges of any basic guideline with which to work. International courts and tribunals have been hesitant in making pronouncements on whether their own previous judgments or those of other courts can be used for interpretative purposes through an application of Article 32.284 It is exactly this situation, which makes the stance of the Tribunal in CCFT v. US all the more interesting. First, the Tribunal reiterates that the term ‘including’ clearly indicates that there are other supplementary means available to the judges, apart from travaux préparatoires and ‘circumstances of conclusion’ of a treaty. So far everything is pretty self-evident. It is logical also to infer, that since the Tribunal focuses on the term ‘including’ it feels that judicial decisions fall neither in the group of travaux préparatoires nor in that of ‘circumstances of conclusion’.285 The Tribunal then, in order to substantiate why judicial decisions are supplementary means within the meaning of Article 32, invokes the Statute of the ICJ. It is, at this point, where things become complex. What the Tribunal is doing is, in essence, an interpretation of Article 32, or if we may be allowed the pun: Ruling on the interpretation of the rule on interpretation. But, how does one interpret Articles 31 and 32? Do they apply to themselves, as such,
in YILC (1964), Vol. I: 275 et seq.; also see ILC, Draft Articles on the Law of Treaties with Commentaries, YILC (1964), Vol.II: 176, at 199–201. 282 Id. 283 For a general overview, see Franciscus A. Engelen, Interpretation of Tax Treaties under International Law: A Study of Articles 31, 32 and 33 of the Vienna Convention on the Law of Treaties and their Application to Tax Treaties (Amsterdam: IBFD, 2004), 293 et seq. 284 this is not to say, of course, that they have refrained from quoting judgments of other courts, a phenomenon more known as judicial cross-fertilization. On this issue, see Francis G. Jacobs, ‘Judicial Dialogue and the Cross-fertilization of Legal Systems: the European Court of Justice’, Texas International Law Journal 38 (2003): 547–56; Anne-Marie Slaughter, ‘A Global Community of Courts’, Harvard International Law Journal 44 (2003): 191–219; Société Française pour le Droit International, La juridictionalisation du Droit International: Colloque de Lille (Paris: Pedone, 2003). 285 we shall return to the latter, infra, when we juxtapose the CCFT v. US with the relevant jurisprudence of the WTO.
canons of treaty interpretation
227
which may lead to a self-referential cycle, going on in perpretuity? The VCLT itself offers a way out, through Article 4: Article 4 Non-retroactivity of the present Convention Without prejudice to the application of any rules set forth in the present Convention to which treaties would be subject under International law independently of the Convention, the Convention applies only to treaties which are concluded by States after the entry into force of the present Convention with regard to such States.286
Ergo, the VCLT and its rules cannot and do not apply to themselves. Consequently, in interpreting Article 32 we would have to resort to customary international law, which, however, as has been pronounced in many judicial decisions is enshrined in Articles 31 and 32.287 Of course, in the present case, these considerations would not be of relevance, as the United States has not ratified the VCLT, therefore customary law would be applicable ab initio. Nevertheless, they would serve for a dispute involving any of the other two Parties to the NAFTA, Canada and Mexico who are bound by the VCLT.288 Based on the above analysis, the Tribunal has to apply customary principles of interpretation. However, since these principles are considered to be codified in Articles 31 and 32, the Tribunal adopts a common approach; it uses the text of the latter as a template for their interpretation of what is to be included in the notion of ‘supplementary means’. Because the Tribunal uses the terminology of Article 32, its interpretation although unusual – the norm is to interpret other provisions, not the provisions on interpretation themselves – does not seem so groundbreaking. However, if we remind ourselves, that the Tribunal is applying not Article 32 as such, but its customary law – equivalent,
286
Article 4 of the VCLT. the reason why this solution breaks the self-referential cycle, is a highly complex issue, a full analysis of which exceeds the scope of this article. However, briefly, the reason is that we abandon the text and go to custom. This is beneficial for two reasons: i) custom may have a different normative content than that of text as stated in the Nicaragua case (Case Concerning Military and Paramilitary Activities in and against Nicaragua (Nicaragua v. United States of America), Judgment of 27 June 1986, 1986 ICJ Rep. 14., paras. 175 et seq.) and more importantly ii) interpretation through custom cannot reach an impasse, because there is always the possibility of resorting to identification of custom i.e. through practice and opinio juris. This last means, that any unresolved issues as to what exactly constitutes ‘supplementary means’ will be resolved at the level of custom and then superimposed on Articles 31 and 32. 288 Canada acceded to the VCLT on 14 October 1970 and Mexico ratified it on 25 September 1974. Information available at: (accessed on 1 May 2009). The treaty entered into force on 27 January 1980, and thus applies to the NAFTA, based on the aforementioned Article 4 of the VCLT, since NAFTA was signed on 17 December 1992 and entered into force on 1 January 1994. 287
228
malgosia fitzmaurice & panos merkouris
then the true importance of this process is revealed. The Tribunal is interpreting custom. We have left the realm of identification of custom (through practice and opinio juris) and entered that of interpretation of custom.289 Let us proceed now with an examination of the interpretative process per se. The first part of the Tribunal’s pronouncement, where it analyses the meaning of the word ‘including’ poses no problem; it is a textual interpretation. But what are we to make of this reference to the ICJ Statute? Although the Tribunal does not explain the reasoning behind this, it must fall within the customary rules on interpretation. The only element that seems to fit the bill is ‘context’ and, more particularly, Article 31(3)(c), or to be exact, its customary equivalent.290 The Tribunal is seeking to determine whether judicial decisions, can be supplementary means within Article 32. If we are to apply the essence of Article 31(3)(c) and refer to “all relevant rules of international applicable in the relations between the parties”, where better to start than with the UN Charter, which epitomises the notion of a treaty with global participation. The ICJ Statute is an integral part of the UN Charter291 and “[a]ll Members of the United Nations are ipso facto parties to the Statute of the International Court of Justice”.292 This would resolve, the issue of ‘applicable between the parties’, even if we were to apply the strictest criterion293 set by the Panel in the Biotech case,294 since all States, parties to the VCLT are parties to the UN Charter, and most importantly if the VCLT as a treaty were to apply. However, in the present case, as was mentioned supra, we are dealing with customary law, which might complicate things depending on the criterion applicable. Nevertheless, even in this case, the universal participation of the UN Charter is sufficient to satisfy even the Biotech case criterion. The only issue, then, which remains is if the ICJ Statute is ‘relevant’ and the Tribunal seems to address the issue when it says: Article 38 [paragraph 1.d.] of the Statute of the International Court of Justice provides that judicial decisions are applicable for the interpretation of public international law as ‘subsidiary means’. Therefore, they must be understood to be also supplementary means of interpretation in the sense of Article 32 VCLT.295 289
very little has been written on this highly interesting topic, and the only references to it are made in passim; see Certain Questions of Mutual Assistance in Criminal Matters (Djibouti v. France), Judgment of 4 June 2008, 2008 ICJ Rep. para. 40; see also Maarten Bos, A Methodology of International Law (Oxford: North-Holland, 1984); Alexander Orakhelashvili, The Interpretation of Acts and Rules in Public International Law (Oxford: OUP, 2008), Ch. 15. 290 because as we analysed supra we are now in the realm of customary law. 291 Article 92, UN Charter. 292 Article 93, UN Charter. 293 which, however, does not seem to reflect either the true meaning of Article 31 as text or the current status of customary law. In more detail see Merkouris, supra note 87. 294 which stated that all parties to the treaty interpreted must be parties to the treaty used by means of Article 31(3)(c); in more detail see supra Section III.B.3. Similarly, the intertemporality issues of 31(3)(c) do not come into play here as the UN Charter precedes the VCLT. 295 CCFT v. US, supra note 253, para. 187.
canons of treaty interpretation
229
Article 38 of the ICJ Statute is considered by many to be an informal outline of the sources of international law, with 38(1)(a)(b) and (c) being the ‘formal’ sources and the rest the ‘material’ or ‘quasi-formal’ sources.296 Judicial decisions, mentioned at Article 38(1)(d) are supposed not to create but to evidence, or indicate the status of international law.297 It is this relevance that the Tribunal was hinting at. The idea may be correct, but the way, however, the Tribunal puts this in words is open to criticism. First, Article 38 does not deal with “interpretation of public international law”, but what the ICJ is to apply in resolving disputes. Secondly, it characterises judicial decisions as “subsidiary means”. Even if we accept this, the Tribunal’s conclusion is a non sequitur. It equates ‘subsidiary’ with ‘supplementary’. Apart from the fact that these are two different terms, they also have also a subtle difference in meaning, both as terms per se as well as within the context in which they are used. ‘Subsidiary’ means “less important than but related to or supplementary to [something]” whereas ‘supplementary’ means “completing or enhancing something”.298 However, even if these definitional nuances seem unconvincing, a contextual approach may be more revealing. It is exactly this point of being ‘less important’ or not that was hotly contested in the travaux préparatoires of the VCLT and was resolved in favour of the adoption of the term ‘supplementary’.299 This term was selected in order to emphasize and to more accurately represent the true function of Article 32, i.e. making complete the interpretation, either through confirmation or clarification, arrived at through Article 31.300 On the contrary, in the travaux préparatoires of Article 38 of the PCIJ Statute, on which Article 38 of the ICJ Statute was entirely based, the term ‘subsidiary’ was employed, exactly in order to stress the subordination of judicial decisions to the other sources of international law; to demonstrate that it was not on par with the other norm creating sources, but its function
296 the latter being a term coined by Sir Gerald Fitzmaurice; Sir Gerald Fitzmaurice, ‘Some Problems Regarding the Formal Sources of International Law’, in Symbolae Verzijl: présentées au Professeur J. H. Verzijl à l’ Occasion de son LXX-ième Anniversaire, Jan H. Verzijl and Frederik M. van Asbeck (eds.) (The Hague: Nijhoff, 1958), 153–76, at 173. 297 Georg Schwarzenberger, International Law Vol. I: International Law as Applied by International Courts and Tribunals (London: Stevens, 1957), 27 et seq. 298 These definitions are taken from the Compact Oxford English Dictionary, <www .askoxford.com/dictionaries/compact_oed/?view=uk> (accessed 1 July 2009). The definition of ‘subsidiary’ may use the term ‘supplement’ i.e. “less important than but related to or supplementary to [something]”, however, the emphasis is on ‘less important’, whereas this element is missing from the definition of ‘supplementary’ itself. 299 for a general overview of the debate see Engelen, supra note 283, Ch. 7, especially 321–6. 300 ILC, Report of the ILC on the Work of its 18th Session: Draft Articles on the Law of Treaties with Commentaries, (A/6309/Rev.1), reproduced in YILC (1966), Vol. II: 218, at 220, para. 10; at 223, para. 19.
[para.10] The fact that article [32] admits recourse to the supplementary means for the purpose of ‘confirming’ the meaning resulting from the application of article [31] establishes a general link between the two articles and maintains the unity of the process of interpretation.
230
malgosia fitzmaurice & panos merkouris
was more that of providing evidence as to the current status of law.301 Consequently, the terms ‘subsidiary’ and ‘supplementary’, functioning in two completely different contexts, do not have the identical meaning that the Tribunal claimed. Despite these shortcomings in the exposition of its ideas, the Tribunal is not necessarily in error when it comes to its final conclusion i.e. that judicial decisions can fall under Article 32. If we read the relevant paragraph under the light of the previous analysis, i.e. as interpretation through Article 31(3)(c) then this theoretical construct seems to stand scrutiny. Of course, taking into account judicial decisions, brings into focus the issues of the binding effect of previous judicial decisions in the realm of international law. International courts are usually quick to assure that precedent is not binding in international law. It comes then as no surprise that the Tribunal, after having acknowledged that judicial decisions can be used as supplementary means of interpretation, makes absolutely certain that it puts to rest any fears of binding precedents, by stating that “it is not obviously clear how far arbitral decisions are of relevance to the Tribunal’s task” and that, in any event, it is “plain that the decisions of other tribunals are not binding on this Tribunal” and tries to clarify the way in which judicial decisions will be used in the sense that the judges will “[consider] other arbitral decisions and the arguments of the Parties based upon them, to the extent ….that they throw useful light on the issues that arise”.302 In the end, after considering a variety of judicial decisions the Tribunal concluded in a characteristically Article 32 – influenced wording that: 223. After a review of the relevant decisions in other cases as supplementary means of interpretation (Article 32 VCLT) it can thus be concluded that some of these decisions provide support to the interpretation the present Tribunal has chosen in earlier sections above of this Award, and that none of these decisions has been found to contradict this Tribunal’s interpretation.303
As was mentioned earlier, international courts, when using other judicial decisions, tend to shy away from incorporating them within a defined theoretical framework, as CCFT v. US did. In the Corn Products case, for instance, the NAFTA Tribunal stated that: The parties have cited a significant number of arbitration awards in other Chapter XI cases and in proceedings brought by investors under other international agreements. There is, of course, no doctrine of binding precedent in international law, nor any reason of principle why the tribunal which has the first word on a subject should necessarily be treated as having spoken definitively thereon.
301 Advisory Committee of Jurists, Procès Verbaux of the Meetings of the Advisory Committee of Jurists: 16 June – 24 July 1920 with Annexes (The Hague: van Langenhuysen, 1920). 302 CCFT v. US, supra note 253, para. 51. 303 Id., para. 223.
canons of treaty interpretation
231
Nevertheless, the views of other tribunals on issues similar to those arising in the present case and on questions of principle which impact upon the issues in these proceedings are necessarily of value to the Tribunal and have accordingly been taken into account.304
This is the common approach, as mentioned above; ensuring that no doubts remain as to the non-binding effect of precedent and utilising judicial decisions without clearly expounding on the theoretical framework within which they will be used.305 A further exception to this rule comes from within the WTO system.306 The EC – Chicken Classification case,307 boasts being one of the very few cases, alongside CCFT v. US to clearly state that judicial decisions can fall under Article 32. However, there is one big difference between them. Whereas CCFT v. US found that judicial decisions are neither travaux préparatoires nor ‘circumstances of conclusion’ of a treaty, thus allowing for a wider discretion as to the decisions that could be made reference to308 the WTO Panel in EC – Chicken Classification considered the judicial decisions in question as falling under ‘circumstances of conclusion’ of a treaty of Article 32, which, in some cases as we will see infra, might create certain temporal restrictions as to the judicial decisions that can be used. In more detail, the issue was whether in the interpretation of Heading 02.10 of the EC Schedule, the Panel could take into consideration the ECJ judgments in Dinter309 and Gausepohl.310 The Panel felt, that this question, had to be split in two parts: i) whether it was, theoretically, possible for judgments to be considered under Article 32 and ii) whether there were temporal constraints as to what judgments could be considered as ‘circumstances of conclusion’ of a treaty. In more detail: 7.390 The first question for determination by the Panel is whether the ECJ judgements, Dinter and Gausepohl, qualify as ‘circumstances of conclusion’ of the EC
304 Corn Products International Inc. v. the United Mexican States, Partial Award on Responsibility of 15 January 2008, ICSID Case No. ARB(AF)/04/01, para. 77, (hereinafter Corn Products v. Mexico), <www.naftaclaims.com/Disputes/Mexico/CPI/CPI-Mexico-MeritsAward .pdf> (accessed on 1 May 2009). 305 noteworthy is the difference of approach between the Tribunals in CCFT v. US and Corn Products v. Mexico, especially if one bears in mind that these decisions are separated not by years but by less than a fortnight. 306 This case is analysed in the NAFTA section and not the WTO one, as the authors felt that a close juxtaposition of these two cases would be more beneficial for the analysis of this issue, rather than their separate presentation within their corresponding Sections. 307 European Communities – Customs Classification of Frozen Boneless Chicken Cuts (30 May 2005), Panel Report, WT/DS269/R & WT/DS286/R (hereinafter EC – Chicken Classification). 308 however, always under the proviso that they shed some light to the issues raised in the case. 309 Case C-175/82, Dinter v. Hauptzollamt Köln-Deutz (1983), ECR 969. 310 Case C-33/92, Fleisch GmbH v. Oberfinanzdirektion Hamburg (1993), ECR I-3047.
232
malgosia fitzmaurice & panos merkouris Schedule within the meaning of Article 32 of the Vienna Convention. The Panel considers that there are two elements associated with this question as it relates to the Dinter and Gausepohl judgements. The first is whether, as a theoretical matter, court judgements can be considered under Article 32. The second is whether the timing of issuance of the ECJ judgements at issue, and more particularly the Dinter judgement, necessarily disqualifies it from consideration under Article 32.311
and by citing the EC – Computer Equipment case, where the Appellate Body had found that the Members’ classification practice and legislation that was applicable at the time of the Uruguay Round should have been taken into consideration under Article 32,312 it held that the list of the Appellate Body was not exhaustive and that …the Appellate Body was merely making a pronouncement on the basis of the facts that were available to it in that case rather than seeking to provide an exhaustive list of items qualifying as ‘circumstances of conclusion’ in all cases. This would suggest that a valid distinction cannot be drawn between, on the one hand, EC legislation and, on the other hand, ECJ judgements for the purposes of Article 32 of the Vienna Convention. Accordingly, the Panel considers that court judgements, such as the Dinter and Gausepohl judgements, may be considered under Article 32 of the Vienna Convention… [FN 681] This conclusion would seem to be particularly valid in relation to the present case where the ECJ judgements in question interpret EC legislation. In our view, it would be an odd situation if such legislation could be considered under Article 32 of the Vienna Convention but not court judgements, which interpret that legislation.313
The next issue the Panel dealt with was that of temporal restrictions. Brazil and Thailand had submitted that the Dinter judgment should not be considered under Article 32 because it was issued in 1983 prior to the Uruguay Round and, therefore, not temporally proximate. The Panel held that “there is no temporal limitation on what may qualify as ‘circumstances of conclusion’ under Article 32 and that ‘relevance’ is the more appropriate criterion”.314 However, this did not mean that there were absolutely no restrictions as to how far temporally distanced a judgment could be with respect to the treaty being interpreted, but that Article 32 itself indicates that “[any] instrument in question must be temporally proximate to the conclusion of a treaty in order for it to be taken into account…under Article 32”.315 In the end, the Panel
311
EC – Chicken Classification, Panel Report, supra note 307, para. 7.390. European Communities – Custom Classification of Certain Computer Equipment (22 June 1998), Appellate Body Report, WT/DS62, 67, 68/AB/R, para. 94. 313 EC – Chicken Classification, Panel Report, supra note 307, para. 7.391 & FN 681 (emphasis added). 314 Id., para. 7.391; and more analytically, para. 7.344. 315 Id., para. 7.392. 312
canons of treaty interpretation
233
considered only the Gausepohl and not the Dinter case, not because the latter was temporally too far316 but because it was not ‘relevant’ to the issue at hand.317 In summary, the Panel considered ECJ judgments to fall under Article 32, but within the notion of ‘circumstance of conclusion’ of a treaty and not within the non-defined other supplementary means. Furthermore, it elaborated a set of criteria: i) relevancy and ii) temporal proximity of the judgments to the treaty being interpreted. Out of these two criteria, the primary one is that of relevancy. Temporal proximity offers a threshold, but quite a flexible one, as long as relevancy is satisfied. The Appellate Body, referred briefly to the theoretical consideration of whether judgments can fall under Article 32. In principle, it agreed with the Panel318 but felt the need to point out that “that judgments deal basically with a specific dispute and have, by their very nature, less relevance than legislative acts of general application”.319 Despite their differences, both the CCFT v. US and the EC – Chicken Classification share a common approach towards acknowledging, explicitly, the possibility of judicial decisions falling under Article 32 of the VCLT and are the forerunners in a, hopefully, growing trend of using judicial decisions in the interpretative process, not tacitly, but within a clearly-defined theoretical framework. E. Conclusions on the NAFTA The analysis of the Trucking Services and Agriculture Products cases, clearly indicates that the Arbitration Panels followed up the letter of the 1969 VCLT provisions on treaty interpretation, including firstly the main rule of treaty interpretation as enshrined in Article 31 (paragraphs 1 and 2) and secondly having recourse to Article 32, in particular to travaux preparatoires. The task of the Arbitral Panels, especially in the second case was particularly daunting, since the arbitral panel had to include in its interpretative task, in addition to relevant Articles of the NAFTA (including its object and purpose), the FTA and the Uruguay Round Agreements. Therefore, it may be said that the Panel explored all the elements that constituted the so-called context of treaty as set out in Article 31 (2) of the 1969 VCLT. In fact, it may be said that the interpretation presented in the framework of the NAFTA is the best possible example
316 although it had been issued in 1983, the Panel thought that it still remained applicable; Id., para. 7.393. 317 Id. 318 EC – Chicken Classification, Appellate Body Report, supra note 82, paras. 290 et seq. and 309. 319 Id., para. 309.
234
malgosia fitzmaurice & panos merkouris
of how the agreement should be interpreted. In the Pope and Talbot Inc. case the Arbitral Tribunal departed from the canons of interpretation as enshrined in 1969 VCLT. It did so in the name of the evolutionary approach to the interpretation. It may be observed that the same approach was adopted by the AB of the WTO, with the observance of the principles of interpretation. Notwithstanding these remarks, it must be said that the reasoning of the Tribunal was very interesting and innovative. Another special feature of the interpretation is the construction of Chapter Eleven of the NAFTA that allows the Free Trade Commission to impose on Tribunals its own interpretation that takes into account the general interests of the Parties to safeguard the proper interpretation of the NAFTA in its entirety. The CCFT v. US case, was remarkable for a variety of reasons, as it, either implicitly or explicitly, dealt with a variety of issues regarding the proper understanding and functioning of Articles 31 and 32 of the VCLT; ranging from the non-existence of a substantive hierarchy both between the elements of Article 31 and between Articles 31 and 32, the relationship between the ‘object and purpose’ of a treaty and that of specific provisions, to a better understanding of the workings and content of 31(3)(a) and (b). Its most important pronouncement, though, remains unquestionably that of the explicit inclusion of judicial decisions within Article 32 of the VCLT and the attempt320 at a theoretical construction that would account for this inclusion. All of the above factors make the analysis of the interpretation techniques adopted in the jurisprudence of Tribunals under Chapter Eleven of the NAFTA fascinating. VI. General Conclusions As evidenced from the above case studies, both the WTO and the NAFTA rely on customary rules of treaty interpretation as enshrined in the 1969 VCLT. Therefore they both apply general rules of international law. This same conclusion was reached by J. Charney, being mindful, however, that the GATT and WTO panels have not always applied the 1969 VCLT in a correct fashion (a point that was forcefully criticised by the WTO Appellate Body, see above).321 For a non-WTO rule to play a role in the interpretative process, the following conditions must be fulfilled: first, the WTO term under scrutiny must be broad and ambiguous in order to justify taking into account other rules; secondly, the other rules must be of assistance in the determination of the meaning of the WTO rule.
320 321
albeit not entirely devoid of errors. Charney, supra note 2.
canons of treaty interpretation
235
This was evidenced in the Shrimp Turtle case, (see above), in which the WTO Panel interpreted the term ‘exhaustible natural resource’ under GATT Article XX paragraph (g) with reference to, among others, the 1982 UN Law of the Sea Convention, CITES and the Convention on Biological Diversity. In general, it must be said that the recent case law of the WTO (especially that of the Appellate Body) indicates that the rules on treaty interpretation as enshrined in the 1969 VCLT are fully acknowledged and adhered to. As many authors have observed this has not always been the case and the earlier jurisprudence of the GATT indicated a certain negligent approach to the 1969 VCLT, and in particular to the canons of interpretation. For example, Kuyper characterised the attitude of the Panels in this respect as “culpable”.322 A related question within the context of this paper is as to whether the WTO system is a self-contained regime. However, in depth discussion of this issue is outside the confines of this article.323 As to the application of the canons of the interpretation in the jurisprudence of the WTO, the above-examined WTO case law demonstrates very clearly that at least the Appellate Body B applies the 1969 VCLT canons of interpretation (the contentious issue remaining the application of Article 31(3)(c). It may be said, as Kuyper advocated in 1994, that previously the WTO paid only lip service to these rules of interpretation, at present there is no doubt that it fully follows and acknowledges these canons. However, certain particularities of interpretation, special to the WTO can still be discerned. As McRae observes, there are some differences in the interpretation adopted by the ICJ and the WTO.324 The ICJ follows the classical approach in interpretation, i.e. relies in most cases on the common intention of parties as it is reflected in the text of the document, however, in the event certain instruments such as unilateral declarations made under the optional cause system (Article 36 (2) of the ICJ Statute), interpretation is based on the intention of the state concerned. The reason for this interpretative approach is the specific character of unilateral declarations, i.e. as they are treated as a ‘unilateral act of sovereignty.’ The institutional character of the WTO is taken into account when interpreting the Agreement. The WTO is a “single undertaking” composed of a series of interlocking agreements, as it was explained in the Brazil-Measures
322 Pieter J. Kuyper, ‘The Law of GATT as a Special Field of International Law, Ignorance, Further Refinement or Self-Contained System of International Law,’ NYIL 25(1994): 229. 323 See on this issue Kuyper, supra note 322, 227–57; Bruno Simma and Dirk Pulkowski, ‘Of Planets and the Universe: Self-Contained Regimes in International Law’, EJIL 17 (2006): 483–529, see in particular 519–523; David Palmeter and Petros C. Mavroidis, ‘The WTO Legal System: Sources of Law’, AJIL 92 (1998): 398. 324 Donald M. McRae, ‘The WTO in International Law: Tradition Continued or New Frontier?’, Journal of Economic Law 3 (2002): 27–41, at 36.
236
malgosia fitzmaurice & panos merkouris
Affecting Desiccated Coconut case (see above). The WTO is an integrated unit and all the agreements have to be interpreted together. The question that has to be taken into consideration in this context is the relationship between the WTO and regional agreements such as NAFTA and the relationship between the WTO and other, subsequent international agreements. This problem is dealt with partly, McRae explains, through the application of Article XXIV of the GATT. This however does not exclude the application of Article 30 of the 1969 VCLT. The general view was also expressed that a holistic approach to the treaty interpretation recognizing hermeneutic value of each individual dimension of general rule encapsulated in Article 31(1) of the Vienna Convention, will moreover allow necessary flexibility for the WTO adjudicating bodies to accommodate broader policy considerations underlying any interpretative exercise (such as for example, internal and external legitimacy). The increased relevance of the WTO legal regime within the system of global governance demands such flexibility, particularly in light of the growing interface …between different societal values, such as the promotion of trade and investment, the advancement of labour standards and human rights, the protection of the environment and cultural diversity.325
The issue of fragmentation of international law brought renewed interest in questions of interpretation of treaties, in particular as regards Article 31(3)(c). Its scope of application had been for a long time a subject of many debates as it relates to the WTO jurisprudence. As Pauwelyn observes the distinguishing factor for rules under Article 31(3)(c) “ought not to be that they are legally binding on all WTO members” but rather that they are an expression of “common understanding” between the WTO members.326 Further he submits that the ambiguity of Article 31(3)(c) does not preclude its broader interpretation, i.e. relevant rules of international law applicable in the relations between the parties, allow their interpretation as rules “legally binding on all parties”.327 Finally, the same author observes that the Appellate Body has already referred in its interpretation of WTO terms to rules that are not legally binding on all the WTO members (such as in the US-Shrimp case, where the Appellate Body noted that GATT Article XX must be read in the “light of contemporary
325 Federico Ortino, ‘Treaty Interpretation and the WTO Appellate Body Report in the US-Gambling: A Critique’, Journal of International Economic Law 9 (2006): 117–48, at 148; see also Michael Lennard, ‘Navigating by the Stars: Interpreting the WTO Agreements’, Journal of International Economic Law 5 (2002): 17–89, at 23. Ortino also observes that the analysis the AB interpretation of the VCLT rules appears to be ‘excessively mechanical’ in particular with regard to evaluating different contextual elements as well as in examining ‘object and purpose’, at 130. He as well makes very valid comments on the use of Art. 32 (supplementary use), and notes that the AB should avoid overuse the ambiguity as of a term as an excuse to have recourse to this Art., which according to him should not be used as a means of the last resort, at 28–9. 326 Pauwelyn, reply to Meltzer’s comments, supra note 87, at 925. 327 Pauwelyn, supra note 87, at 926.
canons of treaty interpretation
237
concerns of the community of nations about the protection of the and conservation of the environment” and also made the reference to other treaties that were not binding on the WTO members, including even the parties to the dispute, in order to conclude that “exhaustible natural resources, include both living and non-living resources”).328 However, Meltzer in his comments on Pauwelyn’s essay warns about the “formal legitimacy” of rules to which the Appellate Body can refer while interpreting WTO terms.329 Pauwelyn agrees that such a warning must be treated seriously and that: [a]rticle 31. 3 (c) should, in the first place, be used to refer to rules explicitly agreed to by all WTO members, especially well-established general principles of law, custom and other treaties or decisions. Opening the door to rules not legally binding on all WTO members, not even on the disputing parties, may be permissible, but it remains a risky step: When and how will the Appellate Body decide that something rises to the level of a ‘contemporary concern of the community of nations’, ‘a widely recognised principle’ or a ‘broad-based recognition of a particular need’? Any such explanation should not be reached lightly and must be explained.330
In the light of the above discussion, it appears that that issue of the application and interpretation of Article 31(3)(c) has the potential of becoming one the most important aspects of the interpretation of the future jurisprudence of the WTO.
328 329 330
Id., at 926. Meltzer, supra note 87, at 922. Pauwelyn, supra note 87, at 926–7.
INTERPRETATION OF (ALLEGEDLY) SELF-JUDGING CLAUSES IN BILATERAL INVESTMENT TREATIES Tarcisio Gazzini I. Introduction In five recent ICSID cases, Argentina has invoked necessity to justify certain emergencies measures – otherwise inconsistent with its international obligations stemming from the bilateral investment treaty (BIT) with the United States – adopted between 1999 and 2001 to tackle a grave economic crisis.1 One of the key issues raised in these cases has been the alleged self-judging character of Article XI of the treaty which reads: This Treaty shall not preclude the application by either Party of measures necessary for the maintenance of public order, the fulfillment of its obligations with respect to the maintenance or restoration of international peace or security, or the protection of its own essential security interests.2
The alleged self-judging character of Article XI deserves a close scrutiny as it has raised several interesting questions from the standpoint of interpretation of treaties in general and of investment treaties in particular. Additionally, in spite of a so far consistent case-law, the issue is likely to be discussed again in the numerous cases against Argentina that are currently pending before ICSID and UNCITRAL tribunals. This essay examines the questions related to the alleged self-judging character of Article XI, bearing in mind the nature of bilateral investment treaties and of related arbitration3. It first describes how ICSID tribunals have dealt 1 CMS Gas Transmission Company v. Argentina, Award of 12 May 2005, ICSID ARB/01/8; LG&E Energy Corp., LG&E Capital Corp., LG&E International Inc. v. Argentina, Decision on Liability of 3 October 2006, ICSID ARB/02/1; Enron Corporation and Ponderosa Assets, L.P. v. Argentina, Award of 22 May 2007, ICSID ARB/01/3; Sempra Energy International v. Argentina, Award of 28 September 2007, ICSID ARB/02/16; Continental Casualty Company v. Argentine Republic, Award of 5 September 2008, ICSID ARB/03/9. In literature, see in particular: Charles Leben, ‘L’Etat de Nécessité dans le Droit International de l’Investissement’, Gazette du Palais 349 (2005): 19; August Reinisch, ‘Necessity in International Arbitration. An Unnecessary Split of Opinions in Recent ICSID Cases? Comments on CMS v. Argentina and LG&E v. Argentina’, Journal of World Investment and.Trade 8 (2007): 191; Théodore Christakis, ‘Quel Remède à l’Éclatement de la Jurisprudence CIRDI sur les Investissement en Argentine? La Décision du Comité ad hoc dans l’Affaire CMS c. Argentine’, RGDIP 111 (2007): 879. 2 The treaty, which was signed on 14 November 1991 and entered into force on 20 October 1994, and is available at <www.unctadxi.org/templates/DocSearch____779.aspx.> (accessed on 1 May 2009). 3 See in particular, Zachary Douglas, ‘The Hybrid Foundations of Investment Treaty Arbitrations’ BYIL 74 (2003): 151.
240
tarcisio gazzini
with the issue and then discusses whether their findings are consistent with Article 31 Vienna Convention on the Law of Treaties (VCLT).4 II. Respondent’s Interpretation of Article XI of the BIT United States – Argentina In all five cases, Argentina argued that Article XI is a self-judging provision and therefore it is for the host State to determine in good faith whether essential security interests are at stake and what emergency measures are necessary. According to Argentina, this interpretation of Article XI is shared by both parties to the treaty and is firmly based on reciprocity. Tribunals cannot therefore depart from this interpretation. They must refrain from performing any substantial review of the host State’s conduct and limit themselves to determine whether the good faith principle has been respected. In order to support the argument that the United States shares this interpretation of Article XI, Argentina maintained that since the Nicaragua decision5 – or after the conclusion of the BIT between the United States and Argentina6 – the United States have consistently treated clauses on emergency measures as self-judging.7 The evidence offered by Argentina included in the first place the BITs concluded by the United States with Russia (1992)8 and Bahrain (1999)9 as well as
4 The International Court of Justice has on several occasions held that Article 31 VCLT reflects customary international law. See Application of the Convention on the Prevention and Punishment of the Crime of Genocide (Bosnia and Herzegovina v. Serbia and Montenegro), Judgment of 26 February 2007, 2007 ICJ Rep., para. 160 and all the other decisions and advisory opinions referred to. Investment tribunals have systematically shared this view. See for instance: Salini Costruttori S.p.A. and Italstrade S.p.A. v. Jordan, Decision on Jurisdiction of 9 November 2004, ICSID ARB/02/13, para. 75; Saluka Investments BV v. Czech Republic, UNCITRAL, Partial Award of 17 March 2006, para. 296; Plama Consortium Limited v. Bulgaria, (Energy Charter Treaty), Decision on Jurisdiction of 8 February 2005, ICSID ARB/03/24, p. 117; Methanex Corp. v. United States, UNCITRAL (NAFTA), Final Award of 3 August 2005, Part II, Chapter B, para. 15. 5 CMS v. Argentina, supra note 1, paras 350; Enron v. Argentina, supra note 1, para. 335; Sempra v. Argentina, supra note 1, para. 379; Continental v. Argentina, supra note 1, para. 186. 6 LG&E v. Argentina, supra note 1, para. 209. 7 For an interesting and comprehensive discussion on the United States position since 1986, see William Burke-White and Andreas von Staden, ‘Investment Protection in Extraordinary Times: The Interpretation and Application of Non-Precluded Measures Provisions in Bilateral Investment Treaties’, Virginia Journal of International Law 48 (2008): 307. 8 In the Protocol to the treaty, reproduced in ILM 31 (1992): 794, at 811, the parties confirmed their mutual understanding that whether a measure is undertaken by a Party to protect its essential security interests is self–judging. The treaty is not in force. 9 Available at <www.unctadxi.org/templates/DocSearch____779.aspx.> (accessed on 1 May 2009). According to Article 14 (1), the treaty
shall not preclude a Party from applying measures which it considers necessary for the fulfilment of its obligations with respect to the maintenance or restoration of international peace or security, or the protection of its own essential security interests.
interpretation of (allegedly) self-judging clauses
241
the current Model Treaty (2004).10 Unlike Article XI of the BIT between Argentina and the United States, all these documents expressly provide for the self-judging character of the provision governing emergency measures. Argentina also relied on several official documents of the United States Congress and in the last two disputes on a letter sent on 15 September 2006 by an official of the Department of State to a former official testifying in the context of a different arbitration. According to the letter, notwithstanding the decision of the ICJ in the Nicaragua case, the position of the U. S. Government is that the essential security language in our FCN treaties and Bilateral Investment Treaties is self-judging, i.e., only the party itself is competent to determine what is in its own essential security interests.11
It is not clear from the decisions to what extent Argentina also relied on provisions contained in BITs concluded since 1991 by the United States which are similar or identical to Article XI of the BIT between Argentina and the United States, but unlike the latter were presented to the Congress as selfjudging.12 III. Claimants’ Interpretation of Article XI of the BIT United States – Argentina The position of the claimants in the five cases can be in good substance summarised in the following points. First, the self-judging character of a provision cannot be presumed but must be expressly stated in the treaty, as it has been the case of Article XXI GATT or certain BITs concluded by the United States.13 To emphasise this point, claimants also relied on the decisions of the International Court of Justice (ICJ) in the Nicaragua and Oil Platforms cases.14
The Letter of submittal to the Congress on this point reads: This Treaty makes explicit the implicit understanding that measures to protect a Party’s essential security interests are self-judging in nature, although each Party would expect the provisions to be applied by the other in good faith. 10 Available at (accessed on 1 May 2009). According to Article 18 (2), a contracting party is allowed to take the measures it considers necessary to protect inter alia its own essential security interests. 11 As quoted by the Tribunal in Sempra v. Argentina, supra note 1, para. 382. See also Continental v. Argentina, supra note 1, fn. 278. 12 See for instance, the letters of submittal to the Congress related to the BITs concluded with Albania (1995) and Bolivia (1998), available at <www.unctadxi.org/templates/DocSearch____779 .aspx.> (accessed on 1 May 2009). 13 CMS v. Argentina, supra note 1, para. 339. In Enron v. Argentina, supra note 1, paras 327 and Sempra v. Argentina, supra note 1, para. 369, the claimants further stressed that self-judging provisions are extraordinary exceptions. 14 See infra notes 26 and 27.
242
tarcisio gazzini
Second, Article XI is concerned with internal or international security emergencies and cannot be invoked in the context of economic crises.15 Third, the attitude of the United States does not demonstrate that Article XI has been consistently considered as a self-judging provision. Quite the contrary, the United States government took the view that the treaty contains “an absolute right to international arbitration of investment disputes”.16 Such a categorical position is hardly compatible with the inclusion of any self-judging provision. Finally, admitting the self-judging character of Article XI would introduce a broad and sweeping exception to the obligations imposed by the treaty in defiance of the latter purpose and object.17 IV. ICSID Tribunals’ Interpretation of Article XI of the BIT United States – Argentina Although the outcomes of the five cases under discussion on the admissibility of the necessity pleas put forward by Argentina were not consistent,18 the five tribunals dismissed the Respondent’s argument that Article XI is a selfjudging provisions and held that the adoption of emergency measures is subject to substantial review and not only control over compliance with the good faith principle.19 15 CMS v. Argentina, supra note 1, para. 340; Enron v. Argentina, supra note 1, para. 330 and Sempra v. Argentina, supra note 1, para. 372; LG&E v. Argentina, supra note 1, para. 211; Continental v. Argentina, supra note 1, para. 170. 16 Letter of Submittal of the United States – Argentina BIT, 13 January 1993, as quoted in Enron v. Argentina, supra note 1, para. 329 and Sempra v. Argentina, supra note 1, para. 371. 17 Enron v. Argentina, supra note 1, para. 330 and Sempra v. Argentina, supra note 1, para. 372. 18 In CMS v. Argentina, supra note 1, paras. 366 ff.; Enron v. Argentina, supra note 1, paras. 324 ff. and Sempra v. Argentina, supra note 1, paras 366 ff., the necessity plea was rejected. In LG&E v. Argentina, supra note 1, paras. 226 ff., the Tribunal accepted that between 1 December 2001 and 26 April 2003 the grave economic and social situation existing in Argentina allowed the adoption of measures on grounds of necessity under Article XI. In Continental v. Argentina, supra note 1, paras. 182 ff., the Tribunal held that the adoption of all measures before it, apart from those concerning the Treasury Bills (LETEs), was justified on grounds of necessity. That economic emergency may fall within the scope of Article XI has been accepted by all five tribunals. In the words of the Tribunal in Continental v. Argentina, supra note 1, paras 174, for example,
actions properly necessary by the central government to preserve or to restore civil peace and the normal life of society (especially of a democratic society such that of Argentina), to prevent and repress illegal actions and disturbances that may infringe such civil peace and potentially threaten the legal order, even when due to significant economic and social difficulties, and therefore to cope with and aim at removing these difficulties, do fall within the application under Art. XI.
See also CMS v. Argentina, supra note 1, para. 359; LG&E v. Argentina, supra note 1, para. 238; Enron v. Argentina, supra note 1, para. 332; Sempra v. Argentina, supra note 1, para. 374. 19 CMS v. Argentina, supra note 1, paras. 332 ff.; LG&E v. Argentina, supra note 1, paras. 207 ff.; Enron v. Argentina, supra note 1, paras. 324 ff.; Sempra v. Argentina, supra note 1, paras. 366 ff.; Continental v. Argentina, supra note 1, paras. 182 ff. In LG&E v. Argentina, supra note 1, para. 213,
interpretation of (allegedly) self-judging clauses
243
The finding is based in the first place on textual considerations.20 It has been held that nothing would prevent contracting parties from leaving each of them to freely determine what is a measure necessary to protect its essential security interests. Indeed, this has occurred in respect to other BITs.21 Three tribunals mentioned also Article XXI GATT 1994. Two of them qualified this provision as self-judging to the effect that it allows a Member of the World Trade Organization to take any action it considers necessary for the protection of its essential security interests.22 A third tribunal, however, took the opposite view and observed that “the very fact that such article has not been excluded from dispute settlement is indicative of its non-self-judging nature”.23 Three Tribunals further held that, considering the exceptional or extraordinary character of the measures adopted on grounds of necessity, the selfjudging character of the relevant provisions contained in BITs cannot be presumed but must be expressed in the text of the treaty.24 Article XI cannot therefore be considered as self-judging due to the silence of the treaty and the absence of any contextual element imposing such an interpretation. Two tribunals considered the object and purpose of the treaty. They held that the object and purpose of the Treaty is, as a general proposition, to apply in situations of economic difficulty and hardship that require the protection of the international guaranteed rights of its beneficiaries. To this extent, any interpretation resulting in an escape route from the obligations defined cannot be easily reconciled with that object and purpose. Accordingly, a restrictive interpretation of any such alternative is mandatory.25 however, the Tribunal noted that substantial review and good faith review would not differ significantly. 20 In Continental v. Argentina, supra note 1, para. 187, the Tribunal held that [a]lthough a provision such as Art. XI, as earlier indicated, involves naturally a margin of appreciation by a party invoking it, caution must be exercised in allowing a party unilaterally to escape from its treaty obligations in the absence of clear textual or contextual indications (footnotes omitted). 21
See supra notes 8 and 9. CMS v. Argentina, supra note 1, para. 339 and Enron v. Argentina, supra note 1, para. 327. A good example of self-judging provision is Article 5 of the North Atlantic Treaty, according to which in case of armed attack against a member of the Organization, each other member shall take ‘such action as it deems necessary’. 23 Sempra v. Argentina, supra note 1, para. 384. 24 CMS v. Argentina, supra note 1; para. 370. In Enron v. Argentina, supra note 1, para. 336, and Sempra v. Argentina, supra note 1, para. 383, it has been held that the provision must be ‘very precise’ in order to establish its self-judging nature. In Sempra v. Argentina, supra note 1, para. 379. The Tribunal further added that 22
[t]ruly exceptional and extraordinary clauses, such as a self-judging provision, must be expressly drafted to reflect that intent, as otherwise there can well be a presumption that they do not have such meaning in view of their exceptional nature. 25 Enron v. Argentina, supra note 1, para. 331; Sempra v. Argentina, supra note 1, para. 373. Similarly, in Continental v. Argentina, supra note 1, para. 187, the Tribunal pointed out that the self-judging character of Article XI “would conflict in principle with the agreement of the parties to have disputes under the BIT settled compulsorily by arbitration”.
244
tarcisio gazzini
The finding on the non self-judging character of Article XI of the BIT between the United States and Argentina was further supported by reference to the ICJ judgments in the Nicaragua,26 Oil Platforms27 and Gabcikovo-Nagymaros28 cases. In the first two decisions, the ICJ held that provisions comparable to Article XI and equally silent on the self-judging character were subject to substantial review. The key passage of the first decision reads: Article XXI [1956 Treaty of Friendship, Commerce and Navigation between the United States and Nicaragua] defines the instances in which the Treaty itself provides for exceptions to the generality of its other provisions, but it by no means removes the interpretation and application of that article from the jurisdiction of the Court as contemplated in Article XXIV. That the Court has jurisdiction to determine whether measures taken by one of the Parties fall within such an exception, is also clear a contrario from the fact that the text of Article XXI of the Treaty does not employ the wording which was already to be found in Article XXI of the General Agreement on Tariffs and Trade. This provision of GATT, contemplating exceptions to the normal implementation of the General Agreement, stipulates that the Agreement is not to be construed to prevent any contracting party from taking any action which it ‘considers necessary for the protection of its essential security interests’, in such fields as nuclear fission, arms, etc. The 1956 Treaty, on the contrary, speaks simply of ‘necessary measures’ not of those considered by a party to be.29
The Gabcíkovo – Nagymaros case, in turn, was relied upon by three tribunals to strengthen the argument of the non self-judging character of Article XI.30 In this decision, the ICJ held that under customary international law the State adopting measures on grounds of necessity is not the sole judge of whether the conditions under which these measures can be adopted have been met.31 As a result, for all five tribunals the meaning attached to Article XI by the parties at the time of the conclusion of the BIT is clear: the provision is not self-judging.32 Three of them, nonetheless, examined the evidence produced by the Respondent on subsequent practice of the parties to the BIT. In this respect, the argument of the Respondent as it emerges from the awards does
26 Case Concerning Military and Paramilitary Activities in and against Nicaragua (Nicaragua v. United States of America), Judgment of 27 June 1986, 1986 ICJ Rep. 14. 27 Oil Platforms case (Iran v. United States of America), Merits, Judgment of 6 November 2003, 2003 ICJ Rep. 161. 28 Gabcikovo-Nagymaros Project (Hungary v. Slovakia), Judgment of 25 September 1997,1997 ICJ Rep. 7. 29 Nicaragua Case, supra note 26, para. 222. 30 CMS v. Argentina, supra note 1, para. 372; Enron v. Argentina, supra note 1, para. 336; Sempra v. Argentina, supra note 1, para. 383. 31 Gabcíkovo – Nagymaros Case, supra note 28, para. 51. 32 CMS v. Argentina, supra note 1, para. 373; LG&E v. Argentina, supra note 1, para. 213; Enron v. Argentina, supra note 1, para. 337; Sempra v. Argentina, supra note 1, para. 385; Continental v. Argentina, supra note 1, para. 187.
interpretation of (allegedly) self-judging clauses
245
not seem consistent. In LG&E v. Argentina, the Respondent apparently argued that after the conclusion of the BIT with Argentina the United States shifted its position with regard to clauses like Article XI and started treating them as self-judging.33 On the contrary, in Enron v. Argentina and Sempra v. Argentina, the Respondent apparently argued that subsequent practice of the United States reaffirmed the self-judging character of Article XI.34 At any rate, none of the tribunals was impressed by the evidence on subsequent practice submitted by the Respondent with a view to demonstrate the self-judging character of Article XI. In two awards, the dismissal of the Respondent’s argument was accompanied by an interesting statement on the possible effects of subsequent practice on the interpretation and modification of BITs. With regard to the alleged self-judging character of Article XI, the two tribunals held that [e]ven if this interpretation were shared today by both parties to the Treaty, it would still not result in a change of its terms. States are of course free to amend the Treaty by consenting to another text, but this would not affect rights acquired under the Treaty by investors or other beneficiaries.35
V. Interpretation of Article XI under VCLT Article 31 (1) Investment treaties are to be interpreted in accordance with the relevant articles of the VCLT.36 Article XI of the BIT between Argentina and the United States, therefore, is to be interpreted in accordance with the ordinary meaning of its terms in their context and in the light of the object and purpose of the treaty, bearing in mind that interpretation is an “holistic exercise that should not be mechanically subdivided into rigid components”.37
33
LG&E v. Argentina, supra note 1, para. 209. Enron v. Argentina, supra note 1, para. 326; Sempra v. Argentina, supra note 1, para. 368. 35 Enron v. Argentina, supra note 1, para. 337; Sempra v. Argentina, supra note 1, para. 385. The statement would have been clearer had it specified in the first sentence that no change would have resulted “for the purpose of this dispute”. The position of the Tribunal in LG&E v. Argentina, supra note 1, para. 213, on this point is rather confused. The Tribunal held that 34
[t]he provisions included in the international treaty are to be interpreted in conformity with the interpretation given and agreed upon by both parties at the time of its signature, unless both parties agreed to its modification. In that case, the date to be considered is November 1991. It is not until 1992, with the ratification of the Russia – US BIT, that the United States begins to consider that the application of the essential security measures are self judging; both instruments post-date the bilateral treaty between the United States and the Argentine Republic and, in both cases, this change was explicitly clarified. 36
See supra note 4. EC – Customs Classification of Frozen Boneless Chicken Cuts (12 September 2005), Appellate Body report, WT/DS269/AB/R, para. 176. According to Ian Sinclair, The Vienna Convention on the Law of Treaties (Manchester: MUP, 1984), 153, 37
[i]nterpretation is a process involving the deployment of analytical and other skills: it cannot be reduced to a few propositions capable of purely automatic application in all circumstances.
246
tarcisio gazzini
With regard to the ordinary meaning to be given to the terms of Article XI,38 the use of the expression “measures necessary” and not “measures the host State considers as necessary”, as in the case of similar provisions in other treaties, clearly militates against the self-judging argument. It is also worth noting that even Article XXI GATT 1994, which expressly provides for the adoption of any action a Member considers as necessary, still remain controversial in spite of its text and the ICJ decision in the Nicaragua case.39 At the time of the conclusion of the BIT between Argentina and the United States, contracting parties had three options in respect of measures adopted on grounds of necessity: (a) including a self-judging clause like Article 14 (1) BIT United States and Bahrain;40 (b) not including any clause at all; and (c) including a clause silent on its self-judging or non-self-judging character like Article XI. Under the first option, the meaning of Article XI would be sufficiently clear and tribunals should limit themselves to determine whether the right to adopt these measures had been exercised in good faith.41 Under the second option, the lack in the treaty of any provision on emergency measures would have not prevented the host State from adopting measures on grounds of necessity under customary international law. As held by the ICJ in Gabcíkovo-Nagymaros, however, an international or arbitral tribunal may substantially review them and determine whether they are consistent with the conditions and limits imposed under customary international law.42 The third option is the most complex as the meaning the parties intended to attach to Article XI is difficult to establish. On one hand, the choice not to go
38 In Territorial Dispute (Libyan Arab Jamahiriya v. Chad), Judgment of 3 February 1994, 1994 ICJ Rep. 6, para. 41, the ICJ confirmed that “[i]nterpretation must be based above all upon the text of the treaty”. See also Sovereignty over Pulau Litigan and Pulau Sipadan (Indonesia v. Malaysia), Judgment of 17 December 2002, 2002 ICJ Rep. 625, para. 37. In US – Import Prohibition of Certain Shrimp and Shrimp Products (12 October 1998), Appellate Body Report, WT/DS58/AB/R, para. 114, the WTO Appellate Body held that “[a] treaty interpreter must begin with, and focus upon, the text of the particular provision to be interpreted”. 39 See supra note 29. In literature, see Dapo Akande and Sope Williams, ‘International Adjudication on National Security Issues: What Role for the WTO?’, Virginia Journal of International Law 43 (2003): 365. 40 See supra note 9. The same result could have been obtained through a protocol like the one concluded between the United States and Russia, see supra note 8. 41 As observed by the ICJ in the Fisheries case (United Kingdom v. Norway), Judgment of 18 December 1951, 1951 ICJ Rep. 116, at 142,
[t]he principle of good faith requires that every right be exercised honestly and loyally. Any fictitious exercise of a right for the purpose of evading either a rule of law or a contractual obligation will not be tolerated. Such an exercise constitutes an abuse of the right, prohibited by law. 42 Supra notes 28 and 31. In Sempra v. Argentina, supra note 1, para. 378, the Tribunal held that “the Treaty itself did not deal with the legal elements necessary for the legitimate invocation of a state of necessity. The rule governing such questions will thus be found under customary law”.
interpretation of (allegedly) self-judging clauses
247
for the first option may indicate a contrario their intention not to introduce in the treaty a self-judging clause. This interpretative technique was used by the ICJ in the Nicaragua case.43 On the other hand, the decision to include Article XI may indicate their intention to depart from customary international law. From this perspective, Article XI would be a potential self-judging provision despite the lack of an express indication in this sense. Yet, the self-judging character of Article XI represents a significant departure from customary international law that can hardly be implied from the mere inclusion in the treaty of a provision on emergency measures. As pointed out in ICSID arbitration44 and in line with ICJ jurisprudence,45 the measures adopted on grounds of necessity being extraordinary, the self-judging character of the relevant clauses must be clearly expressed. The fact that the Nicaragua decision – which can be considered as the catalyst of the United States policy on the self-judging character of necessity provisions46 – was delivered five years before the conclusion if the BIT between the United States and Argentina further militates against the self-judging character of Article XI. Indeed, contracting parties were fully aware that without an express qualification, Article XI was unlikely to be treated as a selfjudging provision.47 The argument put forward by some tribunals that the object and purpose of the treaty requires a restrictive interpretation of Article XI,48 however, is unnecessary and unconvincing. Article XI and investment treaties in general require a balanced interpretation that takes into account, on the one hand, the sovereign rights of the host State and, on the other hand, the legally protected interests of the foreign investor. In this regard, in Mondev v. United States, the Tribunal unambiguously held that there is no principle either of extensive or restrictive interpretation of jurisdictional provisions in treaties. In the end the question is what the relevant provisions mean, interpreted in accordance with the applicable rules of interpretation of treaties.49
43
Supra note 29. Supra section IV. 45 Supra note 29. 46 See CMS v. Argentina, supra note 1, para. 368. In literature, see also Burke-White and von Staden, supra note 7, at 382. 47 In this regard, it is rather surprising that for years the United States concluded several BITs containing clauses similar to Article XI, although their self-judging character was often pointed out in the respective letters of submittal to the Congress. For two examples, see supra note 12. 48 Supra note 25. 49 Mondev International Ltd. v. United States of America, Award of 11 October 2002, ICSID ARB(AF)/99/2 (NAFTA), para. 43. See also Methanex v. United States, supra note 4, para. 105; El Paso Energy v. Argentina, Decision on Jurisdiction of 27 April 2006, Case ARB/03/15, para. 70; Azurix v. Argentina, Award of 14 July 2006, ICSID ARB/01/12, para. 307. In Saluka v. Czech Republic, supra note 4, para. 300, for instance, the Tribunal pointed out that 44
248
tarcisio gazzini
From such a perspective, Article XI aims at striking a balance between the protection of the essential interests of the host State and the interests of the foreign investor. It is therefore not a question of interpreting restrictively Article XI, but rather to ensure rigorous compliance with the conditions under which measures on grounds of necessity can be lawfully adopted. VI. Subsequent Agreement Between the Parties for the Purpose of Art. 31 (3) (a) VCLT In spite of the apparent willingness of both parties to consider Article XI as self-judging, no subsequent agreement on the interpretation of Article XI has ever been concluded between Argentina and the United States for the purpose of Article 31 (3) (a) VCLT.50 This may be explained by the fact that the United States was comfortable with the “strategic ambiguity”51 of Article XI. Be that as it may, it would have been wise for Argentina in the eve of the adoption of the measures on grounds of necessity to seek an agreement on the self-judging character of emergency measures clauses with the States with whom it had concluded BITs. The most obvious form would have been a protocol similar to the one concluded between the United States and Russia. The parties to a treaty enjoy the largest autonomy as far as the form of a subsequent agreement is concerned. Such an agreement, however, cannot be deduced from the conclusion by the United States and Argentina – not to mention by the United States alone – of BITs with third countries containing self-judging clause on measures adopted on grounds of necessity. Every BIT is independent from each other and indeed the great advantage of the existing remarkable network of BITs is precisely the freedom enjoyed by each couple of contracting parties.52 [t]he protection of foreign investments is not the sole aim of the Treaty, but rather a necessary element alongside the overall aim of encouraging foreign investment and extending and intensifying the parties’ economic relations. 50
As observed by the ILC in YILC 18 (1966-II): 221, an agreement as to the interpretation of a provision reached after the conclusion of a treaty represents an authentic interpretation by the parties that must be read into the treaty for purposes of its interpretation.
Anthony Aust, Modern Treaty Law and Practice (Cambridge: CUP, 2007), 239, further observes that “[g]iven that the parties can agree later to modify the treaty, they can also subsequently agree on authoritative interpretation of its terms, and this can amount, in effect, to an amendment”. 51 This is the position of Argentina as referred by the Tribunal in LG&E v. Argentina supra note 1, para. 213. It is worth noting that the protocol to the BIT between the United States and Russia supra note 8, was concluded upon the insistence of the latter, see Burke-White and von Staden, supra note 7, at 384. 52 As noted by Kevin C. Kennedy, ‘A WTO Agreement on Investments: A Solution in Search of a Problem’, University of Pennsylvania Journal of International Economic Law 24 (2003): 77, at 183,
interpretation of (allegedly) self-judging clauses
249
The consequences of a hypothetical subsequent agreement between the United States and Argentina establishing the self-judging nature of Article XI deserve to be briefly examined. It is clear that from the moment the subsequent agreement enters into force, the BIT must be interpreted and applied accordingly. This may be detrimental to the legal protection of the foreign investors’ interests. Yet, foreign investors are the beneficiary of BITs but are also unavoidably exposed to the risk of subsequent authoritative interpretations, possibly inconsistent with previous arbitral decisions, or even amendment of the treaties,53 not to mention their denunciation. Nonetheless, the agreement cannot affect already adjudicated disputes and awards inconsistent with the subsequent agreement cannot be challenged on this ground in the context of annulment proceedings. This would be clearly contrary to the res judicata principle.54 The matter may be more complex in respect of pending cases or future cases related to facts that occurred prior to the entry into force of the subsequent agreement. In all probability, claimants will argue that the agreed interpretation amounts to an amendment of the treaty and as such does not affect the above mentioned categories of disputes. Respondents will object that the subsequent agreement simply clarifies the meaning of the relevant provision and accordingly must be applied to any dispute not already settled. Assuming that the United States and Argentina conclude an agreement interpreting Article XI as a self-judging provision, tribunals may find themselves in an uncomfortable position. On the one hand, they will certainly take into account the string of consistent decisions denying the self-judging
[b]ilateral investment agreements offer the flexibility that is not possible under a multilateral framework. BITs can be tailored to fit country-specific needs in a way that is not possible under a multilateral framework. 53 On the difficulties to distinguish interpretations from amendments, see the debate provoked by NAFTA Free Trade Commission, Notes of Interpretation of Certain Chapter 11 Provisions, 31 July 2001, at <www.naftaclaims.com/files/NAFTA_Comm_1105_Transparency .pdf.> (accessed on 1 May 2009). For a sharp critique of the Commission interpretation see Methanex Corporation v. United States, Second Opinion of Professor R. Jennings, available at <www.naftaclaims.com/disputes_us_methanex.htm.> (accessed on 1 May 2009). As pointed out by Sinclair, supra note 37, p. 138, “[i]t is inevitably difficult, if not impossible, to fix the dividing line between interpretation properly so called and modification effected under the pretext of interpretation”. See also: Jean-Pierre Cot, ‘La Conduite Subséquente des Parties à une Traité’, RGDIP 70 (1966): 633; Mustafa K. Yasseen, ‘L’Interprétation des Traités d’après la Convention de Vienne sur le Droit des Traités’, RCADI 151 (1976-III): 1, Chapter IV; Francesco Capotorti, ‘Sul Valore della Prassi Applicativa dei Trattati Secondo la Convenzione di Vienna’, in Il Diritto Internazionale al Tempo della sua Codificazione: Studi in onore di Roberto Ago, Roberto Ago (ed.) (Milano: Giuffrè, 1987), vol. I, 197; Hugh Thirlway, ‘The Law and Procedure of the International Court of Justice: 1960–1989 (Part Three)’, BYIL 62 (1991): 1, at 48 ff. 54 On the fundamental character of the res judicata principle, see in particular, Industria Nacional de Alimentos, S.A. and Indalsa Perú, S.A. v. Peru, (Previously Empresas Lucchetti, S.A. and Lucchetti Perú, S.A. v. Peru), Decision on Annulment of 5 September 2007, ICSID ARB/03/4, para. 86.
250
tarcisio gazzini
character of Article XI BIT between Argentina and United States. Although not legally binding, precedents remain essential in investment-related arbitration.55 On the other hand, they cannot ignore the interpretation agreed upon by the parties. From a practical perspective, however, the problem may be more apparent than real. Whether the agreement amounts to an amendment of Article XI depends ultimately on the meaning attached to this provision at the time of the alleged breaches of the BIT. If the tribunal believes that Article XI is not a self-judging provision, then the subsequent agreement would alter the meaning of Article XI and as such cannot have any retroactive effect.56 If, on the contrary, the tribunal, departing from existing case-law holds that Article XI is self-judging, then the subsequent agreement would merely confirm the finding.57 In both cases, therefore, the subsequent agreement would have little or no practical impact on the tribunal decision on pending cases. What remains crucial is that the tribunal carefully determines the meaning of Article XI at the time of the commission of the alleged breaches. VII. Subsequent Practice of Parties for the Purpose of Art. 31 (3) (b) VCLT Under Article 31 (3) (b) VCLT, the interpreter should take into account any subsequent practice in the application of the treaty establishing the agreement of the parties regarding its interpretation.58 Subsequent practice of the parties
55 See Gabrielle Kaufmann-Kohler, ‘Arbitral Precedent: Dream, Necessity or Excuse?’, Arbitration International 23 (2007): 357. 56 Yasseen, supra note 53, at 52, observes that
[c]omme toute interprétation authentique, celle qui se dégage d’un accord établi par une pratique ultérieure des parties au traité est rétroactive étant déclarative. Il est néanmoins possible que la pratique ultérieure opère une modification, qui, elle, ne dispose que pour l’avenir. 57 It may be worth noting that during the arbitration proceedings related to a dispute between a Dutch investor and the Czech Republic, the parties to the BIT concluded an agreement on three questions of interpretation (Agreed Minutes dated 1 July 2002). The Respondent argued that the Agreed Minutes were binding “statement on the meaning and the application of the Treaty”, CME Czech Republic B.V. v. Czech Republic, UNCITRAL, Final Award of 14 March 2003, para. 89. While avoiding to take position on the alleged binding character of the Agreed Minutes or on their declarative or amending nature, the Tribunal held that they “support” (para. 437) and “confirm” its finding in respect of res judicata and of the supremacy of international law in case of conflict with national law. 58 See in particular, Case Concerning Kasikili/Sedudu Island (Botswana v. Namibia), Judgment of 13 December 1999, 1999 ICJ Rep. 1045, esp. paras. 49–50. As noted by the ILC in YILC 18 (1966-II): 221,
[t]he importance of such subsequent practice in the application of the treaty, as an element of interpretation, is obvious; for it constitutes objective evidence of the understanding of the parties as to the meaning of the treaty. Recourse to it as a means of interpretation is well-established in the jurisprudence of international tribunals.
interpretation of (allegedly) self-judging clauses
251
may also lead to the modification of the treaty itself.59 Such a possibility is accepted in international law although at the Vienna Conference it was decided not to include in the VCLT a provision on modification of treaties through subsequent practice. It is however important to note that according to the relevant draft article prepared by the ILC, a treaty may be modified by subsequent practice of the parties in the application of the treaty establishing their agreement to an alteration or extension of its provisions.60 An important difference must be pointed out in respect of subsequent practice in inter-States disputes and subsequent practice in investment disputes. In inter-States disputes, the parties to a bilateral treaty can share the view that following subsequent practice a given provision must be interpreted and applied in a certain way. As long as there is agreement between the parties, the tribunal does not need to examine subsequent practice or to determine from which moment on the relevant provision has to be interpreted and applied accordingly. The scenario is different in investment arbitration as the parties to the treaty do not coincide with the parties to the dispute. The parties to the BIT can still agree through subsequent practice on a given interpretation of a treaty provision or even its modification. For the purpose of interpreting and applying a provision in the settlement of a dispute, nonetheless, it is indispensable to determine whether subsequent practice had led to the new interpretation or to the modification of the said provision at the time of the conduct the foreign investor is complaining about. As pointed out in Enron v. Argentina and Sempra v. Argentina,61 the crucial time for considering subsequent practice is the time of the commission of the alleged wrongdoing if the rights acquired under the treaty by the foreign investor are to be adequately protected within a stable and predictable legal framework.62 After temporally limiting subsequent practice relevant for the purpose of interpreting and applying Article XI to the period spanning from the conclusion of the BIT and the adoption by Argentina of the emergency measures, it remains to examine the subsequent practice of both parties in the application of the treaty. Strictly speaking, when dealing with Article XI tribunals should consider the conduct of the organs of Argentina and the United States only to the extent it is directly related to the BIT between the two countries. Such practice may take different forms ranging from official declarations to municipal decisions. 59 See Air Transport Services Agreement Arbitration (United States v. France), Award of 22 December 1963, reprinted in ILR 38 (1969): 182, at 249. 60 Article 38, Draft Articles on the Law of Treaties, YILC 17 (1966-II): 236. 61 Supra note 35. According to Anne M. Slaughter, as referred to by the Tribunal in LG&E v. Argentina, supra note 1, para. 209, the United States considers emergency measures clauses as self-judging ‘presently and retroactively’. 62 See also Plama Consortium Limited v. Bulgaria, (Energy Charter Treaty), Judgment on Jurisdiction of 8 February 2005, ICSID ARB/03/24, para. 162.
252
tarcisio gazzini
BITs concluded by the United States with other States, on the contrary, have little or no relevance as subsequent practice for the purpose of interpreting and applying Article XI. They are completely unrelated to the application of the BIT between the United States and Argentina and they concern only one of the parties to this treaty. Reference to BITs concluded between the United States and other States can at best be an “aid to interpretation”, certainly not a canon or a rule of interpretation.63 Yet, given the res inter alios acta character of treaties, reliance on these treaties must be treated with the greatest prudence,64 especially considering that only a small number of BITs concluded by the United States expressly consider emergency measures clauses as self-judging. Even assuming for the sake of the argument that the United States had consistently concluded a large number of BITs containing self-judging clauses, their significance in respect of Article XI would still remain extremely limited as these treaties would represent the practice of just one party to the BIT between Argentina and the United States. As far as the documents adopted by the United States Congress are concerned, they hardly have any value with regard to Article XI of the BIT between the United States and Argentina since they are related to BITs concluded between the United States and other States. The letters of submittal to the Congress, in particular, might be taken into account exclusively for the purpose of determining the meaning attached by one of the contracting parties to the relevant provisions of the specific BIT the letters are related to. The Letter by the officer of the Department of State deserves a slightly different treatment. It was issued after the adoption by Argentina of the emergency measures and therefore is irrelevant in respect of disputes arisen in respect of these measures. In perspective, however, the document may be qualified as subsequent practice of the United States since it apparently concerns all BITs concluded by the United States. Respondent’s position on subsequent practice therefore is far from convincing for two main reasons. On the hand, the evidence is limited to the United States and neglects the fact that subsequent practice must be the practice of both States. There is no mention to the conduct of Argentina between the conclusion of the treaty and the adoption of the measures on grounds of necessity.65 63
See in general, Thirlway, supra note 53, at 66. In this regard, in Ceskoslovenska Obchodni Banka, a.s. v. The Slovak Republic, Award on Jurisdiction of 24 May 1999, ICSID ARB/97/4, para. 57, the Tribunal warned that 64
[t]he fact that some [BITs] may contain provisions for joint submission of disputes to arbitration does not compel the conclusion that provisions whose wording is at best ambiguous must be interpreted in like manner. Such a construction of the terms of a treaty would be incompatible with the applicable international rules for the interpretation of these types of agreements. (Vienna Convention, art. 31(1) ). 65 This is however without prejudice to potential value as subsequent practice of the unequivocal position taken on the interpretation of Article XI by Argentina in the recent arbitral proceedings.
interpretation of (allegedly) self-judging clauses
253
On the other hand, the subsequent practice of the United States is not related to the application of the BIT with Argentina and therefore hardly relevant for the purpose of Article 31 (3) (b) VCLT. VIII. Conclusions It may at first sight seem strange that Article XI of the BIT between Argentina and the United States has consistently been treated by five ICSID tribunals as a non self-judging provision in spite of the contrary position taken by Argentina in the proceedings and the similar general attitude of the United States. Yet, Article 31 (1) VCLT – applied in line with ICJ jurisprudence – allows to comfortably establish that Article XI is not self-judging and that the measures adopted on grounds of emergency may be subject to substantial review by arbitral tribunals. In this respect, investment treaties do not require any departure from the rules on interpretation established in Article 31 (1) VCLT, nor the adoption of a restrictive interpretation as suggested by two tribunals. Subsequent agreement between and subsequent practice of the parties present interesting issues of interpretation of investment treaties. The noncoincidence between the parties to the BIT and the parties to the dispute implies that the investor is exposed to the risk that the parties to the BIT may adopt an authoritative interpretation of, or even modify Article XI. This is unavoidable since the States remain “the transaction’s exclusive and absolute domini”.66 What can and must be avoided, however, is damaging the acquired rights of foreign investors and undermining the stable and predictable framework investment treaties are concluded to ensure.67 The parties to the BIT can formally agree that Article XI is a self-judging provision and even modify the legal protection of the foreign investor’s interests. However, with respect to pending disputes or disputes related to facts that occurred before the entry into force of the subsequent agreement, the meaning of Article XI must be established at the time of the commission of the allegedly violation of the BIT. A subsequent agreement may be valuable as confirmation of a given interpretation, but cannot modify with retroactive effect the legally protected interests of foreign investors.
66 The expression has been borrowed from Gaetano Arangio-Ruiz, The United Nations Declaration on Friendly Relations and the System of the Sources of International Law (Alpheen aan Rijn: Sijthoff & Noordhoff, 1979), 284–5, esp. note 183. 67 In the dissenting opinion in Industria Nacional de Alimentos, S.A. and Indalsa Perú, S.A. v. Peru, Decision on Annulment of 5 September 2007, ICSID ARB/03/4, para. 9, Berman called for a “particular duty of caution” when interpreting BITs as one of the parties to the disputes was a stranger to the treaty negotiation.
254
tarcisio gazzini
Subsequent practice must be of both States and must be related to the BIT under examination. Equally important, subsequent practice can be taken into account only up to the time of the conduct allegedly contrary to the BIT. Any retroactive effect of subsequent practice is to be excluded. The evidence submitted by Argentina on subsequent practice to prove the self-judging character of Article XI, accordingly, is far from satisfactory and the decision of ICSID tribunals on this point is again correct.
Part III
Interpretation, and Human Rights
INTENTIONALISM AND THE INTERPRETATION OF THE ECHR* George Letsas I. Introduction Looking at the ECtHR case law one finds relatively few references to the rules of interpretations embodied in arts 31–33 of the Vienna Convention on the Law of Treaties (VCLT). Close as its methods are to the general rule of purposive interpretation under Article 31 VCLT, the European Court has created its own labels for the means of interpretation that it uses such as ‘livinginstrument’, ‘practical and effective rights’, ‘autonomous concepts’ etc. What all these methods have in common is the rejection of the idea that the Convention rights must be interpreted in the light of what their meaning was taken to be back in the 1950s. The European Court has repeatedly stressed that the Convention is a ‘living instrument’ which must be interpreted “in the light of present-day conditions”. This is in line with the Vienna Convention on the Law of Treaties (VCLT) which prioritises the ‘object and purpose’ of treaties as a general rule of interpretation and assigns to preparatory works a supplementary role. According to Article 32 recourse to supplementary means of interpretation, such as preparatory works, may be had in order to confirm the meaning resulting from the application of article 31, or to determine the meaning when the interpretation according to article 31: (a) leaves the meaning ambiguous or obscure; or (b) leads to a result which is manifestly absurd or unreasonable.
Yet the connection between drafters’ original intentions and ‘here and now’ interpretation of human rights is not one that we can easily set aside. This is evident in how influential originalist theories of interpretation have been in the context of domestic law, particularly in American constitutional law and theory. The link between drafters’ intentions and judicial interpretation has always been a controversial issue in legal philosophy. This paper addresses the European Court’s approach to drafters’ intentions and the VCLT in the light of this general philosophical question. The paper is divided in two parts. Part I, which is more descriptive, provides an account of the European Court’s case law on the VCLT and drafters’ intentions after a short introduction to originalism. Part II turns to the philosophical foundations of originalism, arguing that its claims are unfounded if not supported by a normative theory about the * The ideas expressed in this Chapter are analyzed in more detail in: George Letsas, A Theory of Interpretation of the European Convention on Human Rights (Oxford: OUP, 2007).
258
george letsas
value of law. In the final section I explore candidate values for the ECHR concluding that the object and purpose of human rights treaties fully justifies the rejection of originalism. II. Originalism in the ECHR A. Originalism in Constitutional Law Originalist1 theories wish to tie interpretation back to the time when the law was enacted. We can distinguish between two types of originalism. The first one, textualism, argues that a legal provision must mean what it was taken to mean originally, i.e. at the time of enactment. Rather than ask ourselves for example whether the right not to be subjected to inhuman and degrading treatment under Article 3 ECHR applies to circumstances of extreme poverty, we should ask: “Did the public at large in 1950 understood extreme poverty to be an instance of inhuman and degrading treatment?” Textualists invite the interpreter to focus on the text enacted and read it in the light of its social and linguistic context at the time of adoption.2 Judges must ‘immerse themselves’3 in the society that adopted the text and understand the text as they understood it then. The second one, intentionalism, argues that a legal provision must apply to whatever cases the drafters had originally intended it to apply. Intentionalism takes legal interpretation to be a form of conversational interpretation. Rather than ask ourselves whether the right not to be subjected to inhuman and degrading treatment should be applied to circumstances of extreme poverty, we should ask: “Did the drafters intend this right to apply to socio-economic conditions?” Intentionalists usually propose the following scheme: If the drafters contemplated a particular situation then they either intended to prohibit it or they did not. If they did not intend to prohibit it then they either intended not to prohibit it or they left the matter open for the courts to decide.4 Intentionalists put more emphasis on drafting history and preparatory works as their task is to retrieve the original understanding of particular legislators.5 1 See Robert H. Bork, The Tempting of America (New York, NY: Free Press, 1989), 143–60; Antonin Scalia, ‘Originalism: The Lesser Evil’, University of Cincinnati Law Review 57 (1989): 849–65; Antonin Scalia, ‘Common Law Courts in a Civil Law System’, in A Matter of Interpretation, Antonin Scalia and Amy Gutmann (Princeton, NJ: Princeton University Press, 1997), 3–47; Thomas Gray, ‘Do We have an Unwritten Constitution?’ Stanford Law Review 27 (1975): 703; David Lyons, ‘Constitutional Interpretation and Original Meaning’, Social Philosophy and Policy 4 (1986): 75. 2 Paul Brest, ‘The Misconceived quest for the Original Understanding’ Boston University Law Review 60 (1980): 204, at 208. 3 Id. 4 Michael J. Perry, ‘Interpretivism, Freedom of Expression and Equal Protection’ Ohio State Law Journal 42 (1981): 261. 5 I should note here that textualism, at least as advanced by Scalia, is also a form of intentionalism: it directs us to the intentions of the public at the time of enactment. The difference
intentionalism and the interpretation of the echr
259
Originalism, in its various forms, has been a very influential doctrine in American constitutional law. It is held by many constitutional theorists as well as judges and it has always been a very controversial doctrine. The controversy over originalism was initiated in the 19th century as the debate over whether constitutional interpretation should be based on common-law doctrinal techniques or on methods of statutory interpretation. Later on, as constitutional adjudication gradually shifted from review of economic measures to civil liberties, originalism became the favourite theory of advocates of judicial selfrestraint. It was used to criticise the Supreme Court’s activism during the Warren period and to offer a theory of interpretation that would prevent judges from deciding evaluative issues. Later in 1987, the discussion on the merits of originalism was re-heated, when President Reagan nominated Robert Bork for the Supreme Court, a judge famous for his extreme originalist views on how to interpret constitutional rights. The issue attracted broad publicity, Bork’s views were finally rejected by the Senate, and his nomination was defeated. B. Golder v. UK: VCLT and the Case of Unenumerated Rights Golder 6 is undoubtedly one of the most important cases in the history of the ECHR. It is not just that it contains the first and perhaps most extensive discussion of the VCLT and the relevant rules of interpretation. It is also the first major case in its early years where the Court had to take a stance on what should be the general theory of interpreting the Convention and the relevance of original intentions. There was, in 1975, no right of direct access to the Court and very little case-law on the substantive rights of the Convention. The VCLT itself had not entered into force. Moreover none of the interpreting methods that the Court is now famous for had at the time been fully advanced. Tyrer7 that inaugurated the ‘living-instrument’ approach was decided in 1978. Engel8 which systematised the theory of autonomous concepts was decided in 1976 and Airey which provided an extensive application of the idea of ‘practical and effective rights’ was decided in 1978. Golder laid the foundations for the interpretive principles that have now become so important for the thousands of applications that the Court receives each year. The legal question in Golder was one that has fuelled the various debates between originalist and non-originalists in the context of American constitutional law, namely that of ‘unenumerated’ rights. These are rights that are not expressly mentioned in the text but it is proposed that they should
between the two lies in the group of people, whose intentions the interpreter aims to discover: textualism is directed at the community at large whereas intentionalism is directed at the drafters. I am grateful to Aileen Kavanagh for this point. 6 Golder v. United Kingdom, EHRR 1 (1975): 574. 7 Tyrer v. United Kingdom, EHRR 2 (1978): 1. 8 Engel and Others v. the Netherlands, EHRR 1 (1976): 647.
260
george letsas
nevertheless be ‘read into’ it. The right in Golder was that of access to court under art 6 ECHR. The applicant, a prisoner serving his sentence, had been denied permission to consult a solicitor with a view to institute libel proceedings against a prison officer. The United Kingdom, which was the respondent state, argued that the ECHR does not confer a right to access to court, given the absence of an explicit provision: if one gets to court she must be given a fair trial, but there is no obligation on the part of the member state to ensure that everyone gets to have his case heard. The argument was not so much to do with the intentions of the drafters as found in preparatory works. Rather it was premised on the idea that the text itself gives a clear indication of drafters intentions: had the drafters intended to create this right they would have done so explicitly by choosing a different wording. In support of this argument the respondent state added that the drafters had clearly thought about the right of access since it is expressly mentioned in Article 5 para 4 and Article 13 ECHR yet they omitted it from Article 6. In its judgment, the Court made reference to arts 31–33 of the VCLT and held that, though not yet in force, these articles enunciate general principles of international law that it has to take into account. In this respect, it first examined the language of article 6 ECHR and whether it confers the rights solely in proceedings that are already pending but not a right to have access to such proceedings. It concluded however that the language does not “necessarily refer only to proceedings already pending” but may well imply “the right to have the determination of disputes relating to civil rights and obligations made by a court or ‘tribunal’ ”.9 In other words it held that the wording itself was neutral between these two options. The Court then elaborated on the ‘object and purpose’ of the Convention, by turning to the preamble of the Convention, as provided for in Article 31 para. 2 of the VCLT. It cited the passage in the ECHR preamble that refers to the “common heritage of political traditions, ideals, freedom and the rule of law” of European countries and noted that “in civil matters one can scarcely conceive of the rule of law without there being a possibility of having access to courts”.10 This account of the rule of law was, on the Court’s view, in accordance with a general principle of law among Contracting States whereby “a civil claim must be capable of being submitted to a judge”. The Court further made the hypothetical point that if the Convention did not guarantee the right to access to court, states could without in breach of that text, do away with courts, or take away their jurisdiction to determine certain class of civil actions and entrust it to organs dependent on the Government.11 9 10 11
Golder v. United Kingdom, supra note 6, para. 32. Id., para 34. Id., para 35.
intentionalism and the interpretation of the echr
261
Such assumptions, the Court held, are “indissociable from a danger of arbitrary power” and would have serious consequences which are ‘repugnant’ to the principle of the rule of law. It concluded that the right of access constitutes an element which is ‘inherent’ in the right stated by Article 6(1), and warned that this is not an extensive interpretation forcing new obligations on the Contracting States: it is based on the very terms of the first sentence of Article 6 para 1 read in its context and having regard to the object and purpose of the Convention.12
It added that there is no need to resort to supplementary means of interpretation as envisaged at Article 32 of the VCLT. This was an important line of reasoning. The Court not only rejected the view that lack of an explicit provision in the text constitutes an obstacle for granting an unenumerated right. It also stressed that the question whether to grant an unenumerated right is not a question whether we should stick to the actual text or extend its meaning through interpretation. For the majority of judges in Golder did not think they added the right of access to court to art 6 ECHR; They insisted that by recognising the right to access to court, they followed an interpretation based on ‘the very terms’ of the first sentence of Article 6(1) and did not force any ‘new’ obligations on the Contracting States. On the contrary, they thought that fidelity to Article 6 ECHR demanded granting this right. “This is what the text says”, they could have said, “its very terms”. But the Court did not just announce that the right of access to court was included within ‘the very terms’ of Article 6. It gave clear reasons why it thought so. Recall the Court’s line of reasoning: 1. In interpretation, one should look at the object and purpose of the law. 2. The object and purpose of the ECHR is to promote the rule of law. 3. One can scarcely conceive of the rule of law in civil matters without right of access to court. 4. The right of access to court is ‘inherent’ in the right to fair trial under Article 6 ECHR. 5. The ECHR protects the right of access to court. Moreover, the Court followed this reasoning without feeling the need to resort to supplementary means of interpretation such as the preparatory works. It felt confident that ‘the object and purpose’ of the ECHR contains the ideal of the rule of law which leaves no ambiguity (necessary for resorting to supplementary means under Article 32 VCLT) as to whether it requires a right of access to court. 12 Id., para 36. The Court added that there was no need to resort to supplementary means of interpretation.
262
george letsas
The above reasoning did not come without objections on the part of some European Court judges. In a lengthy separate opinion, Sir Gerald Fitzmaurice mounted an originalist attack on the majority judgment. He objected that it is unacceptable to read into the text a right which the Convention does not trouble to name, but at the most implies, and which cannot even usefully be implied without at the same time proceeding to a rather careful definition of it, or of the conditions subject to which it operates, and which, by circumscribing it, define it.13
Fitzmaurice followed a typical intentionalist argument: if the drafters did not clearly intend to create a right of access to court then they could not have created one.14 In his view, lack of an express provision and detailed definition of a right of access means that states were not bound by such a right and that the European court should not impose a new obligation on member states by recognising it. Fitzmaurice’s arguments in his separate opinion in Golder merit a more careful discussion. His intentionalist views were based on an argument about the importance of certainty in international law. It is important, the British judge argued, that states have knowledge of the obligations they have undertaken by signing a treaty. “The parties”, he said, “cannot be expected to implement what would be an important international obligation when it is not defined sufficiently to enable them to know exactly what it involves”.15 At this point Fitzmaurice drew a distinction between national and international law. Judicial legislation, he argued, may be acceptable in domestic adjudication,16 but it is totally unacceptable in international adjudication which is based on agreement between states. Even if the lack of a right of access is a regrettable defect of the ECHR – he argued in elegant call for restraint – [i]t is for the States upon whose consent the Convention rests, and from which consent alone it derives its obligatory force, to close the gap or put the defect right by an amendment, – not for a judicial tribunal to substitute itself for the convention-makers, to do their work for them.
The only case where an international tribunal would be justified in substituting itself for the Convention-makers according to Fitzmaurice is if state
13 Golder v. United Kingdom, supra note 6, Separate Opinion of Judge Sir Gerald Fitzmaurice, para. 28. 14 See also Id., para. 40:
It is hardly possible to establish what really were the intentions of the contracting States under this head; but that of course is all the more reason for not subjecting them to obligations which do not result clearly from the Convention, or at least in a manner free from reasonable doubt. 15
Id., para. 30. For a judge of a common law country, Fitzmaurice was surprisingly sceptical of the legitimacy of judge-made law in England. 16
intentionalism and the interpretation of the echr
263
parties clearly intended to delegate this power to it. But he did not find any signs of such intention in the ECHR. C. After Golder: the ECHR as a Living-Instrument Fitzmaurice’s originalist call for judicial restraint did not persuade his fellow judges to change their interpretive methods. In a case that was decided 8 months after Golder, the British judge continued to defend originalism against an idea that was becoming more and more popular with the European Court judges, namely that the Convention should not be understood in terms of the intentions of the parties in 1950. This idea, he insisted, “lacks realism and reason”.17 Yet he could not have been more wrong. In the following 30 years or so, the European Court (and the former Commission) fully endorsed the idea that the Convention is a ‘living instrument’ that must be interpreted in the light of present-day conditions rather than what the drafters thought back in 1950.18 This method, also called evolutive or dynamic interpretation, proved neither unrealistic nor unreasonable. Over time the Court became quite explicit that lack of a clear intention on the part of the drafters is simply irrelevant when considering whether to recognise a right or not. A fine example is found in Matthews v. UK, where the issue was whether elections for the European Parliament fall within the right to vote under the ECHR: That the Convention is a living instrument which must be interpreted in the light of present-day conditions is firmly rooted in the Court’s case law … The mere fact that a body-was not envisaged by the drafters of the Convention cannot prevent that body from falling within the scope of the Convention.19
But the European Court went even further than that. It not only recognised rights that the drafters had not clearly intended to grant, but it also recognised rights that the drafters had clearly intended not to grant. This is most evident in Young, James and Webster.20 The case raised the issue whether socalled ‘closed shops’ in Britain, i.e. the legal requirement that all employees of a certain class are members of a specified trade union, were compatible with freedom of association under Article 11 ECHR. The applicants in this case were rail workers in the British Rail who had refused to enter a closed-shop 17 National Union of Belgian Police v. Belgium, Separate opinion of Judge Sir Gerald Fitzmaurice, EHRR 1 (1975): 578, para. 7. 18 See Tyrer v. United Kingdom, supra note 7, para. 31:
The Court must also recall that the Convention is a living instrument which, as the Commission rightly stressed, must be interpreted in the light of present-day conditions. In the case now before it the Court 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. 19 20
Matthews v. United Kingdom, EHRR 28 (1999): 361, para. 39 (emphasis added). Young, James and Webster v. United Kingdom, EHRR 4 (1981): 38.
264
george letsas
agreement on the grounds that they disagreed with the political aims of the specified trade unions and who were later dismissed fairly under English law. Their argument in Strasbourg was that Article 11 ECHR embodies a negative freedom of association, i.e. a right not to join a trade union if one does not want to, without any negative consequences and that therefore dismissal for failing to be a member of a specified trade union amounts to a violation. The UK government argued that Article 11 ECHR does not confer any right not to be compelled to join an association because this right “had been deliberately excluded from the Convention”, adding in proof the following passage in the preparatory works:21 On account of the difficulties raised by the ‘closed-shop system’ in certain countries, the Conference in this connection considered that it was undesirable to introduce into the Convention a rule under which ‘no one may be compelled to belong to an association’ which features in [Article 20 para. 2 of] the United Nations Universal Declaration (Report of 19 June 1950 of the Conference of Senior Officials, Collected Edition of the ‘Travaux Preparatoires’, vol. IV, p. 262).
It is clear in this passage that the drafters explicitly intended not to confer this right. Yet interestingly, the Court did not find this decisive. Here is what it said: Assuming for the sake of argument that, for the reasons given in the above-cited passage from the travaux preparatoires, a general rule such as that in Article 20 para. 2 of the Universal Declaration of Human Rights was deliberately omitted from, and so cannot be regarded as itself enshrined in, the Convention, it does not follow that the negative aspect of a person’s freedom of association falls completely outside the ambit of Article 11 and that each and every compulsion to join a particular trade union is compatible with the intention of that provision. To construe Article 11 as permitting every kind of compulsion in the field of trade union membership would strike at the very substance of the freedom it is designed to guarantee.22
It then went on to find a violation of Article 11 ECHR on the grounds that there was an illegitimate interference with the substance of the applicants’ right to freedom of association.23 Did the Court simply ignore the drafters’ intentions? Three dissenting judges complained as follows: Reference to the ‘substance’ of freedom of association is not relevant in the present context. Although the Court has often relied on the notion of the substance of the rights guaranteed by the Convention, it has done so only when the
21
Id., para. 51. Id, para. 52. 23 The Court was as always careful to find a violation in the particular case, not regarding closed shops in general and to note that compulsion to join a particular trade union may not always be contrary to the Convention. See Young, James and Webster v. United Kingdom, supra note 20, para. 55. 22
intentionalism and the interpretation of the echr
265
question was what regulation or limitation of a right was justified. It has held that even in cases where regulation or limitations were allowed explicitly or by necessary implication, they could not go so far as to affect the very substance of the right concerned. In the present case, however, the problem is whether the negative aspect of the freedom of association is part of the substance of the right guaranteed by Article 11. For the reasons stated above the States Parties to the Convention must be considered to have agreed not to include the negative aspect, and no canon of interpretation can be adduced in support of extending the scope of the Article to a matter which deliberately has been left out and reserved for regulation according to national law and traditions of each State Party to the Convention.24
Both the government and the dissenting judges put great emphasis on the fact that the matter had been deliberately left out by the drafters, as a decisive factor against the finding of a violation. The Court however insisted that it is the substance of the right that is important, downplaying any significance drafting history may have. However on occasion the European Court has been inconsistent in its rejection of originalism. In Bankovic25, the Court sitting as a Grand Chamber, had to decide whether NATO’s bombing campaign in Former Yugoslavia fell within the jurisdiction of contracting states under Article 1 ECHR. In an unfamiliar fashion and instead of sticking to his preferred method of ‘evolutive interpretation’ the Court rushed into invoking arts 31–32 of the VCLT, including the relevance of preparatory works. The Court never explained in Bankovic why, unlike Golder, the object and purpose of the ECHR does not settle the question of jurisdiction so that there is need to resort to supplementary means of interpretation. Agreeing with the respondent Governments’ submissions (which had been prepared by the United Kingdom) the European Court endorsed an argument which it had repeatedly and expressly rejected since Golder. Had the drafters of the Convention wished to ensure jurisdiction as extensive as that advocated by the applicants, they could have adopted a text the same as or similar to the contemporaneous Articles 1 of the four Geneva Conventions of 1949.26
This is not to say of course that this reasoning was decisive in the Court’s ruling the application inadmissible or that the case was wrongly decided for that reason. But the idea that had the drafters wished to grant a right they could have adopted a different text had certainly not counted against the applicant in Golder. Having said that, the Court’s inconsistent reasoning in Bankovic can
24 Id., Dissenting opinion of Judge Sorensen, joined by Judges Thor Vilhjalmsson and Lagergren, para. 4. 25 Bankovic v. Belgium and Others, ILM 41 (2001): 517. 26 Id., para 75.
266
george letsas
be explained by the fact that it was a highly politicised case involving the thorny issue of member states’ responsibility under human rights law when participating in military operations abroad. III. The Failures of Originalism Is the European Court’s rejection of originalism justified? It sounds counterintuitive to suggest that neither the text nor drafters’ intentions are relevant in solving interpreting difficulties and for good reason. One needs to be careful not to oversimplify the critique against originalism. It is the particular way in which either the text or drafters’ intentions are taken into account under originalist theories that raises concerns. This section aims to show that originalism cannot be the correct theory of interpretation by default but that it is in need of some normative foundation, just like non-originalist theories. Let us first examine more carefully the idea that judges must show fidelity to the text. Many people are inclined to believe that creative interpretation in law is wrong by default because it goes beyond ‘the text’. Faced with someone who insists his novel interpretation is covered by the text, they usually complain: “if this were the case then the text would mean whatever each one wants it to mean”. Yet, we do not think this is a self-evident argument in other forms of interpretation. We do not think for instance that we should interpret food recipes strictly, because otherwise they will mean whatever each one wants them to mean. On the contrary, people usually take it for granted that they can interpret recipes in a way that matches their tastes. Textualism is thus an unfortunate name for a theory of adjudication, if it purports to ground its normative claims on the mere fact that there is a text there to interpret. There are no general theories of textual interpretation. Different theories of interpretation may be applicable to texts of law, literature, food recipes, medical prescriptions, sports play-books, practical jokes, school regulations etc. But more importantly, textualism is an unfortunate label to the extent that it advocates strict interpretation in order to avoid recourse to extra-textual arguments. For surely it is itself an extra-textual theory: very rarely, if ever, the text of the law itself stipulates that interpretation must be strict rather than creative. It is nowhere written in the ECHR that the European Court should interpret the Convention in a strict way. Textualism is not grounded within the ‘four corners’ of the document, so to speak. And even if it did, this would not necessarily be the end of the argument. Yet we would be unfair to textualism if we were to neglect a fact that makes originalist theories possible. This fact is, I believe, indisputable: it is that the circumstances of human life change constantly and along with them, change our beliefs regarding how to understand important things about ourselves and the world we live in. What was yesterday’s pride often becomes today’s shame
intentionalism and the interpretation of the echr
267
and what was yesterday’s crazy claim often becomes today’s widely shared paradigm. Equality is a famous example. People used to believe (including the drafters of the US Constitution) that race discrimination does not violate equality. We, on the other hand, now take race discrimination as a paradigm instance of not treating people as equals. Moreover, we often take changes in belief to constitute substantive discoveries, rather than a mere change in our preferences. We feel we make progress in many areas of human knowledge: we do not simply say that equality prohibits race discrimination because that is how we like to think of equality now. Inevitably, changes in beliefs affect the meaning we feel inclined to attribute to concepts. To be sure, evolution in beliefs takes place slowly and in a piecemeal fashion. Some people might defend old meanings against new beliefs others might feel strongly that the new meaning has always been part of the concept. Disagreement, not just in law but everywhere, is always located within this background of evolution and constant re-interpretation. Originalism insists that the interpretation of the law be insulated from evolution and that substantive disagreement does not contaminate judicial practice. It insists that, in the light of substantive disagreement amongst citizens over fundamental concepts of political morality, judges should not be the ones to decide which interpretation offers the best understanding. In other words, originalism is a theory that defends, not the text or drafters intentions as such but an interpretive option in need of some further normative justification. It is also easy to show that the idea of respecting drafters’ intentions does not by itself call for any particular style of interpretation, be it strict, purposive, evolutive or other. Consider the idea that the interpretation of the law must not change over time but must always remain faithful to the intentions of the drafters. We can first imagine cases where the drafters themselves expressly authorised courts to revise or update the interpretation of the text. Suppose it is enacted (either in a constitution or in an international human rights treaty) that: Inhuman and degrading treatment is prohibited. A treatment is inhuman or degrading only if it is supported by the best moral arguments available to the court at each given time and regardless of drafters’ or the community’s views on the matter.
Clearly, a judge applying this provision would have to attribute different meaning depending on the views of the community at each given time. Hence what was not prohibited as inhuman and degrading in the 18th century may be prohibited now and vice versa. It would be the judge’s duty, out of respect for the drafters to adjust the meaning of the provision to contemporary circumstances. Were she to attribute a fixed meaning she would be disregarding their will. The example purports to make the point that drafters’ intentions come in different kinds and at various levels of abstraction. Take the ECHR. Drafters
268
george letsas
in 1950 had an abstract intention to promote and safeguards human rights in Europe but they also had a more concrete intention about which situations in their view human rights cover.27 To be sure, the ECHR would never have existed if some officials in the Contracting States had not intended to draft, sign and make it legally binding. Which is, however, more crucial, their intention to protect a list of fundamental freedoms of their citizens whatever these may be (intentions of principle), or their intention to protect what they, fifty years ago, believed these freedoms to be (intentions of detail)? Presumably, drafters had the former intention as much as they had the latter: they had a concrete idea of what human rights there are but it was their more abstract belief in the moral objectivity and universality of these rights that led them to draft the ECHR following the Second World War atrocities. Objectivity in rum, is a famously transparent concept: like most people, drafters may have been wrong about morality’s demands. Which intention did drafters take as more dominant? Their concrete intention to be bound by what, they thought, these moral rights were or their more abstract intention of being bound by whatever standards the human rights in the Convention really amount to? One option is to try to discover which of the two (and perhaps at which level) were thought to be more important by the drafters themselves. This is to introduce a third kind of intentions, what is called the drafters’ metaintentions.28 It is possible that the drafters felt more strongly about their abstract intention to protect the fundamental moral rights that people are indeed entitled to rather than their concrete intention to protect those rights that they, fifty years ago, believed individuals are morally entitled to.29 Recall that in his dissent in Golder, Fitzmaurice welcomed the possibility that the convention makers could expressly authorise the Court to expand, through interpretation, the scope of the Convention rights. He just required that their meta-intentions be clearly expressed.30
27 The distinction between abstract and concrete intentions is taken from Dworkin. See Ronald Dworkin, ‘The Forum of Principle’ in A Matter of Principle, Ronald Dworkin (Cambridge, MS: Harvard University Press, 1985), Ch. 2; See also Ronald Dworkin, Life’s Dominion (New York, NY: Vintage, 1993), Ch. 5 and Ronald Dworkin, ‘Comment’, in A Matter of Interpretation, Antonin Scalia and Amy Gutmann (Princeton, NJ: Princeton University Press, 1997), 115. 28 Also called, ‘interpretive’ intentions, see Ronald Dworkin, A Matter of Principle (Cambridge, MS: Harvard University Press, 1985), 52. 29 Dworkin argues that in the case of the American constitution we might invoke various facts in support of this claim: that the rights are framed in abstract rather concrete wording and that the drafters could not have thought their views were the final ones in moral matters. 30 It is worth noting that the distinction between abstract and specific intentions is not foreign to international law. Consider the long-standing controversy about the remedy that should follow the determination of an invalid reservation. Various options are available. First, the state may not be bound by the provision as to which the reservation was made, but continue to be bound by the rest of the treaty. Second, the state may seize to be a party to the treaty altogether
intentionalism and the interpretation of the echr
269
Yet introducing all these kinds of intentions involves a clear issue of circularity. For we begin by distinguishing between the various intentions that states (or drafters) have: concrete, abstract and meta-intentions. We then take their meta-intentions as being the most important ones, which in turn direct us to either their concrete or their abstract intentions. Here is the circularity: we ask what is the way in which drafters’ intentions become relevant, and we answer this question by saying that they become relevant in the way in which drafters intended. Even if states (or drafters) had meta-intentions (which they may not always do), we can only defer to them on the basis of some other reason, not on that of intentions. Suppose that a state’s meta-intention directs us to its concrete rather than abstract intention. Why, one may ask, is its metaintention more important than its abstract intention?31 At this point intentionalism runs out of justificatory resources. The question is not whether states’ (or drafters’) intentions are relevant but which of their intentions are relevant in interpretation. In conclusion, any theory of interpretation for the ECHR (or any international treaty) must at some stage stand outside drafters’ intentions and provide a normative justification based on values of political morality. Dworkin notes this very well in the context of constitutional interpretation: Some part of any constitutional theory must be independent of the intentions or beliefs or indeed acts of the people the theory designates as Framers. Some part must stand on its own in political or moral theory; otherwise the theory would be wholly circular.32 This idea captures the spirit of arts 31–32 VCLT. We cannot know whether (and the extent to which) drafters’ intentions are relevant unless we settle first on the object and purpose of the treaty. How to read drafters’ original intentions is the result of, rather than an alternative to, a theory about the object and purpose of the treaty. Hence the European Court’s rejection of originalism and be no longer bound by it. Third, the reservation may be severed and the state may remain bound the entire treaty, including the provision as to which the reservation was made. Ryan Goodman has proposed to solve this controversy by looking at relevant evidence of what was the ratifying state’s intent: would the state have ratified the convention if they knew that their reservation could be severed? Was in other words the reservation an essential condition for ratifying the treaty? See Ryan Goodman, ‘Human Rights Treaties, Invalid Reservations and State Consent’, AJIL 96 (2002): 531. This is precisely the application of the idea that, since states had both concrete (not to be bound by a specific provision) and abstract intentions (to enter the treaty), we should be looking at their meta-intentions, at what they thought was their dominant intention. Not surprisingly, Goodman’s argument is that this method has the advantage of maintaining respect for state consent as opposed to taking severance of reservations as a compromise to state consent. 31 See Dworkin’s point in A Matter of Principle; Dworkin, supra note 28, at 53, talking about drafters’ meta-intention: Why is it not our view against theirs on a complex issue of political theory, so that if our reasons are good we should not abandon these reasons just because people in another age would have disagreed? 32
Dworkin, supra note 28, at 54.
270
george letsas
is fully justified by the structure of arts 31–32 VCLT, assuming that this is what the object and purpose of the ECHR calls for. Does it however? Could there be a value (an object and purpose) that we may assign to the ECHR which would call for strict interpretation or some other form of judicial restraint advocated by originalists? I turn to this question in the next and final section. IV. The Object and Purpose of the ECHR It is not hard to see that part of what motivates originalist theories is fear of judicial creativity coupled by ideas of democratic legitimacy. In the face of controversy over the application of a constitutional provision it is better to defer to the views of those who had the right to legislate in the first place than leave the matter to be decided by a handful of unelected judges. The originalist could say something analogous about international law and the principles of state consent and equality of sovereign states: in the face of controversy over the application of a treaty, it is better to defer to state consent which is a valid source of state obligation than allow individual judges to impose new obligations upon sovereign states. Indeed this was precisely what Sir Gerald Fitzmaurice advocated in Golder. Yet there are a number of difficulties with these arguments. Often the drafters of a constitution were wholly unrepresentative of the people in the first place (as in the case of the fathers of the American constitution) or have become unrepresentative of the people as it stands now, say because of a change in its territorial boundaries through war or unification with a neighbouring state. Why should their views be taken into account given that they lack or lost a democratic basis? Similarly in the case of international treaties, the original drafting states may now have become a minority of member states. Why should the drafting history be taken into account and not, say, the understanding of the new member states -now forming a majority – either at the time of joining or even their current one? But perhaps the most serious difficulty with the above views is the suggestion that ideas of democratic legitimacy or state sovereignty could justify any legal provision. To put it rather extremely: a legal provision, be it in a constitution or an international treaty, allowing sadistic infliction of pain on little babies of unpopular minority groups would not have any legally binding force even if it were passed by a democratically chosen constitutional convention or even if all contracting states explicitly agreed to it. Dramatic as it may be, the example shows that ideas of democratic legitimacy at domestic level or state consent at international level cannot be the only values when considering how and which legal obligations are created. This is particularly so in the case of human rights. We do not think that the
intentionalism and the interpretation of the echr
271
question whether torture or genocide is permissible is governed by the principle of state consent. Why should we think that state consent is important when interpreting an international human rights treaty? So even if drafters’ concrete intentions or meta-intentions are closer to the notion of state consent, the originalist is wrong to assume that this notion underpins human rights treaties. The nature of fundamental human rights is such that they provide the conditions under which state consent should be respected and states (or parliaments) can legitimately pursue their goals. Hence the originalist argument that evolutive interpretation of human rights might offend state consent is simply irrelevant. It is the nature of human rights treaties, the kind of value they serve that directs us to look at drafters’ more abstract intentions and ignore their concrete or meta-intentions. Is there any other value that the originalist can invoke in support of certainty or strict interpretation of human rights treaties? It is often argued in domestic law that publicity is a distinctive virtue that law should have.33 It is a good thing to have laws that are promulgated officially and are publicly available for everybody to read. This is because people form expectations in their lives and they must be able to know in advance what it is that they are bound by, in order to arrange their behaviour accordingly. It is generally thought wrong to hold people accountable for violating standards they could not possibly know they were bound by. On this view, the idea of protecting people’s expectations calls for both certainty and stability in the interpretation of the law. I have serious doubts as to whether law is an instrument for protecting people’s expectations or people’s welfare in general. But even if it did, the application of this idea to international law would still appear problematic as it is hard to see why surprising contracting states is morally wrong.34 What kind of harm does a state suffer when its expectations regarding its treaty obligations are defeated? It is difficult to think of states as agents who form expectations about their lives so that it becomes morally problematic to surprise them. In domestic law we assign weight to expectations because of the importance they have for the autonomy of the person affected. It is thought that defeating their expectations is problematic because of the value of personhood or agency:35 individuals live a for a relatively short time during which they set higher ends and plans of life to pursue at the expense of day-to-day satisfaction of lower
33 See Lon L. Fuller, The Morality of Law (New haven, CT: Yale University Press, 1964), 34–5 and 43–4. 34 Recall that the argument that states must know exactly what they get themselves into, so to speak, by joining the ECHR was made by Fitzmaurice in his separate opinion in the Golder case. 35 On the connection between this value and human rights see James Griffin, ‘First Steps in an Account of Human Rights’, European Journal of Philosophy 9 (2001): 306.
272
george letsas
desires. This type of autonomy cannot in any sense be attributed to states whose legal personality and continuity is not reducible to that of natural persons.36 True, dynamic interpretation of human rights might surprise contracting states in that they often have to introduce legislative measures in order to comply with their Convention obligations. These measures may be quite extensive involving, among other things, financial costs that will affect the community as a whole. That is why states very often make interpretive declarations to treaties to ensure that certain provisions will be interpreted in the way that state expects them to be interpreted. But again, it is difficult to argue that we must balance this inconvenience suffered by states against the human rights that an individual is entitled to under the ECHR. If the Court is convinced that the applicant is entitled to a remedy as a matter of human rights should the fact that his state will be inconvenienced count against him? The moral force of human rights is precisely that they should not enter into cost-benefit calculations. To have a right is to be entitled to be treated in a certain a way even if the community would be better off by violating it. The purpose of human rights treaties, unlike that of many other international treaties, is to protect the autonomy of individuals against the majoritarian will of their state, rather than give effect to that will. In sum, the values of the ECHR, its object and purpose, fully justify why the Convention must not be interpreted in terms of what the drafters thought in 1950, or to put it accurately, why it should not be interpreted in terms of the drafters’ concrete intentions back in 1950. Having said that, there remains a question as to the merits of evolutive interpretation. For if by evolutive interpretation the Court means interpreting the Convention according states’ current intentions (as opposed to the original ones) found in national legislation or majoritarian beliefs, then the same objections that apply to originalism would apply to it as well. Indeed the Court’s extensive use of consensus as an interpretive value is vulnerable to exactly this criticism.37 Still, the general idea that the ECHR is a living-instrument and the defeat of originalism is something that Europe cannot but be proud of.
36
On the relation between state and personal autonomy see Charles R. Beitz, Political Theory and International Relations (Princeton, NJ: Princeton University Press, 1999, with afterword), 71–83. 37 See George Letsas, ‘The Truth in Autonomous Concepts: How to Interpret the ECHR’, EJIL 2 (2004): 279–305.
CONFLICTING INTERPRETATIONS OF THE ICC STATUTE – ARE THE RULES OF INTERPRETATION OF THE VIENNA CONVENTION STILL RELEVANT?* José Doria1 I. Introduction On 17 July 1998 in Rome (Italy), the Statute of the International Criminal Court (ICC) was adopted to establish the first ever permanent, treaty based, international criminal judicial organ, whose purpose is to promote international peace and security by ensuring that the gravest international crimes do not go unpunished. Indeed the preamble of the ICC Statute inter alia provides that the following were among the main reasons dictating the establishment of Court: Mindful that during this century millions of children, women and men have been victims of unimaginable atrocities that deeply shock the conscience of humanity, Recognizing that such grave crimes threaten the peace, security and well-being of the world, Affirming that the most serious crimes of concern to the international community as a whole must not go unpunished and that their effective prosecution must be ensured by taking measures at the national level and by enhancing international cooperation, Determined to put an end to impunity for the perpetrators of these crimes and thus to contribute to the prevention of such crimes.2
The Statute sets out the Court’s jurisdiction, structure and functions. On July 1 2002 the ICC Statute entered into force, after 60 instruments of ratification were deposited with the UN Secretary-General. Presently the Statute has been ratified by more than 100 States of the world.3 The Court has started its work
* This Chapter was based on the presentation made by the author during the ‘First Annual Conference on International Law: Interpretation under the VCLT’ organized by Queen Mary and Eversheds in 2006, and is current as of that date. 1 The views expressed herein are solely those of the author and do not necessarily reflect those of the Office of the Prosecutor or those of the United Nations. 2 1998 Rome Statute of the International Criminal Court (A/Conf. 183/9): Preamble; adopted on 17 July 1998 [hereinafter Rome Statute]. 3 As of 1st November 2006,103 countries are States Parties to the Rome Statute of the International Criminal Court. On 1st November 2006 Chad deposited its Instrument of Ratification to the Rome Statute. The Statute will enter into force for Chad on 1st January 2007 bringing the total number to 104 States Parties. Data available at <www.icc-cpi.int/statesparties .html> (accessed on 15 December 2006).
274
josé doria
and presently has been dealing with four self—referral situations: Uganda,4 Congo,5 Central African Republic,6 and Cote d’Ivoire.7 In addition the United Nations Security Council (UNSC) has also referred the situation in Darfur (the Sudan a non State party) to the ICC.8 Despite major progress in the implementation of the Statute, it is symptomatic to note the absence of nine out of the ten most representative countries of the world in terms of population. Indeed only one country, Brazil has joined the ICC whereas countries like China, India, the US, Indonesia and Russia, are still out of the Court. It is also noticeable that three of the five permanent members (P5) of the UNSC (China, Russia, US) are out of the Court. All this adds to the ongoing concerns and discussions about the legitimacy of the ICC. The Statute remains one of the major landmark instruments of modern international law and many of its provisions can justly be considered as too far away from an ordinary exercise of progressive development of international law. It is not a surprise therefore that so many divergent opinions exist as well
4 In December 2003, Yoweri Museveni, the President of Uganda self-referred to the ICC Prosecutor the situation concerning the Lord’s Resistance Army. The ICC Prosecutor decided to start investigations into the case and in August and September 2005 the first warrants of arrest were issued against Joseph Koni, Vincent Otti, Raska Lukwyia, Okot Odhiambo, Dominic Ongwen, all senior leaders of the Lord’s Resistance Army. Available at <www.icc-cpi.int/cases/ UGD.html> (accessed on 15 October 2006). 5 In April 2004, the ICC Prosecutor has received a letter signed by the President of the Democratic Republic of Congo (DRC) referring to him the situation of crimes within the jurisdiction of the Court allegedly committed anywhere in the territory of the DRC since the entry into force of the Rome Statute. The ICC Prosecutor decided to start investigations and in February 2006 a warrant of arrest was issued against Thomas Lubanga Dyilo. Available at <www .icc-cpi.int/cases/RDC.html> (accessed on 15 October 2006). 6 In January 2005 the ICC Prosecutor has received a letter sent on behalf of the government of the Central African Republic. The letter refers the situation of crimes within the jurisdiction of the Court committed anywhere on the territory of the Central African Republic. The Prosecutor has not yet decided to start official investigations into this situation. Available at <www.icc-cpi.int/cases/RCA.html> (accessed on 15 October 2006). 7 See ICC – Office of the Prosecutor, Report on the Activities Performed during the First Three Years (The Hague, 12 September 2006), <www.iccnow.org/documents/3YearReport%20_ 06Sep14.pdf> (accessed on 1 May 2009), at 10; noting that
the five situations currently under analysis include the situation in the CAR, following the referral by the CAR Government, and the situation in Cote d’ Ivoire which through a declaration lodged with the Court, accepted the Court’s jurisdiction for crimes committed in its territory since 19 September 2002. 8 In March 2005, the UNSC adopted Resolution 1593 (2005), referring the situation in Darfur (the Sudan) to the ICC Prosecutor, and in June of the same year the Prosecutor decided to start investigations. The Prosecutor hopes that the first warrants of arrests would be issued earlier 2007. Available at www.icc-cpi.int/cases/Darfur.html (accessed on 1 May 2009). See also ICC Prosecutor, Fourth Report of the Prosecutor to the UNSC pursuant to UNSC Res. 1593 (2005), <www.icc-cpi.mt/library/organs/otp/OTP ReportUNSC4-Darfur English.pdf> (accessed on 16 December 2006).
conflicting interpretations of the icc statute
275
among scholars9 as among States10 about the legality of the text thus adopted or the true interpretation to be given to its provisions in order to reconcile it with existing international law. Time and time again, scholars will be asked to review the purposes and mandate of the ICC against the applicable principles of international law. The Statute itself offers in Article 19 (challenges to the jurisdiction of the court), Article 21 (Applicable law) and Article 119 (settlement of disputes), some of the legal instruments at the hands of States for the interpretation of the Statute. Among the interesting issues that merit detailed examination from a law of treaties point of view, this author will address the issue of legality of the ICC by reviewing four interrelated issues: I. The ICC as a novel mechanism to promote international peace and security outside the UN Charter framework; II. The ICC as a novel derivative subject of international law with ceded jurisdiction; III. The ICC as a novel mechanism to protect Member States against international obligations; IV. The ICC as a novel mechanism to enforce international obligations against non-Member States. At the end of the research, the author comes to the conclusion that the apparent contradictions within the text of the ICC Statute could be remedied if the ICC Statute were to be viewed as a treaty based on a concept of presumed passive consent of all members of the international community to be bound by its provisions. II. The ICC as a Novel Mechanism to Promote International Peace and Security Outside the UN Charter Framework Over the last century our understanding of the world order, including the role of international organizations and individuals as subjects of international law has changed dramatically. The collapse of the bipolar world and the end of the cold war has brought about new challenges.
9 See for example the ongoing discussion in US. Among those against the ICC Statute adopted and supporting the US position not to adhere to it, see Ruth Wedgwood, ‘The US and the ICC, the Irresolution of Rome’, Law & Contemporary Problems 64 (2001): 193; Madeline Morris, ‘High Crimes and Misconceptions: The ICC and Non-Party States’, Law & Contemporary Problems 64 (2001): 13. Among those supporting the ICC and condemning the US for not adhering to it see, Michael P. Scharf, ‘The ICC’s Jurisdiction over Non-Party States: a Critique of the US Position’, Law & Contemporary Problems 64 (2001): 67; Thomas Frank, ‘The United States and the ICC: Unilateralism Rampant’, New York University Journal of International Law and Politics 35 (2003): 519. 10 The well known US opposition to the Statute has served as a basis of this analysis in the paper. States in favor are obviously the ones that have so far ratified the Statute and have thus enabled the start of its work.
276
josé doria
It is generally agreed that many of the events that have occurred in the recent past like the egregious violations of human rights in Iraq,11 Haiti,12 Somalia,13 Yugoslavia,14 Rwanda,15 and the mounting of terrorism16 have shocked the conscience of mankind and seriously threatened international peace and security, and attained the magnitude that they have acquired due to the vacuum of power and political pressure created by the collapse of the bipolar world.17 These challenges require a revitalization of international law, in order to cope with the problems and changing demands of the contemporary world. One of such problems concerns the legal foundations and structure of the universal system of maintenance of international peace and security which is being redefined by new developments in the field of prosecution and punishment for the crime of aggression, genocide, serious war crimes, and crimes against humanity which “threaten the peace, security and well-being of the world”.18 In particular with the creation of the new ICC and its relationship model with the UNSC on matters directly related to the maintenance of international peace and security, it is not clear whether this new system is complimentary or concurrent with the UN universal system centered on the powers of the UN Security Council and its primary responsibility for the maintenance of international peace and security. Clearly if the new system is concurrent with the UNSC then the question arises whether it was properly established by way of a multilateral treaty without amending the UN Charter.19 11
See UNSC Res. 688 (5 April 1991) (situation of Kurds in Iraq). See UNSC Res. 940 (31 July 1991) (Haiti). 13 See UNSC Res. 794 (3 December 1992) (Somalia). 14 See UNSC Res. 827 (25 May 1993) (former Yugoslavia). 15 See UNSC Res. 955 (8 November 1994) (Rwanda). 16 The Attacks on the twin towers of the World Trade Center in New York, 11 September 2001. 17 See James Crawford, ‘The ILC’s Draft Statute for an International Criminal Tribunal’, AJIL 88 (1994): 140, at 141, (noting that “…the end of the Cold War coincided with a recrudescence of ethnic or communal violence and mass violations of human rights, for example, in Yugoslavia, Iraq and Somalia.”) 18 ICC Statute, Preamble. 19 During the Rome Conference many States were indeed concerned that some of the far reaching proposals for the jurisdiction of the Court could not be made without amending the UN Charter. Thus Bangladesh affirmed that 12
Regarding the role of the Security Council unless the Charter of the UN itself was amended there was an inescapable link between the crime of aggression and the functions of the Security Council in response to acts of aggression.
Likewise Russia did not believe it was possible in principle for the provisions of the Charter to be amended by other international instrument; those provisions would override any others. Extreme caution was therefore called for in the drafting of the Statute.
See Cherif Bassiouni (ed.), The Legislative History of the International Criminal Court: Summary Records of the 1998 Diplomatic Conference (New York: Transnational Publishers, 2005), (hereinafter The Legislative History of the ICC Statute), at 141, para. 13 and at 186, para. 9 respectively.
conflicting interpretations of the icc statute
277
The existing system of maintenance of international peace and security, arguably allows for many players, not just the UNSC.20 The spirit and the letter of the UN Charter provide that measures that are more appropriate for action by regional or other universal entities should be taken as a rule by such other more appropriate entities, however nothing shall prevent the UNSC from taking action if it is willing to do so. When the UNSC is seized of a situation, other such organs of the system are not encouraged to deal with the same situation, in any way they should not do so in a manner which could be inconsistent with or jeopardize the work being done by the UNSC. Arguably prosecution and punishment of serious egregious acts, is a function which is more appropriate for a judicial organ. However the UNSC may find that doing justice would otherwise jeopardize efforts to achieve peace through other means such as truth and reconciliation commission or amnesties for example. Under Articles 10 and 11 of the UN Charter, the UN General Assembly may discuss any question relating to peace and security and make recommendations to member States and to the UNSC. The ICJ by fulfilling its mandate of principal judicial organ of the UN can and does contribute to maintaining international peace and security. Indeed under Article 33 of the UN Charter member States are required to seek a solution to their disputes “the continuance of which is likely to endanger the maintenance of international peace and security” by resorting to judicial settlement, regional agencies or arrangements or “other peaceful means of their own choice”. Moreover pursuant to Article 95 nothing in the Charter prevents member states “from entrusting the solution of their differences to other tribunals by virtue of agreements already in existence or which may be concluded in the future”. Further Article 52 of the UN Charter, also provides that: nothing in the Charter precludes the existence of regional arrangements or agencies for dealing with such matters relating to the maintenance of international peace and security as are appropriate for regional action, provided that such arrangements or agencies and their activities are consistent with the Purposes and Principles of the United Nations.
Finally under Article 51, UN member States may resort to measures of individual or collective self-defense. Collective measures of self-defense arguably can include mechanisms established for the prosecution of serious war crimes committed during wars of aggression.21
20 During the Rome Conference countries were cognizant of this possibility. Egypt for example opined that under Article 39 of the UN Charter, “the role of maintaining peace and security did not belong only to the Security Council, but also to other UN bodies, notably the General Assembly”. Id., at 182,183, para. 89. 21 Articles 53 and 107 of the UN Charter have in particular legitimized the establishment of the Nuremberg and Tokyo Tribunals in the aftermath of World War II.
278
josé doria
Based on these provisions it can be said that Members States of the international community were entitled to establish either universal or regional permanent judicial organs which are more appropriate in order to punish perpetrators of egregious violations of human rights as a measure contributing to international peace and security, since these objectives are consistent with the purposes and principles of the United Nations. However the establishment and recourse by States to such other specialized mechanisms of promoting international peace and security (universal or regional) are only lawful provided that the primary responsibility of the United Nations Security Council for maintenance of international peace and security is recognized, no matter whether other more specialized mechanisms such as judicial organs exist to punish perpetrators of egregious crimes against humanity. This essentially means that States are entitled to have recourse to other more appropriate means of their choice only until such time as the UNSC itself does not take measures, or does not recommend other appropriate mechanisms to deal with a particular situation.22 Thus under Article 12 of the UN Charter if the Security Council is exercising in respect of any situation endangering peace and security, the functions assigned to it, the General Assembly would not be allowed to make any recommendation with regard to that situation. Upon ceasing to deal with that situation, the Security Council would so notify the General Assembly and Member States. There are no time limits imposed on this notification obligation for example. Likewise under Articles 34 and 36, the UNSC is entitled to deal with any situation in order “to determine whether the continuance of the dispute or situation is likely to endanger the maintenance of international peace and security” and at any stage (irrespective of any measures already taken by member States or international organizations of those states) may make recommendations to them on the appropriate procedures and ways of returning peace and security to a certain country or region. Moreover under Article 53 Member States can have recourse to regional organizations to deal with situations endangering peace however such organizations cannot take enforcement measures without being authorized to do so by the UNSC. And measures of collective self-defense taken under Article 51 should not in any way affect the authority and responsibility of the Security Council under the present Charter to take at any time such action as it deems necessary in order to maintain or restore international peace and security.
Finally one of the essential features of the central role of the UNSC in the system of maintenance of international peace and security is that decisions taken 22
See Articles 12, 34, 36, 39, 51, 53 of the UN Charter.
conflicting interpretations of the icc statute
279
under Chapter VII bind all Members and non Member States of the United Nations23 whereas decisions of regional and other entities bind only State Parties to those organs and entities. Accordingly if the new specialized judicial organ established with the aim of contributing to international peace and security respects the primary role of the UNSC in the global system then it is a welcome and lawful complimentary mechanism which perfectly fits within the existing UN system of maintenance of international peace and security. However if the new specialized judicial organ is given power to override or jeopardize the primary competence of the UNSC to deal with situations endangering international peace and security, and is factually viewed as a concurrent organ with the UNSC, then it would be in serious constitutional conflict with the UN system. Under Article 103 of the UN Charter, the United Nations system would take precedence.24 This could mean that the ICC Statute in what it denies or purports to limit the UNSC exercise of its primary responsibility might become itself a threat to peace and security requiring the UNSC and Member States’ action. In such case the ICC Statute, the Court and its activities could be ultra vires and countries of the world could be entitled to oppose it by any means available to them. This has been a very delicate issue raising conflicting interpretations both by ICC supporters and non supporters. With this overview of the existing global system of maintenance of international peace and security and how it works, we can embark in researching how the ICC subsystem fits within it. A. The Commission of Crimes of Aggression, Genocide, War Crimes, and Crimes against Humanity as a Threat or Breach of International Peace and Security On various occasions the UNSC has identified particularly egregious crimes committed against the civilian population as a threat or breach of international peace and security, and has called for their punishment. Thus in the aftermath of the events on the territory of the former Yugoslavia, the UNSC acting under Chapter VII, determined that the magnitude of the human tragedy in these countries of the former Yugoslavia amounting to
23
See UN Charter, Article 2(6). During the Rome Conference, countries have argued about the need to respect the central role of the UNSC. Kazakhstan for example noted that 24
the Council’s role in such an important issue as the determination of acts of aggression should not be disregarded. It might be true that the Council was not adapted to present-day circumstances and that the Charter should be amended accordingly, but for the time being the Council’s role must be recognized.
See Bassiouni, supra note 19, at 187, para. 13.
280
josé doria
serious war crimes, genocide and crimes against humanity constituted a threat to international peace and security and decided to establish an international criminal tribunal to punish the perpetrators of such atrocities, as a measure to promote international peace and security.25 A similar finding was made in relation to the events in Rwanda and a tribunal was also established to punish the relevant perpetrators.26 The International Law Commission (ILC) in its work on a Draft Code of crimes against the peace and security of mankind has formulated a set of acts of serious gravity and of concern to the international community as a whole that include acts of aggression, war crimes, genocide and crimes against humanity, to which individual criminal responsibility attaches.27 Finally the International Criminal Court was established as a permanent judicial organ with jurisdiction to prosecute acts of aggression, genocide, crimes against humanity and war crimes, which “threaten the peace, security and well-being of the world”.28 In all these instances the international community has easily found the sound interrelationship existing between serious violations of human rights on one side, and its impact on international peace and security on the other, and determined that such acts seriously endanger peace and security for which appropriate measures should be taken, including prosecution and punishment. No one disputes this and commentators have correctly argued that since under the UN Charter, promotion of universal respect for and observance of human rights and fundamental freedoms is necessary for peaceful and friendly relations among nations,29 “allowing massive human rights abuses to go unremedied leads to further violations, and serves to undermine the stability of the world order”.30 The UN Charter accordingly vests on the United Nations
25
See UNSC Res. 827 (25 May 1993), on the establishment of the ICTY. See UNSC Res. 955 (8 November 1994), on the establishment of the ICTR. 27 The ILC Draft Code of Crimes against the Peace and Security of Mankind adopted in 1996 lists the following crimes: crime of aggression (article 16), genocide (article 17), crimes against humanity (article 18), crimes against United Nations and Associated Personnel when committed in a systematic manner or on a large scale with a view to preventing or impeding that operation from fulfilling its mandate (article 19), war crimes when committed in a systematic manner or on a large scale (Article 20). 28 See ICC Statute (preamble); 2004 Relationship Agreement between the ICC and the UN (preamble). Recently the ICC Prosecutor in its Fourth Report to the UNSC on the situation in Darfur, the Sudan, explained that 26
The crimes are particularly serious given that they affect a population that is already suffering greatly due to the violence of previous years; moreover these perpetrators are standing in the way of progress towards peace and security in Darfur and the region.
See ICC Prosecutor, supra note 8. 29 See UN Charter, Article 55. 30 Cedric E. Evans, ‘The Concept of ‘Threat to Peace’ and Humanitarian Concerns: Probing the Limits of Chapter VII of the UN Charter’, Transnational Law & Contemporary Problems 5 (1995): 213, at 224.
conflicting interpretations of the icc statute
281
and in particular the Security Council competence to protect human rights and fundamental freedoms because of the detrimental effect their violation has on international stability, peace and security.31 While the issue of whether egregious violations of human rights amounting to genocide, crimes against humanity and war crimes should qualify or not as a threat to peace or security warranting individual and collective international efforts to prosecute and punish is settled,32 the key issue, as it has been rightly noted by one commentator, is to identify “who determines the characterization of a particular offense and on what bases”.33 The existing system of maintenance of international peace and security, as we noted, allows entities other than the UNSC to play a role in the process of identifying egregious human rights violations as endangering the international peace and security, and to work towards the elimination of such threats with measures that are more appropriate for those specialized entities. It devotes to the UNSC, however, a central role in this system, which means that the remaining entities are complementary, auxiliary mechanisms and not concurrent with the UNSC. The other key players in the system must recognize the leading role of the UNSC and defer to it any such issue when it is seized of any situation, the continuation of which might endanger the international peace and security. They also must pledge themselves to comply with any binding decisions taken to that effect under Chapter VII of the UN Charter. The leading position of the UNSC in the system was reflected in the establishment of the International Criminal Tribunals for the former Yugoslavia and Rwanda pursuant to Security Council resolutions adopted under Chapter VII. It was later endorsed by the 1994 Draft Statute of an International Criminal Court prepared by the International Law Commission (ILC Draft Statute).34 Under the ILC Draft Statute, the Court indeed would not have jurisdiction over situations being considered by the Security Council unless it decided otherwise,35 and there was no place for a Prosecutor with proprio motu powers of initiation of proceedings in relation to a situation.
31 Id.; see also Lois E. Fielding, ‘Taking a Closer Look at Threats to Peace: The Power of the Security Council to Address Humanitarian Crises’, University of Detroit Mercy Law Review 73 (1996): 551, at 560. 32 See the ICC Statute Preamble providing: “Recognizing that such grave crimes threaten the peace, security and well-being of the world” and
Affirming that the most serious crimes of concern to the international community as a whole must not go unpunished and that their effective prosecution must be ensured by taking measures at the national level and by enhancing international cooperation. 33 Timothy McCormack, ‘Selective Reaction to Atrocity: War Crimes and the Development of International Criminal Law’, Albany Law Review 60 (1997): 681, at 728. 34 See ILC, Draft Statute of an International Criminal Court, Report of the International Law Commission on the work of its 46th Session, 2 May-22 July 1994, (A/49/10). 35 ILC Draft Statute (Article 23(3) ).
282
josé doria B. The Role of Key Players in the Prosecution and Punishment of Serious International Crimes as Contributing Factors to Promote International Peace and Security
With the establishment of the ICC it can be said that there are now two key players together with the UNSC with special responsibilities to promote and restore international peace and security in situations of egregious human rights violations amounting to international crimes. The delimitation of their respective roles is however not yet clear cut as it will be seen in this research. 1. The UNSC and the Experience of the Two Ad Hoc Tribunals and Other Internationalized Tribunals The two Tribunals were established by binding UN Security Council resolutions under Chapter VII of the UN Charter. In addressing the objectives pursued with the option of establishing the ICTY, the French representative stated that while prosecuting the guilty was necessary for punishment and deterrence purposes, i.e. to do justice to the victims and to the international community, and to send a clear message to those who continued to commit these crimes that they would be held responsible for their acts, it was also first of all for the UNSC a matter of fulfilling its duty to maintain and restore peace.36 The ICTY Appeals Chamber in Tadic has confirmed that the establishment of the Tribunal was a proper exercise of the UN Security Council’s authority, and that the Security Council has resorted to the establishment of a judicial organ in the form of an international criminal tribunal as an instrument for the exercise of its own principal function of maintenance of peace and security.37
The Appeals Chamber also stated that whether it was the appropriate thing to do, was for the Security Council, “which enjoys wide discretionary powers in this regard”, to decide.38 Some of the principal characteristics of these Tribunals include primacy over national jurisdictions, and the fact that as an enforcement measure under Chapter VII, all States had binding obligations to cooperate with the Tribunals.39 While these characteristics have proved effective in dealing with the situations in the former Yugoslavia and Rwanda representing extreme real
36 See Statement of the French Representative, (1993) (S/PV.3175), UNSCOR, 48th Session., 3175th meeting, at 8. 37 See Prosecutor v. Dusko Tadic, Decision of 2 October 1995 of the Appeals Chamber on the Defence Motion for Interlocutory Appeal on Jurisdiction, ICTY, (IT-94-1-AR72), [hereinafter Tadic case], para.38. 38 Tadic case, para.39. 39 All these features substantially distinguish the Tribunals from the ICC established on a consensual basis and therefore respectful of the Member States’ sovereignty.
conflicting interpretations of the icc statute
283
threats to international peace and security, most countries of the world and scholars were not so convinced about the wisdom of having a permanent Court centered on UNSC powers too. On the basis of this unwillingness were not only objective limitations40 of judicial organs created as an exercise of the functions of the UNSC, but most importantly policy considerations. Essentially the selectivity involved in UNSC decisions (in determining what was a threat to peace and what was not) was viewed as contrary to a sound model of justice which ought to be impartial. States argued a new permanent court should have the essential features and objectives of an ordinary national criminal court first and maintaining international peace and security as a corollary to those objectives. Only this, it was argued could lead to impartiality in the discharge of justice. For these reasons, during the Rome Conference many countries were opposed to any role of the UNSC in the new Court. Libya for example justified its opposition to the UNSC with the fact that “the Security Council and its decisions were influenced by the interests and positions of certain permanent members, so that its resolutions were selective and followed a double standard.”41 Pakistan argued that “the UNSC was primarily a political body, and its decisions were based on political considerations rather than legal principles”.42 Lebanon was also of the view that the UNSC should have no role in it as the Court should not have to wait until the SC took a decision on the question of a military threat, act of aggression or breach of peace.43
Commentators have also criticized the selectiveness involved in tribunals centered on UNSC powers. For example Theodor Meron was concerned about the selectivity involved in a system where the establishment of a tribunal for a given conflict situation depends on whether consensus to apply Chapter VII of the UN Charter can be obtained. What is needed is a uniform and definite corpus of international humanitarian law that can be applied apolitically to internal atrocities everywhere, and that recognizes the role of all states in the vindication of such law.44
40 Commentators have pointed to a number of limitations of the ad hoc Tribunals model. See Philippe Kirsch, ‘The International Criminal Court and the Enforcement of International Justice’, Pace International Law Review 17 (2005): 47, at 49; noting the following:
They are geographically limited. They respond primarily to events in the past. Their establishment involved substantia] costs and delays. And last but not least, their creation depends on the political will of the international community of the day. In the case of an ad hoc tribunal, they required a decision by the Security Council acting under Chapter VII of the Charter. Because of all these factors, both the punishment and the deterrence function of the ad hoc tribunals are limited. 41
Bassiouni, supra note 19, at 135, para.79. Id., at 143, para.41. 43 Id., at 160, para. 17. 44 Theodor Meron, ‘International Criminalization of Internal Atrocities’, AJIL 89 (1995): 554, at 555. 42
284
josé doria
Timothy McCormack agreed that while there is no suggestion here that the alleged atrocities committed in the former Yugoslavia and Rwanda should not be prosecuted, the argument remains that the lack of commitment to enable international prosecution of alleged atrocities in other conflicts (and in the case of Burundi and Zaire, the same conflict as that which engulfed Rwanda) is inconsistent and selective.45
Likewise James Crawford wrote that: It is certainly arguable that the Security Council’s Chapter VII powers extend to the creation of criminal tribunals where necessary as part of an approach to resolving a conflict. But the creation of such tribunals by executive resolution in the exercise of emergency powers is less than satisfactory. The principle of legality is of particular significance in criminal cases. It connotes that criminal courts be established on a secure constitutional base; that the law to be applied be sufficiently defined in advance; that court personnel, the judges especially, have security of tenure, and that they be independent of executive interference in the performance of their judicial functions.46
Other commentators were concerned about the narrowness of the objective of maintaining peace (i.e. the new court should not wait until peace is considered endangered before it takes action), which is the basis of UNSC actions, as opposed to the interest of just doing international justice and enforcing relevant international norms punishing such acts. Thus one commentator has opined that the UN Security Council resolutions creating the ad hoc Tribunals for Yugoslavia and Rwanda indicate that the one place for criminal courts in the UN system is in support of the UN Security Council’s function of preserving and maintaining international peace and security. We should recognize that this is the objective and not expect that it will be best filled by creating international tribunals identical with domestic criminal courts.47
A purely legalistic model of international justice was not however pursued in Rome48 as States realized that in an international set there was a potential real risk of conflict between the mere objective of pursuing justice and punishing perpetrators and the broader objective of ensuring peace which has more of political than judicial elements and therefore is more appropriate for UNSC actions. 45
McCormack, supra note 33, at 727. James Crawford, ‘The ILC adopts a Statute for an International Criminal Court’, AJIL 89 (1995): 404, at 416. 47 Colin Warbrick, ‘The United Nations System: A Place for Criminal Courts?’, Transnational Law & Contemporary problems 5 (1995): 237, at 260. 48 Some commentators still think that what was established was indeed a legalistic model. See John M. Czametzky and Ronald J. Rychlak, ‘An Empire of Law? Legalism and International Criminal Court, Notre Dame Law Review 79 (2003): 55, at 61; noting that: 46
The ICC as an institution is the result of absolute faith in a nonpolitical, legalistic model of justice: where human rights violations have occurred, prosecutions must take place either on the national level or in the ICC. The ICC is, therefore, the apotheosis of ‘legalism’.
conflicting interpretations of the icc statute
285
Cognizant of these problems, but fearing abuses by the UNSC, the States in Rome rejected a UNSC central role in the new permanent system,49 by empowering an independent Prosecutor with powers of a political nature arguably contra balancing against the UNSC (such as the power that the Prosecutor would have to initiate proceedings proprio motu or not to initiate such if he found that not to be in the interests of justice to do so, even if requested by the UNSC).50 It still remains to be seen however whether the model adopted in Rome centered on powers of an independent Prosecutor with wide political discretion is better than having the UNSC doing the same thing, apart the constitutionality problems that it raises. Henry Kissinger for example has suggested that the ICC model adopted in Rome with no moral or political checks as part of its structure is a flawed one which can only potentially cause the Court to become in the long run either irrelevant or dangerous.51
2. The Intended Role of the ICC as the New Key Player in the Global System of Maintenance of International Peace and Security through Criminal Prosecution of Egregious Atrocities The ICC was established on 17 July 1998 with competence to adjudicate cases of egregious violations of human rights, in particular when these crimes threaten the peace, security and well being of the world. While the mention of crimes that “threaten the peace, security and well being the world”52 could also be understood as a reference to the powers of the UNSC to trigger the Court’s jurisdiction under Chapter VII, a closer analysis of the provisions and structure of the Rome Statute indicates that the role assigned to the Court is not a purely legalistic one, as many would have wished.53 Indeed as one respected commentator has once recognized, 49 During the Rome Conference the UK and some other countries argued that such fears were unfounded. See Bassiouni, supra note 19, at 183, para. 95; the UK noting that it
was somewhat puzzled by the fears expressed by some delegations that such referrals would interfere with the independence of the Court simply because the Security Council was a political body: no one had accused the Security Council of interfering with the independence of the tribunals for the former Yugoslavia and Rwanda, which had already been in operation for some time. 50
See ICC Statute, Article 53. Henry A. Kissinger, ‘The Pitfalls of Universal Jurisdiction’, Foreign Affairs (July–Aug. 2001): 86, at 95–6. 52 ICC Statute (Preamble). 53 Colombia for example argued in Rome that 51
“the Court’s independence and impartiality should be clearly reflected in the Statute. While the Court should have an institutional relationship with the United Nations, none of the latter’s bodies ought to have any influence over the Court or be able to obstruct its activities.”
Colombia therefore “viewed with concerns those provisions of the draft which might enable the Council to thwart the jurisdiction of the Court”. See Bassiouni, supra note 19, at 185, para. 6. See also supra notes 41, 42 and 43.
286
josé doria At the fundamental level, the purpose of creating an International Criminal Court is to contribute to maintaining international peace and security, which obviously could not be done while inhibiting the operation -difficult enough at the best of times – of the Charter organ tasked with establishing that very international peace and security.54
States assembled in Rome were indeed in no doubt that the Court was to play an important role in the global system of maintenance of international peace and security. Thus Japan argued that: the relationship between the Security Council and the Court was a key issue. Since crimes within the Court’s jurisdiction would be the most serious crimes of international concern, and since the Council was the principal organ responsible for maintaining international peace and security, it was right that the latter should have a role under the Statute.55
Likewise Russia was of the view that there was no conflict between the ‘political’ role of the Council and the activities of the Court. The Council was intended to have a political impact on States and the Court would be playing an essential role in the maintenance of peace and security.56
It further supported the draft preamble of the ICC Statute referring to the United Nations because the purposes and the principles of the Charter of the United Nations concerning the maintenance of international peace and security and respect for human rights were directly connected with the activities of the future Court.57
At the same time, States had no doubts that it was legally proper to establish the ICC as a new organ with a role in the global system of maintenance of peace and security, since as Ukraine emphasized for example, the UNSC was “not the sole body responsible for the maintenance of international peace and security”.58 The role of the Court in promoting the maintenance of international peace and security became also reflected in the Relationship Agreement of the Court with the UN in which it is noted the important role assigned to the ICC in dealing with the most serious crimes of concern to the international community as a whole, as referred to in the Rome Statute, and which threaten the peace, security and the well-being of the world.59 54 See Sir Franklin Berman, ‘The Relationship between the International Criminal Court and the Security Council’, in Reflections on the International Criminal Court, Herman A. M. von Hebel et al. (eds.) (The Hague: TMC Asser Press, 1999), 173, at 177. 55 Bassiouni, supra note 19, at 183, para. 97. 56 Id., at 186, para. 9 (emphasis added). 57 Id., at 256, para. 32. 58 Id., at 185, para. 4. 59 See 2004 Relationship Agreement between the UN and the ICC.
conflicting interpretations of the icc statute
287
This Statement is not mere assumption it is based on specific provisions of the ICC Statute. In particular this can be seen from four important provisions of the ICC Statute: i. The ICC is given jurisdiction in relation to crimes committed by nationals of non-state parties.60 There is nothing in the ICC Statute otherwise indicating that this provision would apply only if the UNSC would trigger the Court’s jurisdiction; ii. Under the ICC Statute the UNSC is allowed to defer proceedings only for one year renewable,61 after which the Prosecutor is entitled proprio motu to prosecute a case if in his view the situation represents a threat to international peace and security. Moreover under the Statute all traditional Stateto-State conditions apply, including the rules of complementarity, and the immunity regime under Article 98(1), even when jurisdiction is triggered by the UNSC.62 Additionally the wide discretion given to the Prosecutor, includes the power not to initiate an investigation or to suspend such if it is in the interests of justice not do so, even in situations referred to him by the UNSC (for example in cases of amnesties63);
60
See ICC Statute, Article 12 (2) (a). See ICC Statute, Article 16. The UNSC had once attempted to extend this period for an indefinite time by adopting yearly resolutions to protect UN peacekeepers from the Court’s jurisdiction. Countries pressed that it would be illegal, first because the UNSC should only have recourse to Article 16 when there was real (not imaginable) threat to peace and security, and second because the Article was clearly intended to impose time limits to the deferral powers of the UNSC. The UNSC was forced to change its mind. See UNSC res. 1487 (12 June 2003), paras. 1,2 stating that 61
acting under Chapter VII requests consistent with the provisions of Article 16 of the Rome Statute, that the ICC, if a case arises evolving current or former officials or personnel from a contributing State not a Party to the Rome Statute, over acts or omissions relating to a United Nations established or authorized operation, shall for a 12-month period starting 1 July 2003 not commence or proceed with investigation or prosecution of any such case, unless the Security Council decides otherwise… expresses the intention to renew the request in paragraph 1 under the same conditions each 1 July for further 12-month periods for as long as may be necessary. 62
See ICC Prosecutor, supra note 8, at 2; where the Prosecutor notes that In the first report to the UNSC the Office emphasized that the ICC is a court of last resort, only acting in situations where: (I) there has not been any national investigation or prosecution of the case being brought to the Court; or (II) where there is or has been, such an investigation or prosecution, but it is vitiated by an unwillingness or inability to genuinely carry out that investigation or prosecution.
It is true however that in the particular situation of Darfur the UNSC resolution encourages the Court to support local judicial efforts and truth commissions. Therefore the UNSC empowers the Court to do so, but it cannot be assumed for granted that the UNSC would do so in every such situation (as circumstances in which crimes are committed might change). The ICC Statute cannot limit the UNSC powers. Decisions taken under Chapter VII are normally viewed as sanctions against the State concerned (Articles 5, 25, 39 of the UN Charter), and other members of the international community are required to cooperate fully with the UNSC (Articles 25, 43, 48 UN Charter), unless the UNSC decides otherwise. 63 See ICC Statute, Article 53.
288
josé doria
iii. Under the ICC Statute a special regime applies for war crimes. The Court is encouraged to assert jurisdiction particularly when the crimes are a result of widespread or systematic policy,64 and the Prosecutor is required to assess the gravity of the crimes before launching an investigation.65 Further, countries have a right to opt-out of the ICC’s jurisdiction in relation to war crimes66 (it is had in mind war crimes short of being considered crimes against humanity or genocide since these would constitute ipso facto a serious threat to peace and security).67 And finally a strong regime of exempting defenses68 against liability in case of petit war crimes is pushed forward that would if applied further effectively restrict the jurisdiction of the court to those war crimes situations amounting again to a serious threat to peace and security;69 iv. The jurisdiction over the crime of aggression.70 Therefore it can be said that these provisions justify the ICC role as a specialized organ with responsibilities in the promotion of international peace and security in the specific area of prosecution and punishment of serious human rights violations.71 This of course does not deny the more general objectives of
64
See ICC Statute, Article 8(1). ICC Statute (Articles 17(d) and 53(1) (c), 53(2) (c) )./ 66 See ICC Statute, Article 124. 67 During the Rome Conference France justified the need for an opt-out procedure for war crimes and suggested that whereas 65
acceptance of the jurisdiction of the Court could be obligatory for any State becoming a party to the Statute with respect to the crime of genocide and crimes against humanity. War crimes however, as defined in the 1907 Hague Convention and the 1949 Geneva Conventions and additional Protocols thereto, might be isolated acts. A solution must be found to enable States with particular difficulties in that area to be able to become parties to the Statute. It was not a matter of drawing up an a la carte Convention, but of allowing some flexibility. There could be a system requiring consent by the State of nationality of the perpetrator, so that the Court could exercise its jurisdiction.
Bassiouni, supra note 19, at 154, para.49 (emphasis added). 68 See ICC Statute, Articles 31–33. 69 Just to compare, defenses for war crimes under the ICTY jurisprudence are only allowed as a mitigating factor and not as a basis to exclude criminal responsibility. Yet prosecutions at the ICTY are justified since the entire situation in the former Yugoslavia (not just the specific war crimes) was already viewed as a breach or threat to international peace and security, in which many of the war crimes committed, also amounted to serious crimes against humanity or genocide. 70 See Article 5(2) ICC Statute, providing that “ the Court shall exercise jurisdiction over the crime of aggression once a provision is adopted in accordance with articles 121 and 123 defining the crime and setting out the conditions under which the Court shall exercise jurisdiction with respect to this crime. Such a provision shall be consistent with the relevant provisions of the Charter of the United Nations. 71 This view is generally supported by scholars too. See Luigi Condorelli and Santiago Villalpando, ‘Relationship of the Court with the United Nations’, in The Rome Statute of the International Criminal Court: A Commentary, Antonio Cassese et al. (eds.) (Oxford: OUP, 2002), 219, at 221 noting that
the function of the Court is considered to have an intrinsic connection with the purposes of the UN, especially with the Role of the Security Council. In fact, the Court is intended to contribute to the
conflicting interpretations of the icc statute
289
the Court which are punishment, justice and deterrence.72 On the contrary it is through these actions that the Court can play its role in promoting peace, security and justice.73 Obviously playing such a role does not require a preliminary finding of a threat to peace or security by it or any other organ. It would be enough that the Prosecutor found that the specific crimes were grave enough to warrant investigations and prosecutions.74 The problem therefore is not that these provisions in the Rome Statute were clearly intended to give the ICC a role in the global system of maintenance of international peace and security. This would still be perfectly lawful, as we reviewed it in the sections above. However, the problem is that these provisions are potentially conflicting with the UN and general international law. How? We will see it below by critically accessing the “constitutionality” of the four ICC provisions just referred above. 1. For example under the existing system of maintenance of international peace and security, nationals of non-party States can be constrained to maintenance and restoration of international peace, both under a preventive perspective – as its mere existence can hinder potential criminals from committing heinous crimes that threaten the peace, security and well-being of the world – and more directly – by contributing to general reconciliation (and thus effective restoration of peace) through the punishment of those who have committed such crimes. 72
See Kirsch, supra note 40, at 49 noting that: The rationale behind creating international tribunals is that when united systems fail, they can further three objectives: to punish individuals responsible for the most serious international crimes; to bring justice to victims; and, over time, to contribute to the deterrence of potential perpetrators and the building of a culture of accountability. The work of international criminal tribunals is also linked to maintaining security, guaranteeing lasting respect, and the enforcement of international justice.
73 For example the ICC Office of the Prosecutor considers that the role of the Court in contributing to international peace and security will be better accomplished by maximizing the impact of its activities as deterrence to future crimes. The Prosecutor has affirmed this as one of the guiding principles of his prosecutorial strategy and asserts that:
as noted in the Preamble of the Statute, the Court has a role in contributing to the prevention of future crimes. The Office has to maximize the impact of each of its activities, from the analysis of the information, to the beginning of the investigation, to the trial and eventual conviction. Massive crimes are planned; the announcement of an investigation could have a preventive impact. The mere monitoring of a situation could deter future crimes from being committed. It increases the risk of punishment even before trials have begun. Interestingly, this effect is not limited to the situation under investigation but extends to different countries around the world.
See ICC – Office of the Prosecutor, Report on Prosecutorial Strategy (The Hague, 14 September 2006), <www.iccnow.org/documents/ProsecutorialStrategy_06Sep14.pdf> (accessed on 1 May 2009), at 6 74 In the ICC Prosecutorial Strategy the Prosecutor pledges to have adopted a ‘sequenced’ approach to selection, whereby cases inside the situation are selected according to their gravity. Although any crime falling within the jurisdiction of the Court is a serious matter the Statute clearly foresees and requires an additional consideration of “gravity” whereby the Office must determine that a case is of sufficient gravity to justify further action by the Court. In the view of the Office, factors relevant in assessing gravity include: the scale of the crimes, the nature of the crimes, the manner of commission of the crimes.
ICC – Office of the Prosecutor, supra note 73, at 5.
290
josé doria
respect decisions of organs to which their respective States are not a party only through decisions taken by the UNSC under Chapter VII.75 For example the UNSC has established the ICTY and the ICTR with compulsory jurisdiction over anyone who committed crimes on the territory of the former Yugoslavia and Rwanda. When the UNSC decided to establish these Tribunals, Yugoslavia was not even asked, and Rwanda voted against the decision.76 Yet coercive measures taken by the UNSC under Chapter VII are viewed as perfectly lawful. The ICC however is a consensual organ of States parties created outside the UN Charter.77 As such it could in principle only bind citizens of Member States. Instead it has been given powers to adjudicate acts committed by citizens of non-party States too. Curiously though an exception to this rule is actually made, but in relation to citizens of Member States that have specifically requested so under Article 124 of the ICC Statute. This clearly reflects ambitions of a mechanism of ensuring peace and security with a status similar to that of the UNSC, i.e. a concurrent mechanism not a complimentary one. This however could in principle only be done by amending the UN Charter or with the support of the overwhelming majority of the more than 200 States of the world. 2. The ICC Statute is based on two fundamental pillars: the principle of complementarity (Article 17) and the active leading role of an independent Prosecutor for example under Articles 16 and 53 of the ICC Statute. The principle of complementarity for example in the way in which it is being interpreted by the Prosecutor would allow him to dilute the UNSC powers under Chapter VII, by requiring that UNSC referrals be also subject to the checks and balances of the Court’s system, such as unwillingness, immunities, etc.78 75 See UN Charter, Articles 2(6), 32 and 35. During the Rome Conference several States were concerned about this. The US for example stated that it
had grave difficulties with establishing a Court that presumed to have jurisdiction over the citizens of a State that had not ratified the treaty creating it, except in situations where the Security Council had taken enforcement action under Chapter VII of the Charter of the United Nations, which was binding on all Member States.
Bassiouni, supra note 19, at 309, para. 42. 76 See Adoption of the UNSC Res. 955 (8 November 2004), Provisional Verbatim Records, (S/PV.3453). 77 Sweden for example “was not fully convinced by the representative of Germany’s arguments about inserting the Court fully into the system of universal jurisdiction, even with regard to the core crimes’. It stressed that “the Court was being created by a Convention, and some regard must be had to that fact”. Bassiouni, supra note 19, at 148, para. 69. 78 During the Rome Conference States were concerned about these unlawful restrictions on the work of the UNSC. The US for example stated that it was willing to work with others to find a compromise with respect to the Council’s proper role, but the powers and functions of the Council must not be rewritten. Wording was needed that did not impose an obligation on the Council to draft its own resolution with a specified period for its applicability.
Bassiouni, supra note 19, at 309, para. 48.
conflicting interpretations of the icc statute
291
In addition the Prosecutor has under Article 53 a real chance of deciding contrary to the UNSC that prosecution is not the best way of ensuring peace, by allowing for example consideration of amnesties. Suppose a situation like in Sierra Leone where the UNSC came to the conclusion that perpetrators of egregious violations of human rights should stand trial and that the amnesty that was declared in an earlier peace agreement was irrelevant.79 An independent Prosecutor using his discretionary powers could at least in theory conclude under the ICC Model that the declared amnesties should prevail, and that there was no place for prosecutions.80 Further, using the powers assigned to him under Article 16, the Prosecutor could at least in theory conclude at any time after one year (renewable) that the UNSC review of a situation is no longer justified irrespective of what the UNSC thinks and proceed with investigations once suspended by the UNSC under Chapter VII. 3. The opt-out mechanism for war crimes and other crimes for as long as it pretends to legitimize violations of international law considered as grave breaches, is also far beyond what even the UNSC could afford itself to do. Therefore it is also prima facie contrary to the UN Charter and international legality generally.81 4. The disputed role of the UNSC in situations involving the crime of aggression was one of the reasons behind the delayed consideration of the inclusion of this crime within the jurisdiction of the Court.82 Some States had
79 On Sierra Leone see UNSC Resolution 1325 (14 August 2000). See also UN Secretary General, Report of the Secretary-General on the Establishment of a Special Court for Sierra Leone (4 October 2000), <www.afrol.com/Countries/Sierra_Leone/documents/un_sil_court_041000 .htm> (accessed on 1 May 2009) 80 Under Article 53(3) (a) the UNSC is only given the consolation of challenging in Court the appropriateness of its referral decision. However this would be an unprecedented situation to have the UNSC that adopted a biding coercive Chapter VII decision to dispute with the State concerned and the Prosecutor the appropriateness of its decision that prosecutions are the best way of ensuring peace and security in a certain region, and that no amnesties should be recognized. 81 During the Rome Conference the ICRC reacting to the proposed opt-out mechanism for war crimes emphasized that
“any form of additional consent, such as an opt-in precondition to the exercise of the Court’s jurisdiction, might give the impression that States could lawfully protect war criminals from prosecution. That would be a retrograde step for international law and would severely limit the Court’s effectiveness”.
Similarly Switzerland stated that it could not endorse the adoption of models that “took the form of an opt-out mechanism for war crimes”. Bassiouni, supra note 19, at 382, para. 51 and at 343, para. 19 respectively. 82 For example Pakistan opposed “the inclusion of the crime of aggression because of its controversial nature”. It argued that aggression was traditionally considered a crime committed by States whereas Pakistan favoured the principle that the Court’s jurisdiction should be limited only to crimes committed by individuals. That raised the complex problem as to how an individual might be prosecuted and punished for aggression, unless the Security Council first determined the existence of aggression, and that then those
292
josé doria
pretended that the ICC alone be entitled to deal with this issue.83 However to dispute the role of the UNSC in these situations would be a clear violation of the letter of the UN Charter regarding the Chapter VII prerogatives of the UNSC,84 apart from the purely practical problems involved (which will be dealt with in another part of this paper below). These are all provisions of the ICC Statute that introduce considerable changes in the UN System.85 Under Articles 2(6), 5, 24, 25 of the UN Charter decisions of the UNSC under Chapter VII are binding upon all States and entities of the world (irrespective of their affiliation or not to the UN), and are not subject to the checks or challenges by States or any other entities or organizations created by them, unless otherwise decided by the UNSC itself.86 To state that the responsible were identified. In most cases those in authority would be the accused, something which threatened the concept of sovereignty of States.
Bassiouni, supra note 19, at 130, paras. 41–2. 83 During the Rome Conference Syria was against a UK proposal according to which the Court would be unable to exercise jurisdiction with respect to a crime of aggression unless the Security Council had first determined that a State had committed such an act. Syria argued that in over 200 cases dealt with by it, the Council had avoided making such a determination. It had become a club of superpowers, whose right of veto could protect thousands of international criminals by blocking the Court’s procedures.
Therefore, in order to have a Court that would deal with all who committed international crimes, Syria “was against assigning any role to the Security Council”. Bassiouni, supra note 19, at 186, para. 10. 84 Many countries supported this position in the Rome Conference. For example Israel suggested that it was inappropriate at the present stage to include the crime of aggression in the jurisdiction of the Court. However if it was included, determination by the Security Council under Article 39 of the Charter of the United Nations of the existence of an act of aggression should be a precondition to the exercise of the jurisdiction of the Court in so far as acts of aggression were concerned. That function, a basic function of the Council under Article 24 of the Charter, could not be ignored by the Statute, transferred to the Court or shared with the Court.
Likewise Bangladesh dismissed as invalid the suggested dichotomy between the UNSC that was purportedly a political body and the Court that was supposed to be a legal body. Bangladesh argued that when the Security Council acted under Article 39 of the Charter of the United Nations in cases involving the crime of aggression, its determination was the legal characterization of a situation. No Court could escape the binding effect of such a determination.
Bassiouni, supra note 19, at 171, para. 114 and at 186, para. ll, respectively. 85 During the Rome Conference Russia noted that with regard to deferral, it found it difficult to agree with any wording that might be interpreted as modifying the obligations of States under the Charter of the United Nations, in particular under Chapter VII. Moreover, the introduction of any time limit might be interpreted as affecting the Council’s powers under Chapter VII.” Likewise Namibia agreed, with regard to the role of the Security Council that “the Conference could not amend the Charter of the United Nations.
Bassiouni, supra note 19, at 314, para. 115, and at 315, para.128 respectively. 86 See Mahnoush H. Arsanjani, ‘Reflections on the Jurisdiction and Trigger Mechanism of the International Criminal Court’, in in Reflections on the International Criminal Court, Herman A. M.
conflicting interpretations of the icc statute
293
ICC Statute requires respect for such rules is of no avail, since Article 103 states clearly that UN Charter provisions take precedence. As one commentator noted, because participation in the United Nations has become almost universal, this ‘constitutional’ dimension of Article 103 still remains, since it manifests the priority of the Charter over any other commitments which may also be concluded between the Member States. It is especially remarkable that Article 103 applies not only to obligations laid down in the Charter but also to the decisions taken in conformity with it by the competent organs.87
At the end of this analysis it is sound to question the following: Can a treaty impose limitations on the ability of a UN Charter organ to fulfill its obligations under the UN Charter? The quick answer would be that these novel changes that the ICC Statute has purported to introduce into the UN system could only be made in principle by changing the UN Charter itself, or adopting at the very least a treaty with similar universal membership and acceptance like the UN Charter. In any case since under its Article 103, the UN Charter overrides contrary provisions of other treaties, we are faced in the ICC case with a serious ‘constitutional’ problem. C. Is the ICC ‘Constitutional’? Otherwise put the question is whether the novel mechanism can still be properly interpreted as conforming to the principle of legality under the UN Charter and the Vienna Convention on the law of treaties. And what are the implications of making changes into the global system of maintenance of peace and security outside the UN Charter framework? We will analyze this issue first from a pragmatic point of view and then from a legal one. 1. From the Viewpoint of Political Pragmatism The political pragmatism behind the establishment of the ICC with seemingly constitutional problems, can be seen through the eyes of John M. Czarnetzky and Ronald J. Rychlak who are convinced that what sparks faith in the ICC,
von Hebel et al. (eds.) (The Hague: TMC Asser Press, 1999), 57, at 70 correctly noting that: Article 17 on admissibility and Article 90 on the obligation of the third State regarding competing requests by the Court and by another State do not provide any priority for the jurisdiction of the Court even if the matter has been referred to it by the Security Council. The result may not be fully consistent with the original intention of empowering the Security Council with the right of referral which was to avoid the creation of ad hoc tribunals. 87 Pierre-Marie Dupuy, ‘The Constitutional Dimension of the Charter of the United Nations Revisited’, Max Planck Yearbook of the United Nations Law 1 (1977): 1, at 12.
294
josé doria is the dream that such an empire, ruled by law and not fallible human judgment, will largely eradicate the problem of impunity in international affairs through deterrence, and that violations of human rights will become an abnormality. Current political institutions have not stopped human rights violations of the worst sort, so the problem must be with those political institutions. The ICC is thus an attempt to transcend political institutions, both at the national and the international level, and enthrone law and its institutions as the governing authority of the world.88
In this author’s view the model ICC adopted in Rome although championed as the embodiment of impartiality89 is in indeed a reflection of a growing resentment by midlevel powers like Germany,90 and Canada91 (eager to play a major role in international relations too) of the ability of the five permanent members of the UNSC (the P5) not only to shield themselves from international scrutiny, but also to use their powers under Chapter VII to impose obligations on other States that heavily encroach upon their sovereignty, i.e. the unique ability that the P5 have to influence international developments in the way they find necessary. With the end of the cold war and the enhanced chances of agreement among the P5, added to the UNSC over expanding definition of threats to peace and security, the one time remote probability of using the mandatory powers under Chapter VII, has become a palpable reality for countries, and the example of Yugoslavia has only sparked concerns about the ‘unequal’ sovereignty of States. While it was thought that UN reforms were urgently needed to re-invert the perceived misbalances,92 until the situation changed substantial safeguards 88 John M. Czarnetzky and Ronald J. Rychlak, ‘An Empire of Law? Legalism and International Criminal Court’, Notre Dame Law Review 79 (2003): 55, at 62. 89 For example German argued during the Rome Conference that it believed that
in order to ensure the independence of the Prosecutor it was vital to give him or her the power to initiate investigations ex officio, since otherwise prosecutions could only be brought if a State party or the Security Council referred a situation to the Court. Giving the Prosecutor the power to act proprio motu have the advantage of depoliticizing the process of initiating investigations.
Likewise Sweden stated that it did not “wish the Court to be a mere tool of the Security Council; it should be truly effective, and the Prosecutor should be given an ex officio role”; Bassiouni, supra note 19, at 175, para. 14, and at 176, para. 17, respectively. 90 These countries were the undisputed leaders of the so called like-minded group of states at the Rome Conference and were the ones pushing for the most radical jurisdictional mechanism in the ICC, based on universal jurisdiction and with a marginal role of the UNSC. At the same time Germany is also one of the four countries members of the so-called G4 group of states pushing for a permanent seat in the UNSC. See Section C.2.(c)(1) infra. 91 Canada for example vividly supported the universal jurisdiction principle suggested by Germany as the sole basis of the Court’s jurisdiction stating that it was committed to a Court with inherent or automatic jurisdiction over the three core categories of crime: genocide, war crimes and crimes against humanity. An opt-in or State consent regime would allow States to veto Court action and would render the Court ineffective. The number of States whose acceptance was required must be kept to a minimum.
Bassiouni, supra note 19, at 160, para. 11. 92 For example Mexico noted in Rome that the “Conference was taking place at a time when the United Nations was discussing a number of proposals for reform of the Security Council”
conflicting interpretations of the icc statute
295
protecting the sovereignty first of all of the mighty States were suggested for the new ICC.93 It can with certainty be said that the extent of these protective safeguards sparking from complementarity and ending with the immunity provisions will normally ensure that the ICC will be a tool essentially used for the developing world with collapsed judicial systems,94 and never for countries with well developed judicial systems, democracy95 and rule of law traditions, and which can therefore have recourse to proper national proceedings.96 This objective achieved by empowering an independent Prosecutor with the right of scrutiny of the actions of any member of the international community on one side, and by significantly curbing the powers of the UNSC, on
and that “those discussion were relevant for the present debate”. In particular Mexico alerted that “the Conference should not repeat the mistake made at San Francisco by tying the new Court to the organs of the United Nations, like the International Court of Justice”. In its view the “Security Council would be one source of information for the new Court regarding the existence of situations involving aggression, but not the only one.”. Bassiouni, supra note 19, at 179, para. 52. 93 During the Rome Conference Germany approving of the complementarity principle held that “where national criminal justice systems were non-existent or unable or unwilling to prosecute a given serious crime, the International Criminal Court should exercise jurisdiction”; Bassiouni, supra note 19, at 110, para. 55. 94 Three years after the start of its work, the ICC is still uniquely focused on events on one continent (Africa). Reacting to growing criticism, the Prosecutor stated with a dose of cynics that his Office understood the “concerns about geographic focus, but regional balance is not a criterion for situation selection under the Statute”. See ICC Office of the Prosecutor, supra note 7, at 2 95 President Philippe Kirsch has indeed frankly recognized this by stating that In the case of democratic countries that have systems that function perfectly well, there is no reason why any issue should come before the ICC, because … a determination of how a proceeding is conducted does not depend on its outcomes. It does not require a prosecution. It could require an acquittal or even a decision not to prosecute. All that is required is that proceedings are conducted normally.
See Philippe Kirsch, Remarks on Accountability vs. Impunity: The Role of the International Criminal Court, Statement to the Council on Foreign Relations, Washington D.C. (16 January 2004), <www.cfr.org/publication/6696/accountability_vs_impunity.html> (accessed on 23 November 2006). 96 The Prosecutor informed in his first Report on activities that of the approximately 20% of communications warranting further analysis, 10 situations have been subjected to intensive analysis. Of these 3 proceeded to investigation (the DRC, Northern Uganda, and Darfur), two were dismissed (Venezuela, and Iraq), and five analysis are ongoing.
In the case of Iraq where the alleged crimes were committed by nationals of a state party to the ICC (the UK) that was one of the coalition states in the war in Iraq, the communication noted indeed that the information available supported a reasonable basis to believe that a limited number of instances of willful killing and or inhuman treatment had occurred. However the alleged crimes committed by nationals of State Parties in Iraq did not appear to meet the required gravity threshold. Additionally the Prosecutor noted that although it was not necessary to reach a conclusion on complementarity in light of the conclusion on gravity, national proceedings had been initiated in relation to each of the incidents.
ICC Office of the Prosecutor, supra note 7, at 9–10. See also ICC OTP Iraqi Response (9 February 2006), <ww\v.icc-cpi.int/library/organs/otp/OTP_etter_to_senders_re_Iraq_9_February_2006 .pdf> (accessed on 21 October 2006).
296
josé doria
the other, could raise however serious issues of ‘constitutional’ legality as a subtle amendment of the UN Charter. It is in particular difficult to see why ICC supporters are concerned about the negative influence that the political decision-making of the Security Council would have on the ICC,97 but at the same time believe in the ability of the Prosecutor to make the same prudential judgments correctly.98 It is not surprising therefore that many countries were both opposed to a role by the UNSC and to an independent Prosecutor.99 Therefore not being completely an impartial entity as long as it is essentially directed to deal with situations affecting the developing world, the model adopted in Rome could pose serious legitimacy problems for this reason too. In particular from a pragmatic point of view, one may question the appropriateness of such a universal court in lieu of similar regional institutions based on subsidiarity principles, similar to the institutional system of protection of 97
Norway for instance expressed the view that “a qualified and independent Prosecutor would be the best insurance against politicized action by the Court and should be able to deal with criticism in relation to the setting of priorities when there were many possible cases”. Likewise the Netherlands argued that “the Prosecutor should have the full use of all sources of information, from governmental and non-governmental sources as well as from victims’ associations”. Bassiouni, supra note 19, at 173, para. 130; at 173, para. 134 respectively. 98 For example the United States argued during the Rome Conference that the argument that the State and Security Council referral approach would mean a politicized Prosecutor, while the proprio motu approach would ensure an impartial one, seemed simplistic. It would be naive to ignore the considerable political pressure that organizations and States would bring to bear on the Prosecutor in advocating that he or she should take on causes which they championed. Both organizations and States might seek to act politically, but there was a significant difference in the accountability of States, as opposed to individuals and organizations… The proprio motu proposal thus risked routinely drawing the Prosecutor into making difficult public policy decisions he or she was neither well equipped nor inclined to make. Such initial public policy decisions would be best made elsewhere, freeing the Prosecutor to deal for the most part with the law and the facts.
Russia agreed that “if the Prosecutor was given direct power to initiate investigations, proprio motu, both the Prosecutor and the Court would become politicized”. Bassiouni, supra note 19, at 172, para. 127, and at 173, para. 130; at 173, para. 133 respectively. 99 Cuba argued for example that it was opposed to an independent prosecutor because even in this case conflicts of interest and jurisdiction would undoubtedly arise and politically motivated investigations could affect the credibility of the Court. A frank commitment to international cooperation was preferable to the so-called impartiality of one individual.
India that also opposed a UNSC role in the Court was of the view that the necessary cooperation would not be promoted by allowing the Prosecutor to act on his own, on the basis of sources of information, regardless of their reliability. Such an ex officio role for the Prosecutor would jeopardize the principle of complementarity which was generally accepted as the basic foundation for the establishment of the Court.
Iran thought it was premature to give the Prosecutor the power to initiate investigations on his own, the Court would be established on the basis of a multilateral treaty and would be an international criminal Court, but not a supranational Court, justifying the Prosecutor’s having ex officio powers of investigation.
Bassiouni, supra note 19, at 171, para. 118; at 169, para. 105; and at 167, para. 82 respectively.
conflicting interpretations of the icc statute
297
human rights which has both universal and regional components. Th is author asserts that the establishment of regional criminal Courts closer to the place of events, victims and witnesses might be a valid alternative and a more effective instance to combat impunity rather than the universal ICC. Eventually the universal and regional organs could be combined in a well elaborated division of labor. The existence of an already functioning ICC does not necessarily obviate the need to establish such regional criminal Courts. On the contrary the problems that the ICC is confronted with today might serve as the catalyst for the establishment of such other regional Courts in future.100 2. From a Legal Point of View: Is the New System Legal? If the new system is being modeled outside the UN Charter framework, is it still legal? Or it is our understanding of what is legal which needs a reshape? How to react to these new developments challenging the way we interpret the law? An answer to these questions can be given from three different angles (ICC supporters’ position, ICC non supporters’ position, and a neutral position). We will review each of these positions below. (a) Position One: Change the Way We Understand and Interpret the Law (ICC Supporters’ Approach) During the discussions about the role of the UN Security Council in a permanent Court, from the International Law Commission (ILC) to the Rome Conference, the essential divide was among those who worried that conferring too little powers to the UNSC could undermine the role of the Security Council (and therefore raise problems of constitutionality) and those who were concerned that conferring too broad powers on the Security Council (given the possibility of abuse of the veto powers), could undermine the judicial independence and integrity of the Court or the sovereign equality of States. Ultimately the ILC chose to confer to the UN Security Council broad powers, whereas the Rome Conference decided to limit these powers arguably beyond the UN Charter plank, raising conflicting issues of constitutionality. The 1994 ILC Draft Statute of an International Criminal Court provided indeed for the Court’s jurisdiction that depended on State consent but with the UN Security Council having priority jurisdiction under Chapter VII. No independent power of action was envisioned for the Prosecutor.101 This is
100 With the exception of the Americas, all other regions have had experience with internationalized forms of criminal tribunals (Europe (Nuremberg, and the Yugoslavia Tribunals, Kosovo UN administration mixed courts); Africa (Rwanda Tribunal and the Sierra Leone Special Court); Asia (the Tokyo Tribunal, the Extraordinary Chambers of Cambodia, and Timor Leste UN Administration mixed criminal Panels). 101 See ILC Draft Statute Articles 21,22,23 and 25; See also ILC, Report of the Working Group on the Question of an International Criminal Jurisdiction – Report of the International Law
298
josé doria
because it was understood that a permanent International Criminal Court would essentially deal with situations of egregious violations of human rights endangering international peace and security, however it was admitted that in certain circumstances a permanent Court would be entitled to deal with any situation amounting to international crimes falling under the jurisdiction of the ICC.102 Therefore, on one side, it was generally felt that situations in which Chapter VII was at issue ought to be signaled to the Court by the Security Council, but on the other side, that in other than Chapter VII cases, individual States should have referral rights. In addition by analogy with the cases of occurrence of armed attacks that entitle countries to use force in self-defense under Article 51 of the UN Charter until the UNSC has taken measures, here too it was understood, that States should have referral rights until such time as the UN Security Council decides to take measures.103 Regarding the role of the UNSC, the ILC was of the view that despite the possibility of abuse of the veto powers, attributing broad powers to the UNSC was consistent with the UN Charter, and therefore essential to “satisfy the conflicting demands of political realism and legal principle”, as noted by James Crawford.104 It was perceived that only a Court established under these terms would not require a substantial change of the UN Charter, no matter the way in which States would ultimately have chosen to establish the Court (as a UN principal organ (by amending the UN Charter); a UN subsidiary organ by resolution of its organs; or by treaty).105 In Rome the permanent members of the UN Security Council defended the initial ILC draft model as the one consistent with the present day world order
Commission on the work of its forty-fourth session, (4 May-24 July 1992), (A/47/10, Annex) (hereinafter 1992 Working Group Report), para. 4. 102 The ICC Prosecutor has actually recognized that Although any crime falling within the jurisdiction of the Court is a serious matter, the Rome Statute (articles 53(1) (b) 53 (2) (b) and 17(1) (d) ) clearly foresees and requires an additional consideration of ‘gravity’ whereby the Office must determine that a case is of sufficient gravity to justify further action by the Court
Bassiouni, supra note 19, ICC-Office of the Prosecutor, supra note 7, at 6. 103 See Mark S. Stein, ‘The Security Council, the International Criminal Court and the Crime of Aggression: How Exclusive is the Security Council Power to Determine Aggression? Indiana International & Comparative Law Review 16 (2001): 1, at 15; Stein notes that Under Article 51, the Security Council’s power to determine aggression is supreme, but not exclusive. Pending a determination by the Security Council, States can make their own determination of aggression. This arrangement makes it more plausible that the ICC can determine aggression, consistent with the Charter, especially given the Security Council’s authority to suspend ICC proceedings under Article 16 of the ICC Statute 104 See James Crawford, ‘The Work of the International Law Commission’, in The Rome Statute of the International Criminal Court: A Commentary, Vol. I, Antonio Cassese et al. (eds.) (Oxford: OUP, 2002), 23, at 25. 105 See ILC Draft Statute, Report of the ILC Appendix III.
conflicting interpretations of the icc statute
299
centered on the primary responsibility of the UN Security Council to maintain peace and security in the world. Thus during the Rome Conference the US argued that in view of the Security Council’s responsibilities under the Charter for restoring and maintaining international peace and security, recognition of its role in the Statute was vital to the proper functioning of the Court, in accordance with the obligations of Member States under the Charter
and that for this reason it had long supported the proposal in the original International law commission text requiring affirmative action by the Security Council before a complaint concerning a matter under consideration by the Council could be addressed by the Court.106
China also defended that it was essential that the Security Council be empowered to refer cases to the Court since otherwise it might have to establish a succession of ad hoc tribunals in order to discharge its mandate under the Charter. The Security Council should also have the power to determine whether acts of aggression had been committed. The operations of the Court should not impede the Council in carrying out its important responsibilities for maintaining peace and security.107
Similarly France was of the view that there must be consistency between the actions of the Court and the actions of the Security Council where there were situations endangering peace. The Statute should provide for the Security Council to be able to ask the Court to defer action in situations coming under Chapter VII of the Charter of the United Nations.108
However, despite efforts by the P5 and other States, a Statute with a limited role for the UK SC was adopted on 17 July 1998 providing for an independent Prosecutor with the ability to override the UNSC powers under Chapter VII.109 Some of the UNSC permanent Member States were promptly to denounce the treaty as violating the UN Charter. Russia voted in favor and signed it but is unwilling to ratify the treaty whereas China and the US voted against the text thus adopted. Marc Grossman the then Under Secretary of State for political affairs affirmed in 2002 that “the treaty approved in Rome dilutes the authority of the UN Security Council and departs from the system that the framers of the UN Charter envisioned”.110 106
Bassiouni, supra note 19, at 309, paras. 47, 48. Id., at 182, para. 85. 108 Id., at 154, para. 46. 109 See Section II.B.2. supra. 110 Marc Grossman, ‘American Foreign Policy and the International Criminal Court’ (May 6, 2002), <www.state.g0v/p/9949.htm> (accessed on 12 March 2005). 107
300
josé doria
ICC supporters not surprisingly contend that there was nothing abnormal in the establishment of the ICC conferred with such broad powers outside the UN framework. However most of the arguments put forward to substantiate their position are political rather than strictly legal, they essentially focus on issues of sovereign equality of States that would be violated by the ability that the UNSC permanent Member States alone would have had to use their veto powers to bar any prosecutions against their own citizens if the ILC model was adopted. Thus for example Nigeria argued that in setting up the Court, the international community was doubtless mindful of the many problems which had hindered such a move in the past; including the failure of the Security Council to act fairly and decisively in matters of global concern.
For these reasons Nigeria thought that the Security Council “should have no role whatsoever with regard to referral of matters to the Court”.111 Lesotho stressed that if an independent and effective Court was to be established, it was essential that the Prosecutor should have authority to initiate investigations ex-officio. If investigations and prosecutions could only be triggered by States and to some extent by the Security Council, the functioning of the Court would be dependent on the political motivations of those entities and as a result be severely hampered, because in practice States and the Security Council would be reluctant, or unable, to lodge complaints or refer situations to the Court.112
Switzerland considered that the Council should never serve as a filter to prevent matters from being referred to the Court. Nor should the Prosecutor be obliged to notify the Council whenever a State submitted a case to the Court. However, the Council might well wish the Court to defer consideration of a case for a certain period, but that period should not be too long and should not be used to remove or destroy evidence.113
Similarly Iraq argued that the Court must be independent from any political body. It was therefore unacceptable for the Security Council to have a role in the Court, bearing in mind the veto right given to certain States and the Council’s membership and method of voting.114
No one denies the correctness of this type of criticism, however, the point is that this is what the UN Charter ‘constitutional’ law is, and to argue that attaching the ICC to the UNSC would have exposed it to potential abuses does not make it any more legal overriding the UN Charter. This would be a political
111 112 113 114
Bassiouni, supra note 19, at 167, para. 76 (emphasis added). Id., at 167, para. 84. Id., at 308, para. 30. Id., at 165, para. 62.
conflicting interpretations of the icc statute
301
position not a legal one. In order to preserve the principle of legality, changes into the system if sought should be done in principle within the framework of the UN Charter, unless the new treaty adopted could be interpreted in a way which does not conflict with it. On the other side one should not lose sight of the fact that the wide margin of discretion accorded to the UNSC P5 was not intended to be used abusively (although this might have happened in the past more often than it is desirable), but responsibly in situations where one or more Member States found that taking action could actually potentially cause more harm than help.115 Imagine if in the situation of the post -apartheid South Africa, a zealous and independent Prosecutor wanted at all costs to prosecute the crime of Apartheid to replace the work of the truth and reconciliation commission. It is doubtful whether this would have brought more peace and security than caused more harm.116 For this reason some States including the UNSC P5 Members wished to have States and the Security Council with the sole authority to refer matters to the ICC,117 however if neither the Security Council nor any State party believed a situation was grave enough to be referred to the ICC that should speak “volumes against any potential need for the ICC Prosecutor to get involved”.118 (b) Position Two: Resist the Changes and Return to the Status Quo (US and Possibly Russia’s and China’s Approach) While there are a number of concerns for the US and other permanent members of the UN Security Council towards the ICC Model adopted, the most important among them are two: 115
In Rome the US argued that it remained unconvinced by the arguments put forward in favour of a proprio motu Prosecutor, and rejected the idea that the community of States was so lacking in moral and political courage that, when faced with an atrocity meriting the attention of the Court, not one State party would respond. It was wrong to argue that States’ unwillingness to invoke the Court’s jurisdiction was presumptively foreshadowed by the past reluctance of States to take on national prosecution of atrocities. On the contrary, the Court would provide an alternative to overcome the variety of legal, political, practical and resource difficulties which had made States reluctant, if not unable, to take on such prosecutions.
Bassiouni, supra note 19, at 172, para. 126. In the opinion of this author the self-referral cases so far before the ICC at least confirm these US arguments. 116 Of Course ICC supporters would say that Article 53 of the ICC Statute gives a margin of discretion to the Prosecutor to consider the interests of justice in lieu of prosecutions. But then this would be the same ‘politization’ that the ICC supporters wanted to avoid by limiting the Security Council’s role. 117 Kenya for example also argued that it saw no reason why the Prosecutor would require ex officio powers to trigger Court action. The twin triggers of States and the Security Council, subject to appropriate controls, were sufficient to cover all cases which would need to go before the Court.
Bassiouni, supra note 19, at 168, para. 92. 118 See Ambassador David J. Scheffer, ‘War Crimes Tribunals: The Record and the Prospects’, (Address at the Conference Convocation for the Washington College of Law Conference), American University International Law Review 13 (1998): 1389, at 1398.
302
josé doria
First one is that in their view the “treaty adopted in Rome dilutes the authority of the UN Security Council and departs from the system that the framers of the UN Charter envisioned”,119 and therefore is contrary to the UN Charter. And second is that the extension of the Court’s jurisdiction to nationals of non-party States without these States’ consent violates international law. The main issues around the first objection have been sufficiently addressed in the preceding chapters, and those related to the second objection will be addressed in the chapters below. Therefore we will discuss in this session the practical implications of the US and other P5 Members’ policy of resistance to the ICC Statute as adopted. UNSC P5 Member States think that the ICC Statute is an instrument raising serious constitutionality problems it is only normal that they oppose it and are willing to resist the changes and return to the status quo. The United States have been particularly persistent in their objections to the new changes. It is important therefore to review the prospects for their success, and second to find out why so many countries and scholars are steadfastly eager to prove that all the elements of the ICC Statutory regime are lawful120 despite prima facie evidence tending to show the contrary. What then makes countries and scholars so willing to ‘override’ international law in this case? The first argument would be that the opposition to the American demands is of course a reflection of the American stance towards the rest of the world. The American establishment is prone to view international law and international institutions as necessarily instruments for the promotion of the American hegemony in the world.121 One American commentator rightly criticized the US position towards the ICC stating that the Government viewed it as a matter of “everybody but us” justice122 – and rightly warning that this position would “invite the other nations of the world to
119 120
See supra note 110. Portugal stressed for example that it “endorsed the position of the German delegation with regard to States not parties to the Statute. The solution proposed would result in a more effective tribunal and was in harmony with international law”.
Bassiouni, supra note 19, at 164, para. 47 (emphasis added). 121 As one commentator notes: In the initial phases of the deliberations for the establishment of an international criminal court, the United States sought to safeguard its own nationals from prosecution in the ICC by insisting on Security Council authorization of all prosecutions in the ICC. That would enable the United States, as a Permanent Member of the Security Council, to veto decisions of the Council calling for the prosecution of an American national.
See Johan D. van der Vyver, ‘American Exceptionalism: Human Rights, International Criminal Justice, and National Self-righteousness’, Emory Law Journal 50 (2001): 775, at 797. 122 Tina Rosenberg, ‘Conference Convocation, War Crimes Tribunals: The Record and the Prospects’, American University International Law Review 13 (1998): 1406, at 1408.
conflicting interpretations of the icc statute
303
look at the Court as something that the United States has designed for its own purposes.”123 Beyond third world countries, this US attitude has started to irritate even closer American allies such as the UK, France, Germany, Canada and Australia.124 As a result in the battle for a UN Charter ‘friendly’ treaty the US was isolated and had to ally with ideological enemies of the past such as Russia and China, and the US was also among the seven only countries to vote against the final text of the ICC Statute adopted in Rome together with countries traditionally considered as ‘rogue’ ones, like Iraq, Iran, Sudan and Libya. Therefore despite the problems, the broad support that the ICC enjoys in the world is more an opposition to the American hegemonic policy rather than to existing law and institutions. It is no surprise that there are more scholars post factum trying to legitimize the new born institution rather than against it.125 Moreover States themselves are more inclined to find solutions 123
Id., at 1409–10. As Ambassador David Scheffer, Head of the American Delegation to the Rome Conference has frankly recognized 124
On the larger issue of overall protection for the U.S. military, however, we finally had to face the fact that we were barking up the wrong tree, and our military services were not being well-served with losing arguments. I spent many years seeking full immunity for our military forces and their civilian leadership in negotiations that quite frankly sometimes seemed the theater of the absurd. I was given nothing to offer—certainly not signature or ratification—in return for an absolutist carve-out that other governments, particularly our closest allies, found arrogant and hypocritical. I finally successfully lobbied my colleagues in Washington to permit me to offer a ‘good neighbor’ pledge towards the Court in return for full protection. Since the next administration could reverse that political pledge, however, it proved unconvincing.
David J. Scheffer, ‘A Negotiator’s Perspective on the International Criminal Court’, International Military Law Review 167 (2001): 1, at 8–9 (emphasis added). 125 See for example Luigi Condorelli and Santiago Villalpando, ‘Can the Security Council Extend the ICC’S Jurisdiction?’, in The Rome Statute of the International Criminal Court: A Commentary, Antonio Cassese et al. (eds.) (Oxford: OUP, 2002), 571, at 577; where they interestingly state that the extension of the Court’s jurisdiction would result from a resolution by virtue of which the Security Council request (or orders) that the Court take action beyond the terms of the Statute. Such a request per se would imply that the Court would contravene its own constitutive act, thus patently violating the principle of specialty. By virtue of this later rule, the scope of the powers of the ICC are a function of the common objectives that determined its creation and that are specified in its constitutive act. As a consequence the ICC could not be entitled to act beyond the terms of the Statute, except in case of an amendment or review in that sense approved under the procedure provided for in Part 13 or following a subsequent agreement between all the States Parties.
This author finds this and others arguments by the named commentators disputable. The lex specialis theory is not applicable to the present case. Article 103 of the UN Charter clearly provides that in cases when countries enter into treaties overriding the UN Charter, the later takes precedence. Limiting the powers of the UNSC by a treaty outside the UN Charter regime is clearly contrary to it, and therefore the UNSC can in principle use the fullest of its powers under Chapter VII, if compelling reasons warrant doing so. If the ICC Statutory provisions are an impediment, the UNSC could make use of the resources of the same institution on an ad hoc basis, conferring upon it jurisdiction under a separate UNSC resolution extending for example its temporal jurisdiction, abolishing the complementarity and immunity regime and obliging States of the world to cooperate unconditionally with the ICC for that specific purpose stated in
304
josé doria
post factum for the situation created rather than thinking of closing the new institution for alleged unconstitutionality of its Statute.126 This notwithstanding, the US has been adamant in its position against the ICC, and has even passed a law that sanctions countries that have adhered to the ICC, and refused to sign an Article 98(2) Agreement excluding Americans from its jurisdiction.127 The question is how successful the Americans could be. It is undisputable that with US support the Court would have been by far stronger and active in the international arena.128 However, in the opinion of this author the US objection has little chance of leading to the collapse of the ICC,129 or make the situation return to the previous status quo. The US has been unable to distinguish between concerns for preserving the existing system of maintenance of international peace, and those that relate to a more limited aspect of its own ill-perceived national interests.130 Instead of propounding that a security system which is controlled by the great powers has a chance of doing a better job for the entire world (and consequently for the Americans), the discourse has been that the old system has to be preserved because otherwise Americans will come under scrutiny of other countries.131 The veto powers that were conceded to the UNSC P5 to ensure an effective
the resolution, because the UNSC found it essential to maintain international peace and security. See also discussion infra in notes 149–153, and accompanying text. 126 See infra Section C.2.(c). 127 See The American Service Members’ Protection Act of 2002 (ASPA), Pub. L. No. 107–206, 116 Stat. 820. tit. II (2002). This law is also called the ‘The Hague Invasion Act’ for it gives the American President a right to free by force, Americans detained at the ICC Detention Unit in The Hague. 128 As Wedgwood correctly points out, if the US adhered to it, the ICC would have gained “the advantage of American diplomatic, military, and economic power to give teeth to the tribunal’s orders…” See Wedgwood, supra note 9, at 198. 129 John Bolton the then Under Secretary of State for Arms Control recognized once that making the ICC collapse was the key objective of the US anti-ICC policy: We should isolate and ignore the ICC. Specifically, I propose for United States policy–I have got a title for it… I call it the Three Noes: no financial support, directly or indirectly; no collaboration; and no further negotiations with other governments to improve the Statute. … This approach is likely to maximize the chances that the ICC will wither and collapse, which should be our objective.
John Bolton, ‘Is a UN International Criminal Court in the U.S. National Interest?’ (Hearing Before the Senate Subcomission on Foreign Relations: Statement of John R. Bolton, Under Secretary of State for Arms Control, Department of State), 105th Cong. 724 (23 July 1998). 130 Scheffer, supra note 124, at 9; frankly recognizing that: in the eight years of my deliberations in Washington on the International Criminal Court–beginning with the work of the International Law Commission in 1993 and 1994- I do not recall hearing any senior Defense Department official refer to the core purpose of the Court, namely to advance international justice and enforce the law of armed conflict. Every single discussion was dominated by how the Court would impact the United States military (emphasis added). 131
See John Bolton, ‘The Risks and Weaknesses of the International Criminal Court from America’s Perspective’, Law & Contemporary Problems 64 (2001): 167, at 171 (stating that the ICC was “a strategy to assert supremacy over the United States”). See also Senator Jesse Helms, ‘We Must Slay This Monster’, Financial Times, 31 July 1998, at 18 (stating that the ICC is “a threat to US national interests … and it is our responsibility to slay it before it grows to devour us”).
conflicting interpretations of the icc statute
305
maintenance of international peace and security are normally propounded by the US establishment instead as powers that would just shield US citizens from criminal responsibility.132 Under these circumstances, countries view the ICC as a deterrent against super powers that would tend to commit abuses during armed conflicts with no fear of repercussion. Against the weapon of ‘might’, the ‘little’ of this world prefer to shield behind the force of law which is ‘blind’.133 This is not to say that the American and other P5 objections would have no impact on the ICC. Certainly they would, at least on the way in which the ICC Statute will be interpreted and possibly amended (favoring some of the US and other P5 legitimate demands). But in contrast the P5 would also need to make more general concessions to other countries. Distending the conflicting interests of States around the ICC will most surely lead to possible reforms of the United Nations organs in particular the Security Council and the International Court of Justice. And the sooner they are carried out the better. Some of these alternative solutions are reviewed in the session below. (c) Position Three: Adapt Post Factum to the Events That Occurred and Change the Law As a result of the collapse of the Soviet Union and the end of the Cold War, a new multipolar world order is being redefined. Substantial changes occurred also in the type of threats that one can expect today, unanticipated in 1945 when the UN Charter was adopted. If in 1945, it was expected that threats to peace, would take the form of armed attacks of one state against another, today these threats are represented by brutal civil wars or terrorist acts in the name of religious ideologies. This type of new threats was clearly not envisioned by the Security system centered on the veto powers of the P5 members of the UNSC. As one commentator has noted the P5 veto powers, this institutionally disastrous by-product of the Cold War undermined a fundamental assumption on which the post-war system had been based. The Charter
132 John Bolton, the Under Secretary of State for Arms Control and International Security called the ICC a threat “on the independence and flexibility that America’s military forces need to defend U.S. national interests around the world”; Bolton, supra note 131, at 169. 133 Indeed the 1933 Montevideo Convention on Rights and Duties of States, whose provisions are considered today general customary international law and find expression in the UN Charter, provides in Article 4 that
States are juridically equal, enjoy the same rights, and have equal capacity in their exercise. The rights of each one do not depend upon the power which it possesses to assure its exercise, but upon the simple fact of its existence as a person under international law.
1933 Montevideo Convention on Rights and Duties of States, LNTS 165 (1933): 19.
306
josé doria had spelled out a bargain: States would renounce their freedom to use force unilaterally in return for a reliable systemic guarantee of collective security. Profligate recourse to the veto made that bargain untenable because States could no longer rely on the Security Council to actually authorize the collective measures necessary to protect against new threats to the peace or their legitimate national interests. Rather, when a threat arose, the State causing that circumstance could be expected to be immune to collective measures either because it, itself, had a veto, or because it was sheltered under the protective wing of one of the Council’s permanent members.134
Facing the challenges of our time necessarily requires that the institutions established to support the Cold War era status quo are little by little replaced by others that reflect the changes operated at the beginning of the 21st century. The far reaching progressive contents of the ICC Statute is indeed a claim by the State parties for changes in the constitutional foundations of the world order established in 1945 in San Francisco. By limiting the instances of UNSC involvement into the work of the ICC, the Statute indeed attacks the neuralgic centre of the San Francisco system of international security – the veto powers of the UNSC. The ICC Statute has thus at least the merit of showing that the nations of the world are willing to change a system of security that no longer serves their interests, and is viewed as an impediment to progress in one of the most vital areas to promote an effective collective security today.135 Countries understood that the new demands of the multipolar world required a permanent criminal institution, but one which would not be controlled by the UNSC. Countries argued in Rome that submitting the Court to the exclusive jurisdiction of the UNSC would be contrary not only to the idea of justice, but doing so was not required under the UN Charter either. The UNSC as we noted above, has exclusive jurisdiction over a situation only until such time as it is considering that situation. But once a veto decision is cast, countries are therefore free to have recourse to individual and collective measures of self-help.
134
Thomas Frank, ‘Collective Security and UN Reform: Between the Necessary and the Possible’, Chicago Journal of International Law 6 (2006): 597, at 601. 135 For example Mexico stressed during the Rome Conference that the Statute was not the right place to resolve differences of interpretation concerning the powers of the principal organs of the United Nations, particularly in view of the fact that the United Nations itself was now undergoing a process of far-reaching reform which could include changes in the role and powers of the Security Council. To link the Court solely to the Council, many of whose decisions were limited by the right of veto, was in Mexico’s view not only a grave political error but also a decision which was without foundation in law.
Bassiouni, supra note 19, at 399, para. 17.
conflicting interpretations of the icc statute
307
States assembled in Rome have resisted to US demands and empowered an independent Prosecutor with ex officio triggering rights. This fact of itself does not appear to violate the letter or the spirit of the UN Charter as such. It is not the general idea and purposes behind the ICC Statute that violate the UN Charter, but some of its specific provisions that purport to limit the reach of Chapter VII powers of the UNSC. To reconcile the conflicting interests between the UN and the ICC the following measures were suggested: a) change/reform the composition of the UNSC (and the veto powers) by making it a more democratic institution; b) accept the ICC as a new mechanism of maintenance of international peace and security alongside the UN, by signing a relationship agreement; c) enhance the role of the ICJ in disputes of legal character, by vesting it with the rights of a ‘Supreme Court’ of the world. (1) Change/Reform the Composition of the UNSC (and the Veto Powers) One commentator has noted that a contributing factor as to why the United States was unable to accomplish its goals at the Rome Conference was severe tension that exists in the United Nations between underdeveloped countries and the more developed ones, like the United States. In her view the underdeveloped countries are jealous of the Security Council’s expanding power in international security matters, and the recent failed attempt by middle ranking powers to expand the Security Council contributed to this tension. She was of the view that these factors “made it impossible for the United States to preserve an American veto over prosecution decisions through the use of the Security Council”.136 Indeed many view the veto system in the UNSC as an anachronism, that should be abolished because the justifications that there were in the past137 no longer apply nowadays, “because communism is dead, the world is more integrated, and the power structure is multipolar”, which makes political compromise more realistic.138 Abolishing the veto power would however take time to accomplish since the P5 are not even ready to admit more permanent members into the UNSC. A Panel of Eminent Persons established by the UN Secretary-General had proposed two options for the enlargement of the UNSC: (1) increasing the
136
Wedgwood, supra note 9, at 21. These justifications were: first, unanimity which was considered indispensable for peace; second, protection of respective national security interests of the UNSC permanent members; third protection of minority blocs from overbearing majority coalitions; and fourth prevention of rash Security Council decisions. See Keith L. Sellen, ‘The UN Security Council Veto in the New World Order’, Military Law Review 138 (1992): 187, at 235. 138 Id., at 235. 137
308
josé doria
number of both permanent and non-permanent members; or (2) enlarging the Council with new non-permanent members only.139 Based on this report Japan, Germany, Brazil and India, known as the Group of Four or G-4, have put forward an enlargement proposal that would give them, together with two African nations, permanent seats. P5 members were not very much enthusiastic about the idea. The US for example would support Japan’s bid for a permanent seat, but no more than that.140 While the US is “open to considering expansion proposals”, it thinks “the primary purpose of Security Council reform should be to make the Council more effective”, rather than more numerous.141 However, if by ‘effective’ it is meant considering abolishing the veto power altogether instead of expanding the Security Council with new veto holder Member States, that would have been welcome. Otherwise it is clear that in the new multipolar world, as a minimum the veto power needs to be diversified to take account of the geographic distribution of the members of the international community, if the UNSC wants to be trusted in matters like the ICC. As one commentator has rightly put it: Many members of the United Nations with little or no voice in the deliberations of the Council are probably somewhat surprised to find that it may order them to take major steps that they consider contrary to their national interest and that, moreover, are incongruent with expectations created by multilateral treaties to which they are parties. … The more the Council uses these very wide powers, especially in the absence of a broad consensus, the more urgent will be the calls for institutional reform.142
(2) Change the System of Maintenance of International Peace and Security and Accept the ICC as a New Key Player Article 2 of the ICC Statute provides that the Court should be put in relationship with the UN. Based on this, a relationship agreement was approved both by the ICC Assembly of State parties on 7 September 2004 and the UN General Assembly on 13 September 2004.143 On 4 October 2004 it was signed in New York by Mr. Philippe Kirsch, ICC President and Mr. Kofi Anan, Secretary General of the United Nations and entered into force.144 139 United Nations, A More Secure World: Our Shared Responsibility – Report of the SecretaryGeneral’s High-Level Panel on Threats, Challenges and Change, (2004) (A/59/565), at 65. 140 See Nicholas Burns, ‘On United Nations Reform’, (Testimony as Prepared Before the Senate Foreign Relations Committee (July 21, 2005) ) www.state.gOv/p/us/rm/2005/49900 .htm> (accessed on 12 September 2006). 141 Id. 142 Thomas Frank, ‘The Powers of Appreciation, who is the Ultimate Guardian of UN legality?’, AJIL 89 (1992): 519, at 523. 143 See ICC, ICC Press Release of 1 October 2004, <www.icc-cpi.int/press/pressreleases/ 46.html> (accessed on 22 October 2006). 144 See ICC, ‘ICC-UN Agreement Signed’, ICC Newsletter No. 2, October 2004, <www.icc-cpi .int/librarv/about/newsletter/files/ICC-NL2-200410_En.pdf> (accessed on 22 October 2004).
conflicting interpretations of the icc statute
309
The Agreement between the ICC and the UN regulates the working relationship between these two organizations, and establishes the legal foundation for cooperation within their respective mandates. The ICC is an independent international institution with a mandate – in support of the UN Charter—to address crimes that threaten international peace and security. The Relationship Agreement therefore reflects a delicate balance between independence and cooperation, respecting the autonomy and confidentiality of both institutions (emphasis added).145
The relationship agreement “bears in mind the principles and purposes of the UN Charter”, notes the important role assigned to the ICC in dealing with the most serious crimes of concern to the international community as a whole, as referred to in the Rome Statute, and which threaten the peace, security and the well-being of the world
and bears in mind that in accordance with the Rome Statute, the International Criminal Court is established as an independent permanent institution in relationship with the United Nations system.146
Further in Article 2 the Agreement lays down the principles under which the relationship is established. It provides that the relationship will be based on UN recognition of the existence of the Court as an independent judicial institution with legal personality, and on the Court’s recognition of the responsibilities of the UN under the Chapter. The parties also pledge themselves to respect each other’s mandate as stated in their respective constitutive instruments and as recognized under the relationship agreement.147 It is clear that the Agreement was intended to be a recognition by the United Nations of the mandate assigned to the ICC under the Rome Statute of dealing with “the most serious crimes of concern to the international community as a whole”, “which threaten the peace, security and the well-being of the world.”148 However the question is whether this agreement could be viewed as replacing, amending or legitimizing any amendments that the circumstances would require to make in the UN Charter, in other to avoid the ICC Statute being interpreted as overriding it. At least this would appear to be
145 Coalition for the International Criminal Court, ‘The Relationship Agreement between the ICC and the United Nations’, (accessed on 21 November 2006). 146 See 2004 Relationship Agreement between the ICC and the United Nations, (Preamble). 147 Id., Article 2 (3). 148 Id., Preamble (para. 3); Article 2(1, 3). Scholars recognized indeed that the Agreement will establish, in general terms, “the principles of cooperation between the Court and the UN for the purpose of maintaining and restoring international peace and security”. Condorelli and Villalpando, supra note 71, at 223.
310
josé doria
what some commentators would have hoped to see after the conclusion of the relationship agreement.149 This author finds it difficult to agree to such a proposition. It is true that the Agreement represents the recognition of international legal personality upon the ICC as a whole new institution. It is also true that the UN recognizes that the ICC has a mandate in dealing with serious crimes specially those that threaten international peace and security. However the UN, without making it general, appears to have reserved to itself the right to decide on a case by case basis, whether specific provisions of the ICC Statute purport to limit the powers of its organs in a manner which would be inconsistent with the UN Charter. The Agreement makes it explicit that the Court recognizes the UN responsibilities under the Charter and the purposes and principles that it enunciates.150 One of the essential principles enunciated in the Charter (Article 103) states that the Charter takes precedence over conflicting provisions in other treaties. Another principle (Article 2(6) ) provides that the UN shall ensure that even non-party States to the UN Charter comply with the UNSC decisions under Chapter VII, if required.151 Therefore, the Agreement does not validate any aspect of the ICC work that would purport to override the established mechanisms of maintenance of international peace and security centered on the leading role of the UNSC in situations endangering peace and security. This basically means that the UN will use the ICC to the extent that it does not run against the UN Charter. Whenever it runs, it will be ignored. The practice of the UNSC so far that
149
For example Condorelli and Villalpando have asserted that the the Agreement provided for in Article 2 will not only have the effect of tying the Court to the UN system, but it will also legally bind the UN to respect the spirit and the rules of the Statute. In other words, the Agreement will have the legal effect of establishing an obligation for the UN to abide by the provisions of the ICC Statute
and that the UN Charter does not entitle the Security Council to overlook its own obligations deriving from a conventional source (such as the Relationship Agreement concluded by the Organization with the Court) or the conditions established in an autonomous treaty (such as the Statute) regarding its intervention in the activity of a separate legal entity.
Condorelli and Villalpando, supra note 71, at 223; Condorelli and Villalpando, supra note 125, at 580 respectively. See also supra note 125, and accompanying criticism. 150 See Relationship Agreement, Article 2(2), Preamble (para. l). 151 For example on several occasions UNSC Chapter VII decisions have been directly addressed to and considered binding even on private organized armed groups fighting internal wars. See detailed discussion of UNSC sanctions imposed for example against a former Angolan rebel group-UNITA, in Jose Doria, ‘Angola-A Case Study in the Challenges of Achieving Peace and the Question of Amnesty or Prosecution of War Crimes in Mixed Armed Conflicts, Yearbook of International Humanitarian Law 5 (2005): 3.
conflicting interpretations of the icc statute
311
ignores the jurisdiction of the ICC152 or limits it153 in cases in which decisions are taken under Chapter VII is a clear illustration of this policy position. (3) Enhance the Role of the ICJ as Another Probable Actor in the New System As one commentator has once noted the proliferation of judicial institutions in the world dealing each with specific areas of international law, is reason enough to assign a leading role to the ICJ.154 Article 34 of the ICJ Statute that provides that only States may appear before the Court in litigation, means of course that the ICJ “cannot be a court of criminal jurisdiction, where individuals are found to be guilty or innocent”,155 still the ICJ has a potentially universal subject matter jurisdiction. It may adjudicate upon the interpretation of a treaty and any question of international law, including, of course, the laws of war and humanitarian law.156
Arguably the ICJ leading judicial role might find more justification in today’s world, where judicial institutions are being created with jurisdiction over potentially political issues that were normally the realm of political bodies such as the UNSC or the UN General Assembly.
152 For an example see UNSC Res. 1497 (1 August 2003), para. 7, authorizing the deployment of a multinational force of West-African States in Liberia, but deciding under Chapter VII that
current or former officials or personnel from a contributing State, which is not a party to the Rome Statute of the International Criminal Court, shall be subject to the exclusive jurisdiction of that contributing State for all alleged acts or omissions arising out of or related to the Multinational Force or United Nations stabilization force in Liberia, unless such exclusive jurisdiction has been expressly waived by that contributing State. 153
See UNSC Res. 1593 (31 March 2005), referring under Chapter VII the situation in Darfur (the Sudan) to the ICC, exempting from the Court’s jurisdiction 98(2) agreements (preamble (para. 4) and nationals of any operations in Darfur authorized either by the UNSC or African Union (para. 6). 154 See Jonathan I. Charney, ‘The Impact on the International Legal System of the Growth of International Courts and Tribunals’, New York University Journal of International Law and Policy 31 (1999): 697, at 706, noting that: The various international tribunals other than the ICJ,… do have their own agendas. They were formed to serve the interests of the states that established them within the treaty regime for which they were created. The allegiance to that treaty regime may become greater than the allegiance to the international legal system as a whole. These specialized tribunals present the risk that their own centrifugal forces will drive them in directions away from the core of international law. As a result, these specialized tribunals could develop greater variations in their determinations of general international law and damage the coherence of the international legal system. 155 See Rosalyn Higgins, ‘The Relationship between the International Criminal Court and the International Court of Justice’, in Reflections on the International Criminal Court, Herman A. M. von Hebel et al. (eds.) (The Hague: TMC Asser Press, 1999),163, at 163. 156 Id.
312
josé doria
This is particularly the case with the International Criminal Court, which Member States have entrusted with a mandate to prosecute and punish international crimes that threaten peace and security. As we already noted, the problem here is not so much with the appropriateness of prosecution of serious violations of human rights when they are deemed to have occurred, as it is with the powers of essentially a political nature that were conferred upon an independent Prosecutor to decide of the appropriateness of measuresother than prosecution—even where evidence shows that serious acts amounting to serious violations of human rights might have been committed.157 Arguably this is a discretion that the Prosecutor is entitled to use irrespective of what the UNSC position is. Under the terms of the ICC Statute, the UNSC can refer situations and can defer investigations and prosecutions.158 However it cannot control the ultimate outcome of its referral or deferral decisions. It cannot for example decide that in the particular circumstances of a situation in a country international prosecutions should not be carried out. And conversely its referral decisions can also be overridden by the Prosecutor based on the preliminary consent rules enshrined in the ICC Statute (complementarity). Likewise the Prosecutor can also overrule a UNSC deferral, after one year (renewable) irrespective of what the UNSC position is.159 The situation is even more troublesome if one considers that the ICC would have jurisdiction over the crimes of aggression (once it is defined), and that the Prosecutor could trigger of its free will the Court’s jurisdiction in these cases too. In potential situations of aggression the problem is not that the Prosecutor and the ICC could not be allowed in the ordinary process of doing justice, to find that a specific individual was guilty of the crime of aggression. The problem is that such determination necessarily implies a determination that a certain State behind that individual has committed the crime of aggression. Under the UN Charter the determination of whether a State was an aggressor is left normally to the UNSC (if the situation is ongoing) or to the ICJ in
157 Some States were disappointed with this outcome. The US argued for example that the draft Statute “was strong on paper but weak in reality”, particularly since it
burdened the Court with a propria motu Prosecutor, an institutional weakness which would result in the Court being overwhelmed with complaints and embroiled in controversy
Bassiouni, supra note 19, at 399, para. 20. 158 See ICC Statute (Articles 13(b), 16). 159 The argument of certain ICC proponents to the effect that the limited duration of the suspensive effect of Article 16, would “necessary guarantee that the process will be managed with restraint” does not make it anymore legal. Berman, supra note 54, at 178. In this author’s view the essential thing is that it should not be up to States assembled outside the UN Charter framework to impose limitations on the work of the UNSC in contravention of the Charter.
conflicting interpretations of the icc statute
313
any circumstances. Moreover that determination necessarily entails State responsibility.160 In particular the State victim of aggression is entitled to seek compensation for the unlawful acts of war, and other sanctions may be imposed on the aggressor State.161 The ICC has no jurisdiction over legal entities but only over individuals.162 It is clear that allowing the ICC to determine whether anyone was guilty of the crime of aggression would embroil it in serious political controversy and would have far reaching implications contrary to the provisions of its Statute, since it would mean in practice passing too a judgment on the acts of States.163 Conversely, making the jurisdiction of the Court dependent upon a finding of aggression by the UNSC could preempt the work of the Court in a way which is contrary to fundamental principles of due process like the presumption of innocence until guilty is proved, impartiality and so on. However letting the Court to work alongside and independently from the UNSC would also risk having contradictory decisions on one and the same question (imagine the UNSC deciding that there was no aggression and the ICC finding a specific individual (the Prime Minister) of that country guilty of the crime of aggression). The end result would be justice denied. Under these circumstances, the ICJ could represent a viable alternative. There are certainly still some statutory limitations to having the ICJ to play a major role in the work of an impartial ICC, among which the fact that the ICC does not allow States or individuals direct access to the ICJ,164 and the ICJ would not accept direct challenges by individuals.165
160 Military and Paramilitary Activities in and against Nicaragua (Nicaragua v. United States of America), Judgement of 27 June 1986, 1986 ICJ Rep.14, at 146–7; see also Igor Blischenko and Jose Doria, Precedents in International Public and Private Law, (Moscow, 1999), at 345. In Rome the US rightly argued that
only the Security Council could take the forceful measures that were necessary if aggression was to be addressed and remedied. That gave rise to political and other problems that had made it difficult to find consensus in the past, yet the Security Council had an essential role to play.
Bassiouni, supra note 19, at 135, para. 98. 161 See for an example UNSC Res. 687 (8 April 1991), paras. 16–29, establishing a compensation fund, imposing other sanctions on Iraq and reaffirming under Chapter VII that Iraq was liable under international law for any direct loss, damage, including environmental damage and the depletion of natural resources, or injury to foreign Governments, nationals and corporations, as a result of Iraq’s unlawful invasion and occupation of Kuwait. 162 163
ICC Statute, Article 25(1). During the Rome Conference, France justly warned indeed that it would be in the interests of the Court itself to be able to rely on a prior determination by the Security Council to avoid having to pass judgment not only on persons but also on States.
Bassiouni, supra note 19, at 137, para. 115. 164 ICC Statute (Article 119(2) ). 165 Statute of the International Court of Justice (Article 34(1) ).
314
josé doria
But these are not insurmountable difficulties. With sufficient political will States can waive the statutory limitations in both institutions and give the ICJ an enhanced role in a new system of maintenance of international peace and security more consonant with present day needs. III. The ICC as a Novel Derivative Subject of International Law with Ceded Jurisdiction or When May International Organizations Exercise Rights Ceded by States Affecting Non-Member States International Organizations are secondary derivative subjects of international law. They have normally only such rights as they were conferred upon them by those entities that have created them (be they States or other organizations). This normally raises the question of whether those entities that have created such organizations were entitled to do so. For example the ability of the UNSC to establish judicial organs was challenged in the Tadic case.166 The accused argued that the establishment of the Tribunal was ultra vires, since the UN Charter does not confer upon the UNSC the right to establish such organs. The question concerned the scope of measures allowed to the Security Council under Article 41. The Appeals Chamber came to the conclusion that under the UN Charter, the UNSC was entitled to establish the tribunals because, the list of measures the UNSC is entitled to take under Article 41 was not exhaustive, as the list was preceded by the terms “may include”, which the ICTY Appeals Chamber understood as not excluding any other measures that the UNSC deems appropriate.167 This issue could potentially be relevant also in situations where an organization is created by States. For example it is accepted that the ability of States to establish international organizations is limited by the existence of norms jus cogens. States cannot purport to establish a lawful organization whose purpose would be contrary to a jus cogens norm of international law. Article 53 of the Vienna Convention on the Law of Treaties (VCLT) provides that a jus cogens norm is one accepted and recognized by the international community of States as a whole as a norm from which no derogation is permitted and which can be modified only by a subsequent norm of general international law having the same character.
Article 53 of the VCLT also provides that a treaty would be “void if, at the time of its conclusion, it conflicts with a peremptory norm of general international law”. Norms prohibiting for example slavery, racial discrimination, genocide,
166 167
Tadic case, paras. 8, 37, and 38. Tadic case, paras. 35, 37, and 38.
conflicting interpretations of the icc statute
315
or the unlawful use of force in international relations have been recognized as norms jus cogens.168 States cannot purport to establish a lawful organization whose aims would be to promote for example slavery, acts of aggression, or genocide. Such a treaty and the organization created with these aims would be considered void ab initio. Obviously the situation would appear even more illegal if these States decided that the organization would consider as its members ipso facto any other State of the world that has not specifically consented to it. Does it make the situation any less illegal if the organization is not created to promote acts violating jus cogens, but instead to protect norms jus cogens. The fact that the organization intends to pursue a good cause, does it justify conferring to it a binding force reaching any State of world with or without its consent? My preliminary submission would be that since international organizations are consensual organs of States, they should bind obviously only those States that have consented to it either implicitly or explicitly. Articles 34 to 38 of the Vienna Convention on the Law of Treaties (VCLT) regulate the types of relationship that can be established between a treaty and non-Member States. In particular, under Article 34, “a treaty does not create either obligations or rights for a third State without its consent”. Article 38 further provides that “nothing precludes a rule set forth in a treaty from becoming binding upon a third State as a customary rule of international law, recognized as such”. With these preliminary observations in mind we will review the provisions of the ICC Statute regarding third States. The problem of whether the ICC can lawfully exercise jurisdiction over nationals of third States without those States’ consent is at the heart of much of the opposition to it by the three permanent members of the UNSC, but also other States that have so far not yet joined the ICC. It is central to the original jurisdictional mechanism adopted in the ICC Statute. Under its Articles 12 and 13 States ratifying the ICC Statute are bound by it. However in addition under the formulation adopted non-State parties can also be unwillingly forced to accept the jurisdiction of the Court. This happens in two situations: a. when a situation involving nationals of nonparty States is referred to the Court by the UNSC, and b. when a crime is committed by a national of a nonparty State on the territory of a State party.
168 See Barcelona Traction, Light & Power Co. (Belgium v. Spain), Second Phase, Judgment of 5 February 1970, 1970 ICJ Rep. 3, para. 33.
316
josé doria
Both situations raise legitimate concerns about the legal sources of this binding obligation upon non-party States. A. The UN Charter as the Primary Legal Basis of UNSC Referrals Binding Non-Party States to the Court Article 13(b) of the ICC Statute provides that the Court may exercise its jurisdiction with respect to a crime referred to in Article 5 “if…a situation in which one or more of such crimes appears to have been committed is referred to the Prosecutor by the Security Council.” If the situation referred relates to a Member State, then everything should be legally appropriate, since the Member State in question has conceded to be bound to this referral mechanism. Does the situation change, if the State in question is not a Member State of the ICC? Virtually no one disputes that the situation does not change,169 but depends obviously on the basis put forward to justify the lawfulness of the UNSC actions. Provisions of the Vienna Convention on the Law of Treaties allow us unambiguously to clarify this issue. Article 31 of VCLT clearly states that a treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose.
Further 31(3)(c) of the Vienna Convention also states that in interpreting the treaty other factors could be taken into account such as for example any relevant rules of international law applicable in the relations between the parties. Article 32 also allows recourse to supplementary means of interpretation if the interpretation under Article 31 leads to a result which is manifestly absurd or unreasonable. The ICC Statute in line with these provisions of the VCLT provides in its Article 21 that apart the Statute itself, in interpreting the law, the Court shall 169 Although not supported by the remaining States, it is worth noting that at least three States in Rome opposed the legitimacy of having the UNSC to compel non party States to the Court’s jurisdiction. India argued that
As the Council would almost certainly include non-party States among its members, that provision would confer on such States the power to compel both States parties and other non-party States to submit to the Court’s jurisdiction, in violation of the law of treaties, as well as conferring on the Council a role never envisaged for it by the Charter of the United Nations.
Sudan was of the view that the proposal allowing the Security Council to submit complaints to the Prosecutor or refer matters directly to the Court, without the consent of the State concerned being needed, “was dangerous, it was important that the Court should not be weakened”. Finally Syria also argued that any proposed triggering option that would give the Security Council the right to trigger action even with respect to States which were not Parties to the Statute would be a violation of the Vienna Convention on the Law of Treaties.
Bassiouni, supra note 19, at 345, para. 34, at 166, para. 69, and at 147, para. 55 respectively.
conflicting interpretations of the icc statute
317
have recourse to other applicable treaties and the principles and rules of international law. It is clear from the ICC Statute that UNSC referral decisions are based not only on the possibility conferred to it under Article 13(b) of the ICC Statute, but principally on the UN Charter itself. The travaux preparatories show that most States assembled in Rome shared the view that in relation to non-party States the UN Charter would be the primary binding basis of jurisdiction. For example Chile correctly argued that the case of a State not party to the Statue in which heinous crimes had been committed and which had not accepted the Court’s jurisdiction should be discussed in relation to the role of the Security Council. Under Chapter VII of the Charter of the UN, the Security Council could certainly submit a situation involving a State or its nationals to the Court.170
Similarly Russia was unable to agree that an international treaty could create obligations for third parties which were not parties to it. The only way the Court could exercise jurisdiction over a non-party was by means of a Security Council decision.171
The US affirmed that the universal jurisdiction proposal for the Court would represent an extraordinary principle, in conflict with certain fundamental principles of international law, and would undermine the Statute generally. The proposals by Germany and the Republic of Korea would have the effect of applying a treaty to a State without that State’s consent, and in the absence of any action by the Security Council under Chapter VII of the Charter of the United Nations. Even if a State was not a party, the Court would have jurisdiction to judge its official acts and imprison even its head of State. Such a situation could not be justified on the basis of existing law and the US objected to it in principle. An international treaty could not impose itself in that manner on non-party States; the only solution was to reach out to other States through the Charter and the powers of the Security Council that had been created by States under that separate regime.172
Commentators have voiced closer views. For example Sir Franklin Bergman, Legal Adviser of the British Foreign Office, has persuasively stated that it was not in actual fact correct to say that Article 13 ‘gives the Security Council the key’, in the sense of purporting to convey any new power on the Council. It was widely understood from an early stage that this would have been neither legally nor politically possible. Instead what the provision does—correctly—is to stipulate that the powers of the Court are brought into play when the Security Council makes reference to it, in other words to empower the Court, not the Council.173
170 171 172 173
Bassiouni, supra note 19, at 149, para. 5. Id., at 163, para. 40. Id., at 161, para. 23. Bergman, supra note 54, at 174.
318
josé doria
The UN Charter is according to Article 30 of the VCLT the only legal instrument binding both non-party States and State parties to such UNSC referrals to the Court under Chapter VII. This proposition is also supported by provisions of Articles 34, 35 and 38 of the VCLT, since the UN Charter is the only agreement to which both non-party States and State parties to the ICC Statute have consented to. Additionally the problem of the forum convenient chosen to enforce the UNSC decision to initiate criminal proceedings in this case (i.e. the ICC to which a State has not adhered) is also solved by recourse to provisions of the UN Charter, mainly Articles 24, 25, 41, 43 and 48. According to Article 24(1) of the UN Charter, Member states confer on the Security Council primary responsibility for the maintenance of international peace and security, and agree that in carrying out its duties under this responsibility the Security Council acts on their behalf.
Under Article 25, Members States “agree to accept and carry out the decisions of the Security Council.” Regarding the powers to institute jurisdictions to punish serious violations of international law, the so called Chapter VII powers of the UNSC are described in particular in Articles 39, 41 and 48. Article 39 provides that the Security Council shall determine the existence of any threat to the peace, breach of the peace, or act of aggression and shall make recommendations, or decide what measures shall be taken in accordance with Articles 41 and 42, to maintain or restore international peace and security.
Article 41 provides for non military measures whereas Article 42 for the military ones. Finally under Article 48(1) of Chapter VII, the Security Council may determine whether the action required to carry out its decisions is to be taken by all or only some of the UN Member States. Therefore the primary legal basis of the jurisdiction of the Court in situations referred to it by the UNSC is the UN Charter.174 It was also based on this that the UNSC did establish the ad hoc Criminal Tribunals whose decisions are binding on all States of the world.
174 Russia reacting to those States that defended the extreme view that non-party States could not be obliged to the Court’s jurisdiction even by a UNSC resolution under Chapter VII, recalled that the
Security Council was responsible for maintaining international peace and security and that the question of whether a member was or was not a party to the Statute or to any other treaty was not of vital importance, since Article 103 of the Charter would prevail. There would thus be no violation of treaty law.
Bassiouni, supra note 19, at 314–5, para. 116.
conflicting interpretations of the icc statute
319
B. Legitimacy of the Transfer of Territorial Jurisdiction to an International Institution Without the Consent of Affected Third States According to Article 12, the ICC can exercise its jurisdiction if the State of the territory where the crime was committed or175 the State of nationality of the accused is a party to the Rome Statute or has accepted the jurisdiction of the ICC with respect to the crime in question. Therefore under this formulation, persons accused of committing the relevant crimes may be subject to prosecution even if the State of their nationality is not a party to the Rome Statute. This, as we noted, is the most controversial formulation in the ICC Statute. All turns around the right that States have or not to confer to an international organ rights that they enjoy in their mutual relations without the consent of the State affected. While the US176 and some other States177 have argued that based on Articles 34, 35 and 38 of the VCLT, it would be illegal to pass onto an international body rights that States as sovereigns enjoy in their territory without the consent of all involved, ICC supporters say it is possible. However despite claims that the US position is ill-founded they have so far not presented convincing arguments to support their allegations. Often ICC supporters rely on political considerations.178 But still this is not the law. We will critically review below the traditional allegations advanced by ICC supporters: 175 During the final stages of the Rome Conference, the US Delegation attempted to change the ‘or’ by ‘and’ (which would have made nationality states consent mandatory), by arguing that
Article 12 on pre-conditions to the exercise of jurisdiction by both the State on whose territory the crime had occurred and the State of nationality of the accused would be required.
However this amendment was rejected by a “no action” vote proposed by Norway. See Bassiouni, supra note 19, at 400, para. 22; and paras. 24, 31 respectively (emphasis added). 176 In Rome the US argued that it supported the Geneva Conventions of 1949 and acknowledged the importance of universality of jurisdiction in its proper context for the effective vindication of international law. However, the proposed Statute took the principle of universal jurisdiction far outside any acceptable context. Moreover, the attempt to impose the jurisdiction of the Court on States which did not become parties to the Statute would violate an elementary rule set out in the Vienna Convention on the Law of Treaties.
Bassiouni, supra note 19, at 400, para. 21. See also supra note 172. 177 Mexico for example argued that for the court to exercise its jurisdiction, it should be necessary for the State where the accused was and the State of nationality of the accused to have given their consent.
Similarly Iran wished to stress the fundamental importance of the principle of State consent and stated that “the consent of the custodial State, the territorial State and State of nationality should be required”. Bassiouni, supra note 19, at 156, para. 64, and at 157, para. 69 respectively (emphasis added). 178 For example New Zealand argued that the suggestion voiced by some States that it would be necessary for the territorial State and possibly the State of nationality to consent may create a problem by enabling a State whose national had committed serious crimes in another State to withhold its consent and shield the accused. That would not contribute to enhancing peace and security.
320
josé doria
a. First argument: If sovereign States have the right to try nationals of nonparty States in their own territory, then they can also cede such a right to an international institution179 consistent with the lotus principle.180 This position is not convincing, the sovereign rights that States have to try foreign nationals who have committed crimes are reciprocally exercised by them. The same right that a State A has to try nationals of State B is equal to the right that State B has to try nationals of State A. This is derived from the fundamental principle of sovereign equality of States. This is clearly different from ceding such a right to an international organ without the consent of the national State.181 International organizations although
Likewise Switzerland considered that the requirement that the State of nationality of the accused or suspect must accept the jurisdiction of the Court would have the consequence that nationals of non-party States would be outside the jurisdiction of the Court regardless of their whereabouts, whereas currently they were subject to the jurisdiction of States other than their own as soon as they crossed their national frontiers.
Bassiouni, supra note 19, at 159, para.7 and at 343, para.20, respectively. 179 See Gerhard Hafner et al., ‘A Response to the American View as Presented by Ruth Wedgwood’, EJIL 10 (1999): 108, at 117, noting that: “Nationality and territoriality are each an individual and sufficient basis for jurisdiction” and “it is within the sovereign power of a State to allow an international body to exercise jurisdiction in the same way in which that State may exercise jurisdiction”.; See also Hans-Peter Kaul, ‘Precondition to the Exercise of Jurisdiction’, in The Rome Statute of the International Criminal Court: A Commentary, Antonio Cassese et al. (eds.) (Oxford: OUP, 2002), 583, at 587, stating that: the universality approach starts from the assumption that, under current international law, all States may exercise universal jurisdiction over these core crimes. It combines this assumption with the very simple idea that States must be entitled to do collectively what they have the power to do individually. Therefore, States may agree to confer this individual power on a judicial entity they have established and sustain together and which acts on their behalf.
In this author’s opinion, these statements would only be correct if the new entity’s authority engaged only Member States. Those who advocate this position forget that universal jurisdiction for the core crimes is also based on presumed or express consent of all States. Hence transposing this to a new non-sovereign entity can only be done with the same kind of consent of all those States otherwise, its effect would necessarily be restricted to the States that accepted it only. 180 Id., at 592, Hans-Peter Kaul notes that in general, there is no rule in international law which would prohibit such a transfer of a State’s rights to a new entity, under international law, which this State is co-establishing as a Contracting Party and will maintain through its contributions as a State Party.
Kaul’s position is based on the ‘Lotus principle’ articulated in 1927 in the case SS Lotus by the Permanent Court of International Justice which stated that “Restrictions upon the independence of States cannot… be presumed” and that international law leaves to States “a wide measure of discretion which is only limited in certain cases by prohibitive rules”. S.S. Lotus (France v. Turkey), 1927 PCIJ (ser. A) No.10, at 18. In this author’s opinion this statement is correct only as far as it pertains to member States of the new entity, not non-party States. 181 Even the practice of interstate relations shows that States are normally allowed to disregard a request for the exercise of ceded jurisdiction if the person concerned is not a national or is not a person with at least permanent resident status in that country. For example Article 11 of 1972 European Conventions on the transfer of proceedings in criminal matters, states that: The requested State may not refuse acceptance of the request in whole or in part, except in any one or more of the following cases: c) if the suspected person is not a national of the requested State and was not ordinarily resident in the territory of that State at the time of the offence. (Emphasis added).
conflicting interpretations of the icc statute
321
independent subjects of international law, are not sovereign entities and therefore cannot work on the basis of reciprocity with a sovereign State, i.e. the right that an international organization would have to try nationals of non-party States cannot be compensated by a similar right that non-party States would have to try nationals of the international organization. The only way by which a non-party State would have to respond on the basis of reciprocity would be to create with some other countries another international organization B and confer to it the same type of jurisdiction over non-party States that are members of the international organization A. This logic would result in chaos in international relations.182 b. Second argument: The Nuremberg, Tokyo and the ad hoc Tribunals are precedents that legitimize the principle of ceded jurisdiction.183 This position is also wrong. The Nuremberg and Tokyo Tribunals were covered by provisions of Articles 53 and 107 of the UN Charter that legitimize any action against former aggressor States during World War II. In contrast here the ICC would have jurisdiction for future cases, not over aggressor States of the WWII period. The ad hoc Tribunals cannot serve as an example either. They were established under Chapter VII of the UN Charter, and countries of the world have accepted to be bound by coercive measures taken by the UNSC under Chapter VII. c. Third argument: The crimes for which jurisdiction was ceded to the Court are all crimes for which universal jurisdiction applies, i.e. crimes of concern to the international community as a whole or jus cogens prohibitions entailing erga omnes obligations for all. Therefore the ICC should have jurisdiction over such crimes committed by nationals of non-party States.184 182
Ambassador David Scheffer once noted: Official actions of a non-party State should not be subject to the Court’s jurisdiction if that country does not join the treaty, except by means of Security Council action under the UN Charter. Otherwise, the ratification procedure would be meaningless for governments. In fact, under such a theory, two governments could join together to create a criminal court and purport to extend its jurisdiction over everyone everywhere in the world. There will necessarily be cases where the international court cannot and should not have jurisdiction unless the Security Council decides otherwise.
See David Scheffer, ‘Is a UN International Criminal Court in the U.S. National Interest?’ (Hearing Before the Senate Subcomission on Foreign Relations: Statement of David. J. Scheffer, Ambassador-at-Large for War Crimes Issues, Department of State), 105th Cong. 1314 (23 May 1998). 183 Scharf, supra note 9, at 103–9. 184 Seth Harris, ‘The United States and the International Criminal Court: Legal Potential for Non-party States’, Hawaii Law Review 23 (2000): 277, at 303, stating that: Normally, sovereign States can pass their authority to prosecute these crimes against the international community to an international court. The erga omnes obligations imputed from the nature of the subject matter as jus cogens entitles prosecution of individuals by an international court, if the domestic court refuses to prosecute.
See also Scharf, supra note 9, at 76, stating that …where the territorial state gives its consent…in addition to the principle of territoriality, the ICC has a legitimate interest on the basis of the universal nature of the crimes to prosecute the nationals
322
josé doria
This position also fails. First if this was a valid argument the ICC should have been given universal jurisdiction as actually Germany did propose.185 However this proposal was rejected by the same States.186 States clearly distinguished the universal jurisdiction that attaches to such crimes, and the jurisdiction that the Court – a consensual organ of States -should have.187 Second jus cogens, erga omnes even if this was the case would be the norms themselves, but not the jurisdiction of the ICC. Jurisdiction is not jus cogens, it still remains only the act of conferring a right to adjudicate a case or situation. This necessarily raises the question of the forum convenient to do so. And all procedural rules still would apply. For example in the recent Congo v. Rwanda case188 on genocide issues, Congo attempted to have recognized to the ICJ a right to adjudicate the case despite absence of consent by Rwanda to the Court’s jurisdiction, justifying this with the fact that it was a question of a violation of jus cogens and therefore States do not need to consent to be bound by specific jurisdictions in theses cases.189
of non-party states. In this limited context, the jurisdiction of the ICC can be deemed to be based concurrently on the universal and territorial bases of jurisdiction. 185
In Rome Germany justified its proposal to confer universal jurisdiction to the Court by stating the following: The Court would be acting on behalf of the international community as a whole. Since the contracting parties to the Statute could individually exercise universal jurisdiction for the core crimes, they could also, by ratifying the Statute, vest the Court with a similar power to exercise such universal criminal jurisdiction on their behalf, though only of course with regard to the core crimes.
Bassiouni, supra note 19, at 145, para. 48. 186 France argued for example that the international community was perhaps not yet ready for the idea of universal jurisdiction as put forward by Germany. There was no obligation on State not parties to the Statute to cooperate. Generally speaking the State on whose territory the crime had been committed and the State of nationality of the accused or the custodial State would have to be parties to the Statute, or have accepted the competence of the Court for it to be in a position to exercise its jurisdiction.
Bassiouni, supra note 19, at 154, para. 48. 187 For example Israel rejecting the German proposal held that while States had universal jurisdiction in respect of the core crimes, the Court was a judicial organ, exercising its jurisdiction on a consensual basis, subject to the conditions and limitations contained in the Statute.
Bassiouni, supra note 19, at 148, para. 66. 188 See Armed Activities on the Territory of the Congo (New Application: 2002) (Democratic Republic of Congo v. Rwanda), Jurisdiction and Admissibility, Judgement of 3 February, 2006, 2006 ICJ Rep. 6 [hereinafter Congo v. Rwanda case]. 189 Congo asserted among others, 3 bases of jurisdiction which involved claimed breaches of norms jus cogens. First, Congo asserted that by its acts of aggression and serious violations of human rights Rwanda breached its obligations under the 1948 Genocide Convention and contended that Rwanda’s reservation withholding jurisdiction from the ICJ was invalid since it intended to prevent the Court from protecting the norm jus cogens prohibiting genocide. Second, Congo asserted that the Rwanda’s reservation to the Racial Discrimination Convention was invalid since it codifies norms jus cogens. Third, Congo relied on Article 66 of the Vienna Convention on the Law of Treaties, to contend that the Court had jurisdiction in relation to any violation of norms jus cogens. See Congo v. Rwanda case, para. 15.
conflicting interpretations of the icc statute
323
The ICJ rejected this reading and stated that one thing is the prohibition itself that is jus cogens, and another thing is the jurisdiction to entertain such a case that can be exercised in different ways.190 The ICJ was only but one of such ways, and consent to it as the forum convenient need to be taken from all the interested parties.191 Additionally provisions of the ICC Statute denying jurisdiction in certain cases also justify a position according to which the ‘jus cogeness’ of a violation does not justify compulsorily submitting a country to the jurisdiction of a forum not consented to. For example, provisions under Article 98 on immunities and previous agreements that exempt nationals of third States from the ICC’s jurisdiction clearly show that consent of third States is required. Additionally was it the case States parties would not have been allowed to opt-out of the Court’s jurisdiction under Articles 121, 123, and 124. All this shows the consensual basis of the ICC. d. Fourth argument: the ICC Statute imposes obligations not on States but on national individuals.192
190
The Court first held that the erga omnes character of a norm and the rule of consent to jurisdiction are two different things, and that the mere fact that rights and obligations erga omnes may be at issue in a dispute would not give the Court jurisdiction to entertain that dispute.
Applying this dictum to the case at issue the Court held: Rwanda’s reservation to Article IX of the Genocide Convention bears on the jurisdiction of the Court, and does not affect substantive obligations relating to acts of genocide themselves under that Convention. In the circumstances of the present case, the Court cannot conclude that the reservation of Rwanda in question, which is meant to exclude a particular method of settling a dispute relating to the interpretation, application or fulfillment of the Convention, is to be regarded as being incompatible with the object and purpose of the Convention
See Congo v. Rwanda case, paras. 64, 67 (emphasis added). 191 The ICJ held that it “has jurisdiction in respect of States only to the extent that they have consented thereto” and not because of the jus cogens character of the norms involved. See Congo v. Rwanda case, para. 65. 192 van der Vyver, supra note 121, at 818, stating that: the ICC Statute provides a forum for the prosecution of individuals suspected of having committed acts of genocide, crimes against humanity or war crimes. It does not place any obligations upon nonparty states unless such states have consented to cooperate with the ICC.
See also John Murphy, ‘The Quivering Gulliver: US Views on a Permanent International Criminal Court’, The International Lawyer 34 (2000): 45, at 65, stating that: for the court to exercise jurisdiction over the United States, the consent of the U.S. Government would clearly be required. It is an entirely different issue, however, as to whether U.S. consent is required for the Court to exercise jurisdiction over its national.
See also Kaul, supra note 179, at 609 noting that: the jurisdiction of the Court over nationals of non-States Parties to the extent implied in the universal approach should not be seen as imposing an obligation on a non-State Party (or infringing upon a right of such State), but rather as responding to (existing) obligations of the individual under international law.
The last remarks disregard two important facts: that individuals are subject of international law to the extent agreed to by their respective countries, and that there is a difference between the
324
josé doria
This position also fails since it makes useless provisions in the Statute about immunities, and treaties under Article 98; surrender rules under Article 90, and opt-out provisions under Articles 121, 123 and 124. These provisions affect individuals, but require positive action from the State of nationality. Moreover the provisions on complementarity (Article 17) arguably also oblige non-party States to be involved in the work of the Court if and when they wanted to avoid the exercise of jurisdiction over their nationals. So with or without their free will nonparty States are necessarily constrained by the existence of the ICC. These are all provisions arguably protecting individuals, not States, but which necessarily involve States at the end of the day.193 On the other side, the notion of State is always a combination of at least three elements: territory, population and government. To disassociate the population from the notion of State would be to deprive the notion of State of its contents contrary to international law.194 e. Fifth argument: Recent treaties on combating terrorism also provide for jurisdiction in relation to citizens of non-party States, therefore the ICC is not novel.195 This position is not convincing either. The main point is that these treaties do not create international organs to which competence is ceded to try
obligations as such and the mechanisms chosen to enforce these obligations. States have agreed to confer on themselves on a mutual basis the right to enforce the obligations against individuals. A decision to confer in addition such a right to a third organ is one that can only bind the States involved, since it is not absolutely necessary to enforce the obligation. 193 As India rightly pointed out during the Rome Conference, the success of the Court would depend in great measure on cooperation among States aimed at punishing heinous crimes of international concern. While the Court’s jurisdiction would be individual, the nature of the crimes was such that the reputation of Governments would inevitably come under scrutiny.
Bassiouni, supra note 19, at 169, para. 104 (emphasis added). 194 The 1933 Montevideo Convention on the Rights and Duties of States in Article 1 defines indeed that the State as a person of international law should possess the following qualifications: a) a permanent population, b) a defined territory, c) government, d) capacity to enter into relations with other States. 195 Guenady M. Danilenko, ‘The Statute of the International Criminal Court and Third States’, Michigan Journal of International Law 21 (2000): 445, at 461 et seq; citing anti-terrorism Conventions and stating that:
there is still another perfectly valid basis for the Rome Statute’s jurisdictional reach. International law recognizes that State parties to an international treaty may exercise extraterritorial treaty-based jurisdiction over crimes of an international character defined by that treaty.
See also Scharf, supra note 9, at 100 et seq stating that: It has been suggested in this regard that a distinction should be drawn between treaties codifying customary international law and treaties legislating new international crimes…But even if some of these crimes were deemed not to reflect customary international law, there is precedent for State parties to exercise universal jurisdiction created solely by treaty over the nationals of non-party States.
conflicting interpretations of the icc statute
325
nationals of third States. States have conferred to each other rights and obligations to punish terrorist acts that they can exercise towards foreigners on a reciprocal basis and according to the Lotus principle (i.e. nothing prevents a State from exercising non-prohibited rights in its own territory). However, the suggestion to include these treaties under the jurisdiction of the ICC was rejected in Rome.196 Conclusion: Unless provisions of the ICC Statute are read in a way that conforms with fundamental cannons of the Vienna Convention on the Law of Treaties, the same reasons that have prevented countries from adopting the universal basis of jurisdiction should be the same that should have prevented them from adopting the territorial basis of jurisdiction since both are based on the right to try nationals of non-party States. Does this mean that the ICC Statute is an illegal treaty in this aspect? In this author’s view not, but the scale of resistance to it will depend to a large extent on how provisions of the Statute are interpreted. No matter how we are all interested in seeing the ICC functioning, denying that under current international law consent is needed to bind a country to the jurisdiction of an international organ will not help. Denying that the plain text of the ICC Statute appears prima facie to oblige non-party States to it will not help either. The solution to the legality of the ICC Statute is to discover in those provisions the implied passive consent of all member States to be bound by its provisions. The discussion that follows arguably supports this approach to the ICC Statute which is more non-party State and international law ‘friendly’. IV. The ICC as a Novel Mechanism to Protect Member States against International Obligations or Whether State Parties Can Have Fewer Obligations under a Treaty Than Non-State Parties The ICC Statute generally prohibits reservations,197 but allows one specific type of reservation. Under Article 124 (the same situation would apply mutatis
196 Patrick Robinson, ‘The Missing Crimes’, in The Rome Statute of the International Criminal Court: A Commentary, Antonio Cassese et al. (eds.) (Oxford: OUP, 2002), 497, at 517 (noting that at the Rome Conference, among the principal arguments against the inclusion of international terrorism as a crime within the jurisdiction of the ICC were that “some terrorist acts were ordinary crimes and thus not sufficiently serious to warrant prosecution by an international tribunal” and that
it could be more efficiently prosecuted at the national level, though it was acknowledged that there were some terrorist acts that would meet the criteria of gravity required for the inclusion of the crimes in the Statute (emphasis added) ). 197
ICC Statute, Article 120.
326
josé doria
mutandis in relation to Articles 121(5)198 and 123(1) (3)199), States may if they wish declare that they do not accept the jurisdiction of the Court for war crimes for a period of seven years when the crimes are said to have been committed either by its nationals or on its territory.200 This problem is directly linked to the question of whether the ICC binds non-party States. Because it raises the issue of whether these reservations put a State party in a situation of a non-party State or in a third ‘special’ situation. Scholars have different opinions on this issue. Some think that the intention was to relieve the State-party from the jurisdiction of the Court including when its nationals commit a crime in another State-party.201 This reading appears to be consistent with the intention of those who proposed the provision. Otherwise, were the original intention behind that provision just to make them non-party States (i.e. with the possibility of having their citizens tried for crimes committed on the territory of State parties), these States would have just preferred to refuse to join the treaty.202 The plain text of Articles 124, 123 and 121, leaves no doubts that the intention was to exempt the relevant States from the Court’s jurisdiction under
198
Article 121(5) provides: Any amendment to articles 5, 6, 7 and 8 of this Statute shall enter into force for those States Parties which have accepted the amendment one year after the deposit of their instruments of ratification or acceptance. In respect of a State Party which has not accepted the amendment, the Court shall not exercise its jurisdiction regarding a crime covered by the amendment when committed by that State Party’s nationals or on its territory. (emphasis added)
199
Article 123(1)(3) provide respectively: para. 1 Seven years after the entry into force of this Statute the Secretary-General of the United Nations shall convene a Review Conference to consider any amendments to this Statute. Such review may include, but is not limited to, the list of crimes contained in Article 5. The Conference shall be open to those participating in the Assembly of States Parties and on the same conditions
and para. 3 “The provisions of article 121, paragraphs 3 to 7, shall apply to the adoption and entry into force of any amendment to the Statute considered at a Review Conference”. 200 The full text of Article 124 provides: Notwithstanding article 12, paragraphs 1 and 2, a State, on becoming a party to this Statute, may declare that, for a period of seven years after the entry into force of this Statute for the State concerned, it does not accept the jurisdiction of the Court with respect to the category of crimes referred to in article 8 when a crime is alleged to have been committed by its nationals or on its territory. A declaration under this article may be withdrawn at any time. The provisions of this article shall be reviewed at the Review Conference convened in accordance with article 123, paragraph 1. 201 See Stephane Bourgon, ‘Jurisdiction Ratione Temporis’, in The Rome Statute of the International Criminal Court: A Commentary, Antonio Cassese et al. (eds.) (Oxford: OUP, 2002), 543, at 555; stating that
by making the declaration prescribed by the transitional provision, also referred to as the opting-out mechanism, a State stipulates that ‘it does not accept the jurisdiction of the Court.’ This confirms that when the provision was adopted, an overall exclusion of the exercise of jurisdiction by the International Criminal Court was intended. 202 France that proposed the deal threatened not to join the treaty if its suggestion was refused. The US that proposed an initial deal of 10 years both for war crimes and crimes against humanity ended up outside the ICC Statute among others exactly because this was not accepted.
conflicting interpretations of the icc statute
327
Article 12. This regime clearly puts States parties in a third privileged ‘special’ situation different from that of non-party States, and in this way State parties in practice would have fewer obligations under the ICC Statute than non-party States. This would certainly add to the originality of the ICC treaty, as it would amount to an attempt of certain members of the international community to shield their citizens from international responsibility for violation of jus cogens norms, by simply signing a treaty exempting them from it.203 It would also basically mean a treaty in which States have agreed to commit violations of jus cogens with impunity, a reading that therefore would invalidate the ICC Statute ab initio according to Article 53 of the VCLT. Moreover, since this exclusionary rule would in fact make of the ICC Statute a treaty that creates fewer obligations for State parties than for non-party States, then this reading is also perhaps what under Article 32 VCLT leads to “a result which is manifestly absurd or unreasonable”. This is so because it equals to a statement that non-party States had an obligation to enter a treaty if they also wanted to have fewer obligations than they wanted to avoid. This certainly cannot be the case. A reading that conforms to the spirit and letter of the Vienna Convention on the Law of Treaties would necessarily imply that since State parties may opt-out of the Court’s jurisdictions, this ability should the more so be available to non-party States. For non-party States therefore the ICC Statute would only be binding until such time as they do not overtly speak against it. This possibility is arguably given to non-party States in particular under Article 98(2) of the ICC Statute that allows the signing of Agreements with ICC States Parties, exempting their nationals from the Court’s jurisdiction – a practice already followed by the US. V. The ICC as a Novel Mechanism to Enforce International Obligations against Non-Member States or What Is the True Purpose of Article 98(2) Agreements Article 98 (2) agreements are bilateral agreements negotiated between the United States and third States, whether or not those States have ratified the Rome Statute.204 Under these agreements, each State consents not to surrender 203 States assembled in Rome have denounced this approach as inconsistent with international law. See US arguing that
the proposed Statute would also seek to place State parties in a privileged position vis-a-vis nonparty States, by permitting prosecution of officials and nationals of non-parties while shielding officials and nationals of parties from prosecution for the same crimes.
Bassiouni, supra note 19, at 400, para.21. 204 As of 22 November 2006, the US had signed such Agreements with more than 100 countries of the world both State parties and non-party States <www.amicc.org/usinfo/ administration_policy_BIAs.html> (accessed on 24 November 2006).
328
josé doria
the nationals of the other to the ICC without first obtaining the other State’s consent. The language adopted in the Agreement between the United States and Romania205 is typical of the Agreements with State parties. Article 2 of that treaty states: Persons206 of the United States of America present in the territory of Romania shall not, absent the expressed consent of the government of the United States of America, (a) be surrendered or transferred by any means to the International Criminal Court for any purpose, or (b) be surrendered or transferred by any means to any other entity or third country, or expelled to a third country, for the purpose of surrender to or transfer to the International Criminal Court.207 Arguably this practice is contrary to the prevailing interpretation of Article 98(2) of the ICC Statute which has been that Article 98(2) was intended to apply only in situations where a ‘sending State’ enters into an agreement on the status of armed forces on a foreign territory, but not for cases of general agreements signed to shield every one of its citizens.208 This interpretation is in particular based on a narrow reading of Article 98(2). ICC supporters claim that the use of the expression ‘sending States’ was intended solely for situations of persons sent to another country in official capacity, and not to all citizens of a country.209 The US however has been signing general agreements without any connection to any specific forces but in reference to any American citizen, who are
205 2002 Agreement between the government of the United States of America and the government of Romania regarding the surrender of persons to the International Criminal Court (2002), <www.amicc.org/docs/Romania_98_2.pdf> (accessed on 22 October 2006). 206 In the Agreement, ‘Persons’ are defined as “current or former government officials, employees (including contractors), or military personnel or nationals of the United Sates.” See 2002 US-Romania 98(2) Agreement, Article 1, infra note 207. 207 2002 Agreement between the government of the United States of America and the government of Romania regarding the surrender of persons to the International Criminal Court (2002), <www.amicc.org/docs/Romania 98 2.pdf> (accessed on 22 October 2006). 208 Article 98(2) provides that
the Court may not proceed with a request for surrender which would require the requested State to act inconsistently with its obligations under international agreements pursuant to which the consent of a sending State is required to surrender a person of that State to the Court, unless the Court can first obtain the cooperation of the sending State for the giving of consent for the surrender. 209 The Draft Council Conclusions on the International Criminal Court submitted for adoption of the Council of the European Union on 30 September 2002, recommended that States Parties of the ICC members of the European Union by entering into such agreements take in consideration that
any solution should cover only persons present on the territory of a requested State because they have been sent by a sending State cf. Article 98 paragraph 2 of the Rome Statute.
Council of the European Union, Draft Council Conclusions on the International Criminal Court, (30 September 2002) <www.amicc.org/docs/EC9_30_02.pdf> (accessed on 24 November 2006).
conflicting interpretations of the icc statute
329
thus protected if found guilty of a crime, claiming that this was the only way to reconcile the ICC Statute with general international law.210 Was Article 98(2) intended for these general agreements or not? Apart from the purported language in the Article that refers to ‘sending States’ however, there is arguably nothing else in the ICC Statute itself or in the travaux preparatoires that would justify such a narrow reading. The issue actually came up in Rome out of US concerns in relation to the need to ensure prior State consent before jurisdiction of the ICC could be claimed over nationals of nonparty States. The US then argued by giving examples of clear cases when consent was absolutely needed and stated that with regard to the States which must consent, the regime must include a non-State Party whose official actions were alleged to be crimes. That might be the State on whose territory a crime had occurred but, in the case of peacekeeping or international conflict, it might be another State: the State which had sent the troops concerned. That State should be responsible for their prosecution or for consenting to their prosecution by the Court (emphasis added).211
Arguably the use of the terms ‘must include’ and ‘might’ clearly speaks of an illustrative intention, and does not show that the case of those ‘officially sent’ was the only one that the US had envisioned. This was only the most obvious reason why such consent should be requested, but the language of the travaux preparatoires is inclusive not exclusive. If there was any doubt the present US practice of signing general agreements and other remarks during the Conference clearly show that the US by proposing these provisions did not intend the 98(2) agreements to be so restricted. This broader interpretation of Article 98 (2) is also more consistent with other Articles (on the opt-out regime for certain crimes) and the objective and purposes of the ICC Statute (creation of a universal organ to contribute to peace and security by punishing international crimes), which, as it is argued here, is a treaty based on presumed or implied passive consent of all member States of the international community to be bound by its provisions. Indeed it is perfectly defensible to argue that the US general agreements were prompted by provisions of the ICC Statute that on one side purport to bind nationals of non-party States to the jurisdiction of the Court (under Article 12) and on the other, confer on State parties a right to opt-out of the Court’s jurisdiction (Articles 121,123,124). It is more logical and legally sound
210 The US on the last day of the Rome Conference stated that it “deeply regretted that following four years of work, it faced the end of the Conference with profound misgivings”. The US argued that the draft Statute submitted for adoption did not contain the necessary opt-out provision to attract many States, and it attempted to exercise jurisdiction over the official actions of non-parties, a significant departure from the established principles of international law. Bassiouni, supra note 19, at 399, para. 20. 211 Bassiouni, supra note 19, at 162, para. 24.
330
josé doria
to presuppose that if anything, the opt-out provisions of the ICC Statute were intended to equate States parties with the status of non-party States. Therefore if States parties were given a right to exclude for themselves the Court’s jurisdiction, it is only normal that such a right is also available to non-party States. Even if initially Article 98(2) was not intended for such general Agreements (as some would assert), a broader interpretation is consistent with the purposes and reach of the ICC Statute in line with the Vienna Convention on the Law of Treaties. Not surprisingly countries that have voted against the ICC Statute remain adamant in their position that unless the ICC Statute is read in such a friendly way, it would be manifestly contrary to international law.212 Hence these countries view the signing of these 98(2) general agreements as consistent with the ICC Statute and general international law including the Vienna Convention on the law of Treaties. On the other side, it is worth noting that this broad interpretation does not bring significant changes to a situation already covered by the narrow letter of Article 98(2). Arguably if anyone, those sent in official capacity like the members of the military are the ones that most probably would have come under the Court’s jurisdiction (and these are already clearly protected by that Article). Other citizens (with the exception of mercenaries) such as tourists, if anything would most probably qualify just as victims. In this author’s view, accepting a broader reading of the ICC Statute has a far reaching ‘constitutional’ importance, since it would bring back to the same level States parties that opt-out of the Court’s jurisdiction and those that are non-parties to the Statute, that should also therefore be entitled to exempt themselves from the presumed passive consent that links them to the ICC Statute. Hence put on the same level, non-party States and State parties to the ICC Statute could only be submitted again under the Court’s jurisdiction by a Chapter VII referral decision of the UNSC.213
212 Thus reacting to the 2002 EU Draft Conclusions on 98(2) Agreements, the US Department of State insisted that it would continue the practice of signing these general agreements, calling it a ‘friendly’ approach to the ICC Statute, which made it consistent with international law. See Richard Boucher, US Department of State, ‘US Assesses EU Decision on ICC Article 98 Agreements’ (Press Briefing of 1 October 2002) <www.useu.be/Categories/JusticeandHomeAffairs/ Oct0102BoucherEUArticle98.ICC.html> (accessed on 22 October 2006). 213 Certain authors go so far as to admit that the exclusion was intended to be so absolute that even the UNSC would not be able to submit these States back to the Court’s jurisdiction. Bourgon, supra note 201, at 556, noting that
the issue is whether the Security Council, acting in accordance with Article 13(b) and pursuant to Chapter VII of the United Nations Charter could override the transitional provision. Considering the difference between Article 12 and 124, whereby the former allows Article 13(b) to operate independently while the latter does not, it appears that the Security Council cannot override the opting out mechanism.
conflicting interpretations of the icc statute
331
Essentially it would also mean that the US and other non-party States indeed recognize the ICC Statute as passively binding on non-party States unless these States have spoken against its application either generally (by signing in particular these agreements) or by way of using the complementarity procedure in specific cases of prosecutions of their nationals. The principle of presumed or implied passive consent would be then the binding principle of the ICC Statute.214 This reading is important since it has the potential of bringing the ICC back into the ‘constitutional’ framework of the UN system too. VI. Conclusions There is no doubt that under normal circumstances the ICC could not extend its jurisdiction to nationals of non-party States without the consent of those States to be bound by its Statute. The research has shown that the crucial point in the arguments against and in favor of non-party State consent is whether States can establish new subjects of international law to which they cede their rights that affect third states without the consent of these third states. International law and the practice of States do not appear to support such a proposition. Under normal circumstances the only valid mechanism through which a State could be submitted to the jurisdiction of the Court without its consent would be a Chapter VII referral by the UNSC. Only a UNSC decision would bind both State parties and nonparty States that would have opted out of the Court’s jurisdiction. Applying the Rules of interpretation of the Vienna
In this author’s view, this is a mistaken reading based on the assumption that the ICC Statute has rewritten the UN Charter. The ICC Statute only offers the UNSC a possibility of using its resources for the Council’s own purposes of maintaining peace and security as the entity with the primary responsibility in this area. Once a UNSC decision has been made in relation to a State that has excluded for itself the Court’s jurisdiction, this State is considered an ordinary non-party State (for that purpose) and therefore can be submitted back to the Court’s jurisdiction by the UNSC acting under the UN Charter coercive Chapter VII provisions. The obligations of that State were not waived in relation to the UN Charter, obviously not. 214 Presumed passive consent to the ICC jurisdiction has to be distinguished from active forms of consent (permanent or ad hoc). Passive consent binds the State to the Court’s jurisdiction from the moment of the entry into force of the Statute according to Article 11(1) (it however only engages the State for crimes committed by nationals on territory of member States). In contrast active forms of consent bind the State to the Court’s jurisdiction from the moment of the entry into force of the Statute for that State (under Article 11(2) ) with the exception of ad hoc consent given ‘with respect to the crime in question’ (under Article 12(3) ). Only in this case the State is fully bound by all provisions of the Statute including the cooperation regime. For this reason active forms of consent cannot be presumed, must be given in expressis and engage the State for crimes committed not only on its own territory, but also by its nationals on the territory of non-member States (unless it has opted out of the ICC Statute as provided for in certain cases discussed above).
332
josé doria
Convention on the Law of Treaties, this author finds that there are serious reasons to believe that this reading more consistent with international law can be made out of the apparently conflicting provisions of the ICC Statute. In particular there appears to be a conflict between the right conferred upon Members States to opt out of the Court’s jurisdiction and a similar right apparently denied to non-Member States, not willing to be a party to the ICC Statute. To suggest that the only option available to any country wishing to have fewer obligations under the ICC Statute, would be to adhere to it, appears “manifestly unreasonable” and this is certainly a reading one should avoid. Clearly the drafters of the Statute did not intend to confer on Court’s Members more self-excluding rights than to non-Member States. Article 98 read in combination with other self-excluding provisions of the ICC Statute allows a perfectly defensible interpretation of the Statute as a treaty based on presumed or implied passive consent of all members of the international Community. Such a reading would also be consistent with the purported objective of contributing to international peace and security, which is the broader objective that mandates the Court’s work to punish specific egregious acts of violation of human rights and international law. There is no doubt that the discussion above shows the relevance of the rules of interpretation of treaties under the VCLT, 30 years on.
INTERPRETING CONSTITUTIVE INSTRUMENTS OF INTERNATIONAL CRIMINAL TRIBUNALS: REFLECTIONS ON THE SPECIAL COURT FOR SIERRA LEONE Phoebe Okowa I. Introduction In 2000, the United Nations entered into a bilateral agreement with the government of Sierra Leone, which established a Special Court to try those responsible for the most serious atrocities committed during the Sierra Leone conflict.1 In October 2009, the Special Court delivered its last but one decision, in relation to the atrocities arising out of the civil war in that country.2 In the one outstanding case, the court has disposed of the preliminary issues relating to its legal basis under the constitutive treaty.3 It is therefore an appropriate moment to assess the Court’s contribution to the interpretation of its constitutive instrument, in this collection of essays that reflect on 30 years since the entry into force of the Vienna Convention. The competence of the UN to enter into the agreement of this kind, although unprecedented, was not in doubt, the UN having characterised the situation in Sierra Leone as constituting a threat to international peace and security, it was conceivably free to determine what it regarded as an appropriate response in the exercise of its powers under the Charter, including the establishment of an international criminal tribunal.4 Yet as is apparent from the first line of cases considered by the Special Court, the legal basis of the tribunal has not been free from controversy. It has raised in a profound manner, the relationship between the constitutive instrument of an
1
2002 Agreement between the United Nations and the Government of Sierra Leone on the Establishment of a Special Court for Sierra Leone, available at the Special Court for Sierra Leone website: <www.sc-sl.org/LinkClick.aspx?fileticket=CLk1rMQtCHg%3d&tabid=176>. (last accessed on 1 November 2009). 2 Prosecutor v. Issa Hassan Sesay; Prosecutor v. Morris Kallon; Prosecutor v. Augustine Gbao (RUF Case), Appeal Judgment of 26th October 2009, Special Court for Sierra Leone, (SCSL-0415-A-1311). 3 Prosecutor v. Charles Taylor, Appeals Chamber Decision of 31 May 2004, Special Court for Sierra Leone, (SCSL-03-01-I-059). 4 In Prosecutor v. Charles Taylor, the Special Court observed that it was: certain that the power of the security council to enter into an agreement for the establishment of the Court was derived from the Charter of the United Nations both in regard to the general purposes of the United Nations as expressed in Article 1 of the Charter and the specific powers of the Security Council in Articles 39 and 41. These powers are wide enough to empower the Security Council to initiate, as did by resolution 1315, the establishment of the Special Court by agreement with Sierra Leone
334
phoebe okowa
international criminal tribunal and Security Council powers under the UN Charter. The Court was asked to consider the competence of the UN to conclude a treaty on apparently equal terms with a member state, calling for a consideration of the hierarchical relationship between the UN Charter and a bilateral treaty of the kind concluded with the government of Sierra Leone. It has also brought to the fore the extent to which the lex specialis character of constitutive instruments affects their interpretation and whether this may require a departure from accepted canons of treaty interpretation. It also calls for an examination of the extent to which due process concerns, including fairness, assume a particular prominence in the interpretation of treaties in the field of international criminal justice, and whether the liberties of the individual at the centre of these instruments are in principle compatible with other canons of treaty interpretation such as the subsequent practice of the parties. For instance, what is to be the relationship between the nullum crimen (non-retroactivity of penal laws) and the rule that a treaty has to be interpreted in context, which may require the interpretation to take into account subsequent developments in general international law, including changes in the substantive content of the law? The Special Court had to confront this issue in relation to the question of criminality of child soldiers, and whether the lex lata content of international criminal law in this area had crystallized by the time of the Sierra Leonean civil war. The article addresses the legal status of peace agreements, including the legal consequences of an agreement between a state and a rebel organisation. It reflects on the legal significance as a matter of treaty law of having third parties signing a treaty as moral guarantors and the implications for enforcement. The jurisprudence of the Special Court has also highlighted the continuing uncertainty on how international law deals with conflicts of norms especially those that arise in relation to international criminal law and those of a general character under customary international law. The debate whether accountability for international crimes as contained in the applicable treaties should take precedence over the developed law of state immunity is an area where this conflict has been particularly pronounced. The Sierra Leone cases also indirectly raise the question of coherence in the normative framework of international law. They involve a consideration of the development of international criminal law and the extent to which this can be achieved within the ideological milieu of general international law, including principles on treaty interpretation. This chapter is not intended as a critique of the substantive law coming within the Court’s jurisdiction or the application of that law in the jurisprudence of the Court. This has been exhaustively dealt with elsewhere.5 Instead it Prosecutor v. Charles Taylor, supra note 3, para. 37; See also Danesh Sarooshi, The United Nations and the Development of Collective Security: The Delegation by the UN Security Council of its Chapter VII Powers (Oxford: OUP, 1999). 5 Robert Cryer, ‘A Special Court for Sierra Leone?’, ICLQ 50 (2001): 435–46; Cesare P.R. Romano, André Nollkaemper and Jann K. Kleffner (eds.), Internationalized Criminal Courts,
interpretation and the special court for sierra leone
335
focuses on the legal questions generated by the treaty basis of the Special Court in light of the law of treaties and the recent jurisprudence of the tribunal. II. Background For much of the early 1990’s Sierra Leone was involved in a bitter civil war led by the Revolutionary United Front (RUF) a renegade rebel group, supported by the then Liberian President Charles Taylor. The rebellion, although clearly lacking in ideological conviction or popular support, was in part a response to the widespread dissatisfaction with successive post-colonial regimes that were seen as corrupt and indifferent to the needs and aspirations of the general population. The government’s attempt to quell the rebellion by arming informal militia groups (Civil defence forces) incensed the army who responded by attacking these groups.6 There were intermittent attempts to resolve the conflict peacefully. In 1996, general elections were held in which Ahmad Kabbah was elected president. The newly elected leader swiftly moved to conclude a peace agreement with the RUF leader Sankoh, under the terms of which a general amnesty was granted to the RUF. The agreement which came to be known as the Abidjan accord, granted the RUF an immediate ceasefire, and also committed the government of Sierra Leone to a programme of disarmament and demobilization of the rebels.7 This agreement did not last, and within a year, President Kabbah himself was overthrown in a military coup, led by a faction of the Sierra Leone Army, the Armed Forces Revolutionary Council (AFRC). The coup and the subsequent power sharing deal between the RUF and AFRC were widely condemned by the international community, including the Economic Community of West African States (ECOWAS) which imposed an embargo. Following an intervention by the ECOWAS, Cease-fire Monitoring Group (ECOMOG) President Kabbah was reinstated in power in 1998. The horrendous atrocities committed by all parties to the civil war including the serious violations of human rights have been extensively documented elsewhere.8 It is remarkable that the main strategy employed by all combatants in Sierra Leone, East Timor, Kosovo and Cambodia (Oxford: OUP 2004); Sarah Williams, ‘Amnesties in International Law: The Experience of the Special Court for Sierra Leone’, Human Rights Law Review 5 (2005): 271–309; William A. Schabas, ‘The Relationship Between Truth Commissions and International Courts: The Case of Sierra Leone’, Human Rights Quarterly 25 (2003): 1035–66. 6 For a concise account of the background to the conflict see Taru Kuosmanen, Bringing Justice Close: Hybrid Courts in Post-Conflict Societies (Helsinki: Erik Castrén Institute, 2007), at 87. 7 See 1996 Peace Agreement between the Government of the Republic of Sierra Leone and the Revolutionary United Front of Sierra Leone, available at: <www.sc-sc.org/abidjanaccord .html> (last accessed on 1 November 2009). 8 See Sierra Leone Truth and Reconciliation Commission, 2004 Report of the Sierra Leone Truth and Reconciliation Commission (TRC Report), available at: <www.trcsierraleone.org/ drwebsite/publish/index.shtml> (last accessed on 1 November 2009); Ibrahim Abdullah, ‘Bush
336
phoebe okowa
the Sierra- Leonean conflict was the deliberate targeting of civilians. Large numbers of the civilian population were subjected to amputations, protracted acts of torture, forced recruitment of child soldiers, as well as abduction and deliberate destruction of civilian properties. The worst atrocities were committed by both the RUF and AFRC soldiers in 1996. In the face of apathy by the international community, and a largely ineffective response from the Nigerian led ECOMOG peacekeepers, the Government of President Kabbah was forced to enter into a peace agreement with the RUF rebels. Under the terms of the Lome agreement, the RUF rebels were granted a complete and unconditional amnesty, and in return, the RUF agreed to abandon the arms struggle. This was to prove particularly controversial, especially since at the time of its conclusion, the UN representative had apparently made a last minute reservation to the terms of the amnesty provision, stating that from the point of view of international law and of the United Nations, the grant of amnesty did not extend to international crimes of genocide, crimes against humanity, war crimes and other serious violations of international humanitarian law.9 The United Nations was not formally a party to the Lome agreement but had signed along with other states and institutions as moral guarantors of the treaty, raising questions as a matter of treaty law of the binding nature of this reservation and its precise basis.10 The RUF rebels were incorporated into a government of national unity and their leader Foday Sankoh became vice-president as well as the de facto head of a government parastatal with responsibility for mines. There followed a prolonged period when the terms of the ceasefire were not respected by both sides. The United Nations Security Council intervened to help implement the terms of the ceasefire. This was eventually achieved in 2002 when President Kabbah declared the war over. III. Why a Treaty? The agreement of which the Special Court forms a part was negotiated at the behest of the government of Sierra Leone.11 At the end of the civil war and Path to Destruction: The Origin and Character of the Revolutionary United Front/Sierra Leone’, Journal of Modern African Studies 36 (1998): 203–35. 9 UN Secretary-General, Report of the Secretary-General on the Establishment of a Special Court for Sierra Leone, (4 October 2000) (S/2000/915), para. 23. 10 The official signatories to the 1999 Lome Accord are the government of Sierra Leone under president Kabbah and the RUF. The agreement was witnessed by the UN, the OAU and ECOWAS together with the representatives of Togo, Burkina Faso, Ghana, Liberia, Nigeria and Ivory Coast. 11 For the text of the Agreement and Statute of the Special Court, see <www.sc-sl.org> (last accessed on 1 November 2009); 2002 Ratification Act of the Special Court Agreement of 2002, Act, Supplement to the Sierra Leone Official Gazette Vol. CXXX No.2, available at: <www .sc-sl.org/LinkClick.aspx?fileticket=BcKTbFltxZk%3d&tabid=176> (last accessed on 1 November 2009).
interpretation and the special court for sierra leone
337
after the formal conclusion of a peace agreement,12 Sierra Leone’s president Kabbah wrote to the Secretary General of the United Nations asking for the UN’s assistance in establishing a tribunal that would try those responsible for some of the worst atrocities committed during the Sierra Leone conflict.13 There was a strong feeling that if the tribunal was to be perceived as legitimate and provide some element of closure on the worst period in Sierra Leone’s history, it had to be rooted in international ideas of justice but also in Sierra Leone’s legal traditions. This model was in part intended as a corrective to the two ad hoc tribunals for the former Yugoslavia and Rwanda which had been criticised as far removed and therefore out of touch with the communities where the atrocities had taken place. The involvement of the government of Sierra Leone in the structure, composition, and locus of the Special Court was intended to create a special bond between the Court and the people of Sierra Leone, and to help in the development of the institutions of justice. The latter objective was to be achieved through a series of outreach and public education programs.14 As a result, the tribunal is mixed/hybrid, both in terms of its composition as well as the substantive law it is mandated to apply. The judges and personnel were to be appointed by the UN Secretary General, and the government of Sierra Leone respectively. The government of Sierra Leone was empowered to appoint one of the three judges in the trial chamber and two of the five judges in the Appeals Chamber.15 The Court is required to apply generally accepted principles of international law and Sierra Leonean criminal law.16 The Statute specifically permits the prosecution of offences UNDER Sierra Leonean law, such as those relating to the rape of girls as provided for under Sierra Leone’s the Prevention of Cruelty to Children Act of 1926. Prosecution may also be undertaken for offences relating to wanton destruction of property under the Malicious Damage Act of 1861. The Secretary General noted that these two categories of offences had a particular resonance under Sierra Leonean law but were inadequately regulated under general international law.17 12 1999 Peace Agreement between the Government of Sierra Leone and the Revolutionary United Front of Sierra Leone (RUF), signed on 7 July, 1999 in Lome, Togo (Lome Accord); available at: <www.iss.co.za/AF/profiles/SieraLeone/pax.pdf> (last accessed on 1 November 2009). 13 See President of Sierra Leone, Letter from the President of Sierra Leone to the UN Secretary General, (12 June 2000) (S/2000/786). 14 President of the Special Court for Sierra Leone, First Annual Report of the President of the Special Court for Sierra Leone, available at: <www.sc-sl.org/LinkClick.aspx?fileticket=NRhDcb HrcSs%3d&tabid=176> (last accessed on 1 November 2009). 15 See Article 12(1) of the 2002 Statute of the Special Court for Sierra Leone (hereinafter ‘the Statute’), available at: <www.sc-sl.org/LinkClick.aspx?fileticket=uClnd1MJeEw%3d&tabid =176> (last accessed on 1 November 2009). 16 See Art. 1(1) of the Statute of the Special Court. 17 UN Secretary-General, Report of the Secretary-General on the Establishment of a Special Court for Sierra Leone, (4 October 2000) (S/2000/915), para. 19.
338
phoebe okowa
The international crimes provided for under the Court’s statute broadly replicates the range of offences coming within the jurisdiction of the two ad hoc tribunals and forming part of general international law. They include crimes against humanity, violations of common article 3 of the Geneva Conventions and Additional Protocol II, serious violations of international humanitarian law such as the conscription of children and attacks against peacekeepers. One other consequence of this arrangement is that the Special Court is not a UN tribunal and as a result, it is not funded out of the UN budget but by voluntary donations from member states.18 This is a significant shortcoming for two reasons. First, it has placed the Court and its staff in a precarious financial position especially when the donations have not been forthcoming. Second, it means that orders such as arrest warrants directed at member states and any other obligations of co-operation addressed to third states are not mandatory, nor do they have a sense of priority as they do not have the imprimatur of the UN Charter.19 IV. Legal Consequences of the Lome Peace Agreement A. The Immunity Question It has already been noted that the bilateral treaty between the UN and the government of Sierra Leone establishing the Special Court was concluded against the background of the Lome Peace Accord between the government of Sierra Leone and the Sierra Leonean rebel group Revolutionary United Front of Sierra Leone (RUF). The United Nations along with other states also signed as guarantors of the agreement. Article IX of the Lome accord, granted absolute and complete pardon and reprieve to all RUF combatants and collaborators in respect of anything done by them20 “in pursuit of their objectives” for atrocities committed during the Sierra Leone conflict. This agreement replaced an earlier bilateral treaty between the Revolutionary United Front and the Government of Sierra Leone, which had been signed at Abidjan in 1996, but had in fact failed to stop the violence.21 The Abidjan accord had also granted
18 See UN Security Council Resolution 1315, para. 8(c); Report of the Secretary-General, supra note 17, para 68. 19 See in particular Article 25 and 103 of the Charter of the United Nations. 20 See UN Secretary-General, Seventh Report of the Secretary-General on the United Nations Observer Mission in Sierra Leone (1999) (S/1999/836), para. 7. 21 1996 Peace Agreement between the Government of Sierra Leone and the Revolutionary United Front of Sierra Leone, (UN Doc. S/1996/1034). Article 14 of this agreement provided that the
Government of Sierra Leone shall ensure that no official or judicial action is taken against any member of the RUF/SL, in respect of anything done by them in pursuit of their objectives as members of that organisation up to the time of signing this agreement
interpretation and the special court for sierra leone
339
the combatants complete reprieve for any atrocities committed by them before 1996. Unlike the Abidjan accord, the Lome agreement had additional provisions which contained commitments on the part of all parties to disarm, the establishment of a broad-based government of national unity, and it also provided for internationally supervised elections. However, the Secretary General’s Special representative had only signed the agreement, apparently subject to the proviso that the amnesty would not apply to international crimes of genocide, crimes against humanity, war crimes, and other serious violations of international humanitarian law.22 The signature of the Secretary General and the third states who had signed as guarantors did not however qualify the essentially bilateral character of the instrument- in effect an agreement between a state and a rebel movement. This amnesty provision and the disclaimer by the Special Representative of the Secretary General (SRSG) were to trouble the court in some of the most significant cases to come before it.23 What was to be the effect of this amnesty provision and the SRSG proviso within the broad scheme of the Court’s mandate? Moreover the Secretary General was not a party but a moral guarantor raising the question of legal significance if the instrument was to be treated as binding. The existing literature by their silence or ambivalence on the question of the legal status of peace agreements could be seen as an affirmation of the inherently extra-legal character of such agreements.24 Crucially Article 10 of the Statute of the Special Court provides that: An amnesty granted to any person falling within the jurisdiction of the Special Court in respect of the crimes referred to in Articles 2 to 4 of the present Statute shall not be a bar to prosecution
As is apparent from its terms, this provision was directly at odds with article IX of the Lome Agreement. The Court was thus immediately faced with the
22 Curiously the test of this qualification was never appended to the agreement itself and has never been found, Christine Bell, On the Law of Peace, Peace Agreements and the Lex Pacificatoria (Oxford: OUP, 2008), at 247. 23 See Prosecutor v. Morris Kallon and Brima Bazzy Kamara, Decision of 13 March 2004 of the Appeals Chamber on Challenge to Jurisdiction: Lome Accord Amnesty, Special Court for Sierra Leone, (SCSC-04-15-PT-060); Prosecutor v. Allieu Kondewa, Decision of 25 May 2004 of the Appeals Chamber on Lack of Jurisdiction/ Abuse of Process: Amnesty Provided by the Lome Accord, Special Court for Sierra Leone, (SCSL-04-14-T-128-7347); Prosecutor v. Augustine Gbao, Decision of 25 May 2004 of the Appeals Chamber on Preliminary Motion on the Invalidity of the Agreement between the United Nations and the Government of Sierra Leone on the Establishment of the Special Court, Special Court for Sierra Leone, (SCSL-04-15-PT-141); Prosecutor v. Moinina Fofana, Decision of 25 May 2004 of the Appeals Chamber on Preliminary Motion on Lack of Jurisdiction: Illegal Delegation of Jurisdiction by Sierra Leone, Special Court for Sierra Leone, (SCSL-04-14-PT-102). 24 But for an exceptional critical account on the status of these agreements see Bell, supra note 22.
340
phoebe okowa
question of reconciling two conflicting provisions, both claiming lex specialis status, in the context of the specific dispute. Moreover, it was significant that any interpretation carried out had to be coherent with established norms of treaty interpretation under general international law. Significantly, the dispute thus brought to the fore the legal status of a peace agreement concluded in part between a state and non-state actors and the circumstances if any, when its provisions could be regarded as dispositive. Rescinding this agreement in its entirety would have done untold damage to the fragile peace settlement that brought the civil war to an end. Yet giving full effect to it was also clearly incompatible with an emerging international system of justice that required perpetrators of heinous crimes to be brought to justice. Moreover, the argument was made before the Court and elsewhere, that a potential amnesty was in any event of limited effect; as the crimes in question were subject to universal jurisdiction. The alleged perpetrators therefore remained liable to prosecutions in the courts of foreign states not bound by the amnesty arrangements.25 The negotiators had envisaged a potential conflict, and made an attempt to resolve this by inserting a provision into the Court’s Statute which preserved the amnesty in so far as it related to crimes under Sierra Leonian law (Article 5) but deprived it of effect in relation to international crimes (Articles 2–4). In Prosecutor v. Kallon and Prosecutor v. Kamara,26 the first of four cases on the amnesty question, this compromise was challenged as incompatible with the Vienna Convention on the Law of Treaties. Counsel for Kallon argued that reneging on the immunity provisions of the Lome agreement was tantamount to a unilateral attempt by one party to a treaty amending its terms without the consent of the other party. The Special Court was therefore asked not to assert jurisdiction over crimes committed prior to July 1999 when an amnesty was granted by virtue of the Lome Agreement and for a declaration that it would be an abuse of process to allow the prosecution of crimes pre-dating the Lome Agreement.27
This objection to the Court’s jurisdiction in turn raised a preliminary question whether a tripartite peace agreement between a rebel movement and state parties (however designated) could be described as a treaty, the interpretation and implementation of which was subject to the law of treaties. Was the legal quality of the agreement elevated on account of the signature of the UN and
25 Williams, supra note 5, at 271–309; Christine Bell, Peace Agreements and Human Rights (Oxford: OUP, 2000), at 262. 26 Prosecutor v. Morris Kallon, Prosecutor v. Brima Bazzy Kamara, Decision of 13 March 2004 of the Appeals Chamber on Challenge to Jurisdiction: Lome Accord Amnesty, Special Court for Sierra Leone, (SCSL-04-15-PT-060-I & II; SCSL-04-16-PT-033-I & II). 27 Id., para. 1.
interpretation and the special court for sierra leone
341
other states as moral guarantors? Lex lata content of treaty law does not recognise the designation of a signatory as a moral guarantor and the treaty making capacity of non-state groups could also not be stated with authority. The prosecution maintained that the agreement was wholly without consequence on the international plane; that as an instrument between two national bodies, its effect was limited to domestic law and was in any event not intended to cover crimes coming within the jurisdiction of the Special Court. It was held that the agreement was not a treaty as the RUF lacked personality to conclude an agreement – treaties by their nature could only be ratified between states and international organisations.28 The dispute thus involved the consideration of a preliminary point of some significance as to the precise legal status of an agreement concluded between a state and rebel groups. On one view, since as a matter of general international law a state is under no obligation not to fight an insurrection – a matter regarded as purely internal to that state,29 an obligation on the part of a state to cease all armed activities has no binding effect on the international plane. The prosecution’s argument was of course consistent with a particular view of international law within the legal positivist school, where only those rules that have been consented to by states are binding.30 On this view, international law is still an essentially inter-state legal order, and only in very exceptional circumstances are other groups granted capacity under it. Yet there is an increasing tendency to reject this predominant in the literature. The conception of international law as a pluralist legal order with gradations of binding norms, all qualitatively different, and extending to agreements entered into by non-state groups has been persuasively argued in some of the recent literature.31 The existence in many areas of international law such as in the field of environmental protection of soft law norms which are generally taken to be binding is a further indication of the outdated character of the pure form of legal positivism advanced by the prosecution in the Sierra Leone cases. In any case, objections to the justiciability of peace agreements must be set against the generally accepted principle that instruments emanating from a government, and entered to on the international plane are intended to have legal effect as far as possible.32 Moreover, to the extent that coercive enforcement is rarely a
28
Id., para. 48. Brad Roth, Governmental Illegitimacy in International Law (Oxford: Clarendon Press, 1999). 30 Gennadij M. Danilenko, Law-Making in the International Community (Dordrecht: Nijhoff, 1993). 31 Bell, supra note 25; Alan Boyle and Christine Chinkin, The Making of International Law (Oxford: OUP, 2007). 32 Paul Reuter, An Introduction to the Law of Treaties, (London: Pinter, 1989), at 26, para. 74; Robert Jennings and Arthur Watts (eds.), Oppenheim’s International Law: Vol. I Parts 2–4 (London: Longman, 1992, 9th edition), at 1280. 29
342
phoebe okowa
defining hallmark of international agreements, the fact that the peace agreement could not be treated as dispositive in an adjudicatory forum is only marginally fatal to its legal quality. The defence also maintained that the treaty character of the Lome agreement was confirmed by the fact of its ratification by the Sierra Leonean Parliament, a process that under law of Sierra Leone was reserved for binding international instruments.33 The defence also attached particular importance to the fact that six other states had under article 34 ratified the treaty as moral guarantors further confirming the internationally binding character of its provisions.34 In its decision of 13th March 2004, the Court adopted the prosecution’s argument. It denied that the Lome peace agreement was an international treaty the provisions of which had to be taken into account by the government of Sierra Leone before entering into an agreement with conflicting provisions, in the instrument establishing the Special Court.35 The amnesty provisions, the court noted, could constitute a bar to prosecution before national courts but they were wholly without effect on the international arena.36 The court further observed that any amnesty that encompassed crimes against humanity, serious war crimes, genocide, or torture would be wholly invalid under international law.37 But what was the precise legal basis of the invalidity as a matter of general international law? The potential reach of amnesties laws has been a matter of controversy.38 Those arguing that international law in some form should recognise amnesties frequently rely on Article 6 (5) of Additional Protocol II to the 1977 Geneva Conventions. This Article provides that Authorities in power shall endeavour to grant the broadest possible amnesty to persons who have participated in the armed conflict, or those deprived of their liberty for reasons related to the armed conflict, whether they are interned or detained.
But as the ICRC has pointed out in a comprehensive survey of customary law rules, the provision is not an endorsement of a general amnesty but a reflection of a broad principle of combatant immunity- that insurgents should not 33
Prosecutor v. Morris Kallon, Prosecutor v. Brima Bazzy Kamara, supra note 26, para. 25. Id., para. 30. 35 Id., paras. 42, 63 and 86. 36 Id., paras. 67 and 86. 37 Id., para. 71. 38 There is now a vast literature on this subject. See Steven Ratner, Jason Abrams and James Bischoff, Accountability for Human Rights Atrocities in International Law: Beyond the Nuremberg Legacy (Oxford: OUP, 2009, 3rd edition), at 168; Benedict Chigara, Amnesty in International Law: The Legality Under International Law of National Amnesty Laws (Harlow: Logman, 2002); Naomi Roht-Arriaza, ‘Sources in International Law of Treaties of an Obligation to investigate, Prosecute and Provide Redress’, in Impunity and Human Rights in International Law and Practice, Naomi Roht-Arriaza (ed.) (New York, NY.: OUP, 1995), 24–38. 34
interpretation and the special court for sierra leone
343
be punished for the mere act of having taken part in an insurrection.39 It is therefore neutral on the question of immunity. But even assuming that the Lome agreement was legally binding on the international plane, what effect did it have on obligations to prosecute? The Court noted that in a hierarchically structured international system, the prohibition of international crimes such as genocide, crimes against humanity, torture were superior and could not be derogated from by treaty.40 In a recent study the International Law Commission has also concluded that international crimes such as those listed have the status of jus cogens, and therefore priority over all other conflicting international obligations.41 There is a considerable degree of consensus that the priority that attaches to jus cogens norms invalidates international law obligations in conflict. But uncertainty also persists as to whether the invalidated norms must be treaty rules or whether the jus cogens effect also extends to rules of general international law, or even rules which although not formally classifiable as treaties are nevertheless regarded as binding rules on the international plane. The International Criminal Tribunal for the former Yugoslavia in the Furundzija Case had no doubt that it extended to both. The tribunal observed that: Because of the importance of the values it [the prohibition of torture] protects, this principle has evolved into a peremptory norm of jus cogens, that is, a norm that enjoys a higher rank in the international hierarchy than treaty law and even ‘ordinary’ customary rules. The most conspicuous consequence of this higher rank is that the principle at issue cannot be derogated from by states through international treaties or local or special customs or even general customary rules not endowed with the same normative force.42
In the Pinochet Case,43 a majority of the House of Lords, was inclined to deny granting immunity to the former Chilean leader by way of response to an extradition request, on the grounds that the crimes in respect of which the Spanish authorities had requested his extradition had the status of jus cogens, and therefore had to take priority over any claims of immunity under general international law.
39
See Rule 159 in Jean-Marie Henkaerts and Louise Doswald-Beck (eds.), Customary International Humanitarian Law, Volume I: Rules (Cambridge: CUP, 2005), at 611–4. 40 ILC Study Group on Fragmentation, Fragmentation of International Law: Difficulties Arising from the Diversification and Expansion of International Law, Report of the Study Group, finalized by Martti Koskenniemi, (13 April 2006) (A/6N.4/L.682) (hereinafter Final Report of the ILC Study Group): paras. 361–3 and 374; Jochen Frowein, ‘Jus Cogens’ in Encyclopaedia of Public International Law, Rudolf Bernhardt (ed.) (Amsterdam: North-Holland, 1997), Vol. 3, 65; See also Article 53 and 64 of the Vienna Convention on the Law of Treaties. 41 Final Report of the ILC Study Group, supra note 40. 42 Prosecutor v. Furundzija, Judgment of 10 December 1998 of Trial Chamber II, ICTY, (IT95-17/1), para. 153. 43 Regina v. Bow Street Metropolitan Stipendiary Magistrate and others exparte Pinochet Ugarte (No. 3), Judgment of 24 March 1999, House of Lords, ILR 119: 136.
344
phoebe okowa
The denial of legitimacy to blanket amnesties in these cases clearly represent an emerging state practice in support of subordination of any claims of immunity in cases where they conflict with obligations to prosecute under general international law.44 If the Lome agreement did impose obligations on the government of Sierra Leone as argued for by the defence, then these were inherently subordinate to peremptory norms in relation to international crimes. However, does the idea that the Lome agreement was in fact subordinate to jus cogens norms rest on a preliminary finding (even if implicit) that the accord was in fact an international instrument governed by the law of treaties. In the event the court was inclined to treat it as a wholly domestic instrument, with the result that it was incapable of creating any binding obligations on the government of Sierra Leone which had to be taken into account by a subsequent treaty. The suggestion that jus cogens norms are peremptory and override all conflicting treaty provisions is today regarded as incontestable. But if the immunity provisions were not dispositive, what effect did the finding of non-immunity have on the other obligations under the agreement? Could the immunity provisions be severed leaving the other provisions of the Lome intact and therefore binding. B. Other Legal Consequences It has been noted that the central tenet of the Court’s judgment is to support the thesis that the Lome agreement as an instrument between a state and national groups, was of no consequence on the international plane. At one level this is a curious conclusion; the instrument had been negotiated at the international level, and witnessed by no less than the Secretary General’s representative and other State representatives as moral guarantors. Purely domestic instruments of no consequence on the international plane do not as a general rule need the legal imprimatur or endorsement by third states or international agencies. There was a commitment on the part of all parties to disarm and engage with democratic processes, a matter of considerable concern to the United Nations and the regional guarantors in the fulfilment of their broad mandate to maintain international peace and security.45 Moreover, the commitments to representative government were a particular application of undertakings binding on the government on the government of Sierra Leone as a result of participation in other instruments.46 On one view a logical
44 Alexander Orakhelashvili, Peremptory Norms in International Law, (Oxford: OUP, 2006); John Dugard, ‘Possible Conflicts of Jurisdiction with Truth Commissions’, in The Rome Statute of the International Criminal Court-A Commentary, Antonio Cassese and Albin Eser (eds.) (Oxford: OUP, 2002), 693, at 698. 45 Lome Accord, supra note 12, Articles XVI and III respectively. 46 See in particular 1966 International Covenant on Civil and Political Rights, UNTS 999: 171; 1981 African Charter on Human and People’s Rights, ILM 21: 58.
interpretation and the special court for sierra leone
345
consequence of the decisions of the Special Court was to render all these related undertakings of no consequence on the international plane; after all the conclusion that the Lome accord was a domestic instrument of no consequence on the international level had not been qualified in parts. But could the Court’s purported invalidation of the amnesty provisions be taken as a once and for all statement on the status of such agreements in their entirety? The legality of the provisions of peace agreements were technically never in issue and the Court’s conclusions on the non-immunity provisions of the Lome accord did not affect the legality or otherwise of other provisions such as those calling for democratic representation in Sierra Leone’s political processes. On one view the Court was concerned with enforcement in a judicial setting, leaving open the possibility that even if a Court of law could not give effect to an agreement between a State and non -State groups this did not foreclose the issue of enforcement through political processes or other supervisory organs. In other words it is quite possible to read into the Court’s ruling that in general, it lacked manageable standards by which to assess the legality of a peace agreement. Treaty law which the court was purporting to apply in fact left open the possibility that agreements concluded between other subjects of international law were binding agreements. This conclusion is supported by reading both Articles 2 and 3 of the 1969 Vienna Convention on the law of treaties. On the face of it, Article 2 of the Vienna Convention on the face of it would seem to exclude peace agreements. For it defines a treaty as An international agreement concluded between States in written form and governed by international law, whether embodied in a single document or in two or more related instruments and whatever its particular designation.
But Article 3 of the treaty leaves open the possibility that such agreements may in fact be binding. It states that: The fact that the present Convention does not apply to international agreements concluded between states and other subjects of international law, or to international agreements not in written form shall not affect: (a) the legal force of such agreements (b) The application to them of any rules set forth in the present convention to which they would be subject under international law independently of the convention.
The provision clearly envisages the possibility that agreements concluded between states and non-state groups may be binding as a matter of international law However, outside the very narrow confines of international humanitarian law, there is little explicit recognition of non-state groups as subjects of international law.47 47
Liesbeth Zegveld, Accountability of Armed Opposition Groups in International Law (Cambridge: CUP, 2002), at 133 et seq.
346
phoebe okowa
The provisions of common Article 3 of the Geneva Conventions and the two 1977 Additional Protocols implicitly rest on the assumption that these entities have international personality and are capable of having rights and obligations on the international plane. Christine Bell has persuasively put forward the argument in support of international law as a plural legal order in which parallel normative regimes of variable legal quality co-exist. She writes: In the peace agreement context, however, pluralism arises not because of historical accident but because the very concepts of law and politics, domestic and international sphere, are being mutually constructed by the agreement. Only a mixed/political legal system can deal with this, and so a range of different order normative systems emerge with overlapping mandates and jurisdictions. The routes to implementation and development of agreement are multiple and various across. Political, legal, international, hard and soft law mechanisms.48
Bell’s argument in support of a lex pacificatoria clearly endows these peace agreements with legal quality. But she does this by accepting that their preeminently legal character is nonetheless not dependant on their being treated as equivalent to inter-state treaties but as legal documents on the international plane which are nevertheless binding. She makes the argument for a conceptual distinction between treaties and peace agreements and concedes that their enforceability will also not necessarily dependant on the same rules or legal processes.49 It has already been noted that under the terms of the Lome agreement, the government of Sierra Leone undertook to hold elections, disarm militia groups affiliated to it and commit to a programme of national reconciliation which included incorporating the RUF into a broad based government of national unity. Arguably those provisions mandating the government of Sierra Leone to hold elections, and incorporate the RUF into a broad-based national government of reconciliation are commitments on the international plane in respect of which both signatories could be held to account. It is true that the international court of justice has been particularly reticent to accept that undertakings given by a government to hold elections could be enforceable on the international plane. In its 1986 judgment in the Nicaragua Case, the International Court implicitly held that undertakings given by the Nicaragua government to the OAS, to hold elections and institute a program of reform aimed at the respect of human rights were a matter of domestic policy and inherently non-justiciable.50 The Court accepted that there was nothing in the enforceable content of international law to stop a state from making undertakings of this kind but it explicitly rejected the view that commitments 48
Bell, supra note 22, at 191. Id., at Chapters 6 and 7. 50 Case Concerning Military and Paramilitary Activities in and against Nicaragua (Nicaragua v. United States of America), Judgment of 27 June 1986, 1986 ICJ Rep. 14, para. 261. 49
interpretation and the special court for sierra leone
347
to hold elections were binding obligations on the international plane.51 This conclusion attracted a strong dissent from Judge Schwebel who observed that there was nothing to debar a state – or a revolutionary junta from making binding undertakings towards other states in respect of matters which would otherwise be within its exclusive jurisdiction. He regarded obligations undertaken by the Nicaraguan government to the OAS and its members, to govern in accordance with specified democratic standards and policies, including the inclusion of groups which had opposed the previous regime as fully justiciable and enforceable.52 In its judgment in DRC v. Uganda, the court again exhibited a marked reluctance to treat undertakings by the warring parties to the Congolese conflict under the terms of the Lusaka agreement as legally binding and therefore inherently justiciable.53 It would be precipitate to conclude from the Special Court’s decision on the immunity question, that it has settled once and for all the contested issue of the binding nature of peace agreements. The Court was not called upon and did not attempt to develop the general law in this area. Moreover, although the Nicaragua case cast doubt on the legal content of the commitment to hold elections, this had been a unilateral commitment by Nicaragua and not a bilateral or a multilateral one as in the context of a peace agreement. It is also possible to argue that the international lex lata has now moved to incorporate an entitlement of the population to democratic governance. The Congo case comes closest to a near outright rejection of the binding quality of a peace agreement but this too is implicit rather than explicit. V. The Question of Vires and the Charter of the United Nations Counsel for the defendants also raised constitutional arguments about the competence of the United Nations to conclude the agreement establishing the Special Court. These arguments in part mirrored those that had been raised before the Yugoslav and Rwanda tribunals on the competence of the United Nations to establish criminal tribunals in the exercise of its chapter VII powers.54 A more fundamental challenge raised the question whether the
51
Id., paras. 258–64. See Dissenting Opinion of Judge Schwebel, in Nicaragua Case, supra note 50, at 373, paras. 246–7. 53 Armed Activities on the Territory of the Congo (Democratic Republic of the Congo v. Uganda), Judgment of 19 December 2005, 2005 ICJ Rep. 1, paras 99–105; on which see Phoebe Okowa, ‘Congo’s War: The Legal Dimension of a Protracted Conflict’, BYIL 55 (2006): 203, at 235–8. 54 Prosecutor v. Dusko Tadic, Decision of 10 August 1995 of the Trial Chamber on the Defence Motion on Jurisdiction, ICTY, (IT-94-1-T); Prosecutor v. Dusko Tadic, Decision of 2 October 1995 of the Appeals Chamber on the Defence Motion for Interlocutory Appeal on Jurisdiction, ICTY, (IT-94-1-AR72); Prosecutor v. Slobodan Milosevic, Decision of 8 November 2001 of the 52
348
phoebe okowa
Security Council had the requisite powers under the Charter as a constitutive instrument to enter into the specific treaty that formed the basis of the Special Court with the government of Sierra Leone.55 Two related arguments were made.56 The first argument questioned the competence of the Security Council to delegate its treaty making powers to the Secretary General. The second proceeded on the premise that the creation of the Special Court by way of a bilateral treaty was implicitly ultra-vires the Charter, since a bilateral treaty could only be terminated with the mutual consent of both parties, the Security Council’s decision making power, it was suggested, was effectively subordinated to that of the government of Sierra Leone.57 Implicit in this was the argument that the powers conferred on the Security Council under Chapter VII, involved the exercise of a wide measure of discretion on whether a particular situation constituted a threat to the peace or had ceased to constitute such a threat. This required the retention for instance of a residual power to terminate the mandate of the Special Court should it reach the conclusion that the situation no longer constituted a threat to international peace and security, or at that the furtherance of the interests of the international community required the termination of the tribunal’s mandate, or its amendment. In other words, the power envisaged here was in the nature of a prerogative which the Security Council alone could exercise without subordinating its decision making powers to as it were to the wishes of another government. In so far as the “Security Council did not remain empowered to terminate the operation of a tribunal or amend the terms of the statute, the delegation of powers to the Special Court was illegal”.58 A related argument cited as further evidence in support of the contention that the Security Council had divested itself of its express and implied powers under Chapter VII, was in relation to the fact that the overall management and control of the Special Court was vested in a management committee, and although a representative of the Secretary General was part of the committee, Trial Chamber on Preliminary Motions, ICTY, (IT-02-54-PT); Prosecutor v. Joseph Kanyabashi, Decision of 18 June 1997 of Trial Chamber II on the Defence Motion on Jurisdiction, ICTR, (ICTR-96-15-T), para. 27; see William Schabas, The UN International Criminal Tribunals: The Former Yugoslavia, Rwanda and Sierra Leone (Cambridge: CUP, 2006), at 52–3. 55 Prosecutor v. Taylor, Decision of 31 May 2004 of the Appeals Chamber on Immunity from Jurisdiction, Special Court for Sierra Leone, (SCSL-2003-01-I-059), para. 37. 56 See Prosecutor v. Monina Fofana, Decision of 25 May 2004 of the Appeals Chamber on Preliminary Motion on Lack of Jursidiction – Illegal Delegation of Powers by the United Nations, Special Court for Sierra Leone, (SCSL-2004-14-PT-100-6836 & 6846), paras. 1, 5, 9–10. 57 Article 54 of the 1986 Vienna Convention on the Law of Treaties between States and International Organisations or Between International Organisations, (A/Conf.129/15). Article 22 and Article 23 of the Treaty establishing the Special Court provided that an amendment or termination of the treaty must be agreed upon by the parties. 58 Prosecutor v. Monina Fofana, supra note 56, para. 1.
interpretation and the special court for sierra leone
349
there was no provision as such for overall control or oversight by the security council as envisaged by Chapter VII.59 Thus in broad terms, the objection to the Court’s jurisdiction raised the question whether the Security Council’s Chapter VII Powers were reconcilable with the creation of an independent institution by way of treaty. In Prosecutor v. Moinina Fofana60 the Special Court rejected the argument that the delegation of the treaty making powers to the Secretary General was improper. It pointed that the Secretary General had no powers of his own to conclude an agreement but instead acted at the request of the Security Council in his capacity as executive organ.61 The argument that the treaty entered into between the governments of Sierra Leone was ultra vires the Charter as a constitutive instrument was also decisively rejected. The Court pointed out that in Resolution 1315, the Security Council had established that in accordance with Article 39 of the UN Charter the situation in Sierra Leone constituted a threat to international peace and security and that the creation of the Court was an effective collective measure for removing threats to the peace without acting in a coercive manner.62 The Court noted that given the nature of the tribunal as a judicial organ it had to retain a measure of overall independence. External control could therefore only be exercised in relation to policy and administrative matters. In the case of the Sierra Leone Court, the necessity for such control had been envisaged and provided for under Chapter VII of the constitutive instrument. Such necessary control was exercised through the management committee established under Chapter VII of the tribunal. The Court noted that at the time when the agreement was concluded, the Government of Sierra Leone, as member state of the UN was no doubt aware of the SC broad mandate in relation to the maintenance of international peace and security Therefore a unilateral retreat from the treaty obligation by the Security Council would not constitute a breach of good faith in relation to the agreement.63 In the case of a conflict between obligations under the bilateral agreement and the maintenance of international peace and security, the Security Council would act in the fulfilment of a higher ranking obligation under the Charter of the UN.64 The agreement could not therefore override the mandate of the Security Council under Chapter VII of the Charter.
59 See Phakiso Mochochoko and Giorgia Tortora, ‘The Management Committee for the Special Court for Sierra Leone’, in Internationalised Criminal Courts and Tribunals: Sierra Leane, East Timor, Kosovo and Cambodia, Cesare Romano, André Nollkaemper and Jann Kleffner (eds.), (Oxford: OUP, 2004), 141. 60 Prosecutor v. Monina Fofana, supra note 56. 61 Id., paras. 16–7. 62 Id., 22–9. 63 Id., para. 28. 64 Id., para. 27.
350
phoebe okowa
Elsewhere, Dan Sarooshi has noted that it is not necessarily inconsistent with the Charter for the Security Council to create an organ to exercise a power which the Council itself does not possess, and for the Council to be legally bound by the resulting decisions.65 To deny such a power would be to implicitly nullify its authority as a judicial body, making it impossible for it to discharge the functions for which it was created. In his Separate opinion Judge Robertson implicitly conceded that the resulting relationship between the two treaty partners was probably an unequal, but this fact could not have been lost on the government of Sierra Leone as a member of the UN, and under Article 25 of the Charter, it would be obliged to accept and carry out any decisions taken by the Security Council even if this involved terminating the treaty.66 This part of the judgment is important for it unequivocally confirms that in those situations where the Security Council’s sets up an independent organisation to carry out its Chapter VII powers, the presumption remains that the Security Council retains overall control in the fulfilment of its mandate, and the exercise of those may not only necessitate the variation of the mandate of the institution but even its abolition if the requirements of international peace and security dictate.67 There are however limitations to the Security Council’s powers of oversight if criminal tribunals are to carry out their functions as independent judicial bodies, free from political control. For a start it is implicit that the Council will be bound by determinations of the judicial body, and therefore it is not necessarily inconsistent that the Council should create a body and be subject to its authoritative determinations.68 Second even where its powers have been varied or terminated, this cannot affect the binding nature or authority of decisions already given.69 VI. Interpretation and Amendment by Subsequent Practice There has long been a presumption that treaties including those emanating from specialist regimes operate within the broad normative framework of public international law. As a result the general rules of that system including
65
Sarooshi, supra note 4, at 103–5. Prosecutor v. Monina Fofana, supra note 56, Separate Opinion of Judge Robertson, para. 6. 67 Sarooshi, supra note 4, at 104. 68 The ICJ held as much in: Effect of Awards of Compensation Made by ;the United Nations Administrative Tribunal, Advisory Opinion of 13 July 1954, 1954 ICJ Rep. 47, at 61; Prosecutor v. Dusko Tadic, supra note 54, para. 15; Prosecutor v. Tihomir Blaskic, Decision of 18 July 1997 of Trial Chamber II on the Objection of the Republic of Croatia to the Issuance of Subpoena Duces Tecum, ICTY, (IT-95-14), at 11; José Álvarez, ‘Judging the Security Council’, AJIL 90 (1996): 1, at 11. 69 Sarooshi, supra note 4, at 105. 66
interpretation and the special court for sierra leone
351
those of treaty interpretation supplement and validate the lex specialis provisions of those international instruments.70 Yet the application of the general rules such as that contained in Article 31(3)(c) which provides that a treaty is to be interpreted in context, taking into account the broad context in which it was concluded, present particular problems in the field of criminal liability. It is difficult to conceive of how an account of the subsequent practice can be applied without in some form conflicting with other lex specialis rules of interpretation in the criminal law field, in particular the nullum crimen sine lege principle. The principle has been expressly endorsed in the Universal Declaration of Human Rights which provides that “No one shall be held guilty of any penal offence on account of any act or omission which did not constitute a penal offence under national or international law at the time when it was committed”.71 The significance of this principle was acknowledged by the UN Secretary General at the time of the establishment of the International Criminal tribunal for Yugoslavia. The ICTY was only authorised to apply those rules that were conclusively accepted as part of general international law at the time the crimes were committed.72 Like Security Council tribunals on Yugoslavia and Rwanda, the Statute of the Special Court does not give a detailed content of the substantive law to be applied. In Article 1 of the Statute of the Special Court, it is provided that the Court shall have the power to prosecute persons who bear the greatest responsibility for serious violations of international humanitarian law and Sierra Leone Law committed in the territory of Sierra Leone since 30 November 1996.
The substantive crimes deemed binding on the government of Sierra Leone at the beginning of the Court’s temporal jurisdiction ere enumerated in broad outline in Articles 2–4. However, the detailed content was to be determined by way of renvoi to treaty instruments and customary law that were by binding on Sierra Leone in 1996. This method of ascertaining the applicable substantive law raised the question whether subsequent treaties and developments in state practice could be employed in determining both the scope and the interpretation of the applicable substantive law. Briefly, Article 2 endows the Special Court with jurisdiction to punish crimes against humanity and article 3 permitted prosecution of those alleged to have committed or ordered serious violations of Common Article 3 of the Geneva Conventions and AP II. In article 4 a broad generic category was also 70
Final Report of the ILC Study Group, supra note 40, Section C. 1948 Universal Declaration of Human Rights (A/810): Art. 11(2). 72 UN Secretary-General, 1993 Report of the Secretary-General pursuant to para 2 of Security Council Resolution 808, (S/25704); On the application of the non-retroactivity principle see Prosecutor v. Furundzija, supra note 42, para. 14. 71
352
phoebe okowa
included: the Court was to have jurisdiction over other serious violations of international humanitarian law including the offence of conscripting or enlisting children under the age of 15 into the armed forces or using them to participate actively in hostilities. The last category was challenged on the basis that none of the treaty instruments, or customary law binding on Sierra Leone in 1996 had contained a regime of criminal responsibility for recruitment and enlisting of child soldiers.73 True there were a number of treaty instruments enjoining states to prohibit the recruitment of children and put in their place regimes for the protection of children. These included the 1949 GC and its two Protocols, the 1989 Convention on the Rights of the Child and the African Charter on the Rights and Welfare of the Child. However, the language contained in these instruments as well as the institutions for their implementation clearly indicated that these were obligations of States subject to the normal processes of delictual responsibility and monitoring of compliance but not individual criminal liability. In Prosecutor v. Norman, the Prosecution maintained that these treaty instruments when examined in the context of subsequent developments in international law especially Article 8 of the Statute of the International Criminal Court and its impetus on implementing legislation, demonstrated an intention on the part of the international community to create a regime of criminal liability.74 This was contested by the defence, who argued that neither general international law nor treaties binding on Sierra Leone in 1996 had criminalized the recruitment or enlisting of child soldiers. Implicit in the defence argument was a further contention that such modification would in any case be inapplicable as it was manifestly contrary to principle of nonretroactivity of penal laws. Although not expressed in those terms, at issue was whether the customary law processes for treaty modification, in particular through subsequent developments in state practice was applicable to regimes of criminal responsibility. In the Gabcikovo-Nagymaros Case,75 the international Court had stated as a general principle that treaties were not ossified but had to be interpreted in light of subsequent developments in international law. The prosecutor and amicus brief presented by the University of Toronto and UNICEF, argued that the wide acceptance of criminal responsibility in the Rome Statute and the implementing legislation provided evidence that international law in the form of custom had criminalized such conduct. That as a result of the wide ratification of the Rome Statute and its impact, customary 73 Prosecutor v. Norman, Decision of 31 May 2004 of the Appeals Chamber on Preliminary Motion Based on Lack of Jurisdiction (Child Recruitment), Special Court for Sierra Leone, (SCSL-2004-14-AR72(E)-131-7383, 7398, 7413 & 7430), para. 1. 74 Id., para. 2. 75 Gabcikovo-Nagymaros Project (Hungary v. Slovakia), Judgment of 25 September 1997, 1997 ICJ Rep. 7, para. 112.
interpretation and the special court for sierra leone
353
law applicable to the recruitment and enlisting of child soldiers had been modified to provide for a regime of individual criminal responsibility. The court summarily concluded that the treaty instruments binding on Sierra Leone in 1996 envisaged a regime of criminal responsibility without engaging in any meaningful way with these contested positions. There was incontrovertible evidence before the Court and elsewhere, that the drafting history of article 4 did not conclusively support the thesis that the provision was merely incorporating an accepted customary law duty76. Moreover, scholarly opinion is almost unanimous, that the crime of “recruiting and enlisting children” in Article 8 of the ICC statute was new law with no counterpart as such in general international law.77 It was left to Judge Robertson in his dissenting opinion to outline why treaty amendment or interpretation through subsequent practice raised problems of their own when applied to instruments imposing criminal liability. First, he observed that the treaty obligations in question were general and aspirational in character which required no more than a prohibition and due diligence in protecting children as vulnerable categories.78 Central to his argument was the nullem crimen principle. He noted that it was only after the Rome Statute was concluded and adopted by 122 states in July 1998 that one could realistically say that customary law criminalized the recruitment and enlisting of children.79 The Rome Statute and its impact on national laws could not be relied on to retroactively criminalise conduct which although abhorrent had not specifically been defined as a crime at the time of its commission.80 It is regrettable that the Court missed the opportunity to state with some clarity whether the general rules on treaty interpretation contained in article 31 of the Vienna Convention in particular 31(3)(c) was equally applicable to the interpretation of treaties imposing criminal liability. The relationship between the general principles of treaty interpretation contained in the Vienna Convention and the principle of specificity (law attracting penal consequences must be capable of reasonable ascertainment) as well as the non-retroactivity of penal provisions also need clarification.81 It is suggested that the nullem crimen 76 As Judge Robertson points out the initial draft of the Secretary General had provided for the crime of conscripting and abducting children, a proposition that arguably enjoyed a much broader support in state practice. The amendment to include recruitment and enlisting was apparently introduced by the President of the Security Council without providing any evidence that customary international law had criminalized such conduct. Prosecutor v. Norman, supra note 73, Dissenting Opinion of Judge Robertson. 77 Schabas, supra note 54, at 285. 78 Dissenting Opinion of Judge Robertson, supra note 76, paras. 24–5. 79 Id., paras. 31–8. 80 Id., para. 45. 81 Article 15 of the International Covenant on Civil and Political Rights provides that:
No one shall be held guilty of any criminal offence on account of any act or omission which did not constitute a criminal offence under national or international law at the time it was committed.
354
phoebe okowa
principle as a specific rule of treaty interpretation should override the general rules of interpretation contained in Article 31(3)(c) of the Vienna Convention. VII. The Treaty’s Unconstitutionality under Sierra Leone Constitution The constitutionality of the Special Court was also challenged on the grounds that the government of Sierra Leone had failed to comply with domestic constitutional law requirements,82 including a requirement to hold a referendum before establishing new judicial institutions; that as a result the constitutive treaty was invalid as “the violation was manifest and concerned a rule of its internal law of fundamental importance”.83 It was argued that the S. 108 of the constitution of Sierra Leone declared that the Supreme Court of Sierra Leone to be the final Court of Appeal and enjoined the government to hold a referendum before establishing other tribunals or altering this judicial framework and court structure in Sierra Leone. Citing article 46 of the Vienna Convention, it was argued that the bilateral agreement between the UN and the government of Sierra Leone was invalid in so far it was concluded in disregard of a fundamental principle of municipal law. The correctness of the defence argument was in any case rejected by the international court in its decision in Qatar v. Bahrain.84 The competence of the government of Sierra Leone to conclude the agreement was also challenged on the grounds that at the time of its conclusion, it controlled only one third of the territory, the rest being under control of two rebel movements, the RUF and the AFRC.85 On this basis, it was argued that in the face of such limited territorial control, president Kabbah’s government’s claim to be the government of the state was clearly questionable, and as a consequence its competence to conclude the treaty also in doubt. In its judgment, the Court denied that the Special Court was part of the judiciary of Sierra Leone.86 It had been set up as an international institution with its own powers including treaty making powers and not as part of Sierra See also Article 7 of the 1950 ECHR, UNTS 213: 222; Article 9 of the 1969 American Convention on Human Rights, UNTS 1144: 123; and Article 7(2) of the 1981 African Charter of Human and People’s Rights, UNTS 1520: 217. 82 Prosecutor v. Morris Kallon, Prosecutor v. Sam Hinga Norman and Prosecutor v. Brima Bazzy Kamara, Decison of 13 March 2004 of the Appeals Chamber on Constitutionality and Lack of Jurisdiction, Special Court for Sierra Leone, (SCSL-2004-15-PT-059-I & II; SCSL-200414-PT-035-I & II & SCSL-2004-16-PT-032-I & II), para. 8. 83 Citing Article 46 of the Vienna Convention on the Law of Treaties. 84 Case of Maritime Delimitation and Territorial Questions between Qatar and Bahrain (Qatar v. Bahrain), Jurisdiction and Admissibility, Judgment of 1 July 1994, 1994 ICJ Rep. 112, at 121–2. 85 Decision on Constitutionality and lack of Jurisdiction, supra note 82, para. 12. 86 Id., paras. 49–52.
interpretation and the special court for sierra leone
355
Leone’s domestic Court structure. Furthermore, the Court noted that Article 8 of the Statute of the Special Court established its primacy over the national courts of Sierra Leone. As a result, it concluded, that the presumed need to comply with domestic law procedures before new tribunals are set up was clearly inapplicable.87 In relation to the argument that the treaty was invalid on account of being entered into by a government that lacked the requisite control over territory, the court citing Sierra Leone Telecommunications Co. Ltd v. Barclays Bank PLC88 adopted the orthodox position in favour of the presumed legality of the de facto constitutionally elected government of a state.89 The judgment therefore does not deal with the question whether there are situations where a government’s access to public power is deeply contested, with the result that its claim to speak for the whole is in itself in doubt.90 If the test of governmental authority, as a matter of general international law remains effective control and not constitutional legitimacy, then clearly there is a case for arguing that where that control is in itself in issue, then the treating making competence of the government should rightly be called into question. VIII. Conclusions The Sierra Leone cases have demonstrated the significant role played by the general framework of treaty law in interpreting and applying the lex specialis provisions of the constitutive instruments of international criminal tribunals. The cases have confirmed in a practical context that the fragmentation of international law need not necessarily threaten its coherence as a normative system. To that extent they confirm the ILC’s own conclusion that the proliferation of sub-systems within international law is not inimical to the continuing vitality of international law as a unified legal order. The cases however have also demonstrated the significant limitations of a specialised tribunal to deal with large questions of international law which may arise by way of an incidental issue to the main subject matter of the dispute. A large number of the treaty questions that the Special Court had to deal with were clearly uncontroversial, and it is therefore not surprising that they were dealt in a summary manner. However, the cases have also illustrated areas of tension between international criminal justice as lex specialis and other general rules of customary international law.
87 88 89 90
Id., para. 53. Sierra Leone Telecommunications Co. Ltd v. Barclays Bank PLC (1998), All ER 2: 821. Decision on Constitutionality and Lack of Jurisdiction, supra note 82, paras. 75–9. See generally Roth, supra note 29, at 252 et seq.
BIBLIOGRAPHY Books & Chapters in Books Advisory Committee of Jurists, Procès Verbaux of the Meetings of the Advisory Committee of Jurists: 16 June – 24 July 1920 with Annexes (The Hague: van Langenhuysen, 1920). Bos, Maarten, A Methodology of International Law (Amsterdam: North Holland, 1984). Schwarzenberger, Georg, International Law as Applied by International Courts and Tribunals: Vol 1 (London: Stevens, 1957). Andenas, Mads, (ed.), Article 177 References to the European Court – Policy and Practice (London: Butterworths, 1994). Anderson, David, References to the European Court (London: Sweet and Maxwell, 1995). Arangio-Ruiz, Gaetano, The United Nations Declaration on Friendly Relations and the System of the Sources of International Law (Alpheen aan Rijn: Sijthoff & Noordhoff, 1979). Arsanjani, Mahnoush H., ‘Reflections on the Jurisdiction and Trigger Mechanism of the International Criminal Court’, in in Reflections on the International Criminal Court, Herman A. M. von Hebel et al. (eds.) (The Hague: TMC Asser Press, 1999), 57. Ashton, Peter, ‘South African Water Conflicts: are they inevitable or preventable?’ in Water for Peace in the Middles East and South Africa, Green Cross International (ed.) (Geneva: Green Cross International, 2000), 94. Aust, Anthony, Modern Treaty Law and Practice (Cambridge: CUP, 2007). Aust, Anthony, Modern Treaty Law and Practice (Cambridge: CUP, 2000). Badie, Bertrand and Marie-Claude Smouts, Le Retournement du Monde. Sociologie de la Scene Internationale (Paris: Presses de la Fondation Nationale de Sciences Politiques, 1995). Barksdale, Clement, trans., The Illustrious Hugo Grotius of the Law of Warre and Peace with Annotations. III Parts and Memorials of the Author’s Life and Death, (London: printed by T. Warren, for William Lee, and are to be sold at his shop at the signe of the Turks-head in Fleet-Street, 1654). Bassiouni, Cherif (ed.), The Legislative History of the International Criminal Court: Summary Records of the 1998 Diplomatic Conference (New York: Transnational Publishers, 2005). Bederman, David, Classical Canons: Rhetoric, Classicism and Treaty Interpretation (Aldershot: Ashgate, 2001). Bell, Christine, On the Law of Peace, Peace Agreements and the Lex Pacificatoria (Oxford: OUP, 2008). Bell, Christine, Peace Agreements and Human Rights (Oxford: OUP, 2000). Berman, Franklin, ‘The Relationship between the International Criminal Court and the Security Council’, in Reflections on the International Criminal Court, Herman A. M. von Hebel et al. (eds.) (The Hague: TMC Asser Press, 1999), 173. Blischenko, Igor and Jose Doria, Precedents in International Public and Private Law, (Moscow, 1999). Bork, Robert H., The Tempting of America (New York, NY: Free Press, 1989). Bos, Maarten, A Methodology of International Law (Oxford: North-Holland, 1984). Bourgon, Stephane, ‘Jurisdiction Ratione Temporis’, in The Rome Statute of the International Criminal Court: A Commentary, Antonio Cassese et al. (eds.) (Oxford: OUP, 2002), 543. Boyle, Alan and Chinkin, Christine, The Making of International Law (Oxford: OUP, 2007). Bradbury, Ray, Fahrenheit 451 (New York, NY: Harper Collins Publishers, 2004). Bredimas, Anna, Methods of Interpretation and Community Law (Oxford: North-Holland Publishing Co., 1978). Brownlie, Ian, Principles of Public International Law (Oxford: Clarendon Press, 1990). Brownlie, Ian, Principles of Public International Law (Oxford: OUP, 1998). Burton, Steven J., Judging in Good Faith (Cambridge: CUP, 1992). Capotorti, Francesco, ‘Sul Valore della Prassi Applicativa dei Trattati Secondo la Convenzione di Vienna’, in Il Diritto Internazionale al Tempo della sua Codificazione: Studi in onore di Roberto Ago, Roberto Ago (ed.) (Milano: Giuffrè, 1987), vol. I, 197.
358
bibliography
Chang, Yi-Ting, The Interpretation of Treaties by Judicial Tribunals (New York: Columbia University Press, 1933). Charles R. Beitz, Political Theory and International Relations (Princeton, NJ: Princeton University Press, 1999, with afterword). Chigara, Benedict, Amnesty in International Law: The Legality Under International Law of National Amnesty Laws (Harlow: Logman, 2002). Condorelli, Luigi and Santiago Villalpando, ‘Can the Security Council Extend the ICC’S Jurisdiction?’, in The Rome Statute of the International Criminal Court: A Commentary, Antonio Cassese et al. (eds.) (Oxford: OUP, 2002), 571. Condorelli, Luigi and Santiago Villalpando, ‘Relationship of the Court with the United Nations’, in The Rome Statute of the International Criminal Court: A Commentary, Antonio Cassese et al. (eds.) (Oxford: OUP, 2002), 219. Coppée, Henry, Elements of Logic (New York: American Book Company, 1872). Corten, Olivier, Les Conventions de Vienna sur le Droit des Traités: Commentaire Article par Article (Bruxelles: Bruylant, 2006). Cotterell, Roger, The Politics of Jurisprudence: A Critical Introduction to Legal Philosophy (London: Butterworths, 1989). Cottier, Thomas, ‘The WTO and Environmental Law: Three Points for Discussion’, in Trade and Environment: Bridging the Gap, Agatha Fijolkowski and James Cameron (eds.) (London: Cameron May, 1998), 56. Crawford, James, ‘The Work of the International Law Commission’, in The Rome Statute of the International Criminal Court: A Commentary, Vol. I, Antonio Cassese et al. (eds.) (Oxford: OUP, 2002), 23. Crisp, Roger and Michael Slote, Virtue Ethics (Oxford: OUP, 1997). Danilenko, Gennadij M., Law-Making in the International Community (Dordrecht: Nijhoff, 1993). Dashwood, Alan and Angus Johnston, (eds.), The Future of the Judicial System of the European Union (Oxford: Hart Publications, 2001). de Búrca, Gráinne, G. and Joseph H. H. Weiler, (eds.), The European Court of Justice (Oxford: OUP, 2001). de Vattel, Emmerich, The Law of Nations (New York: AMS Press, reprint of 1863 edition). de Vattel, Emmerich, The Law of Nations; or Principles on the Law of Nature: Applied to the Conduct and Affairs of Nations and Sovereigns, (London, 1702). de Vattel, Emmerich, The Law of Nations; or the Principles of the Law of Nature: Applied to the Conduct and Affairs of Nations and Sovereigns (London: G.G.J & J. Robinson, 1792). Degan, Vladimir D., L’Interprétation des Accords en Droit International (The Hague: Martinus Nijhoff, 1963). Donald W. Greig, International Law (London: Butterworth, 1976). Dugard, John, ‘Possible Conflicts of Jurisdiction with Truth Commissions’, in The Rome Statute of the International Criminal Court-A Commentary, Antonio Cassese and Albin Eser (eds.) (Oxford: OUP, 2002), 693. Durling, James P. and Nicely, Matthew R., Understanding the WTO Anti-Dumping Agreement: Negotiating History and Subsequent Interpretation (London: Cameron May, 2002). Dworkin, Ronald, ‘Comment’, in A Matter of Interpretation, Antonin Scalia and Amy Gutmann (Princeton, NJ: Princeton University Press, 1997), 115. Dworkin, Ronald, ‘The Forum of Principle’ in A Matter of Principle, Ronald Dworkin (Cambridge, MS: Harvard University Press, 1985), Ch. 2. Dworkin, Ronald, A Matter of Principle (Cambridge, MS: Harvard University Press, 1985). Dworkin, Ronald, Life’s Dominion (New York, NY: Vintage, 1993). Engelen, Franciscus A., Interpretation of Tax Treaties under International Law: A Study of Articles 31, 32 and 33 of the Vienna Convention on the Law of Treaties and their Application to Tax Treaties (Amsterdam: IBFD, 2004). Fischer-Lescano, Andreas and Gunther Teubner, Regime-Kollisionen: Zur Fragmentierung des globalen Rechts (Frankfurt am Main: Suhrkamp, 2006). Fish, Stanley, Doing What Comes Naturally: Change, Rhetoric, and the Practice of Theory in Literary and Legal Studies (Oxford: Clarendon Press, 1989). Fitzmaurice, Gerald, ‘Some Problems Regarding the Formal Sources of International Law’, in Symbolae Verzijl: présentées au Professeur J. H. Verzijl à l’ Occasion de son LXX-ième
bibliography
359
Anniversaire, Jan H. Verzijl and Frederik M. van Asbeck (eds.) (The Hague: Nijhoff, 1958), 153. Frowein, Jochen, ‘Jus Cogens’ in Encyclopaedia of Public International Law, Rudolf Bernhardt (ed.) (Amsterdam: North-Holland, 1997), Vol. 3, 65. Fuller, Lon L., The Morality of Law (New Haven, CT: Yale University Press, 1964). Fuller, Lon L., The Morality of Law, (New Haven, CT: Yale University Press, 1969). Gaillard, Emmanuel, ‘Investment Treaty Arbitration and Jurisdiction Over Contract Claims – the SGS Cases Considered’, in International Investment Law and Arbitration, Tod Weiler (ed.) (London: Cameron May, 2005), 325. Gaja, Giorgio, ‘Does the European Court of Human Rights use its stated methods of interpretation?’, in Divenire Sociale e Adeguamento del Diritto: Studi in Onore di Francesco Capotorti, Francesco Capotorti (ed.) (Milano: Giuffré, 1999), 213. Gardiner, Richard, Treaty Interpretation (Oxford: OUP, 2008). Gounelle, Max, La Motivation Des Actes Juridiques En Droit International Public: Contribution a Une Theorie de l’ Acte Juridique en Droit International Public (Paris: Pedone, 1979), 79. Greig, Donald, Intertemporality and the Law of Treaties (London: BIICL, 2001). Grotius, Hugo, De Jure Belli ac Pacis [On the Law of War and Peace] (1625). Grotius, Hugo, On the Law of War and Peace, (Oxford: Clarendon Press, 1925). Haraszti, Gyorgy, Some Fundamental Problems in the Law of Treaties (Budapest: Akadémiai Kiadó, 1973). Hart, Herbert L.A., The Concept of Law (Oxford: Clarendon Press, 1961). Henkaerts, Jean-Marie and Doswald-Beck, Louise (eds.), Customary International Humanitarian Law, Volume I: Rules (Cambridge: CUP, 2005). Higgins, Rosalyn, ‘Some Observations on the Inter-Temporal Rule in International Law’, in Theory of International Law at the Threshold of the 21st Century, Jerzy Makarczyk and Krzysztof Skubiszewski (eds.) (The Hague: Kluwer Law International, 1996), 173. Higgins, Rosalyn, ‘The Relationship between the International Criminal Court and the International Court of Justice’, in Reflections on the International Criminal Court, Herman A. M. von Hebel et al. (eds.) (The Hague: TMC Asser Press, 1999), 163. Honig, Bonnie, Democracy and the Foreigner (Princeton, NJ: Princeton University Press, 2001). Hornsby, Albert S., Oxford Advanced Learner’s Dictionary (Oxford: OUP, 1995). ICSID, History of the ICSID Convention. Documents Concerning the Origin and the Formulation of the Convention on the Settlement of Investment Disputes between States and Nationals of Other States (Washington, DC: ICSID, 1968, reprinted in 2001). Jennings, Ivor W., The Law and the Constitution, (London: University of London Press, 1943). Jennings, Robert and Arthur Watts (eds.), Oppenheim’s International Law (London: Longman, 1992). Jennings, Robert and Watts, Arthur (eds.), Oppenheim’s International Law: Vol. I Parts 2–4 (London: Longman, 1992, 9th edition). Kaufmann-Kohler, Gabrielle, ‘Annulment of ICSID Awards in Contract and Treaty Arbitrations: Are there Differences?’, in Annulment of ICSID Awards, Emmanuel Gaillard and Yas Banifatemi (eds.) (New York, NY: Juris Publishing, 2004), 189. Kaul, Hans-Peter, ‘Precondition to the Exercise of Jurisdiction’, in The Rome Statute of the International Criminal Court: A Commentary, Antonio Cassese et al. (eds.) (Oxford: OUP, 2002), 583. Kelsen, Hans, Introduction to the Problems of Legal Theory (Pure Theory of Law) (Oxford: Clarendon Press, 1992). Klabbers, Jan, ‘Reluctant Grundnormen: Articles 31(3)(c) and 42 of the Vienna Convention on the Law of Treaties and the Fragmentation of International Law’, in Time, History and International Law, Matthew Craven, Malgosia Fitzmaurice and Maria Vogiatzi (eds.) (Leiden: Martinus Nijhoff, 2007), 141. Koskenniemi, Martti, Fragmentation of International Law: Difficulties arising from the Diversification and Expansion of International Law. Report of the Study Group of the International Law Commission (Helsinki: Erik Castrén Institute, 2007). Koskenniemi, Martti, From Apology to Utopia: The Structure of International Legal Argument (Cambridge: CUP, 2005). Koskenniemi, Martti, From Apology to Utopia: The Structure of International Legal Argument (Cambridge: CUP, 1989).
360
bibliography
Kuosmanen, Taru, Bringing Justice Close: Hybrid Courts in Post-Conflict Societies (Helsinki: Erik Castrén Institute, 2007). Kutscher, Hans, ‘Methods of Interpretation as seen by a Judge at the Court of Justice’, in Reports of a Judicial and Academic Conference held in Luxembour in 27–28 September 1976 (Luxembourg: Court of Justice of the European Communities, 1976), 1. Legum, Barton, ‘The Introduction of an Appellate Mechanism: the U.S. Trade Act of 2002’, in Annulment of ICSID Awards, Emmanuel Gaillard and Yas Banifatemi (eds.) (New York, NY: Juris Publishing, 2004), 289. Letsas, George, A Theory of Interpretation of the European Convention on Human Rights (Oxford: OUP, 2007). Linderfalk, Ulf, On the Interpretation of Treaties: The Modern International Law as Expressed in the 1969 Vienna Convention on the Law of Treaties (Dordrecht: Springer, 2007). Luhmann, Niklas, Law as a Social System (Oxford: OUP, 2004). Madison, James, ‘The Federalist No. 51’, in The Federalist Papers, Alexander Hamilton, James Madison and John Jay, (New York: Bantam Books, 1982), 261. Marcus Tullius Cicero, De Divinatione, Book II. Marcus Tullius Cicero, De Officiis, Book I. McDougall, Myres S., Harold Lasswell, and James C. Miller, The Interpretation of International Agreements and World Public Order (New Haven: New Haven Press, 1994). McNair, Arnold, ‘Treaties and Sovereignty’, in Symbolae Verzijl: presentees au Professeur J. H. W. Verzijl a l’ occasion de son LXXieme anniversaire, Jan H. W. Verzijl and Frederik M. Van Asberck (eds.) (The Hague: Nijhoff, 1958), 222. McNair, Arnold, The Law of Treaties (Oxford: Clarendon Press, 1961). McRae, Donald, ‘Approaches to the Interpretation of Treaties: the European Court of Human Rights and the WTO Appellate Body’, in Human rights, Democracy and the Rule of Law: Liber Amicorum: Luzius Wildhaber, Luzius Wildhaber and Stephan Breitenmoser (eds.) (Zürich: Dike, 2007), 1407. Mochochoko, Phakiso and Tortora, Giorgia, ‘The Management Committee for the Special Court for Sierra Leone’, in Internationalised Criminal Courts and Tribunals: Sierra Leane, East Timor, Kosovo and Cambodia, Cesare Romano, André Nollkaemper and Jann Kleffner (eds.), (Oxford: OUP, 2004), 141. Oppenheim, Lassa F. L., Robert Y. Jennings, and Arthur D. Watts (eds.), Oppenheim’s International Law 9th ed. (London: Longman, 1996). Orakhelashvili, Alexander, Peremptory Norms in International Law, (Oxford: OUP, 2006). Orakhelashvili, Alexander, The Interpretation of Acts and Rules in Public International Law (Oxford: OUP, 2008). Pauwelyn, Joost, Conflict of Norms in Public International Law (Cambridge: CUP, 2003). Polyaenus, Stratagems, Book III. Polybius, Histories, Book III & XII. Powell, Jefferson, Constitutional Conscience: The Moral Dimension of Judicial Decision (Chicago, IL: University of Chicago Press, 2008). Prudhomme, Auguste, La Loi Territoriale et les Traités Diplomatiques devant les Juridictions des États Contractants (Paris: LGDJ, 1910). Quintus Ennius, Annales (Fragmenta), Book VI. Radbruch, Gustav, Einführung in die Rechtswissenschaft, (Stuttgart: Koehler Verlag, 1980). Ratner, Steven, Abrams, Jason and Bischoff, James, Accountability for Human Rights Atrocities in International Law: Beyond the Nuremberg Legacy (Oxford: OUP, 2009, 3rd edition). Reuter, Paul, An Introduction to the Law of Treaties, (London: Pinter, 1989). Reuter, Paul, Introduction to the Law of Treaties (London: Kegan Paul International, 1995, trans. by José Mico and Peter Haggernacher). Robinson, Patrick, ‘The Missing Crimes’, in The Rome Statute of the International Criminal Court: A Commentary, Antonio Cassese et al. (eds.) (Oxford: OUP, 2002), 497. Roht-Arriaza, Naomi, ‘Sources in International Law of Treaties of an Obligation to investigate, Prosecute and Provide Redress’, in Impunity and Human Rights in International Law and Practice, Naomi Roht-Arriaza (ed.) (New York, NY.: OUP, 1995), 24. Romano, Cesare P.R., Nollkaemper, André and Kleffner, Jann K. (eds.), Internationalized Criminal Courts, Sierra Leone, East Timor, Kosovo and Cambodia (Oxford: OUP 2004). Rosenne, Shabtai, ‘Conceptualism as a Guide to Treaty Interpretation’, reprinted in Shabtai Rosenne, An International Law Miscellany (Dordrecht: Martinus Nijhoff, 1993), 441.
bibliography
361
Rosenne, Shabtai, Breach of Treaty (Cambridge: Grotius, 1985). Roth, Brad, Governmental Illegitimacy in International Law (Oxford: Clarendon Press, 1999). Rousseau, Charles, Droit Itnernational Public, Tôme I: Introduction et Sources (Paris: Dalloz, 1970). Russell, Bertand, ‘Introduction’, in Tractatus Logico-Philosophicus, Ludwig Wittgenstein (London: Routledge Classics, 2005), ix. Sarooshi, Danesh, The United Nations and the Development of Collective Security: The Delegation by the UN Security Council of its Chapter VII Powers (Oxford: OUP, 1999). Satow, Ernest, A Guide to Diplomatic Practice (London: Longmans, 1917). Scalia, Antonin and Amy Gutmann, A Matter of Interpretation (Princeton, NJ: Princeton University Press, 1997). Scalia, Antonin, ‘Common Law Courts in a Civil Law System’, in A Matter of Interpretation, Antonin Scalia and Amy Gutmann (Princeton, NJ: Princeton University Press, 1997), 3. Schabas, William, The UN International Criminal Tribunals: The Former Yugoslavia, Rwanda and Sierra Leone (Cambridge: CUP, 2006). Schauer, Frederick, Playing by the Rules: A Philosophical Examination of Rule-based Decisionmaking in Law and in Life (Oxford: Clarendon Press, 1991). Schwarzenberger, Georg, International Law, Vol. I: International Law as Applied by International Courts and Tribunals (London: Stevens, 1957). Schwebel, Stephen M., ‘May Preparatory Work be Used to Correct rather than Confirm the ‘Clear’ Meaning of a Treaty Provision?’, in Theory of International Law at the Threshold of the 21st Century, Makarczyk, Jerzy, (ed.) (The Hague: Kluwer Law International, 1996), 541. Shahabuddeen, Mohamed, Precedent in the World Court (Cambridge: CUP, 1996). Sinclair, Ian, The Vienna Convention and the Law of Treaties, (Manchester: MUP, 1984). Société Française pour le Droit International, La juridictionalisation du Droit International: Colloque de Lille (Paris: Pedone, 2003). Sorel, Jean-Marc, ‘Article 31 – Convention de 1969’, in Les Conventions de Vienne sur le Droit des Traités: Commentaire Article par Article, Olivier Corten & Puierre Klein (eds.) (Brussels: Bruylant, 2006), 1285. Sur, Serge, L’ Interprétation en Droit International Public (Paris: Librairie Générale de Droit et de Jurisprudence, 1974). Teubner, Gunther, Law as an Autopoietic System (Oxford: Blackwell, 1993). Thirlway, Hugh, ‘The Proliferation of International Judicial Organs: Institutional and Substantive Issues: The International Court of Justice and other International Courts’, in Proliferation of International Organizations: Legal Issues, Nicolaas M. Blokker and Henricus G. Schermers (eds.) (The Hague: Kluwer Law International), 251. Thucydides, The Peloponnesian War, Book III, V & VI. Titus Livius (Livy), Ab Urbe Condita, Book XXI. Twining, William and David Miers, How to Do Things with Rules (London: Butterworths, 1991). Valerius Maximus, Valeri Maximi Factorum et DictorumMemorabilium, Book VII. van Damme, Isabelle, Treaty Interpretation by the WTO Appellate Body (Oxford: OUP, 2009). van der Bossche, Peter, ‘From Afterthought to Centerpiece: the WTO Appellate Body and its Rise to Prominence in the World Trading System’, in The WTO at Ten: The Contribution of the Dispute Settlement System, Giorgio Sacerdoti et al. (eds.) (Cambridge: CUP, 2006), 289. Vermulst, Edwin, The WTO Anti-Dumping Agreement: A Commentary (Oxford: OUP, 2005). Villiger, Mark E., ‘Articles 31 and 32 of the Vienna Convention on the Law of Treaties in the Case-Law of the European Court of Human Rights’, in Internationale Gemeinschaft und Menschenrechte: Festschrift für Georg Rezz zum 70 Geburtstag am 21 Januar 2005, Jürgen Bröhmer and Georg Rezz (eds.) (Köln: Heymann, 2005), 317. Villiger, Mark E., Customary International Law and Treaties (Dordrecht: Martinus Nijhoff, 1985). Villiger, Mark E., Customary International Law and Treaties (The Hague: Kluwer Law International, 1997). Wallach, Lori and Patrick Woodall, WTO: Whose Trade Organization? A Comprehensive Guide to the WTO (New York: New Press, 2004). Welchman, Jennifer, The Practice of Virtue: Classic and Contemporary Readings in Virtue Ethics (Indianapolis: Hacket, 2006).
362
bibliography
Wolfke, Karol, ‘Treaties and Custom: Aspects of Interrelation’ in Essays in the Law of Treaties: a Collection of Essays in honour of Bert Vierdag, Jan Klabbers and Rene Lefeber (eds.), 31. WTO Appellate Body Secretariat, WTO Appellate Body Repertory of Reports and Awards 1995– 2004 (Cambridge: CUP, 2005). Zegveld, Liesbeth, Accountability of Armed Opposition Groups in International Law (Cambridge: CUP, 2002).
Articles ‘L’ environnement dans la jurisprudence de 1’OMC’, Les notes Bleues de Bercy 186 (1–15 July 2000): 2. ‘Metalclad Judicial Review Proceeding. Court’s Reasons for Judgment issued on 2 May 2001’, British Columbia Law Review 89 (2001): 359. Abdullah, Ibrahim, ‘Bush Path to Destruction: The Origin and Character of the Revolutionary United Front/Sierra Leone’, Journal of Modern African Studies 36 (1998): 203. Akande, Dapo and Sope Williams, ‘International Adjudication on National Security Issues: What Role for the WTO?’, Virginia Journal of International Law 43 (2003): 365. Alexandrov, Stanimir A., ‘Breaches of Contract and Breaches of Treaty’, The Journal of World Investment & Trade 5 (2004): 555. Allott, Philip, ‘The Concept of International Law’, EJIL 10 (1999): 31. Álvarez, José, ‘Judging the Security Council’, AJIL 90 (1996): 1. Apostolidis, Charalambos, ‘L’ Affaire d’ Ile de Kasikili/Sedudu (Botswana/Namibie): l’ Arret de CIJ du 13 Decembre 1999’, 45 Annuaire Francaise de Droit International 45 (1999): 434. Appleton, Arthur, ‘Amicus Curiae Submissions in the Carbon Steel Case: Another Rabbit from the Appellate Body’s Hat?’, Journal of International Economic Law 3 (2000): 691. Appleton, Arthur, ‘Shrimp/Turtle: Untangling the Nets’, Journal of International Economic Law 2 (1999): 477. Ascensio, Herve, ‘L’ Amicus Curiae devant les Juridictions Internationales’, Revue Generale de Droit International Public 105 (2001): 897. Bekker, Pieter H.F. ‘Oil Platforms (Iran v. United States)’, AJIL 98 (2004): 550. Bollecker, Brigitte, ‘L’ Avis Consultatif du 21 Juin 1971 dans l’ Affaire de la Namibie (Sud-Ouest Africain)’, Annuaire Francais de Droit International 17 (1971): 271. Bolton, John, ‘The Risks and Weaknesses of the International Criminal Court from America’s Perspective’, Law & Contemporary Problems 64 (2001): 167. Boyle, Alan, ‘Further Development of the Law of the Sea Convention: Mechanisms for Change’, ICLQ 54 (2005): 563. Brest, Paul, ‘The Misconceived quest for the Original Understanding’ Boston University Law Review 60 (1980): 204. Brierly, James L., ‘The Codification of International Law’, Michigan Law Review 47 (1948): 2. Buergnethal, Thomas, ‘Proliferation of International Courts and tribunals: Is it Good or is it Bad?’, Leiden Journal of International Law 14 (2001): 267. Buffard, Isabelle and Karl Zemanek, ‘The ‘Object and Purpose’ of a Treaty: An Enigma?’, ARIEL 3 (1998): 311. Burke-White, William and Andreas von Staden, ‘Investment Protection in Extraordinary Times: The Interpretation and Application of Non-Precluded Measures Provisions in Bilateral Investment Treaties’, Virginia Journal of International Law 48 (2008): 307. Cameron, James and Kevin R. Gray, ‘Principles of International Law in the WTO Dispute Settlement Body’, ICLQ 50 (2001): 256. Canal-Forgues, Eric, ‘Remarques sur le Recours aux Travaux Préparatoires dans le Contentieux International’, RGDIP 97 (1993): 901. Charney, Jonathan I. ‘Is International Law Threatened by Multiple International Tribunals?’ RCADI 271 (1998): 139. Charney, Jonathan I., ‘The Impact on the International Legal System of the Growth of International Courts and Tribunals’, New York University Journal of International Law and Policy 31 (1999): 697. Chester, Brown, ‘The Proliferation of International Courts and Tribunals: Finding your Way through the Maze’, Melbourne Journal of International Law 3 (2002): 453.
bibliography
363
Christakis, Théodore, ‘Quel Remède à l’Éclatement de la Jurisprudence CIRDI sur les Investissement en Argentine? La Décision du Comité ad hoc dans l’ Affaire CMS c. Argentine’, RGDIP 111 (2007): 879. Cot, Jean-Pierre, ‘La Conduite Subséquente des Parties à une Traité’, RGDIP 70 (1966): 633. Crawford, James, ‘The ILC adopts a Statute for an International Criminal Court’, AJIL 89 (1995): 404. Crawford, James, ‘The ILC’s Draft Statute for an International Criminal Tribunal’, AJIL 88 (1994): 140. Cryer, Robert, ‘A Special Court for Sierra Leone?’, ICLQ 50 (2001): 435. Cunningham, Richard and Cribb, Troy, ‘Dispute Settlement through the Lens of ‘Free Flow of Trade’: A Review of WTO Dispute Settlement of US Anti-Dumping and Countervailing Duty Measures’, Journal of International Economic Law 6 (2003): 155. Czametzky, John M. and Ronald J. Rychlak, ‘An Empire of Law? Legalism and International Criminal Court, Notre Dame Law Review 79 (2003): 55. Danilenko, Guenady M., ‘The Statute of the International Criminal Court and Third States’, Michigan Journal of International Law 21 (2000): 445. Doria, Jose, ‘Angola-A Case Study in the Challenges of Achieving Peace and the Question of Amnesty or Prosecution of War Crimes in Mixed Armed Conflicts, Yearbook of International Humanitarian Law 5 (2005): 3. Douglas, Zachary, ‘The Hybrid Foundations of Investment Treaty Arbitrations’ BYIL 74 (2003): 151. Dupuy, Pierre-Marie, ‘The Constitutional Dimension of the Charter of the United Nations Revisited’, Max Planck Yearbook of the United Nations Law 1 (1977): 1. Durling, James P., ‘Deference, but only when Due: WTO Review of Anti-Dumping Measures’, Journal of International Economic Law 6 (2003): 123. Evans, Cedric E., ‘The Concept of ‘Threat to Peace’ and Humanitarian Concerns: Probing the Limits of Chapter VII of the UN Charter’, Transnational Law & Contemporary Problems 5 (1995): 213. Fachiri, Alexander P., ‘Interpretation of Treaties’, AJIL 23 (1929): 745. Fielding, Lois E., ‘Taking a Closer Look at Threats to Peace: The Power of the Security Council to Address Humanitarian Crises’, University of Detroit Mercy Law Review 73 (1996): 551. Fitzmaurice, Gerald, ‘The Law and Procedure of the International Court of Justice, 1951–4: Treaty Interpretation and other Treaty Points’, BYIL 33 (1957): 203. Fitzmaurice, Gerald, ‘Treaty Interpretation and Other Treaty Points, 1951–1954’, BYIL 33 (1957): 203. Frank, Thomas, ‘Collective Security and UN Reform: Between the Necessary and the Possible’, Chicago Journal of International Law 6 (2006): 597. Frank, Thomas, ‘The Powers of Appreciation, who is the Ultimate Guardian of UN legality?’, AJIL 89 (1992): 519. Frank, Thomas, ‘The United States and the ICC: Unilateralism Rampant’, New York University Journal of International Law and Politics 35 (2003): 519. French, Duncan. ‘Treaty Interpretation and the Incorporation of Extraneous Legal Rules’, ICLQ 55 (2006): 281. Gardiner, Richard, ‘Art and Rules’, Opinio Juris website, (accessed on 1 November 2009). Garwood-Gowers, Andrew, ‘Did the ICJ Miss the Boat on the Law on the Use of Force?’, Melbourne Journal of International Law 5 (2004): 241. Goodman, Ryan, ‘Human Rights Treaties, Invalid Reservations and State Consent’, AJIL 96 (2002): 531. Gourgourinis, Anastasios, ‘Delineating the Normativity of Equity in International Law’, ICLR 11 (2009): 327. Gray, Thomas, ‘Do We have an Unwritten Constitution?’ Stanford Law Review 27 (1975): 703. Green, James A., ‘The Oil Platforms Case: An Error in Judgment?’, Journal of Conflict and Security Law 9 (2004): 357. Greenwald, John, ‘WTO Dispute Settlement: An Exercise in Trade Law Legislation’, Journal of International Economic Law 6 (2003): 113. Greig, Donald W., ‘Oppenheim Revisited: An Australian Perspective’, Australian Yearbook of International Law 14 (1992): 227.
364
bibliography
Griffin, James, ‘First Steps in an Account of Human Rights’, European Journal of Philosophy 9 (2001): 306. Hafner, Gerhard et al., ‘A Response to the American View as Presented by Ruth Wedgwood’, EJIL 10 (1999): 108. Halberstam, Malvina, ‘A Treaty is a Treaty is a Treaty’, Virginia Journal of International Law 33 (1992): 51. Harris, Seth, ‘The United States and the International Criminal Court: Legal Potential for Nonparty States’, Hawaii Law Review 23 (2000): 277. Harvard Research, ‘Article 19: Interpretation of Treaties’, AJIL 29 (1935): 961. Heiskanen, Veijo, ‘Oil Platforms: Lessons of Dissensus’, Nordic Journal of International Law 74 (2005): 179. Higgins, Rosalyn, ‘Time and Law: International Perspectives on an Old Problem’, ICLQ 46 (1997): 501. Hollis, Duncan, ‘Art and Auto-Interpretation of Treaties’, Opinio Juris website, (accessed on 1 November 2009). Hyde, Charles Cheney, ‘Concerning the Interpretation of Treaties’, AJIL 3 (1909): 46. Jacobs, Francis G., ‘Judicial Dialogue and the Cross-fertilization of Legal Systems: the European Court of Justice’, Texas International Law Journal 38 (2003): 547. Jennings, Robert, ‘General Course on Principles of International Law’, RCADI 121 (1967/II): 323. Johnstone, Ian, ‘Treaty Interpretation: The Authority of Interpretive Communities’, Michigan Journal of International Law 12 (1991): 371. Kammerhofer, Jorg, ‘Oil’s Well that Ends Well? Critical Comments on the Merits Judgement in the Oil Platforms Case’, Leiden Journal of International Law 17 (2004): 695. Kaufmann-Kohler, Gabrielle, ‘Arbitral Precedent: Dream, Necessity or Excuse?’, Arbitration International 23 (2007): 357. Kaufmann-Kohler, Gabrielle, ‘In search of Transparency and Consistency: ICSID Reform Proposal’, Transnational Dispute Management 2(5) (2005): 8. Kennedy, David, ‘The Turn to Interpretation’, Southern California Law Review 58 (1985): 251. Kennedy, Kevin C., ‘A WTO Agreement on Investments: A Solution in Search of a Problem’, University of Pennsylvania Journal of International Economic Law 24 (2003): 77. Kirsch, Philippe, ‘The International Criminal Court and the Enforcement of International Justice’, Pace International Law Review 17 (2005): 47. Kissinger, Henry A., ‘The Pitfalls of Universal Jurisdiction’, Foreign Affairs (July–Aug. 2001): 86. Klabbers, Jan, ‘International Legal Histories: The Declining Importance of Travaux Préparatoires in Treaty Interpretation?’, NILR 50 (2003): 267. Klabbers, Jan, ‘Interpretation as the Continuation of Politics by Other Means’, Opinio Juris website, (accessed on 1 November 2009). Klabbers, Jan, ‘Jurisprudence in International Trade Law; Art. XX of GATT’, JWT 26 (1992): 86. Klabbers, Jan, ‘On Rationalism in Politics: Interpretation of Treaties and the World Trade Organization’, Nordic Journal of International Law 74 (2005): 405. Klabbers, Jan, ‘Some Problems Regarding the Object and Purpose of Treaties’, Finnish Yearbook of International Law 7 (1997): 138. Klabbers, Jan, ‘The Bridge Crack’d: A Critical Look at Interdisciplinary Relations’ International Relations 23 (2009, forthcoming). Klabbers, Jan, ‘The Commodification of International Law’, Select Proceedings of the European Society of International Law 1 (2006): 341. Klabbers, Jan, ‘The Relative Autonomy of International Law, or The Forgotten Politics of Interdisciplinarity’, Journal of International Law and International Relations 1 (2004–2005): 35. Koskenniemi, Martti, ‘Constitutionalism as Mindset: Reflections on Kantian Themes in International Law and Globalization’, Theoretical Inquiries in Law 8 (2007): 9. Kuyper, Pieter J., ‘The Law of GATT as a Special Field of International Law, Ignorance, Further Refinement or Self-Contained System of International Law,’ NYIL 25(1994): 229. Laursen, Andreas, ‘The Judgment by the International Court of Justice in the Oil Platforms Case’, Nordic Journal of International Law 73 (2004): 160. Lauterpacht, Hersch, ‘De l’Interprétation des Traités: Rapport’, Annuaire de L’Institut de Droit International 43 (1950/II): 366.
bibliography
365
Lauterpacht, Hersch, ‘Les Travaux Préparatoires et l’ Intérpretation des Traités”, RCADI 48-II (1934): 709. Lauterpacht, Hersch, ‘Restrictive Interpretation and the Principle of Effectiveness in the Interpretation of International Treaties’, BYIL 26 (1949): 48. Leben, Charles, ‘L’Etat de Nécessité dans le Droit International de l’ Investissement’, Gazette du Palais 349 (2005): 19. Lenaerts, Koen, ‘Interpretation and the Court of Justice: a Basis for Comparative Reflection’, The International Lawyer 41 (2007): 1011. Lennard, Michael, ‘Navigating by the Stars: Interpreting the WTO Agreements’, Journal of International Economic Law 5 (2002): 17. Letsas, George, ‘The Truth in Autonomous Concepts: How to Interpret the ECHR’, EJIL 2 (2004): 279. Lindroos, Anja and Michael Mehling, ‘Dispelling the Chimera of ‘Self-Contained Regimes’: International Law and the WTO’, EJIL 16 (2005): 857. Lyons, David, ‘Constitutional Interpretation and Original Meaning’, Social Philosophy and Policy 4 (1986): 75. Macauley, Stewart, ‘Non-contractual Relations in Business: A Preliminary Study’, American Sociological Review 28 (1963): 55. Mahoney, Curtis J., ‘Treaties as Contracts: Textualism, Contract Theory, and the Interpretation of Treaties’, Yale Law Journal 116 (2007): 824. Marceau, Gabrielle, ‘A Call for Coherence in International Law. Praises for the Prohibition Against ‘Clinical Isolation’ in WTO Dispute Settlement’, JWT 33 (1999): 120. McCormack, Timothy, ‘Selective Reaction to Atrocity: War Crimes and the Development of International Criminal Law’, Albany Law Review 60 (1997): 681. McLachlan, Campbell, ‘The Principle of Systemic Integration and Article 31(3)(c) of the Vienna Convention’, ICLQ 54 (2005): 279. McNair, Arnold, ‘Comments on the Report of Lauterpacht’, Annuaire de l’ Institut de Droit International 43 (1950): Tome I, 450. McRae, Donald M., ‘The WTO in International Law: Tradition Continued or New Frontier?’, Journal of Economic Law 3 (2002): 27. Mehrisch, Brijesh N., ‘Travaux préparatoires as an Element in the Interpretation of Treaties’, Indian Journal of International Law 11 (1971): 39. Merkouris, Panos, ‘Debating the Ouroboros of International Law: The Drafting History of Article 31(3)(c)’, International Community Law Review 9 (2007): 1. Meron, Theodor, ‘International Criminalization of Internal Atrocities’, AJIL 89 (1995): 554. Morris, Madeline, ‘High Crimes and Misconceptions: The ICC and Non-Party States’, Law & Contemporary Problems 64 (2001): 13. Murphy, John, ‘The Quivering Gulliver: US Views on a Permanent International Criminal Court’, The International Lawyer 34 (2000): 45. Ochoa-Ruiz, Natalia and Esther Salamanca-Aguado, ‘Exploring the Limits of International Law relating to the Use of Force in Self-defence’, EJIL 16 (2005): 499. Okowa, Phoebe, ‘Congo’s War: The Legal Dimension of a Protracted Conflict’, BYIL 55 (2006): 203. Orakhelashvili, Alexander, ‘Oil Platforms (Islamic Republic of Iran v. United States), Merits, Judgment of 6 November 2003’, ICLQ 53 (2004): 753. Orakhelashvili, Alexander, ‘The Impact of Peremptory Norms on the Interpretation and Application of United Nations Security Council Resolutions’, EJIL 16 (2005): 59. Ortino, Federico, ‘Treaty Interpretation and the WTO Appellate Body Report in the US-Gambling: A Critique’, Journal of International Economic Law 9 (2006): 117. Palmeter, David and Petros C. Mavroidis, ‘The WTO Legal System: Sources of Law’, AJIL 92 (1998): 398. Papastavridis, Efthymios, ‘Interpretation of Security Council Resolutions under Chapter VII in the Aftermath of the Iraqi Crisis’, ICLQ 56 (2007): 83. Pauwelyn, Joost, ‘Bridging Fragmentation and Unity: International Law as a Universe of InterConnected Islands’, Michigan Journal of International Law 25 (2004): 903. Pauwelyn, Joost, ‘The Role of Public International Law in the WTO: How Far Can We Go?’ AJIL 95 (2001): 575. Perry, Michael J., ‘Interpretivism, Freedom of Expression and Equal Protection’ Ohio State Law Journal 42 (1981): 261.
366
bibliography
Reinisch, August, ‘Necessity in International Arbitration. An Unnecessary Split of Opinions in Recent ICSID Cases? Comments on CMS v. Argentina and LG&E v. Argentina’, Journal of World Investment and. Trade 8 (2007): 191. Rishikof, Harvey, ‘When Naked Came the Doctrine of ‘Self-Defense’: What is the Proper Role of the International Court of Justice in Use of Force Cases?’, Yale Journal of International Law 29 (2004): 331. Rosenberg, Tina, ‘Conference Convocation, War Crimes Tribunals: The Record and the Prospects’, American University International Law Review 13 (1998): 1406. Sands, Philippe, ‘Treaty, Custom and the Cross-Fertilization of International Law’, Yale Human Rights and Development Law Journal 1 (1999): 85. Scalia, Antonin, ‘Originalism: The Lesser Evil’, University of Cincinnati Law Review 57 (1989): 849. Schabas, William A., ‘The Relationship Between Truth Commissions and International Courts: The Case of Sierra Leone’, Human Rights Quarterly 25 (2003): 1035. Schachter, Oscar, ‘The Invisible College of International Lawyers’, Northwestern University Law Review 72 (1977): 217. Schaffer, Gregory, ‘United States – Import Prohibition of Certain Shrimp Products’, AJIL 93 (1992): 508. Scharf, Michael P., ‘The ICC’s Jurisdiction over Non-Party States: a Critique of the US Position’, Law & Contemporary Problems 64 (2001): 67. Scheffer, David J., ‘A Negotiator’s Perspective on the International Criminal Court’, International Military Law Review 167 (2001): 1. Scheffer, David J., ‘War Crimes Tribunals: The Record and the Prospects’, (Address at the Conference Convocation for the Washington College of Law Conference), American University International Law Review 13 (1998): 1389. Schønberg, Søren and Karin Frick, ‘Finishing, Refining, Polishing: on the Use of Travaux Préparatoires as an Aid to the Interpretation of Community Legislation’, European Law Review 28 (2003): 149. Schreuer, Cristoph, ‘The Interpretation of Treaties by Domestic Courts’, BYIL 45 (1971): 255. Sellen, Keith L., ‘The UN Security Council Veto in the New World Order’, Military Law Review 138 (1992): 187. Shaw, Malcolm, ‘Case concerning Kasikili/Sedudu Island (Botswana/Namibia)’, ICLQ 49 (2000): 964. Simma, Bruno and Dirk Pulkowski, ‘Of Planets and the Universe: Self-Contained Regimes in International Law’, EJIL 17 (2006): 483. Sinclair, Anthony C., ‘The Origin of the Umbrella Clause in the International Law of Investment Protection’, Arbitration International 20 (2004): 411. Slaughter, Anne-Marie, ‘A Global Community of Courts’, Harvard International Law Journal 44 (2003): 191. Small, David H., ‘The Oil Platforms Case: Jurisdiction Through The – Closed-Eye of the Needle’, The Law and Practice of International Courts and Tribunals 3 (2004): 112. Spamann, Holger, ‘Standard of Review for World Trade Organization Panels in Trade Remedy Cases: A Critical Analysis’, Journal of World Trade 38 (2004): 509. Stein, Mark S., ‘The Security Council, the International Criminal Court and the Crime of Aggression: How Exclusive is the Security Council Power to Determine Aggression? Indiana International & Comparative Law Review 16 (2001): 1. Tarullo, Daniel, ‘Paved with Good Intentions: The Dynamic Effects of WTO Review of AntiDumping Action’, World Trade Review 2 (2003): 373. Tavernier, Paul, ‘Observations sur le Droit Intertemporel dans l’Affaire de l’Ile de Kasikili/ Sedudu (Boswana/Namibie);Cour Internationale de Justice, Arret du 13 decembre 1999’, Revue Generale de Droit International Public 104 (2000): 429. Thirlway, Hugh, ‘The Law and Procedure of the International Court of Justice: 1960–1989 (Part Three)’, BYIL 62 (1991): 1. Thirlway, Hugh, ‘The Law and Procedure of the International Court of Justice’, BYIL 69 (1992): 44. Thomas, J. Christopher, ‘Reflections on Article 1105 of NAFTA: History, State Practice and the Influence of Investors’, ICSID Review -FJ1L 21 (2002): 21. van der Vyver, Johan D., ‘American Exceptionalism: Human Rights, International Criminal Justice, and National Self-righteousness’, Emory Law Journal 50 (2001): 775.
bibliography
367
Warbrick, Colin, ‘The United Nations System: A Place for Criminal Courts?’, Transnational Law & Contemporary problems 5 (1995): 237. Wedgwood, Ruth, ‘The US and the ICC, the Irresolution of Rome’, Law & Contemporary Problems 64 (2001): 193. Williams, Sarah, ‘Amnesties in International Law: The Experience of the Special Court for Sierra Leone’, Human Rights Law Review 5 (2005): 271. Wright, Quincy, ‘The Interpretation of Multilateral Treaties’ AJIL 23 (1929): 94. Yasseen, Mustafa K., ‘L’Interprétation des Traités d’après la Convention de Vienne sur le Droit des Traités’, RCADI 151 (1976-III): 1.
Documents 2002 Ratification Act of the Special Court Agreement of 2002, Act, Supplement to the Sierra Leone Official Gazette Vol. CXXX No.2, <www.sc-sl.org/LinkClick.aspx?fileticket=BcKTbFltxZk %3d&tabid=176> (last accessed on 1 November 2009). Bolton, John, ‘Is a UN International Criminal Court in the U.S. National Interest?’ (Hearing Before the Senate Subcomission on Foreign Relations: Statement of John R. Bolton, Under Secretary of State for Arms Control, Department of State), 105th Cong. 724 (23 July 1998). Boucher, Richard, US Department of State, ‘US Assesses EU Decision on ICC Article 98 Agreements’ (Press Briefing of 1 October 2002) <www.useu.be/Categories/JusticeandHome Affairs/Oct0102BoucherEUArticle98.ICC.html> (accessed on 22 October 2006). Burns, Nicholas, ‘On United Nations Reform’, (Testimony as Prepared Before the Senate Foreign Relations Committee (July 21, 2005) ) www.state.gOv/p/us/rm/2005/49900.htm> (accessed on 12 September 2006). CCFT v. US, Transcripts of Hearing of 9 October 2007, <www.naftaclaims.com/Disputes/USA/ CCFT/CCFT-USA-Transcript%20Day-01.pdf> (accessed on 1 May 2009). CIEL, ‘EC-Biotech: Overview and Analysis of the Panel’s Interim Report’ (2006), www.ciel.org/ Publications/EC_Biotech_Mar06.pdf (accessed on 1 May 2009). Coalition for the International Criminal Court, ‘The Relationship Agreement between the ICC and the United Nations’, (accessed on 21 November 2006). Comments by Myres S. McDougal, of the United States Delegation to the Committee of the Whole of the Vienna Conference on the Law of Treaties, in Vienna Conference on the Law of Treaties, Meetings of the Committee of the Whole, 31st Meeting, at 167, para. 38. Comments by Shabtai Rosenne, (A/CN.4/SR.872), reproduced in YILC (1966): Vol. I, Part II, at 200, paras. 34–5. Comments of Alfred Verdross (A/CN.4.SR.728), reproduced in YILC (1964): Vol. I at 33, para. 6. Comments of Antonio de Luna (A/CN.4/SR.766), reproduced in YILC (1964): Vol. I, at 285, para. 36. Comments of Antonio de Luna (A/CN.4/SR.871), reproduced in YILC (1966): Vol. I, Part II, at 194, para. 16. Comments of Jimenez de Arechaga (A/CN.4.SR.728), reproduced in YILC (1964): Vol.I, at 34, para. 11. Comments of Jimenez de Arechaga, in Vienna Conference on the Law of Treaties, Meetings of the Committee as a Whole, 31st Meeting, para. 64. Comments of Jose Maria Ruda, in Vienna Conference on the Law of Treaties, Meetings of the Committee as a Whole, 33rd Meeting, para. 25. Comments of Milan Bartos (A/CN.4/SR.766), reproduced in YILC (1964): Vol. I, at 287, para. 57. Comments of Mustafa Kamil Yasseen (A/CN.4.SR.729), reproduced in YILC (1964): Vol.I at 37, para. 35. Comments of Sir Erick Beckett, ‘De l’ Interpretation des Traités’, Annuaire de l’ Institut de Droit International 43 (1950): Tome I, at 438–40. Comments of Sir Ian Sinclair, in Vienna Conference on the Law of Treaties, Meetings of the Committee as a Whole, 33rd Meeting, para. 8. Commission of the European Communities, European Governance – A White Paper (25 July 2001), (COM(2001)428 final).
368
bibliography
Council of the European Union, Draft Council Conclusions on the International Criminal Court (30 September 2002) <www.amicc.org/docs/EC9_30_02.pdf> (accessed on 24 November 2006). Dispute Settlement Body, ‘Minutes of Meeting held on 19 February 2009’, WT/DS/B/M/265, para.78 <docsonline.wto.org/GEN_highLightParent.asp?qu=%28%40meta%5FSymbol+WT %FCDSB%FCM%FC%2A+and+%40meta%5FMeet%5FDate+19%2F02%2F2009%29& doc=D%3A%2FDDFDOCUMENTS%2FT%2FWT%2FDSB%2FM265%2EDOC%2EHTM& curdoc=3&popTitle=WT%2FDSB%2FM%2F265> (accessed on 1 July 2009). Grossman, Marc, ‘American Foreign Policy and the International Criminal Court’ (May 6, 2002), <www.state.g0v/p/9949.htm> (accessed on 12 March 2005). Helms, Jesse (Senator),‘We Must Slay This Monster’, Financial Times, 31 July 1998, 18. ICC – Office of the Prosecutor, Report on Prosecutorial Strategy (The Hague, 14 September 2006), <www.iccnow.org/documents/ProsecutorialStrategy_06Sep14.pdf> (accessed on 1 May 2009). ICC – Office of the Prosecutor, Report on the Activities Performed during the First Three Years (The Hague, 12 September 2006), <www.iccnow.org/documents/3YearReport%20_06Sep14 .pdf> (accessed on 1 May 2009). ICC Prosecutor, Fourth Report of the Prosecutor to the UNSC pursuant to UNSC Res. 1593 (2005), <www.icc-cpi.mt/library/organs/otp/OTPReportUNSC4-Darfur English.pdf> (accessed on 16 December 2006). ICC, ‘ICC-UN Agreement Signed’, ICC Newsletter No. 2, October 2004, <www.icc-cpi.int/ librarv/about/newsletter/files/ICC-NL2-200410_En.pdf> (accessed on 22 October 2004). ICC, ICC OTP Iraqi Response (9 February 2006), <ww\h;v.icc-cpi.int/library/organs/otp/OTP_ etter_to_senders_re_Iraq_9_February_2006.pdf> (accessed on 21 October 2006). ICC, ICC Press Release of 1 October 2004, <www.icc-cpi.int/press/pressreleases/46.html> (accessed on 22 October 2006). ICSID Secretariat, Possible Improvements of the Framework for ICSID Arbitration (Discussion Paper of 2 October 2004) <www.worldbank.org/icsid/improve-arb.pdf> (accesssed on 1 December 2006). ICSID Secretariat, Suggested Changes to the ICSID Rules and Regulations (Working Paper of the ICSID Secretariat, 12 May 2005) (accessed on 1 December 2006). ILC Study Group on Fragmentation, Fragmentation of International Law: Difficulties Arising from the Diversification and Expansion of International Law – Conclusions of the Study Group (18 July 2006) (A/CN.4/L702). ILC Study Group on Fragmentation, Fragmentation of International Law: Difficulties Arising from the Diversification and Expansion of International Law, Report of the Study Group, finalized by Martti Koskenniemi, (13 April 2006) (A/6N.4/L.682). ILC Study Group on Fragmentation, Fragmentation of International Law: Difficulties Arising from the Diversification and Expansion of International Law, Report of the Study Group, finalized by Martti Koskenniemi, (13 April 2006) (A/6N.4/L.682). ILC Study Group on Fragmentation, Fragmentation of International Law: Difficulties Arising from the Diversification and Expansion of International Law, (29 July 2005) (A/CN.4/L.676). ILC Study Group on Fragmentation, Fragmentation of International Law: Difficulties Arising from the Diversification and Expansion of International Law, Report of the Study Group, finalized by Martti Koskenniemi, (13 April 2006) (A/6N.4/L.682). ILC Study Group on Fragmentation, Report of the Study Group, (2 May 2006) (A/CN.4/L682/ Add.l). ILC, 726th Meeting: Law of Treaties, (A/CN.4/167), reproduced in YILC (1964), Vol. I: 20. ILC, 726th Meeting: Law of Treaties, (A/CN.4/167), reproduced in YILC (1964), Vol. I: 20. ILC, 765th – 766th Meetings: Law of Treaties, (A/CN.4/167/Add.3), reproduced in YILC (1964), Vol. I: 275. ILC, 765th – 766th Meetings: Law of Treaties, (A/CN.4/167/Add.3), reproduced in YILC (1964), Vol. I: 275. ILC, 873rd Meeting: Law of Treaties, (A/CN.4/186 and Addenda; A/CN.4/L.107, L.115), reproduced in YILC (1966), Vol. I Part II: 202. ILC, Draft Articles on the Law of Treaties with Commentaries, YILC (1964), Vol. II: 176. ILC, Draft Statute of an International Criminal Court, Report of the International Law Commission on the Work of its 46th Session (2 May-22 July 1994) (A/49/10). ILC, Guiding Principles Applicable to Unilateral Declarations of States Capable of Creating Legal Obligations with Commentaries, YILC (2006), Vol. II, Part II.
bibliography
369
ILC, Report of the ILC on the Work of its 18th Session: Draft Articles on the Law of Treaties with Commentaries, (A/6309/Rev.1), reproduced in YILC (1966), Vol. II: 218. ILC, Report of the International Law Commission (58th Session, 1 May-9 June and 3 July-11 August 2006) (A/61/10): 413–6. ILC, Report of the International Law Commission (Fifty-eighth session, 1 May-9 June and 3 July11 August 2006) (A/61/10). ILC, Report of the Working Group on the Question of an International Criminal Jurisdiction – Report of the International Law Commission on the work of its forty-fourth session, (4 May-24 July 1992), (A/47/10, Annex). ILC, The International Law Commission Commentary on the draft Vienna Convention, YILC (1966): Vo. II, at 223. Kirsch, Philippe, ‘Remarks on Accountability vs. Impunity: The Role of the International Criminal Court, Statement to the Council on Foreign Relations’, Washington D.C. (16 January 2004), <www.cfr.org/publication/6696/accountability_vs_impunity.html> (accessed on 23 November 2006). NAFTA Free Trade Commission, ‘Notes of Interpretation of Certain Chapter 11 Provisions of 31 July 2001’, at <www.naftaclaims.com/files/NAFTA_Comm_1105_Transparency.pdf.> (accessed on 1 May 2009). President of Sierra Leone, Letter from the President of Sierra Leone to the UN Secretary General, (12 June 2000) (S/2000/786). President of the Special Court for Sierra Leone, First Annual Report of the President of the Special Court for Sierra Leone, <www.sc-sl.org/LinkClick.aspx?fileticket=NRhDcbHrcSs%3d& tabid=176> (last accessed on 1 November 2009). Scheffer, David, ‘Is a UN International Criminal Court in the U.S. National Interest?’ (Hearing Before the Senate Subcommission on Foreign Relations: Statement of David. J. Scheffer, Ambassador-at-Large for War Crimes Issues, Department of State), 105th Cong. 1314 (23 May 1998). Sierra Leone Truth and Reconciliation Commission, 2004 Report of the Sierra Leone Truth and Reconciliation Commission (TRC Report), <www.trcsierraleone.org/drwebsite/publish/index .shtml> (last accessed on 1 November 2009). Sir Humphrey Waldock, Third Report on the Law of Treaties, (A/CN.4/167 & Add.1–3), reproduced in YILC (1964): Vol. II, 5. Sir Humphrey Waldock, Third Report on the Law of Treaties, (A/CN.4/167 & Add.1–3), reproduced in YILC (1964): Vol. II, 5. Statement of the French Representative, (1993) (S/PV.3175), UNSCOR, 48th Session., 3175th meeting, at 8. The American Service Members’ Protection Act of 2002 (ASPA), Pub. L. No. 107–206, 116 Stat. 820. tit. II (2002). UN Secretary-General, ‘Report of the Secretary-General on the Establishment of a Special Court for Sierra Leone’ (4 October 2000), <www.afrol.com/Countries/Sierra_Leone/ documents/un_sil_court_041000.htm> (accessed on 1 May 2009). UN Secretary-General, 1993 Report of the Secretary-General pursuant to para 2 of Security Council Resolution 808, (S/25704). UN Secretary-General, Report of the Secretary-General on the Establishment of a Special Court for Sierra Leone, (4 October 2000) (S/2000/915). UN Secretary-General, Report of the Secretary-General on the Establishment of a Special Court for Sierra Leone, (4 October 2000) (S/2000/915). UN Secretary-General, Seventh Report of the Secretary – General on the United Nations Observer Mission in Sierra Leone (1999) (S/1999/836). United Nations, A More Secure World: Our Shared Responsibility – Report of the SecretaryGeneral’s High-Level Panel on Threats, Challenges and Change, (2004) (A/59/565). United Nations, United Nations Conference on the Law of Treaties, Official Records (1968).
Treaties 1933 Montevideo Convention on Rights and Duties of States, LNTS 165 (1933): 19. 1945 Charter of the United Nations, UNTS 1: XVI.
370
bibliography
1945 Stattue of the International Court of Justice, <www.icj-cij.org/documents/index.php? p1=4&p2=2&p3=0&PHPSESSID=3674b0b2753569a8012620be950d7dd9> (accessed on 1 July 2009). 1948 Universal Declaration of Human Rights (A/810). 1950 European Convention on Human Rights, UNTS 213: 222. 1965 Convention on the Settlement of Investment Disputes between States and Nationals of Other States (the ICSID Convention), ILM 5 (1965): 532. 1966 International Covenant on Civil and Political Rights, UNTS 999: 171. 1969 American Convention on Human Rights, UNTS 1144: 123. 1969 Vienna Convention on the Law of Treaties (1969), 1155 UNTS 1155 (1980): 331; ILM 8 (1969): 679. 1972 European Conventions on the Transfer of Proceedings in Criminal Matters, ETS 73. 1973 Convention on the International Trade in Endangered Species of Fauna and Flora (CITES), UNTS 993: 243. 1981 African Charter of Human and People’s Rights, UNTS 1520: 217. 1981 African Charter on Human and People’s Rights, ILM 21: 58. 1986 Vienna Convention on the Law of Treaties between States and International Organisations or Between International Organisations, (A/Conf.129/15). 1986 Vienna Convention on the Law of Treaties between States and International Organisations or between International Organisations, ILM 25 (1986): 543. 1992 North American Free Trade Agreement (NAFTA), ILM 32 (1993): 289. 1992 US-Russia BIT, ILM 31 (1992): 794. 1994 Agreement Establishing the World Trade Organisation in Final Act Embodying the Results of the 1994 Uruguay Round of Multilateral Treaty Negotiations, ILM 33 (1994): 1125. 1994 Agreement establishing the WTO, ILM 33 (1994): 1144. 1994 Agreement on Implementation of Article VI of the General Agreement on Tariffs and Trade, <www.wto.org/english/docs_e/legal_e/19-adp.pdf> (accessed on 1 May 2009). 1994 Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPs), ILM 33 (1994): 1197. 1994 Dispute Settlement Understanding (DSU), ILM 33 (1994): 1226. 1994 Energy Charter Treaty, ILM 34 (1995): 360. 1994 General Agreement of Trade and Services (GATS), ILM 33 (1994): 1168. 1994 General Agreement on Tariffs and Trade, ILM 33 (1994): 1154. 1996 Peace Agreement between the Government of Sierra Leone and the Revolutionary United Front of Sierra Leone, (UN Doc. S/1996/1034). 1996 Peace Agreement between the Government of the Republic of Sierra Leone and the Revolutionary United Front of Sierra Leone, <www.sc-sc.org/abidjanaccord.html> (last accessed on 1 November 2009). 1998 Rome Statute of the International Criminal Court (A/Conf. 183/9). 1999 Peace Agreement between the Government of Sierra Leone and the Revolutionary United Front of Sierra Leone (RUF), <www.iss.co.za/AF/profiles/SieraLeone/pax.pdf> (last accessed on 1 November 2009). 1999 US-Bahrain BIT, <www.unctadxi.org/templates/DocSearch____779.aspx.> (accessed on 1 May 2009). 2002 Agreement between the government of the United States of America and the government of Romania regarding the surrender of persons to the International Criminal Court (2002), <www.amicc.org/docs/Romania_98_2.pdf> (accessed on 22 October 2006). 2002 Agreement between the government of the United States of America and the government of Romania regarding the surrender of persons to the International Criminal Court (2002), <www.amicc.org/docs/Romania982.pdf> (accessed on 22 October 2006). 2002 Agreement between the United Nations and the Government of Sierra Leone on the Establishment of a Special Court for Sierra Leone, <www.sc-sl.org/LinkClick.aspx?fileticket= CLk1rMQtCHg%3d&tabid=176>. (last accessed on 1 November 2009). 2002 Statute of the Special Court for Sierra Leone, <www.sc-sl.org/LinkClick.aspx?fileticket= uClnd1MJeEw%3d&tabid=176> (last accessed on 1 November 2009). 2004 Central America-Dominican Republic Free Trade Agreement, (accessed on 1 July 2009).
bibliography
371
2004 Chile-US FTA, <www.ustr.gov/trade-agreements/free-trade-agreements/chile-fta/finaltext> (accessed on 1 July 2009). 2004 Model Treaty, (accessed on 1 May 2009). 2004 Relationship Agreement between the ICC and the United Nations, UNTS 2283, II-1272: 196. 2004 Singapore-US FTA, <www.ustr.gov/sites/default/files/uploads/agreements/fta/singapore/ asset_upload_file708_4036.pdf> (accessed on 1 July 2009). 2004 Uruguay-US BIT, ILM 44 (2005): 268.
LIST OF CASES ICJ Aegean Continental Shelf case (Greece v. Turkey), Jurisdiction of the Court, Judgment of 19 December 1978, 1978 ICJ Rep. 3. Aerial Incident of 27 July 1955, (Israel v. Bulgaria), Judgment of 26 May 1959, 1959 ICJ Rep. 127. Anglo-Iranian Oil Company case (United Kingdom v. Iran), Preliminary Objections. Judgment of 22 July 1952, 1952 ICJ Rep. 105. Application of the Convention on the Prevention and Punishment of the Crime of Genocide (Bosnia and Herzegovina v. Serbia and Montenegro), Judgment of 26 February 2007, 2007 ICJ Rep. 1. Armed Activities on the Territory of the Congo (Democratic Republic of the Congo v. Uganda), Judgment of 19 December 2005, 2005 ICJ Rep. 1. Armed Activities on the Territory of the Congo (New Application: 2002) (Democratic Republic of Congo v. Rwanda), Jurisdiction and Admissibility, Judgement of 3 February, 2006, 2006 ICJ Rep. 6. Barcelona Traction, Light & Power Co. (Belgium v. Spain), Second Phase, Judgment of 5 February 1970, 1970 ICJ Rep. 3. Case Concerning Kasikili/Sedudu Island (Botswana v. Namibia), Judgment of 13 December 1999 1999 ICJ Rep. 1045. Case Concerning Military and Paramilitary Activities in and against Nicaragua (Nicaragua v. United States of America), Judgment of 27 June 1986, 1986 ICJ Rep. 14. Case concerning the Frontier Dispute (Benin/Niger), Judgment of 12 July 2005, 2005 ICJ Rep. 90. Case of Maritime Delimitation and Territorial Questions between Qatar and Bahrain (Qatar v. Bahrain), Jurisdiction and Admissibility, Judgment of 15 February 1995, 1995 ICJ Rep. 4. Case of Maritime Delimitation and Territorial Questions between Qatar and Bahrain (Qatar v. Bahrain), Jurisdiction and Admissibility, Judgment of 1 July 1994, 1994 ICJ Rep. 112. Certain Questions of Mutual Assistance in Criminal Matters (Djibouti v. France), Judgment of 4 June 2008, 2008 ICJ Rep., <www.icj-cij.org/docket/files/136/14550.pdf> (accessed on 1 July 2009). Dispute regarding Navigational and Related Rights (Costa Rica v. Nicaragua), Judgment of 13 July 2009, Declaration of Judge Guillaume, <www.icj-cij.org/docket/files/133/15330.pdf> (accessed on 1 November 2009). Effect of Awards of Compensation Made by; the United Nations Administrative Tribunal, Advisory Opinion of 13 July 1954, 1954 ICJ Rep. 47. Fisheries case (United Kingdom v. Norway), Judgment of 18 December 1951, 1951 ICJ Rep. 116. Gabcikovo-Nagymaros Project (Hungary v. Slovakia), Judgment of 25 September 1997,1997 ICJ Rep. 7. Legal Consequences for States of the continued presence of South Africa in Namibia (South West Africa), Advisory Opinion of 21 June 1971, 1971 ICJ Rep. 16. North Sea Continental Shelf cases (Germany v. Denmark and the Netherlands), Judgment of 20 Febrary 1969, 1969 ICJ Rep. 3. Nuclear Tests case (Australia v. France), Judgment of 20 December 1974, 1974 ICJ Rep. 253. Oil Platforms (Iran v. United States of America), Preliminary Objections, Judgment of 12 December 1996, 1996 ICJ Rep. 812. Oil Platforms case (Iran v. United States of America), Merits, Judgment of 6 November 2003, 2003 ICJ Rep. 161, reprinted in ILM 42 (2003): 1334. Rights of Nationals of the United States of America in Morocco (France v. United States of America), Judgment of 27 August 1952, 1952 ICJ Rep. 167. South West African cases (Ethiopia v. South Africa and Liberia v. South Africa), Judgments of 18 July 1966, 1966 ICJ Rep. 6. Sovereignty over Pulau Litigan and Pulau Sipadan (Indonesia v. Malaysia), Judgment of 17 December 2002, 2002 ICJ Rep. 625.
374
list of cases
Territorial Dispute (Libyan Arab Jamahiriya v. Chad), Judgment of 3 February 1994, 1994 ICJ Rep. 4.
PCIJ Advisory Opinion Concerning Nationality Decrees issued in Tunis and Morocco, Sir Ernest Pollock, Pleadings, 1923 PCIJ (Ser. C), No. 2, at 197. Advisory Opinion No.14 relative to the Competence of the European Commission of the Danube, 1927 PCIJ (Series B), No. 14, 28. Advisory Opinion on the Polish Postal Service in Danzig, 1925 PCIJ (Ser. B), No. 11, 6. Case concerning the Employment of Women during the Night, PCIJ (Ser. A/B), No. 50. Diversion of Water from the Meuse Case; Case of Diversion of Water from the Meuse, 1937 PCIJ (Ser. A/B), No. 70. On the Interpretation of the Convention of 1919 Concerning Employment of Women During Night, 1932 PCIJ (Ser. A/B) No. 50. Phosphates in Morocco case (Italy v. France), 1938 PCIJ (Ser. A/B) No. 74. Territorial Jurisdiction of the International Commission of the River Oder, 1929 PCIJ (Ser. A), No. 23. The Case of the S.S. Lotus, (France v. Turkey), 1927 PCIJ (Ser. A), No. 10, 17. The Question of Jaworzina, Advisory Opinion (1923), PCIJ (Ser. B), No. 8. The Wimbledon case (France, Italy, Japan and the UK v. Germany), 1923 PCIJ (Ser. A) No. 1, 15.
PCA Boundaries in the Island of Timor Case, (Netherlands v. Portugal) (1914), RIAA XI: 502. Dispute Concerning Access to information under article 9 of the OSPAR Convention (Ireland v. United Kingdom), PCA, Final Award of 2 July 2003. In the Arbitration Regarding the Iron Rhine (‘IJzeren Rijn’) Railway Between The Kingdom of Belgium and the Kingdom of the Netherlands, PCA, Award of the Tribunal of 24 May 2005. In the Matter of an Arbitration Under the UNCITRAL Rules 1976, Saluka Investments BV (The Netherlands) v. the Czech Republic, Partial Award of 17 March 2006. Island of Palmas (or Miangras) case (United States of America v. the Netherlands) (1928), RIAA 2: 829. Mox Plant Case (Ireland v. United Kingdom), PCA, Request for Provisional Measures Order of 3 December 2001. North Atlantic Coast Fisheries Case (Great Britain v. United States of America) (1910), reprinted in James Scott (ed.), Hague Court Reports (New York: OUP, 1916), 186.
WTO/GATT Brazil – Measures Affecting Desiccated Coconut (17 October 1996), Panel Report, WT/DS22/R. Brazil – Measures Affecting Desiccated Coconut (21 February 1997), Appellate Body Report, WT/DS22/AB/R. Canada – Measures Affecting Exports of Unprocessed Herring and Salmon (22 March 1988), Panel Report, BISD 35S/98. Chile – Price Band System and Safeguard Measures Relating to Certain Agricultural Products (23 September 2002), Appellate Body Report, WT/DS207/AB/R. EEC – Regulation on Imports of Parts and Components (16 May 1990), Panel Report BISD 37S/132. European Communities – Customs Classification of Certain Computer Equipment (22 June 1998), Appellate Body Report, WT/DS62, 67, 68/AB/R.
list of cases
375
European Communities – Measures Affecting the Approval and Marketing of Biotech Products (29 September 2006), Panel Report, WT/DS291.292.293. European Communities – Measures Concerning Meat and Meat Products (Hormones case) (16 January 1998), Appellate Body Report, WT/DS48/AB/R. European Communities – Measures Concerning Meat and Meat Products (Hormones) (13 February 1998), Appellate Body Report, WT/DS26/AB/R. European Communities – Regime for the Importation, Sale And Distribution of Bananas (22 May 1997), Panel Report, WT/DS27/R/HND. European Communities – Regime for the Importation, Sale And Distribution of Bananas (9 September 1997), Appellate Body Report, WT/DS27/AB/R. European Communities – Anti-Dumping Duties on Imports of Cotton-Type Bed Linen from India (12 March 2001), Appellate Body Report, WT/DS141/AB/R. European Communities – Customs Classification of Frozen Boneless Chicken Cuts (30 May 2005), Panel Report, WT/DS269/R & WT/DS286/R. European Communities – Customs Classification of Frozen Boneless Chicken Cuts (27 September 2005), Appellate Body Report, WT/DS269/AB/R & WT/DS286/AB/R. European Communities-Measures Affecting the Approval and Marketing of Biotech Products, Interim Reports of the Panel, 7 February 2006, WT/DS291/INTERIM: WT/DS292/ INTERIM; WT/DS293/INTERIM. Import, Distribution and Sale of Alcoholic Drinks by Canadian Provincial Marketing Agencies (22 March 1988), Panel Report, L/6304-35S/37Japan – Taxes on Alcoholic Beverages (4 October 1996), Appellate Body Report, WT/DS8/AB/R, WT/DS10/AB/R; WT/DS11/AB/R/Section D. Mexico – Tax Measures on Soft Drinks and Other Beverages (6 March 2006), Appellate Body Report, WT/DS308/AB/R, AB-2005-10. Mexico – Tax Measures on Soft Drinks and Other Beveragesl (7 October 2005), Report of the Panel, WT/DS308/R. Thailand – Anti-Dumping Duties on Angles, Shapes and Sections of Iron or Non-Alloy Steel and H-Beams from Poland (5 April 2001), Appellate Body Report, WT/DS122/AB/R. Thailand – Restrictions on Importation of and Internal Taxes on Cigarettes (7 November 1990), Panel Report, DS10/R-37S/200. United States – Anti-Dumping Measures on Certain Hot-Rolled Steel Products from Japan (23 August 2001), Appellate Body Report, WT/DS184/AB/R. United States – Combed Cotton Yarn from Pakistan (5 November 2001), Appellate Body Report, WT/DS192/R. United States – Continued Existence and Application of Zeroing Methodology (19 February 2009), Appellate Body Report, WT/DS350/AB/R. United States – Countervailing Duties on Certain Corrosion-Resistant Carbon Steel Flat Products from Germany (28 November 2002), Appellate Body Report, WT/DS213/AB/R. United States – Countervailing Duties on Non-Rubber Footwear from Brazil (3 June 1995), Panel Report, SCM/94. United States – Final Anti-Dumping Measures of on Stainless Steel from Mexico (20 December 2007), Panel Report, WT/DS334/R. United States – Import Prohibition of Certain Shrimp and Shrimp Products (15 May 1998), Panel Report, WT/DS58/R. United States – Import Prohibition of Certain Shrimp and Shrimp Products (6 November 1998), Appellate Body Report, WT/DS58/AB/R. United States – Imposition of Countervailing Duties on Imports of Fresh and Chilled Atlantic Salmon from Norway (28 April 1994), Panel Report, SCM/153. United States – Measures Affecting the Cross-Border Supply of Gambling and Betting Services (4 July 2005), Appellate Body Report, WT/DS285/AB/R. United States – Prohibition of Imports of Tuna and Tuna Products from Canada (22 February 1982), Panel Report L/5198 – 29S/91. United States – Restrictions on Import of Sugar (22 June 1989), Panel Report, Supp. BISD 36S/331. United States – Restrictions on Imports of Tuna (16 August 1991), Panel Report, ILM 30 (1991): 1594.
376
list of cases
United States – Restrictions on Imports of Tuna (22 February 1982), Panel Report, L/519829S91. United States – Safeguard Measures on Import of Fresh, Chilled or Frozen Lamb Meat from New Zealand and Australia (US – Lamb) (16 May 2001), Panel Report, WT/DS177/R & WT/ DS178/R. United States – Standards for Reformulated and Conventional Gasoline (20 May 1996), Appellate Body Report, WT/DS2/AB/R. United States – Standards for Reformulated and Conventional Gasoline (29 January 1996), Panel Report, WT/DS2/R. US Section 337 of the Tariff Act of 1930 (7 November 1989), Panel Report, L/6439-36S/345.
ECtHR Bankovic v. Belgium and Others, ILM 41 (2001): 517. Bianchi v. Switzerland (2006), Application no. 7548/04. Bosphorus Hava Yollari Turizm Ve Ticaret Anonim Sirketi v. Ireland, EHRR 42 (2006): 1. Engel and Others v. the Netherlands, EHRR 1 (1976): 647. Golder v. United Kingdom EHRR 1 (1975): 574. Loizidou v. Turkey, EHRR 23 (1997): 513. Mamatkulov and Askarov v. Turkey, EHRR 41 (2005): 494. Manoilescu and Dobrescu v. Romania and Russia (2005), Application no. 60861/00. Matthews v. United Kingdom, EHRR 28 (1999): 361. National Union of Belgian Police v. Belgium, Separate opinion of Judge Sir Gerald Fitzmaurice, EHRR 1 (1975): 578. Nikolaus and Jurgen Treska v. Albania and Italy (2006), Application no. 26937/04. Tyrer v. United Kingdom, EHRR 2 (1978): 1. Young, James and Webster v. United Kingdom, EHRR 4 (1981): 38.
ECJ Case 29/69, Stauder (1969), ECR 419, CMLR 19 (1970): 112. Case C-17/96, Badische Erfrischungs-Getränke (1997), ECR I-4617; CMLR 1 (1998): 341. Case C-175/82, Dinter v. Hauptzollamt Köln-Deutz (1983), ECR 969. Case C-199/05, European Community v. Belgium (2006), Opinion of AG Stix-Hackl, ECR: I-10485. Case C-203/07 P, Greece v. Commission (2008), Opinion of AG Mazák. Case C-277/93, Commission v. Spain (1994), Opinion of AG Tesauro, ECR (1995): I-5515; CMLR 1 (1995): 866. Case C-33/92, Fleisch GmbH v. Oberfinanzdirektion Hamburg (1993), ECR I-3047. Case C-500/99 P, Conserve Italia v Commission (2002), ECR: I-867. Case C-80–96, Quelle Schickedanz (1998), ECR I-123; CMLR 1 (1998): 373.
ICTY Prosecutor v. Dusko Tadic a.k.a ‘Dule’, Appeals Decision on the Defence Motion for Interlocutory Appeal on Jurisdiction of 2 October 1995, ICTY, Case No. IT-94-1-AR72. Prosecutor v. Dusko Tadic, Decision of 10 August 1995 of the Trial Chamber on the Defence Motion on Jurisdiction, ICTY, (IT-94-1-T). Prosecutor v. Dusko Tadic, Decision of 2 October 1995 of the Appeals Chamber on the Defence Motion for Interlocutory Appeal on Jurisdiction, ICTY, (IT-94-1-AR72). Prosecutor v. Tihomir Blaskic, Decision of 18 July 1997 of Trial Chamber II on the Objection of the Republic of Croatia to the Issuance of Subpoena Duces Tecum, ICTY, (IT-95-14).
list of cases
377
Prosecutor v. Furundzija, Judgment of 10 December 1998 of Trial Chamber II, ICTY, (IT-9517/1). Prosecutor v. Slobodan Milosevic, Decision of 8 November 2001 of the Trial Chamber on Preliminary Motions, ICTY, (IT-02-54-PT).
ICTR Prosecutor v. Joseph Kanyabashi, Decision of 18 June 1997 of Trial Chamber II on the Defence Motion on Jurisdiction, ICTR, (ICTR-96-15-T).
Special Court for Sierra Leone Prosecutor v. Issa Hassan Sesay; Prosecutor v. Morris Kallon; Prosecutor v. Augustine Gbao (RUF Case), Appeal Judgment of 26th October 2009, Special Court for Sierra Leone, (SCSL-0415-A-1311). Prosecutor v. Morris Kallon and Brima Bazzy Kamara, Decision of 13 March 2004 of the Appeals Chamber on Challenge to Jurisdiction: Lome Accord Amnesty, Special Court for Sierra Leone, (SCSC-04-15-PT-060). Prosecutor v. Morris Kallon, Decision of 13 March 2004 of the Appeals Chamber on Challenge to Jurisdiction: Lome Accord Amnesty, Special Court for Sierra Leone, (SCSL-04-15-PT060-I & II). Prosecutor v. Brima Bazzy Kamara, Decision of 13 March 2004 of the Appeals Chamber on Challenge to Jurisdiction: Lome Accord Amnesty, Special Court for Sierra Leone, (SCSL-0416-PT-033-I & II). Prosecutor v. Morris Kallon, Decison of 13 March 2004 of the Appeals Chamber on Constitutionality and Lack of Jurisdiction, Special Court for Sierra Leone, (SCSL-2004-15PT-059-I & II). Prosecutor v. Sam Hinga Norman, Decison of 13 March 2004 of the Appeals Chamber on Constitutionality and Lack of Jurisdiction, Special Court for Sierra Leone, (SCSL-2004-14PT-035-I & II). Prosecutor v. Brima Bazzy Kamara, Decison of 13 March 2004 of the Appeals Chamber on Constitutionality and Lack of Jurisdiction, Special Court for Sierra Leone, (SCSL-2004-16PT-032-I & II). Prosecutor v. Allieu Kondewa, Decision of 25 May 2004 of the Appeals Chamber on Lack of Jurisdiction/ Abuse of Process: Amnesty Provided by the Lome Accord, Special Court for Sierra Leone, (SCSL-04-14-T-128-7347). Prosecutor v. Augustine Gbao, Decision of 25 May 2004 of the Appeals Chamber on Preliminary Motion on the Invalidity of the Agreement between the United Nations and the Government of Sierra Leone on the Establishment of the Special Court, Special Court for Sierra Leone, (SCSL-04-15-PT-141). Prosecutor v. Moinina Fofana, Decision of 25 May 2004 of the Appeals Chamber on Preliminary Motion on Lack of Jurisdiction: Illegal Delegation of Jurisdiction by Sierra Leone, Special Court for Sierra Leone, (SCSL-04-14-PT-102). Prosecutor v. Monina Fofana, Decision of 25 May 2004 of the Appeals Chamber on Preliminary Motion on Lack of Jursidiction – Illegal Delegation of Powers by the United Nations, Special Court for Sierra Leone, (SCSL-2004-14-PT-100-6836 & 6846). Prosecutor v. Taylor, Decision of 31 May 2004 of the Appeals Chamber on Immunity from Jurisdiction, Special Court for Sierra Leone, (SCSL-2003-01-I-059). Prosecutor v. Norman, Decision of 31 May 2004 of the Appeals Chamber on Preliminary Motion Based on Lack of Jurisdiction (Child Recruitment), Special Court for Sierra Leone, (SCSL2004-14-AR72(E)-131-7383, 7398, 7413 & 7430). Prosecutor v. Charles Taylor, Appeals Chamber Decision of 31 May 2004, Special Court for Sierra Leone, (SCSL-03-01-I-059).
378
list of cases Iran-US Claims Tribunal
Amoco International Finance Corporation v. Iran (1987) Iran-US CTR 15: 189. Iran v. The United States of America, Case No. Bl (Counterclaim), Interlocutory Award of 9 September 2004, Award No. ITL 83-B1-FT.
Arbitration & National Cases A. Goetz v. Burundi, Award, 10 February 1999, ICSID Reports 6: 5. AAPL v. Sri Lanka, Award, 27 June 1990. ADF Group, Inc. v. United States of America, Award, 9 January 2003, ICSID Reports 6: 470. AES Corp. v. Argentina, Decision on Jurisdiction, 26 April 2005. Affaire Concernant L’Apurement des Comptes entre le Royaume des Pays-Bas et la Republique Francaise en Application du Protocole du Septembre 1991 Additionnel à la Convention Relative à la Protection du Rhin contre la Pollution par les Chlorures du 3 Decembre 1976, Sentence Arbitrale du 12 Mars 2004. Aguas del Tunari, S.A. v. Bolivia, Decision on Jurisdiction, 21 October 2005. Air Transport Services Agreement Arbitration (United States v France), Award of 22 December 1963, reprinted in ILR 38 (1969): 182. Amco v. Indonesia, Award, of 20 November 1984, ICSID Reports 1: 413. Amco v. Indonesia, Decision on Annulment, 16 May 1986, ICSID Reports 1: 521. Amco v. Indonesia, Decision on Jurisdiction, 25 September 1983, ICSID Reports 1: 389. Antippa v. Germany, Recueil des Décisions desTribunaux Arbitraux Mixtes VII (1928). Aschberg Hopwood and Crew, Ltd. v. Quaritch, reprinted in Recueil des Décisions des Tribunaux Arbitraux Mixtes V (1926): 332. Azurix v. Argentina, Award of 14 July 2006, ICSID ARB/01/12. Bayindir v. Pakistan, Decision on Jurisdiction, 14 November 2005. Beagle Channel Arbitration (Argentina v. Chile) (1977), ILR 52: 93. Betancourt Arbitration, Moore Arbitration: Vol. V, 939. Bogdanov v. Moldova, Award, 22 September 2005. Camuzzi v. Argentina, Decision on Jurisdiction, 11 May 2005. Canfor v. United States, Tembec v. United States, Terminal Forest Products v. United States, Order of the Consolidation Tribunal, 7 September 2005. Cayuga Indians Claims Case, American and British Claims Arbitration Tribunal, Award of 22 January 1926, reprinted in AJIL 20 (1926): 587. Ceskoslovenska Obchodni Banka, a.s. v. The Slovak Republic, Award on Jurisdiction of 24 May 1999, ICSID ARB/97/4. CME v. The Czech Republic, Final Award, 14 March 2003, ICSID Reports 9: 264. CME v. The Czech Republic, Partial Award, 13 September 2001, ICSID Reports 9: 121. CMS Gas Transmission Company v. Argentina, Award of 12 May 2005, ICSID ARB/01/8. Compania de Aguas del Aconquija, S. A. & Vivendi Universal (formerly Compagnie Generale des Eaux) v. Argentine Republic, Decision on Jurisdiction, 14 November 2005. Compania de Aguas del Aconquija, S. A. & Vivendi Universal (formerly Compagnie Generale des Eaux) v. Argentine Republic, Decision on Annulment, 3 July 2002, ICSID Reports 6: 340. Continental Casualty Co. v. Argentina, Decision on Jurisdiction, 22 February 2006. Continental Casualty Co. v. Argentine Republic, Award of 5 September 2008, ICSID ARB/03/9. Corn Products International Inc. v. the United Mexican States, Partial Award on Responsibility of 15 January 2008, ICSID Case No. ARB(AF)/04/01, <www.naftaclaims.com/Disputes/Mexico/ CPI/CPI-Mexico-MeritsAward.pdf> (accessed on 1 May 2009). David. J. Adams case, Annual Digest (1919–1922), Case No. 243. Dispute concerning the course of the frontier between B.P. 62 and Mount Fitzroy (Laguna del Desierto case) (Argentine v. Chile), Arbitral Award of 21 October 1994, ILR 113: 76. Eastern Bank Ltd. v. the Turkish Government, Recueil des Décisions desTribunaux Arbitraux Mixtes VIII (1929): 188. El Paso Energy v. Argentina, Decision on Jurisdiction of 27 April 2006, Case ARB/03/15. EnCana v. Ecuador, Award, 3 February 2006. Enron Corp. and Ponderosa Assets, L.P. v. Argentina, Decision on Jurisdiction, 14 January 2004.
list of cases
379
Enron Corporation and Ponderosa Assets, L.P. v. Argentina, Award of 22 May 2007, ICSID ARB/01/3. Enron v. Argentina, Decision on Jurisdiction (Ancillary Claim), 2 August 2004. Enron v. Argentina, Decision on Jurisdiction, 14 January 2004. Ethyl Corp. v. Canada, Decision on Jurisdiction, 24 June 1998, ICSID Reports 7: 12. Eureko v. Poland, Partial Award, 19 August 2005. Feldman v. Mexico, Award, 16 December 2002, ICSID Reports 7: 341. Gas Natural SDG, S.A. v. The Argentine Republic, ICSID Case No. ARB/03/10, Decision of the Tribunal on Preliminary Question on Jurisdiction, 17 June 2005. Grand River Enterprises Six Nations Ltd. v. United States, Decision on Jurisdiction, 20 July 2006. Greece (on behalf of Apostolidis) v. Federal Republic of Germany Arbitration, ILR 34 (1960): 242. In the Consolidated Arbitration under Chapter Eleven of the North American Free Trade Agreement (NAFTA) and the UNCITRAL Arbitration Rules between the Canadian Cattlemen for Fair Trade (CCFT) and the United States of America, Award on Jurisdiction of 28 January 2008, <www.naftaclaims.com/Disputes/USA/CCFT/CCFT-USA-Award.pdf> (accessed on 1 May 2009). In the Matter of an Arbitration under Chapter Eleven of the North American Free Trade Agreement between ADF Group Inc. and the United States of America, Award of 9 January 2003, ICSID Case No. ARB(AF)/00/1, <www.state.gov/documents/organization/16586.pdf> (accessed on 1 May 2009). In the Matter of Arbitration under Chapter Eleven of the NAFTA and the UNCITRAL Rules, between Methanex Corporation and the United States of America, Decison. of Tribunal on the Petition from Third Persons to intervene Amici Curiae of 15 January 2001, <www.naftalaw .org/methanex/Methanex-Amicus.Decision.pdf> (accessed on 1 May 2009). Industria Nacional de Alimentos, S.A. and Indalsa Perú, S.A. v. Peru, (Previously Empresas Lucchetti, S.A. and Lucchetti Perú, S.A. v. Peru), Decision on Annulment of 5 September 2007, ICSID ARB/03/4. Italy v. Federal Republic of Germany Arbitration, ILR 29 (1959): 448. Jan de Nul & Dredging International v. Egypt, Decision on Jurisdiction, 16 June 2006. Joy Mining v. Egypt, Award, 6 August 2004. Kummerow Arbitration (1892), Moore Arbitrations: Vol. V, 526. Lauder v. Czech Republic, Award, 3 September 2001. LETCO v. Liberia, Award, 31 March 1986, ICSID Reports 2: 346. LG&E Energy Corp., LG&E Capital Corp., LG&E International Inc. v. Argentina, Decision on Liability of 3 October 2006, ICSID ARB/02/1. Loewen Group, Inc. and Raymond L. Loewen v. United States of America, Award, 26 June 2003, ICSID Reports 7: 442. Loewen v. United States, Decision on Jurisdiction, 9 January 2001. Maffezini v. Spain, Decision on Jurisdiction, 25 January 2000, ICSID Reports 5: 396. Metalclad Judicial Review case, British Columbia Law Review 89 (2001). Methanex Corp. v. United States, UNCITRAL (NAFTA), Final Award of 3 August 2005. Methanex v. United States, Decision on Jurisdiction, 7 August 2002. Mines Aroa Arbitration, (UK v. Venezuela) (1903), reprinted in Moore Arbitrations: Vol. V, 498. Mondev International Ltd. v. United States of America, Award, 11 October 2002, ICSID Reports 6: 192. MTD v. Chile, Award, 25 May 2004. NAFTA Arbitral Panel Established Pursuant to Art. 2008 – In the Matter of Tariffs Applied by Canada to Certain United States Origin Agricultural Products (2 December 1996), Final Panel Report, Sec. File ODA 95-2008-01. NAFTA Arbitral Panel Established Pursuant to Chapter Twenty – In the Matter of Cross-Border Trucking Services (6 February 2001), Final Panel Report, Sec. File No. USA-MEX-9820080-01. National Grid v. Argentine Republic, Decision on Jurisdiction, 20 June 2006. Neer Case (US v. United Mexican States), Decision of 15 October 1926, RIAA 4: 60. Noble Ventures v. Romania, Award, 12 October 2005. Occidental Exploration and Production Co. v. Ecuador, Award, 1 July 2004. Pan American Energy v. Argentina, Decision on Preliminary Objections, 27 July 2006.
380
list of cases
Plama Consortium Limited v. Bulgaria, (Energy Charter Treaty), Decision on Jurisdiction of 8 February 2005, ICSID ARB/03/24. Polyxene Plessa v. the Turkish Government, Recueil des Décisions desTribunaux Arbitraux Mixtes VIII (1929): 224. Pope & Talbot Inc. and the Government of Canada, Award in Respect of Damages of 31 May 2002. Pope & Talbot Inc. and the Government of Canada, Interim Award of 26 June 2000. Pope and Talbot v. Canada, Award on the Merits, 10 April 2001, ICSID Reports 7: 102. Regina v. Bow Street Metropolitan Stipendiary Magistrate and others exparte Pinochet Ugarte (No. 3), Judgment of 24 March 1999, House of Lords, ILR 119: 136. Ronald S. Lauder v. The Czech Republic, Final Award, 3 September 2001, ICSID Reports 9: 66. Rumanian Minister of War v. The Turkish Government, reprinted in Recueil des Décisions des Tribunaux Arbitraux Mixtes, institúes par les Traités de Paix, Gilbert Gidel (ed.) (Paris, Recueil Sirey, 1928): Vol. VII, 996. S. D. Myers Inc. and the Government of Canada, Award of 12 November 2002. Salini Costruttori S.p.A. and Italstrade S.p.A. v. Jordan, Decision on Jurisdiction of 9 November 2004, ICSID ARB/02/13. Salini Costruttori SpA et Italstrade SpA c/Royaume du Maroc, Decision on Jurisdiction, 23 July 2001, Journal de Droit International (2002): 196, ICSID Reports 6: 400. Salini v. Jordan, Decision on Jurisdiction, 29 November 2004. Saluka Investments BV v. Czech Republic, UNCITRAL, Partial Award of 17 March 2006. Sambaggio Arbitration, Moore Arbitration: Vol. V, 666. Sempra Energy International v. Argentina, Award of 28 September 2007, ICSID ARB/02/16. Sempra Energy Intl. v. Argentina, Decision on Jurisdiction, 11 May 2005. SGS Societe Generale de Surveillance S.A. v. Islamic Republic of Pakistan, ICSID Case N° ARB/01/13. SGS Societe Generate de Surveillance S.A v. Republic of the Philippines, ICSID Case N° ARB/02/6. SGS v. Pakistan, Decision on Jurisdiction, 6 August 2003, ICSID Reports 8: 383. SGS v. Philippines, Decision on Jurisdiction, 29 January 2004, ICSID Reports 8: 518. Sierra Leone Telecommunications Co. Ltd v. Barclays Bank PLC (1998), All ER 2: 821. Suez v. Argentina, AWG Group v. Argentina, Decision on Jurisdiction, 3 August 2006. Suez v. Argentina, Decision on Jurisdiction, 16 May 2006, ICSID Case No. ARB/03/17. Tokios Tokeles v. Ukraine, Decision on Jurisdiction, 29 April 2004. United Mexican States v. Metalclad Corp., Judgment, Supreme Court of British Columbia, 2 May 2001, ICSID Reports 5: 236. United Parcel Service of America Inc. v. Canada, Decision of 17 October 2001. United Parcel Service of America, Inc. v. Canada, Award, 22 November 2002, ICSID Reports 7: 288. Waste Management, Inc. v. United Mexican States, Award, 30 April 2004. Wena Hotels v. Egypt, Decision on Jurisdiction, 29 June 1999, ICSID Reports 6: 74. Young Loan Arbitration, (UK, US, France, Belgium and Switzerland v. Federal republic of Germany), Arbitral Award of 16 May 1980, ILR 59 (1980): 494.
INDEX adoption of treaty 66, 257 autonomous concepts 257, 259 bona fide See principles, of good faith Biotech case 175–9, 228 CCFT v. US case 217–32 circumstances of conclusion 158, 208, 210 customary law Vienna rules as 5, 9, 20, 23–30, 65–9, 102–5, 130, 133–4, 153–4 fragmentation 19, 31, 40, 56, 76, 79, 151, 176, 179, 236 Harvard Draft 87 ICSID See International Centre for Settlement of Investment Disputes Institut de Droit International 68 International Centre for Settlement of Investment Disputes 139–51, 239–54 interpretation as art or science 1–13, 17, 25, 105 authentic 62, 112, 114–5, 138, 148, 248 contextual 49–52, 118, 156–7, 204, 219, 228–30, 236, 243 dynamic See interpretation, evolutionary/evolutive effective 132–4 evolutionary/evolutive 119–20, 171, 206, 263, 265, 267, 271–2 grammatical See interpretation, textual of customary norms 7–8, 227–8 restrictive 132–4, 195, 206, 243, 247–8 textual 28–9, 84, 85–6, 121, 154–7, 164, 214, 221, 228, 243, 258–9, 266 teleological 26, 29, 107, 120, 132, 180 virtuous 17–38 Interpretative Note 199 intertemporal law doctrine of 4, 67–8, 228 renvoi fixe 68 renvoi mobile 68 jus cogens 58, 108–9, 314–5, 321–3, 328, 343–4 Kasikili/Sedudu Island 59–74, 154, 157–8, 223–4, 250
NAFTA See North American Free Trade Agreement North American Free Trade Agreement 50–1, 124, 129, 136, 138, 140, 147–8, 153, 196–233, 247, 249 object and purpose 5, 57, 72, 102, 104, 107, 116–7, 130–3, 136, 155–7, 161–2, 165, 167–182, 187, 190, 198, 205, 207, 209, 219, 220–1, 233–4, 236, 243, 245, 247, 257, 260–1, 265, 269–70, 272, 316, 323 ordinary meaning 5, 28, 41, 62–3, 67, 70–2, 88, 102, 116, 118, 130, 154, 157, 159, 167–72, 178, 181, 198–9, 205, 209, 214–5, 224, 245–6, 316 Oil Platforms case 18, 34, 40– 57, 65, 154, 178, 179, 241, 244 PCA See Permanent Court of Arbitration PCIJ See Permanent Court of International Justice Permanent Court of Arbitration 11, 18, 52, 54 Permanent Court of International Justice 62, 76, 78, 80, 83, 87–8, 90, 106, 134, 157, 192–3, 229, 320 subsequent agreement 59–74, 136, 148, 205–7, 209, 222–5, 236, 248–50 subsequent practice 59–74, 104–5, 146, 198, 205, 209–10, 222–5, 244–5, 249–52, 350–3 preamble 104, 126, 131, 133, 157, 170, 172, 197–9, 204, 260, 273, 276, 280–1, 285–6, 291, 309–10 preparatory work 66, 75–95, 137–9, 157–8, 201, 209–10, 225–32, 257–61, 317, 329 principles of good faith 30, 37, 41, 78, 80, 84, 102, 109, 130, 156, 157, 172, 180, 198, 213, 215, 240, 241, 242, 243, 316, 349 of non-retroactivity 67, 190–5, 216–7, 250, 253, 334, 351–3 of systemic integration 11, 34, 48, 56, 57, 179, 34, 39–58, 175–9, 186, 228, 230, 235–7, 316, 351, 353–4
382
index
pacta sunt servanda 37, 52, 156, 198, 213 ut res magis valeat quam pereat 155, 179, 181, 183 res judicata 100, 139–44, 249, 250 travaux préparatoires See preparatory work
treaties amendment of 61, 63, 212, 217, 248–50, 262, 296, 303, 309, 319, 326, 348, 350, 3543 modification of 62, 245, 248–51, 253, 292 reservations of 199–200, 268–69, 322–3, 325–6, 336