PROMISES OF STATES UNDER INTERNATIONAL LAW

Textbooks on international law, dicta of the International Court of Justice

and the International Law Commissions Guiding Principles applicable
to unilateral declarations of states capable of creating legal obligations of
2006, all reflect the fact that in international law a states unilateral declaration can create a legally binding obligation. Unilateral declarations are
common, as a look at the weekly headlines of any major newspaper will
reveal. Many of the declarations made at the highest level are, of course,
vaguely expressed and carry no tangible legal commitment. But others
deliver a very clear message: for instance the USs April 2010 declaration
on its future use of nuclear weapons or Kosovos declaration of independence and pledge to follow the Ahtisaari Plan, are two recent and prominent examples of unilateral declarations at the international level.
The same sources, however, also reveal that while state promises are
accepted as a means for states to create full blown legal commitments,
the law governing such declarations is far from clear. This monograph
fills a gap in international legal scholarship by raising and answering the
question of the precise legal value of such pledges in the realm of public
international law.
After a brief introduction state promises in international law are defined
and contrasted with other unilateral acts of states, and the history of
promises in state practice and court decisions is delineated, together with
scholarly opinion. The book then provides a detailed picture of the international legal framework governing promises of states, and ends with a
brief assessment of the raison dtre for promises as a binding mechanism
in international law, along with their advantages and disadvantages in
comparison with the classical mechanism for assuming international obligations the international treaty.
This is currently the only book to present a comprehensive overview of
the legal effect of promises by states in international law.
Volume 37 in the series Studies in International Law

Studies in International Law

Recent titles in this series
Human Dignity and the Foundations of International Law
Patrick Capps
Sovereignty and the Stateless Nation: Gibraltar in the Modern Legal
Context
Keith Azopardi
The International Court of Justice and Self-Defence in International Law
James A Green
State Liability in Investment Treaty Arbitration: Global Constitutional and
Administrative Law in the BIT Generation
Santiago Montt
Reappraising the Resort to Force: International Law, Jus ad Bellum and
the War on Terror
Lindsay Moir
International Law and Dispute Settlement: New Problems and Techniques
Edited by Duncan French, Matthew Saul and Nigel White
The Democratic Legitimacy of International Law
Steven Wheatley
Reflections on the UN Declaration on the Rights of Indigenous Peoples
Edited by Stephen Allen and Alexandra Xanthaki
Contracting with Sovereignty: State Contracts and International Arbitration
Ivar Alvik
Multi-Sourced Equivalent Norms in International Law
Edited by Yuval Shany and Tomer Broude
The Distinction and Relationship between Jus ad Bellum and Jus in Bello
Keiichiro Okimoto
International Humanitarian Law and Terrorism
Andrea Bianchi and Yasmin Naqvi
Promises of States under International Law
Christian Eckart
For the complete list of titles in this series, see
Studies in International Law link at
www.hartpub.co.uk/books/series.asp

Promises of States under

International Law
Christian Eckart

OXFORD AND PORTLAND, OREGON

2012

Published in the United Kingdom by Hart Publishing Ltd

16C Worcester Place, Oxford, OX1 2JW
Telephone: +44 (0)1865 517530
Fax: +44 (0)1865 510710
E-mail: mail@hartpub.co.uk
Website: http://www.hartpub.co.uk
Published in North America (US and Canada) by
Hart Publishing
c/o International Specialized Book Services
920 NE 58th Avenue, Suite 300
Portland, OR 97213-3786
USA
Tel: +1 503 287 3093 or toll-free: (1) 800 944 6190
Fax: +1 503 280 8832
E-mail: orders@isbs.com
Website: http://www.isbs.com
Christian Eckart 2012
Christian Eckart has asserted his right under the Copyright, Designs and Patents Act 1988,
to be identified as the author of this work.
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission
of Hart Publishing, or as expressly permitted by law or under the terms agreed with
the appropriate reprographic rights organisation. Enquiries concerning reproduction which
may not be covered by the above should be addressed to Hart Publishing Ltd
at the address above.
British Library Cataloguing in Publication Data
Data Available
ISBN: 978-1-84946-232-7
Typeset by Hope Services, Abingdon
Printed and bound in Great Britain by
TJ International Ltd, Padstow, Cornwall

Foreword
by Christian Tomuschat
The principle of sovereignty cannot be equated with freedom from any
legal constraints. Precisely because of their sovereign powers, states are
capable of binding themselves by assuming international obligations.
What the Permanent Court of International Justice said in its very first
judgment in the Wimbledon case,1 applies not only to international treaties, but also to specific unilateral acts. The author of the present book
is not the first one to discover this ground rule of contemporary international law, but he is indeed the first lawyer to explain its raison dtre
in a fully persuasive fashion. When the International Court of Justice in
1974 declared the proceedings in the Nuclear Tests cases moot,2 thereby
relying on statements of a number of high-ranking French governmental
office holders that atmospheric nuclear tests in the Pacific Ocean would
not be continued, the legally binding force of unilateral declarations was
far from unequivocally established. Indeed, unilateral acts do not appear
in the list of legal sources referred to in Article 38 of the ICJ Statute. Here
and there, hints had surfaced in international jurisprudence that a state
was bound to honour promises it had made vis--vis another state. But no
cohesive doctrine had evolved from those bits and pieces.
But the legal position remained opaque. Why should such a promise,
made without a quid pro quo, produce a true legal obligation? Was it not
extremely dangerous to hold a state accountable for purely verbal declarations lacking any formality? Could not states be trapped, particularly
in cases where they attempted to show generosity towards their partners,
but strictly on a political plane? Indeed, international relations are rife
with unilateral acts. On a daily basis, governments have to explain their
choices not only before their own peoples, but also before their foreign
partners and before the international community. It is obvious that openness should not lead to legal imprisonment. The basic fact is that states are
sovereign entities and that any obligations, to the extent that they do not
flow from general international law, must be accepted by them. Their will
must not be misinterpreted. The international legal order would be threatened in its entirety if, on a regular basis, states had to defend themselves
against unintended interpretations of their acts and utterances.
Series A 1, 17 August 1923.
Nuclear Tests (Australia v France) [1974] ICJ Rep 253, 272; Nuclear Tests (New Zealand v
France) [1974] ICJ Rep 457, 478.
1
2

vi Foreword
After it had terminated its main pieces of codification on the law of
treaties and on state responsibility, it was tempting for the ILC to put the
topic Unilateral Acts of States on its agenda in order to clarify the many
questions which the judgments of the ICJ in the Nuclear Tests cases had left
open. Many thought that the new topic could be addressed in parallel to
the law of treaties: the basic axiom, the proposition pacta sunt servanda,
could simply be converted into declaratio est servanda. In the course of the
work of the ILC from 1996 to 2006, it turned out, however, that the problematique was more complex than originally presumed. In particular, the
ILC and its rapporteur, Vctor Rodrguez Cedeo from Venezuela, realised that unilateral acts comprised a panoply of different forms of conduct
each of which had its specific features. For instance, protests, on the one
side, and acts of recognition, on the other, may have as common characteristic their origin as unilateral acts. Nonetheless, their function is so
widely different that it would be hardly possible to conceive of a common
legal regime for them. Accordingly, the topic was subsequently restricted.
The final outcome was in 2006 a short elaboration on Guiding Principles
applicable to Unilateral Declarations of States Capable of Creating Legal
Obligations.
The 10 legal principles encompassed in that statement are far from
addressing all of the legal issues that require to be answered in respect of
unilateral declarations. In general, they refrain from laying down clearcut propositions. The reluctance to provide more than tentative answers
is manifested already in the first Principle which states that declarations
manifesting the will to be bound may have the effect of creating legal
obligations. As far as the crucial problem of revocation is concerned, the
Guiding Principles confine themselves to setting forth that unilateral declarations may not be revoked arbitrarily (Guiding Principle 10). On the
whole, the reader cannot but feel that the legitimate needs of legal practice
have not been fully satisfied.
It may well be that at the present stage of legal development no better
responses could be given. The codification process conducted by the ILC
needs firm support from the solutions found in the actual transactions
among nations. Hence, the outcome of the work of the ILC amounted to
a call to continue the search for the appropriate legal regime of unilateral
acts. The author of these lines wrote a short contribution as a first assessment of the Guiding Principles.3 Additionally, he encouraged Christian
Eckart to proceed to a more deep-going study of the topic. The final
result of this endeavour, the present book, has succeeded in clarifying
the systemic foundations of the legal configuration of unilateral promises. Indeed, quite rightly, the author concluded that he should confine
3
Unilateral Acts under International Law in Droits et culture. Mlanges en lhonneur du
Doyen Yadh Ben Achour (Tunis, Centre de Publication Universitaire, 2008) 1487507.

Forewordvii

his study to that specific sector instead of trying to embrace the field of
unilateral acts in its entirety.
The great achievement of the book is that it has been able to demonstrate
the usefulness and legitimacy of the concept of promise in international
law. International treaties have a much more precise profile. Long-standing
experiences determine their scope and meaning. Unilateral promises, on
the other hand, have the great advantage of allowing for more flexibility.
States do not succumb to the rigidity which a treaty normally engenders
by force of the proposition pacta sunt servanda. Yet, promises given with
the intention to specify the direction of future conduct are able to create a
considerable degree of legitimate confidence with their addressees. Thus,
they are intimately connected to the principle of good faith, one of the
fundamental axioms included in the Friendly Relations Declaration of the
UN General Assembly of 19704 (Principle 7). Necessarily, therefore, promises cannot be revoked on the spur of the moment, abruptly and without
any valid reasons. Following the precedent of Article 56(2) of the Vienna
Convention on the Law of Treaties, and the judgment of the International
Court of Justice in the Nicaragua case,5 the author suggests that generally states should have the right to revoke a promise made by them on
the basis of 12 months notice. This is a compromise solution which, on
the one hand, confirms the legal bindingness of unilateral promises but
wisely avoids any kind of dogmatic rigidity.
The book contains a full discussion of all the issues which unilateral
promises can raise in diplomatic practice. It may thus incite the ILC to take
up the topic again, which in 2006 was brought to a rapid end because the
ILC felt that it had not managed to analyse all of its implications in a sufficiently thorough fashion. In sum, the book may be called an outstanding
complement to the existing legal literature on a key issue of contemporary
international law. It will be indispensable reading for any lawyer interested in the concept of unilateral promises in international law.

GA Resolution 2625 (XXV), 24 October 1970.

Military and Paramilitary Activities in and against Nicaragua (Nicaragua v United States) [1984]
ICJ Rep 392, 420, para 63.
4
5

Acknowledgements
I would like to thank the Deutsche Forschungsgemeinschaft (DFG) and
the Research Training Group on Multilevel Constitutionalism at the
Humboldt University, Berlin for the fellowship which provided me with
the financial means and necessary time to write this book. The Research
Training Groups first generation in the form of its founders, organisers,
guests and especially its members deserve an additional thank you for
their intellectual input, as well as for what turned out to be great company during this otherwise solitary time of research and writing. I would
like to express my gratitude to Professor Christian Tomuschat for having
pointed me towards the International Law Commissions struggle with
Unilateral Acts of States and thereby also the topic of this book, as well as
for his continuing intellectual support and encouragement. I finally wish
to thank my family, especially my mother Ann, on whom I could always
rely to discuss subtle nuances in the English language, as well as my wife
Maria and daughter Emilia, whose zero tolerance for glassy eyes after
work kept me firmly grounded in everyday life.

Contents
Foreword by Christian Tomuschat
v
Acknowledgements ix
Abbreviations xv
Table of Cases
xvii
Introduction 1

I
The Presumed Rareness of Promises
4

II A Legal Framework with Many Open Questions:
A First Look
12

III The Need for a Clear Assessment of the Applicable Law
14

IV Approaching the Topic
17
1

Delimiting the Subject: Promise as a Unilateral Act 18

I
Drawing from the Definition of Unilateral Acts
18
II Distinguishing Promises from Other Classes of Unilateral
Acts
23

A Promise
28
B Recognition
29
C Protest
33
D Waiver/Renunciation
34

E A Note on Notification
36

III A Manifestation of Will to Create a Legal Obligation
38
IV Unilateral
40

A Collective State Action
41

(i) Treaties Providing for Rights for Third States
45

(ii) Joint Declarations Outside Treaties
49

(iii) Concluding Remarks on Joint State Action
54

B The Autonomy Debate
55

(i) Independence of Another Unilateral Act?
59

(ii) Independence of Another Multilateral or
Bilateral Act?
62

(iii) Its Really All about Lex Specialis 67

(a) Acts Governed by the Law of Treaties
68

(b) Declarations Made Pursuant to ICJ Statute,
Article 36(2)
69

xii Contents

(c) Declarations Made During Judicial

Proceedings 75

(d) A Brief Note on Other Acts Usually Excluded 77

V Promises of States Only
78
2 A History of Promises 80

I
Important Cases and State Practice
81
A
Mavrommatis Jerusalem Concessions Case
83
B
Certain German Interests in Upper Polish Silesia Case
85
C
Free Zones Case
87

D Declarations Concerning the Protection of Minorities
88
E
Legal Status of Eastern Greenland Case
93

F Assurances Leading up to the Second World War
101

G Austrias Permanent Neutrality
104

H Egypts Declaration on the Suez Canal
108
I North Sea Continental Shelf Cases
114
J Nuclear Tests Cases
116

(i) Facts of the Case and the Judgment Delivered
by the Court
117

(ii) Analysis and Critical Assessment
123

(iii) The Separate and Dissenting Opinions
132

K Swiss Declaration in Relation to Security Council
Resolution 253
138

L Declarations on Torture and Other Inhumane Treatment 140

M British Pledge on the Five Techniques before the
ECtHR
141

N Swiss Petitpierre Assurance to the United Nations
145

O Declaration on the Use of War Material Exported
out of Austria
147
P 
German Approval Case before the German
Constitutional Court
148
Q
Military and Paramilitary Activities in and against
Nicaragua Case
151
R
Filleting of Fish in the Gulf of St Lawrence Arbitration
155
S
Frontier Dispute Between Burkina Faso and Mali Case
156

T Negative Security Assurances
161
U
LaGrand Case
166
V
Armed Activities on the Territory of the Congo Case
169

W 
Questions Relating to the Obligation to Prosecute or
Extradite Case
172

II Concluding Remarks on the History of Promises
174

Contentsxiii

3 The Law on Promises 176

I
Promises and the Sources of International Law
176

II The ILCs Guiding Principles
183

III Legal Basis for the Bindingness of Unilateral Promises
194

A Presumed Consent
196

B State Sovereignty and Intent
198

C Good Faith and Presumed Reliance
201

IV On the Intention to be Legally Bound
207

A The States Manifestation of Will
208

B Ascertaining a Legal Intent to be Bound
211

(i) Restrictive Interpretation
212

(ii) Interpreting a Declarations Text
214

(iii) Clear and Specific Wording
218

(iv) Supporting Circumstances
222

(v) Concluding Remarks on Interpretation
227

C Distinguishing Promises and Offers
228

V Further Requirements
234

A Competent Representative
234

B No Defective Will
237
C Publicity
239

D A Lawful (and Possible?) Object
243
E Registration?
246

F Impact of the Addressees Reaction
247

G Summarising the Requirements for a Legally Binding
Promise
250

VI The Revocability of Promises
251

A Necessary Limitations
251

B Between the Vienna Convention and a More Flexible
Approach 253

C Guiding Principle 10: Providing an Answer (Only) for
Specific Circumstances
258

D Specific Circumstances Aside: Determining a
Ground Rule on Revocation
263

(i) Towards a More Flexible Approach
263

(ii) Reasonable Notice Requirement
267

E Consolidating the Approach: A Ground Rule from
which to Deviate in Specific Circumstances
273

F Concluding Remarks on the Revocability of Promises 275

VII Modifying Promises
276

xiv Contents

VIII Promises and Estoppel
277

A Estoppel
277

B Promises and Estoppel: Distinct but Interacting Legal
Principles 283

(i) Two Distinct Legal Principles
283

(ii) Promises Triggering Estoppel
286

(iii) Non-Promises Triggering Estoppel?
290

(iv) Recapitulating Remarks on the Interplay of
Promise and Estoppel
294

IX Summarising the Legal Framework
294
4 Looking Ahead: A Promising Future? 299

I
Straitjacket vs Empowering Rule
300

II Unilateral Promises and Bilateral Commitments
307

A Why Allow for Legally Binding Promises at the
International Level?
307

B Circumventing or Complementing Treaties?
310
Concluding Remarks

312

Annex I
Annex II

314
321

Bibliography
Index

323
333

Abbreviations
DRC
ECHR
ECtHR
ICJ
ILC
ILO
IMT
NPT
OAS
OAU
PCIJ
UNCLOS
WTO

Democratic Republic of the Congo

European Convention on Human Rights
European Court of Human Rights
International Court of Justice
International Law Commission
International Labour Organization
International Military Tribunal
Non-Proliferation Treaty
Organization of American States
Organisation of African Unity
Permanent Court of International Justice
UN Convention on the Law of the Sea
World Trade Organization

Table of Cases
(sorted chronologically by date of decision):
ICJ, Case Concerning Questions Relating to the Obligation to Prosecute or
Extradite (Belgium v Senegal), Order of 28 May 2009................................. 173
, Case concerning the application of the convention on the prevention and
punishment of the crime of genocide (Croatia v Serbia), Preliminary
Objections, Judgment of 18 November 2008................................................... 82
, Application of the convention on the prevention and punishment of the
crime of genocide (Bosnia and Herzegovina v Serbia & Montenegro), Merits,
Judgment of 26 February 2007........................................................................ 12
, Armed Activities on the Territory of the Congo case (Democratic
Republic of the Congo v Rwanda), New application: 2002, Jurisdiction
and Admissibility, Judgment of 3 February 2006........................... 25, 170, 186
, LaGrand (Germany v United States of America), Judgment of 27 June
2001, ICJ Reports 2001, 466.................................................... 1668, 173, 225
, Fisheries Jurisdiction Case (Spain v Canada), Jurisdiction, Judgment of
4 December 1998, ICJ Reports 1998, 432................................ 58, 71, 215, 223
, Land and Maritime Boundary case between Cameroon and Nigeria,
(Cameroon v Nigeria: Equatorial Guinea intervening), Preliminary
Objections, Judgment of 11 June 1998, ICJ Reports 1998, 275..... 72, 281, 288
, Legality of the Threat or Use of Nuclear Weapons (Advisory Opinion
of 8 July 1996), ICJ Reports 1996, 226........................................................ 165
, Case concerning the Land, Island and Maritime Frontier Dispute (El
Salvadore v Honduras), Application by Nicaragua to Intervene, Judgment
of 13 September 1990, ICJ Reports 1990, 92........................................ 281, 284
, Border and Transborder Armed Action case (Nicaragua v Honduras),
Jurisdiction and Admissibility, Judgment of 20 December 1988, ICJ
Reports 1988, 69........................................................................................... 179
, Case concerning the Frontier Dispute (Burkina Faso v Mali), Judgment
of 22 December 1986, ICJ Reports 1986, 554................. 12, 15660, 189, 209,
21314, 2224, 2489, 288
Arbitration Tribunal Filleting within the Gulf of St Lawrence between
Canada and France, Award of 17 July 1986, Reports of International
Arbitral Awards 1986, 225.................................................................. 155, 225
ICJ, Military and Paramilitary Activities in and against Nicaragua
(Nicaragua v United States of America), Judgment of 27 June 1986, Merits,
ICJ Reports 1986, 14...............................................vii, 713, 1515, 160, 165,
189, 212, 248, 256, 261, 272, 288

xviii Table of Cases

, Military and Paramilitary Activities in and against Nicaragua
(Nicaragua v United States of America), Merits, Dissenting Opinion
Judge Schwebel, ICJ Reports 1986, 259....................................................... 155
German Federal Constitutional Court (BVerfG), Cruise Missiles
Deployment (German Approval) Case, Judgment of 18 December 1984,
ILR 1984, p. 365 (=BVerfGE 68, 1)......................................................... 1489
ICJ, Delimitation of the Maritime Boundary in the Gulf of Maine Area,
(Canada v USA), Judgment of 12 October 1984, ICJ Reports 1984,
p. 246....................................................................................................... 22, 280
, Military and Paramilitary Activities in and against Nicaragua case
(Nicaragua v United States), Jurisdiction and Admissibility, Judgment of
26 November 1984, ICJ Reports 1984, 392.................... 70, 261, 268, 271, 281
, Aegean Sea Continental Shelf case (Greece v Turkey), Judgment of
19 December 1978, ICJ Reports 1978, 3...................................................... 223
ECHR, Case of The Republic of Ireland v The United Kingdom, (Application
no 5310/71), Judgment of 18 January 1978, Series A, No 25, 175.......... 1412
, Case of The Republic of Ireland v. The United Kingdom, Series B:
Pleadings, Oral Arguments and Documents 1980............................. 1423
, Case of Ireland against the United Kingdom, Series B: Pleadings,
Oral Arguments and Documents 1981..................................................... 143
ICJ Nuclear Tests case, (Australia v France), Judgment of 20 December
1974, ICJ Reports 1974, 253....................................v, 12, 25, 69, 11718, 120,
122, 130, 1789, 186, 188, 203, 2089,
212, 218, 220, 2245, 239, 249, 2845
, Nuclear Tests case, (Australia v France), Separate Opinion of Judge
Forster ICJ Reports 1974, 275...................................................................... 132
, Nuclear Tests case, (Australia v France), Separate Opinion of Judge
Ignacio-Pinto, ICJ Reports 1974, 308.......................................................... 133
, Nuclear Tests case, (Australia v France), Joint Dissenting Opinion
of Judges Onyeama, Dillard, Jimnez de Archaga and Sir Humphrey
Waldock, ICJ Reports 1974, 312.............................................................. 1334
, Nuclear Tests case, (Australia v France), Dissenting Opinion of Judge
De Castro, ICJ Reports 1974, 372........................................................ 118, 134
, Nuclear Tests case, (Australia v France), Dissenting Opinion of Judge
Sir Garfield Barwick, ICJ Reports 1974, 391............................................... 137
, Nuclear Tests case, (New Zealand v France), Judgment of 20 December
1974, ICJ Reports 1974, 457...............................................................v, 12, 117
, Nuclear Tests case (Australia v France), Interim Protection, Order of
22 June 1973, ICJ Reports 1973.................................................................... 99.
, Nuclear Tests case (New Zealand v France), Interim Protection, Order
of 22 June 1973, ICJ Reports 1973.............................................................. 135.
Legal Consequences for States of the Continued Presence of South Africa in
Namibia (South West Africa) notwithstanding Security Council Resolution
276 (1970), Advisory Opinion of 21 June 1971, ICJ Reports 1971, 16....... 110

Table of Casesxix

ICJ, North Sea Continental Shelf cases, Judgment of 20 February 1969,

ICJ Reports 1969, 3...................................................................... 114, 224, 280
, North Sea Continental Shelf cases, Separate Opinion of Judge
Ammoun, ICJ Reports 1969, 101................................................................. 115
, Case concerning the Barcelona Traction, Light and Power Company,
Limited, Preliminary Objections, Judgment of 24 July 1964, ICJ Reports
1964, 6........................................................................................................... 280
, Case concerning the Temple of Preah Vihear (Cambodia v Thailand),
Merits, Judgment of 15 June 1962, ICJ Reports 1962, 6................... 212, 280
, South West Africa cases (Ethiopia and Liberia v South Africa),
Preliminary Objections, Judgment, 1962, 319.............................................. 110
, South West Africa Cases, (Ethiopia and Liberia v South Africa),
Preliminary Objections, Separate Opinion of Judge Jessup, ICJ Reports
1962, 387......................................................................................................... 58
, South West Africa cases (Ethiopia and Liberia v South Africa),
Preliminary Objections, Joint Dissenting Opinion of Judges Spender and
Fitzmaurice, ICJ Reports 1962, 465............................................................... 66
, Case concerning the Temple of Preah Vihear (Cambodia v Thailand),
Preliminary Objections, Judgment of 26 May 1961, ICJ Reports 1961, 17..25
, Case concerning the Temple of Preah Vihear (Cambodia v Thailand),
Preliminary Objections, Separate Opinion of Sir Gerald Fitzmaurice, ICJ
Reports 1961, 52....................................................................... 2789, 287, 293
, Case concerning right of passage over Indian territory, (Portugal v
India), Preliminary Objections, Judgment of 26 November 1957, ICJ
Reports 1957, 125........................................................................................... 72
, Anglo-Iranian Oil Co case, (United Kingdom v Iran), Preliminary
Objections, Judgment of 22 July 1952, ICJ Reports 1952, 93.......... 71, 21516
, Fisheries Case (United Kingdom v Norway), Judgment of 18 December
1951, ICJ Reports 1951, 116.................................................................... 22, 65
The International Military Tribunal, Trial of the Major War Criminals
Before the International Mitlitary Tribunal, 1947.................................... 1023
PCIJ, Minority Schools in Albania, Advisory Opinion of 6 April 1935,
Series A/B, No 64, 1935, 4................................................................. 89, 912
, Legal Status of Eastern Greenland case (Denmark v. Norway),
Judgment of 5 April 1933, Series A/B 53, 22........................ 93102, 154, 232
, Legal Status of Eastern Greenland case (Denmark v Norway),
Dissenting Opinion of Judge Anzilotti, Series A/B 53, 76..................... 94, 98
, Free Zones of Upper Savoy and the District of Gex case (France v
Switzerland), Judgment of 7 June 1932, Series A/B 46, 96.......................... 87
, Serbian Loans case, Judgment of 12 July 1929, Series A 20, 4............... 282
, Certain German Interests in Upper Polish Silesia (Germany v Poland),
Judgment of 25 May 1926, Series A 7, 2............................ 857, 101, 225, 231
, The Mavrommatis Jerusalem Concessions case (Greece v Great Britain),
Judgment of 26 March 1925, Series A 5, 6......................... 837, 100, 225, 231

xx Table of Cases
, Case of the S.S. Wimbledon, Judgment of 17 August 1923,
Series A 1, 15............................................................................................ v, 199
Arbiter Baron Lambermont, Arbitration between Germany and the United
Kingdom relating to Lamu Island, Decision of 17 August 1889, Reports of
International Arbitral Awards 1889, 237.................................................... 82
General Assembly: Resolution 61/34, A/RES/61/34, (2006). (Referred
to as: Resolution 61/34).................................................................................. 3
General Assembly: Resolution 51/160, A/RES/51/160, (1997).
(Referred to as: Resolution 51/160).............................................................. 2
League of Nations: Mandates Proposal (3081) and Annex (1373), (1932).
(Referred to as: Mandates Proposal & Annex).......................................... 90
League of Nations: Minorities in Estonia, League of Nations Official
Journal, vol 4, (1923), 131012.................................................................. 901
League of Nations: Minorities in Lithuania, League of Nations Official
Journal, vol 4, (1923), 93233. ...................................................................... 90

Introduction

HIS STUDY FOCUSES on assurances made by states creating a

legally binding obligation for them to act in future as declared; an
obligation which is brought about not through an offer met with
acceptance, ie the conclusion of a bi- or multilateral treaty, but merely
through a states one-sided, unilateral pledge. The following pages will
hence deal with what is commonly referred to as promises in international law, although the formulation is less than fortunate as it seems to
imply a mere moral obligation.1 Promises form a subcategory within the
much broader and quite diverse topic entitled Unilateral Acts of States on
which a (usually rather short) section can be found in nearly every standard public international law textbook.2 The latter regularly group unilateral acts into categories which may vary slightly from author to author
but typically comprise protest, recognition, waiver and promise.3 As their
textbook tradition already indicates, unilateral acts are neither a new nor
are they a rare phenomenon in international law. Quite on the contrary,
there seems to be a consensus that states frequently make use of unilateral
acts, not necessarily limited to the above-mentioned categories.4 It wasnt

1
It has therefore rightly been identified as a misnomer, cf Lukashuk ILC, Summary
Record of the 2629th Meeting, UN Doc A/CN.4/SR.2629 (2000), para 7. But as it is commonly used and accepted within legal doctrine to describe the legal phenomenon of interest
here, it will be retained.
2
See, eg Ian Brownlie, Principles of Public International Law, 7th edn (Oxford/New York,
2008) 61215; Antonio Cassese, International Law, 2nd edn (Oxford, 2004) 18485; Jean
Combacau and Serge Sur, Droit international public, 5th edn (2001) 9096; Georg Dahm, Jost
Delbrck and Rdiger Wolfrum, Vlkerrecht, 2nd edn (Berlin, 2002) vol I(3), 76473; Wolf
Heintschel von Heinegg, Einseitige Rechtsakte in Knut Ipsen (ed), Vlkerrecht, 5th edn
(Mnchen, 2004), s 18, 23440; Quoc Dinh Nguyen, Alain Pellet and Patrick Daillier, Droit
international public, 7th edn, (Paris 2002) 35966; Lasa Oppenheim, Robert Jennings and
Arthur Watts, International Law 9th edn (Harlow, 1992), vol 1, pts 24, 1187; Paul Reuter, Droit
international public, 6th edn (Paris, 1983) 16374; Georg Schwarzenberger, International Law as
Applied by International Courts and Tribunals, 3rd edn (London, 1957) vol I, 54861; Malcolm
N Shaw, International Law, 6th edn (Cambridge/New York, 2008) 12122; Alfred Verdross
and Bruno Simma, Universelles Vlkerrecht: Theorie und Praxis, 3rd edn (Berlin, 1984) 42431.
3
For the classification of unilateral acts see p 23 et seq.
4
See, eg ILC, Report of the Working Group, UN Doc A/CN.4/L.543 (1997) para 5: In
their conduct in the international sphere, States frequently carry out unilateral acts with
the intent to produce legal effects; Krzysztof Skubiszewski, Unilateral Acts of States in
M Bedjaoui (ed), International Law: Achievements and Prospects (1991) 221: In international
life unilateral acts are as numerous as they are frequent; Christian Tomuschat, Unilateral
Acts under International Law in Droits et Culture. Mlanges en lhonneur du Doyen Yadh Ben
Achour (Tunis, 2008) 1487: Unilateral acts are ubiquitous in inter-State relationships; and
Karl Zemanek, Unilateral Legal Acts Revisited in Karel Wellens (ed), International Law:
Theory and Practice, Essays in Honour of Eric Suy (The Hague, 1998) 210, according to whom
unilateral acts have become the most frequent tool of State interaction.

2 Introduction
a big leap from this assessment to perceiving the absence of a common
legal framework in this area as a lacuna in public international law. And
indeed, the International Law Commission (ILC)5 drew this conclusion
in 1996 and decided to propose the topic to the United Nations General
Assembly as one suitable and appropriate for codification and progressive development. The General Assembly agreed,6 endorsed the proposal
and the ILC embarked upon its new project with high hopes, stressing
the importance of codifying unilateral acts of states. As the Commission
pointed out, not only were states frequently acting unilaterally but:
the significance of such unilateral acts is constantly growing as a result of the
rapid political, economic and technological changes taking place in the inter
national community at the present time and, in particular, the great advances in
the means for expressing and transmitting the attitudes and conduct of States.7

It was consequently:
In the interest of legal security and to help bring certainty, predictability and
stability to international relations and thus strengthen the rule of law, [that] an
attempt should be made to clarify the functioning of this kind of acts and what
the legal consequences are, with a statement of the applicable law.8

This assessment notwithstanding, the ILC made only very little progress over the years to come. It turned out to be extremely difficult to
even agree on a common ground to start on, as Commission members
remained sharply divided on the preliminary question of whether a legal
institution of unilateral acts of states, to which a common set of rules
could be applied, actually existed. Yet, hope persisted amongst the majority to develop draft articles9 divided into a general part, including basic
rules such as, inter alia, on the competence to formulate binding unilat5
The ILC is a subcommittee of the UN General Assembly and entrusted with the promotion of the progressive development of international law and its codification. In this capacity
the Commission has produced drafts which formed the basis of various important treaties,
amongst them the UN Convention on the Law of the Sea (1958), the Vienna Convention on
Diplomatic Relations (1961), the Vienna Convention on Consular Relations (1962), as well
as the Vienna Convention on the Law of Treaties (1969). The Commission is comprised of
34 members who shall be persons of recognized competence in international law (art 2(1)
of its Statute), and are elected for a five-year period, a quinquennium. For more see www.
un.org/law/ilc/index.htm.
6
See GA Res 51/160, 3, para 13, which served as the ILCs mandate by inviting the
Commission to further examine the topic.
7
ILC 1997 Report of the Working Group, n 4 above, 2, para 5.
8
Ibid.
9
Whether the Commission should actually develop draft articles as its final outcome
was not uniformly answered and left open, even though draft articles with commentaries
thereto were held to be the most adequate way to proceed with the study, see ILC, Report
of the Working Group, UN Doc A/CN.4/L.558 (1998) 2, para 7. Some members favoured a
mere expository study (eg Simma, ILC, Summary Record of the 2525th Meeting, UN Doc A/
CN.4/SR.2525 (1998) 42, para 5) while others, including the Special Rapporteur who proposed various draft articles throughout his nine reports, hoped to develop genuine draft
articles for the subject.

Introduction3

eral acts or the lack of formal requirements, and a special part, comprised
of four different sections, addressing acts falling into the categories of
protest, waiver, promise and recognition.10 The reports of the appointed
Special Rapporteur, Vctor Rodrguez Cedeo, included various proposals for draft articles for the general part, but despite a report focusing
solely on recognition,11 one that comprised state practice structured into
the above-mentioned categories12 and one highlighting some especially
relevant cases from within each category,13 the Commission never considered any draft articles for part II of the Code, the one dealing with specific
unilateral acts. Instead, after 10 years, nine reports, many working groups
and continuous debates on the topics codifiability, the ILC finally decided
to end its project in 2006. It cut the Gordian knot by slicing off a major
part of the original topic and presented the General Assembly with 10
Guiding Principles applicable to Unilateral Declarations of States Capable
of Creating Legal Obligations.14 Evidently, the ILC wished to rid its agenda
of a subject matter with which it felt it had wrestled long enough. The
General Assembly took note of the final outcome and expressed its appreciation later the same year.15
Informed by the debates that were held within the Commission and in
light of the severe obstacles it encountered when struggling to achieve
a productive and helpful outcome covering the whole ambit of unilateral state conduct, the approach taken here is far narrower. This study
will, just like the Commissions final product does, concentrate only on
declarations that are capable of creating legal obligations, while focusing on a states concrete obligation to act or refrain from acting in a way
it has pledged to do within the future. This will leave us with the category of promise, one which deals with declarations obliging states in
a way that might be comparable to that of treaties but for which despite
the Commissions efforts, only a rudimentary and in parts obscure legal
10
For this approach see ILC 1997 Report of the Working Group, n 4 above, 5, para 18, as
well as the outline prepared by the Working Group, ibid 57, para 19. The Special Rapporteur
proposed a classification of unilateral acts in his Fourth Report, see Victor Rodriguez Cedeo,
Fourth Report on Unilateral Acts of States, UN Doc A/CN.4/519 (2001) 1022, paras 44100.
11
Victor Rodriguez Cedeo, Sixth Report on Unilateral Acts of States, UN Doc A/
CN.4/534 (2003).
12
Victor Rodriguez Cedeo, Seventh Report on Unilateral Acts of States, UN Doc A/
CN.4/542 (2004).
13
Victor Rodriguez Cedeo, Eighth Report on Unilateral Acts of States, UN Doc A/
CN.4/557 (2005).
14
ILC, Guiding Principles applicable to Unilateral Declarations of States Capable of
Creating Legal Obligations, with commentaries thereto, ILC Report, UN Doc A/61/10
(2006) ch IX, emphasis added. The final outcome deals only with declarations, not with acts,
and only with those which can be said to create legal obligations. Articles on specific types
of unilateral acts are not included. For the question which acts are covered by the Guiding
Principles, see also n 16 below.
15
GA Res 61/34, 2, paras 2(d) and 3.

4 Introduction
framework exists.16 Some Commission members had therefore rightly
identified it as one, if not the most important category of unilateral acts
the ILC had to deal with.17 In focusing on promises this book also follows
Paul De Visschers advice, who in 1984 wrote:
it is to be hoped that legal doctrine will again take up the study of promises in
light of the requirements imposed by good faith, which are neither those of the
rule of pacta sunt servanda nor those of the principle of estoppel.18
I THE PRESUMED RARENESS OF PROMISES

A promise is commonly understood to be a unilateral manifestation of a

states will through which it commits itself in a legally obliging manner vis-vis one or several addressees to act or refrain from acting in a particular
16
Whether the Guiding Principles dealing with declarations that are capable of creating legal obligations are applicable to unilateral acts other than promises is far from clear.
Tomuschat, Unilateral Acts, n 4 above, 1497 considers merely protest to be excluded as
protests clearly do not create any legal obligation. Any act of protest is indeed squarely outside the scope of the Guiding Principles (see also pp 3334), despite the commentarys note
954 which refers to a case that involves a protest in support of one of the Guiding Principles.
Recognition is arguably a declaration capable of creating legal obligations, although it is
more adequately in our view described in terms of opposability of an admitted fact. Also,
it is usually said to be irrevocable unless the very object of recognition dissolves. Guiding
Principle 10, however, merely prohibits arbitrary revocation for declarations within the
Principles reach and it is rather unlikely that the ILC intended to allow states more leeway
in this area. Tomuschat, Unilateral Acts, n 4 above, 1499, hence notes: It may well be that
the ILC considered declarations of recognition as not coming within the scope of Principle
1, which, in the view of the present writer would be an erroneous interference. There must
be even stronger doubts whether a waiver may adequately be described as creating legal
obligations, see pp 3436. A waiver, also, and as a consequence of it having extinguished the
right waived, is considered to be non-revocable and not merely non-arbitrarily revocable as
the Guiding Principles would indicate if held to be applicable. See also Cassese, International
Law, n 2 above, 185, for whom: Promise is the only unilateral transaction giving rise to
international obligations proper, that is, establishing a new rule binding the promising State
towards one or more States; similarly Hafner, ILC, Summary Record of the 2595th Meeting,
UN Doc A/CN.4/SR.2595 (1999) para 32, where he pointed out that a definition which
referred to the intention of acquiring legal obligations implied that only promises will be
taken into account. For more on the different categories of unilateral acts, see p 23 et seq.
17
cf, eg Simma, ILC, 2525th meeting, n 9 above, para 8: The most interesting problem,
however, concerned the binding force, for the States from which they emanated, of unilateral
promises a term which . . . was to be preferred to unilateral declaration, which was too
formal and so capable of encompassing very different acts. See also the largely recapitulative Fifth Report in which the Special Rapporteur mentioned that the ILC had noted that the
work of codification and progressive development may focus, at least initially, on promises,
Victor Rodriguez Cedeo, Fifth Report on Unilateral Acts of States, UN Doc A/CN.4/525
(2002) 5, para 14.
18
My translation of Paul de Visscher, Remarques sur lvolution de la jurisprudence de
la Cour Internationale de Justice relative au fondement obligatoire des certains actes unilatraux in Essays in Honour of Manfred Lachs (The Hague, Boston, 1984) 469: il faut souhaiter
que la doctrine reprenne ltude de la promesse unilatrale la lumire des exigences de
la bonne foi, qui ne sont ni celles de la rgle pacta sunt servanda ni celles du principe de
lestoppel.

The Presumed Rareness of Promises5

way in the future. It requires no acceptance or any reaction from the side
of the addressee(s) to become effective. While promises have been studied
before, no English monograph19 focuses exclusively on this topic and even
most articles do not put their whole thrust behind analysing promises of
states but address them only when dealing with unilateral acts in general.20 In light of the potentially far-reaching consequences of such a unilaterally assumed legal obligation, this is rather surprising; all the more
so as the legal framework operable in this area is far from clear. One of
the reasons for promises having received comparatively scarce attention
might be related to the fact that while unilateral acts of states are widely
perceived as a common phenomenon, those falling into the category of
promise are assumed to be rare.
Surely, the unclear legal framework remained a hindrance for promises
to become a unilateral act of considerable impact and played its part in
forcing many commentators to assume a contractual relationship whenever a state was held to have displayed an intent to commit itself, even if
the construction of the latter had to employ tacit acceptances and establish a meeting of minds which scarcely reflected the actual communication that took place between the parties.
While promises may very well be less frequent than commitments undergone through treaties, it is striking that the diagnosis of state promises actually being a rare phenomenon in international law is regularly bolstered not
so much by an empirical study of state behaviour, but by concluding that
unilateral and legally binding undertakings by a state lacking a quid pro
quo must be infrequent, as no state will be willing to give without receiving anything in return.21 Jumping to this a priori conclusion is, however,
19
Neither did we find a monograph on promises in international law in the German or
French language. In Italian there is Sergio M Carbone, Promessa e affidamento nel diritto internazionale (Milano, 1967).
20
Exceptions are especially Sergio M Carbone, Promise in International Law: A
Confirmation of its Binding Force (1975) 1 Italian J International Law 166; Wilfried Fiedler, Zur
Verbindlichkeit einseitiger Versprechen im Vlkerrecht (1976) 19 German YB International Law
35; Jean-Paul Jacqu, A propos de la promesse unilatrale in Melanges offerts a Paul Reuter
(1981) 32745; and Jean-Didier Sicault, Du caractre obligatoire des engagements unilatraux
en droit international public (1979) 83 Revue Gnrale de Droit International Public 633.
21
This assumption apparently dates back to the time before the Nuclear Tests cases where
the doctrinal foundations of promises stood on unclearer grounds and seems to stem from
Eric Suy, Les actes juridiques unilatraux en droit international public (Paris 1962) 111: Notre
thse est la suivante: les promesses purement unilatrales existent en droit international bien
quelles soient trs rares. Cette raret sexplique facilement tant donn quaucun Etat ne se
prte de bon gr faire des concessions spontanes et gratuites. See also Jean Charpentier,
Engagements unilatraux et engagements conventionnels: diffrences et convergences
in Jerzy Makarczyk (ed), Theory of International Law at the Threshold of the 21st Century (The
Hague/London/Boston, 1996) 373, who considers the absence of reciprocity to explain the
rareness of unilateral engagements. In the same vein, Victor Rodriguez Cedeo, First Report
on Unilateral Acts of States, UN Doc A/CN.4/486 (1998) para 141 and Cedeo, Fifth Report,
n 17 above, para 35. Critically already Sicault, Engagements unilatraux, n 20 above, 638,
for whom on sent encore ici lattraction, pour ne pas dire la fascination, exerce par laccord
de volonts.

6 Introduction
premature as there are many reasons conceivable why a state might want
to make a unilateral promise, without acting upon while not necessarily
discarding utilitarian motives.22
One of promises specific and sometimes possibly advantageous features as a unilateral act is that it does not allow for any bargaining; no
acceptance or any reaction from the addressee(s) is necessary for the legal
obligation to arise. In order to create a new legal obligation, a promise is
thereby the quicker and easier legal mechanism in comparison to treaties,
where only through the participation of another or several other parties is
it possible to seal the deal. Of course, the newly assumed obligation created by a promise will only lie with the declaring state,23 but contrary to
what might be assumed, this may not necessarily be to its detriment.
Through a promise a state might, for example, unilaterally assume a
concrete obligation in order to create a fait accompli, thereby hoping its
addressee(s) will refrain from initiating the often long and painstaking
process which typically precedes the conclusion of a treaty, and instead
accept the benefit so readily promised even if the commitment made
might not go as far as its addressee(s) initially demanded.24 This strategy might be especially useful where a state is not so much facing a
typical commercial exchange situation (as through making a promise it
will receive nothing material in return) but where it is confronted with
the claim to alter its behaviour, grant certain guarantees or even establish a clear legal framework for a certain area. In these cases, the self-
proclaimed commitment, the fait accompli, can not only help to reduce
public as well as political pressure to comply with the claims made, but

22
See also Fiedler, Einseitige Versprechen, n 20 above, 54, as well as Tomuschat, Unilateral
Acts, n 4 above, 1495, the latter emphasising that not receiving anything at the juridical level
does not mean a state will not get anything in return: In political terms, there will always be
some advantage.
23
As Guiding Principle 9 stipulates: No obligation may result for other States from the
unilateral declaration of a State. However, the other State or States concerned may incur
obligations in relation to such a unilateral declaration to the extent that they clearly accepted
such a declaration. The acceptance of another states power to regulate certain matters via
a unilateral declaration with negative legal repercussions for the accepting state, ie limiting
its rights, may, however, be expressed prior to the unilateral declaration made. In these cases
a unilateral declaration has been said to merely activate a states prior obligation. Where,
for example, a coastal state issues a unilateral declaration in which it declares to extend its
territorial waters up to the generally accepted limit of 12 nautical miles measured from its
baseline (see United Nations Convention on the Law of the Sea (UNCLOS), Art 3) it will
extend its sovereignty, and thus require full respect of its national rules and regulations,
in an area previously not under its exclusive control. It may, however, do so as states have
agreed to accept these declarations as decisive in determining the scope of a states territorial
seas (within the limits and exceptions agreed upon in UNCLOS, pt II, s II). For the question
whether these sort of declarations really are unilateral acts, which they are, see p 55 et seq.
24
Egypts unilateral declaration on the use of the Suez Channel after the Suez crisis was
deposited with the UN Secretary-General and is an example in this regard as it successfully
forestalled the holding of an international conference on the matter. For more see pp 10814.

The Presumed Rareness of Promises7

in fact quickly alter the legal landscape and thereby create the legal security demanded.25
Also, a state by assuming an obligation unilaterally can, in fact, even be
reaching out for a quid pro quo, be it only in the form of the goodwill of
another state or a future commitment of the latter, while, however, considering it unwise to establish a formal do ut des relationship via a treaty
offer and demand something in immediate return. Instead, it might be
more sensible to use a unilateral obligation and wait a while before asking, without legally demanding, the addressee to return the favour.26 In
a similar vein, states might also pledge their belief in a certain concept
and hence assume a unilateral commitment (like an immediately effective
ban on the use of certain weaponry), in order to entice others to join in.
These mechanisms become all the more important as states in the
globalised world are bound to meet again and their interdependence is
anything but declining; with it mutual trust as well as good relations are
of ever-increasing importance. Especially in a time such as ours, where
public opinion can arouse not only considerable pressure but influence
consumers, companies and states cash flow, states might feel a strong
incentive simply to look good in the international arena;27 hence, the
use of a non-reciprocal legal mechanism such as an (officially and generously proclaimed) promise might come in handy, be it in order to be perceived as selfless and giving or simply to assume a legal obligation, like
for example the cessation of a criticised practice such as drift-net fishing,28
without any more ado.
In another scenario, a state through its representatives can be willing
to commit itself in a legal way for the very reason of limiting its future
options.29 Just as national constitutions do, international obligations can
25
This does, of course, presuppose the possibility of defining a clear normative framework for promises. For the latter see chapter three. A promises beneficiary can, however,
choose to reject the promise made, see also pp 24750.
26
As Sicault remarks in discussing doctrines sometimes rather desperate efforts to construe a treaty relationshipout of separate unilateral declarations: On ignore ainsi dlibrment le fait quune partie peut logiquement estimer que la route vers des avantages
rciproques passe par des obligations univoques auxquelles elle accepte de souscrire en considration de lobjectif final, Sicault, Engagements unilatraux, n 20 above, 644.
27
Even though public opinion, along with blame and shame tactics, are sometimes
belittled, the evolving field known as corporate social responsibility indicates consumers
impact as well as companies awareness of their vulnerability and dependence on a good
image. States have similar interests and as, for example, the Olympic Games in China in 2008
showed, are indeed ready to invest millions into their image, millions which only seemingly
are spent without receiving anything in return.
28
Japan, on 17 July 1990, announced that it was suspending drift-net fishing in the South
Pacific during 1990 and 1991, a year before the adoption of the UN Resolution on the matter,
see (1991) 95 Revue Gnrale de Droit International Public 155.
29
The precommitment theory comes to mind. For an introduction, the application of the latter to international law and some of the incentives also mentioned here, even if not precommitments in the sense of the theory, see Steven R Ratner, Precommitment Theory and International
Law: Starting a Conversation (200203) 81 Texas Law Review 2055, 205859, 207476.

8 Introduction
be used in order to establish legal boundaries. Fearing that a change of
circumstances (be it war, civil strife, revolution, terrorist attacks, inflation or natural catastrophes) will increase especially inner state pressure,
with desperate times calling for desperate measures, a state might find it
wise to get the international community involved by assuming an international obligation to refrain from certain actions hoping that this legal
restraint will help to render the outlawed actions impossible, no matter
what the future brings. Whereas in the former case the idea is one of selfrestraint (or that of future governments),30 the very same strategy might
be employed to influence and alter a third partys perception of what a
state might do in the future (eg convince investors by assuming a commitment outlawing any future nationalisation of property or the like).31
A state might also direct a promise to the international community32 in
face of political reasons which hinder it from directly addressing or even
sitting down at the table with one of the beneficiaries, or indeed the sole
factual beneficiary of the obligation undertaken. In an article discussing
the Nuclear Tests cases, the landmark decision for unilateral and legally
binding assurances, Thomas Franck referred to Egypts promise which
spelled out a legal regime for the use of the Suez Canal and observed:
This is a most useful step forward in international jurisprudence. It is particularly helpful, at a time when Egypt is indicating a willingness to undertake
binding commitments in respect of Israel but not to enter into an agreement
with Israel, that the theory of law should offer no impediments to such unilateral but legally binding accommodations.33

Last but not least, states might also favour a unilateral declaration over
a treaty obligation because the former is perceived as less binding by
them, in the sense of it being more easily revocable and amendable than a
treaty, especially a multilateral one.34
30
Since the change of circumstances mentioned above can simply be the election of a new
government which the present government would rather see with its hands tied by an international obligation.
31
The same reasons can, of course, lead states to conclude a treaty with similar content.
It is, however, much easier and quicker to assume a unilateral obligation vis--vis the international community by means of a promise as the participation of the addressees does not
have to be organised.
32
The international community is frequently referred to as a possible addressee of unilateral declarations and Guiding Principle 6 reads: Unilateral declarations may be addressed
to the international community as a whole, to one or several States or to other entities.
Nevertheless the term is ambivalent. Does it refer to the United Nations, encompass each
and every single state, or both? What about the European Union and other international and
regional organisations? The most sensible interpretation is to assume that in these cases the
declaration is made to all those international actors affected by it, in other words: to whom
it may concern. See also Andreas L Paulus, Die internationale Gemeinschaft im Vlkerrecht
The International Commmunity in Public International Law: English Summary (Mnchen, 2001)
especially 329 et seq and 44446.
33
Thomas M Franck, Word Made Law: the Decision of the International Court of Justice
in the Nuclear Test Cases (1975) 69 American J International Law 612, 61516.
34
On the revocability of promises, one of the doctrinal core questions, see pp 25776.

The Presumed Rareness of Promises9

This inquiry into a states mind is of course fragmentary and mainly

illustrates that promises cannot be assumed to be and necessarily remain
rare, simply from them lacking a direct and tangible quid pro quo. Surely
it will often rather be a mixture of the above-mentioned motives, along
with additional incentives, which may inspire a state to use a unilateral
assurance and not a treaty. The recent declaration of independence by
Kosovo is an interesting case in this regard.35 The adherence to various
obligations, such as to comply with the Ahtisaari Plan, to abide by the
principles of the United Nations as well as with the Helsinki Act and
other acts of the Organization on Security and Cooperation in Europe,
was unilaterally pledged by Kosovo and not bi- or multilaterally ensured.
Kosovo went as far as to expressly affirm, clearly, specifically, and irrevocably, that Kosovo shall be legally bound to comply with the provisions
contained in this Declaration and declare[d] publicly that all states
are entitled to rely upon this declaration.36 There is indeed little
reason or room to doubt that Kosovo was willing to bind itself legally, the
35
For those recognising Kosovo as a state it is a directly applicable example; for those not
recognising Kosovo it is still of importance as it is nevertheless illustrative of a situation in
which a declarant might take recourse to a unilateral declaration. Even if Kosovo is not seen
as a state, a state might find itself in a similar situation, ie being rejected by numerous parties
of a certain treaty it would like to join.
36
See Kosova Declaration of Independence, available at www.assembly-kosova.
org/?cid=2,128,1635, para 12. The declaration includes, inter alia, the following passages:
We, the democratically-elected leaders of our people, hereby declare Kosovo to be an independent and sovereign state . . . (3) We accept fully the obligations for Kosovo contained in the
Ahtisaari Plan, and welcome the framework it proposes to guide Kosovo in the years ahead.
We shall implement in full those obligations including through priority adoption of the legislation included in its Annex XII, particularly those that protect and promote the rights of
communities and their members. (4) We shall adopt as soon as possible a Constitution that
enshrines our commitment to respect the human rights and fundamental freedoms of all
our citizens, particularly as defined by the European Convention on Human Rights. The
Constitution shall incorporate all relevant principles of the Ahtisaari Plan and be adopted
through a democratic and deliberative process . . . (8) With independence comes the duty of
responsible membership in the international community. We accept fully this duty and shall
abide by the principles of the United Nations Charter, the Helsinki Final Act, other acts of the
Organization on Security and Cooperation in Europe, and the international legal obligations
and principles of international comity that mark the relations among states. Kosovo shall have
its international borders as set forth in Annex VIII of the Ahtisaari Plan, and shall fully respect
the sovereignty and territorial integrity of all our neighbours. Kosovo shall also refrain from
the threat or use of force in any manner inconsistent with the purposes of the United Nations.
(9) We hereby undertake the international obligations of Kosovo, including those concluded
on our behalf by the United Nations Interim Administration Mission in Kosovo (UNMIK)
and treaty and other obligations of the former Socialist Federal Republic of Yugoslavia to
which we are bound as a former constituent part, including the Vienna Conventions on
diplomatic and consular relations. We shall cooperate fully with the International Criminal
Tribunal for the Former Yugoslavia . . . (12) We hereby affirm, clearly, specifically, and irrevocably, that Kosovo shall be legally bound to comply with the provisions contained in this
Declaration, including, especially, the obligations for it under the Ahtisaari Plan. In all of
these matters, we shall act consistent with principles of international law and resolutions
of the Security Council of the United Nations, including Resolution 1244 (1999). We declare
publicly that all states are entitled to rely upon this declaration, and appeal to them to extend
to us their support and friendship.

10 Introduction
problem, however, being that the obligations could not have been assumed
by accession to the relevant treaties as some of the respective international
organisations member states (including those with veto powers within
the Security Council) were (and at the time of writing remain) unwilling to recognise it and would have vetoed any admission. Yet, Kosovo
obviously held them to be of vital importance in order to increase its
chances of surviving in the international arena and included assurances
to abide by these principles in its unilaterally proclaimed declaration of
independence.
A perusal of newspaper headlines in fact reveals that states through
their representatives regularly take recourse to unilateral declarations,
declarations that could fall into the ambit of the legal category of promise
as currently defined.37 States declare their willingness to cancel debts,38
to provide financial, technical or humanitarian aid and assistance,39 espe37
For more examples of pledges in highly sensitive areas, see also p 15; more state practice
is addressed in chapter two.
38
See, eg the declaration by President Chirac as cited in the Special Rapporteurs Seventh
Report, in which he announced that France would write off a total of 739 million francs in
bilateral debt that had been incurred by Guatemala, Honduras, Nicaragua, and El Salvador
for development aid . . . and also promised to negotiate a reduction in their commercial debt
at the next meeting of the Paris Club, see Cedeo, Seventh Report, n 12 above, 12, note 36,
and (1999) 103 Revue Gnrale de Droit International Public 195. If a debt is immediately cancelled via a states declaration, as opposed to it merely pledging its will to cancel it in the
near future, the declaration will constitute a waiver rather than a promise. For the distinction
see also pp 3436. An example of a waiver is the Spanish Head of Governments statement
of 4 April 2000: I should also like to inform you that I have announced that US$200 million
of official development assistance to the main Sub-Saharan African countries is being written off. That is to say, Spain is announcing the cancellation of US$200 million worth of subSaharan African countries indebtedness to our country, cited in Cedeo, Seventh Report, n
12 above, 12, note 36.
39
See, eg the following declaration which, amongst others, is provided in Cedeo, Seventh
Report, n 12 above, 11, note 29: the Embassy of Ireland in Washington released a message
dated 23 March 2003 from its Secretary of State that was worded as follows: I have today
announced that the Government is putting aside 5 million in humanitarian assistance for
the alleviation of suffering of innocent Iraqi civilians. This funding will be distributed to our
partner NGOs and International Agencies who have the capability to respond effectively to
the current crisis. See also Cedeo, Seventh Report, n 12 above, 1314, note 37, inter alia, citing the government of Australia as having announced on 28 October 2003 to make AUS$110
million available for the Iraqi people. See also the examples provided by Fiedler, Einseitige
Versprechen, n 20 above, 40, note 20, who, inter alia, mentions a German assurance to grant
a credit to Chile (citing BT.-Drucks. 7/6306f, 11529ff) which, after the regime in Chile had
changed, was never fulfilled, with the German government arguing that it had never given
a legally binding assurance but had merely announced a possible future course of action
(der Kredit wurde lediglich in Aussicht gestellt und nicht verbindlich zugesagt). When
the then German Foreign Minister, Frank-Walter Steinmeier, visited West-African countries
in 2008 he announced an Aktion Afrika [Action for Africa] which was to lead to an increase
in money paid for cultural exchange. Travelling through Ghana, Togo and Burkina Faso,
Steinmeier was told that his slogan was considered a given word, a promise, which he had
to keep (my translation of the German original: Ihm wurde klargemacht, dass sein Slogan
als gegebenes Wort gilt, als Versprechen, das er zu halten hat); Wulf Schmiese, Begeisterung
in Afrika, Zurckhaltung zu Hause, FAZ, 13 February 2008, 6. During the G8 meeting in
Heiligendamm, Germany, in 2007, the worlds richest nations pledged to give US$60 bil-

The Presumed Rareness of Promises11

cially but not limited to times of crisis and catastrophes; they pledge to
reduce greenhouse gases,40 to refrain from drift-net fishing41 and nuclear
testing.42 Nearly any commitment to follow a certain line of conduct can
be drafted along the lines of a unilateral assurance. Modern media does its
job and transports the messages to addressees around the globe. Whether
it be a written declaration or merely a statement made during a press
conference, secured to videotape or hard drive even the spoken word is
quickly stripped of its ephemeral character.
To be sure, this is not to say that any of these declarations are legal
undertakings, nor to claim that none of them are. Looking at the current legal framework applicable to promises, the main question will be
phrased in terms of whether any of them display an intent to be legally
bound. A states intent will, however, be inferred from its actions, which
again are necessarily judged against the background of legal rules. If we
compare a person waving a hand to a friend at a bus stop with one waving
his/her hand during an auction, the action is the same but on account of a
different legal environment, a legal will is going to be inferred in the latter
case. Transferred to promises, this leads to the question as to what kind of
framework is applicable for unilateral declarations made by states: Which
circumstances will be taken as indicating an actors will to be bound? Is a
solemn proclamation necessary, indicative or even sufficient? Do we need
any special wording, even repeated pledges or does a written confirmation of a precisely crafted pledge establish an intent to be legally bound?
These are only some of the questions that come to mind. In addressing
the law of treaties, the concept of intent has been referred to as a rather
awkward concept in itself;43 nevertheless, at least with written treaties,
lion to fight diseases such as AIDS in Africa. German Chancellor Merkel stressed: We are
conscious of our obligations and want to fulfil the promises we made. And we will do that;
while development campaigners criticised the pledge as vague and deliberately misleading,
see Activists slam G8 pledge on Africa, CNN.com, 8 June 2007.
40
See, eg the declaration made by the US President George W Bush: Our immediate goal
is to reduce Americas greenhouse gas emissions relative to the size of our economy. My
administration is committed to cutting our Nations greenhouse gas intensity, how much we
emit per unit of economic activity, by 18 percent over the next 10 years, cited in (2002) 96
American J International Law 487.
41
cf Japans declaration as mentioned at n 28 above.
42
See the statement of the Chinese government made on 29 July 1996 and annexed to a
letter dated the same day from the Permanent Representative of China to the UN SecretaryGeneral, asking for it to be circulated as an official document of the General Assembly. Its
first paragraph reads as follows: On 29 July 1996, China successfully conducted a nuclear
test. The Government of the Peoples Republic of China hereby solemnly declares that it will
start a moratorium on nuclear testing effective from 30 July 1996. Such an important decision
by China is not only a response to the appeal of the vast number of non-nuclear-weapon
States, but also a concrete action to promote nuclear disarmament, GA Doc A/51/262. See
also the declarations made by France with regard to atmospheric tests as cited in the Nuclear
Tests cases and addressed at p 116 et seq.
43
Jan Klabbers, The Concept of Treaty in International Law (The Hague/Boston/London,
1996) 6595.

12 Introduction
its application, while complicated, remains easier since here a formal ratification procedure exists on usually both the international and national
level. For promises this is generally not the case.
II A LEGAL FRAMEWORK WITH MANY OPEN QUESTIONS:
A FIRST LOOK

The crucial point of departure for the international lawyer will, of course,
be: Can such assurances ever be taken legally seriously, ie considered to
be binding? This question as to whether or not official proclamations can
constitute legal commitments and thereby legally speaking solid ground
for other states to rely on, has been answered in the affirmative by the
International Court of Justice (ICJ). If made publicly and with the intent to
be bound a unilateral declaration becomes binding for its author, said the
ICJ in its well known Nuclear Tests cases:
It is well recognized that declarations made by way of unilateral acts, concerning legal or factual situations, may have the effect of creating legal obligations.
Declarations of this kind may be, and often are, very specific. When it is the intention of the State making the declaration that it should become bound according to
its terms, that intention confers on the declaration the character of a legal undertaking, the State being thenceforth legally required to follow a course of conduct
consistent with the declaration. An undertaking of this kind, if given publicly,
and with an intent to be bound, even though not made within the context of international negotiations, is binding. In these circumstances, nothing in the nature of
a quid pro quo nor any subsequent acceptance of the declaration, nor even any
reply or reaction from other States, is required for the declaration to take effect,
since such a requirement would be inconsistent with the strictly unilateral nature
of the juridical act by which the pronouncement by the State was made.44

Even though it has been labelled a politically motivated decision

not to decide,45 the rulings legal doctrine has survived until today and
was confirmed by the Court in other judgments.46 While some scholars
44
ICJ Nuclear Tests (Australia v France), Judgment (1974) 267, paras 4344 and ICJ Nuclear
Tests (New Zealand v France), Judgment (1974) 472, paras 4647. Both judgments are based on
largely identical facts, were issued on the same day and contain for the most part identical
wording. The citations referred to in the following can therefore be found in both rulings,
unless indicated otherwise. Largely for reasons of both the readers and the writers convenience and in keeping with the treatment of the cases by most commentators, future footnotes
will make reference to Australia v France only.
45
Franck, Word Made Law, n 33 above, 613, for whom the ruling, just like the famous
US Supreme Court decision in Marbury v Madison, demonstrated that cases need not have
monumental outcomes to make monumental law, ibid 612.
46
See ICJ, Application of the Genocide Convention, Judgment (Merits) (2007) paras 37778;
ICJ, Armed Activities on the Territory of the Congo, Judgment (Jurisdiction and Admissibility)
(2006) paras 4553; ICJ, Frontier Dispute, Judgment (Merits) (1986) 57374 paras 3940; ICJ,
Paramilitary Activities in and against Nicaragua, Judgment (Merits) (1986) 132, para 261. The
ICJ here confirmed its doctrine, even though no binding commitment was in the end found
to exist in these cases. For more see chapter two.

A Legal Framework with Many Open Questions: A First Look13

had contemplated the existence of unilateral binding assurances termed

promises long before the Nuclera Tests cases,47 others have been highly
sceptical.48 While some scepticism might persist, today the existence of
promises in international law is met with broad acceptance.49 Its capability as a means for states to oblige themselves legally has further more
been confirmed by the work within the ILC, through its final outcome, the
Guiding Principles. Although the latter do not use the word promise, the
definition contained in Guiding Principle 1 is in the Commissions own
words very directly inspired50 by the pivotal decision in that very area, ie
the Nuclear Tests cases. It reads:
Declarations publicly made and manifesting the will to be bound may have the
effect of creating legal obligations. When the conditions for this are met, the
binding character of such declarations is based on good faith; States concerned
may then take them into consideration and rely on them; such States are entitled to require that such obligations be respected.51

The ILC has hence added some of its weight to the legal doctrine of
promises in international law by choosing to underline the ICJs reasoning.
Even though we are therefore far from facing a legal terra incognita when
47
Such as Garner, Balladore Pallieri, Verdross, De Nova or Suy, to name a few. For more
see chapter two and the scholarly debates depicted therein which arose in respect of the
cases and state practice leading up to the Nuclear Tests cases.
48
See especially R Quadri, Cours gnral de droit international public (1964) (III) 113
Recueil des Cours 245, 363; Alfred P Rubin, The International Legal Effects of Unilateral
Declarations (1977) 71 American J International Law 1. See also Charles de Visscher, Problmes
dinterprtation judiciaire en droit international public (Paris, 1963) 18688.
49
For scholarly support see, eg Cassese, International Law, n 2 above, 185; Carbone,
Promise in International Law, n 20 above, 166; Charpentier, Engagements Unilatraux,
n 21 above, 368, para 4; Vladimir-Djuro Degan, Unilateral Act as a Source of Particular
International Law (1994) 5 Finnish YB International Law 149, 188, para 12; Heintschel von
Heinegg, Einseitige Rechtsakte, n 2 above, 239, MN 17; J-P Jacqu, Elments pour une thorie
de lacte juridique en droit international public (Paris, 1972) 25557 and Jacqu, Promesse unilatrale, n 20 above, 327; Nguyen, Pellet and Daillier, Droit international public, n 2 above, 362;
Rodolfo De Nova, Die Neutralisation sterreichs (1958) 54 Die Friedenswarte 298, 299305;
Reuter, Droit international public, n 2 above, 164; Sicault, Engagements unilatraux, n 20
above, 634; Skubiszewski, Unilateral Acts, n 4 above, 224, 22829; Suy, Actes unilatraux, n 21
above, 109; Tomuschat, Unilateral Acts, n 4 above 1496; Visscher, Remarques, n 18 above,
esp 464; Gian Carlo Venturini, Attitudes et actes unilatraux des tats (1964) (II) 112 Recueil
des Cours 363, 396, 4005. For the Special Rapporteur, see especially Cedeo, Fourth Report,
n 10 above, 1819, paras 7984 and Cedeo, Seventh Report, n 12 above, 721. Rejecting any
classification but not the doctrine according to which unilateral assurances of states pledging a certain future course of conduct can be binding, see Brownlie, Principles, n 2 above,
64142. Simply referring to declarations (which are held apart from notification, protest
and renunciation) in this respect, see also Oppenheim, Jennings and Watts, International Law,
n 2 above, s 577, 1187. (However, as a protest, recognition or waiver, and even offers or an
agreement giving rise to a treaty, may be included in a declaration, the term declaration is
incapable of providing an adequate delimitation of the act here in question. On the term
declaration see also the discussion within the ILC as depicted at p 20 et seq.)
50
ILC Guiding Principles with Commentaries, n 14 above, Commentary on Guiding
Principle 1, para 1.
51
Ibid Guiding Principle 1.

14 Introduction
it comes to promises, a closer look, as will be taken in chapter three, at
the legal framework drawn up by the ILC quickly eradicates the assumption that unilateral declarations manifesting a will to be bound and promising certain future conduct are hence a legally well defined tool, ready
for effective use in inter-state relations. Unlike in the case of the codification of the law of treaties, where the Commission after years of work
concluded its mandate by presenting a precise and rather conclusive legal
framework, it failed to do so in the area of obliging declarations. At the
very outset of its study, the Special Rapporteur had rightly remarked that
the doctrine which had developed in this area was far from being consistent and that progressive development and not mere codification was necessary in this area of law.52 The ILC, however, never reached this stage in
its codification attempt and the finally published Guiding Principles, not
draft articles, hence failed to resolve pressing matters within the very area
of their application. In order to overcome the deadlock, the ILC decided
to rely heavily on dicta of the ICJ, an approach which is likely to appeal
to most international lawyers.53 And while this choice indeed paved the
way to drop the topic, it, however, at the same time secured that the
desired outcome of enhancing predictability in international law was not
achieved. The Commission itself had realised from the very beginning
that while the subject of unilateral acts had been touched upon by several judgments of the ICJ, and especially in the Nuclear Tests cases, . . . the
celebrated dicta leave room for uncertainties and questions.54 As the Guiding
Principles chose to cling to the Courts wording (while even loosening it
in part)55 these uncertainties and questions have now been directly transplanted into the Commissions final outcome. In light of the impasse into
which the ILC had manoeuvred itself, the final result might well have
been the only one feasible. Nevertheless, as is shown in chapter three in
more detail when discussing the precise legal framework, anyone reading through the Commissions final product while looking for guidance
when dealing with declarations of states will soon realise that he or she is
left with Guiding Principles which in core aspects fail to live up to their
names very own promise, as they simply do not guide.
III THE NEED FOR A CLEAR ASSESSMENT OF THE APPLICABLE LAW

Without having moved beyond the ICJs brief dicta, the Guiding Principles
for unilateral declarations have anything but closed the book on the topic.
Cedeo, First Report, n 21 above, 5, para 9 and 6, para 15.
See the analysis provided by Tomuschat, Unilateral Acts, n 4 above, 149293.
54
ILC, Report of the Working Group in (1996) II(2) YB International Law Commission 141,
Annex II, Addendum 3, para 3(b), emphasis added.
55
See pp 18394.
52
53

The Need for a Clear Assessment of the Applicable Law15

Quite on the contrary, the lengthy and thorough debates within the ILC,
the disagreement voiced over many important aspects as well as the various reports published on different issues, including the analysis of state
practice in the Special Rapporteurs Seventh Report, can be taken as a
new and fresh starting point for legal academia and research. Because notwithstanding the criticism of the Commissions final outcome, its call for
the identification of more precise rules in order to allow for more stability, predictability or in short security in international relations, is as
important today as it was when it began its work in 1997. Due to the everincreasing interplay of people as well as problems across borders, states
cannot refrain from interaction and, indeed, they dont. Yet, the fragmented and unclear rules operating within the area of unilateral promises
do not provide for any certainty in international relations quite on the
contrary. This status quo is all the more worrisome as states employ unilateral pledges even in highly sensitive areas. There are unilateral assurances
not to use nuclear weapons56 or to grant immunity from prosecution;57
states promise to allow foreign powers to fly their planes over their territories58 or have foreign nuclear-powered warships enter their ports,59 as
well as to deposit nuclear weaponry on the promising nations territory.60
States unilaterally declare not to execute extradited criminals or to refrain
from employing harsh interrogation techniques61 and torture,62 to name a
few more examples from state practice. As mentioned, a closer look at the
declaration of independence proclaimed by Kosovo reveals a full bouquet
of unilateral pledges, which were obviously considered vital in order to
smooth the way to international recognition. International jurisprudence
For the so-called negative-security assurance, see pp 16166.
See, eg the written declaration of the Minister of Defence of the Netherlands through
which states that participated in a NATO training exercise to be held in the Netherlands were
notified and which is cited as relevant state practice by the Netherlands itself in response
to the questionnaire prepared by the ILC and sent to governments by the UN SecretaryGeneral: The declaration was addressed in particular to the participating non-NATO States,
that is, States which were not parties to the NATO status-of-forces agreement, and contained
a promise to provide to members of their military forces similar facilities, exemptions and
waiver of jurisdiction for crimes and offences as contained in the agreement, in ILC, Replies
from Governments to the Questionnaire, Report of the Secretary-General, UN Doc A/
CN.4/511 (2000) 4.
58
See Tomuschat, Unilateral Acts, n 4 above, 149697.
59
See the declaration made by New Zealand vis--vis the United States in 1982 according
to which the former will not forbid US nuclear-powered warships to enter its ports, cited in
(1983) 87 Revue Gnrale de Droit International Public 405.
60
For more see pp 14851.
61
See the declarations of the English representative before the European Court of Human
Rights, addressed at pp 14145.
62
GA Res A/RES/32/64 of 8 December 1977 called upon Member States to make unilateral declarations against torture and other cruel, inhumane or degrading treatment, urging
them to give these declarations a maximum of publicity. A model unilateral declaration
was annexed. For more see p 140. Interesting in this context also Ashley Deeks, Promises
Not to Torture: Diplomatic Assurances in U.S. Courts, ASIL Discussion Paper (December 2008).
56
57

16 Introduction
and legal commentators alike have provided ample reason for states to
rely on unilateral declarations made. To again cite Thomas Franck:
Thanks to the Courts decision, each state must now recognize that what it solemnly says it will do, or, more important, what it says it will not do, becomes a
part of that trellis of reciprocal expectations on which the fragile international
system grows.63

As presented in chapter two, promises are indeed used by states; they

surface in the international legal debate, are relied upon and adjudicated before the ICJ. Yet, as the beginning of chapter three illustrates, the
Guiding Principles established by the ILC reveal a lack of normative precision in this area. One may be tempted to say: states are currently playing
the game, without really knowing the rules. More accuracy in determining the legal framework could, on the one hand, help to establish trust in
binding declarations made, whilst, on the other hand, indicating in which
cases state action might be taken as a sign of goodwill but will not be considered as binding, and hence not fully trustworthy by an international
lawyer nor will it allow for the legal sanctions available for obligations
breached under international law. While this is not to say that legal commitments are necessarily and always more reliable than merely political
ones, it should not be overlooked that whether a declaration is considered to be binding or not is likely to be one factor influencing the amount
of trust placed on it by the addressee, just as it is likely to influence the
declarant in assessing whether he has to abide by the assurance made.64
A solemn pledge is somewhat weakened if it is based on the addressees
assumption that it is not binding, ie modifiable and revocable at will. If
therefore, in our legalised world,65 the international lawyers answer to
the question: Is State A bound by its declaration?, in light of a rudimentary framework, remains a firm Maybe!, without him even being able
to spell out when this maybe is likely to tilt either way, law can not be
considered as stabilising international relations. Despite numerous rulings on the matter, currently, uneasiness still prevails when dealing with
unilateral declarations.66 While this uneasiness is understandable in light
63
Franck, Word Made Law, n 33 above, 616. Similarly, de Visscher, Remarques, n 18 above,
465, emphasising that states now have to be more prudent when they make declarations.
64
As Martti Koskenniemi, a critical observer of legal argument himself, points out in a
footnote with further references: It is popularly assumed that legal considerations play little part in foreign affairs decision-making. Understandably, the experience of legal advisers tends to be the contrary, Martti Koskenniemi, From Apology to Utopia: the Structure of
International Legal Argument (reissue with a new epilogue, Cambridge, 2005) 27, note 26.
65
As David Kennedy remarks: Although it is easy to think of international affairs as a
rolling sea of politics over which we have managed to throw but a thin net of legal rules, in
truth the situation today is more the reverse. There is law at every corner, David Kennedy,
Of War and Law (Princeton, NJ, 2006) 25.
66
See, eg the reactions by the non-nuclear-weapon states, p 161 et seq. See also Fiedler,
Einseitige Versprechen, n 20 above, 50, who speaks of negative Grundhaltungen, which
may roughly be translated as negative attitudes.

Approaching the Topic17

of the aforesaid, legal analysis alone may indeed help to reduce the grey
areas and provide answers to several questions raised.
IV APPROACHING THE TOPIC

In choosing to work on a topic which the International Law Commission

has left fraught with legal challenges, this study and its author are not
driven by hubris. First, in learning from the effort undertaken by the
Commission, this book will focus only on promises and not on unilateral acts of states in general. Furthermore, this authors or, more generally
speaking, any academics loneliness brings with it an important advantage even compared to a group of highly distinguished experts, especially
as very opposing views were present within the Commission. While having the benefit of being able to access the wealth of material provided by
the Commission and to know what some of the best scholars in the field
have to say about the topic, here there is no need to struggle with many
others for a common opinion which, in the face of insurmountable differences, is likely to boil down to the smallest common denominator, or
indeed a compromise which does not work either way. Instead, the topic
of binding unilateral assurances will be presented as a whole and in a
concerted manner; the fields of debate will be highlighted along the way,
in trying to delimit promises as a more usable and more dependable unilateral act in inter-state relations.
In order to do so, the book will start by delimiting the subject matter,
that is, narrow unilateral state action down to declarations made by states
which fall into the ambit of the legal framework applicable to promises.
Subsequently, a History of Promises is presented in which important
cases as well as examples of state practice are briefly analysed. What follows is a chapter on the legal framework which has evolved for promises and in which an attempt is made to move from the rather shattered
and confusing legal scenario sketched out in an analysis of the Guiding
Principles to a more coherent legal picture. As is shown, this is possible
without necessarily having to delve into considerations de lege ferenda, if
only questions such as the legal basis for the obligation undertaken are
clarified beyond the use of the rather elusive term of good faith. From
there, various answers, such as for example how to take the addressees
reaction into account, can be found much more easily. As the relationship,
interplay and differences between a promise and the estoppel principle
are important, an analysis on promises triggering estoppel is included.
The final chapter provides a brief outlook and deals with the dangers and
prospects of international promises and the question whether there is a
need for further codification and development in this area of international
law.

1
Delimiting the Subject:
Promise as a Unilateral Act

HIS BOOKS CENTRE of attention rests on unilateral declarations

which are held to manifest a states will to commit itself legally to
follow a certain line of conduct in the future. It is this chapters aim
to describe and define this sort of declarations more closely.
I DRAWING FROM THE DEFINITION OF UNILATERAL ACTS

As the above-mentioned state action forms a subcategory of unilateral

acts of states, it would seem methodologically sensible to start from their
overall definition and subsequently identify and simply add on those
characteristics which will finally single out the unilateral act of promise.
The definition of unilateral acts has, however, proven extremely difficult1
and despite the International Law Commissions efforts in this field, it
remains true that no all-encompassing, uniform and precise definition of
unilateral acts exists.2 In its 10 years of work there were various attempts
by the ILC to define unilateral acts of states but the issue remained
unsettled till the very end.3 Yet, surprisingly, the commentary to Guiding
Principle 1 says:
1
Jean-Paul Jacqu, Elments pour une thorie de lacte juridique en droit international public
(Paris, 1972) 320 stated in 1972: La notion dacte unilatral est actuellement trs controverse et les auteurs se divisent quant la dfinition de lacte juridique unilatral en droit international public.
2
Wilfried Fiedler, Unilateral Acts in International Law in Rudolf Bernhardt, Peter
Macalister-Smith and Max-Planck-Institut (eds), Encyclopedia of Public International Law
(Amsterdam/New York, 2000) vol 4, 1018.
3
The Special Rapporteur intended to limit his study to strictly unilateral declarations
in his First Report. A declaration of relevance was to be an autonomous expression of clear
and unequivocal will, explicitly and publicly issued by a State, for the purpose of creating a juridical relationship in particular to create legal obligations between itself and
a third State which did not participate in its elaboration, without it being necessary for
this third State to accept it or subsequently behave in such a way as to signify acceptance,
Victor Rodriguez Cedeo, First Report on Unilateral Acts of States, UN Doc A/CN.4/486
(1998) para 170. After some debate within the ILC and recommendations by the Working
Group (see ILC, Report of the Working Group, UN Doc A/CN.4/L.558 (1998)) the following
Draft Article 2 was proposed in the Second Report: Unilateral legal acts of States. For the

Drawing From the Definition of Unilateral Acts19

The wording of Guiding Principle 1, . . . seeks both to define unilateral acts in the
strict sense and to indicate what they are based on.4

purposes of the present draft articles, unilateral legal act means an unequivocal, autonomous expression of will, formulated publicly by one or more States in relation to one or
more other States, the international community as a whole or an international organization,
with the intention of acquiring international legal obligations, Victor Rodriguez Cedeo,
Second Report on Unilateral Acts of States, UN Doc A/CN.4/500 (1999) para 38. In particular, the elements unequivocal, autonomous and publicly triggered a lively debate
within the Commission whose reconvened Working Group subsequently defined the following concept as the new starting point of the study: A unilateral (autonomous) statement by a State by which such State intends to produce legal effects in its relations to one or
more States or international organizations and which is notified or otherwise made known
to the State or organizations concerned, ILC, Report of the Working Group, UN Doc A/
CN.4/L.588 (1999) para 10. The Special Rapporteur, however, proposed an altered version in
his Third Report one year later, reintroducing the word unequivocal as well as, inter alia,
changing the Working Groups formulation of is notified or otherwise made known. Article
1, now officially titled Definition of unilateral acts, read as follows: For the purposes of the
present articles, unilateral act of State means an unequivocal expression of will which is
formulated by a State with the intention of producing legal effects in relation to one or more
other States or international organizations, and which is known to that State or international
organization, Victor Rodriguez Cedeo, Third Report on Unilateral Acts of States, UN
Doc A/CN.4/505 (2000) para 80. As was to be expected, both aspects met with some criticism from within the ILC (see, eg the reactions of Gaja, ILC, Summary Record of the 2628th
Meeting, UN Doc A/CN.4/SR.2628 (2000) para 21; Al-Baharna ILC, Summary Record of the
2629th Meeting, UN Doc A/CN.4/SR.2629 (2000) para 21; and Simma, ibid. 2629th Meeting,
para 40). The Draft Article was nevertheless forwarded to the Drafting Committee which,
however, never made any proposals or issue a report. The same Article surfaced again in
the Special Rapporteurs Fifth (largely recapitulative) Report in which it was presented to
the (partially) newly comprised Commission, see Victor Rodriguez Cedeo, Fifth Report
on Unilateral Acts of States, UN Doc A/CN.4/525 (2002) para 81. Although the Special
Rapporteurs Sixth Report focused largely on recognition, the ongoing problems concerning the delimitation of the topic were reflected by its rather sombre introduction: It is true
that it has not been clearly established that the institution of unilateral legal acts exists, and
the existence of such an institution is by no means clearly defined in international law, even
though there are major doctrinal and case law elements and even a certain State practice
that could demonstrate the existence of the institution, Victor Rodriguez Cedeo, Sixth
Report on Unilateral Acts of States, UN Doc A/CN.4/534 (2003) para 1. The Working Group
convened that year again tried to delimit the study, recommending the following: For the
purposes of the present study, a unilateral act of a State is a statement expressing the will or
consent by which that State purports to create obligations or other legal effects under international law, ILC, Report of the Chairman of the Working Group, UN Doc A/CN.4/L.646
(2003) 2, recommendation 1. After having focused on state practice in his Seventh and Eight
Reports, the Special Rapporteurs Ninth and final Report offered Draft Guiding Principles
and a new and short definition under Guiding Principle 1: Definition of a unilateral act: A
unilateral act of a State means a unilateral declaration formulated by a State with the intent
of producing certain legal effects under international law, Victor Rodriguez Cedeo, Ninth
Report on Unilateral Acts of States, UN Doc A/CN.4/569 (2006) Guiding Principle 1. This
definition, however, was not adopted by the Commission; instead, it agreed, in its final outcome, on the much more limited Guiding Principles. For the definition included therein see
the text following this note above.
4
ILC, Guiding Principles applicable to Unilateral Declarations of States Capable of
Creating Legal Obligations, with Commentaries thereto, ILC Report A/61/10 (2006) ch IX,
Commentary to Guiding Principle 1, para 1, emphasis added. It reads in full: The wording of Guiding Principle 1, which seeks both to define unilateral acts in the strict sense and
to indicate what they are based on, is very directly inspired by the dicta in the Judgments
handed down by the International Court of Justice on 20 December 1974 in the Nuclear Tests
case.

20 Delimiting the Subject: Promise as a Unilateral Act

As the relevant part of Guiding Principle 1 reads Declarations publicly
made and manifesting the will to be bound may have the effect of creating
legal obligations,5 unilateral acts in the strict sense6 would accordingly
be defined as declarations publicly made and manifesting the will to be bound.
The ILC points out that it was directly inspired by the Nuclear Tests cases
in adopting this definition7 and it is thus not surprising that its wording comes close to describing, without however precisely defining, the
very subcategory of unilateral acts with which the International Court of
Justice was dealing in its ruling, ie that of binding unilateral assurances,
termed promises in international law. For the following reasons, however, it cannot be taken as our starting point and be understood to constitute the all-encompassing, uniform and precise definition of unilateral
acts, the absence of which was lamented before.
Whether the word act may be substituted by the term declaration
had been debated within the ILC and several members correctly voiced
their opposition,8 since declarations are an underinclusive description
of unilateral state action. While a vast majority of unilateral acts of states
might indeed be found in declarations, and promises are held to be limited to declarations, not all unilateral acts are. A declaration relates to the
form in which a manifestation of will is presented and can only be understood to mean an express act; any behavior that might not express yet
imply a manifestation of will can hardly be considered a declaration and
would thereby be discarded as a potential unilateral act of state.9 That
non-express state conduct may constitute a unilateral act is, however, far
from being doubtful. The commencement of diplomatic relations is usually considered as an act of implicit recognition of the state with which
formal diplomatic relations are established even though recognition as
such is not declared.10 The same is true for protests which will usually but
Ibid Guiding Principle 1.
The formulation strict sense relates back to the ICJs choice of words in the Nuclear
Tests cases where the Court spoke of a declarations strictly unilateral nature. See p 12.
7
See the Commentary to Guiding Principle 1, n 4 above.
8
See, eg Pambou-Tchivounda, ILC, Summary Record of the 2525th Meeting, UN Doc
A/CN.4/SR.2525 (1998) para. 1, wondering how the Special Rapporteur came to single
out declarations. Alain Pellet criticised this approach numerous times, for what he himself
termed a firm opposition, see ILC, Summary Record of the 2594th Meeting, UN Doc A/
CN.4/SR.2594 (1999) para 17.
9
To hold otherwise and the word declaration to include implicit action would either
make it meaningless or the definition redundant. Because if every manifestation of will, even
when taken to be merely implied in state conduct, is said to constitute a declaration, then
the word declaration might as well be dropped; the will needs to be manifested but the
form is irrelevant. See also Pellet, ILC, Summary Record of the 2526th Meeting, UN Doc A/
CN.4/SR.2526 (1998) para 39: Again, by confining matters to universal declarations, stress
was being placed on the form. The Commission should arrive at a balanced definition of a
unilateral act, which was a manifestation certainly, but a manifestation of will.
10
On implied recognition see, eg Malcolm N Shaw, International Law, 6th edn (Cambridge/
New York 2008) 462 with further references.
5
6

Drawing From the Definition of Unilateral Acts21

need not necessarily be included in a declaration: calling your ambassador home is a quite frequently used mechanism employed by states in this
regard and warning shots by a military vessel or a jet vis--vis an intruder
into contested waters or airspace may also effectively deliver the protest
message without declaring it.11 Even a waiver, although never presumed,
may according to legal doctrine be made implicitly.12
The term declaration furthermore wrongly excludes mere silence from
ever constituting a legally relevant act. While the state acting implicitly
does not spell out its message of, for example, recognition or protest (and
is therefore silent in this regard), it still acts; the silent state however
refrains from any physical action. The distinction is quite subtle, as legal
doctrine might nevertheless infer a legally relevant will from a state that
remains silent and completely inactive; it will do so in situations which
warrant a protest. In these settings saying nothing is considered as noticeable absence of protest, ie as having said nothing against a certain situation, and is turned into and now termed qualified or eloquent silence.13
A state which in the face of a notorious claim persistently refrains from
11
Yet, protest is often described as an act of formal communication and indeed by
some said to be an express act, see Georg Schwarzenberger, International Law as Applied by
International Courts and Tribunals, 3rd edn (London, 1957) vol 1, 552; Krzysztof Skubiszewski,
Unilateral Acts of States in M Bedjaoui (ed), International Law: Achievements and Prospects
(1991) 227, para 33; Eric Suy and Nicolas Angelet, Rechtsgeschfte, einseitige in Ignaz SeidlHohenveldern (ed), Lexikon des Rechts, Vlkerrecht, 3rd edn (Neuwied, 2001) 320. For protest not having to be explicit see Franz Pfluger, Die einseitigen Rechtsgeschfte im Vlkerrecht
(Zurich, 1936) 21112; Jacqu, Elments, n 1 above, 344 and Wolfram Karl, Protest in Rudolf
Bernhardt (ed), Encyclopedia of Public International Law (Amsterdam, 1986) vol III, 1158, who
besides warning shots mentions protest by implication, eg as a byproduct of countermeasures (despite having defined protest not only as formal communication but communication
which had to be issued by, and addressed to, the parties official representatives); see also
Eric Suy, Les actes juridiques unilatraux en droit international public (Paris, 1962) 50: nimporte
quel acte peut contenir une protestation, pourvu que lintention du protestataire y soit clairement nonce. Suy, however, excludes acts which result in direct hostilities or a war and
which are not covered by United Nations Charter, Art 51, asking comment voudrait-on protger ses droits en violant des autres?, ibid 52. But an act which clearly signals a states will
not to acquiesce and to protect its rights will hardly cease to do so when crossing the line of
legality. While the states responsibility for violating international law will, of course, be triggered in these cases, a state which resorts to force over a matter, even though it is illegal, still
(unduly) expresses its will to protest and cannot be considered to have acquiesced.
12
See Pfluger, Einseitige Rechtsgeschfte, n 11 above, 6970; Suy, Actes unilatraux, n 11
above, 15758. Jacqu, Elments n 1 above, 342; Anne M Trebilcock, Waiver in Rudolf
Bernhardt (ed), Encyclopedia of Public International Law (Amsterdam, 1984) vol IV, 1329.
13
See ICJ, Temple of Preah Vihear (Cambodia v Thailand), Merits, Judgment of 15 June 1962,
[1962] ICJ Rep 6, 23: it is clear that the circumstances were such as called for some reaction,
within a reasonable period, on the part of the Siamese authorities, if they wished to disagree
with the map or had any serious question to raise in regard to it. They did not do so, either
then or for many years, and thereby must be held to have acquiesced. Qui tacet consentire videtur si loqui debuisset ac potuisset. For qualified silence and its prerequisites see, eg Jean Paul
Mller and Thomas Cottier, Acquiescence in Rudolf Bernhardt (ed), Encyclopedia of Public
International Law (1992) vol I, 14; Pfluger, Einseitige Rechtsgeschfte, n 11 above, 197, Suy, Actes
unilatraux, n 11 above, 66; Charles de Visscher, Problmes dinterprtation judiciaire en droit
international public (Paris, 1963) 168.

22 Delimiting the Subject: Promise as a Unilateral Act

protesting,14 will hence be taken to have expressed its tacit recognition
or acceptance in the form of acquiescence.15 A state by remaining silent
can therefore be held to have unilaterally manifested its will in a legally
relevant way, ie to have legally acted, a fact that needs to be reflected in
any definition adopted.16 The word declaration clearly does not do so.17
Also, protest is commonly considered to constitute a unilateral act and
the ILC indeed thought it was.18 But even he who expressly protests does
not make a public declaration manifesting the will to be bound, quite on
the contrary. While protest is a manifestation of will, it is a manifestation
of a will not to be bound, not to acquiesce and is designed to hinder the
creation of any legal obligation. A protesting state is in no way bound by
its declaration, as yesterdays protest in no way restrains todays recognition should a state change its mind.
Finally, it is quite surprising that the purported definition of unilateral acts in the strict sense does not make any effort nor even allude to
the declarations necessarily unilateral, let alone strictly unilateral character. Not every declaration publicly made and manifesting the will to
be bound can be seen as constituting a unilateral act in the strict sense
as the definition now seems to claim. Unilateral acts covered by the Law
of Treaties, such as the acceptance of an offer, the accession or the reservation to a treaty, are often said to be excluded from the regime here in
question because they lack a strictly unilateral character. All of them are,
however, encompassed in the ILCs definition, which uses neither the
14
For the prerequisites of a duty to protest, see also ICJ, Fisheries (United Kingdom v
Norway), Judgment of 18 December 1951, [1951] ICJ Rep 116, 139: The notoriety of the facts,
the general toleration of the international community, Great Britains position in the North
Sea, her own interest in the question, and her prolonged abstention would in any case warrant Norways enforcement of her system against the United Kingdom.
15
On acquiescence as equivalent to tacit recognition by conduct, see especially ICJ, Gulf of
Maine (Canada v USA), Judgment of 12 October 1984, [1984] ICJ Rep 246, 305, para 130: They
[estoppel and acquiescence] are, however, based on different legal reasoning since acquiescence is equivalent to tacit recognition manifested by unilateral conduct which the other party may
interpret as consent, while estoppel is linked to the idea of preclusion, emphasis added. See
also the Qui tacet consentire videtur si loqui debuisset ac potuisset citation in ICJ, Temple
of Preah Vihear, n 13 above, 23, as well as Pfluger, Einseitige Rechtsgeschfte, n 11 above, 196:
auch das Stillschweigen eines Staates [kann] unter besonderen Umstnden als Anerkennung
oder Verzicht ausgelegt werden, with further references. See also Suy, Actes unilatraux,
n 11 above, 67; Pellet, ILC, Summary Record of the 2629th Meeting, UN Doc A/CN.4/
SR.2629 (2000) 137, para 69: While some kinds of silence definitely did not and could not
constitute a unilateral act, others might be described as an intentional eloquent silence
expressive of acquiescence and therefore did constitute such an act.
16
The Special Rapporteur wanted to exclude silence in particular for its alleged lack of autonomy, Cedeo, Third Report, n 3 above, 1819, paras 12633. On autonomy see p 55 et seq.
17
The Guiding Principles in their Preamble acknowledge that silence can result in a binding commitment for a state. For more see also p 184.
18
See, eg the outline in ILC, Report of the Working Group, UN Doc A/CN.4/L.543 (1997)
6 under ch III(iv); Victor Rodriguez Cedeo, Fourth Report on Unilateral Acts of States, UN
Doc A/CN.4/519 (2001) 2122, paras 9496; Victor Rodriguez Cedeo, Seventh Report on
Unilateral Acts of States, UN Doc A/CN.4/542 (2004) 4870, paras 89178.

Distinguishing Promises from Other Classes of Unilateral Acts23

word strictly nor autonomous in order to restrict its scope. While abandoning the material concept of autonomy or that of a strictly unilateral
act as commonly understood is possible through the application of the lex
specialis principle, and indeed preferable,19 the Commissions outcome is
contradictory when it drops any reference to an acts unilateral, let alone
strictly unilateral character in its definition, but via the Commentary
claims to uphold this concept and even define what a strictly unilateral
act is.
Despite its Commentarys claim to the contrary, the adopted Guiding
Principle 1 therefore neither provides a valid definition of all unilateral
acts, as its focus on declarations made in order to be bound is too narrow, nor does it define unilateral declarations (much less acts) in the strict
sense, as in this regard it is drafted too broadly.20
The delimitation of the unilateral act labelled promise will therefore
not start from an accepted overall definition of unilateral acts in general,
for the reason that the latter still cannot be considered to have been found.
While such a general definition would have been a helpful starting point,
it is not a necessary prerequisite as long as the unilateral act of interest
here is found to be clearly delimitable.
II DISTINGUISHING PROMISES FROM OTHER CLASSES
OF UNILATERAL ACTS

Even though no final definition of unilateral acts was found within the
ILC, a perusal of the various definitions proposed21 clearly shows that the
Commission was not going to be concerned with all sorts of unilateral
state action.22 Instead, it chose from the outset, and in conformity with
legal doctrine, to focus only on acts of state which may be considered as
an expression of will, carrying an intention to have a legal effect and to
which international law will attribute the (apparently) intended effect of
creating, preserving, altering or extinguishing a states legal relation to
one or numerous other subjects of international law.23
On autonomy see pp 5578.
Note also that while according to the Commentary unilateral acts in the strict sense are
defined by Guiding Principle 1, the Principles Preamble contains another and different but
no less satisfactory definition. The Preamble in its last paragraph reads: [The ILC] adopts
Guiding Principles which relate only to unilateral acts stricto sensu, ie those taking the form
of formal declarations formulated by a state with the intent to produce obligations under
international laws. In comparison to Guiding Principle 1, the word publicly is missing, the
word formal is introduced, and manifesting the will to be bound is substituted by formulated with the intent to produce obligations.
21
See n 3 above.
22
On how the term unilateral is to be understood see p 40 et seq.
23
This is not the wording adopted by the ILC but rather a description of what the
Commission focused in on. An overview was provided in n 3 above of the various (working) definitions debated as well as the concept adopted within the ILC Working Groups
19
20

24 Delimiting the Subject: Promise as a Unilateral Act

Despite imposing this first filter on unilateral acts of states, the ILC
quickly realised that it was still faced with a great variety of acts, a fact
which made it seemingly impossible to apply a one rule fits all approach
to them. The Commission had nevertheless intended to address a vast
array of state actions and, as already indicated, had anticipated overcoming the obstacle presented by the various acts peculiarities by dividing
the draft articles into a general and a special part, the latter containing
subsections, each of which were destined to deal with a typified sort of
unilateral act.24
In attempting to start with the design of a general part on all unilateral
acts, the Commission acted in conformity with a majority of legal doctrine, which assumes the existence of a coherent category of state action
to be labelled unilateral acts of states and encompassing acts which
share enough similarities to have a common set of rules made applicable
to them. Yet, in light of the great differences between the various acts of
states qualifying for this category, rules applicable to each and every act
can only be elaborated if they are either formulated very broadly or cover
only a limited set of general questions.
The following aspects fit this description and were to be contained in
the General Part on which the Special Rapporteur was working till the
very (abrupt) end: the capacity of states, amongst others, to formulate unilateral acts is commonly answered in the affirmative and was designed as
the starting ground within the General Part. It indeed leaves little room
for controversy as it is of no great regulatory use and even those opposing
of 1997 and 1999. In legal doctrine, often no precise and clear-cut definition is provided
but rather a description of various unilateral acts or various features a unilateral act is said
to have; see, however, the definitions provided by Pfluger, Einseitige Rechtsgeschfte, n 11
above, 28; see also Alfred Verdross and Bruno Simma, Universelles Vlkerrecht: Theorie und
Praxis, 3rd edn (Berlin, 1984) 425, s 662: Einseitige Rechtgeschfte sind Willenserklrungen
eines einzelnen Staates oder einer organisierten Staatengemeinschaft, durch welche die von
diesen Rechtssubjekten gewollten Rechtsfolgen im Rahmen des allgemeinen Vlkerrechts
ausgelst werden [Unilateral acts are manifestations of will of a single state or of an
organised community of states through which the legal consequences willed by these legal
subjects are brought about within the general framework of international law (my translation). The translation is, however, slightly imprecise as the German terminology already differentiates between legal acts in general (Rechtsakte), which include merely wilful action
that is important (only) as a legal fact, and acts which are manifestations of will that create the intended legal effect (ie Rechtsgeschfte). The English language speaking simply of
unilateral acts of states would appear to cover both forms of state action, but the former is
not meant when speaking of unilateral acts in this context (see pp 3840). It would hence
be more accurate, yet quite cumbersome, to speak of a definition of unilateral legal acts as
understood by doctrine]. Suy, Actes unilatraux, n 11 above, 44 concludes by considering
unilateral acts to be unilateral manifestations of will which are imputable to a single subject
of international law, do not depend on another juridical act for their effectiveness, produce
no obligations for third parties, and to which a norm of international law attaches the legal
consequences corresponding with the states manifested will.
24
See, eg ILC, Report of the Working Group, ILC Report 2000, UN Doc A/55/10 (2000)
100, para 621(b); Cedeo, Fourth Report, n 18 above, which provides a classification of unilateral acts.

Distinguishing Promises from Other Classes of Unilateral Acts25

a codification of unilateral acts can at least live with a statement indicating a states ability to act unilaterally. Also, the non-existence of formal
requirements for unilateral acts is fairly settled, especially in the light of
clear ICJ jurisprudence in this regard and the fact that international law
is generally considered to impose no formal requirements unless explicit
stipulations to the contrary exist.25 The question who has the competence
to formulate a unilateral act on behalf of a state is equally practically
uncontroversial, at least and as long as it remains limited to those state
representatives mentioned by Article 7(2)(a) of the Vienna Convention
on the Law of Treaties, ie Heads of State, Heads of Government and
Ministers of Foreign Affairs.26 Also, any manifestation of will issued by a
state is usually held to have to be without vices in order for international
law to give it legal effect, just as consent in the area of treaties has to be,
and hence error, fraud, corruption as well as the coercion of a state or
its representative as codified for treaties in Vienna Convention, Articles
48 to 52 are usually mentioned in this regard. Similarly, there is no visible opposition to holding void a unilateral act violating a norm of ius
cogens. But this is as far as it goes in relation to the bulk of unilateral acts.
Even though an analogy to Articles 46 (and 27) of the Vienna Convention27
is sometimes drawn28 and might indeed be sensible (although currently
25
Referring to unilateral declarations creating legal obligations, the ICJ explicitly states
the following in ICJ, Nuclear Tests (Australia v France), Judgment of 20 December 1974, [1974]
ICJ Rep 253, 26768, para 45: With regard to the question of form, it should be observed
that this is not a domain in which international law imposes any special or strict requirements. Whether a statement is made orally or in writing makes no essential difference, for
such statements made in particular circumstances may create commitments in international
law, which does not require that they should be made in written form. Thus the question of
form is not decisive. See also ICJ, Temple of Preah Vihear (Cambodia v Thailand), Preliminary
Objections, Judgment of 26 May 1961, [1961] ICJ Rep 17, 31: Where . . . as is generally the
case in international law, which places the principlal emphasis on the intention of the parties, the law prescribes no particular form, parties are free to choose what form they please
to provided their intention clearly results from it. ILC Guiding Principle 5 hence reads:
Unilateral declarations may be formulated orally or in writing. The same is true for treaties;
while the Vienna Convention applies only to written treaties, it does not negate the existence
of oral treaties or their legal effects, see Arts 2(1)(a), 3.
26
Again the ICJs jurisprudence supports this position, see ICJ, Armed Activities on the
Territory of the Congo (Democratic Republic of the Congo v Rwanda), New application 2002,
Jurisdiction and Admissibility, Judgment of 3 February 2006, para 46: the Court observes
that, in accordance with its consistent jurisprudence, [references omitted], it is a well-
established rule of international law that the Head of State, the Head of Government and the
Minister for Foreign Affairs are deemed to represent the State merely by virtue of exercising
their functions, including for the performance, on behalf of said State, of unilateral acts having the
force of international commitments, emphasis added.
27
Vienna Convention, Art 46 reads as follows: Provision of internal law regarding the
competence to conclude treaties: 1. A State may not invoke the fact that its consent to be
bound by a treaty has been expressed in violation of a provision of its internal law regarding the competence to conclude treaties as invalidating its consent unless that violation was
manifest and concerned a rule of its internal law of fundamental importance. 2. A violation
is manifest if it would be objectively evident to any State conducting itself in the matter in
accordance with normal practice and in good faith.
28
For more see pp 23437.

26 Delimiting the Subject: Promise as a Unilateral Act

national constitutional norms applicable to unilateral commitments by
states seem to be largely lacking), it is not a norm of general applicability as long as protest is considered to be a unilateral act of state subject
to the general framework, since the latter does not express a consent to
be bound. Going a step further and turning towards the central question
whether unilateral acts are revocable, an aspect intimately related to their
binding force, will also lead us away from norms of general applicability.29
Hence, a differentiated approach is called for and with it a categorisation
of unilateral state action.30
While legal doctrine has shown that it is possible to categorise unilateral state action using various criteria, such as eg their form or their
29
While recognition and waiver are considered irrevocable (for more see pp 2933 and
3436), the regime applicable to promises (as presented at pp 25176) is different. Protests
need not be revoked (see pp 3334).
30
The categorisation of unilateral acts is therefore very common and present in nearly
every publication on unilateral acts, and in our opinion for the above-mentioned good
reasons. As to categorising unilateral acts, see, eg Antonio Cassese, International Law, 2nd
edn (Oxford, 2004) 184, who mentions especially protest, recognition, renunciation, notification and promise; Jean Charpentier, Engagements unilatraux et engagements conventionnels: diffrences et convergences in Jerzy Makarczyk (ed), Theory of International Law
at the Threshold of the 21st Century (The Hague/London/Boston, 1996) 368, focusing on
promise, recognition and renunciation; Vladimir-Djuro Degan, Unilateral Act as a Source of
Particular International Law (1994) 5 Finnish YB International Law 149, 184, addressing, inter
alia, promise, waiver, occupation; Fiedler, Unilateral Acts, n 2 above, 1018, considering recognition, protests, renunciation, notification and at times, acquiescence and revocation to
be accepted and delimitable on the basis of customary law, for promises, however, considerable legal uncertainties remained; Wolf Heintschel von Heinegg, Einseitige Rechtsakte
in Knut Ipsen (ed), Vlkerrecht, 5th edn (Mnchen, 2004) esp 23840, MN 1417, mentioning recognition, protest, waiver and promise; Jacqu, Elments, n 1 above, 335, including
subchapters on waiver, protest and promise plus recognition; Quoc Dinh Nguyen, Alain
Pellet and Patrick Daillier, Droit international public, 7th edn (Paris, 2002) 36162, listing
notification (as a condition for the validity of all others), recognition, protest, waiver and
promise; Skubiszewski, Unilateral Acts, n 11 above, 227, differentiating between recognition, protest, promise and waiver; Suy, Actes unilatraux, n 11 above, 41, classifying acts into
those which create rights, ie promise and recognition, which preserve rights, ie protest, and
which abandon rights, ie waiver/renunciation; Christian Tomuschat, Unilateral Acts under
International Law in Droits et culture. Mlanges en lhonneur du Doyen Yadh Ben Achour (Tunis,
2008) 1496, referring to promise, waiver, recognition and protest as the four main categories
of unilateral acts; Gian Carlo Venturini, Attitudes et actes unilatraux des tats (1964) (II)
112 Recueil des Cours de lAcadmie de Droit International de La Haye 363, identifying promise
(which included the subcategory of recognition), waiver, revocation, declarations of war and
others; Verdross and Simma, Universelles Vlkerrecht, n 23 above, 42631, ss 66471, listing
recognition, protest, waiver and promise, although there was some controversy on the latter.
See also Cedeo, Fourth Report, n 18 above, 1022, paras 44100. Contra any classification,
Brownlie in various statements in the Commission, eg ILC, Summary Record of the 2696th
Meeting, UN Doc A/CN.4/SR.2696 (2001) para 25, as well as in Ian Brownlie, Principles of
Public International Law, 7th edn (Oxford/New York, 2008) 612: analysis in terms of categories of promise, protest, and the like is superficial, and tends to confuse conditioning
facts and legal consequences; similarly Camille Goodman, Acta Sunt Servanda? A Regime
for Regulating the Unilateral Acts of States at International Law (2006) 25 Australian YB
International Law 43, 51. Considering the problematique of promises to be better addressed
as one pertaining to all unilateral acts of states and an extra subcategory to be superfluous, Wilfried Fiedler, Zur Verbindlichkeit einseitiger Versprechen im Vlkerrecht (1976) 19
German YB International Law 35, 6871.

Distinguishing Promises from Other Classes of Unilateral Acts27

dependence,31 a distinction focusing on the material content of the act

has proven most suitable when attempting to group only that sort of conduct together to which the same legal rules can be said to apply.32 And
indeed, as form hardly matters when it comes to unilateral acts33 and
an acts legal effect is primarily crafted according to a states expressed
will, it is only logical that state action with similar content, ie containing
similar manifestations of will, is likely to trigger similar effects under
international law.34 Legal doctrine accordingly usually differentiates
between the four unilateral acts already mentioned: promise, recognition,
waiver and protest.35 Never intended as an exhaustive list, these four are
considered as very common unilateral expressions of states, which are
furthermore not embedded in a special legal framework.36 Some authors
have added or subtracted categories from and to the list37 or established
intermediate categories, placing especially promise and recognition
together.38 Also, notification the act by which a state makes a certain fact
to which legal consequences may be attached known to its addressee(s)39
has rather frequently been identified as an additional typical unilateral
On the dependency or autonomy of an act, see pp 5578.
cf the categories chosen by all authors mentioned in n 30 above, and especially
Skubiszewski, Unilateral Acts, n 11 above, 223. For a perusal of literature and/or various
possible classifications see Suy, Actes unilatraux, n 11 above, 33; Cedeo, Fourth Report, n 18
above, 12; Pfluger, Einseitige Rechtsgeschfte, n 11 above, 64.
33
See n 30 above.
34
Some scholars have distinguished amongst unilateral acts according to their legal
effect as the Special Rapporteur has termed it, see Cedeo, Fourth Report, n 11 above,
paras 55, 60, 61. To do so in the context of elaborating a legal framework would, however,
end in circular reasoning: if we knew an acts legal effect, we would no longer need to
group it under a set of legal norms, because then, we would already have done so. The legal
effect of a unilateral act is only established by a legal norm in the first place. In order to
determine an acts effect it consequently has to be subsumed under a legal norm stipulating
that legal fact A will lead to legal consequence B. The norms might dictate certain formal
requirements, but in the realm of unilateral acts they do not and instead merely relate to the
openly manifested will of a state. In doing so, they declare a unilateral acts content to be of
primary importance, ie the substantive will manifested by a state. By identifying typical
manifestations of will, such as the intention to undertake a new legal obligation to behave
in a certain manner, to recognise or refrain from recognising a certain situation as legally
valid or to renounce a right, we can hence define typical unilateral acts, ie promise, recognition, protest and waiver.
35
Again see n 30 above.
36
As, for example, the acts of offer and acceptance are. On this aspect see the discussion
on autonomy and lex specialis at pp 5578.
37
Venturini, Actes unilatraux, n 30 above, 39293, for example, excludes protest, and
adds declarations terminating a treaty, at 41921, as well as declarations of war, at 42327, to
his discussion of typical manifestations of will.
38
See esp Suy, Actes unilatraux, n 11 above, 43 and Jacqu, Elments, n 1 above, 336.
39
This is a loose translation of the definition provided by Dionisio Anzilotti, Cours de droit
international (translation from the Italian by G Gidel, Paris, 1929) 346: lacte par lequel un
Etat porte a la connaissance dun ou de plusieurs autres Etats un fait dtermin auquel peuvent se rattacher des consquences juridiques and also used by Suy, Actes unilatraux, n 11
above, 82.
31
32

28 Delimiting the Subject: Promise as a Unilateral Act

expression of will of legal importance.40 All five will be briefly addressed
in the following sections, thereby including only a few remarks on notification, as other authors have already convincingly shown that on a closer
look it does not form a separate category of unilateral acts of states.41
Unlike books or articles dealing with the overall topic of unilateral acts,
the point of this exercise is not to provide a full picture of the normative
legal framework and case law existing for each specific unilateral act. The
following pages, especially from chapter two on, will focus on promises
only. Yet the decision to do so was not made until after the various unilateral acts had been briefly analysed, as this process reveals important
differences between the four acts; differences which effectively hinder
any legal analysis from providing satisfactory answers for core doctrinal
matters when considering all acts en bloc. It is these characteristics which
furthermore explain why answers accepted in relation to one sort of unilateral state action are not necessarily transferable to the other acts, merely
on account of them sharing some common features as unilateral manifestations of a states will. In briefly going through the standard categories
of unilateral acts, it should also become clear what kind of unilateral acts
are outside the scope of this book.
APromise
A picture of the legal framework which has evolved over time for promises as well as its doctrinal underpinnings is discussed at length in chapter three. For now it is primarily the difference between promise and the
other three typical unilateral acts which is of interest and requires a first
glimpse at our actual subject matter. As our further enquiry will show, a
promise is commonly understood to constitute a unilateral manifestation
of a states will by which it undergoes a legally binding commitment to do
or refrain from doing something in the future.42 This commitment is mir40
See in particular Pfluger, Einseitige Rechtsgeschfte, n 11 above, esp 22224 and the references provided by Suy, Actes unilatraux, n 11 above, 100. For more recent publications
retaining notification as a category, see, eg Cassese, International Law, n 30 above, 184 and
Francisco Villagrn Kramer, Les actes uniltraux dans le cadre de la jurisprudence internationale in International Law on the Eve if the Twenty First Century: Views from the International
Law Commission (New York, 1997) 140.
41
See the references as provided in n 75 below.
42
See, eg Cassese, International Law, n 30 above, 185: Promise is the unilateral declaration by which a State undertakes to behave in a certain manner; Charpentier, Engagements
unilatraux, n 30 above, 368, para 4: cest lintention exprime par un Etat (ou plus gnralement par un sujet de droit international) de suivre une ligne de conduit dtermine; Suy
and Angelet, Rechtsgeschfte, n 11 above, 320: Das Versprechen ist eine Willenserklrung,
mit der sich das Vlkerrechtssubjekt zu einem bestimmten Verhalten verpflichtet; Jacqu,
Elments, n 1 above, 336, La promesse unilatrale . . . est lacte par lequel un sujet de droit
international sengage se comporter dune manire dtermine.

Distinguishing Promises from Other Classes of Unilateral Acts29

rored by a subjective right on the side of the addressee, allowing the latter
to claim the performance, or better fulfilment,43 of what has been promised. Whether or not a promise is revocable and, if so, under what kind
of conditions, is open to a dispute in which the full scale of possibilities
has been articulated, ranging from free revocability at the declarants will
to no revocability without the addressees consent.44 The debate thereby
reflects the continuing disagreement of why a unilateral pledge made by
a state should indeed be binding upon it and compel it to act as declared
despite having changed its mind. The legal situation is clearer for waiver,
protest and even recognition, the reasons lying in their specific features.
BRecognition
Recognition is the unilateral act most akin to promises and as mentioned
is sometimes even placed within one over-arching category of unilateral
acts which create legal obligations.45 State practice in this area is extensive
and the question is bound to arise how much alike promises and recognition really are: in other words, can we learn how to treat promises of
states by looking at the history of the unilateral act named recognition?
Its rather frequent use by states along with the legal effects flowing from
it have led recognition to be regarded as one of the most important unilateral acts of states.46 As a learned commentator has remarked, it is however
also:
one of the most difficult subjects to define in international law, since it is governed by no clear-cut customary rules, and legal opinion has been divided over
fundamental issues. Variations in usage from one State to another are due to
the fact that recognition has been used by governments as an instrument of
foreign policy, to indicate approval or disapproval when new situations arise in
international relations . . . In addition, the term recognition may denote a number of very different phenomena, so that it becomes even more complicated to
establish legal criteria.47

Its complexity coupled with its frequency might at least have doubled
the volume of this book, and would have turned it into a very different
one were it to assume recognition to in effect be a special kind of obliging
unilateral act, that is a special kind of promise, which however was not yet
so special that it should be excluded and necessitated a legal framework
of its own. The latter, however, is clearly the case and thus what is said on
As, of course, promises to refrain from certain action are possible.
For more see pp 25176.
45
See especially Suy and Jacqu as cited n 11 above.
46
See, eg Visscher, Problmes dinterprtation, n 13 above, 19091.
47
Jos Maria Ruda, Recognition of States and Governments in M Bedjaoui (ed),
International Law: Achievements and Prospects (1991) 449, paras 12.
43
44

30 Delimiting the Subject: Promise as a Unilateral Act

the pages which follow is not per se applicable to acts of recognition. It is
important to note that the features of recognition which have slowly been
shaped by customary international law are special in a way which turn it
into an act not only warranting the attention it has received but calling for
and explaining different responses found to core questions of legal interest here, such as the roots for the compelling and restraining power on the
declarant and the resulting limits of its revocation. A brief look at the legal
effects of recognition should suffice to make this point: After having highlighted the difficulties encountered in pinpointing the actual meaning of
recognition, Jos Ruda continues to define it as:
a unilateral act by which a state acknowledges the existence of certain facts,
which may affect its rights, obligations or political interests, and by which it
expressly states or implicitly admits that these facts will count as determining
factors when future legal relations are established, on the lines laid down by the
same act. In essence, recognition is the acceptance by a state of a new state of
affairs, which may have legal consequences.48

Recognition thereby is composed of two intertwined but distinguishable elements which in combination determine its legal effect. Through
the act of recognition a state makes reference (1) to a factual situation which
it (2) accepts as existent with all the legal repercussions flowing therefrom.
The possible objects of recognition are diverse and might range from
legal entitlements or statuses, via border lines and governments as a
states (effective) representative, to whole states being accepted as such.
Whether recognition actually does its bit in creating its very own object
and has constitutive effect or is of a mere declaratory nature has long been
debated in relation to the recognition of states. In its delimitation from a
promise it is surely helpful but not absolutely mandatory to take a side in
this discussion,49 while more important to note that even those arguing
in favour of the constitutive theory do not consider the act of recognition
to actually create the physical set of facts to which the act of recognition
refers. A state that is being recognised will still have to comply with certain features in the real world to warrant its name, but will, in the opinion
of the constitutional theory, only receive international legal personality
Ibid 449, para 3.
As is often pointed out, the declarative theory is more consistent as a legal theory, even
though the political impact and practical importance of a states recognition by its peers cannot be disregarded. For a short critical analysis of the constitutional theory see PK Menon,
The Law of Recognition in International Law: Basic Principles (Lewiston, 1994) 1218. The position of, inter alia, Charles de Visscher, stressing recognitions double nature, seems accurate:
Recognition is declaratory in the sense that it acknowledges the validity of a situation or
a claim. It is constitutive in so far as it puts an end to a politically uncertain state of affairs,
replacing the latter by a clear-cut situation in law, Visscher, Problmes dinterprtation, n 13
above, 193, English translation taken from Ruda, Recognition, n 47 above, 455. See also Suy,
Actes unilatraux, n 11 above, 206.
48
49

Distinguishing Promises from Other Classes of Unilateral Acts31

through recognition.50 In other words, not only from the fairly dominant
declaratory point of view but even when taking a constitutive stance on
recognition, the act makes reference to a factual situation whose existence
is not conditioned upon the will of the declarant (be it the state as an international person for those following the declaratory view or merely factors
such as government, territory and effective control, an ensemble which
only through additional recognition achieves legal personality for advocates of the constitutive theory).
Recognition hence establishes a nexus between a manifestation of will
and a set of facts by making the situation accepted opposable to the state,
hindering it thenceforth from effectively negating the existence of whatever it chose to recognise.51 Unlike promises, whose effects of creating a
new obligation are stipulated in reference to the will manifested, recognition has legal features which are explainable only by the connection
between the will manifested and the external and not will-related set of
facts to which it refers:
First, legal doctrine quite regularly assumes recognition to have retroactive effect dating back to the moment when, for example, a state started
going about its business, ie fulfilled the traditional elements, or a government was effectively established.52 While this aspect has rightly been
50
Whether or not the traditional criteria are relied upon in defining the physical existence
of a state, or reference is simply made to states without any further description within the
constitutive theory, is less important. As Lauterpacht recalls in his analysis, Jellineck, despite
providing the first modern design of the constitutive theory, considered a part of organised
humanity to enter ipso facto into the general community of states, while emphasising that
recognition was necessary in order to become part of the judicial community of states. See
Hersh Lauterpacht, Recognition in International Law (New York, 1978, reprint of Cambridge,
1947 edn) 38. Lauterpacht, himself advocating the constitutive theory in connection with a
duty to recognise a factually existing state, is very clear in this regard: A state may exist as
a physical fact. But it is a physical fact which is of no relevance for the commencement of
particular international rights and duties until by recognition and by nothing else it is
lifted into the sphere of law, until by recognition it has become a juridical fact. Recognition
declares the existence of a physical, not of a legal, phenomenon. This explains why, although
declaratory of a fact, it is at the same time constitutive of legal consequences, Lauterpacht,
Recognition, 75, s 32, Recognition as Declaratory of Facts and as Constitutive of Rights.
51
It is sometimes said that a state is estopped from doing so or even that recognition produces estoppel, see, eg Suy, Actes unilatraux, n 30 above, 208, citing and relying on Georg
Schwarzenberger, The Fundamental Principles of International Law (1955) (I) 87 Recueil des
Cours de lAcadmie de Droit International de La Haye 253. Yet, while it is correct that as with
estoppel a state is not heard when it makes a claim contrary to what it has recognised before,
it would, at least nowadays and after some ICJ jurisprudence, be misleading to speak of
estoppel in this context, as the latter requires detrimental reliance on the part of the addressee
in order to have its legal effect. For more on this see p 279 et seq. But no such requirement
exists for acts of recognition in order to have their estoppal effect.
52
See, eg Lauterpacht, Recognition, n 50 above, 5960; Ti-Chiang Chen, The International
Law of Recognition with special reference to Practice in Great Britain and the United States (London,
1951) 13386; Menon, Recognition, n 49 above, 208; Clive Parry, John P Grant and J Craig
Barker, Parry and Grant Encyclopaedic Dictionary of International Law, 2nd edn (Dobbs Ferry,
NY, 2004) 424, under recognition, retroactivity of, with further references.

32 Delimiting the Subject: Promise as a Unilateral Act

identified as problematic for both theories,53 it is accepted by both and
clearly relates back to recognitions characteristic of establishing a link
between an existing fact and a legally relevant will.
Also, and importantly, (de jure) recognition is usually considered nonrevocable, at least unless the situation on the ground, ie the factual situation recognised, has changed.54 Indeed, non-revocability in the area of
recognition based on a change of politics, that is a change of heart, is
easier to justify than it is for unilateral promises, as a state trying to revoke
an act of recognition in these scenarios would suddenly argue that whatever it admitted to exist has now ceased to do so, even though the (physical) fact recognised has not been altered.55 To allow for such a clear venire
contra factum proprium would strongly destabilise international relations
and the law decides against doing so. The situation with regard to promises is different, as will be seen, and it is so primarily for two reasons. Not
only are promises, unlike acts of recognition, not backed by a customary rule negating their revocability, but also, the message conveyed by
a promise is different than that communicated by an act of recognition.
While in the latter case reference is made to a fact or facts, the existence of
which might have been under dispute up until now, but whose continuing
existence can be tested whether a state continues to claim their existence
or not, a promise lacks this external point of reference and is visibly based
only on a states free choice to commit itself. A state which has yesterday
claimed to want to fulfil obligation X in the future does not per se act contradictorily if it now says that it does not any longer want to do so and has
changed its mind. Whereas recognition therefore might need a change of
external facts in order to be revoked, promises which are in theory a manifestation of a states will to newly commit itself, will be more easily susceptible to change if the internal decision alone has been altered. As will be
seen, this does not mean that promises are freely revocable, as limitations
53
According to Lauterpacht, Recognition, n 50 above, 59, this principle is obviously an
embarrassment of the declaratory view, as a retroactive legal effect had to presuppose a
constitutive legal effect in the first place; contra Chen, Recognition, n 52 above, 186, according
to whom: The very idea that legal effect can be given to acts of previously non-existent entities is fatal to the constitutivist contention, at 186.
54
See, eg Menon, Recognition, n 49 above, 165; Lauterpacht, Recognition, n 50 above, 349;
Heintschel von Heinegg, Einseitige Rechtsakte, n 30 above, 239, MN 14. See also Institut
de Droit International, Resolutions concerning the Recognition of New States and New
Governments (1938) 30 American J International Law, Supplement 185, art 5.
55
This aspect complicates arguing for revocation especially for followers of the declarative theory. For those considering recognition to be not only constitutive but also an act of
total discretion, contra Lauterpacht, it is easier and the questions of revocability would be
very much those raised in regard to promises. Proponents of the latter view will, however,
need to overcome the incoherencies in law caused thereby and face the criticism that the
very idea that the legal personality of a State or the representative capacity of its government
should be dependent on the continued good will of other States is deemed to be derogatory
to the independence and the dignity of the State and inimical to the stability of international
relations, Lauterpacht, Recognition, n 50 above, 349. See also the short and concise critical
evaluation of the constitutive theory provided by Menon, Recognition, n 49 above, 1218.

Distinguishing Promises from Other Classes of Unilateral Acts33

on free revocability will usually arise in order to protect the addressees

reliance. The act of recognition therefore rightly carries a label of its own
and can look back on a history and customary backing only applicable to
it and not to acts of promises. The normative analysis in chapter three is
therefore strictly limited and not, at least not directly, applicable to acts of
recognition just as, vice versa, the answers found to legal challenges in the
realm of recognition cannot simply be transplanted to that of promises.
C Protest
Protest is an act of state,56 that at least the majority of authors57 dealing
with the subject consider to constitute a unilateral act as understood by
doctrine, which would make it a unilateral manifestation of will to which
the law attributes a states (apparently)58 willed legal effect of creating,
altering or extinguishing its legal relations with other states. This definition, however, needs to be altered by proponents of protest as a unilateral act, who will need to add a protests asserted legal effect to the list
and hence include acts preserving legal rights amongst unilateral acts
of states.59 Even so, protest will undoubtedly remain in a category of its
own and much more so than the other three typical state acts of waiver,
recognition and promise, as it is the only form of state action which does
not bind its author in any way.60 It is precisely in order not to be bound
that a state is said to protest and thereby declare its opposition to a certain situation and take any possible interpretation of its conduct as signalling acquiescence off the table. Protest has consequently been labelled the
opposite of recognition61 as through it a state manifests its will not to
recognise whatever it is protesting against.
With protest, the issue of interpreting a states behaviour as expressing
an intention to be bound therefore does not arise. Also, no legal framework for its revocation is necessary as there is no problem when a state
changes its mind; it may freely recognise what it just protested against.
56
It can, of course, as with all other unilateral acts discussed here, be issued not only by
a state but also by any other international legal person. We are, however, concentrating on
states for the reasons briefly addressed below, p 78.
57
See n 30 above.
58
As a states will is deduced from its actions.
59
It is in our eyes contradictory to retain the afore-mentioned description of the effects a
unilateral act has (ie to create, alter or extinguish legal relations), rightly reject notification on
this basis but consider protest to fit the definition, as eg Suy and Angelet, Rechtsgeschfte,
n 11 above, 31921, do. See also n 63 below.
60
See the text accompanied by and following n 18 above.
61
See, eg Skubiszewski, Unilateral Acts, n 11 above, 227, para 33; Heintschel von
Heinegg, Einseitige Rechtsakte, n 30 above, MN 15, actus contrarius; Verdross and Simma,
Universelles Vlkerrecht, n 23 above, MN 667; Visscher, Problmes dinterprtation, n 13 above,
188.

34 Delimiting the Subject: Promise as a Unilateral Act

While this is not to say that protest is incapable of bringing about challenging legal questions, especially when seen in relation to acquiescence,62
these will be structurally very different from those coupled to manifestations of will creating legal commitments, and it is only the latter on which
we can concentrate here.63
DWaiver/Renunciation
Through a waiver or a renunciation (either terminology is commonly
used),64 a state gives up its right or claim which subsequently either ceases
to exist or is transferred to another subject of international law.65 Whether
the latter case really involves not only a transferral but a distinguishable
waiver on the side of the ceding subject, as is sometimes assumed,66 is
62
The legal framework for protests, as such, is however rather clear. Accordingly,
Tomuschat sees no harm resulting from its exclusion from the Guiding Principles, Tomuschat,
Unilateral Acts, n 30 above, 1497.
63
While it is surely a manifestation of will, there is actually a lot to say against the assumption that protest also constitutes a unilateral act in the sense used by legal doctrine. Unlike
waiver, promise and recognition, it is not simply by manifesting its will that a state has the
capability of producing any legal effect. Even where a state intends to preserve its rights
by protesting (cases in which a state merely protests to criticise another states behaviour, eg
a brutal crackdown on a demonstration in a foreign country, have to be discarded from the
outset), a protest is only understood to create its alleged legal effect in situations in which
the law would otherwise consider a state to have signalled its consent by remaining silent.
But even in these scenarios a state, by signalling its non-acceptance, does not alter the legal
relations to other states, it merely forestalls any alteration that, with time, could have been
brought about through its acquiescence. It usually even has to protest repeatedly in order
for it to be heard and, if possible, has to back its own protest by other permissible measures,
including the pursuit of open legal venues, in order for international law to take it seriously
(no mere paper protests). While its preserving function is claimed to be its legal effect, the
law appears to be more concerned with acquiescence as equivalent to tacit recognition by
conduct (a unilateral act) in the above-mentioned scenarios, than it is with protest as a manifestation of will having direct legal effect. It is acquiescence which brings about the alteration of international legal relations and protest will be considered as a legal fact weighing
against it. Its task has hence been described as to rebut the presumption of acquiescence,
Karl, Protest, n 11 above, 1158; and even in the scenarios where it is of legal relevance it is
therefore in this writers opinion primarily of factual importance, that is, it is a manifestation
of will which the law will take into consideration but not a manifestation of will that the
law allows to determine a legal effect by itself. On this important differentiation see also
pp 3840. Future codification attempts should hence focus more on acquiescence as an
expression of a states will (consent/recognition) and take protest merely as an indicator of
its absence. See also Venturini, Actes unilatraux, n 30 above, 393.
64
The French word is renunciation. Whether the difference in terminology is intentionally used in order to indicate subtle differences is doubtful. Trying to distinguish,
Lasa Oppenheim, Robert Jennings and Arthur Watts, Oppenheims International Law, 9th
edn (Harlow, 1992) vol 1, Introduction and Part 1, 1195, s 580, note 1, who nevertheless
acknowledge it being largely a matter of usage and convenience which term is customarily
employed in particular circumstances.
65
See, eg Trebilcock, Waiver, n 12 above, 1328.
66
See Suy, Actes unilatraux, n 11 above, 15557, relying on Cavaglieris distinction between
rinuncia abdicativa and rinuncia translativa.

Distinguishing Promises from Other Classes of Unilateral Acts35

doubtful at best67 but will not be dwelled upon, as a cession necessarily

calls for a ceding and an accepting party and hence a bilateral, contractual
exchange which takes it into the realm of the law of treaties.68 Through
a unilateral waiver, a state, however, alters the legal landscape by extinguishing the right it has renounced;69 once gone, it is no longer in a states
power to revive it and a waiving state is, as a result, considered not to be
able to revoke its waiver once effective.70
The unilateral act of waiver is therefore at least in theory clearly distinguishable from a unilateral promise to no longer make use of a right.
While the latter merely leads to an obligation not to use what still remains
in the possession of the promisor, the former actually extinguishes the
subjective right in question. A waiver therefore, strictly speaking, does not
merely create a legal obligation.71 While it is true that a state waiving a
right will have to respect its own decision and will from then on no longer
be in a position to invoke the right it waived, this situation is not so much
the result of a special obligation created by the waiver, but a logical legal
consequence of it no longer being the holder of the right it waived. It is
henceforth compelled to refrain from invoking this right no more or less
than any other state which is not, be it only not any more, the holder
of the right in question. The reason why a state may not make use of the
right it waived after it has done so is therefore not the fact that through its
waiver it has created a special obligation not to do so, but rather, that it no
longer possesses what it would like to invoke. This is the legal and indeed
binding effect its waiver has had.72
Due to its far-reaching consequences, a waiver according to legal doctrine is never to be presumed, and while waivers are generally described
as not having to be made by an express declaration, a states conduct is
said to have to be an unequivocal manifestation of its will to actually
renounce a right of which it is in possession.73 Hence, in cases of doubt a
See also Venturini, Actes unilatraux, n 30 above, 415.
See Pfluger, Einseitige Rechtsgeschfte, n 11 above, 254; Suy, Actes unilatraux, n 11 above,
15556.
69
See Pfluger, Einseitige Rechtsgeschfte, n 11 above, 271; Suy, Actes unilatraux, n 11 above,
148; Jacqu, Elments, n 1 above, 343.
70
See Pfluger, Einseitige Rechtsgeschfte, n 11 above, 271; Suy, Actes unilatraux, n 11 above,
187; Jean-Didier Sicault, Du caractre obligatoire des engagements unilatraux en droit
international public (1979) 83 Revue Gnrale de Droit International Public 633, 639. A waiver
may, of course, be void or voidable if the will manifested was somehow compromised, eg by
duress, fraud, corruption, etc.
71
Of other opinion Cedeo, First Report, n 3 above, 13, para 63; Cedeo, Fourth Report,
n 18 above, 1819, para 84. Also Tomuschat, Unilateral Acts, n 30 above, 1497.
72
The German speaking lawyer might take recourse to the distinction between
Verpflichtungs- and Verfgungsgeschft, here we are in the presence of the latter. See also
Biscottini as cited in Suy, Actes unilatraux, n 11 above, 186: la renonciation a un effet rel et
non pas obligatoire puisque la manifestation de volont abdicative nentrane pas lobligation
de ne plus faire valoir ses droits, mais lextinction immdiat de ceux-ci. See Introduction, n 16
on the question whether the Guiding Principles are applicable to waivers.
73
See Suy, Actes unilatraux, n 11 above, 15764 with further references.
67
68

36 Delimiting the Subject: Promise as a Unilateral Act

declaration is likely to be read as a promise not to use a right rather than
as a waiver of the right in question.74
Although a waiver is therefore a unilateral manifestation of will that
has important legal effects, it is thus not simply a special form of promise,
that is, an act obliging a state vis--vis an addressee to comport itself in a
certain manner and granting the latter a right it may invoke. It compels
a state not through a special obligation but by extinguishing the right it
renounced and is for that reason considered irrevocable. These features
do not allow us to consider the reasoning, much less the legal framework
which is applicable to it, to be simply transposable to promises, just as the
law on promises does not govern a states waiver.
E A Note on Notification
In conformity with a number of authors who have written on the topic,
including the Special Rapporteur, notification cannot be considered as
constituting a unilateral act in the legal sense described above and consequently no problems in distinguishing it from promises will arise. As there
seems to be a growing consensus on this point,75 the following passage
highlights mainly an aspect within the debate that is usually not clearly
expressed: the legal importance (yet not independent legal significance)
of notification as a states will to convey information to another state.
Notification is the act by which a state makes a certain fact known to its
addressee.76 In order to qualify as a unilateral act as understood by legal
doctrine, the law would have to attach legal significance to the will manifested within the notification by a state and not merely to the knowledge
created thereby. These are two important aspects that need to be held
apart. The knowledge of whatever was notified to the addressee may itself
have significant legal consequences; international law might, for example, require a protest against the notifications content or require a state
to act if some ecological hazard or atrocity was notified to it in order to
prevent its liability. In these cases, however, legal consequences are not
See also Jacqu, Elments, n 1 above, 343.
Arguing against notification as a unilateral act, Suy, Actes unilatraux, n 11 above,
10307; Venturini, Actes unilatraux, n 30 above, 392; Jacqu, Elments, n 1 above, 336. For
authors in favour of notification as a unilateral act see n 40 above.
76
The definitions provided by different scholars vary slightly. The one provided by
Anzilotti (an act by which a State makes a certain fact to which legal consequences can
be attached known to one or several States, my translation) is relied upon by Suy, Actes
unilatraux, n 11 above, 82, who provides further references, including authors that require
a formal act. Oppenheim, Jennings and Watts, International Law, n 64 above, 1193, s 578,
write: By notification states communicate to other states certain facts and events of legal
importance. Quite narrow is the definition used by Cassese, International Law, n 30 above,
184: Notification is the act by which a State makes other States cognizant of a certain action
it has performed.
74
75

Distinguishing Promises from Other Classes of Unilateral Acts37

attached to a states manifested will to notify the recipient, ie notification

as a unilateral act, but directly to the knowledge that was created through
the act of notification.77 This knowledge might have been brought about
through other means a widely published newspaper article or a headline in the news can have the same result. Official notification might, of
course, be important and helpful in trying to prove that a state did have
knowledge of a certain fact. Again however, and independent from the
question whether notification actually shifts the burden of proof to the
notified state,78 it is not the manifestation of a will to notify that will trigger
the legal effect in question but merely the fact that a letter containing certain information has been sent to state X on day Y.
Nevertheless, a states manifested will to inform another state, ie wanting it to take knowledge of a certain fact, may and will often also be of
legal importance, yet not of an independent one. Especially in connection with unilateral acts, the will to notify another state of, for example, a
promise or an act of recognition, forms an integral part of the actual performance of that respective unilateral act itself. In other words, intentional
notification in these scenarios will serve as the manifestation of a states
will to be bound or assume the legal consequences of whatever has been
notified.79 The will to notify the recipient in these scenarios, however,
remains completely intertwined with the will to actually undertake the
unilateral act of which the addressee is informed, which leads us back to
the categories mentioned above. Notification understood as a legally rele
vant manifestation of will consequently does not produce a legal effect by
its own and is hence not a category of unilateral acts proper.
To recapitulate: acts of recognition, waiver and protest can hence be held
apart from promises and indeed, they have to be kept apart. All three differ in important ways. Protests are not binding and do not need to be
revoked; recognition cannot be taken back, especially as it is understood
and relied upon as a states acknowledgement of the existence of a set of
facts in the real world, whereas waivers are held to be irrevocable as they
extinguish the right in question.
Promises are different, and especially the limits on their revocability cannot be inferred from the will manifested, unless the declaration includes an
77
Emphasising the fact that notification conveys information which by itself is of legal
importance and triggers the legal consequences usually attributed to notification as such,
Suy, Actes unilatraux, n 11 above, 106; Venturini, Actes unilatraux, n 30 above, 39394;
Jacqu, Elments, n 1 above, 33536.
78
For Venturini, a formal notification is merely one proof amongst others to prove know
ledge on the part of the addressee and does not give rise to a legal presumption, Venturini,
Actes unilatraux, n 30 above, 394.
79
Were a state, for example, to promise a certain line of conduct via a written declaration,
its will could hardly be assumed to have been manifested if the document, before being
transmitted, was stolen and only thereby brought to the knowledge of the addressee. The
will of making a commitment known to the addressee(s) is a necessary part of any manifestation of a will to undergo a commitment.

38 Delimiting the Subject: Promise as a Unilateral Act

express stipulation in this regard. The question will hence be what kind of
default rule international law imposes on such declarations.80
III A MANIFESTATION OF WILL TO CREATE A LEGAL OBLIGATION

The legal theory discussed in chapter three comes into play only and
this will constitute a crucial point if a state is considered to have manifested
its will to oblige itself legally.81 The obligation furthermore consists of a commitment vis--vis one or several addressees to comport itself in a certain
manner in the future.
The element of a manifested will to undertake a legal commitment first
of all distinguishes legal action from merely politically relevant state
conduct. Drawing the line between the two is often difficult in practice;
nevertheless, the international legal order assumes a clear difference to
exist between the display of political goodwill through statements of
intent, on the one side, and undertakings which will create a legal tie,
on the other. While the former, as any gentlemens agreement, can be
of major importance and build up a strong pressure of adherence, only
the latter are enforceable before legal institutions. Legal obligations are,
despite many critical claims to the contrary, also understood as a tool to
reliably narrow a states future leeway and make its conduct more foreseeable than merely political undertakings would. Not adhering to a legal
commitment might lead to a very general distrust of a states loyalty to
the rule of law, significantly damage its reputation as a reliable partner
and open an avenue for various countermeasures, even in unrelated fields
as permitted by law.
The unilateral and legally relevant manifestation of will to be bound
is not to be confused with unilateral and legally relevant state conduct
which is merely wilful. Through a unilateral act, a state is understood to
trigger exactly the legal consequences it wanted to achieve as the law
determines these in direct reference to the states declared will. The incorporation of an armed group into a states army, the occupation of a territory or the commencement of hostilities, on the other hand, are examples
for unilateral and wilful state actions with effects under international
law. But here, the situation created through such wilful action becomes a
legally relevant fact itself without the law determining the ensuing legal
This question is addressed in detail at p 251 et seq.
This is true notwithstanding the fact that it is in the end not a states will which turns
a declaration into a binding commitment but the addressees (presumed) reliance on a declaration made. Reliance is, however, merely protected and only in these cases will it result
in a legal obligation for the declarant if a state is considered by a reasonable person in the
position of the addressee, ie objectively, to have manifested its will to be legally bound. See
pp 194 and 207 et seq.
80
81

A Manifestation of Will to Create a Legal Obligation39

consequences in direct reference to a state having willed them.82 There

is no manifestation of a will to be bound, much less to be obliged and
the legal effects are not determined by a states manifested will to create, abolish or modify legal ties vis--vis an addressee.83 This distinction
is clear-cut in many cases but it can become quite subtle. A declaration
of war has, for example, been considered to fall outside the category of
unilateral acts as the legal consequences of such a declaration were predetermined by an international legal rule and not defined in relation to a
states will.84 It can, however, hardly make a difference whether a manifestation of will includes a long list of intended consequences or merely a
technical term which in the legal community is understood to encompass
such a list and sets an according legal regime in motion. Being a unilateral manifestation of will to which international law directly attaches the
intended legal consequences vis--vis its addressee, ie creating a state of
war, it cannot be discarded on these grounds.85 It will, however, not figure
prominently on the following pages as the rules pertaining to it are nearly
exclusively those of the lex specialis regime of international humanitarian
law and furthermore it cannot be seen as a declaration intended to (primarily) undertake a legal obligation in form of a claimable commitment,
ie a promise, as of interest here.
State action infringing another states rights is also wilful and often
unilateral. International law, however, again does not define the resulting obligations in the field of state responsibility by referring back to the
acting states demeanour as having displayed its will to undertake any
of the incurring obligations. Quite on the contrary, state action triggering responsibility is typically wilful but the legal consequences are usually very unwelcome and by no means determined on the basis of being
(apparently) willed by the acting state. These acts are therefore not per se
unilateral acts as understood by doctrine, much less promises as defined
here, but it is also wrong to exclude them a priori, as is sometimes done.86
Nothing of the afore-mentioned implies that an act which constitutes a
promise ceases to do so once it is held to infringe international law and
82
As the occupation of virgin territory included not only the wilful act of taking the latter into effective possession but also a manifestation of will to incorporate it into the state,
Pfluger considers it to be a mixed act, Pfluger, Einseitige Rechtsgeschfte, n 11 above, 288.
83
Suy differentiates between faits juridique and actes juridiques, Suy, Actes unilatraux,
n 11 above, 17. So do others, see Jean Charpentier, La reconnaissance internationale et lvolution
du droit des gens (Paris, 1956) 367, para 2; Venturini, Actes unilatraux, n 30 above, 407.
84
Suy, Actes unilatraux, n 11 above, 20: Une manifestation de volont comme la dclaration de guerre, dont les effets sont dtermins davance par la norme et sans rapport avec la
volont, ne sont pas pour nous des actes juridiques au sens propre du mot.
85
For more see Venturini, Actes unilatraux, n 11 above, 42327.
86
The Special Rapporteur wanted to exclude acts triggering state responsibility, first
because the ILC was considering the topic separately, and then on the basis of a lack of
autonomy, see especially Cedeo, Second Report, n 3 above, paras 410. On autonomy see
p 55 et seq.

40 Delimiting the Subject: Promise as a Unilateral Act

to invoke state responsibility.87 As with acts of state responsibility and the
law of treaties, both regimes must be kept apart and may be simultaneously applicable since they regulate different subject matters.
IVUNILATERAL

It is only to unilateral manifestations of a will to undergo a legal commitment to which the doctrine described in the following chapters is applicable. The term unilateral has been traced back to its Latin origin from unus
meaning one and latus standing for side.88 A unilateral act is thereby
clearly distinguishable from a bi- or multilateral act (a treaty) where only
through an agreement,89 sometimes also referred to as a meeting of minds
(ie a collaborative effort of at least two parties and with each manifesting its
will, that is, consenting),90 can the willed legal effect be brought about. In
a treaty situation, be it bilateral or multilateral, a state undertakes a legal
obligation only if the other side agrees to his offer to undertake an obligation, that is, accepts it. Without the addressees supplementary manifestation of will, the obligation offered will hence not be brought about. While
in written and usually heavily negotiated treaty texts (or lengthy back and
forth communication) the final document might more often than not no
longer reflect a structure of offers and acceptances, the obligations stipulated within the text will nevertheless only arise vis--vis another state if
that state consents to the text; on the other side of the same coin, only then
will a consenting state be able to invoke any of the rights granted within
the treaty.91 On an abstract and simplified level, the treaty situation may
therefore be pictured as shown in Figures 1.1 and 1.2.

87
Whether the mere undertaking of an obligation contrary to international law will
already trigger a states responsibility (ie without it having already acted in fulfilment of
the promise made), will need to be analysed on a case by case basis. Such situations are,
however, conceivable as a state, for example, might by promising a certain line of conduct
which would violate its international obligations force another state to react in a way which
is detrimental to the latter.
88
See Jacques Dehaussy, Les actes juridiques unilatraux en droit international public: propos dune thorie restrictive (1965) 92 Journal du droit international 41, 52; Sicault,
Engagements unilatraux, n 70 above, 640.
89
James Brierly made it clear that defining treaties as agreements, as the Vienna
Convention has chosen to do, serves sufficiently to exclude unilateral declarations, James
Brierly, Report by J.L.Brierly, Special Rapporteur for the Law of Treaties, UN Doc A/CN.4/23
(1950) reproduced in (1950) II YB International Law Commission 227, para 20.
90
For international agreements concluded between States in written form and governed
by international law see Vienna Convention, Arts 2(b), (f), (g), 11.
91
For treaties granting rights to third states, see pp 4549.

Unilateral41
state A Manifestation of Will state B
state A Manifestation of Will state B
= Treaty
Figure 1.1 Bilateral treaty

The same structure operates in a setting where more than two parties establish a contractual relationship; here, again, doctrine assumes an
agreement as a consensual bond established between all parties participating in the treaty.

state A

state C

= Treaty

state B

state D

Figure 1.2 Multilateral treaty (here four parties)

The unilateral act, on the other hand, is of legal effect without a second partys wilful action; there is only one manifestation of will to which
the apparently willed legal effect is immediately attached (see Figure 1.3).
state A Manifestation of Will state B (or numerous addressees)
Figure 1.3 Unilateral act

A unilateral act is therefore not to be misunderstood as constituting an

act which has no addressee(s); it is not a manifestation of will uttered by a
state to itself. What makes it unilateral is not the absence of an addressee
but rather the fact that the addressees manifestation of will is not necessary there is no act of acceptance, no expression of consent to the act of
communication received in order for the obligation assumed to become
legally effective. While this aspect seems rather clear and has led to no visible opposition, the term unilateral has nevertheless led to considerable
disagreement not only amongst scholars in general but especially within the
ILC. There are in principle the following two aspects which have spurred
the debate and need to be kept apart, both remaining equally controversial.
A Collective State Action
Although possibly contrary to natural intuition, the translation of unus
latus as coming from one side turned out not to suffice and solve the

42 Delimiting the Subject: Promise as a Unilateral Act

formal92 question that is raised here, asking how many subjects of
international law may be found on the side from which the promise, or,
generally speaking, unilateral act emanates. The Special Rapporteur provided the following answer:
As accepted in most doctrine, a unilateral act should be understood as an act
which is attributable to one or more States and which creates a new legal relationship with a third State which did not participate in its elaboration. More precisely, a unilateral act is an expression of will which is attributable to one or more
subjects of international law, which is intended to produce legal effects and which
does not depend for its effectiveness on any other legal act.93

And:
With regard to form, the doctrine generally considers that what is involved is a
single expression of will on the part of one or more States. Unilateral acts may
accordingly be classified as individual or collective. The fact that there is a single
expression of will does not mean that the subject of law that performs it is also single. To
think otherwise would preclude recognition of the variety of strictly unilateral
acts. The fact that there is a single expression of will means that the author or
authors are placed on the same side of the legal relationship to which the act
gives rise. It also means that the elaboration of the act is attributable to them.94

In this opinion, the manifestation of will creating legal effects therefore does not have to stem from a single subject of international law. Such
a formal requirement is considered superfluous as an act is held to be
unilateral not on account of the number of its authors but on account of
a manifested will becoming legally effective vis--vis a party that is not
involved in its elaboration. In other words: as a result of being allowed to
remain uninvolved and passive while still receiving a right, the act granting the latter is said to be unilateral, despite being declared by numerous
subjects of international law; the two-sidedness which is characteristic of
the treaty sphere, where the meeting of wills between the promisor and
the promisee is necessary in order to give birth to an obligation and the
corresponding right, is lacking. The promisee therefore remained situated on the other side of the unilateral act irrespective of the number of
subjects granting a right. While this view is indeed held by a number of
authors,95 the Special Rapporteur is a bit quick to assume that it is really
92
Suy, Actes unilatraux, n 11 above, 26, refers to this aspect as the lment formel. So do
others, see, eg Jacqu, Elments, n 1 above, 322.
93
Cedeo, First Report, n 3 above, para 133, emphasis added.
94
Ibid para 134, emphasis added.
95
See Dehaussy, Thorie restrictive, n 88 above, 52, who, relying on Eisenmanns courses
in (French) administrative law, was apparently amongst the first to claim la question de
savoir si un acte est plurilatral ou unilatral se ramne celle-ci: les normes (au sens le plus
large de ce terme) qudicte un acte juridique vont-elles rgir exclusivement les rapports
mutuels de ses auteurs ou, au contraire, des rapports juridiques concernant dautres sujets
de droit que son ou que ses auteurs?. Relying on Dehausy but being more precice, Sicault,
Engagements unilatraux, n 70 above, 640: En effet, ce nest pas le nombre des auteurs

Unilateral43

generally accepted that a unilateral act may emanate from more than one
subject of international law. A number of scholars, including some of the
more influential ones who have studied this area in detail, merely agree
to differ, as for them, unilateral acts are only those stemming from a single
subject of international law.96
The ILC put a question mark on the existence of collective acts in 199697
and continued to debate the matter. The report of the 1997 Working Group
addresses the issue as follows:
The fundamental characteristic of unilateral legal acts is, logically, their unilateral nature. They emanate from one single side (from the Latin latus), in other
words, from one or several subjects of international law acting unilaterally
and the participation of another party is not required. This characteristic, which
is to be seen both in the structure and in the object and content of the act, leaves
plurilateral international legal acts, such as treaties, outside the scope of the
study. But it does not exclude so called collective or joint acts, inasmuch as
they are performed by a plurality of States which do not intend to regulate their
mutual relations by this means, but to express, simultaneously or in parallel
fashion, as a unitary block, the same willingness to produce certain legal effects
without any need for the participation of other subjects or parties in the form
of acceptance, reciprocity, etc.98

The Working Group which was convened in 1998 then noted that a
unilateral statement could be made by one or more states jointly or in a
qui constitue le critre de lunilatralit dun acte, mais la situation des auteurs de lacte au
regard des normes poses . . . Les auteurs de la promesse se retrouveront tous du mme cot
de la ligne sparant dans le mcanisme normatif, le droit, de lobligation. And, for unilateral
promises, Jacqu: Pour nous, le caractre unilatral dun acte ne dcoule nullement du fait
quil soit imputable un seul sujet de droit international. Un acte unilatral est un acte htronomateur, cest--dire une manifestation de volont par laquelle un ou plusieurs sujets de
droit crent des normes applicables des tiers. J-P Jacqu, A propos de la promesse unilatrale in Melanges offerts a Paul Reuter (1981) 331. In his earlier book Jacqu seems to have
taken a different stand when he states that Le critre essentiel de lacte unilatral ne rside
pas dans le nombre de volonts qui participent son laboration, mais dans le nombre de
sujets de droit auquel il est imput. Lacte unilatral est un acte qui ne peut tre imput qua
un seul sujet de droit international, Jacqu, Elments, n 1 above, 325.
96
See especially Suy, Actes unilatraux, n 1 above, 28: Dans lacte juridique unilatral par
contre, nous nous trouvons en prsence dune seule manifestation de volont, cest--dire
dune manifestation de volont manant dun seul sujet de droit. Also Suy and Angelet,
Rechtsgeschfte, n 11 above, 319: die Willenserklrung nur eines Vlkerrechtssubjekts.
In the same vein, Venturini, Actes unilatraux, n 30 above, 412; also Nguyen, Pellet and
Daillier, Droit international public, n 30 above, 359, para 235: Par acte unilatral, on doit
entendre lacte imputable un seul sujet du droit international; and Brierly, Report, n 89, 225,
para 6: wholly unilateral engagements, engagements to the creation of which only one international legal person is a party, are not within the scope of the present [ie Law of Treaties]
draft.
97
See the tentative General Outline for Unilateral Acts of States as presented by the ILC,
Report of the Working Group (1996) in (1996) II YB International Law Commission pt 2, Annex
II, Addendum 3, 141, para 4: 1. Definition and Typology: Distinction from (i) Unilateral nonbinding instruments; (iii) [mistakingly skipping (ii)] Treaties (possibility of plurilateral acts
(collective unilateral acts)?).
98
ILC, 1997 Report of the Working Group, n 18 above, para 10.

44 Delimiting the Subject: Promise as a Unilateral Act

concerted manner,99 whilst the Group of 2003 agreed to concentrate on
a statement having been made by one state; it was, however, quick to
add that while being a necessary criterion it was not to be considered an
exhaustive one. Paragraph 6 of the Commentary provided by the Working
Group, as an annex to its own recommendations made the very same year,
addresses the matter again and reads as follows:
The idea that, in principle, a unilateral act emanates from one State would
appear, prima facie, to be obvious. Some members of the Working Group were,
however, of the opinion that unilateral acts may originate with a group of
States, while others considered that they were mixed acts of a unilateral nature
in respect of their addressees and conventional acts in relations of their authors
inter se. That question will have to be discussed in the commentary to the article defining unilateral acts for the purposes of the draft and decided in the body
of the draft.100

The Guiding Principles, however, do not address the matter and the
ILC therefore left it unresolved. In dealing with the question whether
numerous subjects of international law can really be the authors of one
unilateral act, it seems sensible to slice the problem into small portions in
order to clarify what this dispute is about.
There is no disagreement that a single subject of international law
can be the author of a unilateral act, the question is merely whether
it may only and always has to be a single subject of international law.
Consequently, no problem and no disagreement will arise in situations in
which the collective as such has attained its own international legal personality; because if so, then, in a legal sense, we are not any longer in the
presence of a collective act but in that of a single subject of international
law despite the acting subject itself being comprised of a number of
international legal subjects.101 An international organisation with international legal personality may, for example, issue a unilateral act through
its respective organs. In doing so, it is, however, acting as a single subject
of international law, as it is distinct from the members comprising it.102
These acts are therefore not the joint acts in question. The collective
or joint act subject to debate can only be the one in which a number of
international legal subjects act jointly without the collective itself having
acquired legal personality.
This joint action is conceivable in different ways and especially the
following two are to be held apart: the legal subjects (as always we are
going to concentrate on states)103 might conclude a contract amongst each
ILC, 1998 Report of the Working Group, n 3 above, para 11.
ILC, 2003 Report of the Working Group, n 3 above, Annex, 6, Commentaries, para 6.
101
See also Pfluger, Einseitige Rechtsgeschfte, n 11 above, 31.
102
See also Jacqu, Elments, n 1 above, 323.
103
See p 78.
99

100

Unilateral45

other in which certain rights are granted to a third state,104 which did not
take part in the agreement and does not become a party to the contract.
Alternatively, the joint action might not consist of a contract between
the acting states while still constituting what may be termed a combined effort of numerous states, because a commitment is, for example,
declared in one document; a group of state representatives might promise
a certain line of future conduct to the world at large or certain addressees
as the outcome of an international conference without them having concluded a contract inter se. It is especially in these two scenarios in which
the question will arise whether numerous subjects of international law
are issuing a single unilateral act or not. Each scenario will therefore be
addressed separately.
(i) Treaties Providing for Rights for Third States
In the first problematic setting, multiple (at least two) states come together
to conclude a contract which confers a right on a third party.105 No one
denies the purely contractual relationship between the states conferring
the right. However, by emphasising the emergence of a right conferred
vis--vis a third state from the same legal camp,106 the act of granting
the third party a special right might now appear to be a unilateral act, as
shown in Figure 1.4.
state A contractual relationship with state  B (and possibly others)


P: unilateral relationship vis--via
third state which did not take part in
its elaboration?
third state (receiving a right)
Figure 1.4 Scenario 1
Or, indeed, to numerous states or to another subject or subjects of international law.
The first scenario therefore only applies to rights conferred on a third party, and hence
what is alleged to be a promise vis--vis a third state within a contract. The reason is that
the special rules contained in the law of treaties apply only to this scenario. Waivers, acts of
recognition, as well as protests, if included in a treaty and proclaimed in relation to a third
party, will therefore trigger the problems described in scenario 2.
106
For these authors see n 95 above, speaking of un mme camp, Dehaussy, Thorie
restrictive, n 88 above, 52.
104
105

46 Delimiting the Subject: Promise as a Unilateral Act

From the outset, such a construction can only be contemplated for
cases in which the parties do not intend to oblige themselves vis--vis
each other to only subsequently, individually or collectively,107 grant the
third state108 a certain right or merely a certain benefit, but where it is
their intention to confer an actual right on that third party through their
contract, therefore directly obliging themselves vis--vis this third state.
While it may indeed seem sensible to liken these situations to that of an
addressee being the beneficiary of a declaration made by a single state,109
or, generally speaking, single subject of international law, the situation is
different and, importantly, the law construes it in a different manner.
For treaties providing for rights for third states the Vienna Convention,
Article 36(1), provides as follows:
A right arises for a third State from a provision of a treaty if the parties to the
treaty intended the provision to accord that right either to the third State, or to
a group of States to which it belongs, or to all States, and the third State assents
thereto. Its assent shall be presumed so long as the contrary is not indicated,
unless the treaty otherwise provides.110

The wording and solution found by the Vienna Convention are a little
awkward as they reflect the outcome of a debate between two different
schools of thought on the matter, from which they were supposed to form
a compromise. When the ILC was dealing with the law of treaties, divergent views were expressed on the doctrinal construction of interest here,
ie of the legal bond created with the third state. While some considered it
possible for two states to confer a right directly on a third party merely by
expressing their will to do so,111 others were of the opinion that this could
be done only by establishing a collateral contract and hence through an
offer which had to be accepted by the third party.112 Especially in light of
the fact that supporters of a collateral contract did not consider an express
acceptance to be necessary, while those of an automatically created right
allowed for a denunciation of the latter, the disagreement seemed surmountable and in practice likely to produce different results only in very

This would lead directly to our scenario 2 and hence to p 49.

The phrase third state merely means the state not a party to the contract, even though
the contract may of course be concluded amongst more than two states.
109
See Jacqu, Promesse unilatrale, n 95 above, 330.
110
Vienna Convention, Art. 36(1), emphasis added.
111
Through what is then called a stipulation pour autrui.
112
See ILC, Draft Articles on the Law of Treaties with Commentaries in (1966) II YB
International Law Commission 228, paras 37. The differences between the Vienna Convention
Article as adopted and the draft prepared by the ILC is of no relevance in this respect. See
also Paul Reuter, Introduction to the Law of Treaties (London/New York, 1995) 10204, paras
15661; Pierre DArgent, Article 36, Convention de Vienne de 1969 in Olivier Corten and
Pierre Klein (eds), Les conventions de Vienne sur le Droit des Traits, Commentaire article par article (Brussels, 2006) vol II, 1468.
107
108

Unilateral47

exceptional circumstances.113 The Commission consequently set out to

draft an article, which, it said, left the doctrinal foundation open.114
Yet, not only but even more so from todays point of view and after the
Nuclear Tests cases, it cannot be considered really to have done so. The ICJ,
and with it legal doctrine, is quite clear that for unilateral and obliging
commitments vis--vis an addressee, it is the promisors manifestation
of a will to be legally bound which allows it to create a legal obligation,
with no acceptance or reaction necessary on the part of the addressee.115 It
is therefore a single manifestation of will which in the promisor-promisee
relationship creates the legal bond and obligation, hence not a relationship requiring the addressees consent. Turning back to the Vienna
Conventions regulation for contracts providing rights for third states,
one cannot fail to notice that its wording clearly construes the legal relationship in a different manner. While the text adopted does not use the
word acceptance, it stipulates the necessity of the third states assent in
order for it to acquire a right and thereby establishes a relationship which
is based on the beneficiarys consent. This is also stipulated by Vienna
Convention, Article 34, according to which as a General rule regarding
third states: a treaty does not create either obligations or rights for a third
state without its consent.116
As the ILC points out:
Paragraph 1 [of the draft article] lays down that a right may arise for a State
from a provision of a treaty to which it is not a party under two conditions. First,
the parties must intend the provision to accord the right either to the particular
State in question, or to a group of States to which it belongs, or to States generally . . . The second condition is the assent of the beneficiary State.

113
The main practical difference between the two concepts as identified within the ILC
relates to the exact point in time at which the right is conferred upon the third party. Under a
(simple) collateral contract, the right is created at the moment of the third partys acceptance
of the offer made, whereas in the other case the right is conferred at the will of the contracting parties without it having to be accepted by the third state and will hence arise at once
and exist until and unless disclaimed by the beneficiary state, see ILC, Draft Articles on the
Law of Treaties with Commentaries, n 112 above, 22829, para 5.
114
Ibid 229, para 5: the Commission decided to frame the article in a form which, while
meeting the requirements of State practice, would not prejudge the doctrinal basis of the
rule.
115
That it is more the perceived than the actual will which binds the declarant is of less
importance in this context. For more see p 207 et seq.
116
There seems to be some disagreement on whether this Article works as a general rule,
the details of which are elaborated in Arts 3538, as the present writer would assume, or
whether Arts 3538 form an exception to the presumption in Art 34 as, eg Ian Sinclair, The
Vienna Convention on the Law of Treaties, 2nd edn (Manchester, 1984) 101 assumes. Article 35,
however, quite clearly does not include an exception to Art 34 but rather further entrenches
it. The same is hence very likely to be true for Art 36. As here, TO Elias, The Modern Law of
Treaties (Dobbs Ferry, NY, 1974) 67; Theodor Schweisfurth, International Treaties and Third
States (1985) 45 Zeitschrift fr auslndisches ffentliches Recht und Vlkerrecht 653, 664.

48 Delimiting the Subject: Promise as a Unilateral Act

The assent is therefore a condition, which is not the case for obligations
brought about through unilateral acts. Nevertheless, the ILC claims to
have remained on neutral ground when it continues:
The formulation of this condition in the present tense and the State assents
thereto leaves open the question whether juridically the right is created by the
treaty or by the beneficiary States act of acceptance. In one view, as already
explained, the assent of the intended beneficiary, even although it may merely
be implied from the exercise of the right, constitutes an acceptance of an offer
made by the parties; in the other view the assent is only significant as an indication that the right is not disclaimed by the beneficiary.117

The ILC hence pretends those who consider the right to be created automatically and without the need of the third states consent, could read
Vienna Convention, Article 36(1), not to say A right arises for a third State
if [!] . . . the third State assents thereto but A right arises for a third State.
But it may be disclaimed, as well as Article 34 A treaty does not create
. . . rights for a third State without its consent as meaning A treaty creates
a right for a third State. The latter, however, has the right to disclaim it.
But this is not what the wording says. Despite the ILC paying lip service
to the other theory, Paul Reuter is therefore absolutely right when he concludes: In any event, the collateral agreement is the crucial concept in the
mechanism instituted by the Vienna Conventions.118
By choosing to operate with a (rebuttable) presumption of acceptance
on the side of the beneficiary, the text adopted by the Commission has,
however, rather ingeniously reduced the practical differences between the
two debated constructions to an absolute minimum.119 This is true especially as the Vienna Convention chose to regulate another important field
of application in which the two regimes might have led to a difference.
Its Article 37(2) deals with the modification and termination of the right
conferred upon the third state and establishes a default rule which allows
the two states agreeing to grant a right to revoke or modify it at their will
117
ILC, Draft Articles on the Law of Treaties with Commentaries, n 112 above, 229, para 7,
emphasis added.
118
Reuter, Law of Treaties, n 112 above, 104, para 158. See also Philippe Cahier, Le problme des effets des traits lgard des Etats tiers inRecueil des Cours de lAcadmie de Droit
International de La Haye (Leiden, 1974) vol 143, 638: Sur le plan des principes, et malgr le
commentaire de la Commission qui y voyait un compromise, cet article tranche le dbat
entre ceux qui estiment que le droit nat du trait et ceux qui pensent quil nat du consentement, en faveur de ces derniers; Schweisfurth, Treaties and Third States, n 116 above,
66465: in order that such obligations and rights actually arise for third States, these too
must consent, ie accept the obligation or assent to the right. Thus it is not the treaty
itself which brings into existence obligations or rights for third States but rather the collateral
agreement between the third State and the contracting parties. Contra Jimenez de Arechaga,
International Law in the Past Third of a Century (1978) 159 Recueil des Cours de lAcadmie de
Droit International de La Haye 1, 54.
119
As intended by the Commission, see ILC, Draft Articles on the Law of Treaties with
Commentaries, n 112 above, 229, para 6.

Unilateral49

unless it is established that the third states consent for them to do so was
made a requirement:
When a right has arisen for a third State in conformity with article 36, the right
may not be revoked or modified by the parties if it is established that the right
was intended not to be revocable or subject to modification without the consent
of the third State.

The solution adopted thereby deviates from the rules operable in

relation to consensual relationships, ie treaties (where withdrawal and
modification are generally not possible without the contractual parties
assent),120 as well as from those governing unilateral promises.121
In light of the aforesaid, it is therefore correct to conclude that states
conferring a right on a third state are not acting jointly through a unilateral act if they decide to do so in a treaty. The dogmatic construction
adopted in the Vienna Convention is, as seen, a consensual one; although
the term offer is avoided, there is in effect an offer to assent made by
both, or more precisely, each122 state to the third party, while presuming
the latter to accept. The right is hence not conferred via a unilateral act,
ie a single manifestation of will, but on the basis of the latters consent.
Problems which are bound to arise with regard to the second scenario of
joint state action are thereby largely evaded.123
(ii) Joint Declarations Outside Treaties
Here, no contract amongst the jointly acting states exists. One of the most
common scenarios that comes to mind for states making sometimes rather
precise and solemnly proclaimed assurances, are declarations issued after
international conferences and meetings. Does such a declaration, while
assuming an intent to be bound, then constitute a joint unilateral act made
by these states?
120
See especially Vienna Convention, Arts 54, 56 and 59(1)(a) for termination, withdrawal and denunciation with the parties consent and the exceptions in Arts 6062 (material breach, supervening impossibility and fundamental change of circumstances) and Arts
3941 for any modification.
121
The ILC does not specify whether any restrictions on a conferred rights revocability
exist other than an intended irrevocability without the third states consent as mentioned
in the provision. While the concept of good faith should in our eyes impose similar restrictions as in the area of promises, the Commission was unanimous in thinking that until the
beneficiary State had manifested its assent to the grant of the right, the parties should remain
free to revoke or amend the provision without its consent, ILC, Draft Articles on the Law of
Treaties with Commentaries, n 112 above, 22829, para 5. Again, the Commissions emphasis
on consent is clear. The free revocability until the beneficiary State has manifested its assent
is nevertheless doubtful as the Commission subsequently altered the provision from and
the State expressly or implicitly assents thereto to include the above-mentioned presumption of assent. A manifestation of assent is hence no longer necessary.
122
See scenario 2 and the legal analysis provided there.
123
For readers who are not convinced and still prefer the solution by the ILC to allow for
a single yet joint unilateral act to be issued to the beneficiary, scenario 1 will also lead to the
questions now addressed in relation to scenario 2.

50 Delimiting the Subject: Promise as a Unilateral Act

state A state B state C state D

a cting jointly through a declaration but without concluding a

contract

P: unilateral act vis--vis third state(s)?

third state(s) (receiving a right)

Figure 1.5 Scenario 2

Scholars who consider this sort of combined action to form a unilateral act (ie a legally binding manifestation of will) because it comes from
one side must be aware that in saying so, they in fact declare a plurality of states to manifest a single will, or more precisely to have manifested
its own will to be legally bound.124 This necessarily leads to the question
how a collective can be considered to manifest a legally effective will as a
collective, hence become a legal actor that confers a right and undertakes
an obligation, without having acquired (at least partial) legal personality?
Because, to recall, if the latter is not the case and the collective indeed is a
legal person, then, as indicated above, the problem dissolves as we are left
not with numerous subjects issuing a joint or collective act, but with the
standard scenario of one subject of international law manifesting its will
through one unilateral act. If, on the other hand, there is no legal personality for the group, which is to be assumed especially where the gathering or get-together of the acting states is ephemeral or even singular, and
there is no body or any organ acting to express what in these scenarios
is going to be a unanimous decision supported by all acting states, can
it nevertheless be said to issue a unilateral act, ie manifest its will? This
apparent contradiction and how it is to be overcome is scarcely explained
by those supporting this solution.125 In addressing it, it is important to
124
That a unilateral act is a manifestation of a single will is not disputed and is confirmed
by Sicault as well as the Special Rapporteur, despite both arguing in favour of a unilateral
act which stems from numerous subjects of international law. See also the citations provided
in n 125 below.
125
Sicault evades an answer when he says: Sans entrer ici dans une critique dtaille de
lassimilation de lunicit du sujet lunicit de la manifestation de volont, on peut noter
que partir de lide de ct, de champ ou de partie, permet de prendre en compte les actes
collectifs ou collgiaux, Sicault, Engagements unilatraux, n 70 above, 640. The Special
Rapporteur cites him, while merely adding: The fact that the act is a single expression of
will does not mean that the subject of law that performs it is also single. To think otherwise
would preclude recognition of the variety of strictly unilateral acts, Cedeo, First Report,
n 3 above, para 134.

Unilateral51

note that the dilemma presented will only arise in a limited set of cases, to
which we are going to turn in a moment.
Often, however, states making a joint pledge that might resemble a single unilateral act will in fact be giving rise to not one, but multiple parallel
unilateral acts through which each participating state obliges itself vis--vis
the addressee. This will especially be the case where the commitment is
one which can be traced back to each state and could have perfectly been
declared on its own, but which is now proclaimed together with others
and at the same time; such as, for example, a declaration in which each
state commits itself to reduce greenhouse gases by X per cent compared to
what it emitted at the time Y.
In these scenarios the analysis provided by Gian Carlo Venturini in a
footnote on the topic is the most adequate:
Even unilateral acts may emanate following the intervention of multiple subjects (collective acts), but, in this case, the different manifestations of will have each a
distinct judicial value of their own, and are merely extrinsically united, in the way
that they manifest themselves simultaneously in the same act.126

The use of the word act at the end of this citation is unfortunate as
it might be misunderstood. Of course the law does not attach legal relevance to the document as such the unilateral act is not the piece of
paper which might be destroyed without thereby annulling the legal commitment it helped to express but, as pointed out, the will which has
been manifested through it. Venturini is, of course, aware of this fact, as he
correctly points out that the apparently joint act is not one (single) declaration of will, ie one act, but there are numerous manifestations of will and
each of distinct and proper legal importance, only extrinsically united in the
same formal act, ie the same document. It is the will of each of the collectives members, which one by one have manifested their will unilaterally,
despite them using a common channel, such as a single document. This
sort of joint action therefore gives rise to a series of parallel unilateral acts;
there is not one unilateral act with numerous authors, but numerous unilateral
acts, each of which have an author of their own.127
To assume not one but numerous acts to be present at this time not only
adequately reflects the communication taking place but, at the same time,
avoids the above-mentioned doctrinal contradiction with other important
126
Venturini, Actes unilatraux, n 30 above, 412, note 9, my translation, emphasis added,
of the following original: Mme lmanation des actes unilatraux peut avoir lieu la suite
de lintervention de plusieurs sujets (actes collectifs); mais en pareil cas, les diffrentes
dclarations de volont ont, chacune, une valeur juridique propre et distincte, et ne sont
quextrinsquement runies, dans la mesure o elles se manifestent simultanment dans le
mme acte. See also Pfluger, Einseitige Rechtsgeschfte, n 11 above, 32, where he points out
that if states do not act on a common legal basis which allows them to form a single will,
they will merely be able to issue simultaneous manifestations of will.
127
The ILC Working Groups description during the early stage of the work in 1997 also
spoke of states acting simultaneously or in a parallel fashion, see the text to n 98 above.

52 Delimiting the Subject: Promise as a Unilateral Act

implications in law, for example, if only one or some of the unilateral declarations made in a parallel fashion are held to be void or voidable or were
to be revoked. Each act is, and remains, distinct from the others and can
therefore be treated separately. How a truly joint act, understood as one
unilateral act issued by numerous subjects of international law should
be treated, if only one or some of its authors want to revoke it or invoke
fraud, corruption or error, has not been explained by those assuming so,
and will only lead to more dogmatic contradictions and headaches.
Yet, and much to our chagrin, these headaches cannot be completely
avoided by the above-mentioned description of joint action by states.
Problems are bound to arise when numerous states manifest a will which
can only operate when being attributed to more than a single state. This
might be the case where two or more states can only exercise a right
together. The same will be true when a pledge is made in which the declaring states make an assurance which is expressed only as an obligation of
the group itself without any indication as to which member of the group
will fulfil which part of the commitment. To provide an example of this
second aspect: if eight states legally oblige themselves together to reduce
the amount of greenhouse gases or young fish caught by very small
nets by 20 per cent, we do not know how this 20 per cent is going to be
calculated by the acting states. We cannot simply assume that each state
has expressed its will in a parallel fashion to reduce greenhouse gases or
fishing by one-eighth of the promised 20 per cent reduction. Only the collective as such has pledged to do so, and it might itself decide how to
divide the responsibilities to meet its duties. The logical question hence is:
has the collective issued a single unilateral act in this scenario?
As indicated above, those speaking of the same legal camp, who
generally assume collective action to be a possible source of a unilateral
manifestation of will, hardly even address the problem raised here. In criticising Suy for considering merely one subject of international law as the
possible author of a unilateral act, Jacques Dehaussy points to Eisenmanns
courses in (French) administrative law.128 Pfluger also takes recourse to
figures of internal law but is more explicit in tackling the problem when
he refers to a Gesamtakt (collective act) which was indeed a single manifestation of will issued by numerous states without them constituting a
new judicial person. Pfluger considers this admittedly intermediate configuration129 (between own legal personality and a mere plurality of legal
128
Dehaussy, Thorie restrictive, n 88 above, 52. In note 21 he directs the reader towards
divers cours de doctorat de droit administratif approfondi of Mr Eisenmann.
129
See Pfluger, Einseitige Rechtsgeschfte, n 11 above, 32, note 18: Die mehreren zu einer
Handlungseinheit verbundenen Personen stellen vielmehr eine Zwischenerscheinung
zwischen Einzelperson und juristischer Person oder, auf VRgebiet bertragen, zwischen
Einzelstaat und Staatenbund (als VRperson) dar (which roughly translates: The multiple
persons forming a union ready to act in fact constitute an intermediate position between

Unilateral53

persons each acting for themselves) necessary and applicable in the case
in which there is a legal basis unifying the acting states, providing the
example of a protest being issued by two states which together administer a territory.130 While caution is clearly necessary when advocating the
transferral of (notions of even limited applicability in) national law to the
international arena,131 this would not be sufficient reason alone to dismiss
it, as it might indeed turn out that the international legal system poses the
same problem in merely a different language and consequently warrants
a similar solution. However, referring to a Gesamtakt as a manifestation
of will by states that are acting on a joint legal basis, yet without legal
personality, does not by itself solve the legal paradox of having a group
of states as an admittedly non-legal entity acting in a legally obliging
and rights-creating manner; it only gives the act a name and points more
directly to the problem.
Instead of assuming an intermediate position in these cases, which
will lead to many other questions as to how this collective will be treated
if an obligation is breached by it, taken back or void, or if it is indeed
invoked by the addressee, the following alternatives are possible. First,
it is indeed conceivable and only logical to assume that a legal obligation
by the group, as such, simply cannot be assumed, as long as the collective
does not give itself a more stable structure in order to attain international
legal personality as an international organisation, to be able subsequently
to oblige itself as a group, rather than only each of its members. Despite
a will of the declaring states to assume an obligation as a group, we will
then again merely have, at most, parallel obliging declarations in which
states pledge together to fulfil what has been promised, but with each
state individually being no more obliged than to participate somehow in
the fulfilment of the declared goal. Where two states are administering a
territory together we would assume that they, too, both make unilateral
declarations such as a protest, which only in their dependence on another
are of their intended value, but which remain distinct as legal acts. The
other alterative, as we see it, and indeed a preferable one, as it adequately
reflects an intention of the acting states to commit themselves as a group,
is to go light on the requirements on international personality: As the
collective itself is created of subjects of international law which intend to
assume a legally binding obligation as a collective, they apparently want
a single natural person and a legal one, or, and transposed to the international legal level,
between a single State and a federation of States (as a subject of international law)).
130
Ibid 32. See also Heinrich Triepel, Vlkerrecht und Landesrecht (Leipzig, 1899) 5962.
131
Even in the German national legal system Pfluger in his note 22 notes the opposition to
this (at the time) relatively new concept that was apparently introduced in a book written by
Prof Johannes Kunze in 1892 entitled Gesamtakt, ein neuer Rechtsbegriff (Gesamtakt (joint
action) a new terminus in law, my translation). Today the concept of a Gesamtakt exists in
German law, even though in limited areas and with a Gesamtakt not necessarily constituting
a manifestation of will.

54 Delimiting the Subject: Promise as a Unilateral Act

and need this group to be capable of carrying duties and hence to attain, if
only very partial and limited, legal personality. In any event, either solution is in our eyes a more coherent way out of the dogmatic dilemma in
these, moreover, very limited scenarios.
(iii) Concluding Remarks on Joint State Action
Our reply to the question as to whether promises or, more generally
speaking, unilateral acts will necessarily emanate from a single subject of
international law is hence affirmative. It can be summarised in Figure 1.6.
It is consequently correct to say that a unilateral act, and for our purposes a promise, will emanate only from a single subject of international law.
The possible exception is, as indicated, only conceivable in special scenarios

States 'acting jointly'

with the group having

attained legal personality
itself (eg as an IO):
act as a single subject of
international law and issue
a single unilateral act.

without the group

having attained legal
personality

and which confer a right via a treaty:

do not grant the right unilaterally but
via a consensual link in which the
beneficiarys assent is presumed and
with the VCLT stipulating special
regulations as to the rights
revocability and modifiability.
while the manifested will is
attributable only to the group:

and not concluding a

treaty inter se in which
a right is granted

clearly

An 'intermediate position' is contemplated

(no legal personality yet the group manifests
a legally binding will as such) but it is
dogmatically more coherent to either assume
no unilateral commitment for the group, but
merely parallel unilateral commitments by
each State to act in order to achieve the goal
OR , and preferably, to allow the group to
attain at least partial legal personality (and
again have a single subject of international
law which issues a single unilateral act).

Figure 1.6 Summary of joint state action

and the manifested will

is attributable to each
'member': each issue a
single unilateral act
with the group hence
issuing parallel
unilateral acts.

Unilateral55

and even here, it should be rejected as it remains contradictory in its legal

construction. In these cases the group should be attributed with at least
partial legal personality. Those claiming that a single unilateral act may
stem from numerous legal subjects should be aware of the fact of thereby
defending a legally paradoxical construction, an awareness that seems to
be largely lacking in view of the scarce explanations and naked stipulations
in this regard. The fact that unilateral acts will emanate from only a single
subject of international law is, however, not due to an existing additional
formal requirement, but merely the effect of the legal regime attaching consequences to a legal subjects own and only, hence, single manifestation of
will.132
B The Autonomy Debate
Not all acts matching the above described criteria, ie being manifestations
of a legally relevant will to undergo a new obligation and issued by a single subject of international law, are, however, also considered to actually
fall within the legal framework operable for unilateral acts (and hence that
of promises as their subcategory). In order to indicate which ones are and
which ones are not, the adjective unilateral is sometimes substituted by
or allied with the word autonomous. Although disagreement as to what
this term really means is considerable,133 the division into autonomous and
non-autonomous, that is dependent unilateral acts, has been employed
in order to hold acts related to the law of treaties or other legal regimes
apart from the truly unilateral ones, only to which the rules on unilateral
acts will apply.134 In this view, offers, acceptances or acts of accession, to
name only a few, are considered to be unilateral only in a formal way, ie
Despite the assertions by Sicault and Cedeo to the contrary, see n 125 above.
See especially the debate that arose within the Commission in ILC, Summary Record of
the 2603rd Meeting, UN Doc A/CN.4/SR.2603 (1999).
134
Cedeo, Third Report, n 3 above, para 61: The point is to exclude, by means of this
criterion, acts linked to other regimes, such as all acts linked to treaty law. For the autonomy
concept presented by the Special Rapporteur see Cedeo, ILC, 2603rd meeting, n 133 above,
paras 26 and 62; Cedeo, First Report, n 3 above, para 136 et seq; Victor Rodriguez Cedeo,
Second Report on Unilateral Acts of States, Addendum, UN Doc A/CN/500/Add.1 (1999)
para 46, esp paras 62 and 63; and Cedeo, Third Report, n 3 above, paras 6069. See also
Suy, Actes unilatraux, n 11 above, 30 as well as Venturini, Actes unilatraux, n 30 above,
400, 413 and Sicault, Engagements unilatraux, n 70 above, 64041. Pfluger, Einseitige
Rechtsgeschfte, n 11 above, 6566 mentions the possibility to distinguish between dependent
and non-dependent acts (selbstndige and unselbstndige Rechtsgeschfte) without, however,
wanting to discard dependent acts from the realm of unilateral acts. Verdross and Simma,
Universelles Vlkerrecht, n 23 above, 42627, ss 66465 also merely distinguish between the
two forms. Cautiously also Skubiszewski, Unilateral Acts, n 11 above, 222, para 9, who
does not deny a non-autonomous acts legal effect but only describes, for example, ratifications essential function to go beyond unilateralism, while other acts were by definition,
free of any link or relation to a legal transaction in which another state or states participate,
and that some writers regard only the latter as true unilateral acts.
132
133

56 Delimiting the Subject: Promise as a Unilateral Act

to be manifestations of will issued by a single subject of international law
and intended to create legal effects. But because they allegedly depend on
another manifestation of will to actually have their legal effect, they are
said not to create the latter in an autonomous way. According to some
authors this is also true for declarations made under ICJ Statute, Article
36(2), and acts such as the delimitation of maritime boundaries or, more
generally speaking, all acts which are based on a treaty provision.135 In all
these cases, so they claim, it is not on account of the sole will of the acts
author that a legal effect is triggered and accordingly, the act must be classified as not unilateral in substance.
Yet, and even though there is general agreement that acts such as the
ratification of a treaty are not covered by the legal regime dealing with
unilateral acts but by the specific rules laid down in the law of treaties,
the ILC did not reach a consensus as to the concept of autonomy much
less could it agree to include it in a definition of unilateral acts. While the
Special Rapporteur and some Commission members held it to be of fundamental significance,136 others considered autonomy to be an ambiguous term,137 to have nothing to do with the definition of a unilateral
act,138 found the categorisation into autonomous and non-autonomous
acts of secondary importance,139 its application as advocated within the
Commission even absurd,140 and wanted the term to be eliminated.141
In the various definitions debated within the Commission,142 the adjective appears in the ones proposed in the Special Rapporteurs First
and Second Reports, before being put into square brackets by the 1999
Working Group and finally fading away completely. It does not appear
in the Guiding Principles now adopted. The debate on autonomy was,
however, never settled and despite it being eventually discarded from the
definitions later debated within the Commission, the Special Rapporteur
135
See especially Cedeo, n 134 above and Suy, Actes unilatraux, n 11 above, 31, 142 et seq.
Cedeo, especially in his Second Report, went even further and argued not only for an independence of treaty provisions but even pre-existing customary norms, see Cedeo, Second
Report, n 3 above, para 62 and the analysis provided p 62 et seq.
136
Cedeo, ILC, 2603rd meeting, n 133 above, para 26; for the Special Rapporteurs view
see in addition the references provided in n 134 above. Also in favour of the concept of
autonomy in the Commission, see, eg Lukashuk, ILC, 2603rd meeting, n 133 above, para 18;
Economides, ILC, 2603rd meeting, n 133 above, para 24.
137
Mentioning the ambiguity of the word and the different understanding attributed to
the concept by Commission members, Simma, ILC, 2603rd meeting, n 133 above, para 25.
For Pellet no one knew what autonomy meant, ILC, 2603rd meeting, n 133 above, para
35. See also Nguyen, Pellet and Daillier, Droit international public, n 30 above, 362, criticising
the lack of precision attached to the autonomy criterion, and the missing accord amongst
those supporting it, on which acts were autonomous.
138
Candioti, ILC, 2603rd meeting, n 133 above, para 28.
139
Addo, ILC, 2603rd meeting, n 133 above, para 32.
140
Pellet, ILC, 2603rd meeting, n 133 above, para 19.
141
In the ILCs 2603rd meeting, n 133 above, Pellet, Candioti, Ado, Simma, Duggard and
Melescanu all argued in favour of eliminating autonomy from the definition.
142
See n 3 above.

Unilateral57

in his work (in the eyes of one Commission member seemingly obsessed
with autonomy)143 held on to it as an underlying material concept.144
And indeed, the distinction into autonomous and non-autonomous or
dependent and non-dependent acts can frequently be found in legal literature.145 In order to clarify the importance of the autonomy concept, or
the lack thereof, we would like to start with a rather influential article in
the area of unilateral acts which was written by Fitzmaurice in 1957.146 For
him:
Declarations may be of three kinds:
(i) bilateral or multilateral Declarations, which are unilateral neither in substance nor inform;
(ii) unilateral Declarations, that are unilateral both in form and in substance;
and
(iii) unilateral Declarations that are unilateral in form but not in substance.147

The first group of declarations, clearly bilateral or multilateral ones like

a treaty text developed and adopted mutually by two or more parties, are
of no interest here as they from the outset lack any unilateral element. It
is the distinction between category (ii), the unilateral declaration which is
unilateral both in form and in substance (hence truly unilateral), and the
unilateral declaration under (iii) which is said to be unilateral merely in
form but not in substance which matters here. Fitzmaurice differentiated
between these two as follows:
In case (ii) the Declaration is unilateral not only in form but also in substance, in
the sense that it is not made in return for, or simultaneously with, any specific
quid pro quo or as part of any general understanding . . .
In case (iii), on the other hand, the contractual element is present. The
Declaration is unilateral in form, but it is contractual in substance, either
because it is one of two or more similar Declarations intended to be interdependent and interlocking, or because it is linked to the action of another State,
which either forms the quid pro quo for it, or in respect of which it is itself the
quid pro quo. Such a situation gives rise to a treaty position in which the text
or texts concerned will clearly fall to be interpreted according to the normal
rules of treaty interpretation.148
Pellet, ILC, 2603rd meeting, n 133 above, para 12.
See Cedeo, Third Report, n 3 above, para 69: Although the term autonomous is not
included in the definition submitted in this Third Report, it can be assumed that these acts
are independent in the sense mentioned above, although this issue will have to be discussed
further in the Commission so as to define and delimit such acts.
145
See the references in n 134 above.
146
It seems to have been especially influential in establishing the very existence of unilateral promises, see p 125; see also Alfred P Rubin, The International Legal Effects of
Unilateral Declarations (1977) 71 American J International Law 1, 11.
147
Sir Gerald Fitzmaurice, The Law and Procedure of the International Court of Justice
195154: Treaty Interpretation and Other Treaty Points (1957) 33 British YB International Law
202, 229.
148
Ibid 23031.
143
144

58 Delimiting the Subject: Promise as a Unilateral Act

For Fitzmaurice there are therefore unilateral acts which are manifestations of will emanating from a single subject of law but which are not
unilateral in their substance as they are either intended to form a bilateral
relationship or are linked to other state action which forms, or to which
they themselves form, the quid pro quo.
This either-or distinction (intent or quid pro quo) is, however, not
entirely convincing, as the existence of a quid pro quo can only be established by taking recourse to the parties intentions in the first place. Where
an exchange of some sort has taken place it will be necessary to establish
that both parties indeed intended to create a nexus between the two acts
and wanted the latter to be understood as a contractual do ut des exchange.
Without this inquiry there is no way to differentiate between two separate
contracts each favourable to only one party, two unilateral declarations
(eg the subsequent exchange of promises) and, indeed, a single contract.
An exchange will only turn into a one-for-the-other exchange, ie include
a quid pro quo, if the parties intention to do so has been established. In
this regard, it is important to keep in mind that the common law requirement of consideration has not been incorporated into international law.149
In order to determine whether an interdependency exists, the focus will
therefore always lie on whether the declarants intended their declarations
or actions to be interdependent and interlocking, as Fitzmaurice calls it.
Yet, the question remains whether such an act intended to be inter
locking or interdependent, ie the offer or acceptance, can really be called
not unilateral in substance. Does the will to do something under the
condition that somebody else accepts this declaration really turn the
declaration into a mixed act, unilateral only in form but actually bi- (or
multi-) lateral in substance?
When Eric Suy took up this aspect, now under the actual term of
autonomy (lautonomie), in his important work on unilateral acts (which
apparently strongly influenced the Special Rapporteur, who advocated
an autonomy concept along similar lines within the ILC), he differentiated between two forms of dependency. The substantial question to ascertain an acts unilateral substance was whether a manifestation of will can
be said to become effective independent of other manifestations of will which
149
See, eg Brierly, Report, n 89, 225, para 10: It is not thought that the doctrine of consideration plays any part in international law. Also, ICJ, South West Africa (Ethiopia and Liberia v
South Africa), Separate Opinion of Judge Jessup, [1962] ICJ Rep 387 403: The doctrine of consideration, which plays so large a part in Anglo-American contract law, has not been taken
over into the international law of treaties. See also Hersch Lauterpacht, Private Law Sources
and Analogies of International Law (with special reference to International Arbitration) (London/
New York, 1927) 17778; Oppenheim, Jennings and Watts, International Law, n 64 above,
1211, s 588: The object of treaties is always one or more obligations, either affecting all the
parties, or only some or even only one of them. On consideration and especially the ICJs
treatment of Icelands failure of consideration argument raised in ICJ, Fisheries Jurisdiction
(Spain v Canada), Jurisdiction, Judgment of 4 December 1998, [1998] ICJ Rep 432, see also Jan
Klabbers, The Concept of Treaty in International Law (The Hague/Boston/London, 1996) 8687.

Unilateral59

emanated from other subjects of international law.150 The criterion of legal

effectiveness led him, and the Special Rapporteur, to distinguish between
two forms of dependency that rendered an act non-autonomous. First, an
act could be dependent on another unilateral act to have its legal effect. And
secondly, an act could also be dependent on another bi- or multilateral act to
become effective.151 In both cases the act is said to be not unilateral in substance and hence to fall out of the legal framework operable in relation to
unilateral acts. Each aspect is to be considered separately.
(i) Independence of Another Unilateral Act?
The first prong of the purported independency test stipulates that an act
has to be legally effective independent of another unilateral act, ie manifestation of will. With regard to this aspect, Suy wrote the following (my
translation provided in footnote):
Il se peut tout dabord quun acte unilatral doive rencontrer un autre acte de
la mme qualit afin de pouvoir produire des effets. Pris sparment, les deux
actes nont aucun effet juridique immdiat car lordre juridique objectif ne rattache au premier que des effets juridiques potentiels, cest--dire sous condition
que lacte pos rencontre un acte correspondant destin le complter. Comme
exemple typique de ce phnomne, nous pourrions citer loffre et lacceptation.
Il nexiste en droit international aucune norme stipulant lobligation de maintenir loffre pendant un certain temps moins de dispositions expresses dans
loffre elle-mme. Dautre part, il ne peut pas y avoir dacceptation dune offre
inexistante. Cest--dire quaucun effet nest rattach directement loffre, mais
une rencontre et une concordance entre loffre et lacceptation, ce qui nous
amne dans le sillage du droit conventionnel. Sans doute, chacune des manifestations de volont mane dun seul sujet de droit international; mais elles ne
constituent pas des actes unilatraux parce quelles nont pas dimportance en
elles-mmes, et les effets juridiques sont rattachs non pas chaque manifestation de volont, mais plutt lacte rsultant de lensemble de volont. Cet acte
constitue prcisment laccord.152
150
Suy, Actes unilatraux, n 11 above, 30: Il faut, en outre [besides the formal element]
que la manifestation de volont soit, quant son efficacit, indpendant dautres manifestations
de volont manant dautres sujets de droit, emphasis in the original.
151
See ibid para 30 et seq. Also Cedeo, ILC, Summary Record of the 2593rd Meeting, UN
Doc A/CN.4/SR.2593 (1999) para 12: A unilateral act thus existed when it was formally unilateral, when it did not depend on a pre-existing act (first form of autonomy) and when the
obligation was independent of its acceptance by another State (second form of autonomy).
152
Suy, Actes unilatraux, n 11 above, 30, which may be translated as: First, it is possible
that a formally unilateral act must encounter another act of the same quality in order to be
capable to produce effects. Taken separately, the two acts have no immediate legal effect as
the objective legal order merely attaches potential legal effects to the first act, this is to say
only under the condition that the act encounters a corresponding act designated to complete
it. As typical example of this phenomenon we can cite offer and acceptance. There is no
norm in international law which stipulated an obligation to hold an offer for a certain time
if not for dispositions made within the offer itself. And on the other side there cannot be an
acceptance of an non-existent offer. This is to say that no direct legal effect is attached to an

60 Delimiting the Subject: Promise as a Unilateral Act

For Suy, offer and acceptance are paramount examples of acts which
do not fulfil the independency requirement, as their legal effect is said to
depend on another manifestation of will. Consequently, they are found to
be not unilateral in substance. Interestingly, Suy does not actually claim
an offer to have no legal effect by itself but reduces its effect to being not
immediate and merely potential.153 Yet, while it is possible to describe
unilateral acts as being strictly unilateral or not, in a descriptive way
(some invite another manifestation of will and can form part of a bilateral
process while some do not), a distinction amongst unilateral acts using
autonomy as a legal and material criterion in order to hold different legal
regimes apart will not achieve the desired result. In fact, even an offer, as
the paramount example of a dependent act requiring another manifestation of will, is an autonomous unilateral act if looked at in more detail. It
does not require another manifestation of will in order to have its legal
effect. The legal effect of an offer is, of course, not to oblige the offeror to
actually fulfil what he has offered, this will be the treatys effect which
requires both, offer and acceptance. Yet, an offer has a legal effect, and it
will have this effect the moment it is made.
Looking at the quotation provided above, Suys characterisation of an
offer to be of no (immediate) effect and on its own to be of no import
ance apparently rests on an offers following two characteristics. First,
an offeror by making use of it does not become obliged to follow through
with what he has offered to do (or offered to refrain from doing) unless the
offer is accepted; and secondly, an offer is freely revocable until accepted.
While both aspects correctly describe the nature of an offer, this does not
mean that an offer has no direct legal effect of its own. To be sure, an offer
does not commit the offeror to undertake what has been offered before any
acceptance by the addressee, but only because this is the very condition
laid out or inherent in any offer. The offeror promises to do whatever he
offered only if the offer is accepted. Hence, if an offer is a true unilateral
act, ie a manifestation of will to which the apparently willed legal consequences are attached, the law would impose these exact terms onto the
offeror and it would do so from the very moment the offer has been made
vis--vis the addressee without waiting for any acceptance in order to do
so. And indeed, this is precisely what international law does. Through the
offer, but to a meeting of and a concordance between offer and acceptance, which takes us
into the realm of the law of conventions. Without a doubt, each manifestation of will emanates from a single subject of international law; but they do not constitute unilateral acts
because they do not have a legal importance of their own, and the legal effects are attached
not to the respective manifestation of will but rather to the act resulting from the ensemble of
different manifested wills. This act precisely constitutes the accord (my translation).
153
For Venturini, offer and acceptance for themselves indeed have: aucune effet juridique,
Venturini, Actes unilatraux, n 30 above, 413. See also Coumoul as cited in Pfluger, Einseitige
Rechtsgeschfte, n 11 above, 106, note 4, along with Pflugers critical analysis, in which this
position is rejected, ibid 105 et seq.

Unilateral61

offer, the declarant has established a legal condition that, if triggered, will
lead to a future and further obligation. This unilaterally installed automatism along the lines of accept, and I will be obliged to do X is set up by
the offer and already significantly limits the offerors freedom of action.
The self-imposed limitation becomes most apparent when the offeror
has changed his mind, as he will find himself in the very unwelcome situation (the legal effect created by the offer) that a third state, by its manifestation of will (an acceptance), is able to create a new obligation for the
offeror, even though the latter is not (any longer) consenting.154 In order
to (again) alter the legal situation, the offer needs to be revoked before
the addressee accepts it. This revocation is hardly made in order to cancel
a mere potential effect; just as the sword of Damocles poses an immediate threat, so did the offer immediately create a now very unwelcome
legal situation. It is precisely in order to cut down the sword and cancel
this very direct effect, that the offer is revoked. The scenario is not much
different than it is whenever a commitment is made which includes a
condition. To describe the effect as merely potential is inaccurate as the
condition is legally active from the very minute the offer is made.
This effect is by no means negated if, with Suy, offers are held to be
freely revocable before being accepted in international law,155 as it remains
the reason why an offer needs to be revoked in the first place. Before and
until it is revoked, it will have its effect of signalising consent to be bound
according to the terms stipulated within it. To allow a state to take back
its legal commitment and to require it to do so, does not mean it never
existed.
Indeed, an act of acceptance also has independent legal effect, although
it is much harder to analyse, since it is necessarily the second step in a twostep procedure and makes direct reference in determining its legal effect
to the previously made offer. But as Pfluger has pointed out,156 the fact
that the legal order allows error, fraud and other vices to remain invokable by a party to invalidate its consent (see Vienna Convention, Articles
48 to 52) shows their continuing legal importance in obliging their declarants. They are not a legal zero which being added to another one will in
sum rather miraculously give rise to a contract, but compelling unilateral
acts which together give rise to the full contractual commitment.157
We are thereby not disregarding the fact that offer and acceptance will
together give rise to a contract governed by a legal regime of its own, as
stipulated in the Vienna Convention. Of course, a treaty is no unilateral
154
Usually, states are not able to create an obligation for other states against their will; see
also ILC, Guiding Principles with Commentaries, n 4 above, Guiding Principle 9, also briefly
addressed p 187.
155
Article 16 of the United Nations Convention on Contracts for the International Sale of
Goods includes such a rule for offers in its field of application.
156
Pfluger, Einseitige Rechtsgeschfte, n 11 above, 107.
157
See also Pfluger, ibid 105 et seq.

62 Delimiting the Subject: Promise as a Unilateral Act

act. However, an offer is, as it constitutes a unilateral manifestation of
will which has precisely and immediately the legal effect as willed by its
author.
What is true for the paramount example of an allegedly dependent
act, is similarly true for other acts such as the denouncement of a treaty
on the basis of, for example, Vienna Convention, Article 60. It too is a unilateral act, despite being linked (ie occurring in connection) with treaties,
the only reason for which it was regulated in the rules governing the latter.
Autonomy as a material concept based on an acts effectiveness is therefore
in our eyes not capable of distinguishing between acts covered by various
legal regimes and singling out only those which are truly unilateral.
Interestingly, in a very recent publication, Eric Suy has himself expressed
doubts and begun to wonder if the constituent element of autonomy of
unilateral acts should not be somewhat revised or shaded.158 Indeed, in
this authors opinion it should be given up as a material criterion.159 Doing
so does not, of course, mean that an offer as a unilateral act might not be
covered by special rules, rules that have been shaped for it precisely on
account of some of its peculiarities. These rules are then to be applied on
account of the lex specialis principle and not a lack of autonomy. Secondly,
it clearly remains possible to differentiate between an offer and a promise.
As described, for the present writer both are, however, unilateral acts with
an autonomous legal effect. Yet, only the former asks (if only tacitly) for
an acceptance in order to establish the obligation offered, while the latter
does not do so. This condition, inherent in any offer, thereby clearly distinguishes both acts from each other,160 and as a gateway into the law of treaties has attracted special rules to be applicable to it. But before returning
to this aspect,161 the other source of dependency needs to be addressed.
(ii) Independence of Another Multilateral or Bilateral Act?
In his book, Suy argues for a second prong in the determination of independency and wants to exclude all acts as not autonomous which are
based on a clause included in a treaty, these acts being at their basis nothing but the execution of a bi- or multilateral act.162 Following this understanding, a treaty which lays down a rule according to which a unilateral
158
Eric Suy, Unilateral Acts as a Source of International Law: Some New Thoughts and
Frustrations in Droit du pouvoir, pouvoir du droit: mlanges offerts Jean Salmon (Brussels, 2007)
63435.
159
See also Nguyen, Pellet and Daillier, Droit international public, n 30 above, 362, para 238:
Aucune objection srieuse ne peut tre oppose une dfinition large de la catgorie des
actes unilatraux, si lon ne se place pas sur le terrain des sources du droit. For the question
whether promises are a source of international law, see pp 17683.
160
For more see p 228.
161
For more see p 68.
162
Suy, Actes unilatraux, n 11 above, 31 et seq.

Unilateral63

act can subsequently create, alter or revoke a legal effect is said to be executed through that act; the acts very existence and its effects are understood to have been conditioned and determined by a bi- or multilateral
act. Accordingly, for Suy the act lacks an autonomous existence and is
not legally effective in and of itself. For Suy these acts are in effect the
acceptance of an offer made within the treaty: Ces dclarations unilatrales de forme ne sont quautant dacceptations dune offre faite dans la
disposition conventionnelle qui les prvoit.163 In the same vein, while, as
indicated, even going further, the Special Rapporteur wrote:
Of course, the unilateral acts in question are autonomous or independent of preexisting juridical norms, for, as noted in the first report on this topic, a State can
adopt unilateral acts in the exercise of a power conferred on it by a pre-existing
treaty or customary norm. This appears to be the case with regard to, inter alia,
unilateral legal acts adopted in connection with the establishment of an exclusive economic zone. Such acts, while of domestic origin, produce international
effects, specifically, obligations for third States which did not participate in their
elaboration. Naturally, such acts go beyond the scope of strictly unilateral acts
and fall within the realm of treaty relations.164

On the basis of this part of the autonomy test a great variety of acts
were diagnosed as not unilateral. Amongst these were acts of states establishing an exclusive economic zone, as mentioned in the quotation above,
as well as declarations made according to ICJ Statute, Article 36(2),165
also silence,166 acts leading to the international responsibility of a state,167
countermeasures,168 interpretative declarations,169 and any act made in the
course of proceedings before international tribunals.170
Ibid 33.
Cedeo, Second Report, n 3 above, para 62, emphasis added. See also Cedeo, First
Report, n 3 above, para 27, speaking of the absence of a connection with a pre-existing act or
norm or other manifestation of will.
165
See Suy, Actes unilatraux, n 11 above, 31, 142 et seq; Cedeo, First Report, n 3 above,
para 115, who seems to be excluding these acts more because they depended on another
unilateral act and gave rise to a treaty relationship. For more on this aspect see p 69.
166
Cedeo took the notion of autonomy even further when he stated silence cannot
be an independent manifestation of will, since it is a reaction to a pre-existing act or situation, Cedeo, Third Report, n 3 above, para 131, emphasis added. See also Cedeo, Fourth
Report, n 18 above, para 26 silence cannot be considered an autonomous manifestation of
will, since it is a reaction.
167
As it seemed difficult to conceive of an act which gives rise to the international responsibility of a State without being linked to the violation of a pre-existing norm, particularly
the primary norm which the act in question is alleged to violate, Cedeo, Second Report,
n 3 above, para 9.
168
Cedeo, Fourth Report, n 18 above, para 42, since they were linked to a pre-existing
commitment.
169
Since they formed part of a treaty-based relationship, Cedeo, First Report, n 3 above,
para 99.
170
Ibid para 125 et seq, as they should be placed within the context of the treaty which
founds the jurisdiction of the tribunal concerned, at para 126.
163
164

64 Delimiting the Subject: Promise as a Unilateral Act

In the course of discarding acts on the basis of their lacking autonomy,
this concept was apparently watered down to the extent that reactive
behaviour as such or any acts linked to a pre-existing norm or even situation were to be excluded.171 However, focusing on reactive behaviour
(as argued when excluding silence) would mean employing a useless criterion when it comes to distinguishing amongst unilateral acts of states;
both, protest and recognition are also reactions but widely accepted as
constituting unilateral acts, even autonomous ones, by those relying on
this criterion. The term link also does not provide for any delimitation if
used as a material concept, nor explain an exclusion unless employed in a
merely descriptive way to indicate which acts are likely to be covered by
a lex specialis.172
But even if confined to the above-mentioned quotation, ie the dependency on a pre-existing norm, the suggested definition of dependency
cannot convince. The Special Rapporteur clearly went too far by wanting to exclude even acts based on pre-existing customary norms as nonautonomous, as then practically all unilateral acts would be disqualified.
During the debate on autonomy within the ILC, it was rightly pointed
out that every legal act depended on a pre-existent legal rule in order to
become legally effective.173 And indeed, the Special Rapporteur seems to
have subsequently acknowledged this fact in his Third Report in which
he limited his understanding of autonomy to mean only autonomy from
other manifestations of will:
it can be stated that unilateral acts do not depend on an earlier act, that is to
say, on an earlier expression of will, although it is true that all unilateral acts are
based on international law;174

and:
While it is true that a legal act is linked to earlier rules, particularly rules of
general international law, this very broad approach cannot be the yardstick for
determining the autonomy of the act. The point is to exclude, by means of this
criterion, acts linked to other regimes, such as all acts linked to treaty law.175

This essentially leaves us with the question already brought up by

Suy, whether an act which finds its legal basis in a treaty can be said to
depend on it and should accordingly be seen as not a truly unilateral
but really a consensual act, that is thus regulated by the law of treaties.
The mere fact, however, that a treaty clause allows for a state to issue a
171
See above in relation to the exclusion of various acts of states, especially nn 16668
above.
172
For this (in our eyes) favourable approach, see p 67 et seq.
173
See Pellet, ILC, 2603rd meeting, n 133 above, para 19: In his opinion, there was always
an international rule to which all unilateral acts could be traced.
174
Cedeo, Third Report, n 3 above, 11, para 60.
175
Ibid 11, para 61.

Unilateral65

legal act does not suffice to confer a contractual nature on this sort of state
action or turn it into the acceptance of an offer made within the treaty.176
This point has already and formidably been made by Jacqu, who has
illustrated the consequences the above-described definition would have
for acts of international organisations: as an international organisation is
created through a treaty and therefore derives its legal powers from a consensual framework, its acts would have to be judged as never in fact being
unilateral but always contractual. Despite the fact that we are not going
to deal with unilateral acts of international organisations, the criticism
voiced by Jacqu in this regard is also to the point in the present context
(my translation provided in footnote):
Ce nest pas parce quun trait confre certaines comptences un organe international que tous les actes adopts par cet organe seront des actes conventionnels. Il ne faut pas confondre la convention dont lorgane tient sa comptence
avec les actes de cet organe. La volont unilatrale mise par celui-ci ne rencontre pas la volont des Etats parties la convention et ne forme pas avec elle
un accord de volont. Ce nest pas la convention qui donne naissance lacte
unilatral, elle en prvoit simplement lexistence. Lacte est cr par une manifestation de volont de lorgane comptent, il est unilatral.177

The same is true for a states unilateral act made in the execution of a
treaty provision. The provision itself cannot be said to have issued the
act, it merely establishes the possibility for the contracting state to create
a certain legal effect in the future. It is therefore the legal basis on which
the state may act. If it does so, the legal effects are, however, created by
its own manifestation of will, which merely becomes legally effective
because of the existing treaty norm. It is not a consensual act which establishes the width of a nations exclusive economic zone or territorial sea,
but a states single manifestation of will to do so. To construe this declaration as an acceptance of an offer made within the respective treaty,
misconstrues the latter. States do not offer each other to thereby establish
together the width of various national maritime zones, they merely agree
on abstract rules that are accepted amongst them as legally binding and
which allow for a state to act unilaterally within the margin prescribed.178
176
Even if true, we would, as indicated above, hold acceptance to be a unilateral act, only
covered by a lex specialis.
177
Jacqu, Elments, n 1 above, 325: By conferring certain competences on an international
organ not all acts adopted by said organ will become conventional acts. One should not
confound the convention from which the organ derives its competence with the acts of this
organ. The unilateral will issued by it does not encounter the will of the State parties to the
convention and does not form an accord with them. It is not the convention which gives
birth to a unilateral act, it simply foresees its existence. The act is created by a manifestation
of will of the competent organ, it is unilateral (my translation).
178
See ICJ, Fisheries (1951), n 14 above, 132: Although it is true that the act of delimitation
is necessarily a unilateral act, because only the coastal State is competent to undertake it,
the validity of the delimitation with regard to other States depends upon international law.

66 Delimiting the Subject: Promise as a Unilateral Act

These abstract rules are agreed upon by consensus, not the acts which
follow, which remain distinct from the consensual norm-building process
and still constitute (purely) unilateral acts. Their being based on a treaty
provision as regards to their binding force hence does not turn them into
a consensual act.179 In this regard the following passage contained in the
Joint Dissenting Opinion of Judges Spender and Fitzmaurice in the South
West Africa cases, to which Jacqu and others have already drawn attention, is of interest. In discussing the mandates true legal nature both
judges stated that:
We might add, what it should scarcely be necessary to say, that the fact that
an act is done under an authority contained in an instrument which is itself a
treaty (in this case the League Covenant) does not per se give the resulting act
a treaty character.180

And:
The second assumption is that if an act or instrument follows upon certain antecedent consents, this entails that it is itself an agreement. This is not the case. We
have already cited cases such as those under Article 17 of the Charter; and we
could cite numerous examples drawn from private law, of acts which can follow upon various consents and agreements, but which are themselves of quite
a different character. Even legislative acts can follow upon certain consents, and
there may even be, and often is, a constitutional requirement that these should
have been obtained. Yet when a Head of State issues a Decree or Order, and the
latter recites (as it often does) that it is made by and with the consent of his
Council, or of some other body, this does not impart even a vestige of a contractual character to the resultant act.181

Such clauses will therefore empower a state to act the way it does by
laying down a normative framework but will not in any way classify the
subsequent conduct as consensual or unilateral. The legal basis allowing
a state to act by means of a unilateral act in a legally effective way (the
sources of law question, be it a customary rule, a general principle or a
treaty) must hence be kept apart from the question whether the act itself
is unilateral or not.

179
See also the critique by Alain Pellet, Commentary to Art. 38 ICJ Statute in Andreas
Zimmermann, Christian Tomuschat and Karin Oellers-Frahm (eds), The Statute of the
International Court of Justice: A Commentary (Oxford, 2006) 677792, MN 9, where he criticises
Suys approach as not in line with reality: certainly, a State against which action is taken
by, eg, the Security Council under Arts 41 or 42 of the Charter, cannot be deemed to have
agreed to that measure.
180
ICJ, South West Africa (Ethiopia and Liberia v South Africa), Preliminary Objections,
Dissenting Opinion Spender and Fitzmaurice, [1962] ICJ Rep 465, 491.
181
Ibid.

Unilateral67

(iii) Its Really All about Lex Specialis

What should be recalled at this point is the function the classification of acts
into autonomous ones and dependent ones was supposed to have. This distinction was used in order to identify state action which could be grouped
under a similar judicial regime and to exclude those acts which are governed by a different set of rules, especially by the law of treaties. While this
is a sensible cause, we have seen that to relate to legal effectiveness does
not rule out an offer or denunciation of a treaty from the realm of unilateral acts. Yet, they are in part regulated within the law of treaties. Also, to
stipulate that all acts based on a treaty are automatically outside the realm
of unilateral acts is false, as many may indeed constitute unilateral manifestations of will with legal effect, for which the law of treaties will provide
no adequate answers, as the consensual element is missing in the act itself.
Neither prong of the autonomy test is therefore capable of validly distinguishing between acts which will be governed by the law on unilateral acts,
and those which will be covered by any other different legal regime. In our
eyes, this is not surprising, as the analysed concept of autonomy assumes
a material difference, which separates the truly unilateral acts from those
governed by other regimes. It is one thing to say that offer and acceptance
are peculiar unilateral acts (as they form the integral part of a process which
creates a bilateral legal relationship) with special rules being applicable to
them. It is something very different to stipulate a general divide between
unilateral acts, arguing that some acts are more unilateral than others.
All acts issued by a single subject of international law having the apparently willed legal effect by themselves are and remain unilateral acts. The
question is merely whether the default rules on unilateral acts are going to
be applicable to them, or whether a set of special rules already exists and
applies to the act in question. In the words of Alain Pellet, who argued very
much in favour of such an approach within the Commission:
The Commission, which wanted to produce a general system of unilateral acts
and not a list of special regimes, could leave aside certain unilateral acts, such
as ratification or reservations, because they were governed by special rules, but
not because of their lack of autonomy.182

The matter is therefore one of lex specialis.

182
Pellet, ILC, 2525th meeting, n 8 above, para 43. See also Pellet, ILC, 2603rd meeting,
n 133 above, para 13. The Report by Candiotti as Chairman of the ILC 1998 Working Group,
ILC, Summary Record of the 2543rd Meeting, UN Doc A/CN.4/SR.2543 (1998) para 3, actually proposed an Article which would have adequately reflected this approach: Another
article might be based mutatis mutandis on Article 3 of the 1969 Vienna Convention and refer
to unilateral acts not covered by the draft articles, stating, for example, that the draft articles
did not apply to unilateral acts governed by other specific legal regimes, such as the law of
treaties, the law of the sea, the law of international arbitral or judicial procedure, the rules
relating to neutrality and war, without prejudice to the legal force of such unilateral acts
and the application to them of the rules set forth in the draft articles to which they would

68 Delimiting the Subject: Promise as a Unilateral Act

(a) Acts Governed by the Law of Treaties
Various unilateral acts made in connection to treaties are indeed governed
by rules that are to be found in the Vienna Convention on the Law of
Treaties.183 For example, even though the terms offer and acceptance
are not employed by the Vienna Convention, which sets up a highly formalised procedure through which offer and acceptance will be hard to
hold apart, the manifestation of a will to be bound by a treaty remains a
unilateral act. It is, however, regulated by the Vienna Convention (Article
11 et seq), just as the denouncements or termination of international agreements are (Articles 54, 56 and 60), and declarations indicating the suspension of a treaty (Articles 57, 60). As indicated, we do not hold any of these
to be per se different from other unilateral acts. The reason for finding regulations in the Vienna Convention is simply that all of them are unilateral
acts which will typically arise in the context of the conclusion, termination
or modification of treaties. In a merely descriptive way you could say they
are linked to the realm of the law of treaties and it consequently made
sense to draw up a regulation for them within this area.
It is important to note that dropping the autonomy aspect and following
what might be termed a lex specialis approach does not render the distinction amongst unilateral acts obsolete. Every single rule that develops for a
certain type of unilateral act makes it necessary to differentiate whether the
act in question is one to which the rule will or will not apply. By refraining
from relying on a general criterion such as autonomy, which attempts to
divide all unilateral acts according to their effectiveness into two groups,
it remains necessary to look at the rule in question and the terms it stipulates, in order to establish whether this rule will be applicable or not. Of
great importance to our present study, therefore, is the distinction between
an offer and a unilateral promise, as different rules apply to them. We will
address the problem of holding the two apart in practice a little later,184 the
point being made here is only one of theory. Offer and promise are quite
similar unilateral acts in their appearance with, however, very different
legal regimes being attached to them. An offer is considered freely revocable and promises a further obligation only in the case of it being accepted,
which if done (even tacitly) will give rise to a treaty relationship. A promise,
on the other hand, does not ask for an acceptance but conveys the promibe subject under international law, independently of the draft articles, and to the extent that
the specific regimes in question did not contain any special rules on particular aspects.
The principle lex specialis derogat legi generali has been identified by the ILC as a generally
accepted technique of interpretation and conflict resolution in international law, see ILC,
Conclusions of the Work of the Study Group on the Fragmentation of International Law:
Difficulties Arising from the Diversification and Expansion of International Law in (2006) II
YB International Law Commission pt 2, para 5.
183
See also Tomuschat, Unilateral Acts, n 30 above, 1489.
184
For this aspect see also p 228.

Unilateral69

sors will to create an immediate and unconditional legal obligation, once it

is made known to the addressee. It is in this context that the ICJ has therefore used the word strictly unilateral to describe the nature of a promise.
As already seen, the Court in the Nuclear Tests cases stipulated that:
nothing in the nature of a quid pro quo nor any subsequent acceptance of the
declaration, nor even any reply or reaction from other States, is required for the
declaration to take effect, since such a requirement would be inconsistent with
the strictly unilateral nature of the juridical act by which the pronouncement by
the State was made.185

Merely in this sense to distinguish a promise from an offer which

invites an acceptance the Court employed the term strictly unilateral
(while also making it clear that a unilateral commitment can as such create a binding obligation).186 A unilateral act, an offer, can hence be clearly
delimited from our actual subject matter, a promise, and will therefore as
such not be under further scrutiny in the following pages, except where a
delimitation between the two acts is necessary.
(b) Declarations Made Pursuant to ICJ Statute, Article 36(2)
Declarations made according to ICJ Statute, Article 36(2)187 are also sometimes excluded from the study on unilateral acts. For Suy they are not
autonomous and constitute an example for acts made in the execution of a
treaty.188 The Special Rapporteur agrees and considers them to be formal
unilateral acts which, however, gave rise to a treaty relationship and should
be looked at in the context of treaty law.189 Having rejected distinguishing
between dependent and independent acts in the way both authors have
done, and in view of the fact that declarations made under the Optional
Clause are indeed to be regarded as having been made by states with the
intent to undertake a legal commitment to comport themselves in a certain
ICJ, Nuclear Tests (Australia v France), Judgment (1974), n 25 above, 267, para 43.
This quotation should not be mistakenly understood as establishing the autonomous/
non-autonomous distinction that Suy and the Special Rapporteur have both endorsed. By
using this wording the ICJ has merely underlined that declarations which constitute promises do not ask for nor need any quid pro quo or acceptance from the addressee in order for the
promised obligation to arise, which distinguishes them from offers. In this sense a promise
is strictly unilateral while an offer is not, as it opens the door for a bilateral relationship. The
ICJ, however, has not thereby described an offer as being devoid of immediate legal effect
nor has it held it not to be a substantive unilateral act.
187
ICJ Statute, Art 36(2) reads as follows: The States parties to the present Statute may at
any time declare that they recognise as compulsory ipso facto and without special agreement,
in relation to any other State accepting the same obligation, the jurisdiction of the Court in
all legal disputes concerning: (a) the interpretation of a treaty; (b) any question of international law; (c) the existence of any fact which, if established, would constitute a breach of an
international obligation; (d) the nature or extent of the reparation to be made for the breach
of an international obligation.
188
See p 62 et seq.
189
Cedeo, First Report, n 3 above, para 116 et seq, quoted passage from ibid para 117.
185
186

70 Delimiting the Subject: Promise as a Unilateral Act

manner within the future, they nevertheless may be of interest here. The
important question is whether a regime of lex specialis exists (be it the law
of treaties or one that has developed only for these kind of declarations on
account of their peculiarity), that establishes norms which will supersede
the ones usually applicable to unilateral promises.
The ICJ has had numerous opportunities to describe the legal nature
of these declarations on account of them frequently forming the (only)
basis for its jurisdiction. In doing so, it has left little room to doubt their
unilateral nature:
Declarations of acceptance of the compulsory jurisdiction of the Court are facultative, unilateral engagements, that States are absolutely free to make or not
to make. In making the declaration a State is equally free either to do so unconditionally and without limit of time for its duration, or to qualify it with conditions or reservations . . . However, the unilateral nature of declarations does not
signify that the State making the declaration is free to amend the scope and the
contents of its solemn commitments as it pleases. In the Nuclear Tests cases the
Court expressed its position on this point very clearly.190

The ICJ went on to cite the relevant passage from the Nuclear Tests cases
according to which declarations made with the intention of becoming
bound are turned into a legal undertaking.191 It subsequently continued
as follows:
In fact, the declarations, even though they are unilateral acts, establish a series
of bilateral engagements with other States accepting the same obligation of
compulsory jurisdiction, in which the conditions, reservations and time-limit
clauses are taken into consideration. In the establishment of this network of
engagements, which constitutes the Optional-Clause system, the principle of
good faith plays an important role.192

Subsequently, again, the Court cited the Nuclear Tests cases, this time
as underlining the fact that the need for respect of good faith and confidence in international relations explains not only the pacta sunt servanda
principle but equally the binding character of an international obligation
assumed by unilateral declaration.
To begin with (and more as a side note), one cannot fail to notice how
this passage constitutes another blow for a theory of autonomy used in
order to delimit different legal regimes. After having cited the Nuclear Tests
cases, ie the landmark decision for the bindingness of an (autonomous)
unilateral act, ie a States promise, the ICJ has no problem elaborating that
190
ICJ, Military and Paramilitary Activities in and against Nicaragua (Nicaragua v United States),
Jurisdiction and Admissibility, Judgment of 26 November 1984, [1984] ICJ Rep 392, 418,
para 59.
191
For the Nuclear Tests cases see pp 11638.
192
ICJ, Military and Paramilitary Activities in and against Nicaragua (1984), n 190 above, 418,
para 60.

Unilateral71

these unilateral acts in fact give rise to a series of bilateral engagements with
states accepting the same obligation. Also, by using the Nuclear Tests cases
in analysing the legal effects of declarations made under the Optional
Clause, a treaty provision, the Court clearly rejects excluding declarations which may be considered linked to or based on a treaty from the
legal regime on unilateral acts. As we have indicated above in relation
to a treaty offer, the Court is absolutely right in saying so. As with an
offer, declarations providing for the jurisdiction of the Court are not to be
excluded by declaring them to be non-autonomous. Yet, they may have
special rules applicable to them.
More importantly, though, the cited passage proves the importance of
holding apart the initial question whether a declaration constitutes a unilateral act, from the subsequent one inquiring as to what kind of legal
relationship the act might give rise and whether certain peculiarities or
norms established for it might warrant a legal treatment that deviates
from the one usually applicable to unilateral acts intended to create new
commitments, that is, promises. Here it is necessary to take one step at a
time, though. The declarations in question are clearly unilateral manifestations of a states will to create a legal obligation to comport itself in a
certain manner in the future. The ICJ has been crystal clear on this point in
the above-cited Military and Paramilitary Activities in and against Nicaragua
judgment. This assessment is of importance and has, for example, led it
to emphasise the special importance of the intention of the state accepting the compulsory jurisdiction when interpreting such declarations. The
method of interpretation can for this reason differ from the one applicable to treaties.193 In interpreting a declaration made by Iran in relation to
which a legal dispute arose, the Court also addressed the question whether
an appropriate interpretation would need to attach special importance to
each and every word employed by the declarant. According to the Court:
It may be said that this principle should in general be applied when inter
preting the text of a treaty. But the text of the Iranian Declaration is not a treaty
text resulting from negotiations between two or more States. It is the result of
unilateral drafting by the Government of Iran.194
193
According to the Court the regime relating to the interpretation of declarations made
under Article 36 of the Statute is not identical with that established for the interpretation of
treaties by the Vienna Convention on the Law of Treaties, ICJ, Fisheries Jurisdiction (1998),
n 149 above, 453, para 46. See also ibid 454, para 48: the Court has not hesitated to place a
certain emphasis on the intention of the depositing State. See also ICJ, Anglo-Iranian Oil Co.
(United Kingdom v Iran), Preliminary Objections, Judgment of 22 July 1952, [1952] ICJ Rep.
93, 104: It must seek the interpretation which is in harmony with a natural and reasonable
way of reading the text, having due regard to the intention of the Government of Iran at the
time when it accepted the compulsory jurisdiction of the Court. Emphasising the difference
to be at most only very slight, Christian Tomuschat, Commentary to Art. 36 ICJ Statute in
Andreas Zimmermann, Christian Tomuschat and Karin Oellers-Frahm (eds), The Statute of
the International Court of Justice: A Commentary (Oxford, 2006) 627, MN 65.
194
ICJ, Anglo-Iranian Oil Co (1952), n 193 above, 105.

72 Delimiting the Subject: Promise as a Unilateral Act

The very fact that the declarations in question are to be qualified as unilateral acts has therefore influenced the rules of their interpretation. While
this is an important starting point, the question remains open whether
specific legal rules are applicable to declarations under the Optional
Clause. As with an offer, which is initially governed by the general rules
on unilateral acts (unless otherwise established), the legal setting might
rapidly change and move into the conventional sphere once additional
elements are present. Questions of, for example, revocation or modification would then have to be answered in the face of a legal background
very different from the one present when making a standard promise.
Reasons to consider the legal regime governing the declarations under
the Optional Clause to be different are connected to the fact that the ICJ
Statute has made room for a unilateral declaration which is quite unique.
Through it a state accepts the compulsory jurisdiction only in relation to
any other State accepting the same obligation.195 The ICJ Statute therefore
allows and assumes an element of reciprocity for the commitment undertaken which is not present in a standard unilateral undertaking and might
significantly alter the nature of the act in question. In this regard, declarations under the Optional Clause are not easy to grasp. The Court itself has
interpreted them to include both, an acceptance as well as a standing
offer when it said:
Any State party to the Statute, in adhering to the jurisdiction of the Court in
accordance with Article 36, paragraph 2, accepts jurisdiction in its relations with
States previously having adhered to that clause. At the same time, it makes a
standing offer to the other States party to the Statute which have not yet deposited a declaration of acceptance. The day one of those States accepts that offer
by depositing in its turn its declaration of acceptance, the consensual bond is
established and no further condition needs to be fulfilled.196

Employing the terms offer and acceptance would suggest that we are
in the presence of not only a consensual bond but also a contractual one,
once two matching declarations exist. And indeed, the ICJ, especially
in an earlier judgment, has expressly spoken of a contractual relation
between the parties.197 In its more recent rulings, and especially in its
See the wording of ICJ Statute, Art 36(2), n 187 above.
ICJ, Land and Maritime Boundary between Cameroon and Nigeria (Cameroon v Nigeria:
Equatorial Guinea intervening), Preliminary Objections, Judgment of 11 June 1998, [1998] ICJ
Rep 275, 291, para 25.
197
ICJ, Right of Passage over Indian Territory (Portugal v India), Preliminary Objections,
Judgment of 26 November 1957, [1957] ICJ Rep 125, 146: The Court considers that, by the
deposit of its Declaration of Acceptance with the Secretary-General, the accepting State
becomes a Party to the system of the Optional Clause in relation to other declarant States,
with all the rights and obligations deriving from Article 36. The contractual relation between
the Parties and the compulsory jurisdiction of the Court resulting therefrom are established,
ipso facto and without special agreement, by the fact of the making of the declaration,
emphasis added. See also Hersch Lauterpacht, Law of Treaties, Report of the Special Rapporteur,
reproduced in (1953) II YB International Law Commission 90, 101.
195
196

Unilateral73

opinion in the Military and Paramilitary activities in and against Nicaragua

judgment cited above, it has, however, followed a different approach.
According to it:
the declarations, even though they are unilateral acts, establish a series of bilateral engagements with other States accepting the same obligation of compulsory jurisdiction.

This series of bilateral engagements is, however, not a series of bilateral treaties or engagements covered by treaty law. ICJ Statute, Article
36(1) already allows for a compromis, a special ad hoc agreement by the
parties. Declarations under the Optional Protocol in case of overlap display a consensus amongst two parties on the same issue, that of the Courts
jurisdiction. This consensus is, however, not brought about by a meeting
of minds between the parties concerned; the declarations made are not,
as offer and acceptance are, carefully crafted in order to reach a mutual
accord. Furthermore, each state unilaterally tailors an instrument of its own
choosing and deposits it with the Secretary-General. It is then for the ICJ to
consider the declarations overlap, if any, once a dispute arises. Article 36(2)
has therefore rightly been characterised as having been conues plus pour
viter une rencontre des volonts que pour la provoquer.198
Declarations made under Article 36(2) are consequently of immediate,
direct and full legal effect the minute they are made, which means deposited with the Secretary-General according to Article 36(4). As sovereign
acts of unilateral commitment they are open to all kinds of reservations
without being subject to the limits set out by Vienna Convention, Articles
19 to 22. The difference between the legal relationships they give rise to
from that constituted by a treaty becomes clearest in view of a states
decision to withdraw its undertaking. The ICJ does not inquire whether
the other parties to the Optional Clause system do or have at any time
consented to a states wish to revoke its declaration, as it would have
to if the declarations gave rise to a contractual bond. In the Military and
Paramilitary activities in and against Nicaragua judgment, the Court opined
as follows:
Although the United States retained the right to modify the contents of the 1946
Declaration or to terminate it, a power which is inherent in any unilateral act of
a State, it has, nevertheless assumed an inescapable obligation towards other
States accepting the Optional Clause by stating formally and solemnly that any
such change should take effect only after six months have elapsed as from the
date of notice.199
198
Paul Reuter, Principes de droit international public (1961) 103 Recueil des Cours de
lAcadmie de Droit International de La Haye 431, 575 (designed rather in order to evade a
meeting of minds than to provoke it, my translation).
199
ICJ, Military and Paramilitary Activities in and against Nicaragua (1984), n 190 above, 419,
para 61, emphasis added.

74 Delimiting the Subject: Promise as a Unilateral Act

Note that, were the quotation to finish its sentence after Optional
Clause, it might be misunderstood as pointing to the pacta sunt servanda
principle; the Court, however, continued by stressing the formal and
solemn commitment, thereby remaining very much within the doctrine
of unilateral commitments based on the Nuclear Tests cases to which it
already made reference twice within this ruling. With respect to declarations of infinite duration it adds that:
the right of immediate termination of declarations with indefinite duration is
far from established. It appears from the requirement of good faith that they should
be treated, by analogy[!], according to the law of treaties, which requires a reasonable time for withdrawal from or terminations of treaties that contain no
provision regarding the duration of their validity.200

The resulting legal network has hence been characterised as follows:

through uncoordinated unilateral declarations a system emerges which resembles to some extent a multilateral treaty, but does not provide the same expectations of stability and reliability since it is not placed under the proposition of
pacta sunt servanda or declaratio est servanda.201

Treaty law is therefore not applicable even in relation to two overlapping declarations.202 By pointing out that declarations under the Optional
Clause are not based on a principle of declaratio est servanda, Christian
Tomuschat in the above-cited passage underscores the fact that declarations as unilateral acts,203 while not freely revocable,204 are not governed
by a principle that would establish the same level of irrevocability comparable to the rigid principle of pacta sunt servanda. While being unilateral acts, declarations under the Optional Clause may hence have special
rules applicable to them, especially on the basis of the reciprocity element
established by the ICJ Statute. In chapter three on the Law on Promises,
caution will therefore be used when contemplating whether rules identified by the Court in relation to declarations made under the Optional
Clause are transferable to promises in general, or in fact constitute a lex
specialis applicable only in this limited area. Unlike the assumption of a
lack of autonomy, such declarations, however, are not per se discarded
from the realm of unilateral acts or the reach of our topic.205 The ICJs
Ibid, emphasis added.
Tomuschat, Art. 36 ICJ Statute, n 193 above, 627, MN 64. On the revocability of promises see pp 25176.
202
See also Juan Jos Quintana, The Nicaragua Case and the Denunciation of Declarations
of Acceptance of the Compulsory Jurisdiction of the International Court of Justice (1998) 11
Leiden Journal of International Law 97, 11011.
203
See Tomuschat, Art. 36 ICJ Statute, n 193 above, MN 64.
204
Ibid MN 6469.
205
Considering declarations under the Optional Clause to be a special kind of promise
also Degan, Unilateral Act, n 30 above, 20921. See also Jacqu, Elments, n 1 above, 33335;
Charpentier, Engagements unilatraux, n 30 above, 368, para 4; Quintana, The Nicaragua
Case and the Denunciation of Declarations of Acceptance, n 202 above, 111.
200
201

Unilateral75

findings on declarations under the Optional Clause will therefore not be

cast aside as a priori irrelevant.
(c) Declarations Made During Judicial Proceedings
As the History of Promises will show, assurances made by state representatives before the Permanent Court of International Justice (PCIJ) were
amongst the first to be attributed a legally binding character. The ICJ, in a
similar vein, later placed reliance on declarations made before it. Yet, both
courts have remained silent on how this bindingness was brought about.
As will be seen, the declarations in question can hardly be construed, nor
were they treated by the parties or the courts, as offers that were to be
accepted. States instead proclaimed a certain future conduct that benefited the other party as a given, usually in order to convince the court in
question that certain state action (such as an expropriation) was out of the
question and hence need no longer be contemplated by the judges in their
legal assessment. While these declarations never asked for an acceptance,
there is similarly no indication that either the declarant state or its beneficiary considered the declarations made as being an offer to be subject to
an international treaty that needed to be accepted, thereby possibly giving rise to domestic parliamentary ratification requirements. To assume a
contractual bond between the declarant and the court, a theoretical second
alternative for the declarations binding nature, is nothing more than that:
theoretical and similarly unconvincing. Not only might the courts legal
personality to conclude such a contract be questionable, but any such
dogmatic construction would also hardly be in touch with social realities. Neither the PCIJ nor the ICJ even indicated that they were accepting the declarations as an offer made but instead usually took note of the
declaration and recognised its binding character. The conclusion therefore
seems to be rather straightforward: the declarations are not only (here
there can be little dispute) unilaterally drafted but are indeed unilateral
binding commitments. Yet, the answer provided by legal commentators
is far from unanimous in this regard. While several authors indeed draw
this conclusion and include cases which involve declarations made before
international courts in their account of promises in international law,206
the Special Rapporteur chose to discard such declarations from the outset,
precisely on account of them having been made before a court. But a clear
doctrinal reason for this approach remains hard to grasp. According to his
First Report:
these declarations do not seem to be truly autonomous, even though they may
contain a promise, a waiver or a recognition; rather they should be placed
206
See, eg Suy, Actes unilatraux, n 11 above, 12931; Fiedler, Unilateral Acts, n 2 above,
1019; Venturini, Actes unilatraux, n 20 above, 398.

76 Delimiting the Subject: Promise as a Unilateral Act

within the context of the treaty which founds the jurisdiction of the tribunal
concerned. Moreover, the obligations which a State may assume through such
a declaration are related to the claim or legal position of the other State party to
the proceedings, which makes it difficult to classify them as autonomous from
this point of view.207

But what does this mean? The wording to place such declarations within
the context of the treaty which founds the jurisdiction of the tribunal is
extremely vague. As the short analysis provided above, and the longer
one vis--vis the notion of autonomy, have by now hopefully shown, the
mere connection to a treaty is insufficient to discard an act as unilateral
or not. Moreover, the present writer is not aware that the Statutes of the
PCIJ or ICJ declare statements made before them to be binding upon their
author and might thereby form the basis of a special legal act employable
(only) before these courts. When Alfred Rubin, also, considers all such
declarations as hardly unilateral since they were made in a context of
multilateral formality with the integrity of the tribunal [being] involved
directly,208 he too offers no legal explanation for the bindingness of these
declarations as assumed by both courts. What is the legal rule supporting the legal bindingness, what is the legal regime applicable to declarations made in the context of multilateral formality? Surely an offer made
before a tribunal to the other state will remain an offer, an acceptance an
acceptance, and the legal relationship resulting will be a treaty that is
binding upon the parties not because of some connection to the tribunal
but because there is a legal rule of pacta sunt servanda. Unilateral declarations made before an international court will similarly have to have a legal
basis explaining their bindingness. Indeed, the treatment of such declarations as unilateral acts is not only important when trying to identify legal
precedents for, in our case, promises but would quickly move to the centre of attention if a state wanted to revoke such a declaration qualified
as binding by a court, or, for example, claimed a fundamental change in
circumstances in order to terminate the declarations effect. Authors treating all such declarations as special and different on account of them being
connected to the jurisdiction of the court would need to provide a more
detailed answer in respect of the legal character and norms applicable to
such special declarations, such as why a recognition, protest or promise
before a court might be a different legal act than the very same statement
made outside the courtroom.
The only feasible way of distinguishing unilateral declarations before
tribunals considered as binding from other unilateral declarations, is
not to negate their unilateral character but to argue that it was the legal
forum with its heightened degree of formality which allowed reliance to
207
208

Cedeo, First Report, n 3 above, 26, para 126.

Rubin, Unilateral Declarations, n 146 above, 3.

Unilateral77

be reasonably placed upon oral declarations made by state agents in the

proceedings as actual legal undertakings. The difference between the two
approaches depicted is not as slight as might first seem, as now the focus is
shifted towards a different question, which is whether the required amount
of formality that allows the addressee and the court to infer an intent to
make a binding commitment and to be able to rely on a unilateral declaration can only be achieved through the courtroom setting. In other words: is
the courtroom the only venue in which unilateral acts will become binding
or can the judges bench be replaced through other means, such as, for
example, the deposition of a declaration with the UN Secretary-General,
official notification or a particularly solemn proclamation using clear and
specific wording? As the History of Promises will show, the latter has
come to be the case. That is to say, declarations of recognition or protest,
waivers and promises made before international courts are and remain
unilateral acts. A declaration treated by the tribunal as giving rise to a new
legal obligation is thereby a true promise. The only difference between
declarations in the courtroom and those outside is that the special setting
in the first case will allow us to assume an intention of the declarant to
become legally bound to his words more easily than when no such special
setting is present, since a state representative must be aware of the forum
he or she is speaking in, and is thereby put on notice that even the spoken
word is recorded and has to be very carefully chosen. That, however, does
not mean that only a courtroom setting will allow unilateral declarations
to be relied upon as binding commitments, nor does it exclude statements
made in such venue from the ambit of this study.
(d) A Brief Note on Other Acts Usually Excluded
Not all unilateral state action usually excluded on the basis of the autonomy test will be addressed here in as great a detail as acts falling into
the realm of the law of treaties and declarations made under the Optional
Clause or during judicial proceedings have been. It should by now go
without saying that none of them can, in this writers opinion, be per se
excluded from the legal regime applicable to unilateral acts in general or
that of promises in particular. Declarations delimiting maritime boundaries, if proclaimed at the international level or notified to other states, are
unilateral acts with legal effect in international law. Unless covered by a lex
specialis, the legal framework for unilateral acts will therefore be applicable,
and more precisely that of promises, which will, however, be called upon
only in exceptional cases, since it is not primarily in order to undertake a
new obligation vis--vis an addressee that such declarations are made.209
209
Consider, however, the example provided by Tomuschat, Unilateral Acts, n 30 above,
1490.

78 Delimiting the Subject: Promise as a Unilateral Act

The exclusion of silence on the basis of it being non-autonomous
reactive behaviour has already been declined.210 Yet, silence, if legally
relevant, will be so either as tacit recognition or because it is interpreted
as tacit acceptance of an offer made. It is hardly conceivable to interpret
silence as a promise to undertake a new obligation under international
law. Promises will therefore, unlike the other typically enumerated unilateral acts, be restricted to express acts, ie they have to be declared.
As already pointed out, acts leading to the international responsibility of a state may still be unilateral acts and if so will remain of interest,
which is similarly true for countermeasures. Interpretative declarations
were also ruled out by the Special Rapporteur on the basis of them lacking autonomy and might indeed not come within the legal framework of
promises, as long as they remain merely interpretative without taking on
a new or further obligation.
V PROMISES OF STATES ONLY

As its title already indicates, this study focuses exclusively on declarations made by states and not on those made by any other subject of
international law. Thus, in particular, declarations made by international
organisations are not within its reach. While this approach mirrors that
of the ILC and that of most scholars, who usually differentiate between
unilateral acts of states and those of international organisations,211 this is
of course not to say that only states are capable of committing themselves
through an international declaration. Quite on the contrary, just as international organisations are capable of assuming international obligations
through treaties, they can issue unilateral acts which are comparable to
those of states. Especially in their external affairs, ie when acting vis--vis
non-members, similarities are likely to arise.212 Yet, even here, the respec See p 64.
See, eg, the separate treatment of both subjects in Nguyen, Pellet and Daillier, Droit
international public, n 30 above, 359, 367 et seq along with the references provided therein for
each topic.
212
In their internal affairs unilateral acts are going to be just as frequent but are more
likely to be covered by a lex specialis which is to be found in the organisations own legal
framework. For example, EU Directives and Regulations, just like Security Council
Resolutions are unilateral acts, yet they have a legal framework applicable to them which
is determined through the organisations founding treaties, with the latter allowing for very
special unilateral acts to become effective vis--vis its members. The resulting variety of unilateral acts of international organisations complicates this topic. For more see, eg Michel
Virally, Unilateral Acts of International Organizations in M Bedjaoui (ed), International Law:
Achievements and Prospects (1991) 241 and Nguyen, Pellet and Daillier, Droit international public, n 30 above, 367 et seq. See also Cedeo, First Report, n 3 above, 89, paras 3038, where,
at para 32, he calls attention to the difference between acts of sovereign states inter se as
unilateral acts in the context of the relationship of coordination, on the one hand, and the
different scenario where on the basis of a common basic text the ensuing acts are unilateral
acts in the context of relationships of association.
210
211

Promises of States Only79

tive organisations internal legal structure along with its (limited) competences need to be taken into account in order to assess a declarations
legal value, and might require an adjustment within the legal framework.
Also, the historical background is different and each and every rule which
has developed between states will have to be probed as to its transferability. Hence, and just as with the law of treaties where two different,
yet largely similar, set of rules were developed,213 both actors will be kept
apart, with the present focus lying solely on states. As the example of the
Vienna Convention shows, a clarification of the rules applicable to this
classic actor of international law can, however, serve as a blueprint for a
legal framework applicable to international organisations subsequently to
be drawn up.

213
See the specially drafted Vienna Convention on the Law of Treaties between States and
International Organizations or between International Organizations of 1986. At the time of
writing there are 41 parties to this Convention. But as only 29 of them are states, the treaty
is not yet in force since its Art 85 stipulates a minimum requirement of 35 states parties. For
more, see the UN Treaty Collection at http://treaties.un.org.

2
A History of Promises

RTICLES OR MONOGRAPHS dealing with the law of treaties

are in the privileged position of being able to address the latter
without necessarily having to provide a chapter on the history of
treaties, understood as a closer analysis of past state practice and cases in
order to prove their very point of departure, ie that treaties are actually
a means by which states may assume a legally binding obligation. The
very existence of treaties as a legal institution gives rise to no apparent
dispute, even though questions pertaining to it, such as how to hold apart
non-binding from binding agreements, have never ceased to provide fertile ground for intellectual debate.1 Yet, treaties undoubtedly have a long
tradition in international law and could from the outset largely mirror
and thrive on the consensual legal mechanism present in practically every
domestic legal system: the contract. Also, at least since the late 1960s, the
international lawyer has a treaty on treaties, the Vienna Convention on
the Law of Treaties, to which he or she can point and which, even for its
non-members, as reflective of customary law,2 provides a rather detailed
legal framework for written agreements amongst states.3
When it comes to promises of states, whether written or not, the picture
is radically different: Not only are we lacking a consensus on important
aspects of the normative framework, as chapter three will show in more
detail, but the very existence of promise as a legal institution has been
subject to an ongoing debate, with the concept suffering constant attacks
from scholars since its appearance on the doctrinal stage.4 Practically
1
See, eg Jan Klabbers, The Concept of Treaty in International Law (The Hague/Boston/
London, 1996).
2
On this aspect, as well as the difficult distinction between codification and progressive
development within the Vienna Convention, see, eg Ian Sinclair, The Vienna Convention on
the Law of Treaties, 2nd edn (Manchester, 1984) especially 1021. Of course, even provisions
which were progressive at the time, might now have turned into customary law.
3
The Vienna Convention on the Law of Treaties of 1969 entered into force on 26 January
1980, with 111 parties to the Convention at the time of writing, see http://treaties.un.org.
4
See especially R Quadri, Cours gnral de droit international public (1964) (III) 113
Recueil des Cours de lAcadmie de Droit International de La Haye 245, 361; Alfred P Rubin,
The International Legal Effects of Unilateral Declarations (1977) 71 American Journal of
International Law 1, for whom the ICJ in the Nuclear Tests cases established a new rule of
international law, see at 28; also Alexandre-Charles Kiss, Les actes unilatraux dans la practique francais du droit international (1961) 2 Revue Gnrale de Droit International Public 317,

Important Cases and State Practice81

every example of state practice which has been put forward in order to
highlight the acceptance and use of binding unilateral assurances, has
found at least one published legal commentator rejecting it as an example
that really included both a binding as well as unilateral assurance. In addition, most domestic legal orders do not know of promises as unilateral
legal commitments.5 As even the international lawyer has domestic roots,
a predominantly bilateral mindset has come to prevail amongst the legal
profession, which tends, and sometimes by all means tries, to conceptualise any legal commitment in terms of a treaty relationship. Finally, unilateral acts, much less promises, are not mentioned amongst the sources
of international law which are to be applied by the International Court of
Justice (ICJ) as the principal judicial organ of the United Nations in deciding cases brought before it, even though the Court is to apply and decide
in accordance with international law.6 While this last aspect is treated
separately at the beginning of chapter three, all aspects taken together
make the following History of Promises a necessary ingredient to this
study.
I IMPORTANT CASES AND STATE PRACTICE

The history of promises in international law as a binding, though unilateral act seems to begin somewhere between the First and Second World
Wars, even though a precise date of birth is hard to pinpoint, as will be
seen. Instead, what has taken place resembles more a gradual elaboration of a legal institute which has been shaped primarily not through the
practice of states but that of international tribunals. Apparently, neither
for Grotius (15831645) nor Pufendorf (16321694), could a unilateral
assurance as such give rise to a binding commitment; an act of acceptance

especially 31719; and Charles de Visscher, Problmes dinterprtation judiciaire en droit international public (Paris, 1963) 18688.
5
See also Hugh Thirlway, Concepts, Principles, Rules and Analogies: International and
Municipal Legal Reasoning (2002) 294 Recueil des Cours de lAcadmie de Droit International
de La Haye 267, 34041; Wilfried Fiedler, Zur Verbindlichkeit einseitiger Versprechen im
Vlkerrecht (1976) 19 German YB International Law 35, 5051. The German and Italian legal
systems are sometimes cited as having a comparable legal mechanism. At least with reference to the former, which the present writer knows in more detail, this assessment is not
absolutely accurate. In the German legal system a unilateral promise/assurance will usually
have to be made in contractual form in order to be binding, even if the law might construe
the latter through assuming a tacit acceptance and declare the usually necessary reception of
the acceptance by the offeror as not necessary (German Civil Code, BGB, s 151). Yet, there
are one-sided and obliging legal acts, especially the so-called Auslobung (BGB, ss 657661a)
which regulates the (very) special case of a publicly made promise to provide a reward to
whomever fulfils the conditions laid out in it.
6
See the wording of ICJ Statute, Art 38(1).

82 A History of Promises
was necessary.7 This view was confirmed in an arbitral award rendered
by Baron Lambermont relating to the Island of Lamu in 1889, in which an
oral promise by a Sultan of Zanzibar was not considered to be binding on
account of an analysis which held that there had been no concurrence of
wills, ie no agreement, on all essential elements of the matter in question.
It was apparently assumed that a promise could only become binding, if it
became part of an agreement:
Que, pour transformer cette intention en une promesse unilatrale valant
convention, laccord des volonts aurait d se manifester par la promesse
expresse de lune des parties, jointe a lacceptation de lautre, et que cet accord
de volonts aurait d porter sur les lments essentiels qui constituent lobjet
de la convention.8

Similarly, in the (as far as I am aware) first book written on unilateral

acts in international law, the monograph of Franz Pfluger published in
German in 1936, a thorough and in depth analysis of the subject is presented; promises, however, are not addressed and thus, while not explicitly rejected, thereby implicitly held to be non-existent.9 Later publications
have, however, identified a couple of cases which date back before
Pflugers books date of release in 1936 and which at least by some are
held to include promises. The following chronological journey through
relevant case law10 will start with these cases and highlight important
judicial decisions, as well as some examples of state practice not before
tribunals, up to the present time. As mentioned in the introduction, uni7
See Hugo Grotius, The Rights of War and Peace (Translation, Indianapolis, 2005) vol II,
ch XI, para XIV: But that a Promise may transfer a Right, the Acceptance of the Person to
whom it is made is no less required here, than in the Case of transferring a Property; yet so,
that here also a precedent Request shall be judged to subsist, and to have the Force of an
Acceptance. For Pufendorf a promise was also always based upon the condition that the
addressee would be pleased to accept it, as the cause that moved the promisor was in him
[ie the addressee] alone, and not in other things as well, Samuel Pufendorf, Elementorum
Jurisprudentiae Universalis Libri Duo (translation, Oxford, 1931) vol 2, Def XII, para 10, 81.
8
Arbiter Baron Lambermont, Arbitration between Germany and the United Kingdom relating
to Lamu Island, Decision of 17 August 1889, [1889] Reports of International Arbitral Awards
237, 243. The translation of the relevant decisions head note as provided in the publication
reads as follows: [I]f a unilateral promise is to be considered an agreement, the concurrence
of wills must be indicated through the express promise of one of the parties, along with the
acceptance of the other party, and the concurrence of wills must relate to essential elements
which are the object of the agreement an oral promise is not sufficient to establish a right
to trade preference or priority.
9
See Franz Pfluger, Die einseitigen Rechtsgeschfte im Vlkerrecht (Zurich, 1936), addressing
recognition, protest, notification, waiver and occupation in more detail.
10
There is, of course, no doctrine of stare decisis in international law and the decisions of
the ICJ have according to its own Statutes Art 59 no binding force except between the parties and in respect of that particular case. The Court has nevertheless made it very clear that
it will not depart from its settled jurisprudence unless it finds very particular reasons to do
so, ICJ, Application of the Convention on the Prevention and Punishment of the Crime of Genocide
(Croatia v Serbia), Preliminary Objections, Judgment of 18 November 2008, para 53 and also
para 76.

Mavrommatis Jerusalem Concessions Case83

lateral pledges of states to follow a certain line of action are by no means

a rare event. It was therefore necessary to make a choice in this chapter as
to which declarations were to be analysed more closely. In doing so, up
until the ICJs rulings on the Nuclear Tests cases, the prominent examples,
many of which have been addressed by legal scholars, are included, in
particular as practically all have been attacked as false examples from one
or the other side. From there on, the waters are more uncharted and the
chapter, apart from cases that came before the ICJ, will focus on examples
of state practice in which a states declaration has been considered, by at
least one of those involved (ie the declarant or the declarations addressee)
to be a binding unilateral undertaking, and where preferably in addition a
legal opinion of some authority has addressed the statements legal value.
The reason for this choice is to be found in the purpose of this chapter,
which is to highlight how unilateral assurances of states are understood
and treated within the international legal community. The reactions of
scholars (where existent) are presented and analysed in respect of each
of the following examples, of which a brief overview is also provided in
Annex I to the book.11
A Mavrommatis Jerusalem Concessions Case
The Mavrommatis Jerusalem Concessions case was decided in 1925 by the
Permanent Court of International Justice (PCIJ) and is considered by Eric
Suy the first in which a unilaterally binding assurance, ie a promise, has
been relied upon by an international court. The following had happened:
Greece brought suit before the PCIJ against Great Britain, claiming just
and equitable compensation in money in protection of one of its nationals, Mr Mavrommatis. The latter had concluded an agreement with the
city of Jerusalem on 14 January 1914 in which he was granted the concession for the public distribution of electric power and electric tramways
in Jerusalem, as well as a concession for the construction and exploitation of the works necessary for the supply of drinking water to Jerusalem.
However, before Mavrommatis could begin the actual construction work,
the First World War broke out. Great Britain subsequently became the
Mandatory Power for Palestine under the League of Nations system
which also put it in charge of Jerusalem. Despite a letter by Mavrommatis
to the new government of Palestine seeking to determine how to proceed with the concessions granted to him before the war, Great Britain
concluded a contract with a Mr Rutenberg under which the latter, after
having fulfilled some conditions, was to receive a concession for the
utilisation of the waters of the rivers Jordan and Yarmouk and their
11

See p 314.

84 A History of Promises
affluents for generating and supplying electric energy. Possible conflicts with
previously granted concessions were indeed envisaged by the parties
and covered by a clause included in the agreement which opted for an
annulment procedure to be initiated upon application by the new concessionaire, Rutenberg:
In the event of there being any valid pre-existing concession covering the whole
or any part of the present concession, the High Commissioner, if requested in
writing by the Company [formed by Mr Rutenberg] so to do, shall take the necessary measures for annulling such concession on payment of fair compensation agreed by the Company or, failing agreement, determined by arbitration.12

After some back and forth with the British authorities, who accepted
that Mavrommatis had some rights under the concession granted to
him before the war, Mavrommatis claimed compensation for an alleged
expropriation. The British Crown Council, however, pointed out that the
concession for Rutenberg had not actually been put into operation and
that Mavrommatis was free to choose whether he wanted to proceed with
the concession he received back in 1914, or have it resolved in exchange
for an indemnity to be considered for some of the survey and investigation work already done by him; the more than slight catch being that
if Mavrommatis should choose to proceed with the contract, the concessions terms would not be adapted to the new economic situation but stood
as they were, including the amount of paper money agreed on before the
war. Mavrommatis, anything but amused by this chain of events, agreed
that he should be given the right of choice between the concession and an
expropriation, but in the event of executing the concession, argued that it
should be remunerated at its actual current cost with it being understood
that the tariffs were fixed [not in paper money but] in gold piastres.13 The
Greek authorities, to whom Mavrommatis had maintained close contact,
finally brought the matter before the PCIJ.
The Court in its ruling considered Great Britain to be under an obligation to respect its international obligations in carrying out its mandate in
the administration of Palestine and found obligations to arise especially
out of the Treaty of Lausanne and a Protocol annexed to it. According
to the Protocol, pre-existing concessions were to be maintained. On
this basis, the Court addressed the relation between the Mavrommatis
Jerusalem concessions, on the one hand, and the agreement concluded
between Rutenberg and the Crown Agents, on the other. Here, the Court
noted that the agreement itself did not annul the concession granted to
Mavrommatis but conferred a right on Rutenberg to apply for such an
annulment. The Court went on to describe how Rutenberg, in an exchange
12
PCIJ, Mavrommatis Jerusalem Concessions (Greece v Great Britain), Judgment of 26 March
1925, Series A5, 6, 16.
13
Ibid 25.

Certain German Interests in Upper Polish Silesia Case85

with the British authorities, had made it very clear that he was in fact
not going to make use of this right, but was willing to let Mavrommatis
execute his concession (the reason being that Rutenberg would otherwise
have been obliged under the agreement with the British authorities to
compensate Mavrommatis for his loss, which the latter had estimated at
125,000). The Court then cited a declaration made by the British representative, who, besides pledging Great Britains will to comply with its
obligations under the Lausanne Protocol as determined by the Court, also
gave the following assurance:
That being so, there can be no question of our acting on any request to expro
priate M Mavrommatis. If M Rutenberg was so I do not like to use the term
dishonest, so unreasonable, now as to ask to expropriate him after declaring
that he has no such intention, we should not act upon that request.14

The Court continued its judgment after this citation as follows:

After this statement, the binding character of which is beyond question, the Court
considers that henceforward it is quite impossible that the British or Palestine
Governments should consent to comply with a request for the expropriation of M Mavrommatis Jerusalem concession. The clause in Article 29 of the
Rutenberg concession which gave him the right to demand such expropriation must therefore be regarded as deleted and all that remains of that article is an obligation on the part of M Rutenberg and his successors to respect
M Mavrommatis concessions.15

The statement made by the British representative was therefore considered binding by the Court, which, however, did not provide any more
insight into how this bindingness was brought about or what rules were
to govern such a declaration. This is astonishing insofar as the unilateral
character of the statement just quoted is very clear. The British representative neither asks for an acceptance from the beneficiary Mavrommatis,
nor does he ask the Court for one. Instead, the decision of the British government is presented as a fact to the Court, devoid of any conditionality upon a second manifestation of will: there can be no question of our
acting on any request to expropriate M Mavrommatis. Yet, for the Court,
the binding character of this statement was beyond question.16
B Certain German Interests in Upper Polish Silesia Case
Only a year later, in 1926, the PCIJ again had to deal with the legality
of expropriation, this time directed against property owned directly by
Ibid 37.
Ibid 3738, emphasis added.
16
Some authors have, as mentioned, discarded these declarations simply because they
were made before an international tribunal. On this aspect see pp 7577.
14
15

86 A History of Promises
various German nationals or by companies under their control. Poland,
however, subsequently withdrew several of its notifications of an intention to liquidate certain rural estates, and the Polish representative made
it clear that in respect of some of the property originally notified, there
was no longer any intention of liquidation. In its judgment, the Court
addressed this issue as follows:
The representative before the Court of the respondent Party, in addition to the
declarations above mentioned regarding the intention of his Government not
to expropriate certain parts of the estates in respect of which notice had been
given, has made other similar declarations which will be dealt with later; the
Court can be in no doubt as to the binding character of all these declarations.17

In the course of the judgment, the Court returned to the declarations

made in respect of the individual properties.18 Their precise value does
not, however, become entirely clear in these passages, especially as an
immunity from expropriation seems to follow from a contractual duty
already undertaken by Poland (under the Geneva Convention between
Poland and Germany concerning Upper Silesia signed on 15 May 1922).
Scholars have, therefore, seen room to argue that the declarations were
not intended to assume a new binding commitment.19
Whether the declaration in fact merely assured compliance with an
already existing obligation has also been questioned in respect of the
British declaration in the Mavrommatis Jerusalem Concessions case above.20
There, however, it is harder to make the argument as the Court expressly
left it open whether a right remained under the Protocol for Great Britain
to expropriate Mavrommatis, even though it did consider the clause
included in the contract with Rutenberg to infringe Mavrommatis rights
under his concession (especially as it allowed a private individual to trigger
an expropriation procedure for his own benefit).
Be that as it may, even those commentators who emphasise that these
declarations comprise a confirmation of an existing duty cannot ignore
that the PCIJ has chosen a different approach in both cases by announcing
expressis verbis the binding character of these declarations; it has thereby
clearly not treated them as pledges to comply with already existing con17
PCIJ, Certain German Interests in Upper Polish Silesia (Germany v Poland), Judgment of 25
May 1926, Series A7, 2, 13, emphasis added. For Klabbers, Concept of Treaty, n 1 above, 168
69, the PCIJ by referring to all these declarations, referred also to declarations made outside
the proceedings. If this were the case, the position advocated above (p 7577), according to
which declarations made during judicial proceedings cannot simply be dismissed as special and non-unilateral, would find additional support in this judgment, since the Court
treated all statements the same and as equally binding. From reading the judgment, however, it is hard to tell whether the PCIJ is really also alluding to declarations made outside
the oral and written proceedings.
18
See PCIJ, German Interests (1926), n 17 above, especially 58, 7172.
19
See Quadri, Cours gnral, n 4 above, 366.
20
Besides Quadri, ibid, see Rubin, Unilateral Declarations, n 4 above, 3, note 6.

Free Zones Case87

tractual duties but has instead attributed a binding, ie a legally restraining

power to these declarations.21
C Free Zones Case
In its opinion in the Free Zones of Upper Savoy and the District of Gex case,
delivered 7 June 1932, the PCIJ again emphasised the bindingness of a declaration made by a state representative in the course of the proceedings. In
response to doubts which had been expressed by the French side, whether
a declaration made by the Swiss representative accepting certain terms
without making any reservation for subsequent ratification could really
be considered binding from a constitutional point of view, the Court said
that having regard to the circumstances in which this declaration was
made, the Court must however regard it as binding on Switzerland.22
The PCIJ also and expressly decided to place the declaration on record.23
As such, it is tempting to include this incident in the list of cases in which
promises have received juridical acceptance. But on closer look, it was not
a promise that was relied on here. While a unilateral act, the declaration
was both drafted and treated by the parties as an offer which required
acceptance from France as its addressee. As the Court described it:
In the course of the oral arguments before the Court, the Swiss Agent declared,
on behalf of his Government, that if the zones were maintained, the Federal
Government would, if France so desired, agree to the terms of goods between the
zones and Switzerland being settled by experts, failing agreement with regard
to them by the Parties.24

The declaration was therefore understood by the Court to give rise to the
following legal situation:
If the French Government chooses to avail itself of this offer, it will be able to
obtain in favour of the inhabitants of the zones an arrangement for their produce into Switzerland, which will not depend exclusively on the will of the
Swiss Government.
It is true that, in the course of the recent hearings, the French Agent declared
the Swiss proposal to be inacceptable; but it is also true that he regarded it as an
21
According to Alfred Verdross and Bruno Simma, Universelles Vlkerrecht: Theorie und
Praxis, 3rd edn (Berlin, 1984) 429, the Polish representative who had made the declaration
in the German Interests in Upper Polish Silesia n 17 above, considered it binding for Poland.
Although both authors are of the opinion that the declarations only repeated an already
existing obligation, they concede that the Court apparently considered it to be a unilateral
commitment. Quadri, Cours gnral, n 4 above, 366, also declares the Court to have improprement spoken of a declaration obligatoire.
22
PCIJ, Free Zones of Upper Savoy and the District of Gex (France v Switzerland), Judgment of
7 June 1932, Series A/B 46, 96, 170.
23
Ibid 172.
24
Ibid 169, emphasis added.

88 A History of Promises
offer to conclude a Special Agreement, an offer which, in this form, he had no
power to entertain.25

We are therefore in the presence not of a promise but of an offer which

will compel Switzerland to do what it offered only if France so desires
and accepts it. 26 The Free Zones case is not, therefore, of direct precedential27 value in this context. It does, however, clearly illustrate the need to
differentiate between unconditional assurances, on the one hand, and an
offer requesting acceptance, on the other.
D Declarations Concerning the Protection of Minorities
Moving along and thereby away, at least for now, from declarations made
before international tribunals leads us towards those made by various
countries, especially Lithuania, Estonia and Albania in the 1920s, as well
as Iraq in 1932, vis--vis the League of Nations, all addressing the protection of minority rights. They are indeed amongst the first examples cited
of states using binding unilateral assurances at the international level.28
But again, and as in practically every case that is to follow, there is considerable disagreement whether the declarations really constitute unilateral
commitments. Unlike with other declarations, here, their bindingness is
not as such questioned, but their unilateral character is. Instead of being
compelling as a unilateral promise, the declarations are considered by
some to form part of a bilateral exchange, with the restraining power on
the declarant therefore resulting not from the assurance alone, but from
a treaty concluded, of which it formed merely a part. The assurance, in
other words, is considered to be either an offer that was subsequently
accepted by the League, or was itself the acceptance of an offer the League
had made.
Polled on the unilateral-bilateral question, legal commentators seem
to split into two rather equally sized groups, with some abstentions:
For Balladore-Pallieri, who was amongst the first to attribute binding
force to unilateral promises, the declarations made are indeed examples
of binding promises.29 De Nova also considers at least some of them as
Ibid 170.
For Eric Suy, Les Actes Juridiques Unilatraux en Droit International Public (Paris, 1962)
126, note 51, the offer is one to enter into negotiations. See, however, Klabbers, Concept of
Treaty, n 1 above, 16869, for whom such an interpretation is unduly restrictive; the declaration went beyond an offer to enter negotiations but instead contained concessions on the
part of Switzerland.
27
On the (non-technically speaking) precedential value of PCIJ and ICJ cases, see the
brief comment at n 10 above.
28
There were similar declarations by other countries, amongst them Finland, Latvia,
Greece and Bulgaria. The latter three are briefly addressed by Suy, Actes Unilatraux, n 26
above, 11819.
29
Giorgio Balladore Pallieri, Diritto Internazionale Pubblico, 7th edn (Milano, 1956) 316.
25
26

Declarations Concerning the Protection of Minorities89

proof of the existence of promises in international law.30 The assessment provided by Degan is somewhat more careful: the situation was
not clear-cut but rather ambiguous and there was room to view the commitments to have been established either through a treaty or a unilateral
promise.31 Suy, on the other hand, after examining the history of each
of the above-mentioned declarations made, draws the same conclusion
as Quadri did a few years before him, namely, that none of them are
unilateral commitments.32 Rousseau is a little more cautious and classifies these declarations as leading to quasi-contractual engagements.33
Writing 10 years later than Suy, Jacqu, dissecting Suys reasoning, dis
agrees completely and brings us back full circle to the original assessment of Balladore-Pallieri, ie that the declarations made were binding as
unilateral assurances.34
The factual backdrop for this disagreement is, in a nutshell, the following: after the First World War, minimum guarantees for the protection of minorities, the adherence to which were to be monitored by the
League of Nations, had been set up through several treaties between the
Allied Powers, on the one hand, and newly formed as well as defeated
states in Eastern Europe, on the other. Albania, Estonia, Lithuania (and
also Iraq) were amongst the countries not covered by any such treaty but
the League considered similar guarantees necessary and on 15 December
1920 adopted the following recommendation:
In the event of Albania, the Baltic and Caucasian States being admitted into
the League, the Assembly requests that they should take necessary measures
to enforce the principles of the Minority Treaties, and that they should arrange
with the Council the details required to carry this object into effect.35

The commitments were to be brought about through declarations

which, in order to ensure that the standards considered necessary by
the League of Nations were to be achieved, were actually drafted by
the League itself. They were then forwarded to the respective country which
was asked to sign and formally communicate them (back) to the League
of Nations. The Leagues Council subsequently took note of the particular
Rodolfo De Nova, Die Neutralisation sterreichs (1958) 54 Die Friedenswarte 298, 303.
Vladimir-Djuro Degan, Unilateral Act as a Source of Particular International Law
(1994) 5 Finnish YB International Law 149, 199.
32
Suy, Actes Unilatraux, n 26 above, 11421; Quadri, Cours gnral, n 4 above, 128; similarly Visscher, Problmes dinterprtation, n 4 above, 188.
33
Charles Rousseau, Droit international public, vol I, Introduction et Sources (Paris, 1970)
423.
34
Jean-Paul Jacqu, Elments pour une thorie de lacte juridique en droit international public (Paris, 1972) 25153, analysing the declarations made by Lithuania, Estonia, Iraq and
Albania.
35
See PCIJ, Minority Schools in Albania, Advisory Opinion of 6 April 1935, Series A/B, No
64, 1935, 4, 7.
30
31

90 A History of Promises
declaration through a specific Resolution. Finally, the declarations were to
be ratified by the respective countries and deposited with the League.36
This interaction between the League and the country in question has
led some commentators to assume a treaty relationship as having been
established between the two. As mentioned, Suy, who offers a quite
detailed analysis, draws the conclusion that none of the declarations
could be classified as a promise in international law. In fact, the declaration issued by Lithuania, which the assessment of the other declarations is
said to confirm, was the acceptance of an offer, a proposition emanating
from the League of Nations, the offer being the above-cited recommendation made by the Council.37 The fact that various state representatives
could be quoted as having accepted the text drafted by the League, or
that the representative of Estonia expressed his satisfaction at the agreement which has been reached between the Council and the Estonian
Government regarding the question of minorities in Estonia38 are said to
prove this point.
Yet, as Jacqu has remarked,39 the citations relied on by Suy do not
warrant this assessment. The representatives merely indicated that they
accepted the text worked out by the League of Nations, but in order to
proclaim it as their own commitment. Indeed, the League in the abovementioned Resolution does not offer anything when it recommends
the countries to adopt a certain declaration. Such a recommendation or
request could only be an offer if the League was offering to undertake an
obligation in the case of an acceptance by the countries addressed, which,
however, it does not do. The bilateral relationship can hence only be construed in the opposite way: when Lithuania, Estonia and the other countries are asked to apply a certain scheme of rights to their minorities, they
are in fact asked to make, and subsequently alleged to have made, an offer
for the protection of minority rights vis--vis the League. Understood in
this way, the Leagues subsequently adopted Resolution taking note of the
declaration would function as the offers acceptance.40 While this interpre36
The history of the Albanian declaration signed 2 October 1921, ratified by the Albanian
government on 17 February 1922 and deposited at Geneva on the following 22 March, is the
most easily traceable as it is depicted in the ICJ Advisory Opinion, see ibid 9. For the declaration of Lithuania, dating 12 May 1922, see League of Nations, Minorities in Lithuania, (1923)
4 League of Nations Official Journal 932 and Minitciro Adatci, Petition from Thirty-Four
Persons of Russian Origin Living in Lithuania Regarding the Confiscation of their Lands
under the Lithuanian Agrarian Reform Law of 15 February 1922, Report to the League of
Nations Council (1930) 11 League of Nations Official Journal 179; for that of Estonia made on
17 September 1923, see League of Nations, Minorities in Estonia, 4 League of Nations Official
Journal 131012. For the declaration made by Iraq see League of Nations, Mandates Proposal
and Annex, 121216, 134250 and also Suy, Actes Unilatraux, n 26 above, 119.
37
Suy, Actes Unilatraux, n 26 above, 117.
38
League of Nations, Minorities in Estonia, 4 League of Nations Official Journal 1311.
39
Jacqu, Elments, n 34 above, 252.
40
The League of Nations international personality is thereby assumed; for more on this
question see Bardo Fabender, Die Vlkerrechtssubjektivitt internationaler Organisationen

Declarations Concerning the Protection of Minorities91

tation may stress that the declarations were from the outset designed to
be addressed to the League on its own request, this understanding of the
events cannot be supported by the citations relied upon by Suy (as the
states were making the offer and not accepting it), and there is a good
argument to be made that the declarations were in and of themselves solemn
commitments which the League only recognised as such, thereby emphasising that they were being relied upon as legal commitments. Besides the
fact that the above-described procedure seems rather odd for the conclusion of a treaty, the interpretation of a unilateral act would also be more
in line with what the Estonian (and as will be seen also the Lithuanian)
representative expressly stated before the League:
it must be clearly understood that this declaration forms, together with the
resolution submitted to the Council, an indivisible whole which must not, however, be regarded as constituting a Minorities Treaty.41

Interestingly, two of these declarations, the Lithuanian as well as the

Albanian one, became subject of separate and heightened legal scrutiny
after their proclamation, when both countries were accused of having
infringed the guarantees included therein. The Lithuanian declaration
and the alleged breaches thereof formed the basis for a report by Mr
Adatci,42 appointed as a Special Rapporteur by the League; the declaration of Albania was forwarded to the PCIJ by the Council of the League
asking for an advisory opinion on whether Albania had breached its international obligations when it announced the closure of all private schools,
including those used by the countrys Greek minority.43 To begin with
the Court, it had no doubt that the declaration made was binding upon
Albania, but despite the doctrinal tug-of-war that was to ensue in the
years to come amongst legal doctrine trying to explain the precise nature
of the declaration in question, the Court did not consider it necessary to
discuss, let alone explain, how the declaration made could have this effect.
In its opinion, the Court, never speaks of a treaty having been concluded
but strictly confines itself to the use of the word declaration or act:
The declaration of 2 October 1921 belongs to the numerous category of inter
national acts designed for the protection of minorities.44

The legal construction for the in-any-event binding commitment thus

remains unclear, since a treaty may be termed a bilateral act, a promise a
unilateral act. Yet, for Suy, the following wording used by the Court the
(1986) 37 sterreichische Zeitschrift fr ffentliches Recht und Vlkerrecht 17, 2225, English
summary at 4849.
41
League of Nations, Minorities in Estonia, 4 League of Nations Official Journal 1311.
42
Adatci, Report, n 36 above, 17985.
43
PCIJ, Minority Schools in Albania (1935), n 35 above, 423.
44
Ibid 16.

92 A History of Promises
fact that what the Council of the League of Nations asked Albania to accept,
and what Albania did accept, was a regime of minority protection (emphasis added) showed that the underlying legal construction was that of a
treaty. But similar to what has been said above, the fact that the Council
asked Albania to accept a regime of minority protection does not mean
it was thereby making an offer surely it was not offering the regime
(an obligation for Albania!) to Albania even though it asked the latter to
accept it. At most, it could be assumed that it was offering access to the
League itself in return for the declaration, despite the fact that Albania
was admitted on 17 December 1920 and thereby roughly a year before it
made the requested declaration. In any event, the guarantee seems more
of a requirement set out by the Council than the quid pro quo for an accession.45 What the Council was really doing was asking Albania to undertake an obligation, but whether the latter did so through an offer (the
declaration which was proclaimed as its own undertaking) then accepted
by the Council (through a Resolution) or by using a unilateral assurance
(the declaration alone) is not clear.46
Just as the Court refrained from deciding how the binding nature of
the declaration by Albania was brought about, so too did Adatci in his
report submitted to the League on 15 January 1930 and addressing the
Lithuanian declaration in response to a petition made by members of the
Russian minority in Lithuania, who claimed that their rights had been
infringed upon. Unlike the Court, Adatci, however, saw himself pressed to
the point of addressing the problem discussed here, since the Lithuanian
representative, Mr Voldemaras, (much like the Estonian representative
had done before him) claimed that the engagements:
did not constitute an international treaty, but formed a declaration which had
been read before the Council and of which the Council had taken note.47
45
The League in its initial resolution did not stipulate that it was going to admit the countries if they accepted the minority declarations, ie that it was obliging itself to admit the
countries if they made the declaration. The wording of the resolution instead uses the opposite conditionality, ie if the countries are admitted, the Council requests them to adopt a
minority declaration an understandable approach, as only in the latter scenario will the
Council as the organ of an international organisation be addressing one of its members.
46
The dissenting opinion turns out to be similarly unclear on this point, as the passage
quoted by Suy, Actes Unilatraux, n 26 above, 121, appears to be a mistranslation: While
in the French translation, the dissenters Sir Cecil Hurst, the Count of Rostworowski and
M Negulesco regretted the absence of a preamble to the declaration in the following words:
Cette source ou lon pourrait trouver un guide virtuel pour interprter le trait, fait donc
dfaut, they did in fact only remark: That source of potential guidance in the interpretation
of the instrument is therefore lacking, PCIJ, Minority Schools in Albania (1935), n 35 above,
Dissenting Opintion, 31, emphasis added. While in another passage the dissenters got closer
than the majority to construing the declaration as a treaty (see ibid 28), they never made their
position clear, and repeatedly spoke of the declaration as belonging to the series of instruments (ibid especially 27, here the French version speaks of srie des actes), and not the
series of treaties which were installed for the protection of minorities, and therefore did not
provide a clear answer to the question raised here.
47
Adatci, Report, n 36 above, 183.

Legal Status of Eastern Greenland Case93

Apparently, and significantly, for Adatci, the matter was of minor import
ance:
The last point raised by Mr Voldemaras with regard to procedure refers to the
unilateral character of the engagements entered into by Lithuania with regard
to the protection of minorities . . . Without considering, at this moment, how far
the minority engagements entered into by Lithuania were unilateral or bilateral, it seems to me incontestable that they constitute international obligations
for Lithuania.48

Adatci in the following emphasised that the declaration expressly

spoke of the guarantees as constituting obligations of international interest, that were to be placed under the guarantee of the League, a commitment which was again expressly renewed in a letter which informed the
League that a ratification by the Lithuanian Parliament of the declaration
was not necessary.49 In other words, for Adatci at the time (1930), the declarations, no matter whether unilateral or bilateral (that is to say even if
unilateral), were binding upon Lithuania, as she had expressly proclaimed
her will to assume an international obligation. The rather special formal
procedure intentionally followed in these cases notwithstanding, it is this
assessment, together with the Courts silence on the unilateral or bilateral
(but in-any-event binding) declaration, along with the above cited positions of Estonia and Lithuania, which remains particularly noteworthy.50
E Legal Status of Eastern Greenland Case
The well known Legal Status of Eastern Greenland51 case brought by Denmark
against Norway before the PCIJ and decided by the Court in 1933 (with
Adatci now serving as judge and president of the PCIJ), features a declaration which has received the attention of a vast number of international
legal scholars, the so-called Ihlen declaration. Named after its declarant,
Mr Ihlen, who served as the Norwegian Minister for Foreign Affairs at the
time, it was an oral assurance directed to the Danish Minister at Christiana
and made on 22 July 1919, that the Norwegian Government would not
make any difficulties in the settlement of this question.
The question in the settlement of which Norway gave assurance not to
make any difficulties was Denmarks intention to extend its political and
economic interests to the whole of Greenland. When Norway, roughly
10 years later, published a proclamation in which it declared that it had
proceeded to occupy certain territories in Eastern Greenland, Denmark
Ibid 183.
Ibid183.
50
See in this respect also Gian Carlo Venturini, Attitudes et actes unilatraux des tats
(1964) (II) 112 Recueil des Cours de lAcadmie de Droit International de La Haye 363, 399.
51
PCIJ, Eastern Greenland (Denmark v Norway), Judgment of 5 April 1933, Series A/B 53, 22.
48
49

94 A History of Promises
brought a claim before the PCIJ (on 12 July 1931) as it considered these
territories to be subject to its own sovereignty. The judgment rendered by
the Court provided a lengthy description of Greenlands history reaching back to its discovery in 900 AD and drawing the conclusion that
Denmark indeed had had sovereignty over all of Greenland at the time
the Norwegian proclamation had been made.52 This historical support,
however, does not seem to have sufficed for the majority, as they went
on to bolster their opinion by analysing various undertakings Norway
had allegedly made and which had been advanced by Denmark as additional support of its claim.53 In doing so, the Court began by analysing
various bilateral and multilateral agreements, to which both Denmark
and Norway were contracting Parties54 and which it considered as proof
of a Norwegian recognition of Danish sovereignty over the whole of
Greenland. In addition to the engagements dealt with above the Court
found that it had also to consider the Ihlen declaration. The declaration itself, as well as the request which had prompted it, had both been
recorded in a minute prepared and initialled by Ihlen himself, which read
as follows:
I. The Danish Minister informed me today that his Government has heard from
Paris that the question of Spitzbergen will be examined by a Commission of
four members (American, British, French, Italian). If the Danish Government
is questioned by this Commission, it is prepared to reply that Denmark has
no interests in Spitzbergen, and that it has no reason to oppose the wishes of
Norway in regard to the settlement in question.
Furthermore, the Danish Minister made the following statement: The Danish
Government has for some years been anxious to obtain the recognition of all
the interested Powers of Denmarks sovereignty over the whole of Greenland,
and it proposes to place this question before the above-mentioned Committee
at the same time. During the negotiations with the U.S.A. over the cession of
the Danish West Indies, the Danish Government raised this question in so far
as concerns recognition by the Government of the U.S.A., and it succeeded in
inducing the latter to agree that, concurrently with the conclusion of a convention regarding the cession of the said islands, it would make a declaration to
the effect that the Government of the U.S.A. would not object to the Danish
Government extending their political and economic interests to the whole of
Greenland.

See the Courts conclusion, ibid 64.

The majority was, of course, aware that in his dissent, Judge Anzilotti attacked the
majoritys reliance on a historic claim instead of it requesting proof of effective possession,
which Denmark at the time did not seem to have had over the whole of Greenland and especially not over the territories in question. As a result, the case, in Anzilottis opinion, was to
be decided on the basis of the undertakings made by the parties in respect of the territories
in Eastern Greenland, see PCIJ, Eastern Greenland (1933), n 51 above, Dissenting Opinion of
Judge Anzilotti, 76.
54
PCIJ, Eastern Greenland (1933), n 51 above, Judgment, 68.
52
53

Legal Status of Eastern Greenland Case95

The Danish Government is confident (he added) that the Norwegian
Government will not make any difficulties in the settlement of this question.
I replied that the question would be examined.

14/7-19 Ih.

II. To-day I informed the Danish Minister that the Norwegian Government
would not make any difficulties in the settlement of this question.

22/7-19 Ih.55

Denmark had argued that the declaration (I informed the Danish

Minister that the Norwegian Government would not make any difficulties in the settlement of this question) constituted a binding act for
Norway. Speaking for Denmark, Mr Steglich-Peterson in the pleadings
before the Court expressly labelled the Ihlen declaration a unilateral act,
which was binding upon Norway; yet not as a special new form of a unilateral promise but because it was said to have expressed the Norwegian
recognition of Danish sovereignty over the whole of Greenland.56
Charles de Visscher, he too speaking for Denmark, and pleading a few
months later before the Court, also stressed that the declaration had in his
view not given rise to a treaty. The Norwegian allegations that oral treaties, if they existed, were only curiosits historiques had to be considered
besides the point, as:
Il ne sagit pas ici dun trait, il sagit dun engagement pris pour compte
du Gouvernement norvgien, dune promesse faite pour compte de ce
Gouvernement. Le Gouvernement danois a pris acte de la promesse faite par
M. Ihlen au ministre de Danemark, M. Krag. Il a fait plus, il a agi en consquence,
et cela immdiatement, en donnant son appui diplomatique la Norvge dans
la question du Spitzberg. La pratique internationale de tous les jours nous offre
des exemples constants de promesses de ce genre, dengagements revtant un
caractre obligatoire, sans pour cela assumer la forme dun trait.57
Ibid 6970.
See PCIJ, Eastern Greenland (1933), n 51 above, Judgment, 69 and especially the statement by Steglich-Peterson made before the Court as the Danish representative on 25
November 1932, Publications of the PCIJ, Series C 66/02, 2736, especially 274446: The
Danish Government hold the view that the declaration given on behalf of the Norwegian
Government through Norways lawful international representative, the Foreign Minister,
M Ihlen, constituted the recognition of the sovereignty of Denmark over the whole of
Greenland which according to international law, as it has always hitherto been accepted, is
binding on the Norwegian State. Refuting the allegations that a verbal declaration as such
cannot be binding, the Danish representative said: Almost all the quotations upon which
this Norwegian passage is based refer, as the Danish reply has shown, to treaties and not at
all to unilateral declarations like the one under discussion, ibid 274546, emphasis added.
57
See the reply by De Visscher of 20 January 1933 before the PCIJ, Publications of the PCIJ,
Series C 67, 3454 (my translation: This here, is not a treaty, it is an engagement undertaken
on behalf of the Norwegian government, a promise made on behalf of that Government.
The Danish Government took note of the promise made by Mr Ihlen to the Danish minister,
Mr Krag. It has done more, it has acted upon it, and it has done so immediately, in giving
its diplomatic approval to Norway in the Spitzbergen question. Daily international practice
55
56

96 A History of Promises
According to de Visscher the declaration therefore did not give rise to
a treaty but was a promise of a kind for which international practice was
said to offer numerous examples. De Visscher tried to overcome the precedent set by the arbitration award rendered by Baron Lambermont (see
text to note 8 above) by focusing on the precise wording used by the arbitrator, as the latter did not openly speak of a treaty. He hence continued
his pleadings as follows:
Il ne sagit pas davantage ici dune dclaration unilatrale non accepte ou de
simples intentions exprimes, et la citation qui a t faite ce sujet dun extrait
du baron Lambermont est sans pertinence. Il sagit bien ici dun cas o, pour
reprendre la formule du baron Lambermont laccord des volonts sest manifest par la promesse expresse de lune des Parties jointe lacceptation de
lautre.58

De Visscher hence stressed that the promise had been accepted by

Denmark, it was not a unilateral act non-accept. While true, Baron
Lambermont was quite obviously referring to a treaty when he chose his
words and spoke of an accord des volonts in which a promise had to have
been accepted an interpretation which De Visscher in his pleading had,
as cited, just rejected for the Ihlen declaration in refuting the Norwegian
attacks on the existence of oral treaties. At this point, however, it is less
important to consider whether or not De Visschers pleading was compelling when he argued that his interpretation of the Ihlen declaration as
being binding upon Norway, although it did not give rise to a treaty, was
compatible with the position assumed by Baron Lambermont. What is
instead to be kept in mind is that the nature of the declaration and how it
had become binding upon Norway were two questions which had been
expressly raised and addressed as such in some detail before the Court.
Norway responded by stressing that the declaration could not be binding upon it, as it was said to have been non-official and only of a pro
visional nature. It claimed that it in no way constituted a recognition of
an existing situation but was only a positive response to a notification
of a Danish future project in Greenland that would be dealt with at the
conference.59 The Court, faced with these assertions, first in clear words
refuted the Danish claim that the declaration constituted a recognition of
an existing Danish sovereignty attributable to Norway,60 by emphasising
provides us with constant examples of this kind of promises, undertakings having a binding
character without therefore having to assume the form of a treaty).
58
Ibid (my translation: This here is no longer a non-accepted unilateral declaration or
simple expression of intent, and the citation of a passage by Baron Lambermont which has
in this context been provided is without pertinence. This is clearly a case where, to use the
words of Baron Lambermont, the concurrence of wills is manifested by an express promise
of one of the parties met with the acceptance of the other).
59
See the pleadings of Gidel before the court on 12 December 1932, ibid 3193 et seq.
60
See PCIJ, Eastern Greenland (1933), n 51 above, Judgment, 69, para 3, where the Court is
absolutely clear in this respect: The declaration by M Ihlen has been relied on by Counsel

Legal Status of Eastern Greenland Case97

that it was couched in future terms, ie that Norway would not make (ne
fera pas)61 any difficulties in the settlement of this question.
The Court went on to consider:
whether the Ihlen declaration even if not constituting a definitive recognition
of Danish sovereignty did not constitute an engagement obliging Norway to
refrain from occupying any part of Greenland.62

For the PCIJ, the relevant Danish documents, which had preceded
the Danish Ministers dmarche, had made it clear that the question of
Norways interests in Spitzbergen (which Denmark would not oppose)
and the Danish interests in Greenland (which it wanted Norway not to
interfere with):
were regarded in Denmark as interdependent, and this interdependence
appears to be reflected also in M Ihlens minute of the interview. Even if this
interdependence which, in view of the affirmative reply of the Norwegian
Government, in whose name the Minister for Foreign Affairs was speaking,
would have created a bilateral engagement is not held to have been established, it could hardly be denied that what Denmark was asking of Norway
(not to make any difficulties in the settlement of the [Greenland] question)
was equivalent to what she was indicating her readiness to concede in the
Spitzbergen question (to refrain from opposing the wishes of Norway in regard
to the settlement of this question). The declaration which the Minister for
Foreign Affairs gave on July 22, 1919, on behalf of the Norwegian Government,
was definitely affirmative: I told the Danish Minister today that the Norwegian
Government would not make any difficulty in the settlement of this question.
The Court considers it beyond all dispute that a reply of this nature given
by the Minister of Foreign Affairs on behalf of his Government in response to
a request by the diplomatic representative of a Foreign Power, in regard to a
question falling within his province, is binding upon the country to which the
Minister belongs.63

The reply given by Mr Ihlen was therefore considered binding upon

Norway.64 The problem of this passage in the present context is that
it again leaves room for ambiguities and does not provide a clear legal
explanation or rationale for the bindingness of the declaration made,
instead it merely reiterates what has happened and then proclaims its
for Denmark as a recognition of an existing Danish sovereignty in Greenland. The Court
is unable to accept this point of view. A careful examination . . . shows that M Ihlen cannot
have meant to be giving then and there a definitive recognition of Danish sovereignty over
Greenland, and shows also that he cannot have been understood by the Danish Government
at the time as having done so.
61
This is the French wording, which the Court refers to as having been submitted by the
Norwegian government and which was not disputed by Demark, see ibid 69.
62
Ibid 69, para 3.
63
Ibid 7071.
64
The Eastern Greenland case has also been of importance for the question of how far a
Foreign Minister can bind his or her own country on the international plane irrespective of
national constitutional restraints, a question which is left aside here.

98 A History of Promises
legal conclusion as beyond all dispute. Yet, a dispute surely existed and
a perusal of literature on the topic shows that it continues to exist. Before
turning to the judgments perception by legal scholars, it is worth taking a closer look at what the Court has actually said (and what not) in
the paragraph just cited. Returning to the cited passage and breaking
it into smaller pieces, the Court found the following: in addressing the
question whether an interdependence had been established (the Court
already avoids a clear statement and omits speaking of a quid pro quo)
the PCIJ does not reach a conclusion. The answer it gives is merely that
if it had existed, it would have created a bilateral engagement. But what if it
had not? Here it gets interesting. The Court says that even if not, it could
hardly be denied that what Denmark was asking for was equivalent to
what she was ready to do. This is an accurate description of fact but what
does it mean in legal terms? That an interdependence if not established,
still clearly existed? Yet, if something clearly exists, it is quite clearly
established. Perhaps for the Court, Denmarks concession in a notdirectly-dependent yet related manner merely underlined the importance
of the declaration at hand, thereby discarding Norways interpretation of
it as being devoid of any legal meaning. But this is only speculation. The
pivotal sentence, however, is the following:
The Court considers it beyond all dispute that a reply of this nature given by the
Minister of Foreign Affairs on behalf of his Government in response to a request
by the diplomatic representative of a Foreign Power, in regard to a question
falling within his province, is binding upon the country to which the Minister
belongs.

As the Court deduced the bindingness of the declaration merely

by rephrasing the facts (we have the following elements: (1) a response,
(2) of this nature, (3) by a Minister of Foreign Affairs, (4) on behalf of his
government, (5) to a request by the diplomatic representative of a Foreign
Power, (6) in regard to a question falling within his province) but provided little or no hints as to the legal norm under which it was subsuming
this real life scenario (the Court neither used the words unilateral act or
assurance, nor did it speak of a treaty or an agreement in this passage),
it is hardly surprising that the assessment of the Courts judgment is anything but unanimous in this respect. The lacunae in the Courts assessment are especially remarkable and apparent in light of Judge Anzilottis
dissent which has left no room for doubt how he conceived the legal construction to be. For him, the Ihlen declaration together with the request
formed an oral agreement which was binding upon Norway, even though
the questions of Spitzbergen and Greenland in Anzilottis opinion did not
function as a quid pro quo for each other.65 The Court, however, did not
65
See PCIJ, Eastern Greenland (1933), n 51 above, Dissenting Opinion of Judge Anzilotti,
especially 8891.

Legal Status of Eastern Greenland Case99

use the word agreement but repeatedly talked of an undertaking and

made ample use of the word promise, for example:
The Court is unable to read into the words of the Ihlen declaration in the settlement of this question [ie the Greenland question] a condition which would
render the promise to refrain from making any difficulties inoperative should a
settlement not be reached. The promise was unconditional and definitive.66

Since these ambiguities exist within the ruling, the Eastern Greenland
case has been interpreted very differently. For some commentators it is
an important, if not even the central, case in the PCIJs jurisprudence on
the binding force of promises in international law. The editorial comment
written by James Garner in the judgments immediate aftermath in the
American Journal of International Law in 1933, bearing the telling title The
International Binding Force of Unilateral Oral Declarations positions
itself in the doctrinal debate that it helped to kindle with the following
words:
In the Eastern Greenland case the question involved was not whether an oral
treaty between members of the League was a valid one, but whether a unilateral oral declaration was binding on the party whose minister made it.67

And, that:
The conclusion which may be drawn from this decision is, first, that a unilateral
oral declaration in the nature of a promise, the facts of which are admitted by
the party making it, is internationally binding upon the latter party.68

Garner is hence sometimes referred to as being (amongst) the first international scholar(s) to have identified the PCIJs reliance on a promise as
creating binding obligations under international law.69 In a similar vein,
Verdross in his third edition of Vlkerrecht, published in 1955, in his section on unilateral acts, cites the Eastern Greenland case as proof of the existence of binding promises in international law.70 Jacqu also discusses the
judgment and its ambiguities and draws the conclusion that it was really
more the fear present in a large part of legal doctrine that a promises
binding force may be uncertain which led some to construe a contractual
bond in the Eastern Greenland case; the Ihlen declaration really seemed
to be a unilateral promise.71 According to Degan, the assignment of
66
PCIJ, Eastern Greenland (1933), n 51 above, Judgment, 73, emphasis added, see also ibid
7273: notwithstanding the undertaking of 22 July 1919, by which she promised to refrain
from making difficulties.
67
James W Garner, The International Binding Force of Unilateral Oral Declarations
(1933) 27 American J International Law 493, 49495.
68
Ibid 496.
69
See, eg Suy, Actes Unilatraux, n 26 above, 109; Rousseau, Droit international public, vol I,
n 33 above, 423, para 341.
70
Alfred Verdross, Vlkerrecht, 3rd edn (Vienna, 1955) 133.
71
Jacqu, Elments, n 34 above, 255: La promesse du Ministre Ihlen semble bien tre une
promesse unilatrale.

100 A History of Promises

[a] consensual character to the Ihlen Declaration seems to be a pure fiction
serving no useful purpose.72 For Guggenheim, on the other hand, the declaration is a treaty,73 and a number of commentators agree,74 often emphasising the fact that the declaration was termed a response to a request.75
Unlike most, Suy, who is amongst the treaty group in this case, is, however, again more precise in elaborating how this contractual nature is to be
construed.76 Following Anzilotti, he begins by highlighting that there was
no quid pro quo relationship between the Spitzbergen and the Eastern
Greenland questions: the Danish government did not oblige itself to do
anything, says Suy. For him, the response given by Ihlen was therefore
the acceptance of a proposition, that is the acceptance of an offer.77 Yet,
this understanding of the matter would create a clear contradiction. If
there was no quid pro quo and Denmark did not oblige itself to anything,
it cannot be deemed to have made an offer: Denmark certainly did not
offer to Norway that Norway (!) undertakes to make no difficulties in the
matter in the future. It can only be considered to have asked Norway to
assure what Denmark was trying to achieve and thereby to have asked
Norway to offer not to make any difficulties in the question, an offer which
it was from the outset willing to, and after Minister Ihlen had made it, in
this understanding subsequently did (tacitly)78 accept.
The Ihlen declaration, or more precisely the PCIJs judgment and its
dealing with the declaration, are therefor not only a primary example
of the gradual elaboration of the doctrine of promises, which the Court
undoubtedly helped to trigger, but also of the possibly difficult question
whether a declaration constitutes an unconditional promise or an offer
necessitating an acceptance.79 As seen, legal doctrine is split over the matter in the case at hand and the Court has left room for doubts as to how
it considered the legal obligation to have been brought about. The Ihlen
declaration thereby helped to elevate an until then largely undiscussed
question into full daylight, namely, what legal consequences an assurance, if found to be unilateral, might have, even though this question, as
this chapter has already shown, was not entirely new to the Court.
To recapitulate: unilateral declarations of state representatives made
before the PCIJ had been considered as binding in the Mavrommatis
Degan, Unilateral Act, n 31 above, 197.
See Paul Guggenheim, Trait de droit international public, 2nd edn (1967) vol 1, 138, 276.
74
See especially Suy, Actes Unilatraux, n 26 above, 12124; also Suzanne Bastid, Les traits
dans la vie internationale (Paris, 1985) 351.
75
See, eg Suy, Actes Unilatraux, n 26 above, 124; also Verdross and Simma, Universelles
Vlkerrecht, n 21 above, 429, s 669.
76
See Suy, Actes Unilatraux, n 26 above, 124 and, agreeing with Suy, Quadri, Cours gnral, n 4 above, 365.
77
Suy, Actes Unilatraux, n 26 above, 122.
78
Already for Grotius, Of War and Peace, n 7 above, vol II, ch XI, 720, para XIV, a precedent
Request shall be judged to subsist, and to have the Force of an Acceptance.
79
On this question, see p 228.
72
73

Assurances Leading up to the Second World War101

Jerusalem Concessions case, just as they had been in the Certain German
Interests in Upper Polish Silesia case. The (possibly) unilateral nature of
the minorities declarations was brushed over when the reason for such
an international instrument to be binding was not addressed in the
Advisory Opinion on the Albanian minorities schools. Adatci, who later
became the acting president of the PCIJ in the Eastern Greenland case,
when confronted with the assertion that the Lithuanian declaration was
unilateral had, on the other hand, expressly remarked that in his view the
engagements assumed thereby were international obligations, no matter
whether the declaration was a unilateral act or formed part of an inter
national treaty.80
The Court now, much as before, avoided an answer on the question by
leaving the evidently unclear classification of the legal relationship open
to doubt and by instead choosing to assume the bindingness of the declaration to be beyond all dispute. It did so after having discussed bilateral and multilateral agreements (in which the Ihlen declaration was not
included!) while refraining from utilising the words agreement, treaty
or quid pro quo, although these aspects were problematic and had been
debated before the Court, as well as explicitly addressed in Anzilottis
dissent. Not surprisingly, the judgment has thereby fuelled the debate on
promises, since by choosing the above-depicted approach of assuming
the bindingness to be beyond all dispute, without clearly construing the
relationship in contractual terms, the Court, if not held to have actually
considered a unilateral assurance as binding, comes close to implicitly
acknowledging what Adatci had already expressly declared: that the declaration, irrespective of whether it was to be characterised as unilateral or
formed part of a bilateral treaty, was internationally binding for its author.
At least with hindsight and in view of the cases that were to come, it is fair
to say the plot thickens.
F Assurances Leading up to the Second World War
The late 1930s were witness to a number of assurances which immediately predated the Second World War. France and Britain, for example, assured their military assistance to Poland in case of an attack on
31 March 1939, and so did Britain to Greece and Romania on 13 April
1939. But not only the (about to be) Allies made use of assurances, so did
the Nazi regime in an attempt to camouflage its plans; assurances all of
which were made only to be broken. When the war was over, the Charter
of the International Military Tribunal (IMT), which formed an integral part
of the London Agreement of 8 August 1945 between the United States,
80

See the text to n 48 above.

102 A History of Promises

the Soviet Union, Great Britain and France, listed three crimes as within
the jurisdiction of the IMT: (a) crimes against peace; (b) war crimes; and
(c) crimes against humanity. Crimes against peace were defined as
follows:
(a) CRIMES AGAINST PEACE: namely, planning, preparation, initiation or
waging of a war of aggression, or a war in violation of international treaties,
agreements or assurances, or participation in a common plan or conspiracy for
the accomplishment of any of the foregoing;81

The indictment against Hermann Goering et al in its Appendix C

lists Charges and Particulars of Violations of International Treaties,
Agreements, and Assurances Caused by the Defendants in the Course of
Planning, Preparing, and Initiating the Wars and includes seven charges
of violations of assurances, of which the first two (here cited as an example of all seven) read as follows:
XVI CHARGE: Violation of German Assurance given on 21 May 1935 that the
Inviolability and Integrity of the Federal State of Austria would be recognized.
PARTICULARS: In that Germany did, on or about 11 March 1938, at various
points and places along the German-Austria frontier, with a military force
and in violation of its solemn declaration and assurance, invade and annex to
Germany the territory of the Federal State of Austria.
...
XVIII CHARGE: Violation of German Assurances given on 30 January 1937, 28
April 1939, 26 August 1939, and 6 October 1939 to respect the neutrality and territorial inviolability of the Netherlands.
PARTICULARS: In that Germany, without warning, and without recourse to
peaceful means of settling any considered differences did, on or about 10 May
1940, with a military force and in violation of its solemn assurances, invade,
occupy, and attempt to subjugate the sovereign territory of the Netherlands.82

The assurances to respect the territorial integrity of various countries

were hence treated as binding commitments by the Allied Powers drawing up the Charter, as well as the prosecution in its indictment. Their
breach was considered to go as far as to entail individual criminal responsibility as a crime against peace.
Verdross, writing in 1955, therefore mentions the London Agreement as
the (second) example (besides the PCIJ in the Eastern Greenland case) where
81
Charter of the International Military Tribunal (Nuremberg Charter), Art 6(a).
International Military Tribunal, Trial of the Major War Criminals (1947) 11 (also available at
http://avalon.law.yale.edu/imt/imtconst.asp), emphasis added.
82
Nuremberg Trial Proceedings, vol 1, Indictment: Appendix C, 84 (also available at http://
avalon.law.yale.edu/imt/countc.asp). See also charges XIX (assurances given to Belgium), XX
(assurances given to Czechoslovakia), XXII (assurances given to Norway), XXIII (assurances
given to Luxembourg) and XXVI (assurance given to Yugoslavia), all included therein.

Assurances Leading up to the Second World War103

the bindingness of promises in international law had been assumed.83

Most commentators dealing with this case draw the same conclusion as
Verdross has done by listing the London Agreement as proof of the existence of promises in international law.84 Rubin, on the other hand, in his
highly critical account published shortly after the Nuclear Tests cases, considers the assurances made by Germany to merely underline an already
existing obligation and the indictment not to have drawn any direct legal
consequences from these pronouncements. The IMT, according to Rubin,
also did not give any weight to the assurances made.85 It should, however,
be noted at the outset that this argument does not address the fact that the
four Allied Powers, in drafting the statute, considered an infringement of
an assurance to be a possible crime against peace. This fact alone is noteworthy, independent of whether or not the tribunal subsequently found
an accused guilty of this offence in applying the statute and identifying a
unilaterally binding assurance to have been breached. Moreover, while it
is correct that the Tribunal did not draw any special consequences from
the German pledges, this does not mean that they were considered of
no legal value; the Tribunal merely stated that it considered neither the
various treaties (which, no doubt, included binding commitments) concluded by Germany with the countries it subsequently attacked to be of
special importance for the judgment nor the assurances it made, since
it had already concluded that the wars waged were wars of aggression.
The alternative charge for committing a crime against peace, ie waging a
war in violation of international treaties, agreements or assurances was
therefore of no further relevance for the judgment.86 The indictment, on
the other hand, can hardly be considered not to have drawn direct legal
Verdross, Vlkerrecht, n 70 above, 133.
See, eg Suy, Actes Unilatraux, n 26 above, 13132; Jacqu, Elments, n 34 above, 250;
Rousseau, Droit international public, vol I, n 33 above, 42526.
85
See Rubin, Unilateral Declarations, n 4 above, 34. Denying the precedential value of
this case, also Quadri, Cours gnral, n 4 above, 36668, according to whom the Charters
wording was at least ambiguous, as assurances could be understood as informal agreements. The German assurances, according to Quadri, indeed formed such informal agreements as they had been made in response to demands and requests of the interested states.
See, however, the critique by Jacqu, Elments, n 34 above, 25051, who (in this writers view
rightly) rejects Quadris construction as artificial, as Germany was neither acting to accept an
offer nor was it making one.
86
See the following introductory passage in the section Violation of International Treaties
in the judgment: The Charter defines as a crime the planning or waging of war that is a
war of aggression or a war in violation of international treaties. The Tribunal has decided
that certain of the defendants planned and waged aggressive wars against twelve nations,
and were therefore guilty of this series of crimes. This makes it unnecessary to discuss the
subject in further detail, or even to consider at any length the extent to which these aggressive wars were also wars in violation of international treaties, agreements or assurances,
International Military Tribunal, Trial of the Major War Criminals (1947) 216. See also the final
passage in this section: The Tribunal does not find it necessary to consider any of the other
treaties referred to in the Appendix, or the repeated agreements and assurances of her peaceful intentions entered into by Germany, ibid 218.
83
84

104 A History of Promises

consequences from the breach of the assurances made, as it included
the above-mentioned seven charges which are based precisely on such
infringements.
G Austrias Permanent Neutrality
The next case, or better example of state practice as no tribunal or court
was involved, that had numerous international legal scholars arguing
over whether a promise was involved, is Austrias declaration of permanent neutrality. The facts have been stated in numerous publications and
can briefly be summarised as follows:
To pave the way for the conclusion of the so called State Treaty
through which Austria was to regain its full sovereignty after the Second
World War, Austria had indicated its willingness to undertake an obligation of permanent neutrality similar to the status of Switzerland. This
aspect was apparently of particular importance to the Soviet Union and
after some conversations on the matter between the two governments, a
Memorandum was signed in Moscow on 15 April 1955 in which, inter
alia, the following was agreed upon:
1. In the sense of the declaration already given by Austria at the conference in
Berlin in 1954 to join no military alliances and to permit no military bases on its
territory, the Austrian Federal Government will make a declaration in a form which
will obligate Austria internationally to practice in perpetuity a neutrality of the
type maintained by Switzerland.
2. The Austrian Federal Government will submit this Austrian declaration in accordance with the terms of the Federal Constitution to the Austrian
Parliament for decision immediately after ratification of the State Treaty.
3. The Federal Government will take all suitable steps to obtain international recognition for the declaration confirmed by the Austrian Parliament.
...
II. The Deputy Chairman of the Council of Ministers, VM Molotov and
AI Mikhoyan, made the following declaration in the name of the Soviet
Government with regard to the declarations of the Austrian Government delegation:
1. The Soviet Government is prepared to sign the Austrian State Treaty without
delay.
...
4. The Soviet Government is prepared to recognize the declaration concerning the
neutrality of Austria.87
87
This English translation is taken from (1955) 49 American J International Law (Supplement)
191, emphasis added.

Austrias Permanent Neutrality105

The State Treaty for the Re-establishment of an Independent and

Democratic Austria was signed at Vienna on 15 May 1955 and entered
into force on 27 July of the same year. A few months later, Austria passed a
Constitutional Federal Statute, which in its principal Article I proclaimed:
1. For the purpose of the permanent maintenance of its external independence
and for the purpose of the inviolability of its territory Austria, of its own free
will, declares herewith its permanent neutrality.
2. In order to secure these purposes Austria will never in the future accede to
any military alliances nor permit the establishment of foreign states on its territory.88

The above decision as adopted by Austria was subsequently notified to

other governments, asking them to recognise the proclaimed neutralisation, which a majority of governments, including the four Allied Powers
as well as Germany and Switzerland, expressly did.89
In an article on the matter, Rodolfo De Nova sketched the legal backdrop, the law on neutrality, against which the Austrian conduct was
judged by most commentators at the time.90 As permanent neutrality, ie neutralisation, was understood as encompassing obligations not
only for the neutralising state but also for states dealing with it, which
had to respect and in no way violate this neutrality, neutralisation was
believed to be achievable only by means of concluding a treaty; a unilateral neutralisation was largely considered legally impossible.91 Hence
many (according to De Nova in 1958, indeed, the majority of) authors,
understood the Austrian neutrality to have been brought about through a
treaty. This treaty here there is general agreement was not the Moscow
Memorandum itself; while some consider it to impose no legal obligations
as they view it not as a binding agreement but merely an aide-mmoire,
those who assume its legal bindingness accept the fact that the obligations
flowing therefrom merely compelled Austria to work towards its new status by subsequently making a declaration in a form which will obligate
Austria internationally.
For some scholars, Austria has fulfilled this pledge by in fact concluding
a treaty with each and every state that recognised its declared permanent
neutrality: The note by which it informed the respective governments
of its decision to adopt a permanent neutrality is seen as an offer, which
88
Translation taken from Josef Kunz, Austrias Permanent Neutrality (1956) 50 American
J International Law 418, 420.
89
See Eduard Reut-Nicolussi, Die sterreichische Neutralittserklrung vom 26 Oktober
1955 (1956) Internationales Recht und Diplomatie 15, 1920, citing the wording of various recognitions received by Austria; see also De Nova, Neutralisation sterreichs, n 30 above,
307.
90
De Nova, Neutralisation sterreichs, n 30 above, 298320.
91
For this position see, eg Kunz, Austrias Permanent Neutrality, n 30 above, 418 and the
influential publication of Karl Strupp, Neutralisation, Befriedung, Entmilitarisierung (1933) 176.

106 A History of Promises

the addressees are said to have accepted when they recognised it.92 With
this opinion, the fact that the memorandum which sketched out the pro
cedure as envisaged by its signatories spoke of a declaration . . . which
will obligate Austria internationally that was only subsequently to be
recognised by as many states as possible, is overcome by declaring the
memorandum to be badly drafted and the use of the word recognition
to be inappropriate.93
Other commentators apparently have more trust in the ability of
states legal departments and their use of precise legal terminology. ReutNicolussi in his analysis goes as far as simply dismissing any treaty construction as absurd.94 Since he agrees that a neutralisation can never be
unilaterally assumed, he draws the conclusion that Austrias notification
could and did not alter the legal landscape in any way; there was consequently also nothing for other governments to recognise and Austria
was in no way internationally bound through its actions to act as a neutral state. While Reut-Nicolussis assessment more adequately reflects the
actions undertaken by Austria and the recognising governments when it
rejects viewing them as conclusions of a treaty, it at the same time hinders
Austria from achieving what it had openly claimed its intentions to be,
ie to become internationally bound by its declaration. The reason to do
so is the already depicted assumption that permanent neutrality, seen as
imposing obligations on third parties, cannot be achieved unilaterally. A
third group of commentators, however, agree with Reut-Nicolussi only
in rejecting the treaty construction, which, as Jacqu says, would indeed
push consensualism in international law to an extreme, as neither Austria
nor the states which recognised its neutrality felt they were becoming parties to a contract.95 Venturini also considers it as artificial to attribute the
meaning of offer and acceptance to acts of a different nature.96 A memorandum prepared by the US Office of the Legal Adviser before the United
States answered Austrias request for recognition strengthens this position, as it shows that the United States was well aware of the legal terminology employed:
92
See Coppini ((1957) Revue de Droit Internationale, Sottile 18) as cited by De Nova,
Neutralisation sterreichs, n 30 above, 309; Quadri, Cours gnral, n 4 above, 365 and
also Degan, Unilateral Act, n 31 above, 202, for whom, in the end, the constitutional law
was just an offer which obtained its effects in international law by its acceptance. Some
further references for this position are provided by De Nova, Neutralisation sterreichs,
n 30 above, notes 2 and 33.
93
Criticising the Memorandum, Kunz, Austrias Permanent Neutrality, n 88 above, 421;
labelling the use of the word recognition inappropri, see Coppini as cited by De Nova,
Neutralisation sterreichs, n 30 above, 309. Similarly, Quadri, Cours gnral, n 4 above,
365: la dclaration fut suivie par des actes quon a appels tort de reconnaissance.
94
Reut-Nicolussi, sterreichische Neutralittserklrung, n 89 above, 21 (abwegig).
95
Jacqu, Elments, n 34 above, 251.
96
Venturini, Actes Unilatraux, n 50 above, 405, attribution artificieuse dune valeur
doffre ou dacceptation des actes ou attitudes ayant une autre nature.

Austrias Permanent Neutrality107

In its response to the note from the Austrian Government announcing its neutrality [Note: not offering!], the United States will state that it has taken due
cognizance of that constitutional law and recognizes the perpetual neutrality of
Austria as defined therein. The literal definition of the word recognize is commonly
accepted in terms of to act or take note of. Insofar as it relates to an official acknowledgement of an international status such as neutrality as declared by a sovereign
government it thus requires the recognizing State to refrain from taking those
actions with regard to the other State which might violate the accepted concepts
of a neutral State in the international community. This, then, is the extent of the
obligation assumed by the United States in recognizing the Austrian announcement of its neutrality.97

There is little reason to believe that other governments would have

misunderstood their recognition to form the acceptance to an allegedly
proposed treaty. Since Austria intended to become unilaterally bound by
its declaration as a permanently neutral state, its declaration is not denied
this effect by this third group of authors. The notification to various countries is understood to have elevated a domestic decision of neutrality onto
the international plane,98 forming a binding promise under international
law. This promise of neutralisation is necessarily different in its effect
from a neutralisation achieved through a treaty, since a unilateral act can
indeed only establish duties for its declarant and not force any unconsenting third state to respect this neutrality,99 which is precisely why Austria
is believed to have sought recognition from third states, as it wanted to
ensure that its permanent neutrality would henceforth be internationally respected. In this opinion, which has the merits of closely reflecting
not only what the Memorandums signatories had said they intended to
happen (ie have Austria make an internationally binding declaration which
subsequently is to be recognised) but also the legal terminology later
actually employed by the acting states, we are therefore in the presence
of two unilateral acts: a promise, obliging Austria and Austria only, to
remain neutral, followed by an act of recognition by which this neutrality is accepted by the recognising state, thereby compelling the latter in
future to respect this status. Although there is little intellectual strength in
mere numbers, this assessment, once discarded as an isolated opinion,100
has by now found ample support.101
97
Attorney, Office of the Legal Adviser (Wehmeyer) to the Legal Adviser (Phleger),
Memorandum, Nature of Austrias Neutrality and Legal Implications of United States
Response to the Neutrality Declaration, 16 November 1955, MS Department of State, file
663.0021/11-1655, cited in Marjorie M Whiteman and Green Haywood Hackworth, Digest of
International Law (Washington, 1963) vol 1, 350, emphasis added.
98
On this aspect Suy, Actes Unilatraux, n 26 above, 134.
99
See especially De Nova, Neutralisation sterreichs, n 30 above, 299 et seq.
100
Quadri, Cours gnral, n 4 above, 365.
101
See De Nova, Neutralisation sterreichs, n 30 above, 30511; Jacqu, Elments,
n 34 above, 251; Felix Ermacora, 20 Jahre sterreichische Neutralitt, 2nd edn (1975) 75; Suy, Actes
Unilatraux, n 26 above, 13335; Venturini, Actes Unilatraux, n 50 above, 405; Karl Zemanek,

108 A History of Promises

H Egypts Declaration on the Suez Canal
The Egyptian declaration on the Suez Canal of 24 April 1957 is the next,
again disputed, example of a binding assurance in international law.102
It was made at the end of the Suez Crisis which began to unfold after
Egypt decided to nationalise the Suez Canal Company on 26 July 1956.103
Its doing so prompted a protest by the United States, France and Britain,
which claimed the nationalisation of the Canal Company would infringe
upon the international status of the Canal itself. Two international conferences in London followed in an attempt to come up with a new international system to operate the canal, but both met with strong opposition
from Egypt, which refused to attend. The Security Council, seized in the
matter by Britain and France, finally intervened by issuing a unanimous
Resolution that listed a number of requirements which were to be fulfilled in order to reach a final settlement of the dispute.104 The situation,
however, escalated into a military confrontation when Israel moved into
the Gaza strip and the Sinai-Peninsula on 29 October 1956, quickly followed by an invasion of British and French troops into Egypt in order to
occupy the Canal. Under the political pressure of the General Assembly105
and the agreement that a UN emergency force would be installed in the
region, the invasion came to an end only a few months later, with British
and French troops withdrawing in December and the last Israeli soldiers
in March the following year. Parallel to the reopening of the Canal at the
end of April 1957, Egypt issued the declaration which is of interest here.
Unlike some of the other declarations mentioned, the written Egyptian
declaration included a rather clear statement as to its intended legal bindingness as an international instrument, leaving little room to interpret it
as merely a political pledge. Its final paragraph reads:
This Declaration, with the obligations therein, constitutes an international instrument and will be deposited and registered with the Secretariat of the United
Nations.106
Neutral Austria in the United Nations (1961) 15 International Organization 408, 409. See also
Rudolf L Bindschedler, Die Neutralitt im modernen Vlkerrecht (1956) 17 Zeitschrift fr auslndisches ffentliches Recht und Vlkerrecht 1, 4, according to whom a permanent neutrality can
be established through a unilateral act. Contra, besides those authors mentioned at n 92 above,
also Verdross, for whom the obligation for Austria has arisen only through the acceptance of
(or acquiescence in) its notification, which was not a promise as it asked for recognition, Alfred
Verdross, Die immerwhrende Neutralitt sterreichs, 4th edn (Vienna, 1977) 33.
102
For the complete declaration, see UN Doc General Assembly, A/3576 (also S/3818) or
(1957) 51 American J International Law 673.
103
For a more detailed historical account, see, eg Jos A Obieta Chalbaud, The International
Status of the Suez Canal, 2nd edn (The Hague, 1970) ch I, especially 18 et seq; Yves van der
Mensbrugghe, Les garanties de la libert de navigation dans le canal de Suez (Paris, 1964) 87.
104
See UN Security Council Res S/3675.
105
See UNGA Res A/RES/997 ES-I, A/RES/999 ES-I and A/RES/1002 ES-I.
106
See UNGA Doc A/3576 (= UN Doc S/3818) 5.

Egypts Declaration on the Suez Canal109

The declaration was thereafter indeed deposited and registered with the
Secretariat of the United Nations under Article 102 of the UN Charter and
published in the United Nations Treaty Series.107 Yet, there is hardly any
doubt that the declaration did not constitute a treaty, as it was unilaterally made without the involvement or acceptance, let alone the signature
or ratification, of any other nation. As seen in the briefly sketched history leading up to the military invention, other countries were quite interested in and had actually demanded to work out a multilateral agreement
which would install an international system for the Canal, an agreement
that Egypt tried to avoid by making its unilateral move. The declarations
unilateral character is hence widely accepted as will be seen, for some
countries it was the very reason they rejected it and has been acknowledged by the UN Secretary-General, who during the press conference
announcing the declarations registration, while clearly emphasising his
purely administrative character in the procedure, pointed out that the
word agreement in Article 102 of the UN Charter could in his view be
understood to include unilateral engagements.108 Critics of the declaration
as an example of a binding promise in international law, therefore, have
little trouble in qualifying it as a unilateral act, but raise doubts whether
this unilateral declaration did in fact give rise to new legal obligations for
Egypt. Here, two different problems are seen.
In his chapter on promises, Suy includes the Egyptian declaration in the
section on what he termed pseudo-promises. A pseudo-promise is said
to be not a real promise, as through it a state merely promises its addressee
to do what it is already obliged to do.109 In other words, while dressed as a
real promise, such a declaration merely assures to comply with an already
existing obligation and, in doing so, it at most recognises this obligation
but it cannot newly create it. A convincing example for such a declaration,
which might be labelled a pseudo-promise, is the pledge made by South
107
See United Nations Treaty Series, Registration No 3821, 1957, vol 265, 299. UN Charter,
Art 102 provides that: Every treaty and every international agreement entered into by any
Member of the United Nations after the present Charter comes into force shall as soon as
possible be registered with the Secretariat and published by it.
108
For the Secretary-Generals position during his press conference, see Jacques Dehaussy,
La dclaration gyptienne de 1957 sur le canal de Suez (1960) VI Annuaire Francais de droit
international 169, 172 and Mensbrugghe, Canal de Suez, n 103 above, 299. For more on the
registration of unilateral acts, see p 246.
109
Suy, Actes Unilatraux, n 26 above, 14041. But it should in this respect be kept in mind
that a state is free to assume the very same legal obligation more than once and through
more than one binding legal mechanism. Just as it can conclude numerous treaties including
the same obligation (ie peaceful settlement of disputes, the insurance of human rights, free
trade, etc), it is able to make numerous promises which all assure the same line of action,
even vis--vis the same addressee. If the declarant state is already under an obligation vis-vis the addressee to do what it now assures will be done, this declaration will, however, and
without indications to the contrary, rather be understood and interpreted not as an expression of will to undertake a new and separately valid promise but as the declarants pledge
simply to comply with its existing obligations.

110 A History of Promises

Africa, in which it announced its decision to continue to administer the
territory of South West Africa in accordance with the mandate received
from the League of Nations, even after the latter had ceased to exist. As
the ICJ made very clear, the mandate given to South Africa was still in
force even though the League of Nations no longer existed.110 South Africa
was therefore not free to choose to assume or decline the mandates obligations vis--vis the international community and by declaring to do as it
did, it was hence not creating a new legal obligation, but merely confirming and thereby recognising an already existing one.111
Returning to the Egyptian declaration, it too is classified by some commentators as a pseudo-promise, since it is said merely to reiterate the
Constantinople Convention,112 which the declaration according to its own
first paragraph expressly reaffirms. The question which necessitates a
closer look is, hence, whether or not the Egyptian declaration really only
reiterates the obligations already contained in that Convention. The latter
was concluded on 2 Mach 1888, between Great Britain, Germany, AustriaHungary, Spain, France, Italy, the Netherlands, Russia and Turkey (the
Ottoman Empire) and assures the free navigation of the Suez Maritime
Canal.113 Its Article 1(1) reads: The Suez Maritime Canal shall always be
free and open, in time of war as in time of peace, to every vessel of commerce or of war, without distinction of flag. And it is this free passage
through the Canal which is at the centre of the Conventions attention,
which in its 16 Articles, inter alia, imposes the principle of equality in the
use of the Canal and various prohibitions not to interfere with its free use,
and establishes rules to be applied during war and to vessels of war travelling through the Canal. The declaration, on the other hand, only begins
by reaffirming the terms and spirit of the Constantinople Convention
which Egypt pledges it will continue to respect, observe and implement.
But it cannot be reduced to merely having this function; if the declaration is, figuratively speaking, placed side by side with the Constantinople
Convention of 2 March 1888, the result is not that of a clear confirmatory
match, as the declaration covers aspects not included in the Convention.
While the declarations second Article adds nothing new when it again
pledges observance of the Constantinople Convention and to abide by the
United Nations Charter, and Article 3(a) is also drafted in explicit refer110
See ICJ, South West Africa (Ethiopia and Liberia v South Africa), Preliminary Objections,
Judgment, [1962] ICJ Rep 319, 334; see also ICJ, Legal Consequences for States of the Continued
Presence of South Africa in Namibia (South West Africa) notwithstanding Security Council
Resolution 276 (1970), Advisory Opinion of 21 June 1971, [1971] ICJ Rep 16, 37.
111
ICJ, South West Africa (1962), Preliminary Objections, Judgment, n 110 above, 340.
112
Besides Suy, for this position see also Quadri, Cours gnral, n 4 above, 372.
113
For the text of the Convention see, eg US Department of State, The Suez Canal Problem, 26
July22 September 1956; a Documentary Publication (Washington, 1956) 1620; Mensbrugghe,
Canal de Suez, n 103 above, Annex IV, 38992 (French); Obieta Chalbaud, Suez Canal, n 103
above, Appendix B, 14649.

Egypts Declaration on the Suez Canal111

ence to the Convention in its assurance of free and uninterrupted navigation, the declaration, in the following, deviates from already existing
obligations. Besides some rather vague declarations of intent, it includes
the following:
(i) the assurance that the tolls levied will be increased by no more than
1 per cent per 12 months unless otherwise agreed or failing such
agreement as adjudicated by arbitration, the procedure of which is
stipulated later on in the declaration (Article 3(b));
(ii) the assurance that the Canal is maintained and developed in accordance with the progressive requirements of modern navigation (Article
3(c));
(iii) the assurance that an autonomous Suez Canal Authority will operate
the Canal (Article 4);
(iv) the assurance than any future alteration of the Canal Code affecting the
principles included in the Declaration can be challenged according to
the arbitration proceeding established therein (Article 6);
(v) the assurance that in case of a dispute with the Canal Authority an
arbitration tribunal is available, composed of one nominee of the complaining party, one of the Authority and a third to be chosen by both,
or, in case of disagreement, as determined by the President of the ICJ
with the decisions made by a majority of its members being binding
upon the parties (Article 7 (bd));
(vi) 
the assurance that differences arising between the parties to the
Constantinople Convention will be referred to the ICJ and Egypt will
take the necessary steps to accept the compulsory jurisdiction of the
Court (Article 9 (b)).114
As none of these obligations is to be found in the Constantinople
Convention it is not convincing to label the declaration a mere recognition
of Egypts already existing obligations under it.115 The declaration instead
goes further in its promises and it does so in an attempt to appease the
international community, and especially the Security Council, which in
the already mentioned Resolution had included the fixing of tolls and
charges by agreement between Egypt and the users within the requirements to be fulfilled, just as it had requested unresolved affairs to be settled by arbitration.116
Rubin dismisses the declaration on a different ground when he turns to
the reaction of states to the declaration. Although he admits that Egypts
114
Egypt accepted the compulsory jurisdiction of the ICJ in relation to disputes that might
arise under this declarations paragraph on 18 July 1957, see UNGA Doc A/3576/Add.1
(= S/3818/Add.1).
115
See also Dehaussy, Dclaration gyptienne, n 108 above, 171, 17374; Mensbrugghe,
Canal de Suez, n 103 above, 299; also Degan, Unilateral Act, n 31 above, 19495, 20304.
116
See UN Security Council Doc S/3675.

112 A History of Promises

friends supported the declaration, he emphasises that despite the fact that
[t]he Suez Canal users in fact continued to use the Canal, and Egypt continued to administer the Canal according to the terms of the unilateral
declaration, . . . the users claimed that their actions were not to be construed as an acceptance of the legal force of the Egyptian declaration.117
In turning to the reactions of other states, it would of course be wrong to
look for an act of acceptance, as a unilateral promise neither invites an
acceptance nor need it, technically speaking, be accepted by its addressee.
Yet, in order to be a fruitful example of state practice, it is true that the declaration needs to have been accepted as binding for Egypt. While many
states seem to have abstained from any express reaction, those present
in the Security Council at the time came together to discuss the meaning
of Egypts declaration. Here, Rubins assertion that (all) the Suez Canal
users claimed their actions were not to be construed as an acceptance is
too broad, as the situation appears to have been more complex; reactions
were mixed and, in analysing them for present purposes, it is essential
to hold apart whether a state rejected Egypts declaration on account of
being displeased with the declarations content, or, and more importantly
to us, on account of the declarations unilateral nature. The Chinese representative in the Security Council had in very clear words identified that
the questions raised by the Egyptian declaration fell within these two distinct categories:
The first relates to a unilateral declaration per se. One may ask how binding is
it? How long can its terms be relied on? and what exactly does registration with
the United Nations do to such an instrument? The other category of questions
concerns the contents of this Declaration.118

While China refrained from answering the questions pertaining to the

first category mentioned, other members did not, albeit being of a very
different opinion on the declarations legal effect. Frances representative
in very clear words rejected the declaration not only for its content, but
also for its form, and voiced its countrys concerns especially about the
revocability and modifiability of a unilateral assurance:
What, then, is the real value of this declaration at the present time? We are told
that it constitutes an international instrument which will be deposited and
registered with the Secretariat of the United Nations: however, a unilateral
declaration, even if registered, obviously cannot be anything more than a unilateral act, and we must draw the conclusion from these findings that just as the
Declaration was issued unilaterally, it can be amended or annulled in the same
manner.119
Rubin, Unilateral Declarations, n 4 above, 7.
Representative of China, UN S/PV.777, 26 April 1957, 14, para 59.
119
UN S/PV.776, 26 April 1957, 11, para 59, also printed in Alexandre-Charles Kiss,
Rpertoire de la pratique Francaise en matire de droit international public (Paris, 1962) vol I, 617
18, see also the following comment made by France: It is the nature of the declaration rather
117
118

Egypts Declaration on the Suez Canal113

This position was shared in full by the United Kingdom,120 just as Cuba
stressed that from its point of view:
Declarations made by Governments last only as long as the Governments themselves, whereas treaties, ratified by parliaments in accordance with constitutional provisions, are obligatory for the entire period specified in the treaties.121

While Australia also regretted not having heard a firmer undertaking of

a permanent nature,122 the representative of Columbia began by pointing
out that the declaration, on account of its unilaterality, was indeed open to
amendments by Egypt. This, however, was not to be true for the passages
in which Egypt undertook the obligation to submit any dispute to arbitration and where it undertook to assume the compulsory jurisdiction of the
ICJ; since these obligations were regularly assumed unilaterally, they (in
the eyes of Columbia) were irrevocable undertakings.123 The representative of the Philippines, while regarding it as an interim measure, stressed
the binding nature of the declaration which could be:
lawfully . . . invoked by a user of the Canal, before any United Nations organ,
as binding on the Egyptian Government from the moment of its binding acceptance for registration by the Secretary-General.124

Iraq, too, was particularly pleased to see that the Egyptian Government
considered its Declaration as a solemn undertaking and a binding international instrument.125 The representative of the Soviet Union considered the question whether the declaration could be binding despite its
unilateral character, to be settled after Egypt had itself declared it to
be an international instrument, when asking for its registration with the
UN Secretary-General, who had acceded to Egypts request.126 Sweden,
in a similar vein, accepted the declaration as an internationally binding
instrument.127 The United States said it was not fully satisfied with the
declarations content, but decided to refrain from any final judgment until
the regime had been tried out in practice. Nothing regarding the declarations unilateral character was mentioned as troubling by the United
States.128
than its content which makes it so serious. It is impossible to conceive of a unilateral settlement of the Suez Canal problem; whether or not it was registered with the United Nations
Secretariat, Egypt would always be free to revoke it, UN S/PV.778, 20 May 1957, 8.
120
Representative of the United Kingdom, UN S/PV.777, 26 April 1957, 1819, para 86 and
20, paras 9293; again at UN S/PV.778, 20 May 1957, 18, paras 114 and 118.
121
Representative of Cuba, UN S/PV.776, 26 April 1957, 12, para 65.
122
Representative of Australia, UN S/PV.777, 26 April 1957, 3, para 6.
123
Representative of Columbia, UN S/PV.776, 26 April 1957, 1516, paras 8485 and 17,
para 92.
124
Representative of the Phillipines, UN S/PV.776, 26 April 1957, 14, para 73.
125
Representative of Iraq, UN S/PV.777, 26 April 1957, 8, para 25.
126
Representative of the USSR, UN S/PV.777, 26 April 1957, 12, paras 4445.
127
Representative of Sweden, UN S/PV.777, 26 April 1957, 15, para 65.
128
Representative of the United States, UN S/PV.776, 26 April 1957, 3, para 11.

114 A History of Promises

The case is of value in the present context, as it includes a countrys
very clear attempt to bind itself not through a bilateral but through a unilateral act. Egypt not only expressly declared its written declaration with
the obligations contained therein to be an international instrument but,
in addition, chose to add a layer of formality by depositing it with the UN
Secretary-General. In choosing to do so in a matter which was already in
the spotlight of international attention after the Suez Crisis, Egypt presented the core question relating to all unilateral promises to the international community, ie (and requoting the Chinese representative) How
binding are they? The answers provided by the members of the Security
Council range from not binding at all and freely revocable (United
Kingdom, France, Australia, Cuba) to binding (Philippines, Iraq, Soviet
Union, Sweden) and in parts even irrevocable (Columbia).
These different attitudes notwithstanding, and besides the fact that the
declaration was felt by a majority of users not to fully comply with the
six requirements included in the Security Council Resolution, it seems to
have long outlived the temporal character that had been attributed to it
by many states at the time.129
I North Sea Continental Shelf Cases
The facts giving rise to the North Sea Continental Shelf cases of 1969 find
the legal scholar searching in vain for a declaration that was worth much
discussion on whether it was a promise under international law. It is nevertheless of some interest on account of the fact that a unilateral commitment by Germany had been alleged by Denmark and the Netherlands and
was briefly dealt with by the ICJ. Since Article 6 of the Geneva Convention
on the Continental Shelf, and its so called equidistance method was, as
such, not directly applicable to the Federal Republic of Germany for the
simple reason that the latter had signed but never ratified the Convention,
the Netherlands and Denmark, here in the words of the Court, argued that:
the Convention, or the rgime of the Convention, and in particular of Article 6,
has become binding on the Federal Republic in another way namely because,
by conduct, by public statements and proclamations, and in other ways, the Republic
has unilaterally assumed the obligations of the Convention; or has manifested its
acceptance of the conventional rgime; or has recognized it as being generally applicable to the delimitation of continental shelf areas. It has also been
suggested that the Federal Republic had held itself out as so assuming, accepting or recognizing, in such a manner as to cause other States, and in particular
Denmark and the Netherlands, to rely on the attitude thus taken up.130
129
See Dehaussy, Dclaration gyptienne, n 108 above, 18283, also Mensbrugghe, Canal
de Suez, n 103 above, 30304.
130
ICJ, North Sea Continental Shelf, Judgment of 20 February 1969, [1969] ICJ Rep 3, 25, para
27, emphasis added.

North Sea Continental Shelf Cases115

The Court, however, advised caution on these very broad allegations as

follows:
As regards these contentions, it is clear that only a very definite, very consistent
course of conduct on the part of a State in the situation of the Federal Republic
could justify the Court in upholding them; and, if this had existed that is to
say if there had been a real intention to manifest acceptance or recognition of
the applicability of the conventional rgime then it must be asked why it was
that the Federal Republic did not take the obvious step of giving expression to
this readiness by simply ratifying the Convention. In principle, when a number
of States, including the one whose conduct is invoked, and those invoking it,
have drawn up a convention specifically providing for a particular method by
which the intention to become bound by the rgime of the convention is to be
manifested namely by the carrying out of certain prescribed formalities (ratification, accession), it is not lightly to be presumed that a State which has not
carried out these formalities, though at all times fully able and entitled to do so,
has nevertheless somehow become bound in another way.131

The Court drew the following conclusion:

Having regard to these considerations of principle, it appears to the Court that
only the existence of a situation of estoppel could suffice to lend substance to this contention that is to say if the Federal Republic were now precluded from denying
the applicability of the conventional rgime, by reason of past conduct, declarations, etc, which not only clearly and consistently evinced acceptance of that
rgime, but also caused Denmark or the Netherlands, in reliance on such conduct, detrimentally to change position or suffer some prejudice. Of this there is
no evidence whatever in the present case.132

The ICJ continued by saying that it was therefore not necessary to subject each and every act on which Denmark had relied to full scrutiny, and
after a brief enumeration of these acts dismissed them, as in the result it
appears to the Court that none of the elements invoked is decisive; each is
ultimately negative or inconclusive; all are capable of varying interpretations or explanations.133
Judge Ammoun, in his separate opinion, criticised the fact that the
Court had as quoted assumed only the existence of a situation of estoppel
could suffice to lend substance to this contention. In his opinion:
The Judgment does not take into account a well-settled doctrine that a State
may be bound by a unilateral act.134

As proof of the well-settled doctrine referred to, Ammoun cites the

introductory and final page of the section on the obligatory character of
unilateral promises in Eric Suys book. Based on this doctrine, he dedicates
Ibid 25, para 28.
Ibid 26, para 30, emphasis added.
133
Ibid 26, para 32.
134
ICJ, North Sea Continental Shelf, Separate Opinion Ammoun, [1969] ICJ Rep 101, 120,
para 21.
131
132

116 A History of Promises

a very short passage in his separate opinion to analysing an announcement made by Germany, which is, however, quickly rejected and found to
be irrelevant, since its reference was not even to the Conventions section
on the equidistance principle. Judge Ammoun then turns to acquiescence
by conduct (which he also rejects in its application to the case at hand),
and hence away from unilateral promises in international law.135
These passages within the judgment and the separate opinion show
that the problematique of a unilateral commitment was by now not only on
the parties but also on at least one of the ICJs judges radar at the time.
It is also noteworthy for the Courts position not to lightly presume the
interpretation of a states conduct as signalling the acceptance of a legal
rgime in the light of it not having fulfilled the formal acceptance procedure, which it had helped to put in place by participating in the elaboration of a treaty on the matter. The fact that the majority of the Court chose
not to take into account the allegedly well-settled doctrine of unilateral
acts as criticised by Judge Ammoun should, however, not be overstated in
this context. Whether or not the doctrine could be characterised as wellsettled at the time is doubtful, but not ultimately decisive for the question why the majority chose not to address it. After all, even for Ammoun
there was very little evidence of a possible unilateral commitment in the
case at hand, evidence that he himself refuted very quickly. Also, by moving straight to estoppel, the Court brushed off the table not only unilateral
assurances, but also express recognition and recognition by conduct and
mere silence in the form of acquiescence, surely more as non-applicable to
the facts in the present case (as Ammoun indeed concludes in his separate
opinion) rather than as generally non-existent. The acts before the Court
simply included none which could be considered as including an assurance worthy of the name; they were instead, in the words of the Court,
ultimately negative or inconclusive.
J Nuclear Tests Cases
The ICJ delivered its judgment in the Nuclear Tests cases on 20 December
1974. It is a judgment which has become reasonably well known, yet not for
the Courts opinion on whether atmospheric nuclear testing as conducted
by France in the region of French Polynesia at the time was legal. This not
only legally interesting, but in addition politically explosive, question was
not answered in the Courts opinion, which has therefore been labelled
one of judicial avoidance and has met with considerable criticism. While
only producing a mouse of a decision136 as to the actual questions before
On acquiescence see chapter one, n 15.
Thomas M Franck, Word Made Law: the Decision of the International Court of Justice
in the Nuclear Test Cases (1975) 69 American J International Law 612.
135
136

Nuclear Tests Cases117

it, the Court nevertheless, and en route to its conclusion that there was no
need for any further adjudication, issued the landmark decision on the
bindingness of unilateral assurances, ie promises in international law. For
Thomas Frank, it thereby demonstrated that cases need not have monumental outcomes to make monumental law.137 Others, however, seem to
interpret the ruling more as a monumental mistake. However, the bulk of
criticism is not directed at the passage in the opinion which establishes
the general bindingness of unilateral assurances. But before continuing
any further in assessing the judgments impact and importance, the facts
and the precise content of the decision need to be recalled.
(i) Facts of the Case and the Judgment Delivered by the Court
Australia and New Zealand both instituted (what remained separate
though largely identical) proceedings against France in respect of a dispute concerning the holding of French atmospheric nuclear weapon tests
in the Pacific Ocean.138 Since France considered the Court to be manifestly
not competent139 to adjudicate in the matter, it abstained from taking part
in the proceedings, merely asking the Court through a letter to be so good
as to order that the case be removed from the list.140 The ICJ, however, did
not heed this request and the case remained before it, first resulting in an
order, in which the Court asked France as an interim measure to avoid
nuclear tests causing the deposit of radio-active fall-out on Australian
[respectively New Zealand and in addition the Cook Islands, Niues and
the Tokelan Islands] territory.141 France, nevertheless, conducted two
further series of atmospheric tests in the Pacific, before the Court rendered
its judgment. Requested by Australia:
to adjudge and declare the carrying out of further atmospheric nuclear weapon
tests in the South Pacific Ocean is not consistent with applicable rules of international law.
And to order that the French Republic shall not carry out any further tests142

Ibid.
For the following factual account see ICJ, Nuclear Tests (Australia v France), Judgment of
20 December 1974, [1974] ICJ Rep 253 and ICJ, Nuclear Tests (New Zealand v France), Judgment
of 20 December 1974, [1974] ICJ Rep 457. As stated in Introduction, n 44, the facts underlying
both cases were similar, however, the proceedings were not joined. As a consequence the
Court issued two largely identical judgments on the very same day. Future references are to
Australia v France only, unless otherwise indicated.
139
ICJ, Nuclear Tests (Australia v France), Judgment (1974), n 138 above, 255, para 4.
140
Ibid 257, para 14.
141
ICJ, Nuclear Tests (Australia v France), Order of 22 June 1973, [1973] ICJ Rep 99, 106 and
ICJ, Nuclear Tests (New Zealand v France), Order of 22 June 1973, [1973] ICJ Rep 135, 142.
142
ICJ, Nuclear Tests (Australia v France), Judgment (1974), n 138 above, 256, para 11. The
application from New Zealand was drafted in different wording (see [1974] ICJ Rep 460,
para 11) but the Court nevertheless interpreted both as requesting the same.
137
138

118 A History of Promises

the Court proceeded with its judgment. As indicated in its order, the disputed questions of the claims jurisdiction and admissibility were to be
dealt with first, as preliminary matters. The ruling, however, elaborated
that before these aspects could be addressed, it was necessary to take a
step backwards and examine a question which was essentially prelimin
ary (or pre-preliminary in the words of a dissent):143 on account of its
inherent jurisdiction, the Court had first to ascertain whether a justiciable dispute (still) existed between the parties. The ICJ came to the conclusion that it did not, as follows.
The Court first established that in spite of the application, which asked
for a declaration that the carrying out of further atmospheric nuclear
weapon tests in the South Pacific Ocean was not consistent with applicable
rules of international law, Australia was in fact only interested in a termination of the French tests; the requested declaration of illegality of all future
tests was said to be nothing more than a means to this end.144 In order
to reach this decision, the Court relied on various statements made by
the Australian (just as in the parallel case those of the New Zealand) representative. These statements were of importance to the Court, not only
for its interpretation that the applicants aim was (only) a cessation of the
tests, but in addition as evidence that both would have been satisfied to
have achieved their goal if they had received a firm assurance by France
that it was to refrain from any future atmospheric testing. Both applicants
are cited by the Court as having repeatedly emphasised that they had not
received such a reliable assurance from France.145 The Court continued
by noting that France had in the meantime made numerous new public
statements which, especially in light of the fact that previous French statements had been addressed and rejected as insufficient by the applicants,
143
ICJ, Nuclear Tests (Australia v France), Dissenting Opinion of Judge De Castro, [1974] ICJ
Rep 372, 373, himself quoting Judge Fitzmaurice in the Northern Cameroons case.
144
The same was considered to be true for New Zealand. In light of the fact that its application, unlike that of Australia, did not even ask for an order against France not to carry out
any further tests but for the Court to find that the French conduct constitutes a violation of
New Zealands rights under international law, and that these rights will be violated by any
further tests, the Courts interpretation of New Zealands application goes even further.
145
The Court cites various statements made by the applicants in this regard. For Australia,
see, eg The concern of the Australian Government is to exclude completely atmospheric
testing. It has repeatedly sought assurances that atmospheric tests will end. It has not
received those assurances. The recent French Presidential statement cannot be read as a firm,
explicit and binding undertaking to refrain from further atmospheric tests, ICJ Nuclear Tests
(Australia v France), Judgment (1974), n 138 above, 261, para 27; see also the text to n 149
below. For New Zealand, see, eg New Zealand has not been given anything in the nature
of an unqualified assurance that 1974 will see the end of atmospheric nuclear testing in the
South Pacific, [1974] ICJ Rep 465, para 27, and It should . . . be clearly understood that
nothing said by the French Government, whether to New Zealand or to the international
community at large, has amounted to an assurance that there will be no further atmospheric
nuclear tests in the South Pacific. The option of further tests has been left open. Until we have
an assurance that nuclear testing of this kind is finished for good, the dispute between New Zealand
and France persists, ibid 46566, para 28, emphasis of the Court.

Nuclear Tests Cases119

merited the ICJs attention. Although they had not been made before the
Court, and partly after the oral hearings had ended, the new statements
were, according to the ICJ, in the public domain, known to the applicants
and, consequently, could not be ignored by the Court. The ruling continued by listing the relevant French statements, both old and new, in chronological order.
(1) The first was a statement contained in a communiqu issued by
the Office of the President of the French Republic, a copy of which was
communicated from the French Embassy in Canberra to the Australian
Department of Foreign Affairs (this statement had been made before the
oral hearings and was commented on as insufficient by Australia):146
The Decree reintroducing the security measures in the South Pacific nuclear test
zone has been published in the Official Journal of 8 June 1974. The Office of the
President of the Republic takes this opportunity of stating that in view of the
stage reached in carrying out the French nuclear defence programme France
will be in a position to pass on to the stage of underground explosions as soon
as the series of tests planned for this summer is completed.

(2) The second was a note from the French Embassy in Wellington sent
to the New Zealand Ministry of Foreign Affairs on 10 June 1974 (although
addressed to New Zealand, the Court could not fail to take note of it also
in the Australia v France judgment):
France, at the point which has been reached in the execution of its programme
of defence by nuclear means, will be in a position to move to the stage of underground tests, as soon as the test series planned for this summer is completed.
Thus the atmospheric tests which are soon to be carried out will, in the normal
course of events, be the last of this type.

Five further statements had followed the oral hearings.

(3) The President of the French Republic made a statement as follows at
a press conference on 25 July 1974:
on this question of nuclear tests, you know that the Prime Minister had publicly expressed himself in the National Assembly in his speech introducing the
Governments programme. He had indicated that French nuclear testing would
continue. I had myself made it clear that this round of atmospheric tests would
be the last, and so the members of the Government were completely informed
of our intentions in this respect.

(4) The Minister of Defence made a statement (the precise wording of

which is not cited) in the course of an interview on French television on
16 August 1974 in which he said that the French government had done its
best to ensure that the 1974 nuclear tests would be the last atmospheric
tests.
146

See n 145 above.

120 A History of Promises

(5)The French Minister for Foreign Affairs made a statement on
25 September 1974, addressing the United Nations General Assembly, in
which he said:
We have now reached a stage in our nuclear technology that makes it possible
for us to continue our programme by underground testing, and we have taken
steps to do so as early as next year.

(6) and (7) Two statements (the precise wording of which is not provided
in the judgment) were made by the Minister of Defence, both in a press
conference on 11 October 1974, stating in almost identical terms that there
would not be any atmospheric tests in 1975 and that France was ready to
proceed to underground tests. The judgment emphasises:
When the comment was made that he had not added in the normal course
of events [as was included in statement 2], he agreed that he had not. The
Minister also mentioned that, whether or not other governments had been officially advised of the decision, they could become aware of it through the press
and by reading the communiqus issued by the Office of the President of the
Republic.147

Before turning to the legal principles involved, the Court concluded

from these statements that France had made public its intention to cease
the conduct of atmospheric tests. In doing so, it placed particular import
ance on the statement made by the French President ((3) above) and the
ones made by the Minister of Defence ((6) and (7) above). The ICJ then
digressed from the actual facts of the case in its judgment to dedicate four
paragraphs to a description of the status and scope of unilateral declarations on the international plane. The paragraphs form a separate section within the judgment and are written in an abstract textbook style,
that is to say, they make no reference to the specific declarations before
the Court, but rather present the ICJs perception of the law applicable
to unilateral declarations as creating legal obligations. In them, the ICJ,
in crystal clear words, affirmed that unilateral assurances devoid of any
quid pro quo or acceptance can be binding upon the declaring state. No
special formal requirements exist as long as a declaration is made publicly
and carries the requisite intention to undertake a legal obligation. For the
Court, an assurances bindingness is grounded in the principle of good
faith. The relevant section of the judgment reads as follows:
It is well recognized that declarations made by way of unilateral acts, concerning legal or factual situations, may have the effect of creating legal obligations.
Declarations of this kind may be, and often are, very specific. When it is the
intention of the State making the declaration that it should become bound
according to its terms, that intention confers on the declaration the character
ICJ, Nuclear Tests (Australia v France), Judgment (1974), n 138 above, 26567, paras 3440.

147

Nuclear Tests Cases121

of a legal undertaking, the State being thenceforth legally required to follow a
course of conduct consistent with the declaration. An undertaking of this kind,
if given publicly, and with an intent to be bound, even though not made within
the context of international negotiations, is binding. In these circumstances,
nothing in the nature of a quid pro quo nor any subsequent acceptance of the
declaration, nor even any reply or reaction from other States, is required for the
declaration to take effect, since such a requirement would be inconsistent with
the strictly unilateral nature of the juridical act by which the pronouncement by
the State was made.
Of course, not all unilateral acts imply obligation; but a state may choose to
take up a certain position in relation to a particular matter with the intention
of being bound the intention is to be ascertained by interpretation of the act.
When states make statements by which their freedom of action is to be limited,
a restrictive interpretation is called for.
With regard to the question of form, it should be observed that this is not a
domain in which international law imposes any special or strict requirements.
Whether a statement is made orally or in writing makes no essential difference,
for such statements made in particular circumstances may create commitments
in international law, which does not require that they should be couched in
written form. Thus the question of form is not decisive. As the Court said in
its Judgment on the preliminary objections in the case concerning the Temple
of Preah Vihear: Where . . . as is generally the case in international law, which
places the principal emphasis on the intentions of the parties, the law prescribes
no particular form, parties are free to choose what form they please provided
their intention clearly results from it. (ICJ Reports 1961, p 31). The Court further
stated in the same case: . . . the sole relevant question is whether the language
employed in any given declaration does reveal a clear intention (ibid p 32).
One of the basic principles governing the creation and performance of legal
obligations, whatever their source, is the principle of good faith. Trust and
confidence are inherent in international co-operation, in particular in an age
when this co-operation in many fields is becoming increasingly essential. Just
as the very rule of pacta sunt servanda in the law of treaties is based on good
faith, so also is the binding character of an international obligation assumed by
unilateral declaration. Thus interested States may take cognizance of unilateral
declarations and place confidence in them, and are entitled to require that the
obligation thus created be respected.148

The Court continues by applying these principles to the statements

made by France. However, it first takes note of the fact that the applicants
had both commented on some of the earlier French statements and had
voiced their opinion that they fell short of constituting a binding undertaking to stop all atmospheric nuclear testing. The legal value of one of
the later declarations, that of the French Foreign Minister ((5) above), had
148

Ibid 26768, paras 4346.

122 A History of Promises

in fact been addressed by the Australian Attorney General, when he was
asked about it in the Australian Senate on 26 September 1974, as follows:
this statement falls far short of a commitment or undertaking that there will be
no more atmospheric tests conducted by the French Government at its Pacific
Tests Centre . . . There is a basic distinction between an assertion that steps are
being taken to continue the testing program by underground testing as early as
next year and an assurance that no further atmospheric tests will take place. It seems
that the Government of France, while apparently taking a step in the right
direction, is still reserving to itself the right to carry out atmospheric nuclear
tests. In legal terms Australia has nothing from the French Government which
protects it against any further atmospheric tests should the French Government
subsequently decide to hold them.149

This interpretation notwithstanding, the Court opined that it had to

form its own view on the meaning and scope intended by the author of
a unilateral declaration. In doing so, it found that the declarations had to
be regarded not as separate from each other but, with particular emphasis on declarations (3), (6) and (7) above, as a whole. They were made
publicly and erga omnes and had to be seen not in vacuo, but as having
been made in relation to the tests that were the subject of the proceedings
before the Court. The ICJ drew the following conclusion:
In announcing that the 1974 series of atmospheric tests would be the last, the
French Government conveyed to the world at large, including the Applicant, its
intention effectively to terminate these tests. It was bound to assume that other
States might take note of these statements and rely on their being effective. The
validity of these statements and their legal consequences must be considered
within the general framework of the security of international intercourse, and
the confidence and trust which are so essential in the relations among States.
It is from the actual substance of these statements, and from the circumstances
attending their making, that the legal implications of the unilateral act must be
deduced. The objects of these statements are clear and they were addressed to
the international community as a whole, and the Court holds that they constitute an undertaking possessing legal effect.
The Court considers that the President of the Republic, in deciding upon
the effective cessation of atmospheric tests, gave an undertaking to the inter
national community to which his words were addressed.
It is true that the French Government has consistently maintained, for example in a Note dated 7 February 1973 from the French Ambassador in Canberra
to the Prime Minister and Minister for Foreign Affairs of Australia, that it has
the conviction that its nuclear experiments have not violated any rule of international law, nor did France recognize that it was bound by any rule of inter
national law to terminate its tests, but this does not affect the legal consequences
of the statements examined above.
149
See ICJ, Nuclear Tests (Australia v France), Judgment (1974), n 138 above, 26869, para 48
and 26162, para 28, emphasis added.

Nuclear Tests Cases123

The Court finds that the unilateral undertaking resulting from these statements cannot be interpreted as having been made in implicit reliance on an
arbitrary power of reconsideration. The Court finds further that the French
Government has undertaken an obligation the precise nature and limits of
which must be understood in accordance with the actual terms in which they
have been publicly expressed.150

By finding that the French government had now given the reliable
assurance which both of the applicants had been found to have initially
sought, the Court, as a result, considered the dispute to have disappeared.
Therefore, it concluded, there was nothing left to adjudicate: the claim no
longer has any object and . . . the Court is therefore not called upon to give
a decision thereon.151
(ii) Analysis and Critical Assessment
The judgments reasoning has been perceived by numerous commentators as flawed in one or the other way and depicted as the ICJs way out
of a dilemma of having to declare atmospheric nuclear tests as contrary
to international law, whereby it would risk continued defiance by France
should it decide to continue with its tests, while in all likelihood being
ignored by China, which was also conducting atmospheric nuclear tests
at the time. It was felt that the alternative of declaring such tests as not
(yet) outlawed by customary international law would have flown in the
face of international efforts at the time to forbid such activity and at least
to have hampered the evolution of a customary prohibition in this area.152
While such insights into what drove the Court to reach its conclusion are
certainly of interest, it would be false simply to discard its ruling on this
account as being born out of a very special politically loaded situation and
hence as having no further impact on the evolution of international law.
After all, it is largely irrelevant whether the judges applied certain legal
principles while gritting their teeth, or embraced them wholeheartedly in
order to flee out of a cul de sac, if, and as long as, the legal reasoning on its
own withstands closer legal scrutiny. This is where the emphasis must lie.
Also, as Ian Brownlie stressed within the Commission:
Much law did in fact emerge from cases regarded at the time as narrowly based.
It was true that special circumstances had tempted ICJ to walk off stage in the
Nuclear Tests cases, and there was reasonable scepticism about the application
Ibid 26970, para 51.
Ibid 272, para 59.
For this dilemma, see, eg Pierre Lellouche, The International Court of Justice: the Nuclear
Tests Cases Judicial Silence v Atomic Blasts (1975) 16 Harvard International Law Journal
614; Franck, Word Made Law, n 136 above, 612; Wilhelm A Kewenig, Der Internationale
Gerichtshof und die franzsischen Kernwaffenversuche in Recht im Dienst des Friedens,
Festschrift fr Eberhard Menzel (1975) 34647.
150
151
152

124 A History of Promises

of the principle of good faith to the particular facts. But that was another matter,
and States did currently rely on the Nuclear Tests cases.153

This is precisely what Thomas Franck underlined by, as cited above,

referring to a mouse of a decision capable of monumental law. In the end,
various legal clusters within an opinion can continue to stand for themselves, even if a judgment as a complete whole should fail to convince;
and it is only the issue of the bindingness of unilateral assurances which is
of interest in the present context and merits closer analysis.154
The introductory passage to the Courts depiction of the legal framework applicable to unilateral and binding declarations includes no reference whatsoever to any of its or its predecessors earlier judgments, let
alone to any external source. The Court rather pretends that any such
reference is not necessary since it was well recognised that declarations
made by way of unilateral acts, concerning legal or factual situations, may
have the effect of creating legal obligations. This passage in its choice of
words follows in the footsteps of the PCIJs beyond dispute findings of
the legal bindingness of a unilateral declaration, when it considers the
bindingness of unilateral declarations simply to be well recognised. On
account of its very broad drafting, the passage, however, actually camouflages what the Court is asserting, making it less open to attack; instead
of referring to unilateral assurances or even promises as being generally
accepted as a source of unilateral legal obligations, the judgment instead
speaks of the much broader overall category of unilateral acts concerning factual as well as legal situations, which may give rise to creating
legal obligations. The legal audience is likely to agree with this assertion
since acts of recognition or unilateral acts giving rise to estoppel will come
to mind; here, the conclusion that these acts have a compelling legal effect
is indeed well recognised.
But the ICJ did not use this introduction to dwell on unilateral acts
such as recognition or a declarations possible effect under the principle
of estoppel (which would have necessitated a finding of detrimental reliance by the addressee);155 instead, it went on to pronounce on the bindingness of various unilateral assurances. As has been shown in this chapter,
153
See Brownlie, ILC, Summary Record of the 2526th Meeting, UN Doc A/CN.4/SR.2526
(1998) 49, para 7.
154
For the judgment in its entirety and the various legal questions involved see, eg Brigitte
Bollecker-Stern, Laffaire des essais nuclaires francais devant la Cour Internationale
de Justice (1974) Annuaire Franais de Droit International 299; Kewenig, Die franzsischen Kernwaffenversuche, n 152 above, 32348; R St J Macdonald and Barbara Hough,
The Nuclear Tests Case Revisited (1977) 20 German YB International Law 337; Serge Sur,
Les affaires des essais nuclaires (1975) 79 Revue Gnrale de Droit International Public 972.
Focusing only on mootness, Jos Juste Ruiz, Mootness in International Adjudication: the
Nuclear Tests Cases (1977) 20 German YB International Law 358; on declaratory judgments,
Jean Pierre Ritter, LAffaire des essais nuclaire et la notion de judgement dclaratoire
(1975) 21 AFDI 278.
155
For more on promises and estoppel see p 277 et seq.

Nuclear Tests Cases125

it would go too far to accuse the ICJ of having simply, out of the blue,
invented a new mechanism for states to be subject to a legal commitment,
by assuming a promise to be capable of creating legal obligations, as
there are cases of state and especially court practice predating this ruling
which had already drawn such conclusions. This fact notwithstanding, it
remained a leap for the Court to assume that the general bindingness of
unilateral assurances made by a state was not subject to debate but simply
well recognised. Whether or not the Court was inspired by an article
written by Fitzmaurice in 1957 remains speculation, but there are evident
parallels in the Courts finding and Fitzmaurices conclusion that:
a declaration [not made in return for, or simultaneously with, any specific quid
pro quo or as part of any general undertaking] may or may not create binding
legal obligations for the declaring party, according to its wording and intent,
and the circumstances of its making; but it seems fairly well settled that it can
and will do so if it is clearly intended to have that effect, and is held out, so to
speak, as an instrument on which others may rely and under which the declarant purports to assume such obligations.156

This assertion notwithstanding, the existence of binding unilateral

assurances was still vigorously denied by some commentators at the
time,157 with the ICJ itself up to this point not having taken a clear stand
in this regard, despite the fact that it had repeatedly relied upon unilateral
declarations made before it. The Court would therefore have bolstered
the strength of its argument considerably had it included references to the
evolution of the doctrine it was going to apply. That the latters precise
legal principles were not (and could not be) generally known, is also illustrated by the fact that the Court found it necessary to sketch out in some
detail the law applicable to obligations unilaterally assumed through a
states assurance before it pronounced on the effect of the French declarations. In doing so, it emphasised the following:
(i) it is the intention of the state making the declaration that it should
become bound according to its terms which confers on the declaration the character of a legal undertaking;
(ii) such an undertaking, if given publicly and with an intent to be bound,
is binding;
(iii) no quid pro quo, no subsequent acceptance nor even any reply or
reaction from other states is required for the declaration to take effect;
(iv) the intention is to be ascertained by interpretation of the act;
156
Sir Gerald Fitzmaurice, The Law and Procedure of the International Court of Justice
19514: Treaty Interpretation and Other Treaty Points (1957) 33 British YB International Law
202, 230.
157
See especially Quadri, Cours gnral, n 4 above, 361 et seq. The critical article byRubin,
Unilateral Declarations, n 4 above, published shortly after the Nuclear Tests cases, also proves
this point.

126 A History of Promises

(v) a restrictive interpretation is called for where a statement limits the
states freedom of action;
(vi) there are no formal requirements as the sole relevant question is
whether the language employed in any given declaration reveals a
clear intention to be legally bound;
(vii) the binding character of an international obligation assumed by unilateral declaration is based on good faith.
Nearly as interesting as these abstract findings is how the Court actually
applied them. As has been seen, the ICJ first brushed away the fact that
Australia had not placed any reliance on the declarations made by France,
which it had repeatedly rejected as insufficient.158 The reason the judges
provided for considering this aspect to be of no particular relevance was
that the Court had to come to its own opinion of the meaning and scope
which France had intended its declarations to have. Since courts are generally not bound by the legal interpretation assumed by one of the parties, and assessing the law is their prerogative, this finding might at first
glance not be very surprising. But the Courts approach touched upon
a question which goes to the very heart of the bindingness of unilateral
assurances, which is: Can an assurance be binding upon the declarant vis-vis an addressee if the latter has not relied on it and even openly rejects
the declaration made? In other words: Is (actual or at least presumed) reliance a necessary ingredient to establish the bindingness of a unilateral
assurance? The Court did not provide an express answer to this question,
which will be picked up again when addressing the basis for a promises
legal bindingness in chapter three.159 The pertinent question at this point,
however, is whether the ICJ dismissed reliance as being of any importance
in deciding not to give any weight to the Australian statements which
maintained that Australia was not relying on the French declarations as
binding commitments. If this were the conclusion to be drawn, it would
be a troubling one, because as a result, a state which has changed its mind
after having made a promise in the past would remain bound to adhere
to an outdated policy160 once proclaimed, even though and where the
addressee benefiting from it had, in fact, long dismissed it. Indeed, not
only an Anglo-Saxon lawyer such as Thomas Franck, but most lawyers
will probably feel what (with some understatement) has been identified
by him as mild distress,161 if the Court had actually held that a state has
See especially the statement of the Australian Attorney General, text to n 149 above.
See p 194.
160
See in this respect the criticism voiced by Rubin, Unilateral Declarations, n 4 above,
30, who, however, goes further when he considers us all losers where a state was bound to
adhere to an assurance made, unless another state had detrimentally relied on the declaration
made. Rubin hence opts for dealing with assurances only via the legal principle of estoppel.
For more on promises and estoppel, see p 277.
161
Franck, Word Made Law, n 136 above, 620.
158
159

Nuclear Tests Cases127

to follow a declaration irrespective of whether another state has relied

upon it. Apart from disquieting large parts of the legal profession, such
an interpretation would lead to a manifest contradiction with the ICJs
very own finding that the binding character of an international obligation assumed by unilateral declaration is based on good faith (as chapter three will establish in more detail). Good faith cannot make binding
a declaration in which no faith has (at least presumably) been placed by
another legal subject.
Taking a closer look at the Courts holding in fact reveals that it did not
discard reliance, as the whole judgment hinges on the assumption that
the applicants had received the assurance which they had sought from
the very beginning. Instead of completely dismissing reliance, the ICJ, on
account of the peculiarities of the case, therefore seems rather to have lowered the bar for finding what has been termed constructive reliance162 to
be present and sufficient for the French declarations bindingness in this
case. The assumption that the Court did so derives from the perception
that the ICJ, in terms of the addressees reliance on the declarations, was
in the following dogmatic pickle: according to the Court, the French declarations were made as legally binding assurances by France. Australia and
New Zealand had both sought an unequivocal assurance and had hence
openly signalled their will to rely on an assurance received. This perfect
match notwithstanding, the applicants (under the reasoning of the Court)
had chosen to reject the French statements in drawing the wrong conclusion that they lacked the requisite precision and France consequently any
intent to be legally bound. Had the Court now found that the Australian
display of non-reliance had destroyed any good faith grounds for the declarations to become binding, it would have adequately applied the doctrine it proclaimed as being based on good faith in establishing the legal
bindingness of a declaration (see further chapter three), but at the cost of
ignoring the fact that the interpretation given to the French statements as
intentional legal commitments by the Courts own judgment was likely
to alter the applicants perception and allow them to rely on the French
declarations made. Indeed, after the Court pronounced its judgment, the
Australian Prime Minister openly declared Australia now to be satisfied
with the legally binding assurances made by the French Republic.163 The
Courts solution in light of this situation was hence not that it held reliance to be of no importance, but that it was satisfied that Australia would
have relied, had it correctly interpreted the declarations made by France,
and that it was understood as being still willing to rely on such an unequivocal unilateral undertaking as was now found to exist by the Court.
Notwithstanding this, it was (as Thomas Franck correctly observed) only
162
163

Ibid 619.
See Le Monde, 8 January 1975, as cited in Franck, Word Made Law, n 136 above, 619.

128 A History of Promises

the Court which at that point in time was actually placing reliance on the
declarations made.164 Yet, it considered this to be sufficient, as it created
the basis for Australia (and New Zealand) to find that they had actually
already received what they had been looking for:
The applicant has repeatedly sought from the respondent an assurance that the
tests would cease, and the respondent has, on its own initiative, made a series
of statements to the effect that they will cease.

As a side note, this is not to overlook that the approach chosen by the
Court is highly problematic, if seen as a general principle for unilateral
commitments and through the eyes of a declarant making a unilateral
assurance. A promising state will not know what consequences will arise if
its declaration is rejected by its addressee. Can the latter still later invoke it
against the declarant? With the Courts decision this would be a possibility
as the rejection may turn out to have been made on the false assumption that
the declaration as such was insufficiently clear and not binding. The more
adequate solution for future scenarios in the interest of legal security therefore, even if more painful for the addressees in question, would rather be
to consider the declarant no longer bound by its statement, even if a court
might find it in fact to have reached the requisite threshold in order to be
understood as manifesting a will to be legally bound.
The fact that the Court interpreted the French declarations completely
differently than two of their addressees, Australia and New Zealand, had
done (the declarations were, after all, understood as being erga omnes declarations), draws attention to the important aspect of how the ICJ actually
established that France had in fact intended to be legally bound by its
public statements. The Court, as we have seen, did not go to the trouble of analysing the value of each declaration but instead assumed the
presidential declaration along with those of the members of the French
Government acting under his authority, up to the last statement made
by the Minister of Defence to constitute a whole. While doing so, the
ICJ stressed that acts of the President could undoubtedly be attributed
to France, a position that will cause little dispute in international law, as
Heads of State are generally understood as being capable of committing
their country on the international level.165 The same cannot as easily be
assumed for acts of the Minister of Defence, yet the Court is not the least
concerned with this aspect, as it seems to assume that the declarations
as a whole are simply attributable to France via its Head of State. How
numerous statements can be understood as constituting a whole is, however, hard to grasp, just as why such grouping should matter: if the Court
Ibid 618.
For treaties, Vienna Convention, Art 7(2) provides that, inter alia, Heads of State are
in virtue of their functions and without having to produce full powers considered as representing their state.
164
165

Nuclear Tests Cases129

was to conclude that none of the individual declarations could be understood as a binding commitment (the ICJ did not clarify this), it would
amount to rather awkward reasoning which, expressed mathematically,
would raise the question posed by Rubin, ie whether in international law
the sum of zeros can be more than zero.166
In addition, and although the ICJ laid so much emphasis in its abstract
discussion on the importance of a states intention in creating a legal obligation (a fact that was later repeatedly stressed in the ILC),167 the way in which
the Court came to the conclusion that France in fact intended to undertake a
legal obligation is highly significant. First, the Court made not the slightest
attempt to enquire what France actually wanted to achieve by making its
declarations; it was not concerned with French motives or with what went
on behind the scenes in France. Such insight would, of course, have been
difficult for the Court to gain and not doing so is in keeping with its own
assumption that the intention is to be ascertained by interpretation of the
act itself, its language being of primary importance.
Following this path, the Court first established that the French Republic
had made public its intention to terminate atmospheric nuclear testing.
Indeed, there can be no doubt that France had repeatedly done so. But
how did the Court come to the conclusion that the French statements were
made with an intention to undertake a legal obligation and were not simply
the announcement of Frances political plans? By emphasising that the
actual substance of the statements and the circumstances attending their
making needed to be considered in determining the legal implications of
the unilateral act, the Court offered little concrete help to determine how
it came to the conclusion that: [France] was bound to assume that other
States might take note of these statements and rely on their being effective. What the Court is thereby in effect saying is that it is less significant
what France actually thought of its declarations at the time, but how other
states perceived the declarations made: France was bound to assume
that other states were going to rely on these declarations as effective.
The Court thereby shifted the emphasis from the French intention to the
perception of other states, yet not the actual perception the Court does
not inquire how addressees in fact understood the declaration but rather
that of a reasonable addressee.168
What then can the reasonable addressee infer from the declarations made
by France in answering the question whether it wanted to undertake a legal
obligation? Nothing, really. The wording which the Court has declared to
Rubin, Unilateral Declarations, n 4 above, 28.
For more on this question see p 207 et seq.
168
This common oscillation in legal arguments between a subjective and objective
approach has been pinpointed by Martti Koskenniemi, From Apology to Utopia: the Structure
of International Legal Argument (Reissue with a new Epilogue, Cambridge, 2005), for various
areas of international law, including the ICJs Nuclear Tests cases judgments at ibid 34555.
166
167

130 A History of Promises

be decisive in order to ascertain whether a state intends to undertake a legal
obligation (or better: can be taken to have intended to do so by its addressees) does not provide any clues in the case before the Court. This is true
not only for the statements which the Court used as supplement, but also
for the principal ones made by the President ((3) above) and the Minister
of Defence ((6) and (7) above). When the President had (himself) made it
clear that this round of atmospheric tests would be the last, was this to
be reasonably understood as a legal commitment? As other commentators have stressed, the French position in the Suez Canal case that unilateral
undertakings could just as easily be unilaterally revoked169 could be cited
to lead the reasonable observer to believe that France was hence very likely
not intending to bind itself through its unilateral acts, as it apparently considered a unilateral instrument to be incapable of having such an effect.170
It must, however, be said that the reasonable addressee would also have
to be a very thorough one in its research to draw this conclusion, as it could
not gather this information from the declarations wording itself, but would
have to discover the relevant French position in the minutes of a Security
Council meeting held 17 years prior to the French declarations on nuclear
testing.
The direct circumstances in which the declarations as a whole, and
particularly those which followed the oral hearings, were made, are also
of little help; they were neither proclaimed in a particularly solemn manner, nor did any other formal means (such as the declarations registration
or at least their official notification via diplomatic channels or a face-toface meeting) underscore their importance, which could have served as
indicative of an intended legal obligation.171 By nevertheless assuming a
legal bindingness, the Court can therefore no longer pretend to be taking
itself out of the equation; it is no longer merely in the position of a potential and reasonable addressee, but by interpreting the French declarations as binding, the World Court is in fact establishing what a reasonable
observer may (from now on) assume. The ICJ is laying the groundwork
through its judgment for a position in which subsequently the reasonable
observing state can rely on declarations such as those of France as legal
unilateral undertakings. The Courts motivation in doing so is revealed by
the short passage inserted before it actually moved on to reach its verdict
on the declarations binding force:
See chapter two, text to n 119.
Rubin, Unilateral Declarations, n 4 above, 27.
171
This is true especially for declarations (3), (6) and (7) which the Court found it must
in particular take into consideration, ICJ, Nuclear Tests (Australia v France), Judgment (1974),
n 138 above, 267, para 41. The second declaration cited above was included in a note to New
Zealand (only), and merely announced that atmospheric tests which are soon to be carried
out will, in the normal course of events, be the last of this type, emphasis added. Declaration
(5) was included in a speech before the UN General Assembly but was equally vague in
respect of a future undertaking (we have taken steps to do so as early as next year).
169
170

Nuclear Tests Cases131

The validity of these statements must be considered within the general framework of the security of international intercourse, and the confidence and trust
which are so essential in the relations among States.

Rather than merely protecting this confidence and trust, the Court
seems to be actively promoting it, as the French statements can now be
relied upon and are elevated to binding commitments. The reasonable
state may now therefore infer a legal undertaking from (clear and specific) statements made by high ranking officials and rely upon them as
such, at least where additional circumstances are present, which for the
Court appear to have been present in the form of the ongoing proceedings
before it, as it stressed that the declarations were not made in vacuo.
It is true that the Court, in finding a legal commitment, and by even
assuming France to have undertaken an erga omnes obligation vis-vis the international community, can hardly be seen to be applying its
self-proclaimed restrictive interpretation. On the other hand, clarifications, developments and even complete shifts in a courts jurisprudence
are likely to come as a surprise to the parties before it and the Court, of
course, has the possibility to apply this newly elaborated doctrine (more)
restrictively in future cases, once the general principle is out of the box.
The ICJ has also cushioned the blow for France, as it only found the latter
to be compelled to do something which it apparently was going to do in
any event, that is, refrain from future atmospheric tests and move underground. Most importantly, the Court has, however, shown that the importance of a states actual intent when making a declaration is somewhat
relative and will in fact serve only as a theoretical point of departure, in
what may be described as the Courts detour, taken in order to ensure that
the bindingness of unilateral assurances is as firmly grounded on the principle of state sovereignty, as it is on the principle of protecting the stability
of and confidence in international relations. Often both principles will be
compatible under the ICJs assumption (which is more accurate after its
judgment has been pronounced) that states making clear unilateral statements can be assumed to have been aware that these declarations will be
interpreted as binding, and to have hence intended such interpretation as
the foreseeable result of their actions. What, however, if they have not in
fact done so? The Courts own approach has pointed the way to what is
going to happen where a dispute arises and the two principles clash; here,
the assumption of a states intent to be legally bound will be reduced to
nothing but a mere legal fiction.172 In considering what a state must have
assumed its addressees were going to perceive, the protection is that of

172
Speaking of a convenient but misleading fiction, Macdonald and Hough, Nuclear Tests
Case Revisited, n 154 above, 354.

132 A History of Promises

the confidence and trust which could reasonably be placed on a declaration
at the time, rather than of what a state actually intended.173
(iii) The Separate and Dissenting Opinions
The judgment just described was fully supported by (only) five of the
15 judges sitting on the bench.174 Since they were joined by four of their
colleagues, who in separate opinions and on different reasoning still
concurred in the conclusion of removing the case from the Courts list,
the final decision was supported by a majority of nine. Opposing them
were six dissenters, who voiced their disagreement in one joint dissent
supported by four and two individual dissents. Naturally, it is in these
seven documents (four separate opinions and three dissents), in which
the soundness of the Courts reasoning, including its position on the legal
effect of unilateral commitments, was first to be addressed. In light of the
active role which the judges have played in promoting the interpretation
of unilateral declarations as legally binding, their reactions are of special interest. Has the framework established by the Court been rejected
amongst its concurring and dissenting peers on the bench?
In his very brief separate opinion of only one page, Judge Forster
emphasised that, in his view, the claim had not only subsequently become
without object, but that it had lacked object ab initio, and radically. In
respect of the unilateral declarations, Judge Forster stated:
The recent French statements adduced in the reasoning of the Judgment do no
more than supplement (to useful purpose, I admit) what I conceived to be the
legal arguments for removal of the case from the Courts list.175

Viewed as a supplement to his legal arguments, there is therefore no

criticism of the reasoning on the bindingness of unilateral assurances here.
Judges Gros and Petrn, in their two separate opinions, do not address
the French declarations nor the doctrine on unilateral acts as pronounced
by the Court; in their opinion, the case lacked any legal substance and
should have been dismissed from the beginning. For Judge Ignacio-Pinto,
the same is true: the claims should have been rejected from the outset as
without object considering their strictly political character. He, however,
does address the reasoning provided by the Court in respect of the French
declarations, which he supports:
173
Although the ICJ therefore started from a subjective standpoint, it in effect clearly
ended on an outside view. Rather therefore than actually giving effect to all three considerations: subjective French intent, subjective reliance by Australia and New Zealand and
objective justice and making no preference, as Koskenniemi, From Apology to Utopia, n 168
above, 354 writes, the ICJ did in fact choose sides and paid no more than lip service to the
French intentions.
174
These were President Lachs, Judges Bengzon, Morozow, Nagendra Singh and Ruda.
175
See ICJ, Nuclear Tests (Australia v France), Separate Opinion of Judge Forster, [1974] ICJ
Rep 275.

Nuclear Tests Cases133

The fact remains that, to my mind, the Court was right to take the decision
it has taken today. I gladly subscribe at least in part to the considerations
which have led to its doing so, for, failing the adoption by the Court of my position on the issues of jurisdiction and the admissibility of the Australian claim, I
would in any case have been of the view that it should take into consideration,
at least in the alternative, the new facts which supervened in the course of the
present proceedings and after the closure of the oral proceedings, to wit various
statements by interested States, with a view to ascertaining whether circumstances might not have rendered the object of the Application nugatory. Since,
in the event, it emerges that the statements urbi et orbi of the competent French
authorities constitute an undertaking on the part of France to carry out no more
nuclear tests in the atmosphere, I can only vote in favour of the Judgment.176

Within the separate opinions, there is therefore no criticism of the reasoning as to the bindingness of unilateral declarations. Instead, Judge
Ignacio-Pinto adds his support to the relevant passage and its application
as contained in the Courts majority opinion. Turning to the dissenting
judges on the bench, the opinion of Judges Onyeama, Dillard, Jimnez
de Archaga and Sir Humphrey Waldock offers (in their own words) a
vigorous dissent177 to the Courts decision that the claim no longer had
any object. Indeed, in around 60 pages, the dissenters attack the judgment
on numerous grounds,178 but the passage on the bindingness of unilateral
declarations, or even the treatment of the French declarations as in casu
binding, are not amongst them. In this regard, the dissenters merely point
out that despite the judgments finding, the case should not have been
dismissed:
Moreover, the Courts finding as to that unilateral engagement regarding the
recurrence of atmospheric nuclear tests cannot, we think, be considered as
affording the Applicant legal security of the same kind or degree as would
result from a declaration by the Court specifying that such tests contravened
general rules of international law applicable between France and Australia. This
176
ICJ, Nuclear Tests (Australia v France), Separate Opinion of Judge Ignacio-Pinto, [1974]
ICJ Rep 308, 310, emphasis in the original.
177
See ICJ, Nuclear Tests (Australia v France), Joint Dissenting Opinion of Judges Onyeama,
Dillard, Jimnez de Archaga and Sir Humphrey Waldock, [1974] ICJ Rep 312.
178
In a nutshell, the dissenting opinion argues that the Court erred by in fact revising
rather than merely interpreting the applications, which in the eyes of the dissenters clearly
requested a declaratory judgment on whether atmospheric nuclear tests conducted by
France after the application was filed (and not merely after the judgment was rendered)
were in conformity with international law. The Court was also wrong to consider the case
moot without the applicants having requested the discontinuance of the proceedings or
even having been heard on this matter. In the eyes of the dissenters, the Court had jurisdiction to hear the claims under its Statute, Arts 36 and 37 together with Art 17 of the General
Act of 1928, the relevant provision of which was still in force and to which Frances reservation under the Optional Clause, which excluded disputes in matters of national defence, was
not applicable. The dispute between the parties hence continued to exist and it was a question that could not be dismissed as only political, since legal aspects had been put forward
by the applicants which required the Courts further scrutiny and for it to reach the merits
of the case.

134 A History of Promises

is shown by the very fact that the Court was able to go only so far as to find that
the French Governments unilateral undertaking cannot be interpreted as having been made in implicit reliance on an arbitrary power of reconsideration; and
that the obligation undertaken is one the precise nature and limits of which
must be understood in accordance with the actual terms in which they have
been publicly expressed.179

Without criticising the Court for having pronounced on the bindingness of unilateral declarations in general, or the bindingness of the French
declarations in particular, the four joint dissenters instead touch upon the
important issue of the limits and the revocability of such a unilaterally
assumed commitment, which in their opinion does not equate to a finding
that the conduct in question was barred by general rules of international
law. In point of fact, the Court left the precise limits of the obligations
which were found to exist for France open, just as it did not rule out their
revocation, but only their arbitrary reconsideration. What to make of this
rather vague wording is taken up again in chapter three, where the revocability of promises is discussed.180 For this history, it is instead noteworthy that the doctrine, as such, had up to this point not been criticised by
even one of the ICJs judges.
Judge De Castro filed a separate dissent, which is of special interest as
he dedicated a couple of pages within his opinion to the French declarations. Being amongst the dissenters, Judge De Castro also considered the
Court to err in declaring the claims moot. However, in his view the Court
was correct to take the French statements into consideration:
They are clear, formal and repeated statements, which emanate from the highest authorities and show that those authorities seriously and deliberately intend
henceforth to discontinue atmospheric nuclear testing.181

De Castro continued by emphasising that in analysing the French declarations, it was of primary importance to clearly distinguish between a
declaration expressing the intention to be legally bound and a mere political statement that a declarant intends to follow a certain line of action in
the future:
Upon the Court there falls the task of interpreting their meaning and verifying
their purpose. They can be viewed as the announcement of a programme, of an
intention with regard to the future, their purpose being to enlighten all those
who may be interested in the method which the French authorities propose to
follow where nuclear tests are concerned. They can also be viewed as simple
promises to conduct no more nuclear tests in the atmosphere. Finally, they can
be considered as promises giving rise to a genuine legal obligation.
179
ICJ, Nuclear Tests (Australia v France), Joint Dissent (1974), n 177 above, 320, emphasis
added by the dissent.
180
See p 251.
181
ICJ, Nuclear Tests (Australia v France), De Castro Dissenting Opinion (1974), n 143 above,
373, para 2.

Nuclear Tests Cases135

It is right to point out that there is not a world of difference between the
expression of an intention to do or not do something in the future and a promise envisaged as a source of legal obligations. But the fact remains that not every
statement of intent is a promise. There is a difference between a promise which
gives rise to a moral obligation (even when reinforced by oath or word of honour) and a promise which legally binds the promisor. This distinction is universally prominent in municipal law and must be accorded even greater attention
in international law.182

In emphasising this distinction, De Castro is not in disagreement with

the majority opinion signed by President Lachs, which equally pronounced on the important role of a states intention when it elaborated
its abstract ground rules. De Castro seems to agree with the underlying
doctrine that a unilateral promise can, as such, be binding upon a state,
even though his use of the word promise in the above passage apparently includes offers, since De Castro continued by pointing out that the
law generally required a quid pro quo from the beneficiary to the promisor, and also allowed any promise to be withdrawn at any time before its
regular acceptance by the person to whom it is made. A pollicitatio as De
Castro says, however, formed the exception to both of these principles.
Pollicitatio, the Latin for promise, is thereby depicted as a unilateral beneficial act, which needs neither quid pro quo nor acceptance. De Castro,
hence, not only did not criticise the Courts approach, but supported the
judgment in that a unilateral promise can as such be binding upon the
state making it. There is, however, clear disagreement between De Castro
and the Courts judgment when it comes to the application of this doctrine to the case at hand. For De Castro, the Court was wrong to give an
affirmative answer to the following questions:
Do those statements of the French authorities with which the Judgment is concerned mean anything other than the notification to the French people or the
world at large of the nuclear-test policy which the Government will be following in the immediate future?
Do those statements contain a genuine promise never, in any circumstances,
to carry out any more nuclear tests in the atmosphere?
Can those statements be said to embody the French Governments firm intention to bind itself to carry out no more nuclear tests in the atmosphere?
Do these same statements possess a legal force such as to debar the French
State from changing its mind and following some other policy in the domain
of nuclear tests, such as to place it vis--vis other States under an obligation to
carry out no more nuclear tests in the atmosphere?183

182
183

Ibid 37374, para 3.

Ibid 37475, para 4.

136 A History of Promises

For him:
To these questions one may reply that the French Government has made up its
mind to cease atmospheric nuclear testing from now on, and has informed the
public of its intention to do so. But I do not feel that it is possible to go farther.
I see no indication warranting a presumption that France wished to bring into
being an international obligation, possessing the same binding force as a treaty
and vis--vis whom, the whole world?
It appears to me that, to be able to declare that the dispute brought before it
is without object, the Court requires to satisfy itself that, as a fact evident and
beyond doubt, the French State wished to bind itself, and has legally bound
itself, not to carry out any more nuclear tests in the atmosphere. Yet in my view
the attitude of the French Government warrants rather the inference that it considers its statements on nuclear tests to belong to the political domain and to
concern a question which, inasmuch as it relates to national defence, lies within
the domain reserved to a States domestic jurisdiction.184

According to De Castro, the Court therefore wrongly assumed a French

intention to bring into being an international obligation. The last dissenter,
the Australian ad hoc Judge Barwick, agreed on this point. He too offered
a long and detailed dissent, in which numerous aspects of the judgment
were attacked. Again, however, the basic doctrine that unilateral assurances can as such be binding upon a state is not amongst them; its application to the French statements, however, is:
This is a very important conclusion purporting to impose on France an internationally binding obligation of a far-reaching kind. Nothing is found as to the
duration of the obligation although nothing said in the Judgment would suggest that it is of a temporary nature. There are apparently no qualifications of it
related to changes in circumstances or to the varying needs of French security.
Apparently it is restricted to the South Pacific area, a limitation implied from
the fact that the source of the obligation is the communiqu of 8 June 1974
issued in the context of the imminence of the 1974 series of tests. The purpose
and intention of issuing the communiqu and subsequently making the various statements is to my mind far from clear. The Judgment finds an intention
to enter into a binding legal obligation after giving the warning that statements
limiting a States freedom of action should receive a restrictive interpretation.
The Judgment apparently finds the clear intention in the language used. I regret
to say that I am unable to do so. There seems to be nothing, either in the language used or in the circumstances of its employment, which in my opinion
would warrant, and certainly nothing to compel, the conclusion that those
making the statements were intending to enter into a solemn and far-reaching
international obligation rather than to announce the current intention of the
French Government. I would have thought myself that the more natural conclusion to draw from the various statements was that they were statements of
policy and not intended as undertaking to the international community such a
184

Ibid 375, para 4.

Nuclear Tests Cases137

far-reaching obligation. The Judgment does not seem to my mind to offer any
reason why these statements should be regarded as expressing an intention to
accept an internationally binding undertaking rather than an intention to make
statements of current government policy and intention.185

It is worth taking note of the fact that none of the 15 judges, including its
six dissenters, have therefore criticised the Court for the textbook style
paragraph on the legal force of unilateral declarations, which has been
cited above. Those judges who have addressed the Courts reliance on the
French statements, be it in their separate opinions or as a dissenter, have
either supported the doctrine and its application in full (Judge IgnacioPinto), viewed it as a supplement to the legal arguments to remove the
case (Judge Forster), questioned the precise limits of the unilateral commitment found to be present in the case (Joint Dissent) or the application
of the doctrine to the statements before the Court (Judges De Castro and
Barwick). When taking the group of 15 judges as a pars pro toto for the
international legal community, or at least as indicative of what is widely
accepted therein, it appears that at this point, the position that unilateral
assurances can as such become binding upon a declarant had reached
this status.186 The precise application of this principle to France, however,
was very clearly criticised. As will be seen, subsequent cases indicate
that the Court is willing to take more seriously its self-declared approach
of a restrictive interpretation, where states make statements by which
their freedom of action is to be limited, without, however, questioning
the underlying principles of binding unilateral assurances, ie promises,
as such.

185
ICJ, Nuclear Tests (Australia v France), Dissenting Opinion of Judge Sir Garfield Barwick,
[1974] ICJ Rep 391, 44849.
186
Looking at the responses provided by scholars, the Court has received mixed reviews
when it comes to the part of the judgment discussed here. Positive: Edward McWhinney,
International Law-making and the Judicial Process: the World Court and the French Nuclear
Tests Case (1975) 3 Syracuse J International Law and Commerce 9, 46, who welcomes the
valuable new additions to international legal doctrine; Franck, Word Made Law, n 136
above, considers the Courts decision on this point to be an important but not unconscionable extension of prior decisions respecting verbal statements, at 615; speaking of a clarification of the doctrine, Sergio M Carbone, Promise in International Law: a Confirmation
of its Binding Force (1975) 1 Italian J International Law 166. Kewenig, Die franzsischen
Kernwaffenversuche, n 152 above, 340 et seq, does not question the doctrine as such but
criticises the Courts finding that the French declarations were binding undertakings and
not merely political declarations of intent. The same is true for Macdonald and Hough,
Nuclear Tests Case Revisited, n 154 above, 352. Strong criticism on the other hand is voiced
by Rubin, Unilateral Declarations, n 4 above, and Jerome B Elkind, Footnote to the Nuclear
Tests Cases: Abuse of Right A Blind Alley for Environmentalists (1976) 9 Vanderbilt JTL 57,
5964, both of whom, however, falsely discard all prior cases as providing any support for
the ICJs judgment. Rubin is in addition under the impression that the Court had declared
unilaterally assumed obligations to be irrevocable rather than only not arbitrarily revocable. On revocation see p 251 et seq.

138 A History of Promises

K Swiss Declaration in Relation to Security Council Resolution 253
In keeping with the chronological order, a few interesting examples of
state practice will be mentioned before returning to the World Court and
its subsequent application of the doctrine announced so clearly in the
Nuclear Tests cases.
The statement made by Switzerland in 1977 vis--vis the United
Nations is the first example in this context and includes a states declaration, the binding nature of which has been assessed in a UN memorandum published in the United Nations Juridical Yearbook under the heading
Legal Opinions of the Secretariat of the United Nations.187 The example
that is to follow after it and relating to unilateral declarations made by
various states against torture and other inhumane treatment was equally
addressed by a UN memorandum. Both illustrate how, only a few years
after the Nuclear Tests cases, ambiguous unilateral declarations of states
were open to very different interpretation. In the first case, the legal interpreter at the United Nations seems to assume a legal obligation instead of
a mere political one rather lightly where a state indicates its future course
of action. The second example illustrates a much more restrictive approach
applied to unilateral declarations of states, again by the UN Secretariat.
Reason for the first memorandum to be requested arose when
Switzerland expressed its opinion that its practice of admitting holders
of Southern Rhodesian passports implied no recognition of nationality,
since passports were considered to be no more than travel papers. The
question of implied recognition of nationality notwithstanding, the Swiss
practice was in potential conflict with an express provision contained
in Security Council Resolution 253 of 1968, in which the Council, acting
under Chapter VII of the Charter, inter alia, in paragraph 5 decided that
all states members of the United Nations should prevent the entry into
their territories, save on exceptional humanitarian grounds, of any person travelling on a Southern Rhodesian passport. Although Switzerland
was not a state member of the United Nations at the time, the questions
whether the Security Council intended to impose an obligation on nonmembers and, if so, whether it could do so or required the latters consent, according to the memorandum did not need to be discussed because
Switzerland (referred to as the State concerned in the memorandum)
had made the following statement in a note of September 1968 to the UN
Secretary-General:
In its statement of 10 February 1967 concerning the Security Council resolution
of 17 December 1966, the Federal Council explained that, for reasons of principle (the State concerned) as a neutral State, cannot submit to the mandatory
sanctions of the United Nations. However, independently and without recog187

See United Nations Juridical Yearbook (1977) 19394.

Swiss Declaration in Relation to Security Council Resolution 253139

nizing any legal obligations to do so, it has taken steps to ensure that any possibility of increasing Rhodesian trade is excluded and that the United Nations
sanctions policy cannot be contravened. The Federal Council will maintain this
position. With reference to the latest (253) resolution of the Security Council, it
will attempt independently and always in the context of the legal order (of the
State concerned), to see that Rhodesian trade cannot avoid the Security Council
sanctions through its territory.188

The memorandum continued that Switzerland had reaffirmed and

strengthened this declaration in a note of 17 February 1977, which read:
The . . . Government, would however, independently and without recognizing
any legal obligation in the matter, take care to prevent [its] territory from being
used for the purpose of circumventing the Security Council sanctions.189

In the memorandum the UN Secretariat drew the following conclusion:

It seems clear on the basis of the foregoing, that with respect to the denial of
entry to Rhodesian passport holders the Government concerned has unilaterally
and unreservedly accepted the obligation. To cite the note of 17 February 1977: The
Government would . . . take care to prevent [its] territory from being used for
the purpose of circumventing the Security Council sanctions. This statement
would seem to be unequivocal and subject to no reservations.190

For the Secretariat, Switzerland had therefore unilaterally undertaken to comply with paragraph 5 of Security Council resolution 253 as
the memorandum stated in another part;191 an undertaking with which
the admission of persons holding South Rhodesian passports was not
compatible, as the memorandum concluded. It should be noted that the
interpretation of the Swiss declaration as a unilateral legal undertaking
is, if anything, more in the tradition of the ICJs actual application of its
doctrine to the French declarations than that of the announced restrictive interpretation. There is little to glean from the circumstances in
which the declarations by Switzerland were made in terms of whether it
intended to make a legally binding declaration, which leaves the declarations wording as the sole and primary indicator of the Swiss intentions.
Here, it must be noted that Switzerland had used rather vague wording
and a formulation which a lawyer (in a state department) would rarely
choose when he or she was trying to express their countrys intention to
undertake a legal obligation. In the first declaration, Switzerland stated
that it would attempt to see that Rhodesian trade would not be able to
avoid the Security Council sanctions through its territory. The second
declaration, while slightly stronger in its assurance, still includes not the
hard wording which is to be found in binding resolutions or standard
Ibid 193, para 3.
Ibid 193, para 3.
190
Ibid 193, para 4, emphasis added.
191
Ibid 194, para 5.
188
189

140 A History of Promises

treaties, such as undertakes to, ensures to, but the weaker will take
care to prevent. Also, and although the Secretariat is technically correct
in implicitly considering the Swiss declaration that it recognised no (existing) legal obligation in the matter to be of no hindrance to the country
assuming a new obligation, the Swiss formulation that it would take care to
do something coupled with without recognising any legal obligation in
the matter, should have suggested even more caution in its approach. As
already indicated, the ICJ rulings that were to come have indeed followed
a more restrictive approach, under which a statement such as the one here
can hardly be taken to qualify as a binding unilateral assurance.
L Declarations on Torture and Other Inhumane Treatment
Another UN memorandum dealt with the question whether unilateral declarations deposited by member states in response to General
Assembly Resolution 32/64 of 8 December 1977 relating to torture and
other cruel, inhumane or degrading treatment or punishment constituted
binding undertakings in international law.192 The General Assembly in its
Resolution had called upon member states to reinforce their support of
an earlier declaration annexed to Resolution 3452 on the Protection of All
Persons from Being Subjected to Torture and Other Cruel and Inhuman
or Degrading Treatment or Punishment, by each adopting a unilateral
declaration.193 These declarations were to be deposited with the SecretaryGeneral and a model unilateral declaration had been appended to the
Resolution. It read as follows:
The Government of . . . hereby declares its intention:
(a) To comply with the Declaration on the Protection of All Persons from Being
Subjected to Torture and Other Cruel, Inhumane or Degrading Treatment or
Punishment (General Assembly Resolution 3452 (XXX) annex);
(b) To implement, through legislation and other effective measures, the provisions of the said Declaration.194

After Luxembourg had deposited such a declaration,195 some administrative questions within the United Nations, such as the adequate place
of registration, arose, which also depended on whether the declarations
were binding legal commitments or not. In assessing the declarations
value, the Secretariat began by recalling the formal registration that was
envisaged within the Resolution. However, the memorandum continued:
See United Nations Juridical Yearbook (1978) 19899.
See UNGA Resolution 32/64.
194
As cited in United Nations Juridical Yearbook (1978) 198, note 81.
195
According to Nigel Rodley, The Treatment of Prisoners under International Law, 2nd edn
(Oxford, 1999) 61 with references in his note 49, 33 states are reported to have made such
declarations.
192
193

British Pledge on the Five Techniques before the ECtHR141

the model declaration annexed to the resolution would merely have
Governments declare their intention to comply and to implement, rather than
indicate that they will or undertake to comply and implement. On balance,
therefore, the resolution suggests that no binding obligation is intended.196

Here, the Secretariat therefore adopted a much more restrictive approach

in its interpretation, by emphasising that the legal code employed was not
that usually used in order to indicate the willingness of a state to undertake
a legal commitment. An examination of the circumstances, ie the debates
which led up to the adoption of the Resolution, was held to confirm this
perspective, especially as the principal sponsor of the Resolution, India,
had introduced it with the following words:
the fourth preambular paragraph [of the draft Resolution] recognized the need
for further international action in the form of a legally binding international
convention. The unilateral declarations called for in paragraph 1 would be an
expression of the good faith of Governments and their moral commitment to
the provisions of the Declaration on Torture.197

Although some states are said to have attributed some legal force to
the declarations made, the Resolution adopted, the travaux prparatoires,
and the fact that the ultimate goal of a binding convention had been contrasted with the immediate object of securing moral commitments,198 led
the memorandums author to conclude that no binding obligation was
intended.
As indicated, it is this rather careful assessment in dealing with and
addressing the legal value of a declaration made by a state, closely focusing on the circumstances as well as a statements precise wording, which
has come to be employed in order to assess its nature as legally compelling or not.
M British Pledge on the Five Techniques before the ECtHR
The British representative before the European Court of Human Rights
(ECtHR) in Ireland v United Kingdom199 also made a unilateral pledge, and
one which was clearly inspired by the Nuclear Tests cases judgments. The
case (which has received some attention, as it was the first inter-state case
ever to be brought before the ECtHR) arose in relation to the practice of
extra-judicial arrests as exercised from August 1971 until December 1975
by the Northern Irish authorities against persons suspected to be involved
United Nations Juridical Yearbook (1978) 198, para 3.
As cited ibid 199, reference to UN Doc A/C.3/32/SR.37, para 27, omitted.
198
Ibid para 4.
199
ECtHR, Republic of Ireland v United Kingdom, application no 5310/71, Judgment of 18
January 1978, Series A, No 25, 175.
196
197

142 A History of Promises

with the Irish Republican Army (IRA).200 Shortly after the first arrest operations had been conducted (the so-called Operation Demetrius), Irish and
subsequently Northern Irish and English newspapers published reports
of various people alleging ill-treatment by security forces to have taken
place during their arrest, interrogation or otherwise. Internal investigations of the matter within Great Britain commenced and focused especially
on the use of five disorientation or sensory deprivation techniques201
that had been applied in combination during the in-depth interrogation
of numerous arrested people. The installed Compton Committee issued
its findings on the matter on 3 November 1971, in which it concluded that
the applied techniques, while constituting physical ill-treatment, did not
constitute physical brutality as it understood that term. The strong criticism which followed upon the publication of this finding in Great Britain
led to the announcement made only two weeks later that another committee was going to be set up. Its task was to investigate whether the current
interrogation practice as applied to people suspected of terrorism needed
to be amended. It was in this situation, and with the second expert opinion
still pending, that the Republic of Ireland on 16 December 1971 submitted
an application against the UK government to the European Commission
of Human Rights. Back in Great Britain, the new commissions Parker
Report was subsequently adopted on 31 January 1972. Although divided
on the question whether the techniques should be ruled out on moral
grounds, it unanimously considered (at least some of) the methods to
be illegal under domestic law. On the same day, the UK Prime Minister
stated the following in the UK Parliament:
[The] Government, having reviewed the whole matter with great care and with
reference to any future operations, have decided that the techniques . . . will not
be used in future as an aid to interrogation.

He added:
The statement that I have made covers all future circumstances. If a Government
did decide . . . that additional techniques were required for interrogation, then
I think that . . . they would probably have to come to the House and ask for the
powers to do it.202

This declaration notwithstanding, the European Commission on Human

Rights considered itself not only competent but bound under the European
Convention on Human Rights (ECHR) to express an opinion whether or
not the use of the five techniques constituted a practice in breach of Article
200
For the following facts, see ibid para 11 et seq, and para 96 et seq; and the outline in ECtHR,
Ireland v United Kingdom, Series B (1980), vol 23-I, 24.
201
These consisted of standing for hours in a stress position, hooding, subjection to
noise, deprivation of sleep, deprivation of food and drink, ECtHR, Republic of Ireland v United
Kingdom (1978), n 199 above, para 96.
202
Ibid para 101.

British Pledge on the Five Techniques before the ECtHR143

3 of the ECHR.203 In an extensive report,204 adopted on 25 January 1976,

the Commission, inter alia, found the interrogation measures to have been
applied in 14 cases and considered the combined use of the five techniques
in the cases before it constituted a practice of inhumane treatment and torture in breach of Article 3 of the Convention.205 Whilst the Commission
decided on 9 March 1976 to leave it at that and not to take the case before
the ECtHR,206 the Irish government did not consider the matter settled and
applied to the ECtHR only a day later, on 10 March 1976.
In its written response to the claims brought by Ireland before the
ECtHR, the United Kingdom did not contest the factual findings that
the said techniques had been applied in the 14 cases identified by the
Commission, nor did it (any longer) contest a breach of Article 3. Instead,
and precisely on account of it not doing so, the British representatives
asked the Court to determine the claim to be without object with regard
to the criticised techniques, citing the ICJs decisions in the Northern
Cameroons and the Nuclear Tests cases as jurisprudence of interest.207 The
Irish Attorney General, Mr Costello, in the oral hearings before the ECtHR,
inter alia, responded to the British submission as follows:
It cannot be said that the object of the claim is achieved, when, after an unfavourable, non-legally binding opinion expressed by the Commission, a respondent government does not admit the validity of the conclusions made by the
Commission or its opinion that breaches of the Convention occurred, but merely
states that it will not contest those conclusions or the Commissions opinion. It
cannot be said, Mr President as was said in the Nuclear Tests case that the
object of the applicants claim has been achieved by such a statement by the
respondent Government.208

The above-cited statement of the UK Prime Minister was, according to

Mr Costello, insufficient:
The announcement of discontinuance made to the United Kingdom Parliament
was qualified, however, by the statement that if the five techniques were reintroduced in future, legislation for this purpose would be required. The respondent Government have failed to give an undertaking that such legislation would
not be reintroduced and they have not done so in their counter memorial.209

A day later, on 8 February 1977, the British representative made the following statements before the Court:
ECtHR, Ireland v United Kingdom, Series B (1980), vol 23-I, 40004.
Ibid 30 et seq.
205
Ibid 411.
206
See the letter from the Secretary to the Commission to the Registrar of the Court,
ECtHR, Ireland v United Kingdom, Series B (1981), vol 23-II, 6.
207
See the counter-memorial of the United Kingdom, paras 0.10 and 0.19, 0.20, ibid 10708
and 111.
208
Verbatim report of the public hearings held on 7, 8 and 9 February 1977, ibid 316.
209
Ibid 317.
203
204

144 A History of Promises

In deference to that reasoning [of the Commission, and equally open to the
ECtHR, that Great Britain had not gone far enough in its undertakings] we considered, before the case opened yesterday and before we heard Mr Costello,
whether we could now remove all conceivable qualifications upon our undertakings as to the future and so remove that which the Commission, on the
material before it, evidently felt to be a bar to the avoidance of adverse conclusions. We believe that we can. We are convinced that we can make a statement sufficiently clear, addressed to this Court as representing the international community of
free Europe, so as to enable the Court to hold at this stage that that statement constitutes
an undertaking possessing legal effect. Thus we hope that the Court will be enabled
to face a situation in which the stated objective of the applicant Government has
in effect been accomplished. . . .
In those circumstances I am authorised to give, in relation to the five techniques, the following unqualified undertaking, and I measure my words with
care. The Government of the United Kingdom have considered the question of
the use of the five techniques with very great care and with particular regard
to Article 3 of the Convention. They now give this unqualified undertaking,
that the five techniques will not in any circumstances be reintroduced as an
aid to interrogation.210

The ECtHR took formal note of the undertaking given before it and
cited the declaration (last two sentences provided above) in its judgment
rendered, 18 January 1978. Unlike the ICJ, however, the ECtHR did not
draw the conclusion from this undertaking that the issue was moot. The
reason given by it was not the inadequacy of the commitment or that of
the other safeguard measures adopted by Great Britain to assure the nonrepetition of the criticised treatment, but that the questions raised were of
an interest that extended beyond the dispute between the parties in the
present case:
Nevertheless, the Court considers that the responsibilities assigned to it within
the framework of the system under the Convention extend to pronouncing
on the non-contested allegations of violation of Article 3. The Courts judgments in fact serve not only to decide those cases brought before the Court but,
more generally, to elucidate, safeguard and develop the rules instituted by the
Convention, thereby contributing to the observance by the States of the engagements undertaken by them as Contracting Parties (Article 19).

As a side note, it should be mentioned that the Court subsequently

found the applied techniques violated Article 3 of the ECHR by constituting a practice of inhuman as well as degrading treatment, but did not
amount to torture. This last finding came as a surprise to many commentators, especially in light of the Commissions unanimous finding that the
treatment in its severity did amount to torture.211
Ibid 344, emphasis added.
For more on this aspect, see David Bonner, Ireland v. United Kingdom, Annotation
(1978) 27 International and Comparative Law Quarterly 897, 898902.
210
211

Swiss Petitpierre Assurance to the United Nations145

In the present context, however, it is the declaration made by the British

representative which is an important and interesting example of state
practice for a number of reasons.212 First of all, the British agent, after
emphasising his authority and his careful choice of words, in clear terms
voiced the intention that the declaration should be understood and held
by the Court to be a legal undertaking (as seen in the cases mentioned in
this chapter, most declarants have not made it that easy for their addressees). Secondly, while made to a court and hence in good company with
other similar declarations already addressed, the British representative
not only marked his words as constituting a legally binding declaration,
but also shed some light on why the ECtHR might function as the declarations addressee: it was seen as representing the international community
of free Europe, and the declaration is therefore in fact addressed to the
latter. The declaration should not be misunderstood and downgraded to
constitute only a pseudo-promise, ie the mere assurance to comply with
an already existing obligation,213 which it was not. The British representative did not make a (legally superfluous) declaration that his country
would adhere to the ECHR and hence refrain from applying techniques
that constitute a breach of it. In fact, Britain, just as the Irish Attorney
General had stressed, did not actually admit that the techniques infringed
the Convention and were hence outlawed by it, but merely chose not to
contest the allegations made by the applicant and, further more, tried to
turn the question whether or not the practice actually infringed the ECHR
into an irrelevant one, by not only abandoning the techniques but by giving an unqualified undertaking that they would never be reintroduced.
This unqualified assurance was therefore made irrespective of whether
the methods do or do not infringe the Convention and, as such, must be
qualified as a unilateral and binding promise by Great Britain, existing
independently and in addition to the obligations included in the ECHR.
N Swiss Petitpierre Assurance to the United Nations
The declaration made by the Head of the Swiss Federal Department of
Foreign Affairs, Mr Petitpierre, to the United Nations, according to which
the United Nations and its officials would as a minimum benefit from the
same treatment which was accorded to any other international organisation based on Swiss territory, dates back to 1946. As such it could have
been mentioned much earlier in this chapter, but will only be addressed
212
Refering to the declaration as a valuable addition to state practice on unilateral acts
and as important for the doctrine of sources of international law, see also Albert Bleckmann,
Anmerkungen zum Nordirland-Fall des Europischen Gerichtshofs fr Menschenrechte
(1979) Europische Grundrechte Zeitschrift 188.
213
For more see text to and following n 109 above.

146 A History of Promises

now for the reason that it was not until 1979 that the question whether this
declaration actually constituted a binding unilateral assurance was not
only raised within Switzerland, but also answered by an expert opinion
in a note dated 2 April 1979 of the Public International Law Section within
its Federal Department of Foreign Affairs. The precise legal value of the
declaration had become of importance when the European UN Office
decided to oppose a fiscal decision by the Canton of Geneva, claiming
to profit from the same standards as included in the Swiss arrangements
with the International Labour Organization (ILO) and the World Health
Organization (WTO). Although the relevant provision was not included
in the agreement made with the United Nations, the Petitpierre declaration of 1946 had assured that:
Les autorits suisses sont disposes accorder aux Nations Unies et leurs
fonctionnaires un traitement au moins aussi favorable que celui accord toute
autre organisation internationale sur le territoire suisse.214

The pledge of 5 August 1946 was well documented, as it had been made
in the form of an official press statement (communiqu official remis la
presse) and had already been recalled in a statement of the Swiss Federal
Council to the Swiss Federal Assembly on 28 July 1955. There, the Council
had stressed the fact that the United Nations, as a result of the assurance
given, were in a position which allowed the United Nations to demand to
benefit from any advantage which, while not included in the provisional
arrangement concluded with it, was accorded by Switzerland to another
international organisation.215 In its expert opinion, the Department
of Foreign Affairs came to the conclusion that the declaration of Mr
Petitpierre constituted a binding promise for Switzerland. Its reasoning as
to the bindingness of the declaration is based squarely on the Nuclear Tests
cases. After finding the declaration to be of a unilateral character made by
a competent state representative, the opinion addresses two conditions,
which, it said, were necessary for a promise to become binding: the will
to promise and the promises publicity. These principles were applied as
follows to the declaration:
[The Department does not consider] Mr Petitpierre to have made that promise
without having the intention of executing it . . . Also, one should not ignore that
the declaration was reproduced in the message of 28 July 1955, which indicates
214
Lucius Caflisch, Pratique Suisse 1982 in (1983) XXXIX Schweizerisches Jahrbuch fr
Internationales Recht 183 (which may be translated as: The Swiss authorities are ready to
provide the United Nations and its officials with a treatment which is at least as favourable
as that accorded to any other international organization on Swiss territory).
215
Ibid 183: nous avons donn aux Nations Unies lassurance quelles seraient au bnfice dun rgime au moins aussi favorable, tous gards, que celui accord toute autre
organisation internationale sur le territoire suisse. En dautres termes, les Nations Unies
peuvent demander dtre mises au bnfice de tout avantage, non prvu dans larrangement
provisoire, que nous accorderions une autre organisation internationale.

Declaration on the Use of War Material Exported out of Austria147

the importance that was attached to it. One can therefore not only admit that
the will to promise was real but also that the intention to execute it was certain.
Concerning the promises publicity, it is necessary that it is made known to
those to which it is of direct interest, be it through a diplomatic note or during a discourse. Diffused in the form of a press communiqu, the promise was
without question brought to the attention of its addressee, the United Nations.
The unilateral engagements undertaken by States must be done so in respect
of the principle of good faith which governs international relations. [citation of
and reference to the Nuclear Tests cases omitted]
In sum, the above analysis undertaken by the Department brings it to the
conclusion that the declaration made by Mr Petitipierre on 5 August 1946 to Mr
Trygye Lie is binding.216

On account of the engagement undertaken by its Head of Foreign

Affairs, the note concluded, the favourable treatment accorded to the ILO
and WTO had therefore also to be attributed to the United Nations.217
O Declaration on the Use of War Material Exported out of Austria
The next case of state practice in the timeline followed here similarly illustrates how the doctrine pronounced by the ICJ in the Nuclear Tests cases has
left its imprint on and influenced the practice of states. Under the Austrian
Law on War Material, any export of war material out of its country is subject to a permission of the Austrian state.218 Section 3(2) of the Law allows
the Austrian authorities to require the weapons recipient to specify in a
declaration the materials precise future use (Endverbrauchsbescheinigung).
In a case that was brought to the attention of the Department for Public
International Law (Vlkerrechtsbro) in the Austrian Federal Ministry of
Foreign Affairs, this statement had been made by a General A on behalf
of his country.219 Before addressing whether the General had the power to
216
My translation of the following original: [Le dpartement ne pense pas] que M Petitpierre
[a] fait cette promesse sans avoir lintention de lexcuter . . . De plus, il ne faut pas perdre de
vue que la dclaration fut reproduite dans le message du 18 juillet 1955, ce qui indique toute
limportance que lon y attache. [On peut] ds lors dautant mieux admettre que cette volont
de promettre tait relle que lintention de lexcuter tait certaine. En ce concerne la publicit
de la promesse, il est exig quelle doit tre porte la connaissance de ceux qui sont directement intresss et cela soit par note diplomatique, soit loccasion dun discours. Diffuse
sous forme dun communiqu de presse, la promesse de M Petitpierre est incontestablement
parvenue connaissance de son destinataire, les Nations Unies. Les engagements unilatraux
pris par les Etats doivent ltre dans le respect du principe de la bonne foi qui rgit les relations
internationales . . . En rsum, lanalyse laquelle [il a t] procd ci-dessus conduisent [le
Dpartement] constater le caractre obligatoire de la dclaration de M Petitpierre faite le 5
aot 1946 M Trygve Lie. Ibid 186.
217
Ibid 186.
218
Kriegsmaterialgesetz of 18 October 1977, sterreichisches BgBl no 540/1977, see also
29 sterreichische Zeitschrift fr ffentliches Recht und Vlkerrecht 32527.
219
For the following see Gerhard Hafner, Die sterreichische diplomatische Praxis zum
Vlkerrecht (1981) 32 sterreichische Zeitschrift fr ffentliches Recht und Vlkerrecht 301, 343
45, s 19.1.2.

148 A History of Promises

commit his state, the Department in its memorandum, while referring to
the Nuclear Tests cases, made it clear that unilateral assurances or promises
made by a state gave rise to an international commitment, if the circumstances in which they were made showed that the declarant state had the
intention for its statements to do so. In the present case, the Department
concluded, the declarant must have been aware that the assurance and
the commitment to in future comport itself as declared was a prerequisite
for the conclusion of the (private law) contract in which his country was
interested. The declaration hence had to be understood as having been
made with the intention to undertake a legal commitment. Since Austria
had received a formal note in which the Foreign Minister of Country X
had via its Austrian ambassador assured the Generals legal competence,
the Department considered that his declaration was sufficient to bind his
country and give rise to the commitment required by the Austrian Law.
P German Approval Case before the German Constitutional Court
Another interesting example of state practice involves an approval220 of
the German government to the US government granting the latter the right
to install nuclear-headed Pershing-II and Cruise missiles on German soil.
The final approval by the German government was given in November
1983221 in order to execute part of the NATO Double-Track Decision,
which had been adopted at Brussels on 12 December 1979. The German
Parliament, however, was not asked to give these governmental actions
its blessing in the form of a federal law. In the eyes of the Green Party in
Germany, the far-reaching step taken required legislation and could not
be made by the executive alone; it hence instigated proceedings before
the German Constitutional Court which issued its ruling on 18 December
1984. In it, the precise legal nature of the permission or approval given
by the German government turned out to be of particular importance,
since Article 59(2) of the German Basic Law requires:
Treaties that regulate the political relations of the Federation or relate to subjects of
federal legislation shall require the consent or participation, in the form of a federal law, of the bodies responsible in such a case for the enactment of federal law.222
220
The term used by the German government, the Court and the applicant was
Zustimmung, see 68 Entscheidungen des Bundesverfassungsgerichts (BVerfGE) 1. It can
be translated not only as approval but also as consent, permission or assent. The case is
reported in the International Law Reports, German Federal Constitutional Court (BVerfG),
Cruise Missiles Deployment (German Approval) [1984] ILR 365, with Approval in the case
name, while in the case report the declaration is mainly translated as declaration of consent.
221
The precise date apparently falls between 16 and 22 November 1983 but is not specified
by the Court, see 68 BVerfGE 7476.
222
Translation taken from translation of the Basic Law provided by the German Parliament
and available at www.btg-bestellservice.de/index.php?navi=1&subnavi=68&anr=80201000,
emphasis added.

German Approval Case before the German Constitutional Court149

An important question was hence whether the German declaration had

given rise to and formed part of a treaty that regulated the political relations of the Federation, in which case the German Parliament would have
to pass legislation in the matter. The Constitutional Court, in its judgment,
first established that the approval in question was not merely a declaration
of intent, but had the legal effect of allowing the United States to deposit
the weapons referred to above on German soil. Since the German government had explained that the final operative command, according to NATO
guidelines, lay with the President of the United States, the Constitutional
Court considered the permissions legal effect as equivalent to a partial
transfer of Germanys sovereign right to decide over the use or non-use
of weaponry stationed on its own territory.223 This approval did not form
part of an international treaty, as there were no indications that the parties understood their actions as giving rise to a bi- or multilateral treaty.224
Further more, there were:
no discernible reasons for the assumption that the Declaration of the Federal
Government would be ineffective under international law and hence could not
bind the Federal Republic of Germany.225

After having noted that a Foreign Minister was capable of binding his
or her country under public international law and no defects of the declaration were apparent, the Constitutional Court went on to decide that
declarations of the kind now before it did not fall under Article 59(2) of
the German Basic Law and hence did not require the participation or
approval of the German Parliament. This was true for unilateral manifestations of will226 made in the context of existing treaties just as it was true
for other unilateral manifestations of will on the international level, such
as acts of recognition, those terminating diplomatic relations, on the utilisation of the continental shelf or fishing zones and the denunciation of
treaties. The Court also expressly rejected any extension of Article 59(2) to
unilateral declarations or its analogue application, a step for which it has
received some criticism, since it thereby allowed the executive to make
far-going commitments without any parliamentary involvement, merely
on account of its choosing to use a unilateral declaration and not a treaty
223
German Federal Constitutional Court (BVerfG), Cruise Missiles Deployment (German
Approval) [1984] ILR 376 (= 68 BVerfGE 91).
224
Ibid 371 (= 68 BVerfGE 82).
225
Ibid 371 (= 68 BVerfGE 82), emphasis added.
226
The translation in the ILR speaks of unilateral declarations of intent, ibid 372, but this
is not what the German original (68 BVerfGE 83) says, as there the phrase vlkerrechtliche
Willenserklrung is used, which means manifestation of will. That this is precisely what
the Court intended to say becomes clear also when looking at the examples provided by the
Court for such manifestations of will, such as the act of recognition or the denunciation of a
treaty, as these are not mere declarations of intent.

150 A History of Promises

when doing so.227 As a result of the Courts reasoning, the case was dismissed on the merits.
The decision of the German Constitutional Court is interesting in
this context, as it found Germany to have issued a unilateral binding
approval or permission, although it is regrettable that the Court has not
shed more light on the actual, final declaration made by Germany and the
precise communication which took place between the United States and
Germany. Had it done so, doubts which have been raised as to whether
the assurance should not have been viewed as giving rise to a treaty228
could be more adequately addressed. Proponents of the treaty view have
especially drawn blood from the word Zustimmung, here translated as
approval, which necessitated the existence of something which could
be approved. This something, so the argument ran, could logically not
have been the deployment itself, as it had not yet begun, but only the
United States intention to deploy. Hence, there must have been a manifestation of will by the United States which found a corresponding will on
the German side, with both together giving rise to an oral treaty. The declaration could furthermore not be seen as a promise, since Germany could
not by itself fulfil the approval, given that the missiles to be deployed
actually belonged to the United States.229 Neither argument, however, is
convincing. The latter can easily be refuted as, of course, Germany was
not promising to install the missiles (a promise which it nevertheless still
could have made, yet not have kept) but only to tolerate and allow such
deployment by the United States, should the latter decide to do so. Such
a promise, not to interfere with a certain action and to allow an act which
without such permission would infringe upon the declarants rights, can
surely be fulfilled by the declarant alone. The fact that there must have
been something that was approved similarly does not imply that the
approval occurred in the form of a treaty, as this something could very
well be the future deployment itself, following the decision adopted in
Brussels. Had Germany, for example, in the final approval stated: In
keeping with the decision adopted by NATO, we hereby allow the United
States to deploy the therein specified missiles on our territory, this would
have been a unilateral assurance lacking a corresponding manifestation
of will by the United States. As has been emphasised, the precise communication which has taken place between the parties is not illuminated
but left in the dark by the Constitutional Court. As it stands, however, the
227
See especially the dissenting opinion of Judge Mahrenholz, summary, ibid 388 (=
German original 68 BVerfGE 11132). On this point, see also Christian Tomuschat, Unilateral
Acts under International Law in Droits et Culture. Mlanges en lhonneur du Doyen Yadh Ben
Achour (Tunis, 2008) 1487, 149596.
228
See especially Theodor Schweisfurth, Die Zustimmung der Bundesregierung zur
Stationierung amerikanischer Mittelstreckenraketen in der Bundesrepublik Deutschland
(1974) 22 Archiv des Vlkerrechts 195, 196203.
229
Ibid 20002.

 Military and Paramilitary Activities in and against Nicaragua Case151

judgment forms just as important an example of state practice as the declaration itself, since the Court, which had the last say on the legal assessment of the declaration, left no doubt that it considered the approval to be
binding as a unilateral act upon Germany.
As the terms already indicate, the Court did not speak of a promise
or of an assurance (Versprechen or Zusicherung), which raises the question already touched upon, whether the case can be mentioned at all in
this context or rather involves a unilateral act of a different (and if so, its
own) category, as it can hardly be classified as a recognition of an already
existing legal situation or a (far-going) waiver of sovereign rights,230 let
alone a protest. There is, however, little reason not to include this example amongst the category of promises discussed here. The term approval
characterises only the obligation that has been undertaken, ie to allow
the deployment, but does not say how this approval has been brought
about, ie via a treaty or a unilateral act. If the latter is chosen, as the
Constitutional Court said it was, the unilateral act used is a promise: the
declaration promises that the deployment of, and the command over, foreign arms on German soil will be tolerated.
Q Military and Paramilitary Activities in and against Nicaragua Case
Returning to the ICJ, the next case after the Nuclear Tests cases which allegedly involved a promise was the case concerning Military and Paramilitary
Activities in and against Nicaragua, decided on the merits on 27 June 1986.
The judgment contains a passage in which some assurances made by
the (new) government of Nicaragua are more closely scrutinised, as the
United States considered them to be solemn commitments which had
subsequently been breached. The relevant facts in this respect were established by the Court to have been as follows:
On 23 June 1979, the Seventeenth Meeting of Consultation of Ministers of
Foreign Affairs of the Organization of American States adopted by majority,
over the negative vote of, inter alia, the representative of the then incumbent
Somoza government of Nicaragua, a resolution on the subject of Nicaragua.
After declaring that the solution of the serious problem is exclusively within the
jurisdiction of the people of Nicaragua, the Meeting of Consultation declared:
That in the view of the Seventeenth Meeting of Consultation of Ministers of
Foreign Affairs this solution should be arrived at on the basis of the following:
1. Immediate and definitive replacement of the Somoza regime.
2. 
Installation in Nicaraguan territory of a democratic government, the
composition of which should include the principal representative groups
which oppose the Somoza regime and which reflects the free will of the
people of Nicaragua.
230

For the distinction see pp 3436.

152 A History of Promises

3. Guarantee of the respect for human rights of all Nicaraguans without
exception.
4. The holding of free elections as soon as possible, that will lead to the
establishment of a truly democratic government that guarantees peace,
freedom, and justice.
On 12 July 1979, the five members of the Nicaraguan Junta of the Government
of National Reconstruction sent from Costa Rica a telegram to the SecretaryGeneral of the Organization of American States, communicating the Plan of
the Government of National Reconstruction to Secure Peace. The telegram
explained that the plan had been developed on the basis of the Resolution of the
Seventeenth Meeting of Consultation; in connection with that plan, the Junta
members stated that they wished to ratify (ratificar) some of the goals that
have inspired their government. These included, first:
our firm intention to establish full observance of human rights in our country in accordance with the United Nations Universal Declaration of the
Rights of Man (sic), and the Charter on Human Rights of the Organization of
American States; . . .
A further goal was:
the plan to call the first free elections our country has known in this century,
so that Nicaraguans can elect their representatives to the city councils and to
a constituent assembly, and later elect the countrys highest authorities.
The Plan to Secure Peace provided for the Government of National Reconstruction,
as soon as established, to decree a Fundamental Statute and an Organic Law, and
implement the Program of the Government of National Reconstruction. Drafts of
these texts were appended to the Plan: they were enacted into law on 20 July 1979
and 21 August 1979.231

The ICJ found the position of the United States on these events to be
reflected, inter alia, in a report submitted to Congress by President Reagan,
where it was stated that the United States was seeking the implementation of Sandinista commitment to the Organization of American States
(OAS) to political pluralism, human rights, free elections, non-alignment,
and a mixed economy.232
In a Congress resolution which the Court found also to express the view
of the US President on the matter, the latter being constitutionally responsible for the countrys foreign policy, it was stated that Congress:
supports the Nicaraguan democratic resistance in its efforts to peacefully
resolve the Nicaraguan conflict and to achieve the fulfilment of the Government
of Nicaraguas solemn commitments to the Nicaraguan people, the United
States, and the Organization of American States.233
231
ICJ, Military and Paramilitary Activities in and against Nicaragua (Nicaragua v United States),
Judgment of 27 June 1986, Merits, [1986] ICJ Rep 14, 8990, para 167.
232
Ibid 90, para 168.
233
Ibid 92, para 171.

 Military and Paramilitary Activities in and against Nicaragua Case153

The resolution hence linked US support for the Contras to alleged
breaches of what the United States referred to as Nicaraguas solemn
commitments. The Court therefore turned to the question whether the
conduct of Nicaragua could be held to have legally warranted countermeasures by the United States. The US assertion of a commitment having
been undertaken, according to the Court, first raised the question of the
possibility of a State binding itself by agreement in relation to a matter of
domestic policy, which was, however, not a real problem, as the ICJ could
not discover, within the range of subjects open to international agreement,
any obstacle or provision to hinder a State from making a commitment of
this kind.234 As the reader will have noted, in this passage the Court solely
contemplated a commitment by agreement and made no reference to a
commitment possibly assumed through a unilateral act. As will be seen, it
has, however, subsequently referred also to this possibility. In now turning to whether Nicaragua had in fact entered into a legal commitment by
communicating with and announcing its plan to the OAS, the ICJ found
the following:
However, the Court is unable to find anything in these documents, whether the
resolution or the communication accompanied by the Plan to secure peace,
from which it can be inferred that any legal undertaking was intended to exist.235

In explaining this conclusion, the Court elaborated that the part of the
resolution which listed certain foundations on which the solution of the
Nicaraguan domestic problems should be based included only recommendations to the future government, as it had also declared the solution
of the matter to be exclusively one for the Nicaraguan people. As such,
this passage could not be understood as including any formal offer which,
if accepted, would have become binding upon Nicaragua. In respect of
the Nicaraguan pledge to hold free elections, the ICJ continued as follows:
Nor can the Court take the view that Nicaragua actually undertook a commitment to organize free elections, and that this commitment was of a legal nature.
The Nicaraguan Junta of National Reconstruction planned the holding of free
elections as part of its political programme of government, following the recommendations of the XVIIth Meeting of Consultation of Foreign Ministers of
the Organization of American States. This was an essentially political pledge, made
not only to the Organization, but also to the people of Nicaragua, intended to be its
beneficiaries. But the Court cannot find an instrument with legal force, whether
unilateral or synallagmatic, whereby Nicaragua has committed itself in respect of
the principle or methods of holding elections.236

The Court thereby ruled out the possibility of any form of legal commitment, be it unilateral or synallagmatic. In doing so, the judgment can
234
235
236

Ibid 131, para 259.

Ibid 132, para 261.
Ibid 132, para 261, emphasis added.

154 A History of Promises

be understood in respect of unilateral pledges as already moving towards
the application of a more restrictive interpretation than the one applied in
the Nuclear Tests cases. The Court is, however, in the above-cited section
primarily concerned with the distinction between political and legally
binding declarations, and seems to be concentrating less on whether the
pledge made was unilateral or formed part of a bilateral undertaking. As
the judgment found no legal intention whatsoever to have been included
in the pledges, there was, of course, no need for any more detail in this
respect. The ICJ has nevertheless, as seen, included a brief reference discarding the pledge as a unilateral legal act and has, in its subsequent
ruling in the Frontier Dispute case (see below), expressly referred to the
present finding in defining the standard of interpretation that is to be
applied to unilateral declarations of states after the Nuclear Tests cases
holdings.
Judge Schwebel, in his dissenting opinion, disagreed with the Courts
position on this point. As opposed to the majority he considered Nicaragua
to have, in fact, undertaken a legal obligation, which (although some passages are ambiguous) appears to have been brought about through an
informal treaty. As Schwebel sees it, essentially, in exchange for the OAS
and its Members stripping the Somoza Government of its legitimacy and
bestowing recognition upon the Junta as the Government of Nicaragua,
the Junta extended specific pledges to the OAS and its Members, which
it bound itself to implement. Judge Schwebel, however, considers it
obvious that the Junta did not, by its written undertakings to the OAS
and its Members, conclude an international agreement in treaty form.
But, he continues, and by pointing to the Vienna Conventions Article 3
as well as to the Legal Status of Eastern Greenland case (not, however, to
the Nuclear Tests cases), argues that an international commitment binding
upon a state need not be made in written, still less particularly formal,
form. Instead: The question is simply, did the authority of the State concerned give an assurance, or extend an undertaking, which, in the particular circumstances, is to be regarded as binding upon it?. According to
Schwebel [w]hen a revolutionary government, soliciting recognition, has
given assurances to foreign governments, such assurances have repeatedly been treated by foreign governments as binding the revolutionary
government and its successors. In contrast to the above-mentioned section, these passages sound more as if focusing on the pledge as giving
rise to a unilateral commitment, and indeed the whole following paragraph speaks of the assurances which the Junta ratified, yet, apparently,
only in order to establish the Juntas intention to be legally bound by its
declaration. In the end, for Judge Schwebel, the legal bond seems to have
been brought about rather through an informal treaty than a unilateral
pledge, when he writes: The OAS offered recognition to the Junta on
bases which the Junta accepted. And while [t]he OAS and its Members

Filleting of Fish in the Gulf of St Lawrence Arbitration155

performed; the Government of Nicaragua did not. Not only was the creation of an international obligation clear; so was its breach.237
Especially in contrast to this dissenting opinion, the ICJs majority
judgment is hence of primary significance in highlighting the restrictive
standard of interpretation which will be applied to a states declaration in
ascertaining whether or not it can be taken to have manifested an intent
to be bound.
R Filleting of Fish in the Gulf of St Lawrence Arbitration
The dispute between Canada and France concerning filleting of fish in
the Gulf of St Lawrence is next in the timeline in the adjudication on
unilateral declarations. It was not brought before the ICJ but before an
arbitration tribunal which, however, confirmed the apparently uncontested understanding that statements made by a state representative
before a tribunal can, as such, become binding upon that state. Unlike in
the case of Great Britain before the ECtHR, but much like the other cases
mentioned above, the French representative did not emphasise his countrys intention to enter into a legal obligation when he repeatedly made
references to the existence of quotas, which restricted the amount of fish
that were caught by a filleting boat operating in the Gulf. The tribunal
stated that in the course of the proceedings of 4 June 1986, he had gone so
far as establishing that whether or not the filleting was done on board, the
boats will not capture a kilogram of codfish more than permitted by the
quota.238
The tribunal considered this statement as binding upon France in the
following way:
With regard to the circumstances in which it has been made, the Tribunal must
consider such declaration as engaging France to use all means in its position
to ensure, together with the Canadian authorities, that this declaration is
respected.239

As the ICJ has done with declarations made before it, the tribunal has
thereby imposed a very low threshold in order to establish an intention to
be legally bound in declarations made within formal proceedings.
237
See ICJ, Military and Paramilitary Activities in and against Nicaragua (Nicaragua v United
States), Merits, Dissenting Opinion of Judge Schwebel, [1986] ICJ Rep 259, 38485, paras
24748.
238
Arbitration Tribunal, Filleting of Fish in the Gulf of S. Lawrence Arbitration (1986) 265,
original French wording: ils ne captureront pas un kilogramme de morue de plus que le
quota autoris.
239
Ibid 265: Eu gard aux circonstances dans lesquelles elle a t faite, le Tribunal doit
considrer pareille dclaration comme engageant la France user de tous les moyens dont
elle dispose pour veiller, conjointement avec les autorits canadiennes, au respect de cette
dclaration.

156 A History of Promises

S Frontier Dispute Between Burkina Faso and Mali Case
Also in 1986, the ICJ rendered a judgment in the case concerning the Frontier
Dispute Between Burkina Faso and Mali on 22 December 1986, where it again
analysed an oral statement, this time a declaration made by Malis Head of
State during a press interview.240 The relevant facts were as follows.
The ongoing quarrel between Burkina Faso and Mali as to the precise
delimitation of their border had finally led to a conciliation attempt by the
Head of State of Somalia, then President of the Organization of African
Unity (OAU), and the President of Senegal. At an international meeting between Burkina Faso (then named Upper Volta), Mali and Togo on
26 December at Lom, the formation of a Mediation Commission composed of Togo, Niger, Guinea and Senegal was agreed upon. One of the
Commissions tasks was to seek a solution to the frontier dispute on the
basis of existing legal documents, for which it subsequently installed a
Legal Sub-Commission, entrusted with drawing up an initial proposal
on the matter. In an interview granted by Malis Head of State on 11 April
1975 to the France-Presse agency, he remarked that:
Mali extends over 1,240,000 square kilometres, and we cannot justify fighting for a scrap of territory 150 kilometres long. Even if the Organization of
African Unity decides objectively that the frontier passes through Bamako, my
Government will comply with that decision.241

Shortly after the Legal Sub-Commission had presented its report to the
Mediation Committee, the latter again gathered at Lom, where a final
communiqu was adopted, in which the Presidents of both countries, Mali
and Burkina Faso, participated. In it, the two parties now agreed to install
a neutral technical committee, to determine the exact location of certain
villages in order to establish the frontier and make proposals for its materialization to the Commission. Roughly a month later, both states during
a meeting at Conakry affirmed their common intention to do the utmost
to transcend the results achieved by the Mediation Commission of the
OAS, especially by facilitating the delimitation of the frontier between
the two States in order to place the final seal on their reconciliation.242
The work of the technical commission was, however, stalled when Mali
refused to grant the Commission the authorisation for overflights over its
territory, which were sought in order to make aerial photographs of the
frontier zone. It was in this situation that the case was brought before the
ICJ.
240
For the following account see ICJ, Frontier Dispute (Burkina Faso v Mali), Judgment of
22 December 1986, [1986] ICJ Rep 554,57074, paras 3440.
241
Ibid 571, para 36.
242
Ibid 572, para 37.

Frontier Dispute Between Burkina Faso and Mali Case157

While both parties agreed that the Mediation Committee did not have
the power to make binding decisions and had never actually completed
its work, Burkina Faso, inter alia, alleged that the Report of the Legal
Sub-Commission, as endorsed by the second summit meeting at Lom,
became binding for Mali on account of the above-cited statement, in which
its Head of State was said to have proclaimed his country bound by the
report subsequently adopted by the Mediation Commission. Mali rejected
any such interpretation by explaining that the comment was merely a
witticism of the kind regularly uttered at press conferences, which hence
implied no more than that Mali is anxious to consider the Commissions
recommendations with goodwill and in good faith.243
Faced with these two different interpretations, the Court considered the
legal value of the declaration to be as follows. First, it made it clear that
the declaration was not part of an oral treaty but, if anything, was binding
as a unilateral act upon Mali:
The statement of Malis Head of State on 11 April 1975 was not made during
negotiations or talks between the two Parties; at most, it took the form of a unilateral act by the Government of Mali.244

The ICJ continued by first of all confirming the potentially binding force
of such unilateral declarations:
Such declarations concerning legal or factual situations may indeed have the
effect of creating legal obligations for the States on whose behalf they are made,
as the Court observed in the Nuclear Tests cases. [reference omitted]

Still referring to the Nuclear Tests cases, the ICJ went on to emphasise
that:
it all depends on the intention of the State in question, and the Court emphasized that it is for the Court to form its own view of the meaning and scope
intended by the author of a unilateral declaration which may create a legal
obligation.245

By stressing that it is for the ICJ to conclude what a state intended at the
time, the Court therefore again underscored that it was going to make an
assessment based on the perception of an outside and (we may assume)
reasonable addressee. But as already pointed out in discussing its Nuclear
Tests cases judgments, it is largely the Court itself which, in this still quite
uncharted territory, is in the process of establishing in which cases a legal
commitment may reasonably be inferred from a unilateral declaration,
and where a declaration will count as a mere political pledge. The ICJ first
recalled its own holding in the above-mentioned Military and Paramiliatary
243
244
245

Ibid 573, para 38.

Ibid 573, para 39.
Ibid.

158 A History of Promises

Activities in and against Nicaragua case, in which no legal undertaking was
found to exist in the communication transmitted by the Junta to the OAS.
With this case in mind, the Court held:
The Chamber considers that is has a duty to show even greater caution when it
is a question of a unilateral declaration not directed to any particular recipient.246

This very cautious approach to be taken where a unilateral declaration

is not directed to a particular recipient as announced by the Court, however, inevitably created tensions with its earlier holdings in the Nuclear
Tests cases. Although the Court claimed it was following the latter in full,
there the ICJ had opined that the French statements not directed to a particular recipient were binding erga omnes declarations. But really, if the
rather loosely worded French statements were legal commitments, then
why should not Malis statement according to which its Government will
comply with that decision also be regarded as binding? In order not to
contradict itself, the Court seems to have had two choices: it could either
openly consider its finding of a legal intent in the Nuclear Tests cases to
have been inaccurate and not compatible with a truly restrictive inter
pretation coupled with the very cautious approach necessitated when
dealing with potential erga omnes statements, or it could try to distinguish
the Nuclear Tests cases declarations and the scenario in which they were
made from the statement now before it. As will be seen, the Court chose
the latter approach, yet without being very convincing. It continued as
follows:
In order to assess the intentions of the author of a unilateral act, account must
be taken of all the factual circumstances in which the act occurred. For example,
in the Nuclear Tests cases, the Court took the view that since the applicant States
were not the only ones concerned at the possible continuance of atmospheric
testing by the French Government, that Governments unilateral declarations
had conveyed to the world at large, including the Applicant, its intention effectively to terminate these tests. [reference to the Nuclear Tests cases omitted] In
the particular circumstances of those cases, the French Government could not
express an intention to be bound otherwise than by unilateral declarations. It
is difficult to see how it could have accepted the terms of a negotiated solution
with each of the applicants without thereby jeopardizing its contention that its
conduct was lawful. The circumstances of the present case are radically different. Here, there was nothing to hinder the Parties from manifesting an intention
to accept the binding character of the conclusions of the Organization of African
Unity Mediation Commission by the normal method: a formal agreement on
the basis of reciprocity. Since no agreement of this kind was concluded between
the Parties, the Chamber finds that there are no grounds to interpret the declaration made by Malis Head of State on 11 April 1975 as a unilateral act with
legal implications in regard to the present case.247
246
247

Ibid 574, para 39.

Ibid 574, para 40.

Frontier Dispute Between Burkina Faso and Mali Case159

The problem with this passage, intended to achieve a compatibility

between the holdings in the Nuclear Tests cases and the decision to be
made here, is twofold: while ambiguous in its last part, it is simply erroneous in the findings made at the beginning. To start with the beginning,
it is not convincing when the Court stipulates that France could not have
concluded a treaty with the applicants in the Nuclear Tests cases without
thereby jeopardising its contention that its conduct was lawful. Of course,
France could easily have done so, either by simply acceding to the Partial
Test Ban Treaty248 which outlawed atmospheric tests and had been ratified by New Zealand in 1963 and Australia in 1964 (and, inter alia, also
the United States, Great Britain and Russia), or by concluding a special
agreement with the two applicants. The undertaking of a new obligation
by France would not have had the slightest effect on the legality of its past
conduct. In addition, if the ICJ was correct in its assessment which it was
not the question has to be raised why the unilateral assumption of such
a new obligation should not have run into exactly the same problem. The
circumstances in the Frontier Dispute case before it were in this respect,
therefore, far from being radically different from those in the Nuclear
Tests cases.
This distinction, unconvincing as to the facts, now steered the Court
towards its problematically worded conclusion, which, reduced to its
core, reads as: Since no agreement . . . was concluded between the Parties,
. . . there are no grounds to interpret the declaration . . . as a unilateral act
with legal implications. In other words: no agreement, no unilateral commitment; yet, of course, if an agreement had been concluded, there would
have been no reason to interpret the subsequent declaration as an act
with important legal implications. With this being so, the Courts finding
would be nothing other than a sweeping conclusion, which stipulates that
where an agreement was not concluded although possible, subsequent
statements will not be interpreted as binding unilateral commitments; the
ICJ would thereby, in effect, have restricted the scope of unilateral undertakings to the very limited scenarios where treaties cannot be concluded,
whatever these may be.249 It is, however, highly doubtful whether the
Court really wanted to proclaim such a far-reaching rule of interpretation
and basically abolish the doctrine of unilaterally assumed undertakings
between states. Not only had the ICJ just declared that unilateral declarations may indeed have the effect the Court found them to have in the
248
See the Treaty Banning Nuclear Weapon Tests in the Atmosphere, in Outer Space and
Under Water (registration no A-6964) available at http://treaties.un.org.
249
For Alain Pellet, Commentary to Art. 38 ICJ Statute in Andreas Zimmermann,
Christian Tomuschat and Karin Oellers-Frahm (eds), The Statute of the International Court of
Justice: A Commentary (Oxford, 2006) 677, 707, MN 93, the conclusion is indeed that were this
precedent to be followed, the potential impact of unilateral acts as a source to be applied by
the Court would fade away.

160 A History of Promises

Nuclear Tests cases, but it had already moved towards the second step in
its analysis, that of interpreting a particular states declaration. Here, it
was faced with a press statement that was ambiguous as to whether a
legal undertaking was intended. The question is thus whether the ICJ, in
its above-cited no treaty, no promise conclusion, rather than pronouncing a general standard of interpretation that would discard even the most
clearly worded, formally and solemnly proclaimed pledge as a unilaterally assumed legal commitment, was not, in effect, only paying attention to the cases special circumstances? To recall, the question the Court
had to answer was whether the press interview statement that Malis
Government will comply with that decision was a legally binding commitment, or whether it had to be understood as a political pledge. One
important circumstance of the case was (apparently) the fact that Mali had
agreed to give the Mediation Commission only the power of recommendation but not to issue a legally binding decision. If it had wanted to do
otherwise, then why had it not done so in a binding agreement when it
had the chance? This, similarly to what had been said by the Court in the
North Sea Continental Shelf cases,250 seems to be the point the Court was
actually stressing when it said:
Here, there was nothing to hinder the Parties from manifesting an intention to
accept the binding character of the conclusions of the Organization of African
Unity Mediation Commission by the normal method: a formal agreement on
the basis of reciprocity.

In other words: since Mali had not agreed to provide the Mediation
Commission with the power to make legally binding decisions, its subsequent unilateral declaration could not be understood as having been
intended now to do so. Especially in light of the very cautious approach
now officially adopted by the Court, it is understandable to assume that
the declaration to comply with the decision rendered by a Commission,
which was not given the power to make legally binding decisions, is
unlikely to have been intended as a de facto unilateral conferral of such
legal powers to the Commission, absent a clear(er) and (more) precise
statement in this respect.
The subsequent ruling by the ICJ in the Armed Activities on the Territory
of the Congo case (addressed below) has confirmed that it is a careful legal
assessment which is used when the question is raised whether a unilateral
declaration is legally binding, and not a brief analysis of the question as to
why the parties refrained from concluding a treaty. What the ICJ has again
made very clear by referring to the Military and Paramilitary Activities in
and against Nicaragua judgment, and by in addition expressly underlining the necessity of using caution before finding a state to have made a
250

See p 114.

Negative Security Assurances161

binding (erga omnes) commitment, is that a restrictive reading has come to

govern the interpretation of a states declaration in order to conclude that
it has been made with the intention of becoming legally bound.
T Negative Security Assurances
Before again returning to the ICJ and in keeping with the chronological
order followed here, another interesting example of states employing
unilateral pledges will be considered: the so-called negative security
assurances. The latter have been sought especially by the non-aligned
members of the Non-Proliferation Treaty (NPT) and have been in the
centre of a lively debate, which has stretched over numerous decades.
As such, the circumstances in which the respective declarations were
made are complex, and a detailed account would exceed the space available here. A brief description can, however, be provided, especially as the
relevant facts have already been gathered by various authors and the territory is therefore fairly well charted.251
While states not in the possession of nuclear weapons, and especially
those signing the NPT, have been eager to reach an agreement under
which the nuclear weapon owning states guarantee not to threaten or
use nuclear weapons against them, the official nuclear powers have not
been willing to do so. Instead, they have only gone so far as making
unilateral declarations. A first set of these so-called negative security
assurances (as opposed to also existing positive assurances in which certain action in the event of an attack is promised) were made during the
Tenth Special Session of the UN General Assembly devoted to disarmament in 1978. Here, the Soviet Union, the United Kingdom and the United
States made official statements and China reiterated a declaration it had
already made in 1964. France was only prepared to give such assurances,
in accordance with arrangements to be negotiated but eventually, in 1982,
also followed suit.252 The declarations, however, were far from being uniform and were subject to different conditions. The UK representative, for
example, stated that:
251
See especially Thomas Bernauer, Nuclear Issues on the Agenda of the Conference on
Disarmament (New York, 1991); George Bunn, The Legal Status of U.S. Negative Security
Assurances to Non-Nuclear Weapon States (1997) 4(3) Nonproliferation Review 1; Allan Rosas:
Negative Security Assurances and Non-Use of Nuclear Weapons (1982) 25 GYIL 199; on the
Soviet declaration of 1982 also Henri Meyrowitz, Le dbat sur le non-recours en premier
aux armes nuclaires et la dclaration Sovitique du 15 Juin 1982 (1982) XXVIII AFDI 147.
252
In 1982, France announced that it will not use nuclear arms against a State that does
not have them and that has pledged not to seek them, except if an act of aggression is carried out in association or alliance with a nuclear-weapon State against France or against a
State with which France has a security commitment, cited in Bernauer, Nuclear Issues, n 251
above, 7.

162 A History of Promises

The United Kingdom is now ready formally to give such an assurance. I accordingly give the following assurance on behalf of my Government to non-nuclear
weapon States which are parties to the Non-Proliferation Treaty or other internationally binding commitments not to manufacture or acquire nuclear explosive devices: Britain undertakes not to use nuclear weapons against such states
except in the case of an attack on the United Kingdom, its dependent territories,
its armed forces or its allies by such a State in association or alliance with a
nuclear weapon State.253

The US declaration was similar254 and critics were quick to point out that
the exception included therein, according to which an attack by a state
which had a nuclear weapon state ally could lead to nuclear retaliation,
was very broad and included scenarios in which a non-nuclear attack by
a non-nuclear state was not supported, approved or even actually known
to its nuclear ally.
The declaration of the Soviet Union was different:
From the rostrum of the UN special session our country declares that the Soviet
Union will never use nuclear weapons against those States which renounce the
production and acquisition of such weapons and do not have them on their
territories.255

Yet, different as it was, the exceptions contained therein were felt

to be again problematic, since the non-deployment condition within
this pledge would allow the Soviet Union to employ nuclear weapons
against a state that had not itself used, nor even had control over, nuclear
weapons stationed on its territory by one of its allies. In 1982, the Soviet
Union further clarified its undertaking by adding that it had assumed
an obligation not to be the first to use nuclear weapons.256 China, too,
had in 1978 reiterated that at no time and in no circumstances will it be
the first to use nuclear weapons.257 Still, the non-nuclear weapon states
were not satisfied. As the Austrian Public International Law Department
(Vlkerrechtsbro) stressed in an internal analysis, even these no first use
assurances were not a promise that a non-nuclear weapon state would
253
See Official Records of the General Assembly, Tenth Special Session, Plenary Meetings,
26th Meeting, para 12 as cited in Bernauer, Nuclear Issues, n 251 above, 9 and Hafner,
sterreichische diplomatische Praxis, n 219 above, 301.
254
The United States will not use nuclear weapons against any non-nuclear-weapon State
party to the Treaty on the Non-Proliferation of Nuclear Weapons or any comparable internationally binding commitment not to acquire nuclear explosive devices, except in the case
of an attack on the United States, its territories or armed forces, or its allies, by such a State
allied to a nuclear-weapon State, or associated with a nuclear-weapon State in carrying out
or sustaining the attack, UN Doc A/S-10/AC.1/30, cited in Hafner, sterreichische diplomatische Praxis, n 219 above, 301.
255
Official Records of the General Assembly, Tenth Special Session, Plenary Meetings, 5th
Meeting, paras 84 and 85, cited in Bernauer, Nuclear Issues, n 251 above, 9 and fully in Hafner,
sterreichische diplomatische Praxis, n 219 above, 301.
256
See the citation in Bernauer, Nuclear Issues, n 251 above, 8.
257
Ibid.

Negative Security Assurances163

never have to face nuclear weapons, as they would allow the respective
nuclear power to use its weaponry in a multilateral conflict even against
a non-nuclear weapon country, as long as another state had already made
use of a weapon from within its nuclear arsenal.258 China in 1982 hence has
to be considered as going a step further when it stated that it undertakes
unconditionally not to use or threaten to use nuclear weapons against
non-nuclear countries and nuclear free zones.259
The legal bindingness of all these declarations has been analysed
by various scholars and was evidently of interest to the declarations
addressees. Allan Rosas in his article concludes that the assurance by the
United Kingdom and the United States seems to be intended to be legally
binding260 especially as the declarants had stressed that the formal status
of the assurances would not be enhanced by concluding a convention.261
While authoritative comments showed that the Soviet Union had initially
not intended to commit itself legally, its position is said to have changed
by 1980.262 At least in 1982 this was apparently the case, as the country, as
cited, proclaimed to have undertaken an obligation not to be the first to
use nuclear weapons. The already mentioned international law department
within the Austrian Foreign Ministry, in its opinion published in 1980, drew
the conclusion that the statements made by the United States, the United
Kingdom and the Soviet Union in 1978 had to be considered as binding
commitments.263 Another scholar, Thomas Bernauer, in his analysis of the
debates held in various fora, including the UN General Assembly and the
Conference on Disarmament, also identified the main weakness of the
19781982 declarations not in them being perceived as non-binding, but
in the exceptions included therein, exceptions which, en plus, varied from
declaration to declaration.264 With regards to their unilaterality, states are
said to have clearly favoured a treaty and to have continued to push for it,
as some of them felt that although legally binding, unilateral declarations
could be amended and withdrawn more easily than a treaty could: a treaty
would hence be more binding.265 While the Soviet Union is said to have
been ready to conclude a treaty, the United States is said to have opposed it,
as it would probably be impossible to find common language to cover the
differing policies of the five (official) nuclear weapon countries.266 Western
Hafner, sterreichische diplomatische Praxis, n 219 above, 303.
A/S-12/11 as cited in Bernauer, Nuclear Issues, n 251 above, 7.
260
Rosas, Negative Security Assurances, n 251 above, 208.
261
Ibid, with further references.
262
Ibid.
263
Hafner, sterreichische diplomatische Praxis, n 219 above, 303. As the Chinese declaration of 1978 pointed to an earlier declaration, in the eyes of the department, it did not
qualify as a unilaterally binding declaration; instead, the statement to which it made reference would have to be analysed, see ibid.
264
Bernauer, Nuclear Issues, n 251 above, 9
265
Ibid 10 and 1920.
266
See Bunn, U.S. Negative Security Assurances, n 251 above, 6.
258
259

164 A History of Promises

countries are reported as having accepted this premise, while emphasising that what they had received were firm, credible and reliable commitments.267
Continued efforts by the non-nuclear states in 1995, when the NPT was
extended, led to new (in part merely reiterating) declarations made by
each of the official nuclear powers. These were circulated as official documents of the UN General Assembly and the Security Council,268 which, in
its unanimously adopted Resolution 984 (1995), took note with appreciation of the statements made.269 As France had indicated when submitting
its declaration, the nuclear powers had sought to harmonise the content
of the negative security assurances, with the result that the French, US,
Russian and UK declarations were now practically identical. Yet each still
included an exception which, however, was not as broadly worded as the
original one, since now the attack itself (so it appeared) had to be carried
out or sustained in association or alliance with a nuclear weapon state.
The very similar declarations (as reaffirmed by the United States and
reaffirmed and clarified by France) read as follows (here the Russian
one):
Russian Federation will not use nuclear weapons against non-nuclear-weapon
States parties to the Treaty on the Non-Proliferation of Nuclear Weapons, except
in the case of an invasion or any other attack on the Russian Federation, its territory, its armed forces or other troops, its allies or on a State towards which it has
a security commitment, carried out or sustained by such a non-nuclear-weapon
State in association or alliance with a nuclear-weapon State.270

China remained true to its already assumed broader obligation when it

not only again stated that it undertakes not to be the first to use nuclear
weapons at any time or under any circumstances, but in addition:
China undertakes not to use or threaten to use nuclear weapons against nonnuclear-weapon States or nuclear-weapon-free zones at any time or under any
circumstances. This commitment naturally applies to non-nuclear weapon
States parties to the Treaty on the Non-Proliferation of Nuclear Weapons or
non-nuclear weapon States that have entered into any comparable internationally binding commitment not to manufacture or acquire nuclear explosive
devices.271

The wording within the Chinese declaration is the clearest and most
precise. China undertakes what it has proclaimed not to do and speaks
267
268

155.

See CD/1039, 328 as cited in Bernauer, Nuclear Issues, n 251 above, 20.
See S/1995/261, S/1995/262, S/1995/263, S/1995/264 and S/1995/265 and A/50/151-

269
See S/RES/984 (1995). On the resolutions positive elements and drawbacks see Nabil
Elaraby, Some Reflections on Disarmament in Christian Tomuschat (ed), The United Nations
at Age Fifty: A Legal Perspective (The Hague, 1995) 1920.
270
S/1995/261, 3, Annex II.
271
S/1995/265, 2, para 2.

Negative Security Assurances165

of a commitment in this respect, to be adhered to at any time and under

any circumstances. The other declarations do not go as far, although they
also include a firm will not use in their wording. The long-lasting debates
over the declarations precise wording, along with their very formal proclamation to the General Assembly and the Security Council, would also
indicate that they may be relied upon as legal undertakings. The crucial
question, of course, is which standard of interpretation is to be applied to
declarations, which fall short of expressing outright that a state is hereby
assuming a legal obligation. Clearly, under the Nuclear Tests cases standard, the repeatedly made, clearly worded and formally proclaimed declarations would qualify as legal undertakings. Even under the standard
in the Military and Paramilitary Activities in and against Nicaragua or the
Frontier Dispute cases, where a more restrictive approach has been used,
this is likely to be the case. Especially with regard to the Chinese declaration, one would be hard-pressed to find a declaration which has more
precise and clear wording, unless of course an official announcement to
undertake a legal obligation is considered necessary. The ICJ has, however, never required this, neither in the Nuclear Tests cases, where it found
a legal obligation to have arisen from the French assurances, nor in any
of the other cases in order to reject the bindingness of a statement before
it. Indeed, the ICJ in its Advisory Opinion on the Legality of the Threat or Use
of Nuclear Weapons seems to have placed the declarations, which it analysed in its opinion alongside the respective treaties, on the same footing
as other law and treaty obligations applicable in these scenarios:
A threat or use of nuclear weapons should also be compatible with the requirements of the international law applicable in armed conflict, particularly those
of the principles and rules of international humanitarian law, as well as with
specific obligations under treaties and other undertakings which expressly deal with
nuclear weapons.272

The ICJ had previously addressed the negative security assurances

and they are very clearly included as other undertakings in this quotation.273 But, of course, the argument that the ICJ apparently concluded
that the declarations were legally binding commitments274 is crucially
weakened by its use of the wording should be compatible, which should
then rather read must be compatible. Similarly, a statement made by the
recipients of the 1995 declarations at the Conference of the Parties to the
Non-Proliferation Treaty is support for the fact that at least they did not
perceive the statements of 1995 to be made as legally binding commitments. Decision 2, paragraph 8 as adopted reads:
272
ICJ, Legality of the Threat or Use of Nuclear Weapons, Advisory Opinion of 8 July 1996,
[1996] ICJ Rep 266, para 105(2)(D), unanimously adopted, emphasis added.
273
Ibid 25153, see especially para 63.
274
For this interpretation see Bunn, U.S. Negative Security Assurances, n 251 above, 9.

166 A History of Promises

Noting United Nations Security Council Resolution 984 (1995), which was
adopted unanimously on 11 April 1995, as well as the declarations of the
nuclear-weapon States concerning both negative and positive security assurances, further steps should be considered to assure non-nuclear-weapon States
party to the Treaty against the use or threat of use of nuclear weapons. These
steps could take the form of an internationally legally binding instrument.275

Indeed, according to Cedeo, the attitude of the authors and the positions of most States appear to reflect the political nature of these statements.276 Particularly if the ICJs decisions in the Nuclear Tests cases is to
be followed, this perception would, however, not hinder an international
court from drawing the conclusion that the assurances received, on account
of their being clear and specific commitments formally made, are already
included in an internationally binding legal instrument, namely that of a
promise in international law. Although at present the precise status of these
declarations is thus subject to some debate, under the ICJs jurisprudence,
good reasons speak for interpreting them as binding commitments.
Most recently, in its Nuclear Posture Review Report of April 2010, the
United States under the Obama administration has chosen to further
refine its assurance by including the following passage:
To that end, the United States is now prepared to strengthen its long-standing
negative security assurance by declaring that the United States will not use or
threaten to use nuclear weapons against non-nuclear weapons states that are
party to the Nuclear Non-Proliferation Treaty (NPT) and in compliance with
their nuclear non-proliferation obligations.277

U LaGrand Case
Returning to the ICJ, and now in 2001, Germany in the LaGrand case
against the United States in its fourth submission asked the Court to:
adjudge and declare that the United States shall provide Germany an assurance that
it will not repeat its unlawful acts and that, in any future cases of detention or
of criminal proceedings against German nationals, the United States will ensure
in law and practice the effective exercise of the rights under Article 36 of the
Convention on Consular Relations. In particular in cases involving the death penalty, this requires the United States to provide effective review of and remedies
for criminal convictions impaired by a violation of the rights under Article 36.278
275
NPT/Conf.1995/32 (Part I), Annex, Decision 2, para 8, available at www.un.org/
Depts/ddar/nptconf/2142.htm, emphasis added.
276
Victor Rodriguez Cedeo, Eighth Report on Unilateral Acts of States, A/CN.4/557
(2005) 21, para 115.
277
United States Department of Defense, Nuclear Posture Review Report (April 2010) 15, 17,
available at www.defense.gov/npr/docs/2010 Nuclear Posture Review Report.pdf.
278
See ICJ, LaGrand (Germany v United States), Merits, Judgment of 27 June 2001, [2001] ICJ
Rep 466, 5089, para 117, emphasis added.

LaGrand Case167

As Germany had explained in its memorial, what it was seeking were

assurances and guarantees of non-repetition279 to be provided by the
United States in order to prevent future breaches of the Vienna Consular
Convention. It is the assurances that Germany was asking for which are
of primary interest here, rather than the guarantees that were requested
to be imposed on the United States by the Court. As Simma, agent for
Germany at the time, explained in the oral hearings before the Court:
while assurances are normally given verbally, guarantees of non-repetition go
beyond that and involve certain preventive actions to be taken by the responsible
State and specifically designed to avoid repetition.280

The term guarantee, used in relation to concrete action and not mere
words,281 led to some debate between the parties, and was later dropped
in the fourth submission which, as quoted above, no longer made use of
that term. In substance, however, the application continued to include a
request for the imposition of certain measures, ie guarantees other than
verbal assurances, especially in its second part. As to the assurances
sought, Germany requested formal assurances from the United States:
The German request for formal assurances is appropriate in the present Case
if only because it will be decided by the International Court of Justice after a
formal procedure. In addition, since all informal requests of Germany, and
even the formal Order of the Court on Provisional Measures were ignored by
the United States, Germany cannot be content any longer with mere informal
assurances on the part of the United States.282

The United States responded by stressing in particular that for the ICJ
to follow this submission would, in the eyes of the United States, mean
that it was imposing a new obligation on it: An assurance or guarantee requires the creation of a new obligation with independent legal
significance.283 Hence, the Court should reject Germanys invitation to
confer upon Germany new or additional rights beyond those existing
under the Consular Convention.284
The ICJ responded in its judgment to Germanys fourth submission by
cutting it into two sections, because, so it concluded, only the first part
279
See LaGrand, ibid, Memorial of the Federal Republic of Germany, 16 September 1990,
para 6.60 et seq, available at www.icj-cij.org/docket/files/104/8552.pdf.
280
LaGrand, ibid, Verbatim Record of the Oral Hearing held on 13 November 2000 at 3 pm,
34, para 25, available at www.icj-cij.org/docket/files/104/4653.pdf, emphasis added.
281
Germany said it was following the distinction which is to be found in Art 30 of the
ILC Articles on State Responsibility, see LaGrand, ibid, Verbatim Record of the Oral Hearing
held on 16 November 2000 at 10 am, 36, para 11, available at www.icj-cij.org/docket/
files/104/4667.pdf.
282
LaGrand, ibid, Memorial of the Federal Republic of Germany, 16 September 1990, para
6.70, as cited at n 279 above.
283
LaGrand, ibid, Verbatim Record of the Oral Hearing held on 14 November 2000 at 3 pm,
para 5.24, available at www.icj-cij.org/docket/files/104/4663.pdf.
284
Ibid para 5.25.

168 A History of Promises

of its first sentence could be understood as seeking a straightforward
assurance,285 ie the part which read: the United States shall provide
Germany an assurance that it will not repeat its unlawful acts. With
respect to this request, the Court first noted the various efforts undertaken
by the United States, all of which were insufficient in the eyes of Germany.
It went on to note that the United States had acknowledged the fact that it
had not complied with its obligations under the Convention and had presented an apology. While an apology was insufficient for the Court, it took
note of the fact that the United States had repeatedly and in all phases
made reference to a vast and detailed programme established in order to
ensure compliance with its obligations under the Convention. This led the
ICJ to pronounce the following:
The United States has provided the Court with information, which it considers
important, on its programme. If a State, in proceedings before this Court, repeatedly refers to substantial activities which it is carrying out in order to achieve
compliance with certain obligations under a treaty, then this expresses a commitment to follow through with the efforts in this regard . . . The Court considers
that the commitment expressed by the United States to ensure implementation of
the specific measures adopted in performance of its obligations under Article 36
paragraph 1(b) [of the Vienna Consular Convention] must be regarded as meeting Germanys request for a general assurance of non-repetition.286

With respect to the other assurances (really still guarantees under the
terminology above) sought by Germany, the Court saw no need to impose
any further obligation on the United States, as there was no special muni
cipal law that was inconsistent with the Convention and the latter, as such,
already compelled the United States to allow the review and reconsideration of a conviction, where the Convention had not been followed. The
Court consequently limited itself to the above finding. This was included
in the operative part of the judgment, where the Court:
Unanimously, [t]akes note of the commitment undertaken by the United States
of America to ensure implementation of the specific measures adopted in performance of its obligations under Article 36, paragraph 1(b), of the Convention;
and finds that this commitment must be regarded as meeting the Federal
Republic of Germanys request for a general assurance of non-repetition.287

The ICJ, therefore, rather than choosing to hold the United States
obliged to make a formal assurance of non-repetition, found the United
States to have, by its own repeated referrals to its substantial activities,
expressed such a commitment. That this commitment is an additional
legal undertaking is not made absolutely clear. But the interpretation that
it must be so, is the only one in line with the Courts finding that it met the
285
286
287

See ICJ, LaGrand, Merits, Judgment (2001), n 278 above, 511, para 120.
Ibid 51213, para 124.
See ibid 516, operational para (6).

Armed Activities on the Territory of the Congo Case169

German request for a formal and reliable assurance, as opposed to a mere

informal pledge, followed only by an apology if breached. As seen, the
United States had expressly stated its understanding that the requested
assurance, which was now found to have been replaced by the commitment undertaken, would give rise to a new obligation. By deducing a unilateral legal undertaking from the repeated statements made during the
proceedings, the ICJ has thereby again applied its rather non-restrictive
approach to declarations made before it. It should, however, be noted that
the Court, while apparently finding a legal commitment to exist for the
United States, only found it to be obliged to follow through with specific
measures which it had claimed it had already adopted and nothing more.
V Armed Activities on the Territory of the Congo Case
That the finding of a legal commitment is indeed subject to a more restrictive approach when declarations are made outside the courtroom has
already been shown by the ICJ in Armed Activities in and against Nicaragua
and especially in the Frontier Dispute case. In its ruling on the case concerning Armed Activities on the Territory of the Congo, the Court has confirmed this while shedding more light on how the legal bindingness of a
declaration will be assessed.
The Democratic Republic of the Congo (DRC) had instituted proceedings against the Republic of Rwanda in respect of a dispute that it claimed
to include massive, serious and flagrant violations of human rights and
of international humanitarian law. In order to establish the jurisdiction of
the Court, the DRC relied in particular on Article 36(1) of the ICJ Statute288
in conjunction with various treaty provisions, amongst them Article IX of
the Genocide Convention.289 The problem with this provision as opening
the avenue to the Court was that although Rwanda was a party to the
Convention, it had included a reservation within its instrument of accession, according to which it did not consider itself bound by said Article
IX. The DRC tried to overcome this obstacle by pointing to a Dcret-loi
adopted by Rwanda, its first article announcing all reservations entered
by the Rwandese Republic in respect of the accession, approval and ratification of international instruments are withdrawn. In addition, the DRC
288
ICJ Statute, Art 36(1) reads: The jurisdiction of the Court comprises all cases which the
parties refer to it and all matters specially provided for in the Charter of the United Nations
or in treaties and conventions in force.
289
Article IX of the Convention on the Prevention and Punishment of the Crime of
Genocide provides that Disputes between the Contracting Parties relating to the interpretation, application or fulfilment of the present Convention, including those relating to the
responsibility of a State for genocide or for any of the other acts enumerated in Article III,
shall be submitted to the International Court of Justice at the request of any of the parties to
the dispute.

170 A History of Promises

cited a statement made by the Rwandese Minister of Justice which corroborated her countrys withdrawal of the above reservation as constituting
a unilateral commitment having legal effects in regard to the withdrawal
of that reservation.290
In its ruling, the Court first dismissed the alleged international effect
of the Rwandese Dcret-loi as amounting to a withdrawal of the abovecited reservation. A withdrawal of the reservation would have to have
been notified at the international level, which Rwanda, despite passing
its municipal Dcret-loi, had not done. The Court then turned to the effect
of the unilateral statement made by Ms Mukabagwiza, the Rwandan
Minister of Justice at the time, on 17 March 2005 before the 61st Session of
the United Nations Commission on Human Rights:
Rwanda is one of the countries that has ratified the greatest number of international human rights instruments. In 2004 alone, our Government ratified ten
of them, including those concerning the rights of woman, the prevention and
repression of corruption, the prohibition of weapons of mass destruction, and
the environment. The few instruments not yet ratified will shortly be ratified and past
reservations not yet withdrawn will shortly be withdrawn.291

The Court, in its analysis, began by assessing the defence advanced by

Rwanda that Ms Mukabagwiza, on account of her position as a mere
Minister of Justice, could not have bound her state to lift a particular reservation. As a starting point, the ICJ established that the General Rule
included for the law of treaties in Article 7 of the Vienna Convention on the
Law of Treaties was equally applicable to a unilaterally assumed obligation:
In this connection the Court observes that, in accordance with its consistent
jurisprudence [references to numerous cases, beginning with the Nuclear Tests
cases and including the Legal Status of Eastern Greenland case, as well as others not expressly relating to unilateral acts but to the competence of the acting authority, omitted] it is a well established rule of international law that the
Head of State, the Head of Government and the Minister of Foreign Affaires
are deemed to represent the State merely by virtue of exercising their functions,
including the performance, on behalf of said State, of unilateral acts having the
force of international commitments.292

The Court then noted that:

with increasing frequency in modern international relations other persons
representing a State in specific fields may be authorized by that State to bind
it by their statements in respect of matters falling within their purview. This
may be true, for example, of holders of technical ministerial portfolios exercis290
See ICJ, Armed Activities on the Territory of the Congo (Democratic Republic of the Congo
v Rwanda), New Application 2002, Jurisdiction and Admissibility, Judgment of 3 February
2006, 26, para 45.
291
Ibid 27, para 45, emphasis added.
292
Ibid para 46.

Armed Activities on the Territory of the Congo Case171

ing powers in their field of competence in the area of foreign relations, and even
of certain officials.293

Taking these considerations into account, the ICJ noted that the statement
by Ms Mukabagwiza had been made while she was speaking in her official
capacity before the United Nations. Also, she had indicated that she was
making her statement, as the Court quoted, on behalf of the Rwandan
people.294 Since the area of human rights fell within the purview of her
position, the Court concluded that the possibility cannot be ruled out in
principle that a Minister of Justice may, under certain circumstances, bind
the State he or she represents by his or her statements.295
The ICJ then, with reference to the Nuclear Tests cases and the Frontier
Dispute case, confirmed that in ascertaining whether a declaration was
binding, its actual content as well as the circumstances in which it was
made296 had to be examined. It further recalled that a statement of this
kind can create legal obligations only if it is made in clear and specific
terms.297
Applying these principles, the Court noted that the declaration now
before it was indeterminate in that it had to be understood as referring to
international instruments in a broad sense, and not to a particular treaty
or reservation or at least merely to human rights treaties, as the preceding sentence included a reference not only to human rights but also to the
environment. The Court also observed that by indicating past reservations . . . will shortly be withdrawn, Rwanda had not precisely specified
when the withdrawal was to take place. For the ICJ:
It follows from the foregoing that the statement by the Rwandan Minister of
Justice was not made in sufficiently specific terms in relation to the particular question of the withdrawal of reservations. Given the general nature of
its wording, the statement therefore cannot be considered as confirmation by
Rwanda of a previous decision to withdraw its reservation to Article IX of the
Genocide Convention, or as any sort of unilateral commitment on its part having legal effects in regard to such withdrawal; at most, it can be interpreted as a
declaration of intent, very general in scope.298

After analysing the declarations wording as not being sufficiently

specific, the Court addressed the circumstances of its making, which it
said corroborated its finding, as the context was that of a presentation of
general policy on the promotion and protection of human rights.299
Ibid para 47.
Ibid 28, para 48.
Ibid.
296
Ibid para 49.
297
Ibid para 50.
298
Ibid 29, para 52.
299
Ibid para 53.
293
294
295

172 A History of Promises

By carefully scrutinising the declaration in this way, the Court has made
it slightly more transparent as to how the bindingness of a declaration
will be assessed. As seen, the starting point in the legal assessment was
the declarations wording, analysed in terms of it having to be both clear
and specific. Although the ICJ did not again expressly stress this aspect, it
remained true to the restrictive interpretation and the cautious approach
applicable to potential erga omnes declarations in deciding that the declaration now before it did not live up to the clear and specific standard.
The statement past reservations not yet withdrawn will shortly be withdrawn, is, after all, not really that unclear, even though it leaves room
for the Court to interpret (while not making it impossible for the Court to
specify) what is reasonably to be understood by shortly. Similarly, that
past reservations might relate to all reservations does not necessarily
indicate that the statement is indeterminate or unspecific, as it may have
meant precisely that, ie that all reservations will be withdrawn. Maybe
it did, but it is here where the ICJs restrictive approach kicks in, which
leads it to negate a broad promise and to instead opt for, at most, a
general declaration of intent. In addition and although the statements
direct context, a presentation of general policy, is only cited as the circumstance which further supported the Courts finding, the ICJ is likely
to have already been influenced by the general character of the remarks
surrounding the particular statement before it, when it identified the latters indeterminate character.
W Questions Relating to the Obligation to Prosecute or Extradite Case
In the case concerning Questions Relating to the Obligation to Prosecute or
Extradite, Belgium brought a claim against Senegal before the ICJ. Belgium
alleged that Senegal was violating its obligations under the Convention
Against Torture by refraining from either prosecuting or extraditing
Hissne Habr, the former President of Chad. The latter was held under
house arrest in Dakar by Senegalese authorities, while being sought by
Belgium in order to face charges in particular for torture and crimes
against humanity. On 19 February 2009, Belgium submitted a request for
the indication of provisional measures to the Court. In it, Belgium argued
that it was facing a risk of irreparable prejudice to the rights asserted in its
claim, as the President of Senegal, in an interview given to Radio France
on 2 February 2009, had indicated the possibility that his country might
lift the house arrest of Habr, should Senegal fail to raise the budget necessary to hold his trial. As a release, in the eyes of Belgium, might allow
Habr to evade prosecution altogether, and would make it impossible
for Senegal to comply with its obligations under the Convention Against
Torture, it asked the Court:

 Questions Relating to the Obligation to Prosecute or Extradite Case173

to indicate . . . provisional measures requiring Senegal to take all the steps
within its power to keep Mr H Habr under the control and surveillance of the
judicial authorities of Senegal.300

In the oral proceedings, Senegal, however, emphasised that it had no

intention of letting Mr Habr go and instead intended to keep him under
surveillance. In response to a question by a member of the Court during
the hearings, Belgium:
indicated that a solemn declaration made before the Court by the Agent of
Senegal, in the name of its Government, could be sufficient for Belgium to consider that its Request for the indication of provisional measures had no further
raison dtre, provided that such a declaration would be clear and unconditional,
and that it would guarantee that all the necessary measures would be taken by
Senegal to ensure that Mr Habr did not leave Senegalese territory before the
Court delivered its final Judgment.301

Belgium expressed the wish that the Court should place such a declaration in the operative paragraph of its order. The Co-Agent of Senegal
answered a question by the Court in this regard by making the following
declaration at the end of the hearings:
Senegal will not allow Mr Habr to leave Senegal while the present case is
pending before the Court. Senegal has no intention to allow Mr Habr to leave
the territory while the present case is pending before the Court.302

The Court noted that Senegal had given a formal assurance, both proprio motu and in response to a question put by a member of the Court on
several occasions during the hearing, and taking note of the assurances
given by Senegal, finds that, the risk of irreparable prejudice to the rights
claimed by Belgium is not apparent at the date of this Order.303
In light of the circumstances in which the declaration was made, ie in
order to bring the interim proceedings to an end, and provide Belgium
with an equivalent to a binding interim order by the ICJ,304 there can
be little doubt that the solemn, clear and unconditional declaration that
Senegal would not allow Habr to leave the country while the case was
before the Court, was binding upon it as (in the words of the ICJ) a formal
assurance, ie a promise under international law.

300
ICJ, Questions relating to the Obligation to Prosecute or Extradite (Belgium v Senegal), Order
of 28 May 2009, para 15.
301
Ibid para 33.
302
Ibid para 68.
303
Ibid para 72.
304
On the bindingness of such orders see ICJ, LaGrand, Merits, Judgment (2001), n 278 above,
50106, paras 98109.

174 A History of Promises

II CONCLUDING REMARKS ON THE HISTORY OF PROMISES

The History of Promises has shown a slow evolution, which was primarily driven by the decisions of international courts and has led to the
establishment of a new mechanism for states to create a legally binding
commitment for themselves in the form of unilateral assurances, commonly dubbed promises under international law. What began with findings that unilateral pledges made during court proceedings, despite their
not resulting in the conclusion of a treaty, could be relied on as legally
binding commitments, was followed by a phase in which the unilateral
or bilateral character of an in any event binding declaration was either
declared irrelevant or left open, in holding that a legal commitment had
been brought about, despite the fact that the unilateral or at least nontreaty character of the declarations had been raised by those involved.
The resulting debate within legal doctrine on the bindingness of unilateral
assurances was ventilated by state practice, such as the Allied consensus
that the breach not only of treaties and agreements prohibiting war, but
in addition that of German assurances not to attack was to be punishable
as a crime against peace, or the Austrian declaration of neutrality, and
finally, and most clearly, the Egyptian declaration on the Suez Canal. The
question as to the precise value of such statements and the legal rules governing their execution was by now clearly posed and finally answered by
the Nuclear Tests cases, in the manner above depicted. The rather special
circumstances in which the ICJ decided on the two applications before it
has not hindered the judgment from leaving its clear mark on subsequent
state practice, as following examples in this chapter have shown, and the
Court has further strengthened its holding by confirming its principal doctrinal footing that states may oblige themselves by unilateral assurances.
Two subsequent cases, Frontier Dispute but especially Armed Activities on
the Territory of the Congo, have shed more light on the restrictive interpretative standard that the Court currently applies to declarations made by
state representatives outside judicial proceedings.
This development in international jurisprudence and its impact on
state practice are by now reflected in the writings of numerous scholars
who have identified promises as capable of committing states in their
legal affairs.305 The most recent joint effort in this area was that of the ILC,
whose finally adopted Guiding Principles have been restricted to unilateral declarations capable of creating legal obligations for the declaring
state, which as the Commentary shows, is based mainly on the ICJs decisions in the Nuclear Tests cases as supported, illustrated and interpreted
in its subsequent holdings in the Military and Paramilitary Activities in and
305

See Introduction, nn 49 and 50.

 Questions Relating to the Obligation to Prosecute or Extradite Case175

against Nicaragua and Frontier Dispute cases, as well as the Armed Activities
on the Territory of the Congo case. That the existence of promises understood as a unilateral and legally binding commitment of a state to act or
refrain from acting in a certain manner in the future has by now become an
accepted reality, however, does not mean that the rules pertaining thereto
are clear. The question of which legal norms are applicable to such statements is the subject matter which will be addressed in the next chapter.

3
The Law on Promises

S THE FOREGOING chapter revealed, the ICJ has by now had

the opportunity to express its view on various important aspects
pertaining to the legal framework operable for state promises,
ranging from the lack of formal requirements, the good faith principle as providing the basis of their binding force, to the question which
authorities can commit their state by making unilateral assurances. The
subject, nevertheless, remains fraught with some considerable uncertainties, as becomes apparent when turning to the work conducted by the
International Law Commission in this area, and especially the published
results, as the latter almost exclusively reflect the ICJs dicta. But before
going into more detail in respect of the legal framework, a question which
also raises some methodological concerns needs to be addressed. It has
to be asked how the law on promises may be described and illuminated
in light of the fact that the existence of promises has never been established on the basis of any of the (primary) sources of international law, as
reflected by Article 38 of the ICJ Statute.
I PROMISES AND THE SOURCES OF INTERNATIONAL LAW

Article 38 of the ICJ Statute, known to every public international lawyer,

determines what sources the ICJ is to apply in cases brought before it and,
as such, is widely perceived as reflecting the sources of international law.1
The wording of its first paragraph will be recalled to read as follows:
Article 38(1)
The Court, whose function is to decide in accordance with international law
such disputes as are submitted to it, shall apply:
1
See especially Michel Virally, The Sources of International Law in Max Srensen
(ed), Manual of Public International Law (New York/London, 1968) 116, 121; Alain Pellet,
Commentary to Art. 38 ICJ Statute in Andreas Zimmermann, Christian Tomuschat and Karin
Oellers-Frahm (eds), The Statute of the International Court of Justice: A Commentary (Oxford,
2006) 677, 735, MN 171 et seq. Also Ian Brownlie, Principles of Public International Law, 7th edn
(Oxford/New York, 2008) 5; Malcolm N Shaw, International Law, 6th edn (Cambridge/New
York, 2008) 71.

Promises and the Sources of International Law 177

(a) international conventions, whether general or particular, establishing rules
expressly recognized by the contesting states;
(b) international custom, as evidence of a general practice accepted as law;
(c) the general principles of law recognized by civilized nations;
(d) subject to the provisions of Article 59, judicial decisions and the teachings
of the most highly qualified publicists of the various nations, as subsidiary
means for the determination of rules of law.

Neither unilateral acts in general nor promises in particular are mentioned

by this provision, which is problematic, since subsuming promises under
the sources of law listed is difficult at best. Indeed, it seems, instead, rather
clear that the direct explanation for the binding force of promises cannot
be found within one of the three classical categories of sources, ie treaties,
custom or general principles of law. To begin with, promises are usually
not, and in no way need to be, based on a treaty provision in order to have
their binding effect. Attempting to ground promises on a customary rule is
equally problematic, because, as the depiction of state practice has shown, it
would be something of a far-reaching leap to assume the existence of a general state practice carried by an opinio juris interpreting unilateral assurances
as binding upon the state having made them, at least up and until the PCIJ
and especially the ICJ began attributing binding force to such unilateral acts
as legal undertakings. It is court practice that has had the decisive shaping
legal effect in this area, rather than that of states. Yet, especially after the
ICJs landmark decision in the Nuclear Tests cases, and on the basis of its
continued jurisprudence in this area, the picture might of course change,
and chapter two portrayed some state declarations made or interpreted in
direct reference to the Nuclear Tests cases doctrine. While states therefore
appear to follow the ICJ in its finding that states may assume obligations
by making unilateral promises,2 and promises might eventually come to
be firmly grounded on a general state practice, supported and spawned in
2
Apart from the cases mentioned, the questionnaire circulated by the ILC on the position of states in respect of unilateral declarations provides an additional, yet quite limited,
insight into the position of states in this repect, see ILC, Replies from Governments to the
Questionnaire, Report of the Secretary-General, UN Doc A/CN.4/511 (2000). The questionnaire was drafted very broadly and included relatively abstract questions on the legal framework applicable to all unilateral acts. As such, it is hardly surpising that the response it
triggered was quite meagre. Of the 12 states that responded, 11 provided an answer to the
questions posed (as Luxembourg, while responding, considered that an evaluation could
not be done in the abstract, and not all unilateral acts could be subjected to the same legal
regime). Not one of the states which answered rejected the doctrine of promises in inter
national law, while seven (Germany, Italy, the Netherlands, Sweden, Great Britain, Argentina,
El Salvadore) expressly referred to promises as a unilateral act of state within their answers.
Of the remaining four, three (Austria, Finland and Israel) in answering the questions posed
at some point used the Nuclear Tests cases jurisprudence in explaining the legal effects a unilateral act might have, thereby showing that they were relying on the judgments findings in
this respect. Brownlies observation that States did currently rely on the Nuclear Tests cases
(see the full citation as provided in ch two, text to n 153) therefore appears to be correct.

178 The Law on Promises

part by the Courts jurisprudence, this status has apparently not yet been
fully reached. It is furthermore still the Courts jurisprudence, and not the
practice of states, which has provided and continues to provide answers as
to the precise rules (eg form, revocability, legal bindingness etc) applicable
in this area.3 Just as a customary international law backing therefore cannot
be found or used in delimiting the law in this field, general principles of
law derived from municipal systems are of no avail as support, as a comparable unilaterally obliging mechanism is scarcely present in national legal
systems.4 The ICJ has, however, traced the binding effect of promises back
to good faith and thereby to a general principle of international law, quite
possibly even the most important one.5 As will be recalled, according to
the Court:
One of the basic principles governing the creation and performance of legal
obligations, whatever their source, is the principle of good faith . . . Just as the
very rule of pacta sunt servanda in the law of treaties is based on good faith, so
also is the binding character of an international obligation assumed by unilateral declaration.6

In terms of sources doctrine and in keeping with the position adopted

by the Court itself, good faith, while being a general principle of international law, does not, however, provide us with a satisfying answer. As
the ICJ quotation already indicates, when it says one of the basic principles governing the creation and performance of legal obligations, whatever
their source, is the principle of good faith, the latter is considered to be
more of a background principle7 which does not by itself and directly
give rise to legal obligations proper. In a different ruling the ICJ, explicitly
referring to its Nuclear Tests cases judgments and the above-cited passage,
has made this point very clear:
The principle of good faith is, as the Court has observed, one of the basic principles governing the creation and performance of legal obligations [reference
3
For recognition, protest and (arguably also) waiver the situation is different, as here,
there is a customary international law footing. This fact is overlooked when unilateral acts
are criticised as not being reflected amongst the sources of international law, because those
which are firmly based on custom do not need to be additionally listed as a source of law. See
also Wolf Heintschel von Heinegg, Einseitige Rechtsakte in Knut Ipsen (ed), Vlkerrecht, 5th
edn (Mnchen, 2004) 234, 235.
4
See also ch two, n 5.
5
Shaw, International Law, n 1 above 10304; on good faith and its importance as a general
principle of (international) law, see also M Lachs, Some Thoughts on the Role of Good Faith
in International Law in Declarations on Principles, Liber amicorum diciplorumque (1977) 4755;
Lasa Oppenheim, Robert Jennings and Arthur Watts, Oppenheims International Law, 9th edn
(Harlow, 1992) vol 1, 32; Brownlie, Principles, n 1 above, 19. For more detail see Robert Kolb,
La bonne foi en droit international public. Contribution ltude des principes gnraux de droit
(Paris, 2000) especially 3 et seq and 154 et seq.
6
ICJ, Nuclear Tests (Australia v France), Judgment of 20 December 1974, [1974] ICJ Rep 253,
268, para 46.
7
Shaw, International Law, n 1 above, 104.

Promises and the Sources of International Law 179

to the Nuclear Tests cases omitted]; it is not in itself a source of obligation where
none would otherwise exist.8

If that is so, promises have no legal footing within any of the three principle sources listed in the Courts provision, as neither treaties, custom
nor general principles of (international) law can directly explain their
binding force or the precise legal regime applicable in this area. As has
been seen, the ICJ (and arguably also its predecessor, the PCIJ, for which
a practically identical provision existed in its statute)9 has nevertheless
assumed a legally binding force for promises. In terms of sources doctrine
the Court has, however, offered little explanation, simply considering the
binding force of unilateral declarations to be well recognised.10 The question is thus whether promises are therefore clearly absent from the list
provided in Article 38 and, if so, what the consequences of such a lacuna
might be.
Surprisingly, scholars provide two different responses to the first half of
this question, ie whether promises are missing from the listed sources of
international law. The first is that while states might bind themselves via
promises, promises are nevertheless not a source of international law, they
are instead said to be only a source of international obligations. In following this approach, which can be found in various publications addressing the topic11 and which was also advocated within the ILC (especially,
though not exclusively, by the Special Rapporteur),12 a distinction is
revived which had originally surfaced in the wake of an article published
by Fitzmaurice in 1958. Not referring to unilateral acts, but to treaties, he
deduced from defining law as meaning rules of general validity for and
application to the subjects of the legal system, not arising from particular
obligations or undertakings on their part,13 that:

8
ICJ, Border and Transborder Armed Action (Nicaragua v Honduras), Jurisdiction and
Admissibility, Judgment of 20 December 1988, [1988] ICJ Rep 69, 105, para 94.
9
See Art 38 of the PCIJs Statute, inter alia in Series D, No 1, 20. On that provision and the
only minor touching up of it when Art 38 of the ICJs Statute was drafted, see Pellet, Art 38
ICJ Statute, n 1 above, MN 17 et seq and MN 42 and 47 et seq.
10
ICJ, Nuclear Tests (Australia v France) (1974), n 6 above, 267, para 43. For more on this
judgment see pp 11638.
11
See, eg Maarten Bos, A Methodology of International Law (Amsterdam/New York, 1984)
89; Shaw, International Law, n 1 above, 122; Krzysztof Skubiszewski, Unilateral Acts of
States in M Bedjaoui (ed), International Law: Achievements and Prospects (1991) 22122, para 3.
12
Victor Rodriguez Cedeo, First Report on Unilateral Acts of States, UN Doc A/CN.4/486
(1998) 1214, especially paras 6971 and ILC, Report on its Work of the Fifty-fourth Session,
General Assembly, Official Documents, UN Doc A/57/10 (2002) 215, para 411; see also Simma,
ILC, Summary Record of the 2525th Meeting, UN Doc A/CN.4/SR.2525 (1998) para 10 and
the criticism voiced by Economides, ILC, Summary Record of the 2526th Meeting, UN Doc A/
CN.4/SR.2526 (1998) para 40.
13
Sir Gerald Fitzmaurice, Some Problems regarding the Formal Sources of International
Law in Symbolae Verzijl (The Hague, 1958) 157, note 2.

180 The Law on Promises

treaties are a formal source of international obligation, but (even in the case of
so-called law-making treaties) are not a formal source of international law.14

The driving force behind Fitzmaurices analysis, just as behind those

opinions referring to unilateral acts or promises merely as sources of obligation, is apparently the fact that the obligations created are restricted only
to the (small) number of participating parties (only one state in the case of
a promise). This distinction advocated by Fitzmaurice in reference to treaties has met with some criticism since it was published;15 amongst other
aspects, it has been pointed out that limiting a definition of law to rules
of general validity was in fact begging the question.16 What is, however,
strikingly odd about the debate on promises as sources of obligations and
not law is that, while this distinction does not seem to be very en vogue in
regard to treaties or more generally speaking the sources listed in Article
38, which are commonly referred to simply as sources of law, and not as
sources of law and/or obligations, it can quite frequently be found in relation to unilateral acts. Yet, it should be clear that promises, once accepted
as an existing legal mechanism, must necessarily share the status of treaties
in this respect, at least treaties to which not all states are parties. Surely,
either a mechanism creating only rules of particular applicability is considered not a source of law, but merely a source of (particular) obligations,
and, thereby, promises along with treaties (even most of the ones often
referred to as law-making)17 are discarded from the sources of law, or this
distinction (which would probably also have to exclude regional custom as
law) is rejected, whereby both treaties and promises are sources of law. To
14
Ibid 176 and 15760. For the distinction between formal and material sources, see, eg
Georges Abi-Saab, Les sources du droit international: un essai de dconstruction in Manuel
Rama-Montaldo (ed), El derecho internacional en un mundo en transfomacion, Liber Amicorum en
homenaje al Profesor Eduardo Jimnez de Archaga (Montevideo, 1994) 31; Fitzmaurice, Formal
Sources, n 13 above, 15354.
15
See MH Mendelson, Are Treaties Merely a Source of Obligation? in William Elliott
Butler (ed), Perestroika and International Law (Dordrecht, 1990); Abi-Saab, Sources, n 14 above,
3940; Pellet, Art. 38 ICJ Statute, n 9 above, 7034, MN 8183; HWA Thirlway, International
Customary Law and Codification (Leiden, 1972) 2527; Virally, Sources, n 1 above, 12627 with
further references. It has also found supporters, see, eg Brownlie, Principles, n 1 above, 4; Sir
Robert General Jennings, Course on Principles of International Law (1967) 121(2) Recueil des
Cours de lAcadmie de Droit International de La Haye 325, 331.
16
Speaking of a petitio principii, Pellet, Art. 38 ICJ Statute, n 9 above, 703, MN 82.
17
Some scholars have distinguished amongst treaties and divided them into law-making
treaties and others. Law-making are treaties which create legal obligations the observance of which does not dissolve the treaty obligation, Brownlie, Principles, n 1 above, 13.
Usually, a large number of participants is additionally considered a requirement for a treaty
to become law-making, see Oppenheim, Jennings and Watts, International Law, n 5 above,
32. It has, however, been admitted that there is no clear and dogmatic distinction between
law-making treaties and others, Brownlie, Principles, n 1 above, 14. Also, as even these treaties can only oblige the parties to them and not impose rules of general applicability, they are
not sources of law if Fitzmaurices analysis is followed. See also the criticism by Thirlway,
International Customary Law and Codification, n 15 above, 2627. If they reflect or help to establish customary law, it is, of course, the latter as a separate source which binds non-parties,
not the treaty itself.

Promises and the Sources of International Law 181

treat promises as sources of obligations but all treaties as sources of law is

contradictory. Once this parallelism is accepted, the necessary conclusion
is that promises are indeed missing from the ICJs list, since conventions
(and note the Articles wording, whether general or particular(!)) are on
it. If Fitzmaurices distinction is accepted (even if only in relation to some
treaties), Article 38 already lists not only sources of law but also sources
of obligations; hence, promises, understood as a source of obligation, are
missing. If his distinction is rejected and all conventions (whether general
or particular, whether they impose obligations on both parties or merely on
one of them) remain sources of law, at least for those to whom they apply,
promises should again be on the list, as their function is the same. Whether
they are missing as sources of law or as sources of obligations is, in this
writers opinion, really of secondary importance; their being obliging only
for the declarant is never denied, just as there is consensus that even if
called sources of law, treaties function only inter partes.18
Article 38 of the ICJ Statute has hence been criticised as incomplete,19
with unilateral acts of states being one of the missing sources.20 The
consequences arising from Article 38s lacuna(e)21 have, however, been
described as minimal in practice. As a merely declaratory (treaty) provision of international legal sources, the Article, as such, cannot hinder
the emergence of new sources in international law.22 Its openness is said
18
The question whether this distinction therefore amounts to nothing more than academic
semantics has been raised, with Mendelson, Source of Obligation?, n 15 above, 87, arguing
that Fitzmaurices distinction might lead to the false assumption that treaties were of a lower
rank than custom, as only the latter was a source of law and not merely of obligations.
19
See, eg Abi-Saab, Sources, n 14 above, 33 et seq; Pierre-Marie Dupuy, La pratique de
larticle 38 du Statut de la Cour Internationale de Justice dans le cadre des plaidoiries ecrites
et orales in United Nations Office of Legal Affairs (ed), Collection of Essays by Legal Advisers of
States, Legal Advisers of International Organizations and Practitioners in the Field of International
Law [= Recueil dArticles de conseillers juridiques dEtats, dorganisations internationales et de
praticiens du Droit International] (New York, 1999) 377, 379; Cedeo, First Report, n 12 above,
13, paras 6670; Pellet, Art 38 ICJ Statute, n 9 above, 705, MN 87.
20
See, eg Antonio Cassese, International Law, 2nd edn (Oxford, 2004) 184: This process for
making law, although not provided for in Article 38 of the Statute of the ICJ, is envisaged by
a general rule which has the same rank as those providing for custom and treaty-making;
Heintschel von Heinegg, Einseitige Rechtsakte, n 3 above, 235; Quoc Dinh Nguyen, Alain
Pellet and Patrick Daillier, Droit international public, 7th edn (Paris, 2002) 360; Pellet, Art 38
ICJ Statute, n 9 above, 705, MN 87 and 88 et seq; while slightly sceptical, see also Wolfgang
Graf Vitzthum, Begriff, Geschichte und Rechtsquellen des Vlkerrechts in Wolfgang Graf
Vitzthum (ed), Vlkerrecht, 4th edn (Berlin, 2007) 1, 75, MN 149. For the fact that it is especially the unilateral act of promise which is missing from the sources rather than all unilateral acts in general, see the text to n 3 above. Contra Bos, Methodology, n 11 above 88.
21
Along with unilateral acts of states, acts of international organisations are often considered to be missing from the list.
22
Being a treaty provision and hence included in one of the sources listed, Art 38 cannot, of course, function as the constitutive cornerstone and bedrock of the sources of international law, see also Oppenheim, Jennings and Watts, International Law, n 5 above, 24; also
Fitzmaurice, Formal Sources, n 13 above, 173; Pierre-Marie Dupuy, Droit International Public,
8th edn (Paris, 2006) 268 and Vitzthum, Rechtsquellen, n 20 above, 6, MN 3, who also stresses
that Art 38 does not contain a numerus clausus of international law sources, at 7475, MN 148.

182 The Law on Promises

to ensure the flexibility of the provision and allows the ICJ, whose duty
it remains in the words of Article 38 to decide in accordance with inter
national law, to keep in touch with the evolution of international law and
to stray beyond the margins set by the catalogue included in the provision.23 While this analysis might be correct in principle and will ring true
vis--vis acts of international organisations, it sounds a little awkward
vis--vis promises, in light of the fact that the Court has had its fair share
in elevating unilateral declarations to the level of a legally obliging tool in
the first place. The ICJ has not simply taken account of a development in
international law and gone along with it in its own judgments and despite
its Statutes outdated provision, but has rather strongly been involved in
crafting a new legal source by starting to apply it. It is therefore not surprising that the Court has been criticised as acting ultra vires.24 This harsh
criticism was targeted at the Courts decision in the Nuclear Tests cases and
there is at least a grain of truth to it.
Yet, as the analysis provided in chapter two has shown, it would be
incorrect to assume that the Court had simply out of nowhere declared
assurances to be legally binding. The ICJ and the PCIJ have, instead,
played a decisive part in the gradual shaping of promises as a legal concept, a process which had started long before the Nuclear Tests cases, was
backed by some state practice, and was only spectacularly catapulted into
broad daylight through the Courts decision therein. While in practice the
line between law-making and law applying can become very thin, the process portrayed in chapter two might be described (and criticised) as falling
quite clearly into the category of law-making by the ICJ. But it would be
the wrong conclusion simply to ignore the development of international
law that, even if mainly instigated by the Court, has by now taken place.25
The PCIJs and the ICJs (evolving) rulings, in which they have based
themselves on unilateral assurances of states or analysed their legal bindingness, have not met with the opposition of states. State practice, as portrayed in chapter two, has instead indicated that there are many scenarios
See Pellet, Art 38 ICJ Statute, n 9 above, MN 79 (with further references).
See the evaluation provided by Alfred P Rubin, The International Legal Effects of
Unilateral Declarations (1977) 71 American J International Law 1, 2829.
25
On judicial legislation see, eg Hersch Lauterpacht, The Development of International Law
(Cambridge, 1996 (1st edn 1958)) 153223 and JL Brierly, The Basis of Obligation in International
Law and Other Papers (Oxford, 1958) 98; on the importance of the ICJs (and PCIJs) decisions in forming international (customary) law, see also Paul Guggenheim, Trait de droit
international public, 2nd edn (1967) vol 1, 112; Oppenheim, Jennings and Watts, International
Law, n 5 above, 41; Pellet, Art 38 ICJ Statute, n 9 above, 78890 and Virally, Sources, n 1
above, 15052and La Pense Juridique (Paris, 1960) 16571. The currently adopted narrow
understanding of estoppel (as requiring detrimental reliance) is another example of such a
court-driven development of international law, as it cannot be grounded in the sources listed
in ILC Statute, Art 38. It has thus been referred to as a rule primarily applied and shaped
in international courts and a rule of judge-made international law, Jean Paul Mller and
Thomas Cottier, Estoppel in Rudolf Bernhardt (ed), Encyclopedia of Public International Law,
vol. I, (1992) 118.
23
24

The ILCs Guiding Principles 183

in which states have good reasons and may hence clearly intend to use a
binding, though unilateral, mechanism in adopting a future legal obligation. Similarly, there is an advantage to a recipient in being able to rely on
such undertakings as legally valid. This might explain, in part, why the
rulings depicted, backed by a partial, if not yet general, practice of states,
have left their imprint on international law as practised and perceived by
the international legal community, including those lawyers sitting in the
international legal departments of states. Public international law books
and articles list promises as a possible means for states to oblige themselves26 and discuss the precise legal regime pertaining to them. Finally,
the International Law Commission, in its 2006 Guiding Principles, has
clung to the ICJs dicta in this area and provided additional support to its
doctrine. For the ILC:
Any State possesses capacity to undertake legal obligations through unilateral
declarations.27

This progressive development outside Article 38 can by now therefore

only be denied by closing both eyes to international law as it is currently
practised and understood amongst the legal profession. With the rules on
promises of states being continuously crafted through the rulings of the
World Court and mirrored in numerous publications of some of the most
highly qualified publicists of the various nations, they are now very much
present in what ICJ Statute, Article 38(1)(d) refers to as the subsidiary
means for the determination of rules of law subsidiary means being a
label the practical significance of which, as has been pointed out, should
not be exaggerated.28 Despite Article 38s lacuna in this regard, this study
will therefore take the cornerstones as set by the ICJ and subsequently
adopted by the ILC as the point of departure and move from there to illuminate the current legal framework applicable to unilateral promises of
states on the basis of a systematic and coherent legal analysis.29
II THE ILCS GUIDING PRINCIPLES

The Guiding Principles applicable to Unilateral Declarations of States

Capable of Creating Legal Obligations as adopted by the International
See Introduction, n 49.
ILCs Guiding Principle 2.
28
See Brownlie, Principles, n 1 above, 19. For Fitzmaurice, Formal Sources, n 13 above,
16873, decisions of international tribunals function at least as a quasi-formal source of
international law.
29
On the use of legal reasoning to illuminate underdeveloped areas of law, see also the
short comment by Ian Brownlie, The Responsibility of States for the Acts of International
Organizations in Maurizio Ragazzi (ed), International Responsibility Today, Essays in Memory
of Oscar Schachter (Leiden/Boston, 2005) 357.
26
27

184 The Law on Promises

Law Commission in 200630 have already been addressed briefly in the
introductory remarks to this study. There it was noted that the overall
topic of unilateral acts of states was drastically reduced by the ILC to
unilateral declarations of states capable of creating legal obligations and,
thereby, in effect to those unilateral acts which form the subject of this
book, ie unilaterally binding assurances, termed promises in international
law.31 As such, the ILCs final product is likely to be turned to and understood as reflecting the law which currently governs promises of states,
especially as the ILC based itself nearly exclusively on ICJ dicta. Here,
it will thus be used as a starting point in order to assess the legal framework applicable to promises. As already indicated, however, it can be no
more than a point of departure, since the ILC, while having clearly identified the topics hot spots, left them unresolved. Unable to find a common position, some important aspects fell completely out of the Guiding
Principles, such as, for example, the existence and legal interpretation of
collective acts and the relationship between the Guiding Principles and
unilaterally obliging declarations covered by other legal regimes. Both of
these questions gave rise to numerous debates in the ILC and its Working
Groups, yet neither are even hinted at, let alone actually dealt with, in
the published outcome.32 The Guiding Principles surprisingly also do
not address the consequences which coercion, fraud or error might have
on a unilateral undertakings legal effect, even though taking a parallel
approach to that adopted for manifestations of will in the realm of treaties
by the Vienna Convention on the Law of Treaties would not have seemed
to be very controversial.33 The lack of ICJ dicta in this area might explain
why the ILC nevertheless could not agree to adopt one or more Guiding
Principles covering these aspects. The ILCs final version also does not
address the relationship between promises and the principle of estoppel,
even though questions are bound to arise in this respect and did indeed
surface within the Commission.34 The Preamble does (at least) include references to additional areas of debate. For example, it notes that informal
conduct and silence are capable of binding states, yet it does not say
what silence or informal conduct have to do with the Guiding Principles
subject matter.35 The important question of determining whether it will
be a recipients (legitimate) expectation or the declaring states intention
For the wording of the Guiding Principles see Annex II.
On the question which subject matter is covered by the Guiding Principles, see
Introduction, n 16.
32
In this study both questions have already been addressed when delimiting its subject matter. For collective acts see pp 4155; for the autonomy vs lex specialis question, see
pp 5578.
33
See Vienna Convention, Art 46 et seq. For more see also pp 23739.
34
For more see pp 27794.
35
Silence, where legally relevant, will be so either as tacit acceptance or tacit recognition;
tacit promises are hardly conceivable. See p 78, also ch one, nn 13 and 1517.
30
31

The ILCs Guiding Principles 185

that will determine a declarations legal effect to be bound is similarly left

vague; again, only in the Preamble does the ILC state:
Noting also that in practice, it is often difficult to establish whether the legal
effects stemming from the unilateral behaviour of a State are the consequence
of the intent that it has expressed or depend on the expectations that its conduct
has raised among other subjects of international law.

Even in respect of questions directly addressed by the Guiding Principles,

the adjective guiding has been characterised as a false promise in the
Introduction to this book, since the Principles fail to offer essential guidance in central aspects within their very field of application. The following
example illustrates this point. Used as an introduction to this chapter on
the law of promises, it highlights some of the crucial legal questions which
have been left open and, hence, need to be further addressed and will form
the main subject matter of this chapter. The case to which the Guiding
Principles will be applied is inspired by some of the state practice cited, but
it is fictional, primarily in order to remain as clear-cut as possible if rather
simplistic. After all, without clear answers to the easy cases, it will be even
more difficult to tackle the hard ones.
Imagine a Head of State who, only a couple of months ago, stepped back
from the microphones and cameras of a major international press conference, in which he or she solemnly declared his or her countrys decision to
dismantle all nuclear weapons within the next five years.36 After a restless
week of power politics and with the world suddenly looking rather dark,
our Head of State changes his or her mind and is no longer convinced that
unilateral disarmament can make the world a safer place. But with the
word already out, he or she might wonder, isnt the country now inter
nationally bound to disarm?
Turning to the ILCs Guiding Principles (reproduced in Annex II to the
book), here are the answers the Head of State will find:
Any State possesses capacity to undertake legal obligations through unilateral
declarations

says Guiding Principle 2, which thereby provides an important starting

point, while mirroring Article 6 of the Vienna Convention on the Law
of Treaties.37 According to Guiding Principle 4, a Head of State is one of
36
The declaration might sound like this: In my capacity as Head of State and on behalf
of my Country of X, I am honoured and proud to solemnly declare on this present day, our
countrys decision to dismantle and destroy all nuclear weapons currently in our possession
within the coming five years. By this step we hope to encourage other countries to follow
suit, so that future generations will not have to live with the daily threat of our worlds total
destruction on account of a single mans push of a button. Readers who dislike made up
scenarios might be inspired by other examples cited, for example by the Chinese declaration
announcing a moratorium on nuclear tests as quoted in Introduction, n 42, or some of the
negative security assurances cited at p 161 et seq.
37
Vienna Convention, Art 6 reads: Capacity of States to conclude treaties: Every State
possesses capacity to conclude treaties.

186 The Law on Promises

those vested with the authority to bind a state on the international plane
via unilateral declarations:
A unilateral declaration binds the State internationally only if it is made by an
authority vested with the power to do so. By virtue of their functions, heads of
State, heads of Government and ministers for foreign affairs are competent to
formulate such declarations. Other persons representing the State in specified
areas may be authorized to bind it, through their declarations, in areas falling
within their competence.

Again a certain parallelism to the Vienna Conventions Article 7(2)(a)

can be identified38, but it was in particular the ICJs ruling in the case concerning Armed Activities on the Territory of the Congo, which aided the ILC in
this respect. There, it will be recalled,39 the Court had expressly stated it to
be a well established rule of international law that the above-mentioned
are deemed to represent their state merely by virtue of exercising their
functions, including the performance, on behalf of said State, of unilateral acts having the force of international commitments.40 According to
the Court, furthermore, other persons representing a State in specific
fields may be authorized by that State to bind it by their statements in
respect of matters falling within their purview.41 This finding has now
been reproduced in the Guiding Principle just cited.
The Head of State in our example will notice that the ILC, in conformity
with the ICJs very clear ruling in this regard,42 established that there is
no defence in claiming never to have issued anything in writing. Guiding
Principle 5 states:
Unilateral declarations may be formulated orally or in writing.

Also, the fact that the declaration was not addressed to any specific state
is not an obstacle for its binding nature, as under Guiding Principle 6:
Unilateral declarations may be addressed to the international community as a
whole, to one or several States or to other entities.

While Guiding Principles 2 and 4 to 6 thus reiterate the ICJs findings,

Guiding Principle 8 declares a new (though largely uncontested and
indeed uncontestable) rule,43 when it announces that:
38
Vienna Convention, Art 7 on Full Powers in its para 2(a) reads: In virtue of their functions and without having to produce full powers, the following are considered as representing their State: (a) Heads of State, Heads of Government and Ministers for Foreign Affairs
for the purpose of performing all acts relating to the conclusion of a treaty.
39
See pp 16972.
40
ICJ, Armed Activities on the Territory of the Congo (Democratic Republic of the Congo v
Rwanda), New Application 2002, Jurisdiction and Admissibility, Judgment of 3 February
2006, para 46.
41
Ibid para 47.
42
See ICJ, Nuclear Tests (Australia v France) (1974), n 6 above, 26768, para 45.
43
See also the assessment by Christian Tomuschat, Unilateral Acts under International
Law in Droits et Culture. Mlanges en lhonneur du Doyen Yadh Ben Achour (Tunis, 2008) 1504.

The ILCs Guiding Principles 187

A unilateral declaration which is in conflict with a peremptory norm of general
international law is void.

This poses no obstacle when applied to the declaration in our example.

The same is true for Guiding Principle 9, according to which:
No obligation may result for other States from the unilateral declaration of a
State. However, the other State or States concerned may incur obligations in
relation to such a unilateral declaration to the extent that they clearly accepted
such a declaration.

In this respect, the question should be raised why the ILC considered
it necessary to establish that no obligation may result for an uninvolved
state to which the latter did not consent, instead of merely clarifying that
no obligation can or will arise for an uninvolved third state from a states
unilateral declaration. States simply do not have the power to draw up
obligations for non-consenting others and consequently there is no need
to forbid such acts. However, no harm seems to have been done by the
provision which only forbids the impossible. Unsurprisingly, Guiding
Principle 9 therefore has no impact on the declaration in our example,
to which we return. So far, the rather unproblematic aspects and principles have been addressed, and no obstacle to a legal obligation has arisen.
With the ILCs guidelines, however, we are not much further forward in
assessing whether a unilateral declaration such as the one in our example
will have a legal effect. For this core question, we are left with Guiding
Principles 1, 3 and 7. The last Principle, Guiding Principle 10, addresses
the similarly important issue as to whether a declaration which has been
found to have created legal obligations can later be revoked.
Guiding Principle 1 reads as follows:
Declarations publicly made and manifesting the will to be bound may have the
effect of creating legal obligations. When the conditions for this are met, the
binding character of such declarations is based on good faith; States concerned
may then take them into consideration and rely on them; such States are entitled to require that such obligations be respected.

Although a questionable requirement,44 the declaration in our example was publicly made and it arguably manifested a will to be bound,
although here we lack essential guidance. How do we determine whether
a will to be bound was manifested? From which circumstances might it be
inferred? Is it the reasonable addressees view that is decisive in assessing whether the declarant manifested a will to be bound (do we analyse
the declared will which might, however, not be what a state intended
to declare), or must we actually inquire into the (difficult to assess) real
will the declarant had at the time, and might be able to prove through
44
A questionable requirement if not interpreted to include any statement that is intentionally communicated to an addressee, be it behind closed doors or not. For more see p 239.

188 The Law on Promises

internal documents. Similar questions might, of course, arise in relation
to bi- or multilateral commitments and have been debated in that regard.
However, with unilateral declarations, the will of the declarant seems to
be of special importance for many commentators, both within and outside
the ILC.45 The Commission, in its Commentary, cites the Frontier Dispute
case in this context, in order to stress that it all depends on the intention
of the State in question.46 Guidance in this respect is all the more necessary, as for unilateral declarations a formalised legal procedure at the
international level is missing ratification of unilateral declarations is the
exception rather than the rule.47 And even at the national level there is
usually no formal act or parliamentary involvement, which could somehow be indicative of a states intention to become legally bound by a subsequently made international declaration.
But even if we assume that we can surmount this hurdle and have
found a public manifestation of a states will to be bound, under the ILCs
Principles we will not have got very far. Guiding Principle 1 stipulates
that, in this case, the declaration may have the effect of creating legal obligations. But it may means that it very well may not even more so if
said by a lawyer or incorporated in a legal text. It has to be noted that the
wording chosen by the Commission on this fundamental matter is even
more surprising when compared with the words chosen by the ICJ, to
which the Guiding Principles continually claim to be paying the highest
respect. To recall, the ICJ in the Nuclear Tests cases stated that:
An undertaking of this kind, if given publicly, and with an intent to be
bound, even though not made within the context of international negotiations, is
binding.48

The Principles do not tell us which additional factors, besides a public

statement made with a will to be bound(!), will decide between the may and
the may not, or will tilt the scale towards or away from legal bindingness. The second sentence simply skips this issue when it continues with:
When the conditions for this are met . . ., and apparently assumes that
we will know how to identify the essential (additional) conditions when
we see them.
Those who sympathise with the answer commonly provided by lawyers asked for an opinion, ie well, it depends, might point to Guiding
Principle 3, which reiterates the importance of circumstances when it
states:
See p 207 et seq.
ILC, Guiding Principles applicable to Unilateral Declarations of States Capable of
Creating Legal Obligations, with commentaries thereto, ILC Report, A/61/10 (2006) ch IX,
370, Commentary to Guiding Principle 1, para 1.
47
For more see p 246.
48
ICJ Nuclear Tests (Australia v France) (1974), n 6 above, 267, para 43 (emphasis added).
For the Nuclear Tests cases see p 11638.
45
46

The ILCs Guiding Principles 189

To determine the legal effects of such declarations, it is necessary to take account
of their content, of all the factual circumstances in which they were made, and
of the reactions to which they give rise.

There is, however, little comfort in being directed to take different

aspects into account without being told in which way. The criticism
voiced here, in other words, is not directed at the well, it depends answer
itself, but at the silence in response to the follow-up question on what the
answer might depend: What aspects are we looking for in declarations,
what in the circumstances, what in the reactions to which they gave rise?
It is furthermore noteworthy that the ILC begins its list of what to take
into account by pointing to a declarations content in order to assess a
declarations legal effects, and not its text. But what is the content of the
declaration made by the Head of State in the example provided? It can
itself only to be ascertained by an act of interpretation and is consequently
of little avail as an interpretative tool. Tomuschat, for this reason, critically
remarks that the ILC is thereby attaching increased importance to the overall circumstances in which a declaration is made; Guiding Principle 3 is
therefore found to be based on a doctrine of interpretation in fundamental contrast to the rules of the [Vienna Convention].49 As the Commentary
shows, the ILC has again used the ICJs jurisprudence as a foundation
for its Guiding Principle and displayed no intention to deviate from the
Courts findings.50 Yet, while the ICJ has referred to a declarations substance or content in its judgments, it has, as will be shown in detail,
actually determined the latter by first turning to a declarations text.51 The
ILCs bare reference to a statements content in Guiding Principle 3 hence
in effect falls short of the guidance provided by the Court in this respect.
By shifting the focus primarily to the (in any event important) circumstances and the reactions to which a declaration gave rise, the question
becomes all the more pressing, how they must be taken into account in
assessing a declarations legal effect. While the Guiding Principles remain
silent in this respect, the Commentary points to some examples of state
practice, yet without explaining what legal effect they are supposed to
illustrate. A closer look at the importance of the reactions [presumably
of the addressee(s)] to which [the declarations] gave rise may illustrate
this point. The short and quite meagre commentary offered by the ILC
cites a couple of declarations by states, which are said to have triggered
different responses from their addressees; some took cognisance of the
Tomuschat, Unilateral Acts, n 43 above, 1502.
In explaining the adopted wording the ICJs Commentary claims that the wording of
Guiding Principle 3 is also inspired by a passage in the ICJ Judgments in the Nuclear Tests
cases, to which allusion had been made in the Frontier Dispute case and the Armed Activities
on the Territory of the Congo case. The Military and Paramilitary Activities in and against
Nicaragua case is also used as support, see ILC, Guiding Principles with Commentaries, n 46
above, 371, Commentary to Guiding Principle 3, para 1.
51
For more see p 211 et seq.
49
50

190 The Law on Promises

commitments undertaken,52 others objected53 or challenged them.54 But
what is the legal consequence of these reactions once they are identified?
None of the cases cited by the ILC in this context has ever been brought
before a tribunal which could have elaborated on the impact of the
requisite addressees reaction. Does the latters objection extinguish the
obligation the declarant wanted to assume? Is an acknowledgement by
the addressee necessary for the obligation to arise or is it simply irrelevant? If it were irrelevant, why is it then necessary to take the reaction into
account? The ILC does not tell us, the Commentary to Guiding Principle 3
really only points to the problem without solving it:
Several of these examples show the importance of the reactions of other States
concerned in evaluating the legal scope of the unilateral acts in question,
whether those States take cognizance of commitments undertaken (or, in some
cases, rights asserted), or, on the contrary, object to or challenge the binding
nature of the commitments at issue.55
52
The examples mentioned in the Commentary to Guiding Principle 3 of states which
have taken cognisance of the commitments as opposed to having objected to or challenged
them, are problematic. The Commentary in its note 937 (the footnotes are consecutively
numbered in the ILC Yearbook with the Commentarys footnotes starting at 921) points to
the Egyptian declaration on the Suez Canal as well as to Jordans statement about the West
Bank, and directs the reader to Victor Rodriguez Cedeo, Eighth Report on Unilateral Acts
of States, UN Doc A/CN.4/557 (2005) paras 6364 for Egypts declaration, and paras 48,
5051 for Jordans declaration. While in the Egyptian example some states indeed took cognisance of the obligation undertaken, others objected to or challenged the declaration, see
p 108 et seq. The Special Rapporteurs Eighth Report only cites Frances rejection along with
a Security Council and a General Assembly Resolution for the reaction of third states; both
Resolutions are, however, completely devoid of any reactions to the Egyptian declaration,
which is not surprising, as the latter was made around six months after the Resolutions had
been adopted. The statement by Jordan on the West Bank, which is classified as a waiver
by the Special Rapporteur and therefore falls into a special category of unilateral acts (see
pp 3436) is equally problematic in this respect. According to the Rapporteurs Report, the
waiver merely led to surprise on the side of the PLO and was not mentioned in the declaration
of the Independent State of Palestine, see Eighth Report, para 48. The United States appears to
have rejected the declaration as it declared that the status of the West Bank and the Gaza Strip
could not be settled or established by unilateral acts, ibid para 50. The reactions by other states,
as cited in ibid para 51, are in part reactions to the declarations of independence by the PLO and
not Jordans waiver, or recognitions of the West Bank being administered by the PLO.
53
Here, note 939 of the Commentary to Guiding Principle 3 presents Uruguays refusal of
a donation of vaccines from Cuba as the first example, but it is not a good one considering
the details of the case: Cuba had offered what it called a donation of vaccines to Uruguay,
but nevertheless asked for the reduction of its debt owed to Uruguay in the amount of the
vaccines cost. Uruguay accepted the deal while merely protesting against Cubas fairly
inadequate semantics, and called the exchange by its real name: a commercial transaction.
The other example provided does not concern a declaration that was intended to assume an
obligation, at least not primarily: Russia is cited as having protested against Turkmenistan
in 1993 when the latter wanted to establish its territorial waters in the Caspian Sea according
to the law of the sea, whereas Russia held the law of the sea to be inapplicable as it considers the Caspian Sea to be really a lake, as it lacks any connection to a global ocean. See ibid
1011, paras 3643 and 1819, paras 84105.
54
Here reference is made to the reactions of the non-nuclear-weapon states. For more see
pp 16166.
55
Commentary to Guiding Principle 3, para 3, references omitted.

The ILCs Guiding Principles191

In our fictitious example we had not mentioned any reactions whatsoever, but even if we assumed that some states acknowledged the commitment, others renounced it and a majority remained silent, we would only
know that these reactions are somehow important.
Leaving aside for the moment the ILCs Guiding Principles 1 and 3 and
the indeterminacy included therein, the remaining principle on a unilateral declarations interpretation is Guiding Principle 7, and its first sentence might shed some additional light on whether a declaration (such as
in our example) has created legal obligations or not. In keeping with the
ICJs jurisprudence, it does refer to a declarations text, when it reads:
A unilateral declaration entails obligations for the formulating State only if it is
stated in clear and specific terms.

The declaration of our Head of State has set out the decision to pursue a
precise aim (nuclear disarmament within five years) and therefore might
well be considered clear and specific. Of course, a little more guidance
might again have been helpful in determining what the standard might
be that is to be applied under the clear and specific formula. However,
Guiding Principle 7, second sentence, apparently gives expression to the
ICJs restrictive interpretation when it says:
In the case of doubt as to the scope of the obligations resulting from such a declaration, such obligations must be interpreted in a restrictive manner. In interpreting the content of such obligations, weight shall be given first and foremost
to the text of the declaration, together with the context and the circumstances in
which it was formulated.

On closer look, however, the ILC in this principle opts to apply a restrictive interpretation (only) where there is doubt as to the scope(!) of the obligations resulting from such a declaration and not when assessing whether
or not a state has actually intended to bind itself. Not only the Guiding
Principles structure, in which the determination of a declarations legal
effects is already covered by Guiding Principle 3, but also the wording of
Guiding Principle 7, sentence 2, thus point towards Tomuschats conclusion that what Guiding Principle 7 addresses is not whether a declaration
has been made with an intention to be legally bound; it instead deals solely
with the substantive details of an obligation found to exist.56 As he says,
the if does not come within the scope of Principle 7 but that of Principle
3. And indeed, the wording of sentence 2 of Principle 7 is clearly restricted
to this aspect as it speaks of the scope of legal obligations, thereby presupposing a legal obligation to be already existing. Sentence 1 of Principle 7,
on the other hand, does not contain the same restricted wording, when it
simply says: A unilateral declaration entails obligations for the formulating State only if it is stated in clear and specific terms. If so applied, it
56

Tomuschat, Unilateral Acts, n 43 above, 1503, emphasis added.

192 The Law on Promises

can even lead to a finding of no existing legal obligation, ie where no clear
and specific wording is used. And as seen in chapter two, this is precisely
what happened in the case concerning Armed Activities on the Territory of
Congo,57 which the ILCs brief commentary to Guiding Principle 7 cites
as support. In doing so, Guiding Principle 7 therefore in its first sentence
seems to apply to the if question of whether a state expressed an intention to be legally bound, while its second sentence establishes a restrictive
interpretation (only) for when an obligation has already been found to
exist. The result is not only odd in respect to the structure the Principles
thus appear to have (Principles 3 and again 7, yet only sentence 1, address
the finding of an intention to be bound). The Principles also contradict the
ICJ jurisprudence to which they refer,58 while leaving open what kind of
interpretation should in fact be applied to declarations of states in assessing their intention to be legally bound.
Some members of the ILC, just as some states in their comments,
emphasised that everything depends on context and might applaud the
indeterminacy found to exist, especially in the main Guiding Principles 1
and 3.59 The result, however, is a considerable lack of legal security. The
Principles themselves put it quite nicely in their preambular part, according to which states may find themselves bound by their unilateral behaviour on the international plane. As with every bad compromise, neither
side within the ILC can really be happy with this outcome. The members
of the ILC who were sceptical of drafting rules for unilateral declarations
or at least promises as a legal institution had to concede that states might
very well be bound by unilateral declarations publicly made and manifesting a will to be bound and have, by adopting the Guiding Principles,
underlined the relevance of the ICJs dicta in this area. The proponents
of such rules, on the other hand, have been presented with Guiding
Principles which tell us only very little about when this bindingness is
brought about. This last criticism should not be misunderstood as a call
for a closely determined system, where every act A triggers a precisely
determined legal consequence B; a certain range of indeterminacy is, of
course, vital for most law to be adequately applicable to complex realities.
Yet, the Guiding Principles expressly stipulate that a state may be bound
by its actions, without providing the state with any clear indication as
See pp 16972.
See p 212 et seq.
See, eg the answers given by Finland to the questionnaire which the ILC had sent to
governments, ILC, Government Replies to the Questionnaire, n 2 above, 2. The answers to
the questions as to what formalities unilateral acts are subjected, which types of unilateral
acts existed and what their content might be, and as totheir importance, usefulness and
value or possible duration, were all answered as depending on context. While Finland in
its answer rightly observed that legal doctrines aim is to protect legitimate expectations, it
overlooked the fact that what can be considered as a legitimate expectation will essentially
depend on the normative legal framework applicable to the act in question. For more see
p 207 et seq.
57
58
59

The ILCs Guiding Principles193

to when this might happen. The reason for the Principles not doing so
appears to relate back to the opposite positions within the ILC which led
the Commission first to narrow the subject matter and then focus nearly
exclusively on ICJ dicta. The time pressure at the end of the quinquennium was apparently so considerable that marked ambiguities within the
Principles could not be ironed out, whereby some question marks have in
fact even been added to those which, especially in respect of legal aspects
not yet covered by the Courts jurisprudence, were already present in the
area.
It is thus hard to imagine the Head of State in our example not sitting
back rather puzzled after reading the ILCs Guiding Principles. If he or
she decides to play it safe, he or she might assume they have created a
legal obligation through the statement and look into possible grounds for
revocation, which are to be found in Guiding Principle 10. Yet, here again,
the reader will be confronted with a couple of open questions. Just as the
ICJ in the Nuclear Tests cases found that the French declarations cannot
be interpreted as having been made in implicit reliance on an arbitrary
power of reconsideration, the Guiding Principles prohibit arbitrary
revocation:
A unilateral declaration that has created legal obligations for the State making
the declaration cannot be revoked arbitrarily. In assessing whether a revocation
would be arbitrary, consideration should be given to:
(a) any specific terms of the declaration relating to revocation;
(b) the extent to which those to whom the obligations are owed have relied on
such obligations;
(c) the extent to which there has been a fundamental change in the circumstances.

Unlike the ICJ, the ILC therefore went a small step further, by adding
a couple of aspects that should be taken into consideration in assessing whether a revocation has been arbitrary or not, such as the extent to
which those to whom the obligation owed have relied on it, whether there
were specific terms of revocation included in the declaration, or a fundamental change of circumstances had occurred. As the declaration in our
example included no reference to revocation and a change of mind can
hardly be seen as a fundamental change of circumstances,60 its revocation
will hinge on the question whether and in which way another states reliance might influence the declarants freedom of action. Again, however,
here the questions will only begin: How will we know if other states have
60
The Guiding Principles refer to a fundamental change in the circumstances and not
of circumstances, found in the Vienna Convention on the Law of Treaties. This change in
wording, however, does not seem to imply any difference in application, as ILC, Guiding
Principles with Commentaries, n 46 above, Commentary to Principle 10, para 3, explicitly refers to the strict limits of the customary rule enshrined in Article 62 of the [Vienna
Convention], see ILC, Guiding Principles with Commentaries, 381.

194 The Law on Promises

relied on the statement made? Do they need to express their reliance or
demonstrate it by any sort of action? What if there is mere silence on the
other side? And why should the addressees reliance decide whether a
declarant acts arbitrarily in revoking in the first place? Because even if
reliance is considered to be important, and it is indeed essential in this
context,61 it is hard to see how good reasons leading a declarant to revoke,
turn into arbitrary ones in the face of reliance.62 Also, and more import
antly, the ILC is silent on how reliance will play out in the equation: Will
the addressees reliance completely hinder the revocation of a unilateral
declaration or does it merely impose certain (and if so, what kind of) limits and restrictions on the declarant?
The questions raised in dealing with this short example are those which
are not solved (and are in part produced) by the ILCs Guiding Principles.
We are going to leave the fictitious statement here. While the ILCs Guiding
Principles have, as seen, the merit of pointing to the relevant dicta in the
area of state promises and have incorporated some of them into a few
rather clear-cut guidelines in the more unproblematic areas, they regrettably reveal major inaccuracies and are of little help when trying to interpret
a concrete declaration as legally binding or not. On account of the very
different views held by its members, the ILC was, in addition, not able
to move forward to illuminate how to fill some of the existing gaps and
answer open questions that the Court has left for potential promises of
states. The questions resulting from the Guiding Principles lacunae and
encountered by the Head of State in our example are now to be addressed
in more detail on the following pages, in order to provide some clearer
answers.
III LEGAL BASIS FOR THE BINDINGNESS OF
UNILATERAL PROMISES

In doing so, it is essential to begin by defining why a unilateral assurance

is considered as becoming binding upon a declarant. What legal norm
or principle, in the end, is it based on and why is that principle held to
compel a state to adhere to its pledge? Far from being a merely academic
exercise, it is one of the keys to finding a solution to several of the open
questions identified above.63
See p 194 et seq.
Reliance should therefore rather stand beside arbitrariness and on its own independent
footing as a ground influencing the possibility of revocation. For more see p 251 et seq.
63
Within the ILC, Simma, ILC, Summary Record of the 2593rd Meeting, UN Doc A/
CN.4/SR.2593 (1999) para 73, rightly stressed that determining the basis of a unilateral acts
binding force, ie the philosophical foundation of the problem, could have an impact on the
solution of very practical topical issues. In the field of promises, Koskenniemis finding that
the unresolved dispute about the basis of obligation emerges each time a State denies that
61
62

Legal Basis for the Bindingness of Unilateral Promises 195

Following the debates within the ILC and in view of the opinions
expressed by legal commentators in publications on the matter, three
answers are nowadays provided with some regularity to explain why a
state will be bound by its promise made. In addition, two more are mentioned, if only rarely. Amongst the more common, the first line of argument
takes recourse to state sovereignty and notes that sovereign states are free to
restrict their freedom as they like, and that there is no apparent reason why
the legal framework should impose an impediment in this respect for them
to do so unilaterally. Closely connected and indeed intertwined with this
understanding, a second section of legal doctrine emphasises the pivotal
importance of a states intent: if a state wants to bind itself unilaterally, then
its will to do so can and must give rise to a legal obligation once it has been
expressed. Finally, and probably amongst the majority of authors, the principle of good faith is put forward as the cornerstone on which the bindingness of a unilateral assurance can be based. It is, however, not always clear
what good faith is actually meant to protect.
While sovereignty, a states intention and good faith are dominant in
the arguments on the bindingness of international promises and will be
addressed shortly and in detail, two other legal concepts have also been
advanced. Especially in slightly older publications, the principle of estoppel in international law has been used to explain a promises legal effect.
As the estoppel principle, along with its relation to the doctrine of promises, is indeed important when picturing the legal framework that might
be triggered by a states unilateral pledge, it is addressed in more detail
below.64 In the present context, it may suffice to emphasise that the bindingness of a promise for the following reasons cannot be based on estoppel
as it is presently conceived to function in international law. The principle of estoppel has a distinct legal effect apart from that which an assurance, as such, might trigger when interpreted as an international promise.
Estoppel acts as a (complete) shield to protect the addressee against the
alteration of a declarants behaviour which is judged to be inconsistent.
Being a shielding legal defence, unlike promises, it hence does not serve
as the basis of an executable and claimable obligation (ie it is no sword
that can be used against the promisor); there is no estoppel obligation
that, if breached, will trigger a states responsibility. To be invoked, estoppel in addition requires detrimental reliance on the part of the addressee in
order to come into play. Only on account of this reactive behaviour on the
part of the addressee, and in order to protect the latter from potential detriment, will estoppel hinder a state from altering its behaviour. A promise,
however, has been held and is generally understood to be binding once
a rule can be applied to it, that is, in any international dispute, is particularly to the point,
Martti Koskenniemi, From Apology to Utopia: the Structure of International Legal Argument
(Reissue with a new Epilogue, Cambridge, 2005) 332.
64
See p 277 et seq.

196 The Law on Promises

it is made and without having to trigger any reaction from the state(s) to
which it was directed. Neither on the grounds of what is necessary for
promises to become binding, nor of their actual legal effect, can a promise
therefore be explained by the estoppel principle, at least if the latter is not
rewritten and stripped of those features which currently define it.65
Beyond estoppel and the trio of sovereignty, intent and good faith, to
which we will turn in a moment, a presumption of consent has also been
mentioned as explaining an assurances binding force. It will, therefore, if
only very briefly, be scrutinised first.
A Presumed Consent
Basing the binding nature of promises, or even that of all unilateral acts,
on the presumption of the addressees consent, is not very common in publications on the topic. However, it is sometimes mentioned in reference to
a side note by James Brierly, written while he was Special Rapporteur for
the law of treaties within the ILC, and as such will not be left unaddressed
here. For Brierly :
A possible explanation of the binding force of so-called unilateral declarations
creative of legal rights against the declarant is to be found in the theory of presumed consent of the beneficiary.66

The problem with Brierlys approach, which if still referred to is usually criticised and discarded,67 is quite obviously that it does not explain
a unilateral acts legal effect: if the unilateral act became binding only on
the basis of the addressees consent, even if presumed, there is no longer
a unilaterally binding act. This is presumably why James Brierly referred
to so-called unilateral declarations in the above passage, as according
to this solution, it is not the unilateral declaration as a single unilateral
manifestation of will which is binding but the consensual bond established
on the basis of a (presumed) meeting of minds, ie two manifested wills.
Rubin, in discussing this theory as applied to the Nuclear Tests cases scenario, rightly stressed that:
It . . . assumes an approach by which a unilateral declaration delivered publicly
and with no particular addressee creates powers in all states as implied offerees to accept by silence the offer contained in the declaration.68
For details and references see p 277 et seq.
James Brierly, Report by JL Brierly, Special Rapporteur for the Law of Treaties, UN Doc A/
CN.4/23 in (1950) II YB International Law Commission 227, para 20.
67
See A Gigante, The Effect of Unilateral State Acts in International Law (1969) NYU
School of Law J International Law and Politics 333, 342; Rubin, Unilateral Declarations,
n 24 above, 11; and Camille Goodman, Acta Sunt Servanda? A Regime for Regulating the
Unilateral Acts of States at International Law (2006) 25 Australian YB International Law 43,
6061.
68
Rubin, Unilateral Declarations, n 24 above, 11.
65
66

Legal Basis for the Bindingness of Unilateral Promises 197

We would thereby simply be back in known treaty waters, as Brierly

acknowledged when he continued by saying: If this theory be correct
the categorization of such declarations as treaties, whether for the purposes of Article 102 of the Charter of the United Nations or otherwise, is
permissible.69
And indeed, as seen in chapter two, in many cases legal commentators have argued that the legal commitment in question had been brought
about, not through the unilateral assurance as such, but instead through
an oral treaty thanks to the addressees tacit assent. Yet, as some of the
cases have illustrated and numerous commentators have never tired of
stressing,70 this strictly consensualist approach has to close both eyes
to social realities, in cases where none of those involved intended to or
actually acted under the impression of concluding an informal treaty:
the declarant, now christened offeror, did not ask for an acceptance of
his declaration, nor did the addressees want to provide one. As seen, the
presumed-consent approach is also not the approach taken by the World
Court, with the Nuclear Tests cases decisions leaving no more room for
doubt. Deciding long after Brierlys comment, as portrayed above, the
Court found that [i]n these circumstances, nothing in the nature of a quid
pro quo nor any subsequent acceptance of the declaration, nor even any
reply or reaction from other States, is required for the declaration to take
effect, since such a requirement would be inconsistent with the strictly
unilateral nature of the juridical act by which the pronouncement by the
State was made. To presume such acceptance would, of course, mean that
a second manifestation of will was necessary in order for the declaration
to become binding, hence the presumption; with the latter, the bindingness would consequently not be an international obligation assumed by
unilateral declaration based on good faith, but an international obligation
assumed by consensus and binding since pacta sunt servanda. Presumed
consent has therefore been rejected by the ICJ and it is at least implicitly
rejected by every author or tribunal that has attributed legal force to a
unilateral declaration as such.

Brierly, Report, n 66 above, 227, para 20.

Vladimir-Djuro Degan, Unilateral Act as a Source of Particular International Law
(1994) 5 Finnish YB International Law 149, 17071; Paul de Visscher, Remarques sur
lvolution de la jurisprudence de la Cour Internationale de Justice relative au fondement
obligatoire des certains actes unilatraux in Essays in Honour of Manfred Lachs (The Hague/
Boston, 1984) 459, 464; Jean-Didier Sicault, Du caractre obligatoire des engagements unilatraux en droit international public (1979) 83 Revue Gnrale de Droit International Public 633,
641; Gian Carlo Venturini, Attitudes et actes unilatraux des tats (1964) (II) 112 Recueil des
Cours de lAcadmie de Droit International de La Haye 363, 400, especially 40405.
69
70

198 The Law on Promises

B State Sovereignty and Intent
But why then should a unilateral assurance, in which a state promises
to do or refrain from doing something in the future, be binding upon
that state? Because it is its sovereign decision to become bound, seems
to be at least part of the answer provided within the legal community. As
Herdocia Sacasa, for example, stressed in the ILC: if a universal norm had
to be found as the basis for the binding legal effect of all unilateral acts,
then it must be sought in the area of state sovereignty.71
The sovereignty argument in addition seems to be at the heart of the
second, somewhat related finding, that it was the intention of a state, and
therefore its will, to become bound which conferred a binding nature
upon an assurance made. The latter position surfaces with some regularity within the debate. The Special Rapporteur in his First Report, for
example, stated the following:
The State which formulates the declaration is bound to fulfil the obligation
which it assumes, not because of the potential juridical interest of the addressee
but because of the intention of the State making the declaration.72

Jean Charpentier, in his article on unilateral engagements, in a similar

vein concludes that good faith was only of subsequent importance in dealing with the execution of an obligation already undertaken and not with
the obligations birth. The answer to the question as to the actual foundation of a unilateral engagements binding force was instead to be found
in the autonomy of the [declarants] will.73 For Krzystof Skubiszewski
also: What is decisive is the intention of the State: the State is bound by its
unilateral act because such has been its intent.74
The problem with this understanding is that despite the assertions
made, it is incapable of explaining why a state should have created an
obligation and become bound by its assurance. While state sovereignty can
serve as the theoretical starting point for explaining a states ability to
undertake legal obligations in the words of the PCIJ in 1923: the right
[and it might be added capability75] of entering into international engage71
Herdocia Sacasa, ILC, Summary Record of the 2527th Meeting, UN Doc A/CN.4/
SR.2527 (1998) para 14.
72
Cedeo, First Report, n 12 above, 29, para 160. Ibid para 161, Cedeo mentions the
need to create greater confidence in international relations as another justification for the
binding nature of unilateral declarations.
73
Jean Charpentier, Engagements unilatraux et engagements conventionnels: diffrences et convergences in Jerzy Makarczyk (ed), Theory of International Law at the Threshold of
the 21st Century (The Hague/London/Boston, 1996) 367, 374.
74
Skubiszewski, Unilateral Acts, n 11 above, 231, para 51. Although the principle of good
faith and the reliance on an act influenced a declarations application, they did not confer the
binding force on the act, ibid 232, para 52.
75
For sovereignty as a description of legal competence see Brownlie, Principles, n 1 above,
291 and 105 et seq.

Legal Basis for the Bindingness of Unilateral Promises199

ments is an attribute of State sovereignty76 it alone cannot explain why

the sovereign should be bound to adhere to its own previous decisions
even against its present will. Because only where a state has changed its
mind will the nature of an obligation and the bindingness of a declaration be
of actual relevance and prove its existence. An account of the legal basis
of a promises obligation therefore has to explain why a past statement
should restrain a states current actions, in other words: why a state is
obliged to do what is has assured, irrespective of whether it presently
wants to do so or not. The explanation for this phenomenon termed bindingness is not provided by the concept of sovereignty alone quite on
the contrary, really. If it is a states sovereignty that allows it to confer a
binding nature upon an assurance at will, then why should this very same
power not allow a state to later act as now intended, and thereby (even
implicitly) to revoke its old decision? Free that is, unrestrained revocability which is the logical result of a (solely) sovereignty-based argument,
and a concept of bindingness, however, do not go together.77
What is true for an argument based solely on state sovereignty is, of
course, also true for the position according to which a states will is the
reason compelling it henceforth to the line of action declared. Yet, as seen,
for the Special Rapporteur: The State which formulates the declaration is
bound to fulfil the obligation which it assumes, not because of the potential juridical interest of the addressee but because of the intention of the
State making the declaration. But again, the question must be raised why
an old will of a state should prevail over its new will? It is this scenario of
a change of mind which puts the theories to the test; but by pitting a states
will at time 1 against the very same states will at time 2, there is really no
reason to see why arguing on the basis of a states intentions should lead
to a preference of past decisions over present ones. To do so would instead
be entirely arbitrary and compel a state to maintain a decision against its
present will merely because it once said so, and even where the obligation
undertaken is of no interest to anyone and possibly even to the detriment
of the declarant or that of other states. Unless an external position is introduced which warrants protection, everything instead militates in favour of
allowing a state to decide what is presently best for it, and not force a longgone decision upon it simply because it was adopted in the past.
Arguments directed against founding legal obligations solely on the will
of a (single) state are of course anything but new and have been exchanged
by lawyers in various contexts. The general theory of auto-limitation as
the foundation of international law, which had been advanced especially by
Georg Jellinek78 at the end of the nineteenth and beginning of the twentieth
PCIJ, SS Wimbledon, Judgment of 17 August 1923, Series A 1, 15, 25.
See also the references provided in n 86 below.
78
To be found in various of his publications, especially Georg Jellinek, Die Rechtliche Natur
der Staatenvertrge. Ein Beitrag zur Juristischen Construction des Vlkerrechts (Wien, 1880).
76
77

200 The Law on Promises

centuries, has been criticised with similar arguments from very early on.
Laband, for example, in 1906 wrote:
A self restraint is without legal force; because it can be resolved at will; if it is
followed, it is not because one may not, but because one does for reason of
prudence, decency, fear etc not want to resolve it. This is true for the State just
as it is for the individual.79

Brierly, amongst others,80 has also criticised the idea of states auto-
limiting themselves based solely on their own will, by stating that:
However we may choose to define law, an essential part of the function of law
must be to limit the wills of those to whom its precepts are addressed, and its
binding force cannot possibly be derived from the will that it limits.81

Here, our business is not that of establishing the foundation of international law but only finding an answer to the question why a unilateral
assurance as such is held to become binding upon the state making it.
Yet, the answer has to take into account what has been emphasised in the
above quotations: Especially in the context of unilateral acts, where it is
only the will of a single state which is said to create a legal obligation, the
words must ring true: the obligation would indeed be non-existent, were
it to be grounded on and explained in reference to the will of the declaring
state, because then it could be changed accordingly, ie at will.82
As already hinted above, some authors, such as Jean Charpentier, have
contemplated setting the question of the initial obligation apart from that
of a declarations revocability, and grounding the initial obligation on a
states will, whereas only the subsequent obligation not to revoke this
declaration and the regime governing its further existence is grounded
on good faith.83 For Skubiszewski, similarly, reliance may decide the acts
revocability in doubtful cases but it does not confer the binding force on
the act.84 Yet, and apart from the fact that the ICJ (as will be recalled and
79
Paul Laband, Besprechung zu: Mrignhac, Trait de Droit Public International (1906)
20 Archiv fr ffentliches Recht 302, 304, and my translation of the following original: Eine
Selbstbeschrnkung ist ohne rechtliche Kraft; denn man kann sich nach Belieben frei machen;
tut man es nicht, so geschieht dies nicht weil man nicht darf, sondern weil man aus Klugheit,
Sittlichkeit, Anstand, Furcht u.s.w. nicht will. Dies gilt vom Staat wie vom Einzelnen. For a
thorough and recent discussion of Jellineks theory and criticism of it, see, eg Jens Kersten,
Georg Jellinek und die klassische Staatslehre (Tbingen, 2000) 409.
80
See, eg, also Alfred Verdross, Le fondement du droit international (1927) 16 Recueil des
Cours de lAcadmie de Droit International de La Haye 24, 26667, whose position (formulated in
French) can loosely be translated as follows: If the legal provision is nothing but the product
of a free will, it, at its basis, is not binding; it is at the discretion of the States which have created it and which will no longer have to take it into account, once they decide it not to be
binding any longer. What has been created by will, can be undone by a will to the contrary.
81
Brierly, Basis of Obligation, n 25 above, 14.
82
See also Koskenniemi, From Apology to Utopia, n 63 above, 310: Limits on State freedom
which are merely willed and capable of being altered at any change of will are not normative
limits at all.
83
Charpentier, Engagements unilatraux, n 73 above.
84
See n 74 above.

Legal Basis for the Bindingness of Unilateral Promises 201

seen in more detail below)85 has emphasised that the creation of the obligation itself is governed by good faith, the alternative concept proposed
would lead to the very contradiction that was just depicted. While it is, of
course, possible to address a declarations bindingness separately from
its revocability, both subjects remain intimately related.86 With free revocability as the result of referring solely to a states will, there is nothing
which can adequately be referred to as an initial obligation much less
a declaration having binding force. Because, what is the promising state
obliged to do, to what is it bound to adhere if it is its will that reigns? The
resulting free revocability instead means that a state does not really have
to perform what it has promised in any way, it is not obliged to do so.
To base the initial obligation or even a declarations bindingness on a
states will and only that of its execution on good faith, hence either hollows out the words obligation and bindingness and leaves them devoid
of any meaning, or is contradictory. Obligation, just as binding declaration, express nothing else than that a states freedom of action is restricted
and limited in a certain way by the declaration made.
With only the two ingredients of sovereignty and a states intent in
the mix, the result will therefore never be that of a binding obligation.
In order to explain a declarations binding force, it is hence necessary to
move further and towards good faith, as many authors and the ICJ have
done. As will be seen, the intention of a state is thereby not entirely taken
out of the picture, however, the bindingness of an assurance cannot be
grounded on it.
C Good Faith and Presumed Reliance
While the ICJs Nuclear Tests cases judgment stressed the importance of a
states intent in assuming a legal obligation, a fact to which we will return,
See p 202 and also n 87.
The connection between the two concepts has been stressed by numerous authors, see,
eg Wilfried Fiedler, Zur Verbindlichkeit einseitiger Versprechen im Vlkerrecht (1976) 19
German YB International Law 35, 58: Nicht zufllig muss sich die Frage einstellen, welche
rechtliche Bedeutung jener Verbindlichkeit noch beizumessen sein soll, wenn sie ohne weiteres durch einen Widerruf wieder aufgehoben werden kann. Schon hieraus wird deutlich,
dass die Formulierung einzelner Verbindlichkeitskriterien stets auch die Mglichkeit des
Widerrufs im Blick behalten muss, wenn nicht gleich zu Begin ein unzutreffender Eindruck
geschaffen werden soll; Sergio M Carbone, Promise in International Law: A Confirmation
of its Binding Force (1975) 1 Italian J International Law 166, 171, who in addressing the ICJs
arbitrary revocation prohibition considers it to undermine once and for all those theories
that, while expressly aimed at determining when a promise may be revoked, eventually
lead to denying its binding force. See also the comment made by Goco during an early ILC
meeting on the Rapporteurs First Report: The binding character of a unilateral act would
be illusory if the legal relationship the act created were to be terminable unilaterally and at
the will of the author State, ILC, Summary Record of the 2524th Meeting, UN Doc A/CN.4/
SR.2524 (1998) para 42.
85
86

202 The Law on Promises

it also clarified that it is not the intention which serves as the basis of a
unilateral declarations bindingness. Instead, for the Court, the principle
governing not only the execution but also the creation of an obligation is
that of good faith:
One of the basic principles governing the creation and performance of legal
obligations, whatever their source, is the principle of good faith . . . Just as the
very rule of pacta sunt servanda in the law of treaties is based on good faith, so
also is the binding character of an international obligation assumed by unilateral declaration.87

The ILC has thus followed the ICJs line of reasoning, when in Guiding
Principle 1 it also assumed the binding character of such declarations is
based on good faith. Yet, the term good faith, as such, is elusive enough
still to camouflage what is actually protected by it. Sicault has highlighted
the various meanings attributable to the legal88 expression of good faith,
ranging from being faithful to the law and ones undertakings, especially
when executing the latter, to a states erroneous state of mind to be acting
in conformity with the law. Neither of these interpretations are, of course,
meant when referring to good faith in the present context. Instead, to hold
a state to its promise on the basis of good faith is widely perceived to
mean to protect the addressees legitimate reliance (= the confidence and
trust placed) on the unilateral assurance given.89 This is what the ICJ was
driving at when it emphasised:
87
How Rubin could interpret the Court thereby to presume that an international obligation may be assumed by way of a unilateral declaration independently of the principle of
good faith and apparently to have taken the view that good faith merely prevents unilateral revocation of the international obligation, created by unilateral declaration is difficult
to understand in light of the Courts own wording according to which the creation of the
legal obligation as such is governed by good faith. See also the similar critique by Sicault,
Engagements unilatraux, n 70 above, 67879.
88
That it is anything more than a moral principle in this area has been doubted especially
by Elisabeth Zoller, La bonne foi en droit international public (Paris, 1977) ss 34750, for whom
at xxvi it cannot found juridically the obligation to respect the legitimate beliefs of others.
Yet, and as will be seen, in the realm of promises it does quite clearly operate as a legal principle; see also the criticism by Sicault, Engagements unilatraux, n 70 above, 681 et seq. See
also Kolb, La bonne foi, n 5 above, 379, note 171 with further references.
89
See Sicault, Engagements unilatraux, n 70 above, 683: il ne sagit plus du devoir
de loyaut de lauteur de lengagement unilatral, mais de la confiance lgitime des destinataires dudit engagement, qui ne doit pas tre due.Formulations vary but reflect the
same understanding, see Rodolfo De Nova, Die Neutralisation sterreichs (1958) 54 Die
Friedenswarte 298, 304: Auf diese Weise legt [die Bekanntgabe des Versprechens] den Grund
zu Erwartungen, die an die Erklrung geknpft werden, und schafft das in sie gesetzte
Vertrauen; und in diesem Vertrauen kann man den Verpflichtungsgrund des Versprechens
erblicken; Wilfried Fiedler, Unilateral Acts in International Law in Rudolf Bernhardt,
Peter Macalister-Smith and Max-Planck-Institut fr Auslndisches ffentliches Recht und
Vlkerrecht (eds), Encyclopedia of Public International Law (Amsterdam/New York, 2000) vol
4, 1021: the acceptance of binding force is intended for the protection of those States which
have had cause to rely upon the declaration; Dinesh Khosla, Nuclear Test Cases: Judical
Valour v. Judicial Discretion (1978) Indian YB International Law 343: This decision, it may be
submitted, clarifies the real sociological basis of obligation in international law. By asserting

Legal Basis for the Bindingness of Unilateral Promises 203

One of the basic principles governing the creation and performance of legal
obligations, whatever their source, is the principle of good faith. Trust and confidence are inherent in international co-operation, in particular in an age when
this co-operation in many fields is becoming increasingly essential.90

To hold a state bound to its words and consider it obliged to follow

through with what it has proclaimed, irrespective of its change of heart, is
therefore considered necessary not in order to enforce a mere because you
said so, or you once wanted so against the declarant state, which would
rather senselessly limit its present freedom of action, but in order to protect
a legitimate interest which is weighed against the declarants freedom to act
as presently intended; this interest is the trust and confidence legitimately
placed on an assurance by the addressee. Only with this external point of
reference can the bindingness of the declaration be explained. But if this is
true, and we believe it is, a number of questions come to mind.
The first consideration is, how the conclusion that only qua reliance will
a binding commitment come into being, can be compatible with the ICJs
finding that a promise required not only no quid pro quo or acceptance
but not even any reply or reaction from other states . . . to take effect.
That the addressees reaction is not necessary for the promise to become
binding has not only been unequivocally asserted by the Court, but is also
generally accepted within legal doctrine. As Fiedler writes in his article
for the Encyclopaedia of Public International Law:
It is not necessary, for the recipient State to have already been induced into
taking action, from which adverse consequences would ensue, should the
State having made the declaration later take the opposite view . . . Rather, it is
sufficient if the declaring State has created a legally relevant situation of trust
which in the specific circumstances of the case entails legally binding force.91
that it is the resultant expectation of an act or behaviour of a nation-state upon which reliance is placed by other nation-states, it helps us discard a not very satisfactory basis of
obligation in international law, namely, the doctrine of consent; Paul Reuter, Droit international public, 6th edn (Paris, 1983) 164: Cette affirmation est base sur le principe de la
bonne foi et plus spcialement sur lobligation de respecter les convictions que lon fait natre par son comportement; Eric Suy, Les Actes Juridiques Unilatraux en Droit International
Public (Paris, 1962) 151: Alors que loffre doit tre accepte pour tre obligatoire, la promesse
lest dun moment ou elle est la base de confiance du sujet favoris. Cest dans cette confiance dans la parole donne que trouve le fondement de validit de la promesse; similarly Venturini, Actes unilatraux, n 70 above, 40203; Alfred Verdross and Bruno Simma,
Universelles Vlkerrecht: Theorie und Praxis, 3rd edn (Berlin, 1984) 430, para 670. Despite classifying the effect as a notion destoppel(in our eyes this is incorrect, see pp 19596), see also
B Bollecker-Stern, Laffaire des essais nuclaires francais devant la Cour Internationale de
Justice (1974) Annuaire Franais de Droit International 299, 330, for whom the protection of
la confiance sucite chez les autres Etats par la declaration unilatrale is the reason qui
empeche son auteur dune part en contester la valeur juridique, dautre part de la retirer
arbitrairement; similarly Jrg P Mller, Vertrauensschutz im Vlkerrecht (Cologne/Berlin,
1971) 112 and Jrg Leutert, Einseitige Erklrungen im Vlkerrecht Ein Beitrag zur Lehre vom
Vertrauensschutz (Diessenhofen, 1979).
90
ICJ, Nuclear Tests (Australia v France) (1974), n 6 above, 268, para 46, emphasis added.
91
Fiedler, Unilateral Acts, n 89 above, 1021.

204 The Law on Promises

Further down in his article, Fiedler even goes a step further when he
says that:
it is sufficient that the recipient State has the possibility of taking notice. This
condition follows from the good faith principle, as only on this basis can a situation of trust be created. It is not in fact necessary for the third State to develop this
trust.92

So although we have concluded that it is the protection of trust and

confidence placed on a declaration by the addressee which is the reason
to hold a state to its word and, hence, for the assurance to be binding, a
states promise will reach this status without the addressee having reacted
and, at least according to the above statement, even without requiring
him to have actually developed any trust. Sicault also positions himself
against the necessity of actual reliance, when he explains his understanding to be as follows:
La confiance dont il sagit ici nest pas, en effet, la confiance effective des destinataires de la promesse (reliance) mais la confiance quils doivent pouvoir avoir
dans le caractre obligatoire de la promesse.93

It is not effective, or actual, reliance which is thus said to be protected,

but according to Sicault, it is the confidence that the addressee(s) must
be able to have in a promise as being binding. Robert Kolb has been cautiously supportive in respect of this approach, which he characterised to
mean that:
il suffit que lacte soit susceptible dune confiance abstraite, cest--dire que
compte tenu des circonstances, tout autre sujet de droit international ou tout
destinataire plus particulier ait pu lgitimement, de bonne foi, tabler sur la
dclaration telle quelle a t faite.94

It is hence only the abstract confidence which, in this opinion, forms the
basis of an international legal obligation. In other words, the faith which an
addressee must be able to place, and not the faith that it is in fact placing,
on a declaration is supposed to make it binding, the overall goal being to
stabilise international relations. This solution, however, is, in this writers
opinion, not convincing. Despite rightly identifying trust and confidence
placed on a declaration as the basis behind good faith, which allows a state
Ibid 1022, emphasis added.
Sicault, Engagements unilatraux, n 70 above, 684. (It may be translated asThis confidence is, in effect, not the addressees actual confidence (reliance) but the confidence which
they have to be able to have in a promises binding character).
94
Kolb, La bonne foi, n 5 above, 335, emphasis added, and continuing: Cest peut-tre avec
ce critre quun quilibre aussi idal que possible entre libert et stabilit est atteint. (My
translation of the quotation: it is sufficient that the act is open to abstract confidence, that
is to say that, with regard to the circumstances, every other subject of international law or
every specific addressee could reasonably, in good faith, have relied on the declaration as it
has been made).
92
93

Legal Basis for the Bindingness of Unilateral Promises 205

to be held to its word, by discarding effective reliance as being of import

ance, its logic assumes a declaration to be binding even where no effective
reliance exists, simply because somebody could have relied on the statement
in question. But while it is one thing not to require actual and positive
proof of the addressees reliance on a unilateral assurance received, it is
something else to consider a declaration to be binding, even where reliance
is not assumed to exist. The point is that reliance which is not there, even
if labelled abstract (confidence) as it could have existed, does not require
protection in order to stabilise international relations; good faith cannot
and need not protect faith which does not exist.95 Were the opposite held to
be true, we could again end up with a situation that was criticised above
and was rightly dreaded by Rubin,96 a situation in which a declaration is
construed as restricting a states freedom of action against its present will,
even though the states concerned warrant no protection as they are not
relying and do not count on the declaration made. That the addressees
actual non-reliance is of considerable impact is in line with the finding of
numerous authors that where an addressee either expressly or by conduct
rejects a states promise, the declarant will no longer be bound vis--vis
that state.97 For Suy this is the result of clausula rebus sic stantibus, but it is
simply the logical consequence of basing a declarations binding character
on the principle of good faith: with no faith being placed on a declaration by the state(s) addressed and benefiting from it, the foundation for a
95
Venturini, Actes unilatraux, n 70 above, 40203 rightly remarked that le caractre
obligatoire de la promesse doit prsupposer que ces destinataires ont confiance en cette
promesse, en un mot, quils se fient elle.
96
Rubin, Unilateral Declarations, n 24 above, 11.
97
See Jean-Paul Jacqu, A propos de la promesse unilatrale in Melanges offerts a Paul
Reuter (1981) 327, 340: Dans le cadre de lanalyse de la Cour, si la promesse trouve son fondement dans la confiance lgitime que les Etats doivent saccorder, le bnficiaire peut fort
bien renoncer tout moment la protection que lui accordait le droit international.Mais
tant que cette renonciation nest pas intervenue, la promesse doit tre maintenue; despite
arguing for abstract reliance, see also Kolb, La bonne foi, n 5 above, 335: Ou bien il pourra
se librer par voie de rvocation unilatrale parce que personne ne se sera fond en confiance sur son acte; see also Suy, Actes unilatraux, n 89 above, 151; Venturini, Actes unilatraux, n 70 above, 40203, as the passage already cited in n 95 above reads in full: Lattitude
de ces mmes Etats [the addressees] prend une importance certaine, tant donn que le
caractre obligatoire de la promesse doit prsupposer que ses destinataires ont confiance
en cette promesse, en un mot, quils se fient elle. Within the ILC, see, eg Simma, ILC,
2593rd Meeting, n 63 above, para 68, according to whom other States could refuse such a
gift if they did not want it. See also the citation of a state representative within the Sixth
Committee as provided by Victor Rodriguez Cedeo, Third Report on Unilateral Acts of
States, UN Doc A/CN.4/505 (2000) para 65: It was unimaginable that a unilateral act would
have legal effects in the relations between its author and another subject of international law
if the latter had raised objections. Contra Rosenstock, ILC, 2593rd Meeting, n 63 above, para
72, for whom the other party was irrelevant as to whether there was a commitment. For
the scenario in the Nuclear Tests cases and the interpretation of the Courts ruling, see p 116
et seq. If the Courts precedent is followed, the misinterpretation of the declaration as not
being specific enough to be binding might form an exception, as constructive reliance will
be enough. For the problem this approach causes for the declarant, see p 128.

206 The Law on Promises

promises bindingness necessarily collapses.98 Where a state displays its
non-reliance upon an assurance, there is no reason to hold the declarant
state bound to its words should it change its mind. This effect, however,
is incompatible with an explanation which founds a declarations binding
force on abstract confidence. Because, as the latter opinion holds actual
reliance to be unnecessary for a declaration to be binding, then why should
its absence suddenly be of any effect on the statement made?
The missing link which explains a unilateral declarations bindingness
once declared and in the absence of the addressees reaction is therefore to
be sought elsewhere. To recall, without reliance there is no reason why a
state should be compelled to adhere to its wording on a good faith basis.
Yet, the ICJ (and commentators agree) stated that an addressee does not
have to react and display its reliance in any way in order for the assurance to have its compelling effect. However, if a promise is rejected, it
will no longer be binding upon the promisor vis--vis the promisee. What
this means is that the legal order in fact presumes that the addressee will
place and is placing reliance upon a promise the very minute it has been
received. Not only is presumed reliance the key to explaining a promises
legal feature, but such a presumption is in addition adequate in light
of the actual communication, which typically takes place between the
promisor and the promisee(s). Through a promise, the addressee receives
a declaration which, legally speaking, is always strictly beneficial for it,
as it is granted a right with only the declarant state undertaking a new
obligation. The latter, furthermore, unlike when making an offer, does
not ask for an acceptance (which, in gratuitous offers where no quid pro
quo is requested, is in practice reduced to nothing but an open display
of reliance), but instead declares its intention to establish an undertaking as unconditional and effective at the moment it is communicated; the
addressee is therefore not even asked to react. Where state A, for example, declares that it will henceforth refrain from drift-net fishing or will
open a canal to all foreign vessels, it makes sense to assume that the states
informed of this declaration will rely on it, even though they remain
silent. Based on this assumption of trust and confidence being placed on
the declaration received, the statement becomes binding upon the declarant the minute it is made. Yet, where the addressee expressly or by conduct rejects the assurance received, the presumption is rebutted and the
declarant state is no longer faced with an interest in need of being taken
into account, protected and weighed against free revocation, should the
promisor change its mind. Hence the above-declared effect according to
which an assurance that has been rejected is no longer binding, ie can be
taken back at will. What this means is not that effective reliance, as such, is
discarded as important, but only the display of reliance in the form of any
See also Jacqu, Promesse unilatrale, n 97 above.

98

On the Intention to be Legally Bound 207

positive reaction, as the addressees actual reliance on a statement which

is beneficial and not rejected is presumed to exist. Only this interpretation
allows the basing of a promises bindingness on good faith while at the
same time requiring no reaction from the addressee.
It is worth pointing out that by focusing on reliance as protected by good
faith in order to explain a promises binding force, legal doctrine does not
sacrifice a promises unilateral character.99 While it is true that via reliance
an aspect of bilaterality is introduced,100 it in no way diminishes the qualification of the binding act as being unilateral, since the legal relationship
is, and remains to be, brought about only through a single manifestation of
will. This, and nothing else, is what the term unilateral act means.101 That
there is at least one more state within the picture is clear from the outset, as
any legal relationship presupposes at a minimum two actors, and a promise
or assurance, as every act of communication, an addressee. As stressed in
defining a unilateral act, it is not labelled unilateral because it is muttered
by a state to itself. The question merely is whether a single manifestation of
will triggers an immediate legal effect, or whether it calls for and needs to
encounter a second manifestation of will, ie an acceptance, in order to do so.
Since promises are binding as such and as a single act of communication102
vis--vis the addressee, they are part of the overall category of unilateral
acts of states. They nevertheless need to be adhered to on account of the
trust and expectations which they have created, as the latter are legally protected under the doctrine of promises in international law.
IV ON THE INTENTION TO BE LEGALLY BOUND

The basis of a promises bindingness has been identified to reside not in

the declarants intent to undertake an obligation but in the addressees
99
As Sacasa, ILC, Summary Record of the 2695th Meeting, UN Doc A/CN.4/SR.2695
(2001) para 32 emphasises: At all events it seemed correct to say that the bilateral nature of
the relationship did not affect the unilateral character of the act. See also Venturini, Actes
unilatraux, n 70 above, 403: Une condition de ce genre [ie reliance or trust], videment
diffrente dune acceptation, puisque la dclaration unilatrale suffit crer des obligations
juridiques dans le cas concret.
100
See Fiedler, Unilateral Acts, n 89 above, 1021: The establishment of trust presupposes
at least a second subject of international law. To this extent it is justified to speak of a specific
reciprocal relationship.
101
See pp 4041.
102
Indeed, even from a completely trust-centred approach which rejects the focus on
manifestations of will and criticises dealing with promises under the heading of unilateral acts, such as Mller, Vertrauensschutz, n 89 above, 112, a promise can be referred to as
unilateral and understood as a different form of inter-state relations than those established
through a treaty undertaking, as it remains possible to differentiate between the acts of communication which are necessary in order to establish trust as a basis for the legally binding
obligation. Whereas a treaty relationship requires back-and-forth communication, a promise
is valid as a single communicative act, announced to the addressee in order to create legally
protected, ie legitimate, reliance.

208 The Law on Promises

legitimate reliance upon a promise made reliance that is presumed to
exist unless the promise is rejected. To find that the law (presumes and)
protects the addressees reliance on a promise as legitimate thereby moves
the spotlight towards the question inquiring which state declarations will
actually be considered as constituting a promise under international law.
Because only then will an addressees actual reliance on a pledge made
also be legally protected as legitimate; and it is here where a states intention to be bound is of (some) importance.
A The States Manifestation of Will
The question when a declaration may be considered and relied upon as a
legally binding undertaking is answered by the ICJ and most scholars by
referring to a states will to commit itself legally. For the ICJ it was clear
that, [o]f course, not all unilateral acts imply obligation; but a State may
choose to take up a certain position in relation to a particular matter with
the intention of being bound.103
And (only) where the state chooses to do so, is the declaration to be
considered a legally binding undertaking:
When it is the intention of the State making the declaration that it should
become bound according to its terms, that intention confers on the declaration
the character of a legal undertaking, the State being thenceforth legally required
to follow a course of conduct consistent with the declaration.104

A promise has accordingly and in conformity with legal doctrine been

referred to as a unilateral manifestation of a states will to bind itself
legally to follow a certain course of future action throughout this study.105
By focusing on a states manifested intention to undertake a legal obligation in defining statements which are legally binding, ie promises, a line is
therefore drawn between mere political statements, on the one hand, and
legal undertakings, on the other. Where a statement is held to be one of
mere intent, a political expression of a states future plan, that of its representatives political dream or aspiration, any reliance on it and the expectations of its fulfilment will not be legally protected. In other words, and
irrespective of whether an addressee has factually relied on such pledges,
in the latter cases any request for the declaration to be fulfilled will remain
unenforceable before an international court, and its breach will not be
sanctioned as a breach of a legal commitment (ie not lead to a states legal
ICJ, Nuclear Tests (Australia v France) (1974), n 6 above, 267, para 44.
Ibid para 43.
105
For definitions see ch one, n 42. By including promises within the category of unilateral acts of states, the understanding is expressed that they are manifestations of a states
will which will have the legal effect their declarant apparently intended them to have, see
pp 3840.
103
104

On the Intention to be Legally Bound 209

responsibility and reparations or lawful countermeasures). The distinction

which is known from the law of treaties, according to which gentlemens
agreements have long been set apart from legally binding commitments,
just as political statements and declarations of intent nowadays are,106
therefore also exists in the realm of unilateral assurances.
To concentrate on whether a state intended to commit itself legally in
defining a statement as a binding promise is, of course, meant to protect
the promising states sovereignty and allow it to decide for itself when its
future freedom of action is going to be limited through a new legal commitment. While an understandable aim, there should be no doubt that it
cannot always be fully achieved, because how is the declarants intention
going to be detected? According to the ICJ, the intention is to be ascertained by interpretation of the act and in recalling the ICJ cases depicted
in chapter two,107 it is important first to note what the ICJ has not done
when interpreting the act: it has not simply asked the declarant what its
intentions were when it made its declaration, nor has it used other means,
such as the disclosure of internal documents, in order to ascertain what
the state might have actually intended yet not openly expressed, at the time
it made its statement. Instead, the Court has emphasised that it had to
form its own opinion and, in interpreting the declaration, has relied upon
its substance and thereby particularly its text,108 and the information that
was openly available to it just as to the addressee at the time, as circumstantial evidence: It is from the actual substance of these statements, and
from the circumstances attending their making, that the legal implications
of the unilateral act must be deduced.109
Dogmatically speaking, the Court has therefore taken what many
commentators have stressed is the only sensible approach in assessing a
statements value, it has looked at the declared intent or better manifestation of a states intent (as circumstances in which a statement is made
will also play a role), rather than searched for the possibly hidden actual
intention the declarant had at the time.110 Not only will the latter be
106
Providing an overview and addressing the parallel questions for the law of treaties,
Jan Klabbers, The Concept of Treaty in International Law (The Hague/Boston/London, 1996)
especially 15 et seq, and on intent 65 et seq.
107
Especially its Nuclear Tests cases, as discussed at p 116 et seq, the Frontier Dispute case,
discussed at p 156 et seq and the Armed Activities on the Territory of the Congo case, discussed
at p 169 et seq.
108
For more see p 211 et seq.
109
ICJ, Nuclear Tests (Australia v France) (1974), n 6 above, 269-70, para 51; see also ICJ,
Frontier Dispute (Burkina Faso v Mali), Merits, Judgment of 22 December 1986, [1986] ICJ Rep
554, 574, para 40: In order to assess the intentions of the author of a unilateral act, account must
be taken of all the factual circumstances in which the act occured and ICJ, Armed Activities on
the Territory of the Congo, Jurisdiction and Admissibility (2006), n 40 above, para 49.
110
See especially Sicault, Engagements unilatraux, n 70 above, 648 et seq; also Thomas M
Franck, Word Made Law: the Decision of the International Court of Justice in the Nuclear Test
Cases (1975) 69 American J International Law 612, 61617; Kolb, La bonne foi, n 5 above, 33536;
Suy, Actes unilatraux, n 89 above, 149; Goodman, Acta Sunt Servanda?, n 67 above, 5657.

210 The Law on Promises

difficult to ascertain (nobody knows what a human being, much less a
state as an abstract and composite legal entity, really and actually intended
when making a statement),111 but inner motives and motivations must be
largely112 irrelevant, where a legal framework attempts to stabilise trust
in statements made and needs to adjudicate upon conflicting interests
resulting from an act of communication issued by one state and relied
upon by another. Otherwise, even the clearest statement expressly voicing
a states decision to enter into a legal obligation could not be trusted, as
upon closer analysis, it could be unmasked by the declarant as not having
been made with an actual intention to undertake a legal commitment at
the time but for other reasons, be it just to appease or fool the addressee.
For trust and security to be placed in international declarations, such an
apologist view113 would be unacceptable; statements made will therefore
have to be judged according to their appearance and at face value. The
latter, of course, is not limited to the declarations wording but includes
the context and the surrounding circumstances in which a statement has
been made.
While necessary, this shift in perspective to the outside world and the
eyes of the reasonable addressee as assumed by the Court, however, entails
the (for the declarant rather unwelcome though unavoidable) consequence
that a declaration may come to be held to express an intent the declarant
never actually had. Where a court (or the addressee) determines to see
for itself what a state intended when making its declaration, the outcome
of its assessment might not be in line with what a state actually intended,
especially where a states intentions are inferred from its actions. The fact
that in these scenarios a declaration made is interpreted as a manifestation
of a will to be bound, although and irrespective of the fact that it did not
actually exist, means that in all these case the manifested will is reduced
to nothing but a mere legal fiction;114 a fiction that is, however, upheld in
order to protect the trust and confidence placed on a declaration by its
addressee as legitimate. With this in mind, it is misleading to label a states
intention to be an important condition of validity of a unilateral promise,

111
For this question in respect of treaties see again Klabbers, Concept of Treaty, n 106 above,
65 et seq, particularly at 69 et seq. Klabbers stresses that the English maxim according to which
not even the devil knows what is inside a mans head was a useful point of departure and
that we are hence not looking for subjective intentions, but for manifest intentions, see
ibid 6970. Since treaties require recourse to the intentions of not only one but at a minimum
that of two states, there the picture is even more complicated.
112
The declarants provable and fundamental error as specified in Vienna Convention, Art
48 for treaties and within the limits prescribed by that Article, is, however, likely also to be
accepted as an exception for promises, see pp 23739.
113
To use the terminology of Koskenniemi, From Apology to Utopia, n 63 above.
114
See the analysis of the ICJs judgment rendered in the Nuclear Tests cases at p 123 et seq
and particularly the text to ch two, n 172.

On the Intention to be Legally Bound 211

as is sometimes done.115 Strictly speaking, this is not the case, as the interpretation of a statement by a court might not reflect the declarant states actual
state of mind when making the statement. While the law, by focusing on a
states manifested intention to be legally bound, thus tries to ensure that the
expectations raised by a declaration in the addressee are legally protected
only where a state also intended to undertake a legal obligation, this aim cannot always be fulfilled. Such a discrepancy between outside interpretation
and actual intention is likely to appear in scenarios in which the promising state has acted malevolently, recklessly or negligently by making statements which, in light of their circumstances and wording, may (reasonably)
be understood as expressing a legal commitment. It is, however, not limited
thereto if the rules according to which an assurance is judged are unclear
and only elucidated after the declaration has already been made (France
arguably underwent this experience in the Nuclear Tests cases). Determining
the legal framework and the exact standard of interpretation applied to
state pledges is hence of pivotal importance in order to protect states from
being taken by surprise by misreading their announcements. The interpretative rules will thus inform not only the addressee of when it may rely
on a declaration made, but also the declarant as to when its actions will be
understood as legally compelling. The clearer the legal canvas is on which
state declarations are judged, the less likely will it be for a state to see trust
that was placed on its declaration also be protected as reasonable in a case
where it actually had no intention of becoming legally bound (malevolent,
reckless and negligent cases aside). After all, a state will usually have
intended what (under a clear interpretative standard) are simply the natural
consequences of its actions when they are interpreted as a manifestation of
a will to be legally bound. Elucidating the standard of interpretation applied
to unilateral declarations of states in order to ascertain whether or not an
intention to be bound has been manifested is therefore of central importance.
B Ascertaining a Legal Intent to be Bound
What then are the rules of interpretation applied to declarations of states
by the ICJ and legal doctrine? While it is clear that only statements made
by a competent state representative will qualify as a legally binding
115
See, eg Kolb, La bonne foi, n 5 above, 333, for whom lintention nest quune condition
dexistence de lacte juridique unilatral, non le fondement de sa validit. Suy, Actes unilatraux, n 89 above, also speaks of la volont de promettre as being a condition. Both authors,
however, subsequently make it very clear that they are not actually considering a states
intention to be the condition of a promises bindingness but only the finding of a states manifestation of will. The latter may, however, be found to exist even where the declarant did not
actually intend to commit itself but acted in a manner which led the reasonable addressee
to assume otherwise, see Kolb, La bonne foi, n 5 above, 33536; for Suy, Actes unilatraux, n 89
above, 149.

212 The Law on Promises

promise,116 the question remains which ones will be taken to constitute a
manifestation of will to be bound in the sense just elaborated. In drawing on the analysis provided by the Court and discussed in chapter two,
the territory is not fully mapped, but neither is it uncharted. The following can be said.
(i) Restrictive Interpretation
To begin with, the question needs to be addressed whether the ILC was
right to assume that the ICJ has opted for a restrictive standard of interpretation to be applied when the scope of a legal obligation resulting from
a declaration is assessed, and not already during (or in addition to) the
preceding assessment of whether a declaration has triggered a legal obligation in the first place. As will be recalled,117 the Guiding Principles are
in fact silent in respect of the interpretative standard to be applied during
the first initial stage, asking if a will to be bound has been manifested.
Yet, Guiding Principle 7, second sentence, says that where the scope of
the obligation resulting from such a declaration was in doubt, such obligations must be interpreted in a restrictive manner. In explaining its finding, the ILC has referred directly to the ICJ whose dicta in the Nuclear Tests
cases are used as support. But the Court actually said the following:
Of course, not all unilateral acts imply obligation; but a State may choose to
take up a certain position in relation to a particular matter with the intention
of being bound the intention is to be ascertained by interpretation of the act.
When States make statements by which their freedom of action is to be limited,
a restrictive interpretation is called for.118

Although slightly ambiguous, the quotation, in the eyes of this writer,

argues for a restrictive interpretation to be applied to phase one, ie when
assessing whether or not a legal obligation exists. Despite the fact that
the ICJ in the Nuclear Tests cases has been criticised for not having in fact
applied a very restrictive approach in this matter,119 in the above quotation
it nevertheless announced that the intention [of being bound(!)] is to be
ascertained by interpretation of the act, a finding that is immediately followed by a call for a restrictive interpretation. As already seen in chapter
two, the ICJ has in addition in its later rulings (ie Military and Paramilitary
Activities in and against Nicaragua,120 and especially the Frontier Dispute121
116
Here the ICJ and the ILC have been very clear and have left little room for doubts as to
which rules will apply, for more see p 234.
117
On the Guiding Principles see pp 18394.
118
ICJ, Nuclear Tests (Australia v France) (1974), n 6 above, 267, para 44.
119
See especially the dissenting opinion of Judges De Castro and Barwick, as presented
amongst the separate and dissenting opinions of the Nuclear Tests cases, discussed at p 132
et seq.
120
See p 151 et seq.
121
See p 156 et seq.

On the Intention to be Legally Bound 213

and Armed Activities on the Territory of the Congo122 cases) in practice interpreted state declarations restrictively. In all these cases, however, it has
done so when conducting the initial determination as to whether or not
a state has manifested a legal intention. The ILC in its Commentary nevertheless cites not only the Nuclear Tests cases in support but also points
to the ICJs finding in the Frontier Dispute case, according to which great
caution has to be used by the interpreter where a declaration has no particular addressee.123 In fact, however, this case is the clearest in respect of
both the standard of interpretation announced and that actually applied
by the Court: caution was used by the ICJ not in dealing with a declaration
which had already been found to include a legally binding commitment,
the scope of which had to be ascertained, but instead when evaluating
whether the declaration of Malis Head of State was, as such, a legally
binding statement. As will be recalled, the Court established its cautious
approach after having referred to its Military and Paramilitary Activities
judgment, where it:
was unable to find anything in that communication from which it can be
inferred that any[!] legal undertaking was intended to exist [reference to the
Military and Paramilitary Activities case omitted]. The Chamber considers that
it has a duty to show even greater caution when it is a question of a unilateral
declaration not directed to any particular recipient.124

If the ILC is to be taken seriously in its references to the Courts jurisprudence, it is therefore more likely than not (possibly on account of the
rather rushed attempt to present some conclusion on the topic at the end of
its quinquennium) that it drafted its second sentence too narrowly.125 The
resulting (rather odd and in part redundant) structure of the Principles
(which then address how to interpret a declaration and evaluate its legal
bindingness in Guiding Principles 1, 3 and 7) will nevertheless remain as
a problem for future interpretation. For the latter, the emphasis should
remain rather on the ICJs jurisprudence than on the ILCs wording, as
the Commission has made it very clear that it intended to be following
the Court.
Accordingly, a restrictive interpretation will have to be applied when
ascertaining whether a state intended to bind itself legally, which in
practice means that in cases of doubt, triggered especially by ambiguous and unclear wording, no legal intention can be assumed to have been
expressed.
See p 169 et seq.
ILC, Guiding Principles with Commentaries, n 46 above, Commentary to Principle 7,
para 2.
124
ICJ, Frontier Dispute, Merits (1986), n 109 above, 574, para 39, emphasis added.
125
See also Tomuschat, Unilateral Acts, n 43 above, 1505, emphasising that [u]nder such
circumstances, full technical accuracy cannot be expected.
122
123

214 The Law on Promises

The ICJ in the Frontier Dispute case, as just cited, has in addition emphasised that it was necessary to show even greater caution when it is a question of a unilateral declaration not directed to any particular recipient. The
underlying reason for the argument of the need to be particularly careful
in these scenarios has apparently been that the resulting obligation, once
found to exist, would, as in the Nuclear Tests cases, be an erga omnes obligation. Yet, it should be said that while it is sensible to assume that states
are reluctant to undertake an obligation vis--vis the entire international
community (which is presumably to be understood as all states and international legal actors possibly concerned), the underlying restrictive interpretation could, of course, have led the Court to find that the declaration
in question was not in fact directed to no particular recipient. Instead, it
could have assumed that a declaration which is made on a subject matter
which is of direct concern only to a limited number of states (or possibly
even only one state as, for example, Burkina Faso in the Frontier Dispute
case) is reasonably to be understood as being directed only to the state(s)
directly involved. The Court, however, took this approach in neither
the Frontier Dispute case nor the Nuclear Tests cases; even though in the
latter case, it stressed that the French declarations had to be seen in connection with the proceedings instigated by New Zealand and Australia.
Consequently, at present, a declaration that does not directly address and
thereby identify its beneficiary, but is instead made openly to the mass
media, will trigger the Courts heightened scrutiny, as it is understood to
have been directed to the whole international community. As such it will,
more often than not, qualify only as a political statement.
(ii) Interpreting a Declarations Text
While therefore assuming a restrictive, and for potential erga omnes declarations an additionally cautious approach, the question remains as to when
a state can nevertheless be taken to have manifested its intent to commit
itself legally. According to the ICJ in the Nuclear Tests cases (see above text
to n 109), the intention to be bound will be deduced from a declarations
actual substance as well as the circumstances surrounding its making. In Armed
Activities on the Territory of the Congo, the Court spoke not of a declarations
actual substance but its actual content,126 and it is a declarations content which has to be taken into account also according to the ILCs Guiding
Principle 3.127 However, and irrespective of whether we concentrate on a
declarations actual substance or its content (actual or not) in order to
decipher a states declaration, a problem is bound to arise which Tomuschat,
in addressing Guiding Principle 3, has identified as follows:
126
ICJ, Armed Activities on the Territory of the Congo, Jurisdiction and Admissibility (2006),
n 40 above, para 49. For more see p 169.
127
See p 183. For the text of Guiding Principle 3, see Annex II.

On the Intention to be Legally Bound 215

It is remarkable that Principle 3 refrains from directing the reader to the text of
a declaration as the primary source for ascertaining its meaning, but focuses on
the content which, of course, will have to be elucidated first.128

Indeed, neither a declarations content nor its substance are openly

identifiable but will themselves have to be ascertained via an act of interpretation, which explains why both will be of little help in that very
process. The Court, in Armed Activities on the Territory of the Congo, has
provided the most transparent assessment of a declarations legal value
so far. There, after having stated that it must examine the actual content
as well as the circumstances in order to determine a declarations legal
effect, it continues by emhasising:
On the first point [ie that of content], the Court recalls that a statement of this
kind can create legal obligations only if it is made in clear and specific terms.129

Content therefore took the Court right to a declarations terms. Despite

the reference to a declarations content or substance it is thus indeed its
text which will be the primary focus of analysis in order to determine
whether it is legally binding or not. Here, there can be little doubt that the
wording used will have to be interpreted in accordance with its ordinary
meaning as is laid down for treaties in Vienna Convention, Article 31.
As the declaration is an act of communication which will be taken into
account and relied upon by the addressee, its interpretation cannot simply be based on the declarants perspective, for example, its special use of
wording, unless the latter is known to the addressee and could hence be
understood as deviating from its normal meaning. While the parallelism
in interpreting the wording of a unilateral act and a treaty will therefore be
strong, a statements unilateral character may nevertheless lead to some
differences, as the ICJs holding in the Anglo-Iranian Oil Co case130 has illustrated. In it, both parties before the Court argued about the meaning of the
Iranian unilateral declaration under ICJ Statute, Article 36(2). Irrespective
of the fact that these unilateral undertakings might be governed by some
lex specialis,131 the Courts finding is of interest for all unilateral declarations, since it focused particularly on the unilateral character of the act.
The Court found that it had to go beyond a purely grammatical reading of
the declaration before it, as the latter brought no conclusion as to which of
the two different interpretations adopted by the parties was correct. What
the ICJ found it had to seek was an:
Tomuschat, Unilateral Acts, n 43 above, 1502, para 3.
ICJ, Armed Activities on the Territory of the Congo, Jurisdiction and Admissibility (2006),
para 50.
130
See ICJ, Anglo-Iranian Oil Co (United Kingdom v Iran), Preliminary Objections, Judgment
of 22 July 1952, [1952] ICJ Rep 93, as well as the more recent ICJ, Fisheries Jurisdiction (Spain v
Canada), Jurisdiction, Judgment of 4 December 1998, [1998] ICJ Rep 432, 453, especially para
46 et seq, where the Court confirms its earlier findings.
131
See p 69 et seq.
128
129

216 The Law on Promises

interpretation which is in harmony with a natural and reasonable way of reading the text having due regard to the intention of the Government of Iran at the
time when it accepted the compulsory jurisdiction of the Court.132

The Court in addition emphasised that it had to interpret the declaration

as it stands, having regard to the words actually used.133 While a natural and reasonable way of reading the declaration has thereby been confirmed by the Court as guiding it, it also referred to the Iranian intention as
having to be considered with due regard. In looking for a manifestation
of a states will to be legally bound, it would, of course, lead us only in
circles were we to first stipulate that the latter had to be ascertained on the
basis of a declarations text, only to then declare that the text itself had to
be interpreted in light of a states intention. Such an approach as in fact
proposed by the Special Rapporteur134 has therefore rightly been criticised
within the ILC and is nowhere to be found in the Guiding Principles.135
However, stressing the importance of a states intention has led the ICJ in
the Anglo-Iranian Oil Co case to pay special attention to the circumstances
in which the declaration was made (circumstances to which we will turn
in a moment) while in addition slightly altering the rules of interpreting
the statements terms. It is this aspect which merits some closer attention
here. The United Kingdom had argued that if the Iranian interpretation
was adopted, the declaration would contain superfluous wording, while
a legal text should be interpreted in such a way that a reason and a meaning can be attributed to every word of the text.136 The Court, agreeing with
the consequences which the Iranian interpretation would have for the text,
however disagreed that such a reading should be eschewed:
It may be said that this [the UK suggested] principle should in general be applied
when interpreting the text of a treaty. But the text of the Iranian Declaration is
not a treaty text resulting from negotiation between two or more States. It is the
result of unilateral drafting by the Government of Iran which appears to have
shown a particular degree of caution when drafting the text of the Declaration.
It appears to have inserted, ex abundanti cautela, words which, strictly speaking,
may seem to have been superfluous.137
ICJ, Anglo-Iranian Oil Co (1952), n 130 above, 104.
Ibid 105.
134
See Victor Rodriguez Cedeo, Fourth Report on Unilateral Acts of States, UN Doc A/
CN.4/519 (2001) para 154, Art (a), para 1: A unilateral act shall be interpreted in good faith
in accordance with the ordinary meaning given to the terms of the declaration in their context
and in the light of the intention of the author State; see also Victor Rodriguez Cedeo, Fifth
Report on Unilateral Acts of States, Addendum, UN Doc A/CN.4/525/Add.1 (2002) para 135.
135
For criticism, see, eg Gaja ILC, 2695th Meeting, n 99 above, para 14. In the light of the
intention of the author State was no longer mentioned by the Special Rapporteur within
his proposal on the interpretation of a unilateral act in his ninth and final Report, see Victor
Rodriguez Cedeo, Ninth Report on Unilateral Acts of States, UN Doc A/CN.4/569 (2006)
Principle 11.
136
ICJ, Anglo-Iranian Oil Co (1952), n 130 above, 104.
137
Ibid 104.
132
133

On the Intention to be Legally Bound 217

As unilateral acts will usually have been unilaterally drafted by the

declarant,138 the result is that their terms will usually not be the result of
long negotiations in which every word is debated, often redrafted and
by this very process precisely determined. Such declarations may thus be
redundant and not live up to the standard that is to be found in a treaty.
However, and on account of a natural and reasonable reading being
attributed to a states use of words, the standard of interpreting the text
of a unilateral declaration will, apart from that, not differ much from
the standard applied to treaties.139 This being said, there is still another
aspect where the two regimes will be different, because unlike in the case
of treaties, in the case of unilateral declarations additional textual support cannot usually be gained from a declarations Preamble, its Annex
or the travaux prparatoires.140 The Special Rapporteur felt differently, however, and sparked a debate on the question by proposing to lay down in a
draft article that the context for the purpose of interpretation of a unilateral act shall comprise . . . its preamble and annexes. He also wanted to
allow recourse to be taken, inter alia, to a declarations preparatory work
as supplementary means of interpretation.141 But both propositions were
not approved by the majority of ILC members,142 and for good reasons.
Preambles and Annexes will usually be missing for unilateral declarations
and there is thus no need to make special reference to them. In the rare
event where they might be present, they can then be taken into account
as part of a statements context. The problem posed by a declarations
138
An exception may be found at p 88 et seq where the declarations on the protection
of minorities were drafted not by their declarants but by the League of Nations and then
passed on to the various states in order to be proclaimed by them as their own unilateral
undertaking.
139
As Tomuschat has emphasised for declarations under ICJ Statute, Art 36(2): Thus, if
there is any departure form the general rules of treaty interpretation in international law,
the distance can only be slight, Christian Tomuschat, Commentary to Art. 36 ICJ Statute in
Andreas Zimmermann, Christian Tomuschat and Karin Oellers-Frahm (eds), The Statute of
the International Court of Justice: A Commentary (Oxford, 2006) 589, 627, MN 65. Of those states
which replied to the ILCs questionnaire some emphasised that the principles of interpretation were similar, such as Finland, Israel, Italy, El Salvador, the Netherlands and, to a certain extent, also Sweden (speaking of a useful guide). Argentina and Austria referred to the
ICJs rulings which were mentioned above and the (slightly) different standard established
therein; see the answers given to questions 1 and 7 by the respective countries, as published
in ILC, Government Replies to the Questionnaire, n 2 above.
140
For treaties see Vienna Convention, Arts 31(2) and 32.
141
Cedeo, Fourth Report, n 134 above, para 154, Art (a) and (b). See also the largely recapitulative Fifth Report, Cedeo, Fifth Report, n 134 above, para 135, Art (a), para 1.
142
For criticism see especially Pellet, ILC, 2695th Meeting, n 99 above, para 5 et seq; also
Gaja, ibid paras 1617 and ILC, Report of the Fifty-fourth Session, A/57/10, n 12 above,
in which the Commissions debate on the Fifth Report is summarised, para 404 et seq. In
response to the criticism voiced in the Commission, Cedeos Ninth Report and the principle of interpretation proposed therein contains no further such references, see Cedeo,
Ninth Report, n 135 above, Principle 11 and the explanation provided by Victor Rodriguez
Cedeo, Ninth Report on Unilateral Acts of States, Addendum 1, UN Doc A/CN.4/569/
Add.1 (2006) para 158.

218 The Law on Promises

preparatory work is, on the other hand, more serious, as even where it
exists, it will usually have been accessible and known only to the declarant and not the addressee. Were it therefore subsequently to be taken into
account, for example during judicial review or simply to reassess a statement, declarations publicly made and apparently manifesting a will to be
bound could still not be trusted as such, since hidden evidence of a states
intentions could come to light and threaten to reframe the statement made.
As it is the reasonable addressees reliance which is protected under the
doctrine of promises, meant to promote trust and confidence in state declarations, the preparatory work will hence have to be left aside, with the
possible exception where it was easily accessible for those addressed.143
(iii) Clear and Specific Wording
So far it has been found that a declaration will have to be interpreted
restrictively, with the primary focus lying on its text, ie the wording actually used, interpreted in a natural and reasonable way. In doing so, the
Court has emphasised that in order to be legally binding, a declaration will
have to be made in clear and specific terms.144 But when is this the case?
To begin with the more clear-cut cases, a reasonable addressee or a
court will find it relatively easy to deduce a states intent to be bound
from its wording, where it has gone as far as expressly addressing its intention to create a legal obligation. While such statements may be rare, the
declaration of the British representative before the European Court of
Human Rights (ECtHR) asking that court to identify the declaration as
constituting an undertaking possessing legal effect145 shows that they
may exist. The Egyptian declaration is another fairly clear-cut example
in this regard, as it expressly spoke of obligations and qualified itself
as an international instrument, to be registered as such.146 Where a state
includes express references to legal obligations, undertakings with legal
effect, etc, there can be little doubt that its statement will be taken to have
actually expressed its will to undertake a legal commitment. The ICJ and
legal doctrine along with it have, however, gone beyond such scenarios
and have not limited their focus to declarations which are expressly
labelled as legal undertakings by their declarants. The Court otherwise
could not have held the French statements or any of those made before it
during proceedings to be binding, nor would its findings that a particular
declaration before it was non-binding have warranted more than a single
As proposed by Gaja, ILC, 2695th Meeting, n 99 above, paras 1617.
See especially ICJ, Armed Activities on the Territory of the Congo, Jurisdiction and
Admissibility (2006), n 40 above, para 40 and ICJ, Nuclear Tests (Australia v France) (1974),
n 6 above, 267, para 43, together with 269, para 51. See also Guiding Principle 7, first sentence, as already addressed at p 191 et seq.
145
For more see p 141 et seq.
146
For more see p 108 et seq.
143
144

On the Intention to be Legally Bound 219

sentence by which it was dismissed simply because it lacked any selfcharacterisation as a legal commitment. As is typical for municipal and
other areas of international law, there is therefore no need for the legal
subjects to use legal terminology in order for their declarations to have
legal effect, and to expressly identify their legal bindingness. While a declaration doing so will thus be understood as expressing a states intent, its
absence will not, as such, be a hindrance to the consideration of a states
nevertheless having manifested its will to be bound. It is therefore by no
means a necessary requirement.
There are additional textual indicators which, although not necessarily
required, will also support the finding that a state has conveyed its intent
to legally oblige itself to the reasonable addressee. Klabbers, in his dissertation on the concept of treaty, has analysed a few of these indicators
in order to test whether a bilateral act will be taken to constitute a binding
treaty or not.147 Although none of them were found to constitute a necessary or a watertight indicator, they are nevertheless of importance and are
all likely to tilt the scale further towards a finding of legal bindingness
where present. Here, there is no apparent reason why a finding that is
important for the interpretation of a potential bilateral commitment will
not have a similar impact if included in a unilateral declaration. To begin
with, where a declaration speaks of its own entry into force, it more often
than not will be taken to allude to its own legal nature, unless there is
evidence to the opposite.148 Similarly, where a states declaration as to a
certain future conduct allows for judicial settlement, it thereby includes a
strong indicator that the pledges contained can be treated as legally effective, ie binding. Also, where a declaration includes remedies or allows for
sanctions where it is breached, this, too, is a textual indicator of an intention to be legally bound.149
Yet, as can be inferred not only from the Nuclear Tests cases but practically every example in chapter two where a statement was found to be
binding by a court, the minimum requirements for a statement to be clear
can be fulfilled without alluding to its own legal nature in the above way.
What is apparently necessary, however, for a statement to qualify in this
regard is, first of all, that it expresses a states unconditional decision to
follow a certain line of future action. If we recall what has been stated in
the UN memorandum on unilateral declarations against torture and other
inhumane treatment written by the UN Secretariat,150 it stressed that they
could not be taken as expressing a legal commitment, since states merely
Klabbers, Concept of Treaty, n 106 above, 68 et seq.
For the parallel treaty scenario see the quite critical analysis in this respect by Klabbers,
ibid 7577.
149
Where such assurances are followed by completely vague pledges, they will, however,
be of little avail. For the parallel problematique in respect of treaties, see again ibid 7778.
150
See pp 14041.
147
148

220 The Law on Promises

stated their intention to comply and to implement, rather than indicate
that they will or undertake to comply and implement. According to
this opinion, in other words, a legal commitment will only be deducible
where a state has decided to act (or to refrain from acting) in a certain
manner and indicates this through appropriate wording (ie will do,
undertakes to, although in the latter case there is, arguably, already evidence of an express reference to a legal obligation undertaken) and not
where the action is merely intended.
To make distinctions along these very precise lines is sensible in light of
the fact that lawyers have developed a code on which to rely, and one that
helps to distinguish legal undertakings from political statements in this
regard. Be it in civil law contracts, legal texts or the law of treaties, precise
wording is what lawyers will concentrate on first for lawyers, a shall
will not be taken to have merely meant should, to name but one typical
and fine yet fundamental distinction. A (state department) lawyers common legal code will therefore be of importance in this area, particularly
for prepared speeches and declarations and any written declaration notified, as the latter will be understood as having been carefully scrutinised
and the words to have been chosen with care. Terms which indicate an
unconditional decision such as will, hereby ensure to, or shall, guarantee to or even undertake to, will therefore form a necessary requirement in order to assume a legal commitment and hold a statement to be
clear in this respect.151
The ICJ, and the ILC in its Guiding Principle 7, have, in addition, stipulated that a declaration must also be specific. Although the ICJ does not seem
to have greatly differentiated between the two terms in the Nuclear Tests
cases,152 it appears to have done so in the Armed Activities on the Territory
of the Congo case, where it recalled the requirement of a statement having to be made in clear and specific terms. A declaration will therefore
not only have to be clear in that it announces a states firm decision to act
in a certain way, but the conduct announced will, in addition, have to be
sufficiently specified. In this respect, the ICJ has been shown to indeed
apply a restrictive interpretation. As will be recalled, the Rwandese statement under scrutiny before the Court in Armed Activities on the Territory
of the Congo was that past reservations not yet withdrawn will shortly
be withdrawn. Although the Rwandese Minister of Justice had therefore
used clear terms as to her countrys decision to do something (will be
withdrawn), this something was in the eyes of the Court not precisely
151
As seen, the ICJ in the Nuclear Tests cases simply avoided a precise analysis by considering the statements to constitute an artificial whole. On this basis it concluded that the
declarations clearly conveyed the decision to move to underground testing.
152
ICJ Nuclear Tests (Australia v France) (1974) 269, n 6 above, para 51, where it found
[t]he objects of these statements are clear, while it referred to legally binding declarations
possibly and often being very specific at 267, para 43.

On the Intention to be Legally Bound 221

enough determined to allow the statement to be understood as a legal

commitment. As seen, the Court stressed that, in its opinion, the statement had to be understood as referring not only to the reservations made
by Rwanda to the human rights treaties but to any treaty concluded. The
restrictive interpretation surfaced again as the Court, apparently considering it highly unlikely that such a broad undertaking had been intended,
on this basis moved to hold the declaration therefore to be of an indeterminate character. The time-frame incorporated was, in addition, found
not to be sufficiently specific as it only stipulated that reservations would
shortly be withdrawn. In the words of the Court the statement by the
Rwandan Minister of Justice was not made in sufficiently precise terms in
relation to the particular question of withdrawal of reservations in order
for it to assume a legal undertaking. The statement could therefore at
most be interpreted as a statement of intent, very general in scope.
As has already been emphasised when discussing the case,153 the ICJ
could, of course, just as well have found the declaration to be (at a minimum) related to human rights treaties and the word shortly to be indeterminate but not indeterminable. To refrain from determining the rather
ambiguous in applying a restrictive approach, however, makes sense,
especially where there are no other compensating factors, which strongly
support a finding that a state nevertheless intended to establish a legal commitment. Otherwise, the clear and specific standard would be rendered
meaningless by the Court which, by filling in what has quite possibly been
deliberately left vague, could move practically every declaration into the
realm of legal bindingness. The precise undertaking would then, however,
be crafted rather by the mind of the Court (or the addressee were it allowed
to do so) than by that of the state which has made the statement. Under
the restrictive approach, a reasonable addressee will hence need to refrain
from interpreting a vague announcement as being binding. Many declarations of states which are reported in the daily media will not meet the
required threshold in this respect. Where states, as so often, pledge to move
forward, to respond to a threat, to fight against, be it corruption, hunger
or AIDS, to improve their efforts or to unleash the creative power of its
best scientists, engineers and entrepreneurs,154 the statements made simply
do not contain specific enough undertakings. As politicians more often than
not, in weighing their words, refrain from being precise, a clear announcement of a states future course of action will trigger particular attention.
That it may in fact do so is thus confirmed and supported by the standard
of interpretation applied to unilateral pledges of states.
See p 169 et seq.
In his speech US President Obama at the United Nations Climate Change Summit on
22 September 2009 announced that the United States: will unleash the creative power of
our best scientists, engineers and entrepreneurs to build a better world, available at www.
america.gov.
153
154

222 The Law on Promises

Yet, a caveat is required. Even where a state has made a clear and
specific statement determining its future course of action, further legal
analysis can come to conclude that the declaration may nevertheless not
be relied upon as a legal undertaking. It may, for example, turn out to be
truly a mere witticism as Mali had in fact argued in the Frontier Dispute
case, or an announcement of a domestic decision intended to remain a
purely domestic matter, it can also be merely a states declaration of goodwill or a purely personal pledge of the state representative who made it.
To differentiate between these and a states truly binding undertakings,
the circumstances in which a declaration was made will be decisive.
(iv) Supporting Circumstances
Even where a clear and specific statement in respect of a states future
conduct is provided, there is, as just mentioned, still some room for uncertainty as to whether the declaration is to be interpreted as a truly legal
undertaking; here, the circumstances in which it was made will be crucial,
a fact that was stressed by the ICJ and also the ILC.155 The problem with
any reference to the circumstances of a case is, of course, that a factual
situation is usually not only complex but, in addition, unique and hence
different every time. By addressing the special circumstances in which a
statement was made, a black box is therefore opened and with it the door
to some quite possibly unforeseeable legal interpretation of a states declaration. Yet, on account of the complexities of social interaction, a precise
and all-including description of the circumstances which will influence
a reasonable addressee in the evaluation of whether a declaration may
be relied upon or not may not be expected. What can and should be provided is, however, an explanation of typical and influential factors, that
will be factored into the assessment of whether a clear and specific decision as announced by the declarant can be relied upon as a legal undertaking in light of the circumstances of a case. In this regard, the Courts
jurisprudence draws attention to the following factors.
When discussing how to interpret a declarations text, it has been
pointed out that Preambles and Annexes will usually be lacking in unilateral declarations, but, where present, will of course be used in ascertaining
the declarants manifested intentions. The same is true for accompanying
statements which, where present, will necessarily form part of the wider
context or circumstances in which a declaration was made, provided that
they were furnished alongside or before the statement in question was
made. The Court has made this point in relation to reservations made to
declarations under Article 36(2) of the ICJ Statute:
155
See the references to the ICJs jurisprudence provided in n 109 above and the ILCs
Guiding Principle 3.

On the Intention to be Legally Bound 223

regard must be paid to the intention of the Greek Government at the time when
it deposited its instrument of accession to the General Act; and it was with that
jurisprudence in mind that the Court asked the Greek Government to furnish it
with any available evidence of explanations of the instrument of accession given
at that time.156

Besides ministerial statements and press communiqus the ICJ, again

in addressing the scope of a reservation made to a declaration under
Article 36(2), has in the Fisheries Jurisdiction case referred even to parliamentary debates and legislative proposals as explaining the intention the
Canadian government had when making its declaration. In doing so, the
Court, however, referred to material that was openly available and indeed
known to Spain as the addressee in question; Spain had referred to the
Canadian parliamentary debates in order to bolster its own interpretation
of the Canadian reservation.157 Where such statements are openly made
or internal discussions have become generally known, for example via
the mass media, they will therefore be of importance as pertaining to the
circumstances in which a declaration was made.
As will be recalled, in the Armed Activities on the Territory of the Congo
case, the Courts finding did not hinge on the circumstances of the case,
as it had already found the statement before it to be indeterminate. It
nevertheless added that it found the presentation of general policy on the
promotion and protection of human rights before the UN Commission
on Human Rights to corroborate its finding that what was proclaimed
was not a legal undertaking. Despite the rather formal address given by
the Minister, the overall character of the speech and the direct context in
which the words under closer scrutiny were made, therefore signalled to
the Court that the latter could not be relied upon as expressing a legal
commitment. The ICJ can hardly be taken to have thereby announced that,
in order for a certain passage to be found to constitute a legal undertaking, the whole text in which it is contained will have to convey an overall
legal character; a legal undertaking can surely be included in an otherwise
more general speech. Yet, where the speech is held and appears to be a
general policy presentation, a statement will have to stand out in a particular way in order for it to be classified as different in character than the
rest of the statements made.
In the Frontier Dispute case, the declaration before the ICJ was clear and
specific in that Malis Head of State had said that if the Organization of
African Unity Commission decides objectively that the frontier line passes
through Bamako, my Government will comply with that decision. The
Court, however, did not analyse its wording, but jumped straight to the
156
ICJ, Aegean Sea Continental Shelf (Greece v Turkey), Judgment of 19 December 1978, [1978]
ICJ Rep 1978, 3, 29, para 69, emphasis added.
157
See ICJ, Fisheries Jurisdiction (1998), n 130 above, 463, para 75.

224 The Law on Promises

factual circumstances in which the act occurred158 in order to conclude
as it did, ie that Mali could not be understood as having conveyed binding force upon the Mediation Commissions conclusions through its statement, in light of the fact that it had already refused to do so in a treaty. The
facts which have lead up to a states declaration are, of course, part of the
context that will be taken into account. Where an ambiguous declaration
interpreted as a unilateral undertaking would require the state to have
changed its mind and now follow a new course of conduct, the standard
will presumably be higher and a commitment much more doubtful, than
where such an interpretation is fully in line with the recent signals sent
out by that state. This cannot only be deduced from the Frontier Dispute
case, but also from the North Sea Continental Shelf cases, where the Court
said:
In principle, when a number of States, including the one whose conduct is
invoked, and those invoking it, have drawn up a convention specifically providing for a particular method by which the intention to become bound by the
rgime of the convention is to be manifested namely by the carrying out of
certain prescribed formalities (ratification, accession), it is not lightly to be presumed that a State which has not carried out these formalities, though at all
times fully able and entitled to do so, has nevertheless somehow become bound
in another way.159

Although no potential promise was involved in that case, the above principle is as important as clear: since Germany had participated in drawing
up the convention that was open for ratification, its decision to refrain from
doing so was apparent. That it had now changed its mind and wanted to
become bound by the regime, but by alternative means, is of course possible, but improbable and hence could not be lightly presumed.
In the Nuclear Tests cases, the ongoing judicial proceedings before the
ICJ were of importance in judging the French declarations, especially
as Australia and New Zealand had made it clear that they had never
received a firm assurance from France in this matter. As the ICJ emphasised, the French statements had to be seen in this context, they were not
made in vacuo.160 Apart from statements made in direct reference to an
ongoing case, the courtroom setting has proved to be of particular import
ance amongst the set of circumstances in which a declaration was made.
It has, as such, allowed courts to assume a legal commitment rather easily
where a declaration announcing a states future conduct has been made
before them. Although the courts never elaborated upon the reasons for
doing so, they obviously assume that a state representative must be aware
of the attention paid to his or her words and therefore to have adopted
ICJ, Frontier Dispute, Merits (1986), n 109 above, 574, para 40.
ICJ, North Sea Continental Shelf, Judgment of 20 February 1969, [1969] ICJ Rep 3, 25,
para 28.
160
ICJ Nuclear Tests (Australia v France) (1974) 269, n 6 above, para 50.
158
159

On the Intention to be Legally Bound 225

them with particular care, as a result of which they may be relied upon as
firm commitments. The PCIJ has hence considered assurances to refrain
from expropriation and nationalisation as beyond question and no
doubt binding.161 The arbitration tribunal in the Filleting of Fish in the Gulf
of St Lawrence arbitration162 applied the same approach, and while pointing to the importance of the context in which they were made, ie the
formal arbitration setting, concluded that they had to be considered as
an engagement for France. It will be recalled that the ICJ in the La Grand
case163 went as far as holding that:
If a State, in proceedings before this Court, repeatedly refers to substantial
activities which it is carrying out in order to achieve compliance with certain
obligations under a treaty, then this expresses a commitment to follow through
with the efforts in this regard164

and unanimously took note of this commitment undertaken and held it

to meet Germanys request for a general assurance of non-repetition. In
the case concerning Questions relating to the Obligation to Prosecute or Extradite,165
the Court, as seen, also referred to the statements made by Senegal as
being formal assurances. A state representatives clear and specific statement that its state intends to do or refrain from doing something in the
future, if made during formal judicial proceedings, is therefore understood as expressing a legal undertaking.
The importance attached to the formal courtroom setting emphasises
another aspect which will surely be of importance in interpreting a declarations bindingness, which is that of formality in general. As shown, the
ICJ has been clear, and the ILC has agreed, that no formal requirements
exist for unilateral declarations, and that whether a statement is made
orally or in writing makes no essential difference, for such statements
made in particular circumstances may create commitments in inter
national law.166 But this does not mean that the fulfilment or lack of form
has no impact on the interpretation of state declarations as manifestations
of their will to be bound. The fulfilment of formalities is not required in
order for a declaration to give rise to a binding commitment, as particular circumstances may underline its bindingness (eg having been made
in the courtroom). Informal oral declarations are hence a quick and flexible mechanism, capable of creating full-blown legal commitments. Yet, a
161
See the Mavrommatis Jerusalem Concessions case, discussed at p 83 et seq and the Certain
German Interests in Upper Polish Silesia case, discussed at p 85 et seq.
162
See p 155 et seq.
163
See p 166 et seq.
164
ICJ, LaGrand (Germany v United States), Merits, Judgment of 27 June 2001, [2001] ICJ Rep
466, 51213, para 124.
165
See p 172 et seq.
166
ICJ, Nuclear Tests (Australia v France) (1974), n 6 above, 267, para 45. For more see ch
one, n 25. See also the ILCs Guiding Principle 5.

226 The Law on Promises

heightened degree of formality, where present, will nevertheless still be
indicative of a states intention that its declaration should be taken very
seriously and, as such, be of importance in ascertaining a declarations
legal value: a particularly solemn form of public proclamation, an official
notification of a declaration to its addressee, or even the registration of a
declaration with the United Nations (as in the case of Egypts declaration
on the Suez canal) are thereby likely to be considered as part of the factual circumstances which, similarly to the courtroom setting, will be of
importance in assessing a declaration.
The Egyptian case, in fact, raises the particular question as to what can
be gained from the registration of a unilateral declaration.167 For treaties,
the act of registration at the international level is usually considered to
be an important indicator as to the required intent to be legally bound.168
Although Klabbers, in his analysis, draws the conclusion that it is not, as
such, an airtight indication169 of whether or not an agreement reflects the
parties intention to be legally bound, there is apparently no evidence of
any document which was in fact registered nevertheless being perceived
as lacking the required legal intention. What Klabbers illustrates, however, is that where registration is lacking, such non-registration cannot be
taken as proving the absence of a legal intent. For unilateral acts, a similar
conclusion can be drawn: as shown especially by the Suez Canal declaration, the registration of unilateral engagements with the United Nations is
likely to be allowed by the Secretariat and, where registered, this very act,
similarly as for treaties, will constitute an important factor indicating a
states intention to be legally bound by that declaration. However, where
registration is lacking, Klabbers finding for treaties is all the more true
for unilateral acts, where registration is very uncommon. Its absence will
therefore not have the slightest (negative) impact when assessing whether
a state has manifested its intention to become legally bound.
The comparison between unilateral acts and the law of treaties in
respect of formalities is of general interest, because although oral treaties are possible, they have apparently been of little impact in the past.170
The overwhelming majority of treaties dealt with by international lawyers are written and, as such, covered by the regime found in the Vienna
Convention. The main reasons for this phenomenon are surely not only
the fact that the content of written undertakings can more easily be
proved, and that they can be undertaken with more precision (especially
for long and complicated undertakings), but also that the fulfilment of
formalities might in addition signify the documents special, ie legal,
167
On whether unilateral declarations need to be registered under UN Charter, Art 102,
see p 246.
168
See Klabbers, Concept of Treaty, n 106 above, 79 et seq with further references.
169
Ibid 84.
170
See also Tomuschat, Unilateral Acts, n 43 above, 149596.

On the Intention to be Legally Bound 227

importance. Formalities are usually time-consuming and a state will

have, or can at least be understood as having, weighed its words more
carefully when it has taken some additional steps; a written document
officially notified to its declarant might for this reason be of some importance as such. Formalities thus warn the declaring state while informing
the addressee that the statement made can be considered more reliable.
It is true that an apparently carefully designed speech may have the very
same impact, and with regard to assurances, oral declarations have been
much more important than their written counterparts. Especially for declarations made erga omnes, intended for to whom they may concern, the
oral unilateral undertaking is useful. However, the lack of formality is
not a necessary characteristic of promises, where written undertakings
are of course equally possible. Formalities are therefore not required, yet
they may be of further indicative value in assessing a states intention to
become bound.
The last point in addition illustrates that whether a state representative
has had the time to choose his or her words carefully and reflect upon
their impact or not, will probably also influence the reasonable addressee,
and hence the Court, in interpreting a statement which is ambiguous as
to the required manifestation of intent. If we contrast a carefully crafted
speech with an oral declaration made on the spur of the moment, as for
example in response to an unforeseen question put forward in an interview, then provided it contains clear and specific wording, the former will
be of greater impact. Where a statement is repeated numerous times and
at the highest level, as in the Nuclear Tests cases, the state can also be taken
to have considered its statement well (the Court, as seen, went as far as
simply grouping the declarations together to form an artificial whole). The
Ihlen declaration, too, is interesting in this regard.171 Notwithstanding the
debate about its precise character (promise or offer) it was oral, yet is generally considered to have conveyed an intention to be legally bound. The
fact that it is held to do so is in part connected to the fact that, although
oral, it was by no means made on the spur of the moment. Minister Ihlen,
instead, asked for some time to consider the matter, and only over a week
later provided his response.
(v) Concluding Remarks on Interpretation
The above paragraphs represent what can currently be said about the
standard of interpretation applied to unilateral declarations in ascertaining whether they are state promises in international law. Although we
may hope for further international jurisprudence to elucidate and build
on the established principles (especially as the ILC has discontinued its
171

See p 93 et seq.

228 The Law on Promises

work), the present framework is not entirely devoid of appropriate guidance. Under it, states which intend to make use of a unilateral assurance,
be it for one of the various reasons mentioned in the Introduction to this
study, will be able to do so. They can expressly mark their declaration as
a legal commitment through appropriate wording, even register it with
the United Nations, or merely use very clear and specific wording, the
importance of which is underlined by an appropriate context, such as a
very solemn proclamation or an official notification.
As always when having to interpret whether a declarant has manifested
an intent to be legally bound, in the municipal civil law or the international
context, and for consensual undertakings or unilateral ones, the result is
a sliding scale. There will be relatively clear cases in which both wording
and circumstances enforce each other, more ambiguous ones where contradictory signs have to be interpreted, and finally statements which clearly
fall short of the required standard. In this respect, possible promises are
not that exceptional. In addition, the restrictive standard that has been confirmed will be used by the Court has made it unlikely that a state will stumble into an unforseeable commitment without its at least having held itself
out as intending to be bound, ie by having made a statement through one
of its official representatives at the highest level or one specially empowered, that was clear and specific about a future course of action and directed
to one or numerous states a statement which, in addition, was made in
a context that on account of its formality, or due to other circumstantial
evidence, stressed the declarations importance and thereby allowed for a
legal effect to be reasonably deduced from it. Where a declaration is made
to the world at large, the reasonable addressee will have to be even more
cautious. Where, on the other hand, a clear and specific position is assumed
as such and solemnly proclaimed, even under a restrictive standard of
interpretation there is no reason not to take a state at its word.
C Distinguishing Promises and Offers
The analysis of various declarations made in this study has repeatedly
provided us with the opportunity to emphasise that only by ignoring the
realities of social interaction between the actors involved could the statement be identified as giving rise to a treaty commitment. Indeed, numerous scholars have stressed that to take note of the existence of unilateral
and, as such, binding promises permitted a relinquishment of the completely artificial interpretation of offers and (usually tacit) acceptances in
cases where this structure is not to be found in the interaction between
the parties.172 For example, if we recall the declarations concerning the
172

References provided at n 70 above.

On the Intention to be Legally Bound 229

protection of minorities as described in chapter two,173 the Estonian representative went as far as expressly stating that his declaration was not to
be understood as giving rise to a treaty. Not surprisingly, the League of
Nations, to whom the declaration was addressed, also did not speak of
accepting an offer (which the representative said he had not made) but
instead took note of the undertaking, which hence had to be considered
already to exist as such. Similarly, and in keeping with the memorandum
signed at Moscow that it would make a declaration in a form which will
obligate Austria internationally, the latter made a declaration in the form
of a written notification to various states of its decision to remain permanently neutral. Although Austria asked for its new position to be recognised and not to be accepted by the states notified (and at least in
the case of the United States there is proof that this was exactly how the
request was understood, as the latter considered the legal effect which
the unilateral act of recognition has under international law), some legal
commentators still interpreted the scenario as giving rise to a treaty,174
despite thereby having effectively to redraft the states inadequate wording. The driving force behind any such interpretation was apparently less
what states actually did in the scenarios just described, but what scholars
perceived them to be legally capable of doing. Imposing an entirely bilateral mindset onto reality, in which even one-sided legal commitments are
brought about only through two matching manifestations of will, meant
that every scenario including a states manifested will to be legally bound
had to be pressed into the treaty structure in order to become effective, as
otherwise it would apparently have to be ignored. Since promises are by
now accepted by the ICJ, the ILC and, generally speaking, international
legal doctrine, as capable of obliging states in their international affairs,
such forced treaty constructions, while never convincing, are in addition
no longer necessary in order to adequately reflect an intent to be bound,
despite not having been made as an offer that was accepted as such.
What has, however, become necessary through legal doctrines acceptance of the existence of promises is to hold a unilateral undertaking in the
form of a promise apart from a bilateral, consensual commitment, especially as different legal frameworks will subsequently apply. This is true
not only at the international level, where a distinction between a promise
and a treaty undertaking will be important, especially for the question of
the commitments revocability, but it is likely to be of considerable impact
also at the national level, where often no parliamentary involvement is
required for a countrys executive to assume a new obligation by unilateral means.175 In interpreting a states manifested will to be legally bound,
See p 88 et seq.
See p 104 et seq.
175
For the situation in Germany where this is, for example, the case, see the holding of the
German Constitutional Court as addressed at p 148 et seq. Denmark and Sweden adopted
173
174

230 The Law on Promises

a promise will therefore need to be identifiable and held apart from an
offer, which forms the gateway to treaty undertakings.
At least in theory, an offer and a promise can be distinguished quite
easily, since an offer is always only a conditional undertaking, the condition being that it requires the recipients additional manifestation of will,
the acceptance, in order to oblige the declarant to do as offered. Where
a statement expressly asks to be accepted, no problems in interpretation
will arise. Similarly, an offer will be easily identifiable where, as is often
the case, it is made with the intention to create an obligation not only for
the declarant but also for the addressee, in other words, where it asks for
a quid pro quo. Even where the latter is not openly requested, a closer
analysis of the communication which has taken place between the parties
might reveal an exchange situation and the existence of reciprocal under
takings agreed upon in mutual dependence. If so, the declaration in question cannot be a promise. While in all these cases an offer is easy to spot,
the situation is often more complex, as international law does not require
the existence of a quid pro quo in order for a treaty to be valid as such; it is
indeed nowhere mentioned in the Vienna Convention and treaties obliging only one side are deemed possible.176 But, if this is so, then it is also
true for offers, in which only a one-sided obligation for the declarant is
proposed. This is why and where the determination of whether a declaration constitutes a treaty offer or a promise can become tricky. On account
of their different legal characteristics, the answer will, abstractly speaking, need to be found by highlighting the realities of social interaction
which have been referred to above; the question that has to be answered,
in other words, is whether a state has manifested its will to undertake an
obligation the moment its statement is communicated. Does it present the
latter as a fact and a given or does the declarant only propose a possible
undertaking, the coming into being of which is put at the disposition of
the addressee and is, hence, intended to hinge on a positive decision of
acceptance?
With these questions in mind, a few declarations that have been identified as unilateral assurances, ie promises, will be recalled. If we begin
something like a middle ground between complete non-involvement of the Parliament
and full parliamentary approval. According to the Danish Constitutional Act, s 19 (3), The
Folketing [Parliament] shall appoint from among its Members a Foreign Affairs Committee,
which the Government shall consult prior to the making of any decision of major import
ance to foreign policy, see the English translation available at www.eu-oplysningen.dk.
Swedens Constitution requires parliamentary involvement only for the conclusion of
agreements, but establishes an Advisory Council on Foreign Affairs composed of the
Parliaments (Riksdags) Speaker and at present 18 members (nine full + nine deputy members) of Parliament, chaired by the King of Sweden. According to ch X, art 6, para 2, second
sentence of the Swedish Constitution: In all foreign policy matters of major significance,
the Government shall confer with the Council, if possible, before making its decision; see
the English translation at www.riksdagen.se/templates/R_PageExtended____6327.aspx.
176
See ch one, n 149.

On the Intention to be Legally Bound 231

with the statement by the British representative before the PCIJ not to
expropriate Mavrommatis, this pledge was presented as an unconditional
fact: there can be no question of our acting on any request to expropriate
M Mavrommatis.177 As such, it is not presented as a possible undertaking,
which, as in favour of Greece, is left at the latters discretion. Britain will
not expropriate Mavrommatis period. The same is true for the statements made by the Polish representative in Certain German Interests in
Upper Polish Silesia,178 where it was assured that certain properties would
not be expropriated. Any such action was ruled out by the declarants
through an unconditional statement, which the Court, without hearing
the beneficiarys position on the matter, interpreted as binding. The very
same obligation could, of course, have been presented as an offer, but
the declarants, apparently motivated by their wish to impress and influence the Court,179 chose not to do so. The German assurances under Nazi
Germany180 were, of course, also not presented as an offer not to attack
and understandably none of the addressees appear to have been under
the impression that such an offer had to be accepted to become binding. Instead, Germany announced (and pretended) that an attack was out
of the question and that the German decision never to attack was firmly
established and as such assured. The Austrian declaration on neutrality
has already been recalled, in which Austria similarly notified its neutrality as a given.181 It only asked for its recognition as thereby it would be
provided with the advantages of its neutrality having to be respected by
the recognising state, an effect that its unilateral undertaking alone could
not have had. However, it did not ask for its declaration to be accepted in
order for its permanent neutrality to become effective, a position which
it had already domestically enshrined in a constitutional law. Egypt
too, proclaimed its regime for operating the Suez Canal by depositing its
unilateral document with the United Nations as in place and effective as
proclaimed.182
In all these cases, the obligation is presented by the declarant and understood by the addressees to have already been adopted as an undertaking.
There is not sufficient space to review all the examples that have already
been discussed in chapter two, but the point should be clear: the question as to how to differentiate between a gratuitous offer and a promise is
whether a state in its declaration announces a proposition that it is willing
to follow only once accepted, or whether it announces an obligation as
177
PCIJ, Mavrommatis Jerusalem Concessions (Greece v Great Britain), Judgment of 26 March
1925, Series A 5, 6, 37.
178
See p 85 et seq.
179
For other reasons possibly motivating a state to undertake a unilateral obligation
through a promise, see pp 610.
180
See p 101 et seq.
181
For more see p 104 et seq.
182
See p 108 et seq.

232 The Law on Promises

thereby undertaken and presents the latter as such. Under this perspective, a declaration such as the Chinese pledge vis--vis the entire commun
ity of states to undertake unconditionally not to use or threaten to use
nuclear weapons against non-nuclear countries and nuclear free zones183
is clearly not an offer. And if one of these statements, to take the Japanese
announcement to suspend drift-net fishing as another example,184 met
with another states We accept, the reaction would indeed come rather as
a perplexing surprise than as a required and expected response, the reason being that the acceptance has not been asked for by the declarant. If
France declares that it will move to underground nuclear testing at a certain time, it does not ask Australia and New Zealand for their acceptance
of this plan, nor does it therefore ask for an acceptance of this statement to
be binding, once interpreted as an assurance, on account of it manifesting
an intent to be legally bound. Of course, a beneficiary may indicate that
it welcomes another states undertaking or that it considers it to be adequate to bring a conflict of interest to an end. Non-technically speaking, it
might therefore accept such a statement, but only as settling a matter or
as welcome, yet it has no power to turn the statement made into an offer
and by accepting it, will not create a treaty relationship.
Analysing what the declarant has proclaimed will allow for a distinction to be made as easily as in the examples described above in many,
but not all cases. As with any field of law, there are the hard cases which
need to be considered. The scenario in the Eastern Greenland case has highlighted an area where the call is a close one, and it has consequently given
rise to ample debate.185 The PCIJ chose rather not to answer the question
of the legal nature of the commitment before it, while scholars who have
done so are split into two approximately equally sized, yet fundamentally
opposed, camps. In chapter two, the History of Promises, this writer
was able to refrain from providing an opinion on how the case should be
decided, as there the emphasis lay rather on analysing the position taken
by the Court and the response it triggered amongst scholars, in order to
highlight the judgments importance in the development of the doctrine of
promises in international law. At this point, however, an answer as to how
to address such a scenario has to be found. By such a scenario the following is meant. The reason why the characterisation of the Ihlen declaration
still divides legal opinion is, first of all, that it cannot simply be identified
as involving an offer, as it does not include a quid pro quo. It is (of course)
not only the proponents of promises who accept this as a fact, but also the
treaty supporters who follow Judge Anzilottis analysis. It will be recalled
that Anzilotti, in his dissenting opinion, rejected the understanding
according to which the declaration gave rise to a Spitzbergen-Greenland
183
184
185

For these so-called negative security assurances see p 161 et seq.

As already referred to in Introduction, n 28.
For details and references see p 93 et seq.

On the Intention to be Legally Bound 233

deal, yet for him it nevertheless constituted an agreement. If so, it was

an agreement which imposed obligations only on Norway and none on
Denmark. Those joining him in his analysis have placed special emphasis
on the fact that the Ihlen declaration constituted a reply in response to
a request, as the Court has stated in its ruling. And indeed, when the
above-stated examples of fairly clear unilateral promises are compared
with the declaration made by Mr Ihlen and the situation in which it was
given, the latter is exceptional on account of it forming an integral part of
a bilateral negotiation. If we return to the relevant passage in the minute
which recorded the dialogue and the statement made, it reads:
The Danish Government is confident (he added) that the Norwegian Government
will not make any difficulties in the settlement of this question. I replied that the
question would be examined.

The Danish confidence expressed in respect of Norways future and

uncertain course of action is nothing but a question posed by Denmark
in this respect, the request, and equivalent to it asking directly: Can we
be sure that you, Norway, will not make any difficulties in the settlement
of this question?. This, in this writers understanding, was clearly not an
offer, as Denmark was offering no sort of obligation on its part.186 It instead
asked and thereby invited Norway to assume a clear position and even
undertake a legal obligation in the matter. Norways representative then
responded: Today I informed the Danish Minister that the Norwegian
Government would not make any difficulties in the settlement of this
question. The question for us is whether this answer was presented as
an unconditional commitment by Norway, as a promise, or whether it
conveyed the message that it was only made as dependent on the (continued) Danish acceptance. Here, the fact that the statement itself makes
direct reference to the request (for the settlement of this question) and was
declared only and directly to the country asking for Norways position
(with no other audience), allows Ihlens statement to appear as a concession wholly motivated by the Danish request and adopted by Norway
only if Denmark should still so require. On a closer look it is, therefore, in
this writers opinion, indeed best classified as an offer, and not a promise,
not to make any difficulties in the matter just what Denmark had hoped
to receive and hence, as signalled before, accepted.
What is true for the Ihlen declaration, and made the assessment particularly difficult as it involved neither a clear-cut promise nor a clearcut offer, will be true for most negotiated declarations directed solely at
the negotiating partner after a back-and-forth communication has taken
place. Similarly, in the case of written declarations and statements which
include only a one-sided obligation (otherwise there is surely a consensual
186
On the offer-acceptance interpretation in this scenario, see the analysis provided at
p 99 et seq.

234 The Law on Promises

bond), but which are worked out (and possibly even signed) by more states
than the one undertaking the obligation, the latter will usually have been
adopted only under the condition of the beneficiarys consent. If so, the
obligation flowing therefrom is a consensual treaty undertaking rather than
a unilateral promise.
To conclude, most offers are hence easily discernible as such via the
requested quid pro quo. Unilateral promises will similarly be often easily
identifiable as such. In dealing with the hard cases and in determining
whether a declaration is really a promise and not a gratuitous offer aiming at the conclusion of a treaty that obliges only the declarant, particular
focus will have to lie on the actual communication between the parties.
What has to be ascertained is whether or not the declarant has asked for
its announcement to be accepted as such, and has made it clear that it
intends an obligation to exist, only if and once the declaration is accepted
by its addressee. Where this is the case, and the obligation undertaken is
hence merely proposed and left to be brought to life by the addressee, it
will be an offer; if it is assured as a given that will no longer be subject to
debate, the declaration with the obligations included therein constitutes a
unilateral promise.
V FURTHER REQUIREMENTS

For a declaration which manifests an intention to be legally bound to

actually compel the state whose representative is making the statement,
there are the following further requirements.
A Competent Representative
As already briefly touched upon when addressing the ILCs Guiding
Principles, a declaration will first of all need to be imputable to a state
in order to be able to constitute a legal commitment for the latter. As will
be recalled, the state representatives considered competent and empowered under international law to compel their state through treaties, as
enshrined in the Vienna Convention, Article 7(a), are also deemed competent to adopt legal obligations unilaterally for their states. It is furthermore generally accepted that a state may allow other personnel of its own
choosing to do so. In this respect, the ILCs Guiding Principle 4 is clear,187
following the ICJs wording in the Armed Activities on the Territory of the
Congo case, which it reflects:
187

See pp 18586.

Further Requirements 235

the Court observes that, in accordance with its consistent jurisprudence [references omitted], it is a well-established rule of international law that the Head of
State, the Head of Government and the Minister for Foreign Affairs are deemed
to represent the State merely by virtue of exercising their functions, including
for the performance, on behalf of the said State, of unilateral acts having the
force of international commitments . . .
The Court notes, however, that with increasing frequency in modern international relations other persons representing a State in specific fields may be
authorized by that State to bind it by their statements in respect of matters falling within their purview. This may be true, for example, of holders of technical
ministerial portfolios exercising powers in their field of competence in the area
of foreign relations, and even of certain officials.188

While this position has ample support,189 neither the Guiding Principles
nor the ICJ have expressed an opinion on the question of how a lack of
municipal empowerment and actual competence might affect the unilateral commitment proclaimed at the international level. Here, various
scholars have argued that the provision found in Vienna Convention,
Article 46 should be applicable, mutatis mutandis, to unilateral acts.190
According to that Article, provisions of internal law regarding the competence to conclude treaties, if breached, cannot be invoked to invalidate
the consent given, with the exception of cases where such violation was
manifest and, in addition, concerned a rule of internal law of fundamental
importance.191
A similar, though not entirely parallel, provision for unilateral acts,
which was not to be restricted to internal norms regarding only the competence to conclude unilateral acts but was to apply to all norms of fundamental importance, had been proposed within the ILC through the
Special Rapporteurs Second Report192 and, changed to be even broader,
his Third Report. According to the latter, there were grounds to invalidate
a unilateral act if it conflicts with a norm of fundamental importance to
188
ICJ, Armed Activities on the Territory of the Congo, Jurisdiction and Admissibility (2006),
n 40 above, 27, paras 46, 47.
189
Besides the ILCs Guiding Principle 4, see also the answers provided by Argentina,
Austria, El Salvador, Finland, Netherlands and Sweden to question 2 in ILC, Government
Replies to the Questionnaire, n 2 above.
190
See, eg Skubiszewski, Unilateral Acts, n 11 above, 230; Sicault, Engagements unilatraux, n 70 above, 65859.
191
Vienna Convention, Art 46 reads: (1): A State may not invoke the fact that its consent to
be bound by a treaty has been expressed in violation of a provision of its internal law regarding competence to conclude treaties as invalidating its consent unless that violation was
manifest and concerned a rule of its internal law of fundamental importance. (2) A violation
is manifest if it would be objectively evident to any State conducting itself in the matter in
accordance with normal practice and in good faith.
192
See the proposed Draft Article 7(7): A State may invoke the invalidity of a unilateral
act: . . . 7. If the expression of a States consent to be bound by a unilateral act has been in
clear violation of a norm of fundamental importance to its domestic law, Victor Rodriguez
Cedeo, Second Report on Unilateral Acts of States, Addendum, UN Doc A/CN/500/
Add.1 (1999) 9.

236 The Law on Promises

the domestic law of the State formulating it.193 Various Commission members, however, rightly stressed that such a wide concept, allowing fundamental municipal legal provisions to penetrate deep into the international
sphere and cancel any commitment undertaken, was untenable.194 Besides
not referring to norms relating to competence, the Third Report did not
even mention a clear violation, as the Second Report had done. In his
Ninth Report, Cedeo proposed a draft article which mirrored the Vienna
Convention more closely, as it required a manifest contradiction in conflict
with a norm of fundamental importance.195 Cedeo nevertheless retained
the broader approach referring to all fundamental norms of domestic law
in this respect and not only to those regarding competence. However, it
is not only international lawyers with a dualist world view who should
reject such a construction, as it would require an addressee of a unilateral
act to delve into the provisions of the declarants national law, even if only
its fundamental, ie constitutional provisions, and to ascertain whether
the declarant has acted in contradiction of any of them. What may be a
relatively clear-cut procedure for fundamental norms of competence will
become quite complex if all norms of fundamental importance (including, for example, fundamental rights which usually allow for a balancing procedure on a case-by-case basis) are to be included. To reduce the
necessary screening to be undertaken by the addressee only to a lack of
competence is therefore reasonable, especially where it remains limited
to a finding which is objectively evident to any State conducting itself in
the matter in accordance with normal practice and in good faith, ie where
such a contradiction is manifest as defined in Vienna Convention, Article
46(2). Where a representative is competent according to this standard and
represents his or her state, it should be his or her task and not that of
the addressee to ensure that the commitment undertaken is in conformity
with its municipal provisions. As seen above in discussing the Guiding
Principles, the proposed article has not been adopted by the ILC. In light
of the criticism voiced within the Commission for good reasons, the proposals to adopt a parallel approach for unilateral acts, such as the one
taken by the Vienna Convention in respect of treaties, is most convincing.
193
The proposed Draft Article 5(8) read in full: A State may invoke the validity of a
unilateral act: . . . 8. If the unilateral act formulated conflicts with a norm of fundamental importance to the domestic law of the State formulating it, Cedeo, Third Report, n 97
above, 22, para 167.
194
See the doubts expressed by most members of the Commission as summarised in ILC,
Report on the Work of its Fifty-second Session, General Assembly, Official Records, Fiftyfifth Session, Supplement No 10, UN Doc A/55/10 (2000) 204, para 602.
195
See Guiding Principle 6: A State that has formulated a unilateral act may not invoke
as grounds for invalidity the fact that the act conflicts with its domestic law, unless it conflicts with a norm of fundamental importance to its domestic law and the contradiction is
manifest, Cedeo, Ninth Report, n 135 above, 2. No definition of manifest (as included in
Vienna Convention, Art 46(2)) was, however, proposed.

Further Requirements 237

This finding notwithstanding, it should be noted that at least at present,

any norm referring to the municipal regime and its regulations in respect
of the competence of a state representative can have only little practical
impact, because, as already stated, such regulations are to our knowledge
lacking in most constitutions for unilateral undertakings.
The Special Rapporteur did not put forward a parallel provision to
Vienna Convention, Article 47 (according to which specific restrictions
on the authority of a state representative need to be notified in order to
become effective), stressing that the unilateral nature of the act and the
absence of any negotiating parties rendered it meaningless in that area.
However, in the rare (but not impossible to conceive) event that a parallel scenario should arise and a specific restriction on a representatives
authority was imposed, the requirement of a restriction having to be notified in order to be opposable will, on account of the good faith basis of
such undertakings, surely apply.
B No Defective Will
That a states manifested will to be legally bound needs to be free of any
defects and will, mutatis mutandis, be subject to the same grounds of invalidity as those which are applied to a states manifestations of will to be
bound by a treaty is widely acknowledged.196 And rightly so, as there is no
reason why a state obliging itself, for example, under the impact of an error
within the (strict) limits laid down in Article 48 of the Vienna Convention
or procured through the corruption of its representative (Article 51), should
be treated differently than a state which has consented to the very same
obligation in a treaty. Generally speaking, there is no significant difference
between a manifestation of will giving rise to a treaty obligation and a unilateral promise in this regard. The Special Rapporteur, consequently, began
by proposing a draft article on the matter in his Second Report.197 The provision, with some alterations and new proposals, reappeared in his Third,198
196
See, eg Jean dAspremont Lynden, Les travaux de la commission du droit international
relatifs aux actes unilatraux des tats (2005) 109 Revue Gnrale de Droit International Public
163, 183; Degan, Unilateral Act, n 70 above, 18788; Fiedler, Einseitige Versprechen, n 86
above, 59; Goodman, Acta Sunt Servanda?, n 67 above, 58; Sicault, Engagements unilatraux, n 70 above, 66668; Skubiszewski, Unilateral Acts, n 11 above, 230, para 46 and Suy,
Actes unilatraux, n 89 above, 149. See also the replies to question 1 in the questionnaire circulated by the ILC that were given by Argentina, Israel and Sweden (Vienna Convention as
a useful guide) and the statement made by the Polish representative in the Sixth Committee
who stressed that Since the consent to be bound by a treaty and the consent of a unilateral
commitment were both expressions of the will of State, it seemed logical that the same reasons for invalidity should apply, as cited by Cedeo, Third Report, n 97 above, 19, para 135.
197
See Draft Article 7, Cedeo, Second Report, Addendum, n 192 above, 9, para 109.
198
See Draft Article 5, Cedeo, Third Report, n 97 above, 22, para 167.

238 The Law on Promises

Fifth199 and then, finally, Ninth Reports,200 where through the omission of
some new propositions, and the inclusion of some further refinements,
which included the distinction between void and only voidable unilateral
undertakings, it came to closely mirror the Vienna Conventions Articles
48 to 53. Grouped together under one over-arching principle with six paragraphs, the Ninth Reports Guiding Principle 7 hence read as follows:
Principle 7 Invalidity of unilateral acts
1. (a)A State that is the author of a unilateral act may not invoke error as
grounds for declaring the act invalid, unless the act was formulated on the
basis of an error of fact or a situation that was assumed by the State to exist at
the time when the act was formulated and that fact or that situation formed
an essential basis of its consent to be bound by the unilateral act.
(b) The foregoing shall not apply if the author State contributed by its own
conduct to the error or if the circumstances were such as to put that State on
notice of the possibility of such an error.
2. Fraud may be invoked as grounds for declaring a unilateral act invalid if
the author State was induced to formulate the act by the fraudulent conduct of
another State.
3. Corruption of the representative of the State may be invoked as grounds for
declaring a unilateral act invalid if the act was formulated owing to the corruption of the person formulating it.
4. Coercion of the person who formulated a unilateral act may be invoked as
grounds for declaring its invalidity if that person formulated it as a result of
acts or threats directed against him or her.201
5. Any unilateral act formulated as a result of the threat or use of force in violation of the principles of international law embodied in the Charter of the United
Nations is invalid.

The last paragraph (6, here omitted) dealt with an act infringing ius
cogens and thereby with the only subject matter which was subsequently
included within the ILCs Guiding Principles, as Principle 8. To put in
writing that acts infringing ius cogens are void was quickly accepted as
posing no obstacle within the Commission; however, unlike paragraphs 1
to 5 of the proposed article, it does not relate to the question what impact
a states defective will might have, as it prohibits a legal effect actually
willed by a state, but which infringes a norm of paramount international
law. None of the above cited provisions listing the grounds that might
199
See Draft Article 5(a)(h), Cedeo, Fifth Report, Addendum, n 134 above, 9, para
119 as corrected by Victor Rodriguez Cedeo, Fifth Report on Unilateral Acts of States,
Corrigendum, UN Doc A/CN.4/525/Add.1/Corr.2 (2002).
200
Guiding Principle 7, Cedeo, Ninth Report, n 135 above, 23.
201
Here the proposed principle was wider than Vienna Convention, Art 51, as it would
allow a state to decide whether or not it intends to invalidate an act formulated by its
coerced representative.

Further Requirements 239

vitiate a states will were, however, adopted in the ILCs final Principles.
The lacunae in this regard are regrettable, as there are good reasons and
indeed broad support for the application of the above-depicted grounds
of invalidity. But the ILCs decision to refrain from adopting a Guiding
Principle on this aspect should not be over-emphasised, as it was probably rather the absence of legal dicta from the ICJ, coupled with a lack
of time to find a consensus on the newly refined principle, rather than
fundamental opposition to the grounds of invalidity proposed, which led
the final Working Group not to dwell further on the matter. Although the
above-cited principle therefore does not have the Commissions blessing
and has not been shaped with complete precision by a drafting committee, it nevertheless comes very close to an adequate solution in this area
by mirroring the Vienna Convention.
CPublicity
The publicity of a unilateral act is regularly referred to as a special
requirement for it to be legally valid, and promises would hence have to
be made publicly in order to oblige their declarant. For Suy, for example,
and writing before the Nuclear Tests cases, promises, besides being subject
to the general requirements, had to fulfil two criteria: they had to manifest
a will to be legally bound, and they had to be made public: la publicit
de la promesse was hence a condition for it to be legally binding.202 For
the ICJ in the Nuclear Tests cases, similarly: [a]n undertaking of this kind,
if given publicly, and with an intent to be bound, even though not made
within the context of international negotiations, is binding.203
Whether a declaration necessarily has to be made in a manner observable by or in a place accessible to the public (as the Merriam-Webster Online
Dictionary204 defines publicly), in other words openly, is, however, more
than doubtful. Why should a declaration made by one state representative to another behind closed doors fail to become binding, where made
with the requisite intent? The number of those who have or could have
witnessed the act can hardly be decisive in this regard; just as treaties may
be concluded in closed sessions, there is no reason not to allow a promise
which has been made in a similar manner to become legally binding, or
even one made by written yet non-public notification. For the manifestation of will to become binding, the requirement should only be that the
declarant has communicated its statement to the addressee, and nothing more.
But returning to Suy, this, surprisingly, is (approximately) what he meant
202
203
204

Suy, Actes unilatraux, n 89 above, 149.

ICJ, Nuclear Tests (Australia v France) (1974), n 6 above, 267, para 43, emphasis added.
See www.merriam-webster.com/dictionary/publicly.

240 The Law on Promises

when speaking of publicity, which he explains as requiring the declaration to be brought to the knowledge of at least those states vis--vis whom
the engagement is undertaken.205 Whether the legal framework should
concentrate on the actual knowledge of the addressee might (and will in
a moment) be questioned, but it is important to begin by confirming that,
at least for Suy, no open publication of the declaration was necessary for
it to have been made publicly. As already seen in chapter two, the Swiss
Federal Department of Foreign Affairs equally saw no problem in finding
a statement made directly to another state and not before the eyes of the
general public to meet the requirements imposed by the term publicly,
when it wrote: Concerning the promises publicity, it is necessary that
it is made known to those to which it is of direct interest, be it through a
diplomatic note or in the event of a discourse.206
While this assessment, which confirms the binding force of a non-open
declaration communicated to its addressee, is to be approved, it is rather
odd to refer to the declarations publicity in such cases. Not surprisingly,
the question which term to use in this context surfaced within the ILC, and
it did so as early as in the discussions on the Special Rapporteurs First
Report, in which Cedeo had referred to declarations publicly issued as
falling within the scope of his study.207 For Ian Brownlie:
the criterion of publicity . . . was certainly relevant in terms of evidence and of
the identification of those to whom the act was addressed. It was not, however,
a necessary condition for the act to produce legal effects. Many declarations, for
instance, between ministers of foreign affairs, were made in camera, but were
nonetheless binding on their authors.208

Cedeo nevertheless retained the word publicly in his Second Report,

but described it as Suy and the Swiss Federal Department have done, to
be fulfilled where the addressee was made aware of the declaration.209
The Working Group convened that year again voiced its doubts on the
use of the word and in its recommendation spoke instead of a statement
205
Suy, Actes unilatraux, n 89 above, 150: La promesse devra en outre tre porte la
connaissance des Etats tiers ou, en tout cas, de ceux des sujets de droit envers lesquels elle
comporte un engagement et qui sont intresss directs: il faut que ceux-ci en aient connaissance. La promesse est donc un acte unilatral soumis rception.
206
See p 147.
207
See the definition provided by Cedeo, First Report, n 12 above, para 170 and cited in
ch one, n 3.
208
ILC, 2527th Meeting, n 71 above, para 15. See also Lukashuk, ILC, 2593rd meeting,
n 63 above, para 49; Al-Khasawneh, ILC, Summary Record of the 2696th Meeting, UN Doc
A/CN.4/SR.2696 (2001) para 9; Simma, ILC, Summary Record of the 2594th Meeting, UN
Doc A/CN.4/SR.2594 (1999) para 60. Also Pellet, stressing that it was clear that publicity
was required only vis--vis the addressee, ILC, 2594th Meeting, para 20. See in addition the
criticism by Tomuschat, Unilateral Acts, n 43 above, 150001.
209
See Victor Rodriguez Cedeo, Second Report on Unilateral Acts of States, UN Doc A/
CN.4/500 (1999) Draft Article 2, para 38 and Cedeos annotations thereto provided ibid
para 55.

Further Requirements 241

which is notified or otherwise made known to the State or organisation

concerned.210 Cedeo in his Third Report then altered his course, but he
went a step further by not focusing on the communication to the addressee,
as had been proposed (ie notified or otherwise made known) but found
a declaration to be a unilateral act, where it is known to the addressee.211
The Commission was not satisfied212 with this new approach, and rightly
so, as shifting the focus to the factual knowledge of the addressee in order
to hold a promise (just as a waiver or an act of recognition) to be binding
would only introduce a factor of uncertainty into the legal framework,
as the moment of factual knowledge is usually not known to the declarant; the same would consequently be true for the moment at which the
legal obligation arises. Focusing on actual knowledge would, in addition,
include declarations which had not been actively communicated to the
addressee by the declarant, but had become known by other means, even
by espionage as a Commission member noted, whereas it was necessary
that the state which was the author of the act took some steps to actually
make it known to the addressee.213 Once these steps have been taken, ie
the statement has been communicated, the declaration will have its effect.
Despite this criticism, Cedeos unaltered proposal was forwarded to the
drafting commission and left for the latter to deal with,214 which, however,
apparently did not produce any result. In Cedeos Fifth Report in 2002,
the criticised term (is known) appeared again.215 The Working Group
of 2003 remained unconvinced, as it chose to include no reference to a
required communication of the act in its recommendation, which referred
merely to a statement expressing the will or consent by which that State
purports to create legal obligations.216 Cedeo, in his final Report, in a
similar vein spoke simply of a unilateral declaration formulated by a
State with the intent of producing certain legal effects under international
law.217 That a declaration will nevertheless have to be communicated to
the addressee was, of course, clear, yet strictly speaking there is no necessity to expressly stipulate an extra requirement in this regard, as without
such action by the declarant, it will not actually have manifested its will to
be legally bound on the international plane.218 A provision clarifying this
ILC, Report of the Working Group, UN Doc A/CN.4/L.588 (1999) para 11.
See Cedeo, Third Report, n 97 above, para 80.
212
See ILC, Report of its Fifty-second Session, A/55/10, n 194 above, 94, para 558 et seq.
213
Ibid para 559.
214
See Cedeo, Fifth Report, n 134 above, 12, para 50 and 18, para 81.
215
Ibid 18, para 81.
216
ILC, Report of the Chairman of the Working Group, UN Doc A/CN.4/L.646 (2003) 2,
para 6, Recommendation 1.
217
Cedeo, Ninth Report, n 135 above, 1, Guiding Principle 1.
218
See also dAspremont Lynden, Lest travaux, n 196 above, 17879, for whom: Epurer
la dfinition de lacte unilatral de toute rfrence a sa publicit ou la connaissance quen a le
destinataire tait donc raisonnable, at 179.
210
211

242 The Law on Promises

requirement would, on the other hand, do no harm, especially as the
common use of the word publicly remains misleading. The ILC, however, in the end did the exact opposite. Casting aside the criticism that
had been voiced, and ignoring the evolution which had taken place in
the terminology, it simply went back to the very beginning and, focusing
solely on the ICJs dicta, again referred to declarations publicly made as
capable of creating legal obligations in Guiding Principle 1.
The ICJ in the Nuclear Tests cases was, of course, faced with public statements and where a statement is publicly made, it will meet the requirement of having been communicated to the addressee(s); but the Court did
not say that only public statements manifesting a will to be bound would
be binding. However, as even those authors for whom publicity is a
requirement in the end do not interpret it as necessitating a truly public
declaration, but consider it to be fulfilled where a statement has been communicated to the addressee, there appears to be no real controversy in this
regard. Appropriate wording would thus certainly have strengthened the
Guiding Principles guiding character, but even with the rather curious
requirement219 of publicity included, it appears to be generally understood and accepted that a state will only need to communicate its statement to its addressee(s) for it to become binding vis--vis those thereby
addressed, be it in the open and truly publicly, behind closed doors or by
a written and direct (ie non-open) notification. As Alain Pellet has said,
with publicity, the matter was hence not one of substance, but merely
one of terminology and there was hardly any need to broadcast a unilateral act if it was intended only for the other State.220
Tomuschat, Unilateral Acts, n 43 above, 1500.
Pellet, ILC, 2594th Meeting, n 208 above, para 20; also Brownlie, It would be regrettable to use a term such as public, as had been proposed, because that would unjustifiably
restrict the scope of the study. The fact that a declaration was public could be sufficient,
but as practice currently stood, it was certainly not necessary, ILC, Summary Record of
the 2543rd Meeting, UN Doc A/CN.4/SR.2543 (1998) para 48. See in addition Fiedler,
Unilateral Acts, n 89 above, 1022, according to whom there is virtually complete agreement that an expression of intent by the declaring State alone is not sufficient, but that the
declaration must be brought to the notice of the subject of international law concerned. If a
rule of customary law had formed at all on the basis of juridical and State practice, then it
merely requires appropriate disclosure to those concerned; it is sufficient that the recipient State has
the possibility of taking notice, emphasis added. See also Carbone, Promise in International
Law, n 86 above, 170, who considers that a promise has to be duly communicated by the
promisor; Charles Rousseau, Droit international public, vol I, Introduction et Sources (Paris,
1970) 424, finds the reception par la bnficiaire to be necessary; in this sense also Verdross
and Simma, Universelles Vlkerrecht, n 89 above, 427, para 665, for whom the declaration
was empfangsbedrftig. Since focusing on the actual knowledge of the addressee in a definition is problematic as explained above, definitions which refer to a declaration having to
be brought to the knowledge of or made known to the addressee (besides Suy, see, eg
Charpentier, Engagements unilatraux, n 73 above, 372: officiellement porte pas forcement par une notification la connaissance de son destinataire, or Sicault, Engagements
unilatraux, n 70 above, 671: il faut que lengagement unilatral soit port a la connaissance dau moins un autre sujet de droit international) might be misunderstood, unless
they emphasise that once communicated the declaration will be considered known by the
219
220

Further Requirements 243

D A Lawful (and Possible?) Object

Two more requirements often appear in articles addressing the conditions
for the validity of unilateral acts, both dealing with the object of the obligation undertaken: it has to be lawful as well as materially possible in
order for the declaration to be valid.221
In terms of the undertakings lawfulness, Cedeo made numerous
remarks in his reports,222 without however proposing a draft article. In
illuminating the lawfulness of its object and purpose referred to, Cedeo
distinguished between two scenarios: in the first, a unilateral act was in
conflict with a peremptory norm, ie ius cogens; in the second, it was incompatible merely with another, previously undertaken obligation, be it by an
earlier unilateral or conventional act.223
How the first scenario should be legally treated appeared, and remains,
fairly obvious in light of the broad consensus according to which any
undertaking, conventional or unilateral, which infringes a peremptory
norm must be void ab inito. Although, as seen, the ILC has confined itself
nearly exclusively to the dicta of the ICJ, which has not covered this
aspect, the position was generally accepted and, on account of it being
uncontroversial, even made it into the Guiding Principles.224
With respect to the second scenario, in which the unilateral undertaking
is incompatible with another undertaking previously adopted by the
declarant, the situation is not as straightforward. For Cedeo, partly relying on Skubiszekwsi, this case too was apparently to be treated as invalidating the undertaking, which infringed the earlier obligation.225 Yet, as
Sicault has argued,226 there is little reason not to treat the situation where
addressee and legal doctrine will not, as the debate in the ILC has underscored, attempt
to assess the moment of factual knowledge. Instead, the possibility of taking notice (as
Fiedler writes) of a declaration communicated by the declarant, will suffice.
221
See, eg Goodmann, Acta Sunt Servanda?, n 67 above, 5455 who, relying on Dionisio
Anzilotti, Cours de droit international (translation from the Italian by G Gidel, Paris, 1929),
writes: The second condition is an appropriate object of the unilateral act, for no manifestation of will can make good an object or purpose that is materially impossible or forbidden
by law. See also Krzysztof Skubiszewski, Les actes unilatraux des tats in Mohammed
Bedjaoui (ed), Droit international: bilan et perspectives (Paris, 1991) vol 1, 231, 242, para 44;
Fiedler, Einseitige Versprechen, n 86 above, 59: rechtliche und tatschliche Mglichkeit der
Erfllung; Suy, Actes unilatraux, n 89 above, 149 speaking of an objet appropri. On the
required legality of a unilateral acts object see also J-P Jacqu, Elments pour une thorie de
lacte juridique en droit international public (Paris, 1972) 16162. See also the analysis provided
by Sicault, Engagements unilatraux, n 70 above, 66164.
222
See, eg Cedeo, Second Report, n 209 above, para 53; Cedeo, Second Report,
Addendum, n 192 above, paras 11416; Cedeo, Fifth Report, n 134 above, paras 9495.
223
Cedeo, Fifth Report, n 134 above, para 94.
224
See Guiding Principle 8.
225
See Cedeo, Second Report, Addendum, n 192 above, para 116; Skubiszewski, Actes
unilatraux, n 221 above, 242, para 44.
226
See Sicault, Engagements unilatraux, n 70 above, 162, who rejects the position
adopted by Jacqu, Elments, n 221 above, 16162.

244 The Law on Promises

the fulfilment of a new undertaking is incompatible with an earlier one
in the same manner as in the law of treaties.227 There, no special treatment
for such undertakings is included and the new obligation is not declared
invalid; a state, by undertaking an incompatible new obligation, will
hence have created a valid obligation. The fulfilment of the latter (or the
former), however, will either directly infringe the other undertaking (eg
where the earlier obligation was not to do something that is now done) or
make it impossible for it to be fulfilled (eg where an action was promised
that can be performed only once, like delivering a specific object). The
resulting breach will, in both cases, therefore invoke the states responsibility vis--vis the state to whom the obligation breached was owed. Were
the new obligation instead automatically invalid, as in the ius cogens scenario, the non-fulfilment could not trigger the declaring states responsibility, as an invalid obligation cannot be breached. In the ius cogens setting
this result is indeed intended, as no state may rely on undertakings which
infringe a peremptory norm, and any damage resulting from it choosing to do so will not be compensated. But where an earlier obligation is
simply infringed, the scenario is (usually) very different, and treating a
subsequent obligation as automatically illegal and invalid would leave
the state vis--vis which that new obligation was undertaken unprotected and unable to claim compensation. The better view, therefore, is
to assume a parallel approach to that found in the Vienna Convention. A
new obligation, merely contradicting an old one, will hence (usually)
not be invalid.
As the term usually implies, in this authors opinion there should (ie
de lege ferenda) be an exception to the rule. The question must be raised
whether there are other obligations which, once undertaken, also necessarily require any subsequent contradictory obligations to be invalidated,
since the threatened breach cannot simply be left to be compensated by
the principles of state responsibility. Obligations of fundamental import
ance, where generally known to have been previously undertaken (eg
where the treaty has been registered), should for this reason, even if not
flowing from a peremptory norm, also render subsequent undertakings
void. This might be true for obligations undertaken in order to protect
individual human rights, and also for other undertakings which establish mutual obligations in sensitive security areas, such as for example the
undertakings included in the Nuclear Non-Proliferation Treaty. Here, any
obligation to compensate the damage caused by a subsequent breach will
necessarily fall dramatically short of having a fully compensatory function for those directly involved. Therefore, where a state makes a promise which necessarily means that it will (and hence it announces that it is
going to) breach such an obligation, this promise should be considered
227

See also dAspremont Lynden, Lest travaux, n 196 above, 18587.

Further Requirements 245

void. That a grain of instability would thereby be added to the system

(already present via the ius cogens principle, the precise content of which
remains difficult to ascertain) is true, as obligations having such a special status would have to be determined. But to determine the indeterminate is a task which lawyers have to perform every day. In the end, an
addressee should not be able to successfully claim either the fulfilment of,
or compensation for the breach of, an obligation which is in flagrant violation of eg a previously assumed human rights obligation of the declarant.
Reliance on such undertakings should not be fostered by the international
legal system.
Besides having to be legal it is sometimes stated that unilateral undertakings also require a materially possible object in order to be valid.
If mentioned, this aspect is presented as a no brainer, requiring little
argumentative backing.228 Yet, it may be doubted whether it really is
a logical requirement. Looking at the law of treaties, the latter does not
declare undertakings which are impossible to automatically evaporate.
Instead, Vienna Convention, Article 61229 identifies the problem as falling
within its Part V, Section 3 and pertaining to the termination and suspension of the operation of treaties. There, the supervening impossibility of
performance will in the case of a permanent disappearance or destruction of an object indispensable for the execution of the treaty, allow it to
be terminated. The treaty is, hence, not automatically invalid. In addition,
since the Article speaks of disappearance or destruction, a state which
undertakes what is from the outset impossible will not be able to rely on
paragraph 1. Under Vienna Convention, Article 61(2), even supervening
impossibility of performance will not always allow the state to free itself
from its obligations, as it may not do so where the impossibility is the
result of a breach by that party, either of an obligation under the treaty,
or of any other international obligation owed to any other party to the
treaty. Unilateral undertakings that are from the outset impossible, yet
promised by a state, could therefore be treated similarly and not automatically allow a state to terminate its undertaking. That the obligation
cannot be fulfilled is clear, yet the existing obligation now breached (not
228
See, eg Sicault, Engagements unilatraux, n 70 above, 661: Il faut donc tout dabord
que la ralisation de ces droits et obligations ne soit pas matriellement impossible, car,
si tel tait le cas, lengagement unilatral serait dpourvu de validit juridique. See also
Goodman, Acta Sunt Servanda?, as cited in n 221 above.
229
Article 61 reads as follows: Supervening impossibility of performance: (1) A party
may invoke the impossibility of performing a treaty as a ground for terminating or withdrawing from it if the impossibility results from the permanent disappearance or destruction
of an object indispensable for the execution of the treaty. If the impossibility is temporary, it
may be invoked only as ground for suspending the operation of the treaty. (2) Impossibility
of performance may not be invoked by a party as a ground for terminating, withdrawing
from or suspending the operation of a treaty if the impossibility is the result of a breach by
that party either of an obligation under the treaty or of any other international obligation
owed to any other party to the treaty.

246 The Law on Promises

terminated) will allow the beneficiary to claim reparations. In the case of
supervening impossibility, a state could, however, be allowed to free itself,
unless the impossibility is brought about by the promising state itself.230
In order to allow for such nuanced differentiations, the question should
not, in any event, be one of automatic invalidity but rather one connected
to the possibility of terminating the obligation undertaken. As such, it
could be held to be encompassed by the rule allowing a state to terminate
or revoke its undertaking on account of a fundamental change of circumstances, as already accepted for unilateral undertakings,231 even though
the Vienna Convention included its own specific article in this respect.
ERegistration?
The registration of treaties is a duty under UN Charter, Article 102 for all
UN members. According to the latter:
Every treaty and every international agreement entered into by any Member of
the United Nations after the present Charter comes into force shall as soon as
possible be registered with the Secretariat and published by it.

But the penalty for its breach is not the voidability or even nullity of the
agreement, which is a concept that was abandoned after the League of
Nations demise.232 Under its paragraph 2, a non-registered treaty or international agreement instead may not be invoked by a party to it before
any organ of the United Nations, which specifically includes the ICJ as
one of the UNs principal organs.233 Despite the fact that this sanction has
apparently not been applied strictly in the realm of treaties,234 the question, as to whether unilateral declarations shall also be registered under
UN Charter, Article 102(1) may arise. As this duty exists for every treaty
and international agreement entered into by any member of the United
Nations, the answer should be a straightforward no, as unilateral assurances simply do not constitute a treaty or agreement. Attention has,
however, been drawn to the fact that the drafters of the Charter apparently considered unilateral engagements of an international character
which have been accepted by the State in whose favour such an engage-

230
Jacqu, Promesse unilatrale, n 97 above, 343 also contemplates an analogue application of Art 61.
231
For more on revocation see p 251 et seq.
232
Article 18 of the Covenant read: Every treaty and international engagement entered
into hereafter by any Member of the League shall be forthwith registered with the Secretariat
and shall as soon as possible be published by it. No such treaty or international engagement
shall be binding until so registered.
233
See UN Charter, Art 7(1).
234
Klabbers, Concept of Treaty, n 106 above, 84.

Further Requirements 247

ment has been entered into,235 as covered by this Article.236 This is rather
surprising not only on account of its wording as such, but all the more
so in light of the Articles predecessor, Article 18 of the Covenant of the
League of Nations, which had spoken of every treaty and every international engagement entered into. While the latter would have included
promises as unilateral international engagements, the word agreement
now introduced makes any such subsumption more than difficult and
strongly indicates the conclusion that under UN Charter, Article 102,
no such obligation can exist. Although Klabbers may be right when, in
briefly addressing the matter, he considers that agreement was a generic
term intended to embrace all those agreements that would possibly fall
through the cracks if reference could only be made to treaties,237 a promise simply is not an agreement, and it need not be accepted by the State
in whose favour it was made, as the drafters envisaged.238 State practice,
too, underscores this finding, as the registration of unilateral assurances
is a rare event, with the Egyptian declaration being the one common
example in this area.239 The latter shows that the registration of a unilateral
declaration is possible and likely to be accepted by the United Nations; it
is even desirable under the spirit of Article 102, as it will make commitments undertaken by states public. Any such action will, as mentioned, in
addition be indicative and helpful in assessing a declarants actual intention to be legally bound.240 However, especially in light of the Articles
clear wording underscored by a lack of state practice or judicial decisions in this regard, a unilateral commitment does not presently need to
be registered. In other words, not doing so is not punishable under UN
Charter, Article 102(2).
F Impact of the Addressees Reaction
As will be recalled, the ILC in its Guiding Principle 1 considers that
[d]eclarations publicly made and manifesting the will to be bound may
have the effect of creating legal obligations. This use of may has been
235
UNCIO Docs, vol 13, Commission IV, 705, as cited by Michael Brandon, Analysis of
the Term Treaty and International Agreement for Purposes of Registration under Article 102
of the United Nations Charter (1953) 47 American J International Law 49, 53.
236
See Klabbers, Concept of Treaty, n 106 above, 82.
237
Ibid 82.
238
See in this respect Brandon, The Term Treaty and International Agreement in Art.
102 UNC, n 235 above, especially 5354. See also the criticism voiced by Dehaussy and
Mensbruegghe in respect of the Secretary Generals position that the term agreement might
be understood to cover unilateral engagements and also unilateral acts such as the Egyptian
declaration, as cited in ch two, n 108.
239
Apart from declarations under the Optional Clause which have to be registered under
ICJ Statute, Art 36(5).
240
See p 226.

248 The Law on Promises

criticised above,241 especially on account of the fact that the Commission
has provided no guidance as to the additional factors which are necessary
for a promise to be binding, besides that of a states publicly manifested
will to be bound. The ILCs finding, so it was said above, is especially dubious as it claims to be very directly inspired by the dicta in the Judgments
handed down by the International Court of Justice on 20 December 1974
in the Nuclear Tests cases.242 Yet, for the ICJ, statements made publicly
and manifesting the will to be bound actually are binding.243 Although the
Commission chose not to offer any explanation in this regard, there are
in fact conceivable scenarios in which the ILCs finding will be correct, ie
where a state makes a public statement manifesting an intent to be bound,
but without the statement having the desired effect. The first scenario in
which this will be the case (and here there will presumably be little opposition from the ICJ) is where the promised undertaking is incompatible
with ius cogens, as already addressed above. Strictly speaking, the ILCs
may is hence correct; such scenarios will, however, not only be rare but
the Commentary provides no indication that it is only these cases which
were meant. Statements made by an incompetent representative strictly
speaking do not reveal a states intent and the same can be said where the
statement is vitiated by error, fraud or other means. Guiding Principle 3
might, however, shed some light on an additional setting, where a declaration is considered publicly made and manifesting a states will to be
bound, but is still not binding upon the declarant, and which falls into the
subject matter of this chapter: the reactions of the addressee(s). It will be
recalled that Guiding Principle 3 elaborates that:
To determine the legal effects of such declarations, it is necessary to take account
of their content, of all the factual circumstances in which they were made, and of
the reactions to which they gave rise.244

The Commentary refers directly to the Nuclear Tests cases, the Frontier
Dispute case, as well as the Armed Activities on the Territory of the Congo
and the Military and Paramilitary Activities in and against Nicaragua cases
in order to support this proposition. And indeed, as elaborated above,245
the declarations content along with the circumstances of its making will
play an important part in analysing whether the declarant actually manifested an intent to be legally bound. The reactions of the addressees have,
however, not been mentioned by the ICJ in this regard and, of course,
they can hardly be referred to in determining whether the declaration has
See p 188.
ILC, Draft Articles on the Law of Treaties with Commentaries, n 123 above, Commentary
to Guiding Principle 1, para 1, first sentence.
243
See p 188.
244
Guiding Principle 3, emphasis added.
245
See p 211 et seq.
241
242

Further Requirements 249

manifested an intention to be bound or not. The addressee cannot turn a

vague and ambiguous statement into one considered as clearly manifesting a legal will, nor vice versa. The reactions will only reflect how the
addressees understood and interpreted the statement and reveal whether
or not the statement was considered welcome or not. The last two cases
before the ICJ referred to above confirm this. In both the Frontier Dispute
and the Armed Activities on the Territory of the Congo cases, the addressees claimed that the statements made manifested an intent to be legally
bound, but the Court opined otherwise. This finding is all the more true
if the Commentarys reference to reactions of other States concerned is
not meant to be restricted to the beneficiaries of a statement addressed
thereby. The Principle is drafted very broadly, as it directs attention simply to the reactions to which they [the declarations] gave rise. A state
which has not been addressed may of course provide its opinion on a
statement, but its position cannot have any impact on the undertakings
legal effect. For the addressee, the situation is, however, slightly different.
While it cannot influence whether or not a state has manifested an intent
to be bound, its behaviour may have an impact where such an intention
has been found to exist.
The ILC has differentiated in the commentary between states taking
cognisance of commitments undertaken, and states challenging or rejecting them, but there is, of course, another possibility, that of mere silence.
Unfortunately, the Commentary only tells us that the reactions of states
concerned are important, but does not indicate in which way.246 The
answer to this question has, in part, already been provided when the basis
of the undertakings legal bindingness was elaborated, but since the ILC
mentioned the reactions (of the addressee) in the same terms as the content
and the circumstances in determining a statements legal effect, there shall
be no room for doubt left in this respect: To begin with, the ICJ has been
very clear in stating that a positive reaction of the addressee is in no way
required in order for the declaration to become binding, as nothing in the
nature of a quid pro quo nor any subsequent acceptance of the declaration,
nor even any reply or reaction from other States is required for the declaration to
take effect.247 In other words, if the addressee remains silent and inactive, the
statement, if manifesting an intent to be legally bound, will remain binding.
Similarly, where the addressee chooses to accept the declaration, this reaction will have nothing to add to the already binding nature of the unilateral undertaking. The acceptor cannot turn the promise into an offer and,
thereby, the undertaking into a treaty commitment, although the addressee,
as stated, may of course welcome a statement and (non-technically speaking) accept it as settling a dispute or the like. A positive reaction by the
246
247

See p 189 et seq.

ICJ Nuclear Tests (Australia v France) (1974), n 6 above, 267, para 43, emphasis added.

250 The Law on Promises

addressee going beyond a mere statement and consisting of reasonable
actions made in direct reliance on the statement, might furthermore (as is
addressed in more detail below) have an impact under the estoppel doctrine and impose additional restraints on a promises revocability.248 The
third possible reaction, that of rejecting an undertaking, will, however, have
a direct impact on the bindingness of the statement made which in no way
alters the finding of the ICJ that a reaction is not required for the statement
to be binding. Since the bindingness of a unilateral undertaking has been
grounded on the principle of good faith and, more precisely, the legitimate
expectations raised in the addressee that the declarant will adhere to the
word given, the basis for the acts compelling nature has been described to
necessarily collapse where the addressee displays that no such expectations
exist. In other words, and as elaborated in more detail above, the presumption of trust, which explains how even with no reaction from the addressee
a declaration may become binding, as the addressee will usually take it into
account and rely upon it, is rebutted where a declaration is rejected by its
addressee.249 The situation thereby becomes comparable to one where the
state has only promised itself to do something. Should it change its mind, it
will be allowed to act accordingly, in the absence of any reliance on the declaration being worthy of protection and requiring adherence to the promise
given. Where a declaration is rejected, there can be no legal constraints on
the statements revocation on a good faith basis.250
In summary: the addressees non-reaction has no impact; its affirmative
action, unless going as far as triggering an estoppel effect, adds nothing to
the declarations binding nature; where its undertaking is rejected, however, the declaring state is not compelled to uphold what it has declared.
G Summarising the Requirements for a Legally Binding Promise
The requirements to which a promise is therefore subject have been found
to be the following: a declaration has to be a unilateral manifestation of
will to be legally bound, made by a states competent representative and
communicated to the addressee. It should not, as such, be subject to any
defects, although amongst the latter not all will immediately result in the
declaration being void, some only give rise to its voidability and leave the
final decision to the state which has made the declaration. The undertaking must not, in addition, infringe ius cogens (and in this writers opinion
any other obligation, known to have been undertaken by a state, which
248
For more on an undertakings revocability see p 251 et seq; on promises and estoppel see
p 277 et seq.
249
See p 201 et seq, particularly pp 20407.
250
For the peculiarities which allowed the ICJ in the Nuclear Tests cases to hold otherwise,
see p 127 et seq.

The Revocability of Promises 251

is of fundamental importance and as such should be interpreted as not

tolerating the existence of a subsequent and legally valid contrary obligation) and it must not have been rejected by the addressee.
VI THE REVOCABILITY OF PROMISES

Once a legally binding undertaking has been identified, both the declarant and the addressee are likely to raise what is a core question to be
answered in the law on promises, which is whether such undertakings
may be revoked and, if so, subject to what kind of restrictions.
A Necessary Limitations
Despite its being a legal question of pivotal importance, there is a lack
of state practice and jurisprudence in this regard and scholars have not
come to a consensus. Legal commentators, instead, offer the full range of
possible answers to the question posed, including that of promises being
freely revocable.251 Skubiszewski, in a footnote to his finding that as a
rule, the State can modify or revoke its unilateral act at will and at any
time,252 at least unless general international law or treaty law imposed
some barriers, in this context emphasises that:
Revocability and modification should not, of course, be confused with the binding force of the unilateral act. [Internal cross-reference omitted]. The State may
unilaterally give a legal undertaking without depriving itself of the right to terminate or modify it for some future time. The position has some analogy to that
of the State which is bound by a treaty but retains the right to denounce it or to
withdraw from it.253

And surely, revocability and modification should not be confused with

a unilateral acts binding force, as an undertaking can be considered
binding, without thereby having to be irrevocable and unmodifiable for
all future time. But, although not to be confused, it is, on the other hand,
not to be ignored that the questions of a unilateral undertakings bindingness and its revocability are intimately related and interdependent. As
has been said when discussing the basis of a states obligation assumed
through a promise, once a state is considered able to revoke an under
taking as it pleases and without external constraints, there will be nothing
left of the acts bindingness.254 The above-cited finding of a states power
to revoke an undertaking at will and at any time does therefore, and
251
As will be seen below, this position was also advocated within the ILC; see also Wilhelm
Wengler, Vlkerrecht (Berlin, 1964) vol I, 304 and especially 308.
252
Skubiszewski, Unilateral Acts, n 11 above, 234, para 64.
253
Ibid 238, note 50.
254
See p 194 et seq.

252 The Law on Promises

despite the footnote arguing to the contrary, touch directly upon a unilateral acts bindingness. In a legal framework in which a state may terminate its undertaking at will and at any time, it could at all times freely
decide whether or not it presently has to adhere to the promise made; it
would therefore not be bound by what it has once promised, nor could
its addressee trust that the behaviour promised would actually be performed. Revocation and bindingness therefore cannot be treated as two
completely separate issues, and some limitations on the former as a result
of the latter need to apply.
Skubiszewski intends to overcome this problem by taking recourse to
good faith, which could impose possible limits on a states power of revocation, since it speaks in favour of the maintenance of a states act which
did not display the authors intention of retaining its full freedom of action
and where another State relies upon the act in its relations with the acts
author.255 But while the protection of expectations raised in the addressee
has been identified above as the basis of the acts legal bindingness, the
assumption of free revocability altered by good faith is problematic. Where
the ground rule is one of unconstrained revocability at will and at any
time, the addressee may not legitimately (and where this rule is known
usually will not) put any faith on the unilateral declaration as creating a
legally binding obligation. It is, in this writers opinion, therefore contradictory to lay down unconstrained revocability as a ground rule, and then
modify the latter on the basis of whether or not the addressee chose to rely
upon the declaration made. The law must first stipulate whether or not
the addressee may rely on a statement as a legally binding undertaking, a
stipulation which, with free revocability, would have been answered in the
negative. To take an example from German civil law, the latter stipulates
that a contract in which immovable property is sold requires notarisation
in order to be valid. If this is the rule, the fact that one of the parties relied
on a merely written contract to be fulfilled, cannot alter it and the good
faith principle as enshrined in the German Civil Code will not change this
finding.256 The opposite approach has rightly been criticised by Rubin, who
in his article on the Nuclear Tests cases stressed:
If the international community were not misled by the unilateral declaration
and did not conceive it as creating a direct legal obligation, no significant question of good faith would seem to arise. The sole legal question would seem to
be a mere technical one of whether an express revocation should be required
prior to a states acting in a manner inconsistent with its unilateral declaration,
or whether the inconsistent action implying revocation should be conceived to
be sufficient to satisfy the legal requirements, if any, flowing from the general
obligation of good faith.257
255
256
257

Skubiszewski, Unilateral Acts, n 11 above, para 66.

Extreme cases of inequity aside.
Rubin, Unilateral Declarations, n 24 above, 1011.

The Revocability of Promises 253

But once the legal bindingness of a promise is accepted as a starting

point, such a finding implies a necessary limitation on a states power of
revocation. An addressee may consequently rely on the declaration made
and expect such reliance to be protected under the legal principle of good
faith. In altering Rubins citation to reflect the present status of inter
national law if the international community is led to believe that a unilateral declaration creates a direct legal obligation (as is by now clearly
the case with the ICJ, the ILC and the majority within legal doctrine
emphasising the legal bindingness of promises in order to allow states to
be able to trust unilateral assurances) the result is that such expectations
are now protected under the legal principle of good faith.
A finding according to which a states declaration is binding, hence,
implies that its power of revocation must somehow be limited, as the expectations of its fulfilment are now protected under the principle of good faith.
The precise limitations on a states power of revocation, however, still
remain to be determined.
B Between the Vienna Convention and a More Flexible Approach
While most commentators have adopted this position and reject an undertakings unrestrained revocability, the answers differ as to how much trust
should be protected under the good faith principle. A number of scholars
have gone as far as to adopt the position at the very extreme opposite
of the spectrum, arguing that an assurance given by unilateral means is
irrevocable.258 But although asserting that promises cannot be taken back,
these authors seem to allow for exceptions and, in fact, to accept the position that unilateral undertakings can hardly be more binding than their
conventional counterparts.259 The limited grounds of termination which
are applicable in the law of treaties are, hence, usually accepted as being
applicable, mutatis mutandis, to promises where an assurance is classified
as irrevocable. Jacqu has expressly said so, and for Suy, who considers
that states may not go back on their legal undertaking, this consequence
is equally clear, as he wants to enlarge the understanding of the word
pactum in pacta sunt servanda to include promises.260 Although doing
258
See Suy, Actes unilatraux, n 89 above, 152, who, in adopting Paul de Visschers choice
of words as Agent for Honduras in the arbitration before the King of Spain (award rendered 23 December 1906) wrote: Les Etats doivent savoir que la vie internationale requiert
la scurit et que pour cette raison il ne leur est pas permis de revenir sur leurs propres
dclarations lorsque celle-ci ont dtermin un autre Etat leur accorder confiance et crdit;
Jacqu, Elments, n 221 above, 256: la promesse ne pourra tre retire partir du moment
o son destinataire en a eu connaissance; see also Venturini, Actes unilatraux, n 70 above,
42123, especially 422.
259
As emphasised by eg Fiedler, Einseitige Versprechen, n 86 above, 67 and Rubin,
Unilateral Declarations, n 24 above, 10.
260
Suy, Actes unilatraux, n 89 above, 4446 and Eric Suy, Unilateral Acts as a Source of

254 The Law on Promises

so has not found many followers in the literature on the subject, there
is considerable support for the thesis that the rules on the terminability of treaty obligations should be applied by analogy also to unilateral
assurances.261 Those authors who go into more detail in this respect consequently describe promises as being revocable in accordance with any
terms incorporated in the promise itself or where the addressee consents
thereto. Other than that, situations as envisaged by Vienna Convention,
Article 61 (supervening impossibility of performance)262 or Article 62 (fundamental change of circumstances) are said to allow for the undertakings
revocation.263
While the transferral of the rules of the Vienna Convention to promises
in this area of fundamental importance might be considered welcome by
the scholars referred to above, it should not be ignored that to do so is
presently simply a proposition. There is neither state practice to point to,
nor is this position backed by the ICJs dicta, which could again serve as
a guiding light in this area, at least unless rejected by a state consensus
to the opposite. Irrevocability is in addition by no means a necessary
or a logical result flowing from the application of the good faith principle. Identified as protecting the expectations placed on a promise as a
legally binding undertaking, it must necessarily impose some restrictions
on a promises revocability in order to do so, but it does not necessitate
applying the very same (and strict) rules to assurances which are applied
to treaties as flowing from the pacta sunt servanda principle.
Given the lack of state practice and strong dicta in this respect, the
revocability of promises is therefore an area in which a broad consensus
within the ILC would have been of great help. But it was apparently precisely on account of these two factors that such a consensus could not be
established. In turning to the Commission, it may first be observed that
International Law: Some New Thoughts and Frustrations in Droit du pouvoir, pouvoir du droit:
mlanges offerts Jean Salmon (Brussels, 2007) 639. Jacqu, in his article Promesse unilatrale,
n 97 above, 344 considered [l]es conditions de rvision et de rvocation de la promesse unilatrale sont fort proches de celles prvues dans le cadre du droit des traits pour la suspension ou lextinction des obligations nes dune convention internationale.
261
See Carbone, Promise in International Law, n 86 above, 172, for whom the ICJ in the
Nuclear Tests cases seems to point towards a rgime under which the discipline of revocation
of promise is similar to that in force on the denunciation of treaties and for whom [s]uch
a conclusion seems to be both well-balanced and worthy of acceptance. See also Francisco
Villagrn Kramer, Les actes uniltraux dans le cadre de la jurisprudence internationale
in International Law on the Eve of the Twenty First Century: Views from the International Law
Commission (New York, 1997) 157 and, as just cited in n 260 Jacqu, Promesse unilatrale,
n 97 above, 34344. Similarly, Sicault, Engagements unilatraux, n 70 above, 650 et seq, for
whom the grounds of revocation are presently rather feeble (mince, 654), but, apart from
revocation according to the terms included in a promise or by mutual consent, the rebus sic
stantibus principle applied; agreeing with Suy, also Charpentier, Engagements unilatraux,
n 73 above, 378.
262
See p 243 et seq.
263
See Jacqu, Sicault and Charpentier as cited n 261 above. Similarly, Goodman, Acta Sunt
Servanda?, n 67 above, 69, last two paragraphs.

The Revocability of Promises 255

the above depicted positions found in the literature on the topic all surfaced quickly and at an early stage of its work amongst its members, even
though the revocability of unilateral undertakings was not addressed in
more detail by the Special Rapporteur until his Ninth and final Report.
The revocability question was nevertheless raised especially within the
1998 debate on Cedeos First Report, and again in 2002 after some of the
Commission members had changed, and the ILC was presented with a
largely recapitulative overview of the topic in the Special Rapporteurs
Fifth Report. The relevant part of the discussion held in 1998 is summarised in the ILC Yearbook as follows:
A number of members made observations regarding the problem of the revocation of unilateral promises. Some commented that the ability of a State to revoke
a unilateral promise which it had made should depend, at least in part, upon its
intention when it performed that act. Thus, if it had intended that its promise be
revocable, then it should be susceptible of revocation, subject to whatever conditions or restrictions that State might have imposed upon itself in that regard
. . . Conversely if the State which had made the promise had intended that it be
irrevocable, then it should not, in principle, be subject to revocation.264

Turned into a legal principle and applied in practice, this position would
probably amount to emphasising the importance of special terms included
in a states declaration and addressing the undertakings revocability; as
held by many scholars these terms would then govern the future execution of the act. Yet, as states are usually reluctant to make promises which
immediately draw attention to the possibility of their future revocation,265
the more important question will be how to treat statements in which any
such indications are lacking. In this respect, and in continuing with the
summary provided in the ILC Yearbook, the following was said:
With regard to those cases in which it was not possible to identify any intention
on the part of the declarant State, one member expressed the view that, since
the legal relations created by a unilateral promise were not reciprocal in nature,
such a promise should be presumed to be revocable at will by the State which
had made it.266

Revocability at will was therefore proposed in the Commission, yet it

triggered a negative response along similar argumentative lines as presented above:
Several members, however, were opposed to this conclusion, stating that, were
this so, the binding nature of such acts under international law would be quite
illusory and the expectations of those States in whose favour they were made
264
ILC, Report on the Work of its Fiftieth Session, General Assembly, Official Records,
Fiftieth Session, Supplement No 10, A/53/10 (1998) 5758, para 185.
265
See p 264 et seq.
266
ILC, Report of its Fiftieth Session, A/53/10, n 264 above, 58, para 185.

256 The Law on Promises

would lack protection.267

ILC members, however, did not automatically jump to the conclusion

that promises should necessarily be considered irrevocable unless the
addressee consented, as:
On the other hand, it was observed that, were unilateral promises to be presumed to be of indefinite duration and not to be susceptible of revocation
without the consent of the State or States in whose favour they were made,
then States would be reluctant ever to make such promises. Alternatively, they
would find themselves compelled in certain circumstances to resile from them.
Accordingly, there should be some, albeit not unlimited, ability on the part of a
State which had made such a promise to revoke it.268

In the following, it was also remarked within the Commission that the
law of treaties might provide some guidance in this respect, yet attention
was drawn especially to the ICJs holding in the jurisdictional phase in
Military and Paramilitary Activities in and against Nicaragua, where the Court
pronounced in respect of a declaration made under the Optional Clause
(the judgment will be addressed in more detail below). It is interesting
to see that once the Commission had, to a certain extent, been reshuffled
in 2002, a very similar discussion arose. It, too, was articulated along the
lines first of free revocability, triggering a response in favour of irrevocability (ie treaty rule transferral), only to prompt a call that it was necessary to establish a middle ground. As the 2002 ILC Yearbook summarises:
There was also a discussion in the Commission about the termination of the
obligation created by a unilateral act. It was noted that in the case of a treaty
there was a procedure and an agreed methodology which must be respected,
whereas, in the case of a unilateral act, only estoppel, acquiescence or the
existence of a treaty, custom or other obligation prevented an equally unilateral
termination.
However, according to another view, a unilateral act could not be revoked at
any time because a State which had unilaterally expressed its will to be bound
was, in fact, bound . . . Unilateral acts, like treaties, lead to situations in which
States were caught against their will; once expressed, their commitment was
irrevocable, yet the treaty or act had no effect unless invoked by other States.
Nonetheless, the point was also made that a unilateral act could be terminated in good faith and that the technique of revocation deserved its place in
the study of means of terminating unilateral acts.269

The Special Rapporteur opted for the general irrevocability of obligations

undertaken by promises, a position which was apparently to be understood
Ibid 58, para 186.
Ibid. For the different positions and their exact wording see especially the remarks
made by Yamada, Hafner, Goco, Pambou-Tchivounda and Herdocia Sacasa at the end of
ILC, 2526th Meeting, n 12 above, para 69 et seq.
269
ILC, Report of the Fifty-fourth Session, A/57/10, n 12 above, 20607, paras 36162.
267
268

The Revocability of Promises 257

as the irrevocability position presented above, ie allowing for a unilateral

undertaking to be amended or revoked only by consensus, according to
the terms incorporated in it or on account of a fundamental change in circumstances or in the case of supervening impossibility of performance.270
This position was never completely given up, but his Ninth Report sent
some mixed signals,271 indicating that Cedeo eventually moved closer to
what Tomuschat has described as the mainstream that preferred a solution
purporting carefully to balance the different interests at stake.272 According
to the Special Rapporteur, the undertaking need not be regarded as a perpetual obligation from which the State can never free itself.273 Instead, and
citing Charles de Visscher, the special circumstances of a case should allow
a relative, flexible position to be adopted.274 Guiding Principle 8, which the
Special Rapporteur proposed, however, provided no room for a more flexible revocation or termination.275 Instead, it first listed the rather unproblematic aspects of an acts automatic termination, where a specific time limit
or condition had been included in the promise, or where a new peremptory
norm had emerged.276 Only two scenarios allowing for subsequent willed
revocation were mentioned, ie where the subject matter had ceased to exist,
or where a fundamental change of circumstances had taken place, although
the latter, oddly, only referred to cases where the fulfilment had thereby
become impossible. The question if and how a promise may be revoked
which does not announce its own lifespan was therefore not expressly
addressed, but Cedeos proposal contained no clause that allowed for
270
See Victor Rodriguez Cedeo, Fifth Report on Unilateral Acts of States, Addendum 2,
UN Doc A/CN.4/525/Add.2 (2002) 11, para 184 and Victor Rodriguez Cedeo, Sixth Report
on Unilateral Acts of States, UN Doc A/CN.4/534 (2003) 2728, paras 11619, with the latter
not being restricted only to recognition and according to which: In sum, unilateral acts can
be said to be unmodifiable in the broad sense of the term, unless the opposite can be inferred
from the act itself or derived from circumstances or conditions provided for therein, or, as
we shall see below, from external situations, para 118.
271
See Cedeo, Ninth Report, Addendum, n 142 above, 30, para 87 and 32, para 92, contra
31, para 88 and especially 35, para 101.
272
Tomuschat, Unilateral Acts, n 43 above, 1498.
273
Cedeo, Ninth Report, Addendum, n 142 above, 35, para 101.
274
Ibid.
275
While the Special Rapporteur had suggested distinguishing between automatic termination of a unilateral undertaking and an acts revocation in the form of a new wilful act
(see Cedeo, Ninth Report, Addendum, n 142 above, 29, para 83 and 35, para 101), the suggested principle followed this maxim only half-heartedly, as it provided that an act may be
terminated or revoked by the formulating State on the grounds listed therein, see Cedeo,
Ninth Report, n 135 above, 3, Guiding Principle 8.
276
A parallel approach was thereby proposed to that followed in Vienna Convention, Art
64. In the latter, the continued dissent amongst Commission members as to whether a treaty
was to be considered void once a new peremptory norm emerged or merely terminated is,
however, still reflected in the Articles wording which simply pretends that both approaches
are compatible and announces that in such scenarios a treaty . . . becomes void and terminates; for more see Anne Lagerwall, Article 64, Convention de Vienne de 1969 in Olivier
Corten and Pierre Klein (eds), Les Conventions de Vienne sur le droit des traits (Brussels, 2006)
vol III, 2299, 233539.

258 The Law on Promises

revocation on other than the above-mentioned grounds. His task was, of
course, complicated by the fact that he was still dealing with the entire
subject of unilateral acts and not only that of promises. A more flexible
position on revocation would therefore have opened the gates to easier
ways of revocation also for waivers and acts of recognition, even though in
these areas, there is much less reason to doubt their irrevocability.277
For this reason alone it is not surprising that the ILCs final outcome
deviates substantially from the Special Rapporteurs proposal. As will
be seen, however, it simply abstained from answering the question and
deciding between a more flexible and a rigid bar on revocation. Instead,
a couple of specific circumstances and the effects they will have on the
revocability of promises are listed, but the treatment of all others, ie the
average promise, is left open to question.
C Guiding Principle 10: Providing an Answer (Only) for
Specific Circumstances
Turning to the Guiding Principle finally adopted by the ILC takes us (as
in other areas of their application) directly to the ICJs dicta contained in
the Nuclear Tests cases. In respect of the French undertakings, the ICJ had
stated:
The Court finds that the unilateral undertaking resulting from these statements
cannot be interpreted as having been made in implicit reliance on an arbitrary
power of reconsideration.

The ILC copied this scrap of dictum into its Guiding Principle 10,
according to which a unilateral declaration which has created legal
obligations for the state making it hence can not be revoked arbitrarily. To bar an arbitrary revocation, of course, immediately triggers the
question how the term arbitrary is to be understood. Herodocia Sacasa
referred to the ICJs wording in the ILC by interpreting non-arbitrary as
indicating a not unlimited power of revocation,278 which is in line with
the above finding that necessary limitations must exist, but it does not
help in determining them. As Jacqu has remarked, referring to arbitrariness reminds a lawyer of the prohibition on the abuse of a legal right,279
as a result of which the limits imposed on a state to revoke its declaration
would, however, be anything but strict. For Jacqu, it is possible to speak
of an abuse (in the rare cases) where a state intends to revoke its undertaking without a motive, merely to abdicate its responsibility or in order to
277
278
279

On the revocation of a waiver see p 35, on that of recognition see p 32.

See ILC, 2526th Meeting, n 12 above, 57, para 77.
Jacqu, Promesse unilatrale, n 97 above, 343.

The Revocability of Promises 259

harm the beneficiary.280 But although the ILC could not reach a consensus
which allowed it to go beyond the Courts not very informative terminology in this respect, it did agree on some indicators and an open-ended list
of criteria,281 which are to serve in assessing whether a revocation would,
in fact, be arbitrary or not. Guiding Principle 10 reads in full:
A unilateral declaration that has created legal obligations for the State making
the declaration cannot be revoked arbitrarily. In assessing whether a revocation
would be arbitrary, consideration should be given to:
(a) any specific terms of the declaration relating to revocation;
(b) the extent to which those to whom the obligations are owed have relied on
such obligations;
(c) the extent to which there has been a fundamental change in the circumstances.

The ILCs Commentary emphasises that the ICJ had not excluded a
states power to terminate a unilateral act, but only its arbitrary withdrawal or amendment. There could hence:
be no doubt that unilateral acts may be withdrawn or amended in certain specific circumstances. The Commission has drawn up an open-ended list of criteria
to be taken into consideration when determining whether or not a withdrawal
is arbitrary.282

Although aspects (a) (specific terms) and (c) (a fundamental change

in the circumstances) are, as just seen,283 regularly mentioned in articles
or debates on the matter as allowing a declarations revocation, they are
usually not debated under the heading of a declarations arbitrary or nonarbitrary withdrawal. Yet, at first sight, there appears to be nothing wrong
with doing so. Indeed, where a fundamental change of circumstances
within the strict limits of the customary rule enshrined in Article 62 of
the 1969 Vienna Convention on the Law of Treaties, as the Commentary
says,284 has occurred, the declarant hardly acts arbitrarily in revoking the
declaration. A unilateral declaration may hence be rescinded in this case,
as the Commentary says.285 Similarly, where a state acts according to the
terms stipulated in its undertaking, it is not behaving arbitrarily, but in
fact only keeping within the undertakings expressly announced limits.
The ILCs decision to include the three above-cited aspects under a
common principle in assessing a revocations arbitrariness is, nevertheless,
280
Ibid. Surprisingly, the non-arbitrary revocation scenarios subsequently mentioned by
Jacqu are, however, only those which are also codified in the law of treaties.
281
ILC, Guiding Principles with Commentaries, n 46 above, 380, Commentary to guiding
Principle 10, para 2.
282
Ibid, emphasis added.
283
See p 253 et seq.
284
ILC, Guiding Principles with Commentaries, n 46 above, Commentary to Guiding
Principle 10, 381, para 3.
285
Ibid.

260 The Law on Promises

problematic, as it might lead to the misunderstanding that the Commission
had found a coherent concept in determining a promises revocability. The
ILC, after all, determines that arbitrary revocation is prohibited and then
lists different scenarios, with the Commentary explaining that there could be
no doubt that unilateral acts may be withdrawn or amended in certain specific circumstances. This finding could easily be interpreted as meaning that
unilateral acts may be withdrawn or amended only in certain specific circumstances. If this were true, the controversial positions within the Commission
would have been overcome in favour of a consensus on a default rule,
according to which a promise usually (ie absent special circumstances)
cannot be revoked. And indeed, why should the fundamental change of
circumstances allowing revocation be expressly mentioned, if promises
were revocable even where the circumstances (including a states mind)
had merely changed. The e contrario argument of including the rebus sic
stantibus clause hence seems to indicate that promises are usually not revocable. The Commission would then have answered the question on the
revocation of promises by setting strict limits and by, effectively, adopting a
Vienna Convention-like approach. As a closer look at the Commentary on
paragraph (a) and especially on paragraph (c) shows, such an interpretation of Guiding Principle 10 would, however, be erroneous.
To begin with aspect (a), the special terms relating to revocation and
included in a declaration: the oddly worded commentary here does not
emphasise the possibility of a state revoking its declaration according to
the terms incorporated, as is usually done when referring to a declarations wording in this regard. Instead, it stresses the restraining power
of such a stipulation, as the declaration cannot be revoked unless the
conditions laid out have been met.286 Here, the Commission considers a
state obliged to remain within the self-imposed limitations. This would,
however (and again e contrario) mean that where no such special terms
hindering revocation have been included, a declaration must be more
easily revocable; otherwise, the inclusion of special terminology would
not matter. Whether the ILC therefore really assumed a general rule of
non-revocation, as the clausula rebus sic stantibus seems to indicate, already
becomes doubtful; and the analysis of aspect (b) confirms that it did not
do so, since it turns out to militate strongly for the opposite ground rule
according to which promises are in principle (though not necessarily
immediately) revocable. Paragraph (b), as cited, refers to the extent to
which those to whom the obligations are owed have relied on such obligations and the Commentary explains that the ILC contemplated a situation
where its addressees have relied on it in good faith and have accordingly
been led detrimentally to change position or suffer some prejudice.287
286
287

See the clarification included in note 982 within the Commentary, ibid 380.
Ibid Commentary to Guiding Principle 10, 38081, para 3.

The Revocability of Promises 261

The Commission, in the last part of this sentence, quotes the ICJ and
adopts the wording it used in the Military and Paramilitary Activities in and
against Nicaragua case. There, however, the Court referred to the estoppel
principle, which explains why it referred to a detrimental change in position on the part of the addressee. The ICJ said in full:
Furthermore, as the Court pointed out in the North Sea Continental Shelf cases
[reference omitted] estoppel may be inferred from conduct, declarations and the
like made by a State which not only clearly and consistently evinced acceptance by that State of a particular regime, but also had caused another State or
States, in reliance on such conduct, detrimentally to change position or suffer some
prejudice.288

While estoppel is addressed below in more detail, it is at this point only

important to note that in an estoppel scenario, a state is (suddenly) no
longer allowed to alter its earlier position on account of the addressee
having not only relied on the statement made, but having in addition
detrimentally changed its position or having suffered some prejudice,
as the Court stressed in the quotation above. Faced not only with simple but with detrimental reliance on the part of the addressee, the estoppel
scenario hence limits the declarants freedom to go back on its word in
order to protect the addressee from actually suffering the impeding detriment. This means that by pointing to estoppel, paragraph (b), via its com
mentary, refers to special circumstances in which a declaration may not
be revoked. Yet, if the non-arbitrary principle already rigidly barred revocation (absent a fundamental change of circumstances or other special
factors such as the addressees consent), there would be little need to stipulate that the exceptional scenario, in which the addressee has relied to
its detriment, will have the very same effect of barring revocation. There
would, in other words, be no need for the estoppel protection as an exception to the rule, where the rule is already that of non-revocability.
A closer analysis, therefore, provides a contradictory result, if we examine the ILC Principles in order to discern whether a promise may usually,
ie absent special circumstances, be revoked or not. Emphasising the effect
of a fundamental change in circumstances, e contrario, seems to imply that
a promise may usually not be revoked, while stressing the limiting power
of special terms and the restraining force resulting from the estoppel principle both indicate that the revocation of promises absent these factors
is not already barred. What is thereby revealed is that the controversial
debates within the ILC as to whether promises are freely revocable, or
whether they must, in this respect, be subject to a treaty or a more flexible regime, ended in no consensus, which is, hence, reflected in Guiding
288
ICJ, Military and Paramilitary Activities in and against Nicaragua (Nicaragua v United
States), Jurisdiction and Admissibility, Judgment of 26 November 1984, [1984] ICJ Rep 392,
415, para 51, emphasis added.

262 The Law on Promises

Principle 10. The different opinions remained unresolved and have only
seemingly been tied together under one concept, by using the ICJs reference to arbitrariness. Such a common concept, however, does not exist
and what the Commission has in fact done, is to simply address special
scenarios on the legal treatment of which most scholars, irrespective of
their opinion as to the ground rule in this area, can agree. For those who
generally opt for free revocability, the exceptional estoppel scenario must
remain of particular relevance in protecting the addressee from suffering
considerable detriment in extreme cases. For proponents of a treaty rule
transferral, revocation without the addressees consent is already barred
in these cases, yet mentioning the estoppel principle can do no harm, as
the latter will merely lead to the same result, ie that a promise cannot be
revoked. For the Vienna Convention rule advocates, revocation, in light
of a fundamental change of circumstances, is possible (as an exception),
while a supporter of free revocability can also live with this finding, which
will for him or her only make the undertaking all the more revocable.
Those opting for a more flexible approach could, apparently, also agree
that these special scenarios represented reasonably clear-cut cases, as they
lie at both ends of the conceivable spectrum. Hence, it appears that what
the ILC members could agree on was how to treat such special settings,
but on nothing more. How revocation of the average promise, ie one
where no special terms are included, no fundamental change of circumstances has taken place and the addressee is not protected by estoppel, is
to be dealt with, is therefore still open to question. With the Commentary
on the aspects mentioned, in effect, pointing in both directions, a systematic reading of the term arbitrary is not possible in order to illuminate
this question.
While a ground rule on revoking promises has therefore not been
established, at least the legal treatment of the following special scenarios
seems to be covered by a consensus:
(1) where specific terms are included in the declaration, it can be revoked
(only)289 according to the terms included;
(2) where a fundamental change of circumstances has taken place as
defined in Vienna Convention, Article 62, a declaration can be revoked;
(3) where the addressee is protected under the estoppel principle on the
grounds that it would suffer detriment in case of revocation, the latter
is barred.
Although not included in the Guiding Principles, there seem to be a
couple of other scenarios for which also a consensus on how they are to
be treated exists:
289
The collision with other special scenarios such as a fundamental change in circumstances probably set aside.

The Revocability of Promises 263

(4) where the addressee consents to revoke the undertaking, it may be

revoked;290
(5) where a declaration includes a condition or a time limit, it will terminate as declared;291
(6) where the addressee rejects the undertaking, the declarant is no longer
bound vis--vis the latter by its promise.292
While the legal treatment of promises and their revocation (or automatic termination) in these specific circumstances is fairly clear, the
underlying, and central, question as to whether promises may generally
and outside such settings be revoked or not, is not answered and will now
be addressed in more detail.
D Specific Circumstances Aside: Determining a Ground Rule on
Revocation
As seen, the ILC reached no consensus on an existing or future ground
rule as to whether or not the average promise may be revoked, ie a
promise which does not fall within the above-mentioned specific scenarios that either clearly allow or hinder a revocation. To recall, Guiding
Principle 10 indicates that a declaration which contains specific terms
may only be revoked accordingly. It also emphasises that where a fundamental change of circumstances has taken place, an undertaking may be
rescinded. In an estoppel scenario, on the other hand, revocation will not
be possible. But what Guiding Principle 10 does not answer is what will
happen if none of these special scenarios are fulfilled. In other words: can
a promise including no special terms be rescinded by a state on account of
a change of policy, provided that the addressee will suffer no considerable
detriment293 thereby?
(i) Towards a More Flexible Approach
The answers provided by scholars and ILC members to this question have
already been cited as ranging from unrestrained revocability, through
general irrevocability to a more flexible position. The position allowing
the termination of an undertaking, at will and any time, has been rejected
as incompatible with the binding nature of a unilateral undertaking. The
adoption of the strict rules applicable to treaties would answer the above
Or is terminated, if the parties conclude a treaty to this effect.
Since thereby the undertaking will simply have the legal effect it was announced to
have.
292
Since then the basis of the obligation will collapse. On the impact of the addressees
reaction on the promise made see p 247 et seq, also p 205 et seq.
293
For more on estoppel as protecting the addressee, see p 277 et seq.
290
291

264 The Law on Promises

question in the negative, and even outside an estoppel scenario, a promise
therefore (special terms or a fundamental change of circumstances aside)
could not be revoked. The proposition to treat obligations assumed by
promises in a parallel fashion to their consensual counterpart is, however,
not necessitated on the basis of the good faith principle. The latter only
requires some (yet not necessarily the same rather strict) limitations in
order to establish a promises bindingness and protect the addressees
reliance thereon. Similarly, by following the ICJs dicta, only an arbitrary revocation is prohibited, which is fully compatible with allowing
a state to revoke an undertaking even outside the scenarios mentioned
in the Vienna Convention. Not backed by state practice or the ICJs dicta
and in the absence even of a consensus amongst scholars (the ILC debates
well illustrating the differences in opinion), the adoption of treaty provisions is presently only a proposition for progressive development in this
area. The advantages and disadvantages of doing so should therefore be
carefully assessed, and some of the more negative aspects of a general
ground rule of irrevocability have been raised within the Commission.
Especially those who opted for a more flexible approach stressed that
states may be reluctant ever to undertake a unilateral obligation through
a promise should this mean that their hands would remain tied for all
future times.294 It is, of course, true that even under an adoption of the
Vienna Convention rules, states could evade this problem by proclaiming
a specific regime for revocation in the declaration itself. But as the depiction of unilateral declarations in chapter two has shown, states might very
well be reluctant to declare an undertaking only to immediately draw its
beneficiarys attention to the undertakings revocability. In answering the
questionnaire circulated by the ILC, Sweden went as far as considering
the absence of such regulation within the declaration itself as being part of
the nature of a unilateral act:
While treaty regimes usually contain provisions on issues like termination, suspension and withdrawal, it is in the nature of unilateral acts that they do not
regulate the corresponding issues. For this reason, there is a need for general
rules on the subject.295

While it is not necessary to go quite as far, such stipulations are undoubtedly rare and probably for good reason, as they might be understood as
emphasising and highlighting rather the limits of the obligation solemnly
announced than the obligation itself. It therefore appears to make a (psychological) difference whether a solemn promise is held to be revocable
according to a general legal framework, or whether an undertaking is
usually irrevocable and a state, in order to alter it, will hence have to sol See the text to n 268 above.
See the Swedish reply to question 9 on the possible revocability of a unilateral act in
ILC, Government Replies to the Questionnaire, n 2 above, 21.
294
295

The Revocability of Promises 265

emnly announce not only the undertaking, but continue by setting out
the regime for its revocation. Although the obligation undertaken may be
the same in both scenarios, the effect will be different; even a (civil) marriage can end in divorce according to the law codified in this respect, yet
the I do would surely lose some of its power in the eyes of the beloved
addressee, were the ground rule altered to be that of complete irrevocability, whereby the very same vow would now be complemented by an
express statement listing all possible grounds for the wedding bonds
termination. Requiring states to expressly announce the limits of their
undertakings might, therefore, be directly opposed to what states actually want to achieve by their solemn promise.296 If necessary, states might
therefore choose not to adopt the undertaking at all. The argument for
express terms to be included as regards altering the scope of an obligation works better the opposite way, as even where an undertaking can be
revoked subject to a more flexible regime, a state remains free to add on
to its obligation express terms limiting the grounds of revocability.
In addition, as chapter two has shown, in the realm of promises it is
in particular the oral promise which has been at the centre of attention.
As such, it constitutes a quick and informal tool to assume an immediate
obligation. What may be an advantage for states looking for a mechanism
to do so has nevertheless also been identified as one of a promises more
dangerous characteristics.297 Along with the impact a hastily assumed
undertaking will have, this danger will only increase if the obligation
undertaken is considered to be basically irrevocable. Finally, Tomuschat
in his article has raised awareness of another consideration which speaks
in favour of a more flexible approach:
At a time when the democratic principle is gaining ground as a decisive factor
for political decision-making both at the domestic and international levels, it
seems to be even less justifiable to consider as set in stone a promise that was
made in a specific historical context.298

The problem identified with a regime of general irrevocability is, hence,

that it will compel the population of a state to remain perpetually bound
by an undertaking which its predecessors representatives have once
assumed, provided that the latter did not insist on a unilateral exit strategy when adopting the undertaking. Unless they have done so, under
296
See also Sicault, Engagements unilatraux, n 70 above, 65253, who in respect of a
time limit integrated into the declaration made says: En effet, le recours une telle limite
dans le temps pourra souvent tre interprt de manire ngative par les destinataires et,
ds lors, lengagement unilatral mis avec de telles restrictions irait lencontre du but
recherch par son auteur. For declarations under the Optional Clause this practice has,
nevertheless, become rather common, see, eg Juan Jos Quintana, The Nicaragua Case
and the Denunciation of Declarations of Acceptance of the Compulsory Jurisdiction of the
International Court of Justice (1998) 11 Leiden J International Law 97, 10607.
297
See Tomuschat, Unilateral Acts, n 43 above, 1496.
298
Ibid 1497.

266 The Law on Promises

a ground rule of irrevocability, present and future generations therefore
could not simply opt out of the promise made (the limited exceptions
of a fundamental change of circumstances and the beneficiaries consent
aside). In pointing to this problem, Tomuschat, of course, sees the other
side of the coin and the danger which lies in a rule allowing for an easier
revocation of obligations undertaken as potentially destabilising the stability and the trust that can be placed in an international undertaking.
He accordingly suggests a compromise should be achieved between a
states freedom of action, on the one hand, and the expectations raised
in the addressee as having received a binding undertaking, on the other.299 And surely, if every change of government allowed for a complete
reassessment of the obligations undertaken, trustworthy long-term agreements would become impossible. The ability to place trust in an undertakings continued existence ranging beyond the next election period in a
given country is a necessary precondition in many areas of international
cooperation. Many legal commitments, be they targeting areas of nuclear
disarmament, climate change, decreasing fishing stocks or investment
regulation, would lose their effectiveness if made subject to permanent
re-evaluation by every single state, despite it having pledged continued
adherence to them.
In this dilemma or relationship of tension between a populations freedom of continued self-determination and the necessity for stable obligations, the difference between the unilateral and the bilateral instrument
chosen for the obligation to arise might become decisive. In other words:
what is true for the agreements discussed above need not necessarily
apply to unilateral undertakings. In addition, since the pacta sunt servanda
principle, with its limited possibility of revocation, already establishes a
rather inflexible yet thereby very stable mechanism for a state to assume
an obligation and ensure its continued cooperation with one or several
states, the unilateral assurance could fill the gap by presenting itself not
only as a more flexible counterpart in assuming an obligation, but also
as a more flexible one in rescinding it, at least as a ground rule. A state
would nevertheless remain free to expressly assume an irrevocable or
largely irrevocable obligation, similar to that contained in a treaty, by
announcing its decision to do so in express terms. Where a state has done
so, the addressee will be allowed to trust in the perpetual existence of the
undertaking assumed and its trust will be legally protected. But where
no such announcement has been made, an undertaking could not generally be relied upon as being irrevocable. Here, a compromise between the
declarants freedom and the necessity to allow for trust and stability in
international assurances would lead to a more flexible regime. But what
in fact does this mean?
299

Ibid 1497.

The Revocability of Promises 267

(ii) Reasonable Notice Requirement

Sicault, in taking the mechanism incorporated in some treaties as a blueprint, considered a double delay to be a sensible solution. According
to this idea, a state cannot revoke its undertaking for a certain limited
period, after which revocation becomes possible but again will only be
effective after another time limit has passed. For Sicault, such a solution
does not place an excessive burden on the declarant while simultaneously
protecting the rights of the addressee.300 And indeed, it appears to be
equally balanced and to form a middle ground between the two extreme
positions of unrestrained and complete irrevocability. The addressee is
allowed to rely on the statement made, yet under certain limits which
nevertheless guarantee the obligation assumed by the declarant through
unilateral means, but one from which it can be released should it consider
a revocation to become necessary. But to achieve this result, a double delay
is not strictly necessary and can instead be boiled down to only a single
one, according to which a state cannot revoke its undertaking with immediate effect. As long as the time span before a revocation becomes effective is long enough to allow the addressee to readapt its position to the
developing legal scenario, it will be adequately protected; especially so
as the estoppel doctrine will, in any event, serve as an ultimate safety net
for extreme situations. In other words, should the addressee, while acting reasonably, have changed its position to now suffer considerable detriment once the undertaking is revoked, estoppel will hinder a state from
revoking its obligation.301 But only in these cases would the undertaking
become de facto irrevocable, a consequence which does not put an unjustifiable strain on the declarant, since reasonable reactions to its statements
are usually, as such, also foreseeable ones.
Although this more flexible approach has here been introduced as a
proposition to counter the negative implications an adoption of the Vienna
Convention rules might have, it is, in fact, backed up by ICJ dicta. While
the ICJ in the Nuclear Tests cases might have already hinted at imposing a
more liberal regime when it only barred arbitrary revocation, it was not
very precise in what it actually meant. But despite this lack of precision
on the revocability of undertakings in the Nuclear Tests cases, it would be
wrong to present the Courts judgments as being otherwise devoid of any
guidance in this matter. Rather, the Military and Paramilitary Activities in
and against Nicaragua judgment on the jurisdiction and admissibility of the
claim is particularly important and telling on the issue of the revocability
of unilateral promises. Reference has already been made to that holding
when discussing the treatment of a unilateral declaration made pursuant to ICJ Statute, Article 36(2). In order to fully grasp the extent of the
300
301

See Sicault, Engagements unilatraux, n 70 above, 653.

See p 277 et seq.

268 The Law on Promises

answer provided by the Court to the question addressed here, the following important facts need to be recalled.302
The United States had accepted the ICJs jurisdiction in making a declaration pursuant to ICJ Statute, Article 36(2) on 14 August 1946, in which it
had stipulated that the declaration will remain in force for a period of five
years and thereafter until the expiration of six months after notice may be given
to terminate this declaration.303 This declaration of 1946 (which expressly
announced a double delay as just discussed) was followed by a second
declaration on 6 April 1984. That day, the United States deposited a notification with the UN Secretary-General, according to which all disputes
with Central American states were to be excluded from the ambit of the
obligation undertaken in 1946. In respect of the six-months notice period
proclaimed in 1946 (the only one that mattered as the five years after
its proclamation had already lapsed), the new notification stipulated the
following:
Notwithstanding the terms of the aforesaid declaration, this proviso shall take
effect immediately and shall remain in force for two years, so as to foster the continuing regional dispute settlement process.304

Since Nicaragua, as a Central American state, had filed its application

with the Court three days after the United States had notified the SecretaryGeneral, the decisive question was whether or not the United States had
the right to (immediately) terminate their original undertaking. The ICJ
approached this question as follows. According to the Court, declarations
under the Optional Clause, and hence those now before it, were:
facultative, unilateral engagements that States are absolutely free to make or
not to make. In making the declaration a State is equally free either to do so
unconditionally and without limit of time for its duration, or to qualify it with
conditions or reservations. In particular, it may . . . specify how long the declaration itself shall remain in force, or what notice (if any) will be required to
terminate it.305

The judgment thereby affirmed that a state may introduce numerous

qualifications into its freely made unilateral undertaking, including those
pertaining to the duration of the undertaking made. The Court continued:
302
For comments on the Courts judgment see, eg HW Briggs, Nicaragua v. United States:
Jurisdiction and Admissibility (1985) 79 American J International Law 373; Thomas M Frank,
Icy Day at the ICJ (1985) 79 American J International Law 379; for the aspects relevant here,
see especially Quintana, The Nicaragua Case and the Denunciation of Declarations of
Acceptance, n 296 above; also Francisco Orrego Vicua, The Legal Nature of the Optional
Clause and the Right of a State to Withdraw a Declaration Accepting the Compulsory
Jurisdiction of the International Court of Justice in Liber Amicorum Judge Shigeru Oda (2002)
vol 1, 46379.
303
ICJ, Military and Paramilitary Activities, Jurisdiction and Admissibility (1984), n 288 above,
398, para 13, emphasis added.
304
Ibid, emphasis added.
305
Ibid 418, para 59.

The Revocability of Promises 269

However, the unilateral nature of declarations does not signify that the State
making the declaration is free to amend the scope and the contents of its solemn
commitments as it pleases. In the Nuclear Tests cases the Court expressed its
position on this point very clearly [citation and reference omitted].306

The judgment continues by citing the passage from the Nuclear Tests
cases judgment according to which unilateral acts made with the
intention to be legally bound establish a legal undertaking for the state
making them.307 In addressing the revocability and modifiability of unilateral undertakings under the Optional Clause, the Court, in other words,
referred directly to its finding of the binding force of unilateral under
takings in the Nuclear Tests cases. As has already been illustrated in chapter one,308 it nevertheless continued by explaining that the declarations,
while being unilateral acts, establish a series of bilateral engagements
with other States accepting the same obligation of compulsory jurisdiction. With a series of bilateral engagements established by the declarations under the Optional Clause, it may be worth briefly (re)considering309
whether the ICJs findings which followed are relevant for the average
promise, which does not give rise to such bilateral engagements. Once
posed, the answer must clearly be yes, it is relevant, because the Court
did not treat the engagements established by the various unilateral declarations as treaty engagements, but saw the network as being established
by various unilateral undertakings. Otherwise, there would have been no
need to refer to the Nuclear Tests cases in the first place pointing to pacta
sunt servanda would have been enough and more effective (yet misguided
as the undertakings are created by a unilateral act). The Court, in addition, continued by emphasising that good faith played an important role
in this network of (unilateral) engagements, only to again refer directly
to its finding in the Nuclear Tests cases, by incorporating into its judgment
the citation according to which good faith serves as the basis of a unilateral acts bindingness. The Court, in these two passages and in the following, therefore does not establish the limits of the declarations revocation
in reference to any bilateral entanglement created, but in direct reference
to its finding on the bindingness of unilateral declarations in the Nuclear
Tests cases.310 What it thereby has in fact done is to answer the question
which is pressing here, since it established how the good faith principle
Ibid.
See pp 12021.
308
See p 70.
309
See p 69 et seq.
310
See again p 69 et seq and in particular the references provided in ch one, n 205 and
accompanying text. For an analysis of this part of the Nicaragua judgment see also Quintana,
The Nicaragua Case and the Denunciation of Declarations of Acceptance, n 296 above, for
whom at 111: declarations under Article 36, paragraph 2 to 5 of the Statute of the Court
constitute unilateral acts of states and as such are governed by a special set of rules of international law that accord a privileged place to the principle of good faith.
306
307

270 The Law on Promises

will limit a states freedom in terminating its unilateral undertaking. The
Courts finding must consequently be of general relevance for all declarations falling under the ambit of its holdings in the Nuclear Tests cases, from
which the Court started its doctrinal analysis, ie for all promises.
After having emphasised the importance of good faith in international
relations in reliance on its Nuclear Tests cases holding, the ICJ went on to
address whether the United States was consequently free simply to disregard its six months notice period for any termination of its 1946 declaration. The US argument, according to which the new notification had not
actually contained a termination but only a modification, which as such
did not fall under the self-imposed notice requirement, had already been
rejected by the Court in the following words:
The truth is that it is intended to secure a partial and temporary termination,
namely to exempt, with immediate effect, the United States from the obligation to subject itself to the Courts jurisdiction with regard to any application
concerning disputes with Central American States, and disputes arising out of
events in Central America.

In now assessing the consequences of the US-announced notice requirement, the Court started by establishing that the United States, by its 1946
declaration, had entered into an obligation vis--vis other states parties to
the Optional Clause system. The latter are, indeed, the addressees of any
Optional Clause declaration and there will be no doubt that an obligation
hence existed. In respect of the precise regime allowing its termination,
the Court found the following:
Although the United States retained the right to modify the contents of the 1946
Declaration or to terminate it, a power which is inherent in any unilateral act
of a State, it has nevertheless assumed an inescapable obligation towards other
States accepting the Optional Clause, by stating formally and solemnly that any
such change should take effect only after six months have elapsed as from the
date of notice.

The first of two important findings contained in this passage is that, in

the opinion of the Court, the United States, by making its unilateral declaration, retained the right to modify its content or even terminate it, as the
latter was a power which is inherent in any unilateral act of a State. An
obligation assumed by a unilateral undertaking is, therefore, not irrevocable in the eyes of the ICJ. In determining the remaining question whether
and, if so, which limits exist on a declarations revocability, statements
which include terms as to their own revocability and those which do not
should be distinguished. As seen, the Court went on to address the first
category in the above quotation, in respect of which it made its second
important finding within this passage. The ICJ considered the United
States to have assumed an inescapable obligation vis--vis its addressees in respect of the six months notice requirement by formally and

The Revocability of Promises 271

solemnly declaring that it would follow this procedure. The ICJ thus held
the expressly assumed limit on revocation to be inescapable and therefore
not itself modifiable or revocable. For unilateral promises in general this
means that the bar imposed under good faith in the eyes of the Court is,
indeed, that self-imposed limits on a unilateral undertakings revocation
cannot later be ignored. In this respect, the Courts ruling backs the ILCs
finding, according to which the limiting character of special terms governing a declarations revocation have to be respected.
The six-months notice period was therefore found by the Court to be
an obligation on the United States; but thanks to the peculiarities of the
Optional Clause system, the argument as to whether Nicaragua could
invoke the US declaration in a suit before the Court was not fully closed
thereby. The United States claimed that since the declaration made by
Nicaragua included no time limit for revocation, that declaration was, as
a result, freely revocable,311 ie subject to a right of immediate termination
at the will of the Nicaraguan state (and not irrevocable as Nicaragua had
claimed).312 On the basis of this argument, the United States invoked the
reciprocity element established by the ICJ Statute, as under Article 36(2), a
state undertakes the obligations included in its unilateral declaration only
vis--vis any other state accepting the same obligation. The United States
hence argued that since the Nicaraguan declaration was freely revocable,
Nicaragua had not accepted the same obligation in its statement and,
for this reason, could not benefit from the more far-reaching US undertaking. Instead, the United States argued, it should be allowed, vis--vis
Nicaragua, to benefit from the right of immediate termination flowing
from the non-specific Nicaraguan declaration. While the Court already
dismissed the US argument invoking reciprocity in order to depart from
the terms used in its own undertaking, it nevertheless proceeded to identify that the US argument also hinged on the understanding that declarations which contain no provision as to their length are freely revocable. In
this respect, the Court made the following, important finding:
Moreover, since the United States purported to act on 6 April 1984 in such a way
as to modify its 1946 Declaration with sufficiently immediate effect to bar an
Application filed on 9 April 1984, it would be necessary, if reciprocity is to be
relied on, for the Nicaraguan Declaration to be terminable with immediate effect.
But the right of immediate termination of declarations with indefinite duration is far from
established. It appears from the requirements of good faith that they should be treated, by
analogy, according to the law of treaties, which requires a reasonable time for withdrawal
from or termination of treaties that contain no provision regarding the duration of their
validity. Since Nicaragua has in fact not manifested any intention to withdraw its
own declaration, the question of what reasonable period of notice would legally
311
See ICJ, Military and Paramilitary Activities, Jurisdiction and Admissibility (1984), n 288
above, 416, para 55.
312
Ibid 417, para 56.

272 The Law on Promises

be required does not need to be further examined: it need only be observed that
from 6 to 9 April would not amount to a reasonable time.313

According to the ICJ, unilateral declarations of infinite duration

therefore may not be terminated immediately. Instead, and following from
the requirements of good faith, they require a reasonable time for withdrawal. Since three days notice was clearly not a reasonable period, the ICJ
did not have to go into more detail at this point. In the just cited paragraphs,
the Court has therefore found declarations under the Optional Clause to
be unilateral declarations which on account of good faith requirements
cannot be freely revoked or amended. Instead, a reasonable time for any
withdrawal is necessary. What the ICJ has thereby said is that flowing from
good faith (and in direct reference to the Nuclear Tests cases): (1) unilateral
declarations are revocable, (2) where special terms have been included on
their revocability these have to be followed, and (3) unilateral declarations
not containing any provision can only be terminated after notice has been
given beforehand and after a reasonable time has passed.
It is interesting to see that the Court has argued by analogy to the law
of treaties in order to establish the reasonable period of revocation. It has
thereby drawn a very different consequence from the analogy than most
scholars have done, for whom a parallelism to the law of treaties has usually been thought to impose much stricter limitations. The reason for this
difference in the outcome of the treaty analogy is that the ICJ has focused
only on and jumped straight to Vienna Convention, Article 56(2), when it
referred to unilateral declarations of infinite duration to be treated similarly to treaties that contain no provision for their termination. Article 56
reads in full:
Denunciation of or withdrawal from a treaty containing no provision regarding
termination, denunciation or withdrawal
1. A treaty which contains no provision regarding its termination and which
does not provide for denunciation or withdrawal is not subject to denunciation
or withdrawal unless:
(a) it is established that the parties intended to admit the possibility of denunciation or withdrawal; or
(b) a right of denunciation or withdrawal may be implied by the nature of
the treaty.
2. A party shall give not less than twelve months notice of its intention to
denounce or withdraw from a treaty under paragraph 1.

Under Article 56(1), treaties which contain no provision regarding termination, denunciation or withdrawal are hence usually not subject to
denunciation or withdrawal. Only where it was otherwise intended or
313
ICJ, Military and Paramilitary Activities in and against Nicaragua (Nicaragua v United
States), Merits, Judgment of 27 June 1986, [1986] ICJ Rep 14, 41920, para 63, emphasis added.

The Revocability of Promises 273

implied by the nature of the treaty may a party denounce or withdraw

from a treaty under the 12 months notice requirement in Article 56(2).
The ICJ, nevertheless, went straight to paragraph (2) in its analogy, which
is in keeping with its finding that a unilateral undertaking may generally
be revoked. The ICJ thereby appears to be assuming an analogy to Article
56(1)(b) and hence to treaties where a right of denunciation or withdrawal may be implied by the nature of the treaty. The nature of the unilateral act, in other words, opened the gateway for applying Article 56(2)
by analogy. Even though the ICJ was, as seen, not required to pronounce
on the exact time-span required under a reasonable notice period, its
allusion to paragraph (2) of this Article therefore points to a minimum
requirement of 12 months notice.314
In continuing to follow the ICJs jurisprudence as guidance in this field,
the more flexible approach to the revocation of promises, which imposes
less restraints on the revocation of a promise than those found in the
Vienna Convention for the typical treaty, can be fleshed out in the way just
shown.315 In light of the apparent consensus on how to treat a number of
specific scenarios, it is their interplay with these findings which remains
to be determined.
E Consolidating the Approach: A Ground Rule from which to Deviate
in Specific Circumstances
As seen, good reasons militate in favour of a more lenient regime on revocation of obligations undertaken by promises than that in place for treaty
undertakings under the Vienna Convention.316 In following the Courts
314
On this aspect see also Quintana, The Nicaragua Case and the Denunciation of
Declarations of Acceptance, n 296 above, 11518, who emphasises that the determination of
what is reasonable will depend on the circumstances of the particular declaration. Quintana
is himself inclined to consider a shorter period of six to nine months, with a minimum of
three months, a reasonable period for termination of declarations under ICJ Statute, Art
36(2) but admits that there is strong evidence which suggests that an analogical application
of this standard [ie requiring 12 months notice] to Article 36 declarations would meet with a
favourable response from the Court, at 116.
315
Among the relatively few states (12) that have responded to the ILCs first questionnaire, it is especially the reply by Israel to the (very broad) question on the revocability of
(all) unilateral acts which reflects such a flexible position on the revocation of promises.
For Israel, a state can limit its right of revocation, but where it does not do so the possibility
of revocation, subject to certain conditions, should be accepted . . . In this regard it is worth
examining whether the principle of good faith, for example, should require that reasonable
notice be given prior to revocation, though such a condition may not be practical in every
instance, in ILC, Government Replies to the Questionnaire, n 2 above, 2021.
316
Besides the arguments and references provided above, see also the evaluation of the
present legal regime on declarations under the Optional Clause after the Nicaragua judgment by Quintana, The Nicaragua Case and the Denunciation of Declarations of Acceptance,
n 296 above, 119, which, mutatis mutandis, also rings true for promises in general: the Courts
position reflects a compromisory, equilibrated view on the subject, which denies validity to
both the unreal proposition that Optional Clause declarations embody perpetual commitments,

274 The Law on Promises

elaboration on the limits imposed by good faith on the revocation of
unilateral declarations under the Optional Clause (and established in
direct reference to the Nuclear Tests cases), the compromise to be achieved
in protecting the declarants freedom of self-determination and the bene
ficiaries expectations thereby looks as follows.
Functioning as its cornerstone is the rule that promises are revocable,
yet not freely and at any time. Revocation instead requires reasonable notice to be given to the addressee, in which a state announces its
intention to revoke its promise. This prior warning and the period which
follows until a revocation can become effective protect the addressee by
allowing it to adapt to the declarants new position. In following the consensus reflected in Article 56(2) for generally terminable treaty obligations
to which the ICJ has pointed, this period appears usually to require a state
to announce its intention to revoke (at a minimum) 12 months before its
undertaking is to cease having its effect. As actus contrarius, such notice
can be delivered in the same way as the original undertaking. For erga
omnes obligations this would in particular mean that another public statement is sufficient, and required, in which a state declares its intention to
rescind its undertaking, effective (at the earliest) 12 months after the pro
clamation has been made.
While this is the ground rule, the revocation of promises will be
treated differently where specific circumstances are present. To begin
with, where special terms are included in the original declaration, these
will govern an undertakings revocability and thereby impose tighter or
lesser limitations. A fundamental change in circumstances, the treatment
of which appears to be generally agreed upon, may, according to the ILC,
allow for a state to rescind its obligation. Under the depicted framework
such a fundamental change is not a necessary requirement for a declaration to be revoked; however, the fundamentally changed circumstances
(which should include a supervening impossibility of performance) might
allow the declarant to revoke its declaration without having to adhere to
the notice period usually required.317 In a similar vein, and at the opposite
side of the spectrum, revocation may still be completely barred under the
distinct (but in this respect important) international law principle of estoppel, whereby an addressee who has altered its position reasonably and to
its detriment will be protected.318 Under the no arbitrary revocation rule,
the rare cases of clearly abusive behaviour (ie revoking an undertaking
in order to harm the addressee) will also remain prohibited. The ground
frozen to eternity, and the somewhat cynical view according to which states can withdraw their
acceptances of the jurisdiction just as easily as they have made them. While the former would
be dangerously idealistic and could even give origin to a dbandade from the Court, the latter
would render completely illusory the notion of a compulsory jurisdiction.
317
For the Vienna Conventions approach, see Art 65 et seq.
318
For more see p 277 et seq.

The Revocability of Promises 275

rule of allowing revocation when coupled to a reasonable notice period is

therefore fully compatible with the ILCs Guiding Principle 10 which, as
seen, provides exceptions for specific scenarios at both ends of the spectrum, however, the lacuna in respect of how to treat the average promise
is now filled. In the assessment whether or not a declaration was arbitrarily revoked, the question whether or not reasonable notice has been
given would therefore have to be added. The ILCs Guiding Principle 10
could hence be altered to read as follows:
Guiding Principle on the Revocability of Promises:
1. A promise can be revoked, but not arbitrarily. In assessing whether or not a
revocation is arbitrary, consideration should be given to:
(a) whether the declarant has given reasonable notice to its addressee before
revoking the undertaking;
(b) whether the declarant acts in accordance with specific terms included in
its declaration and relating to revocation;
(c) whether the declarant can claim a fundamental change of circumstances
as having taken place, which should allow it to terminate its undertaking
even without adhering to the declarations terms or the usually required
reasonable notice period.
2. A promise cannot be revoked where the addressee has not only relied on it, but
has in addition reasonably been led to alter its position in a such a way according
to which it would now suffer detriment should the undertaking be revoked.

Since the estoppel principle has little to do with the question whether or
not the declarant has good reasons to revoke its undertaking, and hence
whether that state acts arbitrarily in revoking its declaration or not,
estoppel has received its own paragraph in the amended Principle presented above. Besides these grounds, which allow for an undertakings
revocation, there are other aspects which will lead to the automatic termination of a unilateral promise. They have already been addressed above
as being fairly unproblematic and will be recalled to include the following three scenarios: where (1) the right received through the declarants
undertaking is rejected by the addressee, (2) the beneficiary consents to
terminate the undertaking, or (3) a condition or time limit was included in
the declaration itself.319
F Concluding Remarks on the Revocability of Promises
There can be little doubt that, when depicting the law on promises, the ice
we are moving upon becomes at its thinnest when addressing the power
of a state to revoke the obligations assumed by its unilateral pledge. The
319

See pp 26263.

276 The Law on Promises

above approach has progressed step by step by first presenting the areas
where the ground is relatively stable. Whether a state is allowed to revoke
its undertaking or not in the special scenarios covered by the Guiding
Principles is, as seen, much clearer than whether it may generally do so or
not. Once the ground rule for promises was addressed, the waters therefore became muddier and the solution provided above cannot claim to
reflect the law, which is, as such, already firmly entrenched. In this respect
there is simply too much disagreement, too little state practice and only
limited jurisprudence. However, what can and has been done is to fill out
the blanks resulting from an agreement only on how to deal with special
scenarios, in depicting the direction the road is currently taking. We have
done so not only by using (in this writers opinion) the best arguments
as guidance, but by following a strong tendency that is visible within the
ILC (as has been seen, it has even been identified as the mainstream position therein)320 which opted for a more flexible approach. The latter is
furthermore strongly supported and was fleshed out with meaning by the
ICJs decision in the Military and Paramilitary Activities case on jurisdiction
and admissibility, a decision which in the context of a promises revocation has not received its deserved attention. As such, the above depicted
amended Guiding Principle 10 presents a legal scenario which, as to its
ground rule, would therefore benefit from more support, especially from
state practice and jurisprudence. Its outline is, however, clearly discernable and the resulting legal framework appears to be well balanced.
VII MODIFYING PROMISES

On the issue of the modification of promises, the finding by the ICJ in

response to the US argument that it was merely modifying and not revoking its undertaking in the Military and Paramilitary Activities in and against
Nicaragua case (as cited above)321 can be taken to have led the way. Where a
state intends to modify its promise only in order to avoid the original obligation, this modification will be treated as a partial or temporary revocation that is consequently governed by the rules applicable to a promises
revocation. As the same reasoning on limiting a states power of revocation applies for these sort of amendments, the ICJs approach is indeed
the only sensible one. Although Guiding Principle 10 expressly addresses
only limits on a unilateral declarations revocation, the Commentary
thereto also backs this position, as it considers a states power of arbitrary
withdrawal (or amendment) to be excluded.322 The criteria considered
See Tomuschat, Unilateral Acts, n 43 above.
See p 270.
322
ILC, Guiding Principles with Commentaries, n 46 above, 380, Commentary to Guiding
Principle 10, para 1.
320
321

Promises and Estoppel 277

applicable to an undertakings revocation will hence be equally applicable

for amendments which only restrict the scope of an undertaking without,
however, completely rescinding it.323 Where a state modifies its under
taking by extending its obligations, it may do so freely and with no other
restrictions than those imposed on all promises under international law
(ie especially no ius cogens infringements).
VIII PROMISES AND ESTOPPEL

The principle of estoppel has surfaced not only during the work and
debates within the ILC,324 but also in various scholarly writings on the
legal effects of unilateral acts in general or promises in particular.325 As
just seen above in addressing the question of a promises revocability, this
study makes no exception. To some extent, the following pages can be
read as an extended footnote to the (informed) finding above, according
to which estoppel will hinder a state from revoking its promise (only) in
the (exceptional) cases where its (below listed) prerequisites are met. That
an entire section rather than an actual footnote is required in this respect
is explained by the fact that the uncertainties connected to the estoppel
principle, characterised as a concept in evolution326 that lacks particular
coherence,327 are probably as numerous as those raised by the existence
of legally binding promises themselves. In order to describe the interplay
of the two principles, it is therefore necessary first to briefly identify and
summarise the characteristics of estoppel, before proceeding to illustrate
its relationship to promises.
AEstoppel
The history of the international law principle of estoppel (sometimes
also referred to as the rule of preclusion) and its evolution to its present
See also Sicault, Engagements unilatraux, n 70 above, 665.
For the Special Rapporteur see Cedeo, First Report, n 12 above, 2425, paras 12831;
Cedeo, Second Report, n 209 above, 34, paras 1114; Cedeo, Third Report, n 97 above,
6, paras 2327; Victor Rodriguez Cedeo, Seventh Report on Unilateral Acts of States, UN
Doc A/CN.4/542 (2004) 7677, paras 196201. For discussions within the Commission and
the different opinions expressed therein, see, eg ILC, Report of its Fiftieth Session, A/53/10,
n 264 above, 55, para 158 et seq; ILC, Report of its Fifty-second Session, A/55/10, n 194
above, 92, para 539 et seq.
325
See, eg Fiedler, Einseitige Versprechen, n 86 above, 4648; Gigante, Unilateral State
Acts, n 67 above, especially 347 et seq; Jacqu, Promesse unilatrale, n 97 above, 33539and
Jacqu, Elments, n 221 above, 256; Suy, Unilateral Acts as a Source, n 260 above, 64041.
326
Mller and Cottier, Estoppel, n 25 above, 116.
327
Brownlie, Principles, n 1 above, 644.
323
324

278 The Law on Promises

form has been described in various scholarly writings.328 Coming from the
Anglo-Saxon domestic concept of the same name, it is now established
and accepted in international law as a distinct and independent international legal principle. The ICJ, too, has recognised it by discussing its
application to the facts before it in various cases and has thereby, as will be
seen, helped to identify more closely the contours of estoppel under international law. Despite the introductory caveat according to which estoppel
is still evolving and lacks coherence, there is hence ample material available in order to try to establish its characteristic features. A perusal of the
doctrine and jurisprudence reveals the following.
Where applicable, estoppel will hinder the state against which it is
invoked from acting in a manner deemed inconsistent with its previously made allegations or earlier behaviour. As such, the typical effect
of estoppel has been identified as barring a state without regard to truth
and accuracy, from adopting successfully different subsequent statements
on the same issue.329 Where a state should try to do so, it is estopped,
or precluded, from acting as intended. Whether estoppel functions only
as a procedural principle which hinders inconsistent behaviour to be
relied on before a court, or as a material rule of substantive international
law, has been subject to some disagreement. Its substantive material value
has, however, been described as now generally accepted.330 Even if this
finding appears to be a slight exaggeration in light of various scholarly
writings to the opposite,331 the general tendency in legal doctrine indeed
appears to move in that direction.332 The ICJs jurisprudence has, in addition, been convincingly shown to reflect an understanding of estoppel
that goes beyond its use as a merely procedural tool which has effect only
by hindering a certain position, and allowing evidence of its truth to be
328
See, eg DW Bowett, Estoppel before International Tribunals and its Relation to
Acquiescence (1957) 33 British YB International Law 176; C Dominic, A propos du principe de
lestoppel en droit des gens (1968); Antoine Martin, Lestoppel en droit international public prcd
dun apercu de la thorie de lestoppel en droit anglais (Paris, 1979); IC MacGibbon, Estoppel in
International Law (1958) 7 International and Comparative Law Quarterly 468. See also Mller,
Vertrauensschutz, n 89 above, 5 et seq; Kolb, La bonne foi, n 5 above, 35793. Further references
provided by Mller and Cottier, Estoppel, n 25 above, 11819.
329
Mller and Cottier, Estoppel, n 25 above, 116.
330
Ibid 117.
331
See especially Martin, Lestoppel, n 328 above, 26371 and the further references provided by Kolb, La bonne foi, n 5 above, 383, note 179. As Kolb empasises, Martin, however,
indeed seems to draw his conclusion that estoppel operated only as a procedural defence
largely from its defensive character. Yet, [l]a question de savoir quel niveau opre cette
dfense, celui de la preuve ou celui du droit subjectif, est distincte, Kolb, La bonne foi, n 5
above, 385.
332
For scholarly support see especially Kolb, La bonne foi, n 5 above, 384 with further references. Fitzmaurice, writing in 1961, and referring to the rule of preclusion as the nearest
equivalent in the field of international law to the common law rule of estoppel, wrote that
it was certainly applied as a rule of substance and not merely as one of evidence or procedure, see ICJ, Temple of Preah Vihear (Cambodia v Thailand), Preliminary Objections, Separate
Opinion Fitzmaurice [1961] ICJ Rep 52, 62. See also MacGibbon, Estoppel, n 328 above, 512.

Promises and Estoppel 279

heard by a tribunal.333 By considering estoppel to be a substantial principle its defensive, shielding character is, of course, not denied it remains
a defence, but one applicable as a substantive rule of international law.334
In respect of estoppels requirements, two different concepts have been
advanced and it is essential to distinguish a wide understanding of
estoppel from the (now dominant) more restrictive interpretation of the
same principle. Under the former, estoppel functions as a general rule
of inconsistency which has its legal effect irrespective of any detriment
found to exist on the part of any third and interested state. Understood
this way, a state, in the words of MacGibbon, ought to maintain towards
a given factual or legal situation an attitude consistent with that which it
was known to have adopted with regard to the same circumstances on
previous occasions.335 The result would be that a state which has adopted
and made known a position at a given time will have to stick to it, as any
subsequent change of position would be barred as inconsistent behaviour. On the basis of this understanding, state action could and would
regularly trigger estoppel, which may in part explain why Brownlie has
found a tendency among writers to refer to any representation and conduct having legal significance as creating estoppel, precluding the author
from denying the truth of the representation, express or implied.336
Under the more sensible337 restrictive approach, a state is estopped from
changing its behaviour only where doing so would prejudice or result in
a detriment for another state, because the latter has altered its behaviour
in legitimate reliance on the former states conduct.338 As estoppel is said
to focus on the detriment which a declaration causes to the addressee in
relation to the declarant, it will also apply where the declaration results
merely in an advantage for the declarant state.339 Under the narrower
See the analysis provided by Kolb, La bonne foi, n 5 above, 38385.
See also Kolb, La bonne foi, as cited in n 331 above.
335
MacGibbon, Estoppel, n 328 above, 512.
336
Brownlie, Principles, n 1 above, 643.
337
For criticism of the wider notion see, eg Kolb, La bonne foi, n 5 above, 36970 and Mller
and Cottier, Estoppel, n 25 above, 117.
338
This altered behaviour might also consist of having done nothing, ie having refrained
from acting, where a state would otherwise have acted.
339
In this respect the following passage taken from Fitzmaurices separate opinion in the
Temple of Preah Vihear case is commonly referred to: The essential condition of the operation
of the rule of preclusion or estoppel, as strictly to be understood, is that the party invoking
the rule must have relied upon the statements or the conduct of the other party, either to
its own detriment or to the others advantage. The often invoked necessity for a consequent
change of position on the part of the party invoking preclusion or estoppel is implied in
this. A frequent source of misapprehension in this connection is the assumption that change
of position means that the party invoking preclusion or estoppel must have been led to
change its own position, by action it has itself taken consequent on the statements or conduct
of the other party. It certainly includes that: but what it really means is that these statements
or this conduct, must have brought about a change in the relative positions of the parties,
worsening that of the one, or improving that of the other, or both, ICJ, Temple of Preah Vihear,
Preliminary Objections, Separate Opinion Fitzmaurice (1961), n 332 above, 63.
333
334

280 The Law on Promises

concept it is hence merely such detrimental behaviour which is barred,
and it is barred precisely in order to protect the reliance legitimately
placed on the conduct or declarations of the other state. In other words:
[c]lear and unequivocal representation, prejudice or detriment are not
simply addenda; they trigger the very justification for specific protection
of settled expectations.340 Not every change of position is therefore immediately considered inconsistent behaviour, as the reason for hindering a
state from changing its position lies less in the changed behaviour as such,
but primarily in the protection of the legitimate expectations raised in the
other state. If the latter has legitimately placed reliance on another states
conduct, it should not suffer any harm therefrom.341
Between these two competing notions of estoppel, legal doctrine, after
leaning towards the wider version, has now taken a fairly clear stand in
favour of the narrower concept,342 which is also strongly supported by
numerous, and more recent, ICJ judgments. It was in its 1969 judgment
in the North Sea Continental Shelf cases (later, in 1984, self-characterised
by the Court as apparently . . . the judgment that gave the most precise
definition of the conditions for invoking the doctrine of estoppel)343 that
the ICJ very clearly opted in favour of a more restrictive understanding of
the estoppel principle. There, it said:
it appears to the Court that only the existence of a situation of estoppel could
suffice to lend substance to this contention that is to say if the Federal Republic
were now precluded from denying the applicability of the conventional regime,
by reason of past conduct, declarations, etc, which not only clearly and consistently evinced acceptance of that regime, but also had caused Denmark or the
Netherlands, in reliance on such conduct, detrimentally to change position or
suffer some prejudice.344
Mller and Cottier, Estoppel, n 25 above, 117.
See Mller, Vertrauensschutz, n 89 above, 11: Es kommt zum Ausdruck, da die estoppel-Doktrin eben genau betrachtet nicht bei einem generellen Verbot widersprchlichen
Verhaltens anknpft, sondern bei der sozialen Notwendigkeit, da eine Partei, durch das
nach den Umstnden berechtigte Vertrauen in das Handeln der andern nicht geschdigt
oder sonst benachteiligt werden darf. (My translation: What this means is that the estoppel
doctrine, when regarded up-close, does not actually focus on a general prohibition of inconsistent behaviour, but concentrates on the social necessity that a party may not be harmed
or disadvantaged by its, in view of the circumstances, legitimate reliance on the actions of
others).
342
Mller and Cottier, Estoppel, n 25 above, 117.
343
ICJ, Delimitation of the Maritime Boundary in the Gulf of Maine Area (Canada v United
States), Judgment of 12 October 1984, [1984] ICJ Rep 246, 309, para 145.
344
ICJ, North Sea Continental Shelf (1969), n 159 above, 26, para 30. For earlier jurisprudence
already moving in this direction see especially ICJ, Barcelona Traction, Light and Power Co,
Ltd, Preliminary Objections, Judgment of 24 July 1964, [1964] ICJ Rep 6, 25, where the Court
in respect of the estoppel claim before it was not able to hold that any true prejudice was
suffered by the Respondent. For other and older judgments, some of which are in favour
of the restrictive understanding of estoppel, some in support of the wider notion, and the
various views expressed in the separate opinions in ICJ, Temple of Preah Vihear (Cambodia v
Thailand), Merits, Judgment of 15 June 1962, [1962] ICJ Rep 6, see Martin, Lestoppel, n 328
above, 73173.
340
341

Promises and Estoppel 281

The Court confirmed this finding in the Military and Paramilitary

Activities in and against Nicaragua judgment, using the words to which the
ILC has made reference in its comment on a promises revocability and
which have already been mentioned in that context. The relevant paragraph will be recalled to read as follows:
Furthermore, as the Court pointed out in the North Sea Continental Shelf cases
[references omitted], estoppel may be inferred from conduct, declarations and
the like made by a State which not only clearly and consistently evinced acceptance by that State of a particular regime, but also had caused another State or
States, in reliance on such conduct, detrimentally to change position or suffer
some prejudice.345

Since then, the ICJ has continued to confirm the requirement of detrimental reliance in further holdings and it has done so by adopting very
clear wording.346 In light of this development it is therefore no exaggeration to speak of a consistent jurisprudence in this respect.
Under the narrower concept, estoppel is hence described by legal doctrine to be applicable where the following factors are present:347
(1) a states conduct or a declaration which is clear and unequivocal;
(2) triggering the effective, legitimate reliance in good faith of another
subject of international law on that conduct or declaration and inciting it to act or refrain from acting in a way which
(3) would result in a detriment or prejudice were the state against which
estoppel is invoked now allowed to alter its position.
In respect of criterion (1), it is the declaration rather than the conduct
which is of primary interest here, as promises necessarily require an
express declaration.348 Having an eye already on the relation between
promises and estoppel, it may be questioned whether a single declaration as such can give rise to an estoppel, or whether numerous consistent declarations would always be required (see the ICJ quotations above).
345
ICJ, Military and Paramilitary Activities, Jurisdiction and Admissibility (1984), n 288
above, 415, para 51.
346
See ICJ, Land, Island and Maritime Frontier Dispute (El Salvadore v Honduras), Application
by Nicaragua to Intervene, Judgment of 13 September 1990, [1990] ICJ Rep 92, 118, para 63:
So far as Nicaragua relies on estoppel, the Chamber will only say that it sees no evidence
of some essential elements required by estoppel: a statement or representation made by one
party to another and reliance upon it by that other party to his detriment or to the advantage of the party making it; ICJ, Land and Maritime Boundary (Cameroon v Nigeria: Equatorial
Guinea Intervening), Preliminary Objections, Judgment of 11 June 1998, [1998] ICJ Rep 275,
303, para 57: An estoppel would only arise if by its acts or declarations Cameroon had consistently made it fully clear that it had agreed to settle the boundary dispute submitted to
the Court by bilateral avenues alone. It would further be necessary that, by relying on such
an attitude, Nigeria had changed position to its own detriment or suffered some prejudice.
347
For the following see especially Martin, Lestoppel, n 328 above, 294 et seq; also Kolb, La
bonne foi, n 5 above, 359 et seq with numerous further references.
348
Whereby it forms an exception to most other unilateral acts which may be made implicitly, see p 78 and p 29 et seq.

282 The Law on Promises

As Kolb emphasises, however, repetition will only be one of numerous
means to strengthen the impact of a states declaration or behaviour:
[l]es actes sans assise temporelle doivent compenser par leur solemnit,
leur importance et dautres circonstances labsence continuit.349 A single,
solemn declaration may hence be sufficient.
In respect of the declaration itself, international law does not distinguish between declarations of fact and declarations of law, and will apply
estoppel to both sort of statements made.350 The Court, as quoted, has
indeed applied estoppel to the question whether a state had accepted a
certain legal regime. In order for a state to be estopped by a declaration
or action, the latter will have to be attributable to it. In other words, and
similar to what has been said about promises, for a declaration to have
an estoppel effect, it needs to have been issued by a competent organ. It
must also have been made freely, ie not under illegitimate pressure or
other influences rendering the will expressed defective.351 Under criterion
(2) effective legitimate reliance is required.352 Whether reliance has been
legitimate or not, is, of course, a judgment call and here again an objective analysis will be required along the lines of whether or not a state
could have reasonably relied on the statement made.353 This important
question will be taken up again, when turning to the interplay of estoppel
and promise below.
Criterion (3) is, as seen, also of heightened importance. Under it, the
addressee must, as a result of its legitimate reliance, suffer a prejudice
should the declarant be allowed to alter its position. Whether estoppel
necessarily requires a material damage, or may take into account immaterial damages suffered, is not entirely clear, yet there is a good argument
to be made that a monetary disadvantage is not presently required for
estoppel to be triggered.354 The detriment must in any event not be too
remote. In order for the requirement of detrimental reliance to function
as a true criterion, it is furthermore clear that the discontinuance or nonperformance of the expected (beneficial) behaviour as such is not enough,
because if the non-fulfilment of a declaration that indicated a position
349
Kolb, La bonne foi, n 5 above, 360. The PCIJ in the Serbian Loans case indeed spoke of a
clear and unequivocal representation: it is quite clear that no sufficient basis has been shown
for applying the principle [of estoppel] in this case. There has been no clear and unequivocal representation by the bondholders upon which the debtor State was entitled to rely and
has relied. There has been no change in position of the debtor State, PCIJ, Serbian Loans,
Judgment of 12 July 1929, Series A 20, 4, 39.
350
For this question see Kolb, La bonne foi, n 5 above, 361 et seq; Mller, Vertrauensschutz,
n 89 above, 10, note 22; Elihu Lauterpacht, The Development of the Law of International
Organization by the Decisions of International Tribunals (1976) IV Recueil des Cours de
lAcadmie de Droit International de La Haye 377, 462; also Martin, Lestoppel, n 328 above, 274.
351
See Bowett, Estoppel, n 328 above, 190; Martin, Lestoppel, n 328 above, 27686 and
Kolb, La bonne foi, n 5 above, 37377, the latter with further references.
352
See Kolb, La bonne foi, n 5 above, 36465.
353
For more see Martin, Lestoppel, n 328 above, 28993.
354
For more see Kolb, La bonne foi, n 5 above, 366 et seq.

Promises and Estoppel 283

favourable to another state was itself to serve as the required detriment,

the criterion would lose its value, whereby the difference between the narrow and the extensive understanding of estoppel would be diminished.
The idea is that estoppel will protect the addressee from suffering a detriment as a result of having relied on the statement (detrimental reliance);
in other words, a detriment which it would not have suffered had the
declaration never been made. This is why the declarant state, which chose
to announce its position, is not allowed now to harm the other state by
altering it. What is required is therefore a detriment other than the simple
loss of the advantage provided in the declaration itself.355 Against such a
detriment, the beneficiary is protected through estoppel.
B Promises and Estoppel: Distinct but Interacting Legal Principles
On the basis of the above sketched features, the relation of the estoppel
principle to the doctrine of promises can now be addressed.
(i) Two Distinct Legal Principles
To begin with, it is clear that the position held by Schwarzenberger in
1957, for whom the typical minimum effect of a unilateral act was that of
an estoppel,356 is by now untenable, as otherwise the evolution towards a
general acceptance of the requirement of detrimental reliance for estoppel to be invoked would simply be ignored. Schwarzenbergers analysis
at the time was squarely based on the broad understanding of estoppel
as a general rule of inconsistency.357 Where only some sort of impending prejudice will allow a state to invoke estoppel under the (now dominant) narrower concept of that principle, not every unilateral act (protests
completely aside) can be said to have an estoppel effect. Surely, not every
promise will therefore lead to an estoppel.
For this reason alone, it is also not convincing to go as far as to ground
the binding effect of promises, or that of all unilateral acts, on the estoppel principle. When Gigante did so in his note written in 1969, he, too,
assumed the requirement of detrimental reliance not to be a necessary
prerequisite and a broadened concept of estoppel to be applicable in
See especially ibid 369.
Georg Schwarzenberger, International Law as Applied by International Courts and Tribunals,
3rd edn (London, 1957) vol 1, 553.
357
For Schwarzenberger, whom Kolb characterised as being particularly liberal in the
application of estoppel, see Kolb, La bonne foi, n 5 above, 366, note 67, under estoppel: a
subject of international law is precluded from denying the truth of a statement made earlier
by a duly autorized representative of the existence of a fact in which such representative has
by word or conduct led others to believe, Georg Schwarzenberger, A Manual of International
Law, 5th edn (London, 1967) 631.
355
356

284 The Law on Promises

international law. Only on the basis of this finding, which was supported
by other scholarly writings at the time and some jurisprudence, could
Gigante conclude that a unilateral act, without resulting in detrimental
reliance by the addressee or benefit to the declarant can, nevertheless,
lead to an estoppel.358 At the time of writing, Gigante, however, did not
have the ICJs ruling in the North Sea Continental Shelf cases (issued that
very year) before him, as it does not appear in his analysis of the Courts
jurisprudence; nor was he, of course, able to foresee the Courts and legal
doctrines subsequently clear positioning in favour of the criterion of detrimental reliance.359 Once it is accepted as one of the essential elements
required by estoppel,360 the bindingness of a unilateral act, and especially
that of promises, cannot be based on estoppel, since no such requirement
exists in order for such unilateral undertakings to become binding. In
respect of the binding nature of promises, the ICJs jurisprudence leaves
no room for doubt in this regard, as it considers no reaction of any sort,
let alone any comportment to its detriment, to be necessary on the part
of the addressee: nothing in the nature of a quid pro quo nor any subsequent acceptance of the declaration, nor even any reply or reaction from
other states, is required for the declaration to take effect.361 The Court, in
addition, had at no point argued for the effect of foreclosure or estoppel
in the Nuclear Tests cases. Instead, it understood the declarations to have
resulted in a binding obligation, as they were found to evince their declarants intention to become bound.
While Mller and Cottier in their article on estoppel concede that the
ICJ in the Nuclear Tests cases did not refer to estoppel, they propose that:
estoppel could provide a more specific scheme to explain the binding effect of
these statements. It is suggested that such declarations were intended to induce
a certain conduct by the Court and the other parties, which, if the declarations
were not observed, would be detrimental to the addressees.362

But even this proposal is more than doubtful, because the question
under estoppel would not so much be whether the declarations were
intended to induce a certain conduct, but whether they reasonably could
and actually have encouraged another state to alter its behaviour. No such
alteration, and hence detrimental reliance, was, however, present in the
case before the Court. As Australia and New Zealand considered the declarations made by France as insufficient, they clearly did not make any
additional dispositions (or refrained from taking certain action) in the belief
Gigante, Unilateral State Acts, n 67 above, 351.
See p 279 et seq.
360
ICJ, Land, Island and Maritime Frontier Dispute (El Salvadore v Honduras), Application to
Intervene (1990), n 346 above, 118, para 63.
361
ICJ, Nuclear Tests (Australia v France) (1974), n 6 above, 267, para 43, emphasis added.
362
Mller and Cottier, Estoppel, n 25 above, 118.
358
359

Promises and Estoppel 285

that the statements would be fulfilled. Even with reliance assumed, there
is also no indication that either state would have suffered an additional
detriment as a result of having trusted France, should the latter threaten
to begin another series of tests, other than that their expectations in the
declarations made would have been disappointed.
In this context, a remark on the considerable drawbacks which would
result from estoppel forming the basis of a promises legal effect should
be made. Estoppel is a very strict all-or-nothing principle, as it leads,
once triggered, to (complete) foreclosure. A state which was up until now
free to alter its behaviour is, via estoppel, suddenly hindered completely
from doing so, as contrary claims are ignored; estoppel thereby provides
for no middle ground, nor does it gradually increase the protection of the
addressee by, for example, only temporarily hindering the declarant from
altering its behaviour. For promises, estoppel where applicable will result
in full irrevocability in practice, as every alteration is barred in view of the
detriment caused as the adoption of a contrary position is ignored and
cannot become legally effective. Were the binding legal effect of promises
based on estoppel, they would, as a result, have very different legal features from the ones depicted above: promises would, as such, be devoid
of a legal effect when made, and they could not simply be relied upon
by an addressee.363 Instead, their (then very strict) legal effect would only
be triggered where the addressee alters its behaviour in a particular way
(ie to its detriment). This last aspect underlines estoppels primary raison
dtre which is not so much that of fostering trust in state declarations as
such but to keep imminent damage at bay.364
The ICJ has therefore taken a very different dogmatic route when it
assumed a unilateral pledge to become binding as such and to create an
international obligation which can be trusted by the addressees who are
entitled to require that the obligation thus created be respected.365 The
Court not only refrained in the Nuclear Tests cases from taking recourse to
estoppel, but it has in its subsequent decisions continued to distinguish
between the two lines of reasoning. In none of the judgments in which it
referred to the Nuclear Tests cases in analysing whether a states declaration was a binding assurance has the Court raised the question whether
a detriment had been suffered by the addressee, nor has it used the lack
See also Fiedler, Einseitige Versprechen, n 86 above, 47.
See ibid 48, for whom promises in the ambit of the estoppel principle might receive
additional, validating support therefrom but whose bindingness cannot be wholly based on
that principle: Die Begrndung der Verbindlichkeit einseitiger Versprechen aus Treu und
Glauben oder estoppel mu sich der dogmatischen Folgen und vor allem der praktischen
Auswirkungen bewusst sein. Insgesamt gesehen kann die Lokalisierung des vlkerrechtlichen Versprechens in der Einwirkungszone von estoppel nur eine ergnzende, flankierende Absicherung bewirken, nicht jedoch die Verbindlichkeit vlkerrechtlicher Versprechen
selbst umfassend belegen.
365
ICJ, Nuclear Tests (Australia v France) (1974), n 6 above, 268, para 46.
363
364

286 The Law on Promises

thereof in order to explain its finding that no obligation existed. In all
these cases, its jurisprudence on the principle of estoppel was not referred
to in establishing the legal principles involved; and vice versa the same is
true: in none of the above-cited judgments which have addressed estoppel is a reference to the ICJs own jurisprudence on promises included.
And rightly so, as they are, as seen, distinct in their legal features.
The fact that most of legal doctrine366 has followed the Court by not
taking recourse to estoppel in order to explain a promises binding force
is, for the above-mentioned reasons, therefore to be supported. Two questions, however, remain to be answered: the first is whether a promise can
nevertheless lead to estoppel, provided that the latters prerequisites are
met; and the second is whether a declaration which indicates a states
future action, but is not a promise under the interpretative standard
described above, may nevertheless estopp the declarant from later changing its behaviour.
(ii) Promises Triggering Estoppel
Turning to the first question, whether a declaration that is held to be a
promise can also lead to an estoppel, might at the very outset trigger the
counter question whether this inquiry is not really superfluous. Since
promises are already binding undertakings, what would an additional
estoppel effect have to add? Fitzmaurice in his separate opinion in the
Temple of Preah Vihear case stressed that where an obligation exists, there is
no need for estoppel, in the following words:
However, in those cases where it can be shown that a party has, by conduct
or otherwise, undertaken, or become bound by, an obligation, it is strictly not
necessary or appropriate to invoke any rule of preclusion or estoppel, although
the language of that rule is, in practice, often employed to describe the situation. Thus it may be said that A, having accepted a certain obligation, or having
become bound by a certain instrument, cannot now be heard to deny the fact,
to blow hot and cold. True enough, A cannot be heard to deny it; but what this
really means is simply that A is bound, and, being bound, cannot escape from
the obligation merely by denying its existence. In other words, if the denial can
366
As portrayed at p 201 et seq, the vast majority of scholars assume a promises bindingness to be based on good faith with no reaction and no detrimental reliance being required
on the side of the addressee. Estoppel is thereby at least implicitly rejected. Expressly rejecting the understanding that the legal bindingness of promises might be based on estoppel,
see, eg Fiedler, Einseitige Versprechen, n 86 above, 4748; Jacqu, Promesse unilatrale,
n 97 above, 339. Even though the development in legal doctrine towards the requirement
of detrimental reliance, as driven by the Courts more recent and now consistent jurisprudence, is not mentioned, see also Goodman, Acta Sunt Servanda?, n 67 above, 6264. For
the Special Rapporteur, who also distinguished the two concepts, and various comments
made in the Commission, see the references provided in n 324 above. Contra Bollecker-Stern,
Essais nuclaires, n 89 above, 330.

Promises and Estoppel 287

be shown to be false, there is no room or need for any plea of preclusion or
estoppel.367

This analysis by Fitzmaurice is to the point, as long as the effect of the

existing obligation can be said to be the very same as that which would also
flow from estoppel. If so, there is indeed no need to raise estoppel. Where
a state is under a treaty obligation to follow a certain course of conduct,
there is nothing to be gained from invoking estoppel should it now intend
to take a position inconsistent with that obligation. It is enough to point to
the obligation itself. The same can be said where a state has issued an act of
recognition: the legal effect of that unilateral act needs no additional support from estoppel. But for promises, the picture might be different, as the
latter, unlike acts of recognition or the standard treaty obligation, can be
revoked within the above elaborated framework. As this is the case, invoking estoppel in respect of a promise is not necessarily a senseless undertaking as it would then entirely preclude a state from altering its behaviour.
The additional legal effect due to estoppel protecting the addressee from
suffering any detriment would thereby effectively hinder the revocation of
the obligation undertaken. Estoppel would hence not simply be invoked in
order to underline an already existing obligation but, in effect, to limit the
possibility of its revocation to the advantage of the addressee.
The question whether a promise may estopp the declarant is hence not
merely of academic interest but may result in a very palpable outcome for
the declarant as well as its addressee(s). Checking the estoppel requirements in respect of a declaration that has been identified to constitute a
promise will reveal the following. In order for a declaration to later estopp
a state, it has been said that it must be clear and unequivocal.368 Promises
in order to be qualified as such will already have met the very same
threshold (as declarations need to be clear and specific),369 and there will
consequently be no obstacle posed in this respect. It has already been said
that in international law not only statements of fact but also those of law
are capable of triggering estoppel,370 and importantly in this respect, that
the international legal concept of estoppel is understood as capable of
being triggered by declarations addressing the future conduct of a state;371
367
ICJ, Temple of Preah Vihear, Preliminary Objections, Separate Opinion Fitzmaurice
(1961), n 332 above, 63. See also Jacqu, Promesse unilatrale, n 97 above, 337, who, in
following Fitzmaurice, hence considers: Ds lors, si lauteur de la promesse a clairement
entendu se lier, point nest besoin dinvoquer lestoppel, il suffit que le destinataire tablisse
la ralit de lengagement pour lui faire produire des effets.
368
See especially Martin, Lestoppel, n 328 above, 274 et seq; Kolb, La bonne foi, n 5 above,
360 et seq, with further references in his note 26. Also Bowett, Estoppel, n 328 above, 18485.
369
See p 218 et seq.
370
See n 350 above.
371
According to Martin, Lestoppel, n 328 above, 274: la question de savoir si pour reprendre la terminologie anglaise lestoppel ne peut resulter que dune representation ayant
port de faon dtermine sur des existing facts, et non dune dclaration dintention, telle
une promesse de faire quelque chose, qui vise des vnements futurs, donc incertains, fort

288 The Law on Promises

promises are regularly referred to as potentially leading to an estoppel
by scholars372 and the ICJ has not hesitated to take declarations as to a
states future conduct into account in its test of whether estoppel could
be invoked in the case before it. In the Land and Maritime Boundary case
between Cameroon and Nigeria, for example, the Court considered that:
an estoppel would only arise if by its acts or declarations Cameroon had consistently made it fully clear that it had agreed to settle the boundary dispute
submitted to the Court by bilateral avenues alone. It would further be necessary
that, by relying on such an attitude, Nigeria had changed position to its own
detriment or suffered some prejudice.373

As seen, the Court in the Frontier Dispute case between Burkina Faso
and Mali was faced with a similar scenario. In the latter, the question was,
however, whether Malis Head of State had accepted the outcome of the
Mediation Commissions decision as binding by its unilateral declaration, and was now compelled to follow through with the promise made,
and not merely barred from ignoring its declaration as a result of estoppel. The question which is raised by employing estoppel in relation to
such declarations (whereby it also appears to lose some of its merely and
strictly shielding character), is hence that of the interplay between the two
principles. Once a declaration is held to be a promise, it will, as seen, have
overcome the initial hurdle for a state to be subsequently estopped from
altering its behaviour, as it will have made a clear and unequivocal declaration announcing its future actions.374 Whether estoppel is triggered by a
promise or not will consequently come to hinge on whether the addressee
has reacted to its own detriment in reliance on the declaration made. Yet,
and importantly, the question asked here is not simply a factual one, ie
whether or not there has been a reaction that will prejudice the addressee.
Instead, a judgment call is introduced at this point, as the addressees reliance has been said to have to be legitimate. The question, in other words,
controverse dans les pays de la common law (problme du promissory estoppel), ne semble
gure avoir retenue lattention au plan international.
372
See especially the authors mentioned in n 325 above; also Martin, Lestoppel, n 328
above, 258, note 7, in which Martin agrees with Jacqu, Elments, n 221 above, 256, that
promises might become irrevocable where they lead to an estoppel, as held also by Mller
and Cottier, Estoppel, n 25 above, as cited in the text to n 362 above. It should in addition
be recalled that the ILC in paragraph 3 of its Commentary to Guiding Principle 10 on revocability cites the ICJs passage on estoppel from the Military and Paramilitary Activities in and
against Nicaragua case in explaining a scenario that will hinder revocation, see pp 26061.
373
ICJ, Land and Maritime Boundary (Cameroon v Nigeria), Preliminary Objections (1998),
n 346 above, 303, para 57. The ICJ, in the North Sea Continental Shelf and Military and
Paramilitary Activities in and against Nicaragua judgments in the passages cited above, has
similarly analysed whether state declarations had signalled the acceptance of a particular
(legal) regime and had thereby triggered estoppel. For more see the text to nn 344, 345 above.
374
For the impact of a finding that a declaration is not a promise on the application of the
estoppel principle, see p 290 et seq.

Promises and Estoppel 289

is whether the declaration was of such a nature as to reasonably lead the

addressee to alter its behaviour as it did.375
In now assessing whether the addressee has acted reasonably or not,
the framework applicable to promises cannot simply be ignored, where
the declaration is one in which a state assures that it will comport itself
in a certain manner in the future. For a statement which constitutes a
binding promise, the addressee necessarily has to take into account that
according to the legal framework operable in that area, a promise can be
revoked where a state has given reasonable notice of its intention to do
so. If this is the case, the recipient of a promise cannot therefore simply
make far-reaching arrangements to its future detriment in reliance on the
perpetual existence of the promise as such. Should it decide to do so and
thereby ignore the legal framework attributable to promises, it cannot
later invoke estoppel to protect itself, as the amount of reliance placed on
the undertaking would not be legitimate. The evaluation of the addressees behaviour in this regard would be similar to where, for example, a
treaty over a lease of territory has been signed between states A and B,
which allows for it to be terminated with 12 months notice. Should state
B, in possession of the lease, rent the territory or property thereon out to
a private contractor for a longer fixed period, it will not be able to shout
estoppel in its defence, should the lease be terminated by state A according to its terms. That state B will suffer damage (as it will breach its lease
with the private contractor), because it has altered its behaviour (rented
out the property) in reliance on the continuing existence of the lease, will
not help, as its reliance on the obligation continuing was not legitimate.
That the addressee may not ignore a promises revocability, however,
does not alter the fact that the declarant has announced a certain future
behaviour and even compelled itself to follow through with it by creating
a legal obligation. Whether the addressee has acted reasonably in altering
its position will, apart from the limits imposed by the legal framework,
therefore depend primarily upon the nature of the action announced.
Where the pledged future conduct (rather than the belief in its perpetual
existence) will, as such, have a considerable effect on the addressee (its
territory, population, wildlife, financial assets, etc), the latters immediate
and potentially detrimental shift in position may very well be reasonable.
Where this is the case, it will, as such, also be foreseeable for the declarant, who consequently cannot be considered to be unduly burdened by
subsequently facing a claim of estoppel were it to announce an alteration
of behaviour. In these cases it is, as seen, the damage that the addressee
would otherwise suffer by having reasonably relied on the undertaking
which will allow it to invoke estoppel.
375

For more see Martin, Lestoppel, n 328 above, 286 et seq.

290 The Law on Promises

However, it should be clear that on account of its rather far-reaching
effect, estoppel will only be triggered as a last resort, as the reasonable
notice period introduced by the ICJ for revoking a promise functions as
the first safety valve in protecting an addressee from actually suffering
any damage. As described above, the word reasonable has been said
to point towards the 12 month period laid down in Vienna Convention,
Article 56(2). Yet, understood as a minimum requirement, a reasonable
period which provides the addressee with the possibility to readapt
may of course be considerably longer, depending on the circumstances.
Before a state is therefore completely barred from adopting a new position
via estoppel, it must be ascertained whether a longer revocation period
would not adequately protect the addressee by providing it with more
time to alter its position in the face of an intended revocation. Only in (the
probably rather rare) scenarios where the detriment cannot otherwise be
avoided, will an undertaking therefore become de facto irrevocable for the
declarant (at least unilaterally)376 under the estoppel principle. But even in
these cases, there is no reason not to provide the declarant with the possibility to revoke its undertaking provided that the damage suffered by
the addressee can be and has been compensated by the declarant. Should
it decide to do so, the detriment the addressee was facing would vanish
and, with it, one of the underlying elements that trigger estoppel. As a
result, the obligation could therefore again be revoked within the normal
promise framework (ie a reasonable notice period provided).
Estoppel therefore does not simply steamroll over a promises legal
effect by applying to every promise made while ignoring the legal framework established above. Instead, it will have its additional protective
effect only where it is required as an ultimate last resort, an ultimate safety
net in place to protect the addressee from suffering uncompensated damage as a result of having legitimately relied on the undertaking made. The
reference to it in the Commentary to the ILCs Guiding Principle 10,377 just
as its express inclusion in the altered version of that Principle on the revocability of promises as proposed above, should be understood as such.378
(iii) Non-Promises Triggering Estoppel?
What remains to be addressed is the impact which a finding that a declaration announcing the future action of a state is not a promise will have
on the assessment whether it may later estopp the declarant. Mller and
Cottier seem to be saying that it is in particular where an intention to be
A consensual cancellation is, of course, possible.
See ILC, Guiding Principles with Commentaries, n 46 above, 38081, Commentary to
Guiding Principle 10, para 3, referred to and cited at p 260.
378
See p 275.
376
377

Promises and Estoppel 291

bound cannot be established that the estoppel effect can be of importance,

when they write:
Of course the doctrine of estoppel may be dispensable if the intention to be
bound seems to be well established in a particular case. However, it is usually
very difficult, if not impossible, to prove intention as a subjective element of a
partys behaviour.379

For Jacqu, too, recourse to estoppel is possible where the intention of a

promises author is uncertain.380 And indeed, at first it appears to be only
logical that since estoppel does not make any reference to a states manifested intent to be bound, a finding that the latter is lacking should be of
little impact when applying estoppel to the case at hand. Yet, in light of
the way in which a states manifestation of will is ascertained when determining whether it has made a promise or not, this conclusion appears to
be inaccurate. To return to the quotation from Mller and Cottier as provided above, the finding in its last sentence that proving intention is usually
impossible, is certainly true. But the standard of interpretation applied to
declarations of states in order to assess whether they are promises or not
has shown (and precisely for the reason that the addressee does not know
what the declarant really intends) that this proof is not actually required.
It is instead a reasonable outside evaluation that will be conducted by
the addressee, just as by a court, in determining whether a declaration
may be relied upon as a legal undertaking. As will be recalled, under such
an objective (ie outside) analysis, it will have to be ascertained whether,
under a restrictive interpretation, a declaration of a state constitutes a
clear and specific indication of its future actions that is supported by circumstances which emphasise the declarations importance, before allowing for the latter to be interpreted as a legal undertaking. Only where this
is the case will a state be understood as having manifested an intention to
be bound by its statement. If this analysis comes to the conclusion that the
declaration in question was too vague, or the circumstances of its making
indicated that no legal undertaking can be assumed to exist, the result will
be that the statement in question was, at most, a political sign of goodwill.
The question raised here is therefore whether such statements, while not
promises, can nevertheless have an estoppel effect on their declarant.
While it is true that estoppel does not ask whether a state intended to be
bound, it does, however, enquire whether a state has legitimately relied on
a declaration. But, and solely in reference to declarations which indicate a
states future(!) behaviour (ie potential promises), this means that the very
same questions as those which have already been raised and answered
when ascertaining whether a declaration qualified as a promise or not,
379
380

Mller and Cottier, Estoppel, n 25 above, 118.

Jacqu, Promesse unilatrale, n 97 above, 337.

292 The Law on Promises

must arise. The question is not simply whether or not the addressee has
relied on a declaration made to its detriment, as the fate of the declarant
would then be placed fully in the hands of its addressee. Instead, it asks
not only whether or not the addressee altered its position in response to
the declaration, but in addition whether this reaction by the addressee
was reasonable, ie whether it can be called legitimate. As already seen
above when focusing on the basis of a promises bindingness, under a
trust or reliance-centred approach, the judgment call has to be made as
to whether a states declaration was trust-worthy: could the declaration
by state A indicating its future behaviour be legitimately taken by state B
as a ground for altering its behaviour? It is therefore no surprise that for
estoppel to apply (and within the literature on estoppel), all the questions
which needed to be addressed within the promise analysis spring back
to life: Was the state representative competent?381 Did he or she act under
pressure, was there fraudulent conduct, ie was the declaration possibly
defective?382 And especially: was the declaration clear and unequivocal
in respect of the future conduct which it announced?383 But if here a different standard was to be applied than when evaluating whether or not a
statement was clear and specific under the promise analysis, the result
would lead to a striking incoherence in the legal system. In effect, the
widely accepted and relied upon distinction between binding legal commitments and non-binding declarations (of intent, goodwill, etc) would be
circumvented by estoppel, as the distinction would be rendered devoid
of much of its importance if via the back-door of estoppel any statement
could become de facto binding. The answers provided to the reliance
questions now dressed in estoppel clothing will consequently have to be
the very same as when they were posed within the promise context. If
declarations, which indicate a states future and beneficial behaviour to
an addressee, have in the promise context been considered as not meeting the adequate threshold and hence not worthy as being relied upon
as legal undertakings which determine a states future actions, the same
result will have to be applied here. The addressee, strictly legally speaking, cannot legitimately base itself on such non-promise indicators of a
states future behaviour and alter its position accordingly. Should it do
so, it will not be able to claim estoppel vis--vis the other state, which has
merely indicated what it considered it would (possibly) do.
While the assessment conducted under estoppel vis--vis declarations
which indicate a states future actions will necessarily run parallel to that
conducted when concluding that a declaration was not clear and specific
enough to be a promise, the matter is different for declarations in which
381
For estoppel and this question see the analysis by Martin, Lestoppel, n 328 above, 276
86 and Kolb, La bonne foi, n 5 above, 27686.
382
See the text to n 351 above.
383
See text to n 368 above.

Promises and Estoppel 293

a state analyses its own past behaviour. It is here where the primary field
of application of estoppel to non-promises will lie. If a state makes its
interpretation known to another state, then these additional, and subsequent remarks, made after a non-promise has been made, might become
important under the estoppel principle. In this respect, the same is true as
for non-treaty obligations. The above citation of Fitzmaurices opinion in
the Temple of Preah Vihear judgment continues by making this point:
Such a plea [of estoppel] is essentially a means of excluding a denial that
might be correct irrespective of its correctness. It prevents the assertion of
what might in fact be true. Its use must in consequence be subject to certain
limitations. The real field of operation, therefore, of the rule of preclusion or
estoppel, stricto sensu, in the present context, is where it is possible that the
party concerned did not give the undertaking or accept the obligation in question (or there is room for doubt whether it did), but where that partys subsequent conduct has been such, and has had such consequences, that it cannot be
allowed to deny the existence of an undertaking, or that it is bound.384

But here it is not the non-obliging declaration itself that therefore triggers estoppel, rather it is the declarants subsequent own conduct or its
representations judging its own, past behaviour that will do so. Taking the
Armed Activities on the Territory of the Congo case as a hypothetical example
to illustrate this point, it means the following. It will be recalled that the
declaration made by Rwanda to withdraw all its reservations was considered by the ICJ to be, at most, a declaration of intent, very general
in scope,385 which could not be understood as a legal commitment. Had
the Democratic Republic of the Congo now claimed to have detrimentally
shifted its position in reliance on this statement, this reliance alone would
not have been enough to estopp Rwanda from subsequently deciding to
withdraw its declarations; the reason being that the question whether the
DRC was able to rely on this statement in determining its future actions
must be answered in the negative. The announcement was too vague to
base trust on it. If otherwise assumed, the whole analysis of whether a
statement is a binding undertaking or not would be undermined, as a
result of which the declarant would be in the constant danger of having to follow statements which were not legal undertakings but rather
announcements of intent, very general in scope. However, had Rwanda,
after its declaration, continuously referred to or otherwise by its conduct
made it clear that it considered its declaration as having given rise to an
obligation to withdraw its reservations, then this subsequent conduct could
have given rise to an estoppel, barring Rwanda from suddenly adopting
384
ICJ, Temple of Preah Vihear, Preliminary Objections, Separate Opinion Fitzmaurice (1961),
n 332 above, 63, emphasis added.
385
ICJ, Armed Activities on the Territory of the Congo, Jurisdiction and Admissibility (2006),
n 40 above, para 52, for more see p 169 et seq.

294 The Law on Promises

the opposite position of asserting that such an obligation had never been
undertaken, given, of course, detrimental reliance on the part of the DRC.
This, however, would involve something other than applying an estoppel
effect directly to the declaration made, which was held not to be a promise
itself.
(iv) Recapitulating Remarks on the Interplay of Promise and Estoppel
As seen, the principle of estoppel has not only a legal effect but also prerequisites of its own, as a result of which a promises bindingness is not
based on estoppel, nor will each and every promise have an estopping
effect for the state making it. Their area of application may, however,
overlap and a promise can therefore also estopp the declarant if the pre
requisites for estoppel are fulfilled. The undertaking will thereby become
de facto irrevocable. But as the addressee has to take the general revocability of promises into account, and is protected first and foremost from
suffering damage via a reasonable period of prior notice before a revocation is effective, estoppel will only be triggered as an ultimate last resort
and in cases where impeding damage cannot otherwise be deflected and
has not been compensated. Where a declaration which indicates a states
future action is held to be a mere declaration of intent or goodwill, it cannot as such legitimately be taken as a reliable announcement of future
action, let alone one on which reliance may be placed in a form that, if
disappointed, will cause additional detriment. Any such behaviour by the
addressee will hence not bar the declarant from later altering its position.
Where a non-binding announcement has been made, however, a state can,
by its subsequent conduct, of course still trigger an estoppel effect.
IX SUMMARISING THE LEGAL FRAMEWORK

Before continuing to chapter four which provides a brief outlook and

some thoughts on the future of promises as a legal institution, it may be
helpful first to summarise the findings included in this chapter on the
Law of Promises. It is, after all, on this basis that the usefulness of promises as a tool for a state to commit itself will be addressed.
As will be recalled, the present chapter began by illustrating that promises are indeed missing from the list in ICJ Statute, Article 38(1), which
is commonly understood as listing the sources of international law.
Its silence in this regard, however, has been shown to have prevented
neither the ICJ nor scholars (or most of the states which answered the
ILCs questionnaire) from very clearly asserting that states can undertake obligations through unilateral pledges. The cases depicted in chapter
two illustrated that states indeed make use of unilateral assurances, the

Summarising the Legal Framework 295

capacity of which to create legally binding obligations is not doubted. The

lacuna in the Statute notwithstanding, the ILCs conclusion that states currently possess the capacity to undertake legal obligations through unilateral declarations was therefore found to adequately reflect international
law as presently applied by courts and tribunals, just as it is understood
by practitioners (including, but not limited to, those arguing before the
ICJ) and the overwhelming majority of scholars, alike. Even without the
Courts Statute being revised to reflect legal realities, the lacuna presented
by promises not being listed amongst the sources of law might fade away
over time, should the principles which have primarily been established by
the World Courts case law continue to be accepted and thereby entrenched
as customary international law through the practice of states.
A closer analysis of the ILCs Guiding Principles, based primarily on
the ICJs judgments in the Nuclear Tests cases, revealed that various and
indeed central aspects were left unclear in the Commissions final version, thereby severely hampering the Principles capacity to be truly guiding. Some of these aspects have already been addressed in chapter one
of this study when determining the nature of a promise, which necessitated in particular identifying the features of a declarations unilateral
character. The present chapter has focused on open questions, including
in particular the precise basis for a promise to be binding, the standard of
interpretation to be applied to unilateral pledges of states, the impact of
the addressees reaction to the undertaking proclaimed, general and additional requirements for a promise to be binding and, last but not least, the
revocability of promises and the impact the principle of estoppel might
have in this area of law.
In so doing, the identification of the principle of good faith as explaining the constraining force and hence a promises bindingness for the
declarant has been confirmed (in rejecting in particular an argument based
on a states sovereign decision or intent to commit itself), while further
arguing that this implied that the reliance of an addressee on the promise
received is protected. This reliance does not have to be expressed by the
state to which the promise is directed, as this chapter has gone on to argue
that it is presumed to exist, apart from the addressees indications to the
contrary. The legal framework thereby adequately reflects the social interaction which has taken place between the states involved: the addressee is
presented with what is (for it legally speaking) always an advantageous
declaration as through it an additional right is acquired, invokable against
the declarant. Since the declarant does not ask for any positive reaction in
the form of the addressees acceptance, no such reaction can be expected
from the addressee, who will nevertheless usually take note of the new
undertaking and rely upon it in its future actions. This presumption of
reliance, of course, still allows the addressee at all times to rebut it and
reject the gift made.

296 The Law on Promises

The follow-up question triggered by this finding was to ask on which
declarations a state may in fact rely as a binding undertaking. In other
words, in which scenarios will a states reliance on a declaration be protected as legitimate? It is here where the distinction of political and legal
undertakings has been said to come into play and with it the focus shifted
to what can be interpreted as a states manifestation of will to commit
itself legally. Judged from an outside perspective of a reasonable observer,
the analysis will use a restrictive interpretation and concentrate on the
text of a declaration and its surrounding circumstances. A state will need
to use clear and specific wording in announcing its future course of
action. And where this is the case, a declaration will, in addition, need
some sort of supporting context underlining its importance, in order for
it to constitute a manifestation of will to be legally bound. However, the
clearer the wording in respect of a legal obligation having been intended,
the less the requirements in this respect will be. Where, as is usual, a state
does not expressly flag its announcement as including a legal commitment, the circumstances will be of considerable importance in order to
assess whether a declaration is a mere political pledge and sign of goodwill, or a legally binding promise. As seen, the courtroom setting has to
be understood as putting state representatives on notice that their spoken
word will be relied on as a legal assurance. Declarations made in direct
reference to ongoing proceedings will in addition, as the ICJ has said, not
be considered in vacuo but as being made also for the ears of the Court
and hence as reliable, ie legal, commitments. Solemn announcements,
official notifications or the fulfilment of additional formalities (be it even
the registration of a unilateral declaration with the United Nations) will,
for a reasonable observer, also be taken as stressing the heightened impact
a state intends its declaration to have. Coupled to a clear and specific
statement, even where restrictively interpreted, the result will hence be a
legal obligation undertaken in the form of a promise. Where this is so, a
promise may therefore be invoked against the declarant before legal institutions and will trigger the latters responsibility (for compensation or
giving rise to countermeasures) should it be breached.
The study moved from there to making it clear that the declaration will
only be a promise where an acceptance is not requested by the declarant, which necessitated a closer look at the differences in law and in the
declaration itself (as an act of communication) between an offer and a
promise. It moved on to identify a couple of further requirements which,
once a manifestation of will to enter into a legal commitment has been
found, need to be fulfilled. A promise will, as such, need to be made by a
competent representative and, in order not to be void or voidable, must
not suffer from any defect. In order to become binding, a promise has
to be communicated to the addressee, a requirement which, in fact, is
an integral part of the manifestation of a states will to undertake a legal

Summarising the Legal Framework 297

obligation vis--vis the addressee. Although frequently and misleadingly

referred to as requiring a publicly made declaration, the intentional act
of communicating the promised course of conduct to the addressee is,
however, sufficient, even if made behind closed doors. Other than that,
an undertaking has to be legal in that it may not infringe a norm of ius
cogens without becoming void. (It has in this respect been proposed that
the same should be true for previously and openly adopted legal obligations of special importance, which allow for no contradicting subsequent
obligations to be valid, as the damage caused cannot simply be compensated by means of state responsibility.)
The study continued by pointing out that the registration of a unilateral declaration is by no means a requirement in order to invoke the latter
before a UN institution, as UN Charter, Article 102 will not apply; but
that such registration might, nevertheless, be important in the aboveaddressed assessment as to whether a state has manifested an intention
to be legally bound by making its declaration. In assessing the impact
of the addressees reaction, its silence has been identified as being of no
consequence for the declarations bindingness. A positive reaction will
usually merely underline and manifest the addressees actual reliance,
which has already been presumed to exist and will, as such, be of separate
and heightened importance only in the exceptional case where it triggers
estoppel. Where the addressee, however, chooses to reject the promise
made, the latter will no longer be compelling for the declarant; with no
reliance being placed on the declaration made, there is nothing left to protect under the good faith principle.
Having established the requirements for a binding legal undertaking
in the form of a promise, this led to the question whether a promise may
be revoked, which has, as seen, been answered affirmatively. A state can
revoke its promise, but the expectations raised in the addressee that a state
will comport itself in a certain manner need to be respected and protected.
Special terms as to revocation included in the declaration and a fundamental change in circumstances aside, international law has been said to
protect the faith placed on an undertaking by providing the addressee
with time to readapt to the new legal situation. It therefore requires the
declarant to provide reasonable notice of its intention to revoke the
undertaking made. In mirroring Vienna Convention, Article 56(2), to
which the ICJs jurisprudence points, the minimum requirement in this
respect, as a rule of thumb, appears to be at least a 12-month delay. Where
the dispositions which the addressee has made have been reasonable as
well as considerable, the reasonable notice period may, however, require a
longer readaptation period for the addressee. Only as a last resort, where
the resulting damage or disadvantage for the addressee legitimately relying on a declaration that has encouraged it to act cannot be compensated
by the declarant, will a promises revocation become de facto impossible

298 The Law on Promises

under the safety net provided by the estoppel principle. The regime in
place for revocation thereby carefully balances the two interests at stake.
This, in a nutshell, is what this chapter has presented the Law on
Promises to be. Its aim was not to move towards new concepts to be
implemented de lege ferenda, but to establish the present legal framework
by using the ICJs dicta and the strong support of the latter as reflected in
the ILCs Guiding Principles as guidance, while adhering to an interpretation of systematic coherence. It is in respect of the revocability of promises
where the territory appeared the most uncharted. The above-presented
approach is, however, backed by more than merely (good) scholarly arguments, but as seen supported in particular by the ICJs interpretation of
good faith as establishing the limits on the revocation of unilateral declarations under the Optional Clause in direct reference to the Nuclear Tests
cases. The interplay between promise and estoppel, to which Guiding
Principle 10 in light of its commentary points, in addition, can only be
taken duly into account where a more flexible approach for the revocation
of promises is otherwise followed. It is on the basis of these legal features
that the future of promises as a legal institution will be briefly addressed.

4
Looking Ahead: A Promising Future?

HE INTRODUCTION TO this book found that while promises

are commonly mentioned amongst unilateral acts in public inter
national law textbooks,1 scholars who have attempted to elaborate
the precise legal rules governing such undertakings are far from numer
ous. The primary explanation for this lack of interest has been said to arise
from the understanding that binding unilateral assurances of states, ie
promises, are a rare phenomenon in international law.2 The assumption
that states have no motive to undertake obligations by unilateral means
has already been rejected as unfounded,3 but the underlying bilateral
mindset expressed in this finding appears to have multifarious roots,
which explains why it has remained difficult for the doctrine of promises
to develop further on the international plane. It is likely that the fact that
promises are not provided for in most municipal legal systems will (even
if only subconsciously) have had an impact on international lawyers, who,
after all, are usually professional hybrids, equally at home in at least one
municipal legal system. Their understanding in such a fundamental mat
ter of how legal commitments between subjects of law can be construed
is therefore likely to have influenced their perception of binding under
takings at the international level. The fact that the dogmatic construction
of unilaterally obliging assurances within international law is relatively
new only enhances this effect.
As the History of Promises has shown, the early part of the twenti
eth century has seen a number of rulings in which international courts
relied on unilateral assurances as giving rise to binding commitments,
and there is state practice coupled to scholarly support from this time.
However, it was not until the ICJ in the Nuclear Tests cases set out its posi
tion in crystal clear words that the existence of binding promises of states
could no longer be (successfully) marginalised as a misguided minor
ity interpretation of cases and state practice within the legal profession.
Although the Courts holding, along with its subsequent decisions that
refined the doctrine elaborated therein, have had an impact on both state
See those listed amongst the scholarly support, Introduction, n 49.
See p 4 et seq.
3
Ibid.
1
2

300 Looking Ahead: A Promising Future?

practice and scholarly writings, the judgments nevertheless do not appear
to have unleashed the ground-breaking effect a true Marbury v Madison
international equivalent4 would have been expected to provoke. One of
the principal answers to the question as to why they failed to do so is to be
found in the fact that many features of the legal framework applicable to
promises appeared to remain in the dark. Particularly as the uncertainties
reached into areas of pivotal importance (for example, revocation), prom
ises have remained capable of prompting sceptically raised eyebrows,
even in debates amongst public international lawyers.
However, as more and more of the question marks with which the topic
is peppered are deconstructed, this critical perception is likely to con
tinue to change, and states, for the reasons mentioned in the Introduction,
might very well increase their use of promises. In other words, the more
the Courts jurisprudence continues to grow and the consensus on the
existence of state promises as unilaterally obliging acts comes to include
also the precise legal framework applicable thereto, the more likely it is
that the use of such assurances by states will intensify. What this chapter
will therefore address is the impact such a development might have on
the international plane. The following pages therefore provide a brief out
look and some thoughts on the prospects, as well as the (alleged) dangers,
connected to promises, should they indeed become a more prominent tool
in inter-state relations.
I STRAITJACKET VS EMPOWERING RULE

When the ILC set about its work on unilateral acts, Simma warned his
peers about creating a straitjacket for states by adopting guidelines which
deprive a unilateral act of its flexibility, and thereby states of the flexible
instrument said to be currently at their disposal.5 Meant as a warning for
an attempt at codification, this caveat can be directed at any develop
ment in the area of promises. Whether it is more jurisprudence in this area
or merely a chapter in a book such as this one, which strives to enhance
the understanding of promises legal features, the warning points to the
practical consequences which the application of legal rules to unilat
eral pledges will have, be they newly developed or only elucidated as
already existing. Although the term is rarely used, this straitjacket con
cern appears to be looming in the background and, where this is the case,
will fuel scepticism in respect of any law regulating unilateral assurances
4
Thomas M Franck, Word Made Law: the Decision of the International Court of Justice in
the Nuclear Test Cases (1975) 69 American J International Law 612.
5
ILC, Summary Record of the 2524th Meeting, UN Doc A/CN.4/SR.2524 (1998) 41, para
82.

Straitjacket vs Empowering Rule301

of states.6 If a state really has to recognize that what it solemnly says it

will do, or more importantly, what it says it will not do, becomes a part
of that trellis of reciprocal expectations on which the fragile international
system grows7 (to quote Thomas Franck), the result might be that states
simply refrain from making pledges on the international level.8 Were this
true, the evolution which has elevated pledges outside treaties from mere
indicators of goodwill to binding commitments would hence rather ham
per than improve international cooperation. The perception that the doc
trine of promises might function as a millstone around a politicians neck
when acting on the international stage may, in part, also be motivated by
a feeling that promises are a dangerous legal device,9 a loose cannon,
which can be turned against an unwary declarant who may not see the
blow coming. Promises could function rather as a trap than a useful tool,
and one which can close on a state should it not manoeuvre in the inter
national arena with exceptional care. But while this warning is to be taken
very seriously and into consideration, particularly with regard to any pro
gressive development in this field, and it is both true and important to
emphasise that states need to weigh their words carefully, it would be a
misconception to view the law which is presently in place and described
above to already function as a straitjacket. Were such a position adopted,
it would have to ignore numerous factors while simultaneously basing
itself on some false assumptions.
To begin with, it would be false to address promises on the assumption
of an existing dichotomy between a (hastily made) oral unilateral pledge,
on the one hand (ie the promise), and a (carefully drafted) written treaty
on the other, because there is a lack of formal requirements for treaties, just
as for promises, and promises might, of course, be made in written form:
the Suez Canal declaration, the Austrian notification of its neutrality, the
declaration on the use of war material vis--vis Austria and some of the
negative security assurances circulated in the General Assembly were all
made in written form. And just as a unilateral promise can deliberately
be issued in writing, so can treaties be concluded orally. The argument
6
See Wilfried Fiedler, Zur Verbindlichkeit einseitiger Versprechen im Vlkerrecht
(1976) 19 German YB International Law 35, 62: Unausgesprochen im Hintergrund steht
daher die Befrchtung, durch allzu starre Verrechtlichung internationer Beziehungen deren
Anpassunsfhigkeit zu mindern und die zwischenstaatliche Kooperation entscheidend zu
hemmen.
7
Franck, Word Made Law, n 4 above, 616.
8
Rubin, for example, in the direct aftermath of the Nuclear Tests cases judgments warned
that states might be discouraged from publishing their intentions as they could be miscon
strued, see Alfred P Rubin, The International Legal Effects of Unilateral Declarations (1977)
71 American J International Law 1, 30.
9
Christian Tomuschat, Unilateral Acts under International Law in Droits et Culture.
Mlanges en lhonneur du Doyen Yadh Ben Achour (Tunis, 2008) 1487, 1496, has referred to
unilateral declarations as a dangerous legal device, especially as [d]ecisions may be taken
lightly, without any of the protective mechanisms of treaty-making becoming operational.

302 Looking Ahead: A Promising Future?

that arose in particular in respect of the declaration by Minister Ihlen in
the Eastern Greenland case, as will be recalled, has not focused on whether
the undertaking was unilaterally binding vs bilaterally non-binding. The
bindingness resulting from these oral declarations was accepted as such,
and the question was merely whether the declaration was an oral offer
that had been accepted or whether it was an oral promise, binding as
such. The perceived danger for a state of hastily issuing a binding decla
ration which will give rise to a far-reaching legal obligation is, therefore,
already present on the international level, as oral offers are accepted.
It is, however, true that oral declarations have played a more important
part in the evolution of the doctrine of promises than they appear to have
occupied in the area of treaties.10 The danger for states of assuming an
obligation orally could hence be considered (at least) heightened by the
doctrine of promises which allows even an assurance not asking for an
acceptance to give rise to a binding obligation. Yet, any such argument
would, of course, have to assume that the legal profession would other
wise reject an assurances bindingness even where the latter is solemnly
made in clear and specific words, instead of construing the communica
tion in these scenarios as an offer coupled to a tacit acceptance.11 Where
this is the alternative, the interpretation of the statement as a legally bind
ing promise will, in fact, provide less of a straitjacket in light of the more
flexible revocation regime in place for promises made. The warning that
states might find themselves trapped by their declarations against their
intentions was, in addition, especially understandable before and in the
direct aftermath of the Nuclear Tests cases, as the Courts finding of a bind
ing unilateral assurance was widely felt to have been applied to state
ments that should not have been (nor should in the future be) held to
express an intention to create a legally binding obligation. In line with the
view expressed by various dissenters, the declarations appeared to be too
vague for any such assumption. But the scenario has changed since then.
As the framework presented in chapter three has shown, it would be mis
leading to look at the standard of interpretation solely through the Courts
Nuclear Tests cases holdings. Since then, more than 30 years have passed
and the ICJ has had the opportunity for some fine-tuning of the standard
of interpretation applied to unilateral pledges of states. Under the restric
tive approach, which concentrates on a clear and specific announcement
of a future course of conduct expressed in a declaration, the importance
of which is underlined by additional circumstantial factors, the danger of
politicians stumbling into an unforeseen legal obligation has been effec
tively minimised. A legal straitjacket is therefore not currently in place.
See also ibid 1495.
For attempts to construe a unilateral assurance as either an offer or acceptance see espe
cially the debate on the Austrian declaration of neutrality at p 104 et seq; see also the discus
sion on the declarations on the protection of minorities at p 88 et seq.
10
11

Straitjacket vs Empowering Rule303

In addition, where the rules applicable to declarations of states are

made clear and not left vague, the result is not only that the addressee
may rely on a statement made, but that the declarant can differentiate and
assess beforehand what impact its declaration is likely to have. The clearer
the legal framework is, the less of a trap there is for a state to walk blindly
into. With a clear standard of interpretation, and especially a restrictive
one as presently in place, which bases itself strongly on a textual analysis
of whether a state proclaims a certain course of action in clear and specific
terms, the evaluation provided by a Court becomes much more foresee
able. It is, in fact, rather the absence of clear guidelines which is danger
ous for states, as their actions may then be found to trigger all kinds of
unforeseen and unforeseeable legal consequences. The danger would be
all the greater where a principle according to which states may assume
legally binding obligations through their unilateral declarations is already
in place (as with promises), but its specific features are left unclear. Any
such scenario would be problematic not only for the acting state and its
addressee, but also for the international legal order and in particular the
ICJ, because where a courts judgment is unpredictable through the rele
vant law remaining in the dark, states will be reluctant to put themselves
into the hands of international judges to solve their disputes. To acheive a
strengthening of international law and the increasing use of international
jurisprudence, the remaining black holes within the international legal
system need to be illuminated. Defending them as creating more flexibil
ity for states will, in fact, have the opposite result.
While illuminating the legal principles governing the interpretation
of assurances of states is just as important as the restrictive standard
found to be applied in this respect, neither alleviate another problem
promises are diagnosed to have: that they require no ratification. As gen
erally understood, ratification can refer to the final act creating an under
takings bindingness at either the international or national level. At the
international level, there is indeed no requirement of ratification as an
extra, formal step, which will provide the document with its final legal
validity on the international plane, such as a deposition of the declara
tion with a registrar.12 But not all treaties in fact require ratification at the
international level, as revealed in Vienna Convention, Articles 11 and 14,
which list ratification as only one amongst other means to become bound.
In addition, with a promise, a state representative remains always free to
expressly assume an obligation under the condition of further ratification
at the international level, pending the approval of the undertaking by a
competent body within its own state.
The problem identified when emphasising the lack of ratification proce
dures hence lies more at the national level: most municipal legal systems
12

See pp 24647.

304 Looking Ahead: A Promising Future?

do not seem to require any parliamentary involvement in order to create a
binding legal obligation through a unilateral act, even where the very same
obligation in a treaty would need its blessing. This danger of bypassing the
national (democratic) legislative body is, however, not one which is inher
ently connected to promises it exists similarly for other unilateral acts, such
as acts of recognition or waiver and, of course, it is not a problem caused
by international law. It remains for the national level to decide whether its
executive is allowed to unilaterally assume an undertaking without parlia
mentary involvement. As the case before the German Constitutional Court
in respect of the cruise missiles deployed on its soil has shown (or, more
precisely, the criticism which that judgment triggered),13 there is indeed a
good argument to be made that national ratification procedures should be
extended to cover any far-reaching obligation, and irrespective of whether
a unilateral or bilateral mechanism is used by the executive. A state repre
sentative could, under an altered municipal law, very well be compelled to
seek parliaments approval before making a unilaterally binding promise.
In contemplating any such requirement of parliamentary involvement, it
should, however, be taken into account that promises might thereby lose
some of their flexibility, as they could no longer be as swiftly employed. The
intermediate model adopted, for example, in Sweden and Denmark has
already been mentioned,14 under which such undertakings do not need full
parliamentary approval but a hearing before a Council or Committee com
prised of some parliamentary representatives is sufficient. It should also be
kept in mind that under the Law on Promises as described above, a state
will usually not trigger an obligation of the same rigidity as a non-qualified
treaty obligation. In this respect, the wider flexibility in adopting the under
taking is mirrored by the framework allowing its revocation.
While the dangers posed by the doctrine of promises therefore seem
manageable, it provides states with a power they do not have without it.
In this respect, the doctrine has features in common with the pacta sunt
servanda principle in place for treaties. Despite its general acceptance, the
latter could similarly be described as restraining a state in its future actions
and compelling it to act as offered once the offer is accepted, whereby
its freedom to do as it likes is limited. But to consider pacta sunt servanda
to form a straitjacket for states would, of course, mean looking at only a
small fraction of the bigger picture, as the principle functions essentially
as an empowering rule for states.15 Once in place, states may assume a legal
For a discussion of the case see p 148 et seq.
See ch three, n 175.
15
Speaking of a power conferring rule for pacta sunt servanda also Roberto Lavalle,
About the Alleged Customary Law Nature of the Rule Pacta Sunt Servanda (1982) 33
sterreichische Zeitschrift fr ffentliches Recht und Vlkerrecht 9, 2021. Lavalle, however, goes
too far by assuming that this function is pacta sunt servandas only purpose, as a consequence
of which Lavalle, relying on Hart, draws (in this writers opinion) the wrong conclusion that
the fulfilment of treaty obligations does not constitute compliance with that norm. A power
13
14

Straitjacket vs Empowering Rule305

obligation which they otherwise could not do. As long as the legal frame
work is reasonably clear, this is not a burden, as a state remains free to
choose whether it comports itself in a manner which creates an obligation
or not. The big advantage it now has, therefore, is that it can assume a
legal obligation where intended. Without the pacta sunt servanda principle,
a state could make all sorts of pledges that it intends to fulfil the terms
agreed upon, but there would be no legal certainty for the addressee that
it will really do so. Once the rule is introduced, the full legal apparatus
(ie the principles of state responsibility, the provision for countermeas
ures, the ability to claim fulfilment before international institutions such
as the ICJ, and a states general reputation for adhering to the law) can
be called on as support by the state making(!) the declaration, in order to
persuade the addressee to believe that the offer/acceptance can indeed be
relied upon.
And the very same is true for promises:16 where a state intends to
announce a future course of action which it wants its addressee(s) to
believe in, to rely on and place trust in (for some of the reasons mentioned
in the Introduction and illustrated in the cases in chapter two), the doc
trine of promises now allows for it to do so and establish a legally reliable
commitment. A state is not of course forced to do so, yet through adequate
wording and proclamation, it can now adopt a unilateral commitment of
its own choosing.
A rather recent example of the result of a lack of such an empowering
tool might serve as an illustrative side-note in this context. It presented itself
in German municipal law in the context of the worldwide financial crisis in
2008. In the face of various banks staggering on the brink of bankruptcy, the
conferring rule, after all, may very well as its flipside also lay down a norm of conduct
which can be breached and the obligations resulting from which are to be followed as such.
16
That a parallel rule of acta sunt servanda for all unilateral acts (also called declaratio est
servanda), or promissio est servanda for promises, has sometimes been put forward is there
fore not surprising; for such proposals see, eg Cedeo, ILC, 2524th Meeting, n 5 above, 35,
para 25; Victor Rodriguez Cedeo, Fifth Report on Unilateral Acts of States, Addendum 2,
UN Doc A/CN.4/525/Add.2 (2002) 26, including a Draft Article 7 entitled Acta sunt serv
anda; also Camille Goodman, Acta Sunt Servanda? A Regime for Regulating the Unilateral
Acts of States at International Law (2006) 25 Australian YB International Law 43, 65 et seq. No
such formula has been used here. The reason not to do so is that by grounding the concept
of binding promises on a latin phrase will as such add nothing to the doctrine presently in
place. While it could nevertheless be used in a merely descriptive way, it does not have the
historical backing of that long-standing bedrock principle for contracts and treaties to which
it alludes. As such, it is likely to trigger more confusion rather than help in determining the
legal framework for promises. See also Lukashuk, ILC, 2524th Meeting, n 5 above, para 47:
The Special Rapporteur, guided by practice, rightly defined the rule giving rise to the bind
ing force of the unilateral act as the principle of good faith. So there was no need to invent
any special rule such as declaratio est servanda, proposed by the Special Rapporteur at para
157 [of his First Report]. The principle of good faith was enough. Candioti, ILC, Summary
Record of the 2526th Meeting, UN Doc A/CN.4/SR.2526 (1998) para 58, considered the
development of such a principle albeit unobjectionable also unnecessary, and the basis of
good faith to be sufficient. Critically also Tomuschat, Unilateral Acts, n 9 above, 149495.

306 Looking Ahead: A Promising Future?

German government feared its citizens could lose their trust in the banking
system altogether. The dreaded worst case scenario was that of thousands
of people storming their banks in order to withdraw their money. In order
to avoid such a run on the banks, which could severely destabilise the whole
banking system, and to calm the German population as quickly as possible,
Chancellor Merkel and her financial minister, Steinbrck, appeared before
the national press and promised their population that its money was safe
and guaranteed by the state.17 The problem which not surprisingly was
quickly identified, especially by the opposition, was that the legal value of
this oral intra-state assurance was, at best, completely unclear. Calls to put
in place a law to secure the legal validity of this (wide-ranging) financial
assurance hence quickly became loud and frequent. And indeed, despite
the Merkel-Steinbrck pledge, the level of distrust amongst the German
population and the fear of losing their money remained considerable,18
even though it did not in fact get out of hand and lead to the extreme reac
tions feared. However, had the position, emphasised especially by the leftwing German party Die Linke, gained more ground, according to which
the oral announcements were not worth the microphone which they had
been coughed into,19 the result could have been very different. The prob
lem which Merkel and Steinbrck were facing when they wanted to estab
lish the German citizens immediate trust in the stability of their savings
(drafting a law is after all not a matter of minutes), was that although they
made a promise in this regard, the intra-German legal framework in fact
destabilised the trust that could be placed in the announcement made, as it
appeared to qualify it as being no more than a mere political pledge.
Rules allowing the establishment of a firm legal commitment hence
have not only a limiting effect. Besides protecting the addressee who has
placed trust in a statement made, they also aid the declarant, where the
intention is to create a reliable commitment. For international promises,
the doctrine will therefore function also as an empowering tool for states,
bolstering an assurances reliability by providing it with a legal backing.
While false promises will now hurt a state more than where the doctrine
17
Chancellor Merkel said: Wir sagen Sparerinnen und Sparern, dass ihre Einlagen sicher
sind. Auch dafr steht die Bundesregierung ein (We tell all depositors that their savings
are secure. The government does also vouch for this., my translation), see Spiegel-Online,
5 October 2008, Merkel und Steinbrck im Wortlaut, Die Spareinlagen sind sicher, avail
able at www.spiegel.de/wirtschaft/0,1518,582305,00.html.
18
A survey conducted shortly after the announced guarantee revealed that only 55 per
cent of those questioned trusted that their savings and bank accounts were secure, while 37
per cent said they were not, see ZDF Politbarometer (October 2008), cited in Deutsche ver
lieren Glauben an Sicherheit der Spareinlagen, Focus-Online, 10 October 2008, available at
www.focus.de/politik/deutschland/finanzkrise-deutsche-verlieren-glauben-an-sicherheitder-spareinlagen_aid_339613.html.
19
As claimed by a member of the executive committee of the Die Linke party, see Frau
Merkel muss Garantiezusagen fr private Spareinlagen per Gesetz absichern, 8 October
2008, available at www.linksfraktion.de/wortlaut.php?artikel=1566944463.

Unilateral Promises and Bilateral Commitments307

is not in place (as its responsibility is triggered should it breach its decla
ration), the state which intends its audience actually to believe and trust
its words and act accordingly may create a more stable expectation that it
will act as announced.
II UNILATERAL PROMISES AND BILATERAL COMMITMENTS

As the doctrine of promises allows states to commit themselves legally

through a unilateral assurance, two further questions are bound to arise:
the first is why, in light of the pacta sunt servanda principle, an additional
legal mechanism was created and accepted on the international level,
even though most municipal legal systems do not contain a comparable
tool. Secondly, with promises operating along-side treaties, it might be
questioned whether there is not a danger of states using unilateral prom
ises to circumvent the more rigid rule in place for treaties, for which,
after all and apart from a subsequent inter-party consensus to the oppo
site, pacta sunt servanda.
A Why Allow for Legally Binding Promises at the International Level?
That states make use of unilateral declarations, and that their bindingness
has become accepted, are two findings which as such are reason enough
to focus on the applicable legal framework, and they were the driving
force to write this study. An entirely different, but similarly interesting,
question is why this has come to be the case. As mentioned, especially in
contrast to most municipal legal systems, it may be asked why binding
unilateral promises have evolved and come to be accepted on the inter
national level. The answer to this question may be considered irrelevant
for the application of the legal framework elaborated above: international
courts adjudicate in reliance on binding unilateral assurances, states refer
to them and scholars assert their existence. But for the future application
of the Law on Promises, the question why this development has taken
place is probably going to be at the back of the mind of many international
lawyers. Where the finding that legal systems without unilaterally bind
ing assurances seem to work just fine prevails in judging promises, the
resulting scepticism may have very practical legal consequences. The
question is hence worth some thought: why might the doctrine of unilat
eral assurances slowly but surely have arisen on the international plane
since the last century?
The question whether or not there is actually a need in international law
to allow states to assume unilateral commitments by a unilateral act has in
fact been addressed in various sections of this study, although until now

308 Looking Ahead: A Promising Future?

not under a heading of its own. An answer to the question might begin by
asking why states might intend to make use of a unilateral assurance as a
nevertheless legally trustworthy pledge in light of the fact that they might
as well conclude a treaty, even an oral one, and even one which includes
only a one-sided, unilateral obligation (yet nevertheless requiring a mani
festation of will by both states in order for it to arise). This question has
already been raised in the Introduction, where an analysis of what might
be influencing a state has been provided. To recall, a couple of factors
have been mentioned. A position can be asserted more easily and quickly
through a promise not asking for acceptance than through a treaty, as
another states active participation is not necessary. Where a promise is
made, no bargaining is involved by the beneficiary, but instead, a certain
position is set before the addressee whereby a fait accompli can be pro
duced. Promises might in addition (publicly) signal a states goodwill and
form a gift which is immediately provided. They further allow a state to
express a firm and immediate commitment to a (good) cause and irrespec
tive of whether other states participate in the declaration or even adopt the
same course of action. A firm announcement of a clear and specific course
of action might also be used as a means of future self-restraint, to act as
declared no matter what comes. The statement can similarly be directed
to a very large audience and a declaration even to the whole international
community can as such be swiftly undertaken. It might be directed to the
community of states, or indeed to a hidden audience, be it private inves
tors, a court or a political opponent, with whom direct communication is
presently and politically impossible.20
As chapter two, the History of Promises, has shown, there are for this
and other reasons numerous declarations which do not reflect the offerand-acceptance structure necessitated in order for a declaration to be con
sidered as binding under a treaty analysis. Many of them nevertheless
display what, from an outside perspective, appears to be a clear intention
to create a reliable undertaking.21 That international law through the legal
doctrine of promises assumes such declarations to be more than a sign
of goodwill has therefore in the foregoing pages been seen to stabilise,
instead of undermine, the expectations which such one-sided pledges are
intended to bring about. After all, they are made in order to be believed
and to influence the perception and behaviour of their addressees. To
support the statements made thus bolsters the impact of the declaration,
just as it provides their addressees with an additional layer of security
For more see p 4 et seq.
Of the cases and state practice discussed in ch two, see the various pledges made before
courts, or such assurances as the one given by Austria of its neutraliy, Egypt of its legal
regime for the Suez canal, the unnamed foreign country vis--vis Austria on the use of war
material, and some of the negative security assurances, especially (though by no means ony)
the ones made by China.
20
21

Unilateral Promises and Bilateral Commitments309

in protecting the belief and reliance on the fulfilment of the undertaking

made.
If this is so, however, how is it that a similar tool of legal commitment
is not present in most national legal systems, where the unilateral pledge
of one citizen will usually not create a legally biding undertaking vis--vis
another? With the municipal contract and international treaty parallelism
in mind, the municipal gap in this respect has been said possibly to raise
some doubts as to the concept of unilaterally obliging promises as such.
However, a couple of differences between the international system and
the intra-state civil law situation are noteworthy. Unlike a countrys popu
lation, which usually numbers millions or more, states remain part of a
group with a limited and relatively small amount of members. This is true
despite the growing importance of other international actors which may
be capable of making similar declarations and are potential addressees.
Such club members move about on the same plane, interact and inform
each other either directly or are informed of their actions and statements
via modern media. On account of the (ever-increasing) interdependence
of states and their populations, the future course of action of one will often
have a significant impact on that of many other states, whereby the fore
seeability of another states conduct is of more importance than ever. It is,
in this respect, no coincidence that states frequently make declarations in
which a certain course of future conduct is openly announced, a behav
iour which is not as commonly seen amongst private actors.22 In particu
lar, declarations which are made by a states highest representative (or
those especially empowered) will more often than not be of a considerable
importance for their addressees (and hence a large number of people).
In addition, and unlike private citizens acting and interacting on all
kinds of social levels, states are in contact with each other usually in a
much more formal way. Individuals performing their roles on both sides
are not only speaking on their own behalf, but with limited powers. It is
therefore no coincidence that politicians speak the way they do, and as
any regular news viewer will have noticed, politicians have made it an
art of their own to counter even the simplest yes or no question with an
answer such as let me put it this way, only to end in a lengthy statement
which gives nothing away. Public declarations of states are similarly often
dissatisfying for those searching for a clear position. If politicians, there
fore, usually employ vague and uncertain terms in their statements on
the international plane, there is a good argument to be made that, when
they refrain from doing so and adopt a clear and specific statement in
22
The continuous growth of importance of other actors on the international plane (eg
international organisations, NGOs or even large corporations) may make parallel legal
norms ever more applicable in those areas. The parallelism will be at its strongest especially
for international organisations, see pp 7879.

310 Looking Ahead: A Promising Future?

matters which fall into their area of expertise and in circumstances which
have put them on notice that the statement will be understood as having
been carefully phrased, this announcement is to be relied upon as more
than a mere political sign of goodwill, ie as a legal commitment. While the
communication of private individuals might indeed be in some danger
of being muzzled if a unilateral pledge was turned into a legally binding
commitment (even here there may be room for debate), the interaction on
the international plane is hence clearly different. Here, the legal frame
work under a restrictive standard of interpretation will respond to these
concerns by supporting (only) those statements with an adequate legal
backing which, on account of their clear message (usually in a important
matter), signal that they want to be taken seriously and may therefore be
understood by the addressee to indeed mean what they say.
B Circumventing or Complementing Treaties?
Although subject to limitations, the legal framework on the revocation of
promises has been said not to be as rigid as the one imposed on obliga
tions assumed by a treaty (at least where both include no terms as to their
revocability). The question which might therefore be asked is whether
providing states with the tool of promises could not undermine the future
existence and use of treaties as a more stable means for states to under
take legal obligations. In this respect it is, however, important to recall
that by a promise a state can only undertake a one-sided obligation for
itself. While it has been said that promises as a result are not necessarily
rare, it would be arguing for the opposite extreme to now assume that
promises could be used to circumvent treaties and will become exceed
ingly common: where a direct quid pro quo is sought by a state, a uni
lateral promise is from the outset of no use. A competition between the
two mechanisms of treaty and unilateral promise would hence only arise
where a state wishes to receive no obligation in direct return. Such an obli
gation may be assumed through an offer asking for nothing more than an
acceptance (ie no quid pro quo), or it can be proclaimed as a hereby and
herewith effective undertaking in the form of a promise. Even in these
cases, the mechanism adopted by a state is likely to depend less on the
regime of revocation triggered thereby than on what the declarant state
hopes to achieve. Should it be interested in establishing an obligation for
its own good (ie to restrain itself, to convince a hidden audience behind
the addressee that a certain conduct is outlawed, to create a fait accompli
that prevents further negotiations and debates, to fulfil an earlier pledge
or obligation that it will commit itself, etc), it is likely to proclaim an
obligation as created through the declaration itself. Especially where
a statement is made to the international community as a whole, there

Unilateral Promises and Bilateral Commitments311

might, in addition, be practical reasons which make the use of a promise

more convenient than that of an offer asking for acceptance.
Even in the limited number of scenarios where competition might
arise, the motivation behind the offer or promise made is hence usually
going to be different. However, where a state has as such no preference as
to whether to adopt an immediate obligation through a unilateral promise
or merely offer one, it might, of course, decide to adopt a promise pre
cisely and solely on account of the above-depicted more flexible regime
governing its revocation. But where it chooses to do so, the message sent
to the addressee is also clear: you may rely, yet not as strongly and per
petually as if the same declaration had been included in a treaty with
no clause as to its revocation. With the limits of a commitment chosen
by a state being clear, the result is not that of bypassing the rule of pacta
sunt servanda. With promises there is now rather merely another means
available for a state to use, depending on which signal it intends to send.
Promises consequently do not circumvent, but rather complement, trea
ties in the toolbox available to states wishing to undertake a new legal
commitment.

Concluding Remarks

N THE WAKE of the International Law Commissions sixtieth anniversary in 2009, various publications have addressed the
Commissions merits and also some of its weaknesses.1 In the context
of this books topic, it will be recalled that the Commissions final output with respect to unilateral declarations capable of obliging states has
met with considerable criticism on the above pages. The Commissions
overall effort in this field has, nevertheless (and in particular thanks to
the ILCs habit of precisely and openly documenting its every working
step), served as a treasure chest of information and inspiration which was
invaluable in reaching many of the findings in this book.
In following in the footsteps of the International Law Commission, the
initial plan for this study had, in fact, been to cover the entire subject of
unilateral acts of states under international law, in order to provide an
up-to-date description of that topic. Studying the ILC debates and the
documents it produced during its endeavour, however, quickly revealed
a major problem such an approach was bound to encounter: while some
aspects within the topic appeared to be long settled and gave rise to little disagreement (such as, for example, the existence and consequences
of protests or the legal effects of a waiver), others were still fraught with
uncertainties. The close attention that the latter required could not be provided where the task was so broad as to cover the law on all unilateral
state action. As some of the subtopics still required full-fledged analysis,
the Special Rapporteur found himself confronted with the task of having to construe the big picture even though some of the essential pieces
of mosaic to be used still lacked clear determination. The books focus
hence changed and became narrower, to deal solely with promises of
states, as the latter were of great potential impact and clearly necessitated
further and intensive study. With the wealth of material produced by the
Commission, its final product, which underscored the importance of the
available and continuously growing ICJ dicta and some scholarly writings, provided a fertile ground for further academic analysis.
1
See Georg Nolte (ed), Peace through International Law: the Role of the International Law
Commission, a Colloquium at the Occasion of its Sixtieth Anniversary, Beitrge zum auslndischen
ffentlichen Recht und Vlkerrecht, vol 211 (Berlin, 2009), and the articles included in the Focus
Section on The International Law Commission: Sixty Years of Progress in Codification in
(2006) 49 German YB International Law 77, particularly Christian Tomuschat, The International
Law Commission: an Outdated Institution? (2006) 49 German YB International Law 77.

Concluding Remarks313

This path, once chosen, led to a journey which nevertheless from the
very beginning necessitated dealing with some questions in need of being
answered for all unilateral acts. In the delimitation of the subject matter it was especially the adjective that defines this kind of state behaviour, ie its unilateral character, which turned out to lead to a field of
vast debate. After the questions raised thereby had been addressed and
the unilateral act which was to be analysed had been delimited, the next
problem encountered was that of a strong disagreement within the legal
community on how to interpret different cases, as well as state practice,
in respect of promises of states. While some authors pointed to examples
dating back to the inauguration of the League of Nations, others considered the Nuclear Tests cases to have had nothing to build on, and each
and every example provided has come under dogmatic fire from one or
the other scholar. In order to provide the full picture, make up our own
mind and allow the reader to judge for him or herself whether this position was accurate or not, the History of Promises has dealt with these
examples in considerable detail. Another pivotal reason for doing so was
that the law in this area has been shaped primarily through jurisprudence,
which hence had to be portrayed. Despite the fact that, especially after the
ICJs positioning within the debate in its Nuclear Tests cases judgments
and its subsequent jurisprudence, both states and scholars were found
to agree on the potential bindingness of unilateral assurances of states,
the open questions within the legal framework appeared to be considerable. On closer look, however, it was not so much a bold move forward
into completely uncharted territory which was necessary in order to illuminate the remaining blind spots. Systematic legal analysis with an eye
on coherence allowed us to reach a number of the above-presented findings. With regard to the revocability of promises, the ICJs application of
the principle of good faith in the Military and Paramilitary Activities in and
against Nicaragua case has been decisive in order to put flesh on the bones
of a more flexible approach, which, as seen, has received the support of
numerous scholars within and outside the ILC. It is especially in this area
that future state practice, jurisprudence and scholarly writings should be
closely monitored in order to ascertain whether the above-depicted findings continue to be further entrenched.
By stepping back to look at the bigger picture presented in this study,
the law on state promises can be seen not to be as sketchy and fragmented
as first thought. In addition, the framework in place does not put states
on too tight a leash, while nevertheless ensuring that what a state promises others in clear and specific words today, will not simply turn to dust
overnight.

(Free Zones
case)

PCIJ
Mavrommatis
Jerusalem
Concessions
case

Case or
State
practice

Issue

Declaration of
state representative
made before
the Court

Date

26 March 1925
(judgment)

(Discussed in the book but omitted here as it was expressly classified as an offer
by the PCIJ)

The British representative had stated that there can be no question of our acting on any request to expropriate M Mavrommatis. If M Rutenberg was so
. . . dishonest, so unreasonable, now as to ask to expropriate him . . . we should
not act upon that request, which led the Court to the following finding: After
this statement, the binding character of which is beyond question, the Court
considers that henceforward it is quite impossible that the British or Palestine
Governments should consent to comply with a request of the expropriation of
M Mavrommatis Jerusalem Concession.

Significance

(Beginning with the inauguration of the League of Nations in 1920 today and
as discussed in detail in chapter two)

Promises: Important Cases and Some Prominent

Examples of State Practice

ANNEX I

25 May 1926
(judgment)

15 January 1930
(Adatcis
Report)

6 April 1935
(Courts
Opinion)

5 April 1933
(judgment)

PCIJ
Certain
German
Interests in
Upper Polish
Silesia case

PCIJ
Advisory
Opinion
Minority
Schools in
Albania

PCIJ
Legal Status
of Eastern
Greenland
case

Oral declaration
confirmed in
writing by the
Norwegian
Minister
for Foreign
Affairs, Ihlen,
made vis-vis the Danish
Minister

Declarations
concerning
the protection
of minorities made
vis--vis the
League of
Nations

Declaration of
state representative
made before
the Court

The Ihlen declaration was considered binding upon Norway by the PCIJ. Again
the legal construction remained ambiguous. The Court used the word promise
numerous times which prompted parts of legal doctrine to begin to consider
promises as binding in international law. The Court, indeed, never spoke of a
treaty, agreement, offer or acceptance when it referred to the declaration, whose
binding nature it simply considered to be beyond dispute.

The legal interpretation of the declarations made by various countries vis-vis the League is ambiguous. The treaty nature of their assurance was, however, explicitly rejected by two state representatives (Estonia and Lithuania).
In an expert opinion prepared for the League, the unilateral character of the
Lithuanian declaration was addressed and left open, since the declaration was
considered to be binding on that country in any event. In its advisory opinion,
the PCIJ did not discuss the nature of the Albanian declaration (also referred to
as an instrument or act but never a treaty), while clearly considering it to be
binding upon Albania.

The judgment contains the following passage in respect of assurances given by

the Polish representative that certain property will not be expropriated: The
representative before the Court of the respondent Party, in addition to the declarations above mentioned . . . has made other similar declarations which will
be dealt with later; the Court can be in no doubt as to the binding character of
all these declarations.

Annex IPromises: Important Cases and Examples of State Practice 315

Date

8 August 1945
(London Treaty)

1955

24 April 1957

Case or
State
practice

International
Military
(Nuremberg)
Tribunal

State practice

State practice

Table cont.:

Egyptian
declaration
on the Suez
Canal

Austrian
declarations
of neutrality

German
Assurances

Issue

Egypt deposited a written unilateral declaration in respect of the Suez Canal

with the UN Secretary-General. The declaration included a reaffirmation of the
Constantinople Convention but also numerous new assurances and a paragraph
according to which the declaration with the obligations therein constitutes an
international instrument.

In fulfilment of a memorandum in which Austria had assured its intention to

make a declaration in a form which will obligate Austria internationally to practice in perpetuity a neutrality of the type maintained by Switzerland, Austria
enacted a constitutional law and subsequently notified governments around the
world of its decision to maintain a permanent neutrality, asking them to recognise this self-declared permanent neutrality.

Article 6 of the IMTs Charter listed the planning, preparation, initiation or

waging of a war of aggression, or a war in violation of international treaties,
agreements or assurances, or participation in a common plan or conspiracy
for the accomplishment of any of the foregoing as a war against the peace.
The indictment against Herman Goering et al included seven charges of such
violations of assurances.

Significance

316 Annex IPromises: Important Cases and Examples of State Practice

20 February
1969
(judgment)

20 December
1974
(judgments)

ICJ
North Sea
Continental
Shelf cases

ICJ
Nuclear Tests
cases

French
declarations
to cease
atmospheric
nuclear
testing

Germanys
conduct and
statements in
respect of the
continental
shelf

The ICJ considered it to be well recognised that declarations made by way of

unilateral acts may have the effect of creating legal obligations. When it was
the intention of the state making the declaration that it should become bound
according to its terms, that intention conferred on the declaration the character
of a legal undertaking, the state being thenceforth legally required to follow a
course of conduct consistent with the declaration. In these circumstances, nothing in the nature of a quid pro quo nor any subsequent acceptance of the declaration, nor even any reply or reaction from other states, was required for the
declaration to take effect, since such a requirement would be inconsistent with
the strictly unilateral nature of the juridical act by which the pronouncement by
the state was made. The intention to be bound had to be ascertained by interpretation of the unilateral act taking note of the declarations substance as well
as the surrounding circumstances while applying a restrictive interpretation.
The Court found that no formal requirements existed. The binding character of
an international obligation assumed by unilateral declaration is said to be based
on good faith.

The Court did not consider whether Germany had made a unilateral binding
assurance as it found only estoppel to be worthy of a closer analysis since: In
principle, when a number of States, including the one whose conduct is invoked,
and those invoking it, have drawn up a convention specifically providing for a
particular method by which the intention to become bound by the rgime of
the convention is to be manifested namely by the carrying out of certain prescribed formalities (ratification, accession), it is not lightly to be presumed that
a State which has not carried out these formalities, though at all times fully able
and entitled to do so, has nevertheless somehow become bound in another way.
Judge Ammoun in his separate opinion criticised the Court for not taking into
account a well-settled doctrine that a State may be bound by a unilateral act.
He himself, however, dismissed it as applicable to Germany in this case.

Annex IPromises: Important Cases and Examples of State Practice 317

17 February
1977
(note)
8 December
1977
(UN memorandum)

13 January
1978
(UN memorandum)

8 February
1977
(declarations)

18 January
1978
(judgment)

State practice
(state declaration assessed
by the United
Nations)

State practice
(state declaration assessed
by the United
Nations)

State practice
ECtHR

Ireland
v United
Kingdom

Date

Case or
State practice

Table cont.:

Declaration by the
British representative made before
the ECtHR and in
direct reference to the
Nuclear Tests cases
pledging not to reintroduce certain interrogation techniques

Unilateral declarations
on torture and other
inhumane treatment
made by states pursuant to GA Res 32/64
which had a unilateral
model declaration
annexed to it

Declaration by
Switzerland to the
United Nations

Issue

Inspired by the Nuclear Tests cases and having decided to make a statement sufficiently clear, addressed to this Court as representing the international community of free Europe, so as to enable the Court to hold at
this stage that that statement constitutes an undertaking possessing legal
effect, the British representative stated the following before the ECtHR:
I am authorised to give, in relation to the five techniques, the following unqualified undertaking, and I measure my words with care. The
Government of the United Kingdom have considered the question of
the use of the five techniques with very great care and with particular regard to Article 3 of the Convention. They now give this unqualified
undertaking, that the five techniques will not in any circumstances be
reintroduced as an aid to interrogation. The ECtHR took formal note of
the undertaking given before it.

Applying a more restrictive approach, the UN Secretariat in a memorandum considered the declarations not to constitute legal undertakings
as: the model declaration annexed to the resolution would merely have
Governments declare their intention to comply and to implement, rather
than indicate that they will or undertake to comply and implement.
On balance, therefore, the resolution suggests that no binding obligation
is intended. This was found to be confirmed by the debates which lead up
to the adoption of the Resolution.

In a note to the United Nations, Switzerland stated that it would . . .

independently and without recognising any legal obligation in the matter, take care to prevent [its] territory from being used for the purpose of
circumventing the Security Council sanctions. In a memorandum, the UN
Legal Secretariat drew the (very far-going) conclusion that Switzerland
had thereby unilaterally and unreservedly accepted the obligation not to
allow the holders of Rhodesian passports entry into the country.

Significance

318 Annex IPromises: Important Cases and Examples of State Practice

27 June 2001
(judgment)

ICJ

Declarations
before the ICJ

Various declarations made by

the five official
nuclear weapons
powers assuring
not to use nuclear
weapons in future
against countries
party to the NPT

Table continued on the following page:

LaGrand case

Negative
security
assurances

1978, 1982, 1995

and 2010
(declarations)

State practice

The Court emphasised that [i]f a State, in proceedings before this Court,
repeatedly refers to substantial activities which it is carrying out in order
to achieve compliance with certain obligations under a treaty, then this
expresses a commitment to follow through with the efforts in this regard.
In its ruling it unanimously took note of this commitment and held that it
must be regarded as meeting the Federal Republic of Germanys request
for a general assurance of non-repetition.

The declarations have, at least in part, been considered to be binding unilateral commitments by legal commentators and (some) of their addressees. Their wording is both clear and specific. China, in addition, spoke of
an unconditional undertaking, and all statements are marked by a high
degree of formality. The ICJ in its advisory opinion unanimously observed
that any future behaviour should be compatible, inter alia, with these
undertakings. The NPT parties, however, have expressed their wish that
these assurances should be contained in a legally binding instrument
and, therefore, appear to interpret them as presently not legally binding.

Annex IPromises: Important Cases and Examples of State Practice 319

(Order)
Questions
relating to the
Obligation to
Prosecute or
Extradite case

ICJ

28 May 2009

3 February 2006
(judgment)

ICJ
Case concerning Armed

Activities on
the Territory
of the Congo

Date

Case or
State practice

Table cont.:

Declaration made
by the representative of Senegal
before the ICJ

Statement made
by the Minister of
Justice of Rwanda
before the
United Nations
Commission on
Human Rights

Issue

Belgium indicated that a solemn declaration made before the Court by the
Agent of Senegal could be sufficient for Belgium to consider that its Request
for the indication of provisional measures had no further raison dtre, provided that such a declaration would be clear and unconditional, and that it
would guarantee that all the necessary measures would be taken by Senegal
to ensure that Mr Habr did not leave Senegalese territory before the Court
delivered its final Judgment. The ICJ consequently took note of the formal
assurance repeatedly made by Senegal which during the proceedings had
solemnly declared that Senegal will not allow Mr Habr to leave Senegal
while the present case is pending before the Court.

The Court further elaborated on its Nuclear Tests cases doctrine and
observed that in accordance with its consistent jurisprudence it is a well
established rule of international law that the Head of State, the Head of
Government and the Minister of Foreign Affairs are deemed to represent
the State merely by virtue of exercising their functions, including the performance, on behalf of said State, of unilateral acts having the force of
international commitments. The Court noted that other persons too might
be authorised to do so and could not rule out that this might be true for
a Minister of Justice. It recalled that for a declaration to be binding its
actual content as well as the circumstances in which it was made had to be
analysed. In respect of content, it further recalled that a declaration could
create legal obligations only if it was made in clear and specific terms. The
declaration before it, however, was not held to be specific enough in order
to view it as binding. Instead it was of an indeterminate character and
lacked a precise time-frame by pledging that reservations would shortly
be withdrawn. The context of a presentation of general policy was said
to confirm this finding.

Significance

320 Annex IPromises: Important Cases and Examples of State Practice

ANNEX II
The ILCs Guiding Principles
Guiding Principles Applicable to Unilateral Declarations of
States Capable of Creating Legal Obligations1
The International Law Commission,
Noting that States may find themselves bound by their unilateral behaviour on the international plane,
Noting that behaviours capable of legally binding States may take the form
of formal declarations or mere informal conduct including, in certain situations, silence, on which other States may reasonably rely,
Noting also that the question whether a unilateral behaviour by the State
binds it in a given situation depends on the circumstances of the case,
Noting also that in practice, it is often difficult to establish whether the legal
effects stemming from the unilateral behaviour of a State are the consequence of the intent that it has expressed or depend on the expectations
that its conduct has raised among other subjects of international law,
Adopts the following Guiding Principles which relate only to unilateral
acts stricto sensu, ie those taking the form of formal declarations formulated by a State with the intent to produce obligations under international
law.
1. Declarations publicly made and manifesting the will to be bound
may have the effect of creating legal obligations. When the conditions for
this are met, the binding character of such declarations is based on good
faith; States concerned may then take them into consideration and rely on
them; such States are entitled to require that such obligations be respected.
2. Any State possesses capacity to undertake legal obligations through
unilateral declarations.
1
Text adopted by the ILC at its Fifty-eighth session, in 2006, and submitted to the General
Assembly as a part of the Commissions report covering the work of that session (A/61/10).
Available at http://untreaty.un.org/ilc/guide/9_9.htm, to be reproduced in (2006) II (2) YB
International Law Commission.

322 Annex II The ILCs Guiding Principles

3. To determine the legal effects of such declarations, it is necessary
to take account of their content, of all the factual circumstances in which
they were made, and of the reactions to which they gave rise.
4. A unilateral declaration binds the State internationally only if it is
made by an authority vested with the power to do so. By virtue of their
functions, heads of State, heads of Government and ministers for foreign
affairs are competent to formulate such declarations. Other persons representing the State in specified areas may be authorized to bind it, through
their declarations, in areas falling within their competence.
5. Unilateral declarations may be formulated orally or in writing.
6. Unilateral declarations may be addressed to the international community as a whole, to one or several States or to other entities.
7. A unilateral declaration entails obligations for the formulating State
only if it is stated in clear and specific terms. In the case of doubt as to
the scope of the obligations resulting from such a declaration, such obligations must be interpreted in a restrictive manner. In interpreting the
content of such obligations, weight shall be given first and foremost to the
text of the declaration, together with the context and the circumstances in
which it was formulated.
8. A unilateral declaration which is in conflict with a peremptory norm
of general international law is void.
9. No obligation may result for other States from the unilateral declaration of a State. However, the other State or States concerned may incur
obligations in relation to such a unilateral declaration to the extent that
they clearly accepted such a declaration.
10. A unilateral declaration that has created legal obligations for the
State making the declaration cannot be revoked arbitrarily. In assessing
whether a revocation would be arbitrary, consideration should be given
to:
(i) any specific terms of the declaration relating to revocation;
(ii) the extent to which those to whom the obligations are owed
have relied on such obligations;
(iii) the extent to which there has been a fundamental change in the
circumstances.

Bibliography
Books and Articles
Abi-Saab, Georges, Les sources du droit international: un essai de dconstruction in Manuel Rama-Montaldo (ed), El derecho internacional en un mundo en
transfomacion, Liber Amicorum en homenaje al Profesor Eduardo Jimnez de Archaga
(Montevido, 1994) 2949
Adatci, Minitciro, Petition from Thirty-Four Persons of Russian Origin Living
in Lithuania Regarding the Confiscation of their Lands under the Lithuanian
Agrarian Reform Law of 15 February 1922, Report to the League of Nations
Council (1930) 11 League of Nations Official Journal 179
Anzilotti, Dionisio, Cours de droit international (translation from the Italian by
G Gidel, Paris, 1929)
Arechaga, Jimenez de, International Law in the Past Third of a Century (1978)
159 Recueil des Cours de lAcadmie de Droit International de la Haye 1
Balladore Pallieri, Giorgio, Diritto Internazionale Pubblico, 7th edn (Milan, 1956)
Bastid, Suzanne, Les traits dans la vie internationale (Paris, 1985)
Bernauer, Thomas, Nuclear Issues on the Agenda of the Conference on Disarmament
(New York, 1991)
Bindschedler, Rudolf L, Die Neutralitt im modernen Vlkerrecht (1956) 17
Zeitschrift fr auslndisches ffentliches Recht und Vlkerrecht 1
Bleckmann, Albert, Anmerkungen zum Nordirland-Fall des Europischen
Gerichtshofs fr Menschenrechte (1979) Europische Grundrechte Zeitschrift 188
Bollecker-Stern, B, Laffaire des essais nuclaires francais devant la Cour
Internationale de Justice (1974) Annuaire Franais de Droit International 299
Bonner, David, Ireland v. United Kingdom, Annotation (1978) 27 International and
Comparative Law Quarterly 897
Bos, Maarten, A Methodology of International Law (Amsterdam/New York, 1984)
Bowett, DW, Estoppel before International Tribunals and its Relation to
Acquiescence (1957) 33 British YB International Law 176
Brandon, Michael, Analysis of the Term Treaty and International Agreement
for Purposes of Registration under Article 102 of the United Nations Charter
(1953) 47 American J International Law 49
Brierly, James, Report by J.L. Brierly, Special Rapporteur for the Law of Treaties, A/
CN.4/23 reproduced in (1950) II Yearbook of the International Law Commission 222
The Basis of Obligation in International Law and Other Papers (Oxford, 1958)
Briggs, HW, Nicaragua v. United States: Jurisdiction and Admissibility (1985) 79
American J International Law 373
Brownlie, Ian, The Responsibility of States for the Acts of International Organizations
in Maurizio Ragazzi (ed), International Responsibility Today, Essays in Memory of
Oscar Schachter (Leiden/Boston, 2005) 35563

324 Bibliography
Brownlie, Ian, Principles of Public International Law, 7th edn (Oxford/New York,
2008)
Bunn, George, The Legal Status of U.S. Negative Security Assurances to NonNuclear Weapon States (1997) 4(3) The Nonproliferation Review 1
Caflisch, Lucius, La pratique suisse en matire de droit international public 1982
in Schweizerisches Jahrbuch fr Internationales Recht (1983) vol XXXIX 177268
Cahier, Philippe, Le problme des effets des traits lgard des Etats tiers
inRecueil des Cours de lAcadmie de Droit International de la Haye (Leiden, 1974)
vol 143, 589736
Carbone, Sergio M, Promessa e affidamento nel diritto internazionale (Milano, 1967)
Promise in International Law: A Confirmation of its Binding Force (1975) 1
Italian J International Law 166
Cassese, Antonio, International Law, 2nd edn (Oxford, 2004)
Cedeo, Victor Rodriguez, First Report on Unilateral Acts of States, A/CN.4/486
(1998)
Second Report on Unilateral Acts of States, A/CN.4/500 (1999)
Second Report on Unilateral Acts of States, Addendum, A/CN/500/Add.1
(1999)
Third Report on Unilateral Acts of States, A/CN.4/505 (2000)
Fourth Report on Unilateral Acts of States, A/CN.4/519 (2001)
Fifth Report on Unilateral Acts of States, A/CN.4/525 (2002)
Fifth Report on Unilateral Acts of States, Addendum, A/CN.4/525/Add.1 (2002)
Fifth Report on Unilateral Acts of States, Corrigendum, A/CN.4/525/
Add.1/Corr.2 (2002)
Fifth Report on Unilateral Acts of States, Addendum 2, A/CN.4/525/Add.2
(2002)
Sixth Report on Unilateral Acts of States, A/CN.4/534 (2003)
Seventh Report on Unilateral Acts of States, A/CN.4/542 (2004)
Eighth Report on Unilateral Acts of States, A/CN.4/557 (2005)
Ninth Report on Unilateral Acts of States, A/CN.4/569 (2006)
Ninth Report on Unilateral Acts of States, Addendum 1, A/CN.4/569/Add.1
(2006)
Charpentier, Jean, La reconnaissance internationale et lvolution du droit des gens
(Paris, 1956)
Engagements unilatraux et engagements conventionnels: diffrences et
convergences in Jerzy Makarczyk (ed), Theory of International Law at the Threshold
of the 21st Century (The Hague/London/Boston, 1996) 36780
Chen, Ti-Chiang, The International Law of Recognition with Special Reference to
Practice in Great Britain and the United States (London, 1951)
Combacau, Jean and Sur, Serge, Droit international public, 5th edn (2001)
DArgent, Pierre, Article 36, Convention de Vienne de 1969 in Olivier Corten and
Pierre Klein (eds), Les conventions de Vienne sur le droit des traits, commentaire
article par article (Brussels, 2006) vol II, 146569
DAspremont Lynden, Jean, Les travaux de la commission du droit international relatifs aux actes unilatraux des tats (2005) 109 Revue Gnrale de Droit
International Public 163
Dahm, Georg, Delbrck, Jost and Wolfrum, Rdiger, Vlkerrecht, 2nd edn (Berlin,
2002) vol I, 3


Bibliography325
De Nova, Rodolfo, Die Neutralisation sterreichs (1958) 54 Die Friedenswarte 298
Deeks, Ashley, Promises Not to Torture: Diplomatic Assurances in U.S. Courts, American
Society of International Law Discussion Paper Series (December 2008)
Degan, Vladimir-Djuro, Unilateral Act as a Source of Particular International
Law (1994) 5 Finnish Yearbook of International Law 149
Dehaussy, Jacques, La dclaration gyptienne de 1957 sur le canal de Suez (1960)
VI Annuaire Francais de Droit International 169
Les actes juridiques unilatraux en droit international public: propos dune
thorie restrictive (1965) 92 Journal du droit international 41
Dominic, Christian, A propos du principe de lestoppel en droit des gens (1968)
Dupuy, Pierre-Marie, La pratique de larticle 38 du Statut de la Cour Internationale
de Justice dans le cadre des plaidoiries ecrites et orales inUnited Nations Office
of Legal Affairs (ed), Collection of Essays by Legal Advisers of States, Legal Advisers
of International Organizations and Practitioners in the Field of International Law =
Recueil dArticles de conseillers juridiques dEtats, dorganisations internationales et
de praticiens du Droit International (New York, 1999) 37794
Droit international public, 8th edn (Paris, 2006)
Elaraby, Nabil, Some Reflections on Disarmament in Christian Tomuschat (ed),
The United Nations at Age Fifty: A Legal Perspective 9-24 (The Hague, 1995)
Elias, TO, The Modern Law of Treaties (Dobbs Ferry, 1974)
Elkind, Jerome B, Footnote to the Nuclear Tests Cases: Abuse of Right A Blind
Alley for Environmentalists (1976) 9 Vanderbilt JTL 57
Ermacora, Felix, 20 Jahre sterreichische Neutralitt, 2nd edn (1975)
Fabender, Bardo, Die Vlkerrechtssubjektivitt internationaler Organisationen
(1986) 37 sterreichische Zeitschrift fr ffentliches Recht und Vlkerrecht 17
Fiedler, Wilfried, Zur Verbindlichkeit einseitiger Versprechen im Vlkerrecht
(1976) 19 German YB International Law 35
Unilateral Acts in International Law in Rudolf Bernhardt, Peter MacalisterSmith and Max-Planck-Institut fr Auslndisches ffentliches Recht und
Vlkerrecht. (eds), Encyclopedia of Public International Law (Amsterdam/New
York, 2000) vol 4
Fitzmaurice, Sir Gerald, The Law and Procedure of the International Court of
Justice, 195154: General Principles and Sources of Law (1953) 30 British YB
International Law 1
The Law and Procedure of the International Court of Justice 195154: Treaty
Interpretation and Other Treaty Points (1957) 33 British YB International Law 202
Some Problems Regarding the Formal Sources of International Law in
Symbolae Verzijl (The Hague, 1958) 15376
Franck, Thomas M, Word Made Law: the Decision of the International Court of
Justice in the Nuclear Test Cases (1975) 69 American J International Law 612
Icy Day at the ICJ (1985) 79 American J International Law 379
Garner, James W, The International Binding Force of Unilateral Oral Declarations
(1933) 27 American J International Law 493
Gigante, A, The Effect of Unilateral State Acts in International Law (1969) 2 NYU
School of Law J International Law and Politics 333
Goodman, Camille, Acta Sunt Servanda? A Regime for Regulating the Unilateral
Acts of States at International Law (2006) 25 Australian YB International Law
43

326 Bibliography
Grotius, Hugo, The Rights of War and Peace (translation, Indianapolis, 2005) vol II
Guggenheim, Paul, Trait de droit international public, 2nd edn (1967) vol 1
Hafner, Gerhard, Die sterreichische diplomatische Praxis zum Vlkerrecht
(1981) 32 sterreichische Zeitschrift fr ffentliches Recht und Vlkerrecht 301
Heintschel von Heinegg, Wolf, Einseitige Rechtsakte in Knut Ipsen (ed),
Vlkerrecht, 5th edn (Mnchen, 2004) 23440
Institut de Droit International, Resolutions Concerning the Recognition of New
States and New Governments (1938) 30 American J International Law, Supplement
185
Jacqu, Jean-Paul, Elments pour une thorie de lacte juridique en droit international
public (Paris, 1972)
A propos de la promesse unilatrale in Melanges offerts a Paul Reuter (1981)
32745
Jellinek, Georg, Die Rechtliche Natur der Staatenvertrge. Ein Beitrag zur Juristischen
Construction des Vlkerrechts (Wien, 1880)
Jennings, Sir Robert, General Course on Principles of International Law (1967)
121(2) Recueil des Cours de lAcadmie de Droit International de la Haye 325
Karl, Wolfram, Protest in Rudolf Bernhardt (ed), Encyclopedia of Public International
Law (Amsterdam, 1986) vol III 115759
Kennedy, David, Of War and Law (Princeton, 2006)
Kersten, Jens, Georg Jellinek und die klassische Staatslehre (Tbingen, 2000)
Kewenig, Wilhelm A, Der Internationale Gerichtshof und die franzsischen
Kernwaffenversuche in Recht im Dienst des Friedens, Festschrift fr Eberhard Menzel
(1975) 32348
Khosla, Dinesh, Nuclear Test Cases: Judical Valour v. Judicial Discretion (1978)
Indian YB International Law 322
Kiss, Alexandre-Charles, Les actes unilatraux dans la practique francais du droit
international (1961) 2 Revue Gnrale de Droit International Public 317
Rpertoire de la pratique Francaise en matire de droit international public (Paris,
1962) vol I
Klabbers, Jan, The Concept of Treaty in International Law (The Hague/Boston/
London 1996)
Kolb, Robert, La bonne foi en droit international public. Contribution ltude des principes gnraux de droit (Paris, 2000)
Koskenniemi, Martti, From Apology to Utopia: the Structure of International Legal
Argument (Reissue with a new epilogue, Cambridge, 2005)
Kramer, Francisco Villagrn, Les actes uniltraux dans le cadre de la jurisprudence internationale in International Law on the Eve of the Twenty First Century:
Views from the International Law Commission (New York, 1997)
Kunz, Josef, Austrias Permanent Neutrality (1956) 50 American J International
Law 418
Laband, Paul, Besprechung zu: Mrignhac, Trait de Droit Public International
(1906) 20 Archiv fr ffentliches Recht 302
Lachs, M, Some Thoughts on the Role of Good Faith in International Law in
Declarations on Principles, Liber amicorum diciplorumque (1977) 4755
Lagerwall, Anne, Article 64, Convention de Vienne de 1969 in Olivier Corten and
Pierre Klein (eds), Les Conventions de Vienne sur le droit des traits (Brussels, 2006)
vol III, 229945


Bibliography327
Lauterpacht, Elihu, The Development of the Law of International Organization by
the Decisions of International Tribunals (1976) IV Recueil des Cours de lAcadmie
de Droit International de la Haye 377
Lauterpacht, Hersch, Private Law Sources and Analogies of International Law (with
special reference to International Arbitration) (London/New York, 1927)
Law of Treaties, Report of the Special Rapporteur, reproduced in (1953) II Yearbook
of the International Law Commission 90
Recognition in International Law (New York, 1978, reprint of Cambridge, 1947
edn)
The Development of International Law (Cambridge, 1996 (1st edn 1958))
Lavalle, Roberto, About the Alleged Customary Law Nature of the Rule Pacta
Sunt Servanda (1982) 33 sterreichische Zeitschrift fr ffentliches Recht und
Vlkerrecht 9
Lellouche, Pierre, The International Court of Justice: the Nuclear Tests Cases
Judicial Silence v. Atomic Blasts (1975) 16 Harvard International Law J 614
Leutert, Jrg, Einseitige Erklrungen im Vlkerrecht Ein Beitrag zur Lehre vom
Vertrauensschutz (Diessenhofen, 1979)
Macdonald, R St J and Hough, Barbara, The Nuclear Tests Case Revisited (1977)
20 German YB International Law 337
MacGibbon, IC, Estoppel in International Law (1958) 7 International and
Comparative Law Quarterly 468
Martin, Antoine, LEstoppel en droit international public prcd dun apercu de la thorie de lestoppel en droit anglais (Paris, 1979)
McWhinney, Edward, International Law-making and the Judicial Process: the
World Court and the French Nuclear Tests Case (1975) 3 Syracuse J International
Law and Commerce 9
Mendelson, MH, Are Treaties Merely a Source of Obligation? in William Elliott
Butler (ed), Perestroika and International Law (Dordrecht, 1990)
Menon, PK, The Law of Recognition in International Law: Basic Principles (Lewiston,
1994)
Mensbrugghe, Yves van der, Les garanties de la libert de navigation dans le canal de
Suez (Paris, 1964)
Meyrowitz, Henri, Le dbat sur le non-recours en premier aux armes nuclaires et
la dclaration Sovitique du 15 Juin 1982 (1982) XXVIII AFDI 147
Mller, Jean Paul and Cottier, Thomas, Acquiescenc in Rudolf Bernhardt (ed),
Encyclopedia of Public International Law (1992) vol I, 1416
Estoppel in Rudolf Bernhardt (ed), Encyclopedia of Public International Law
(1992) vol I, 11619
Mller, Jrg P, Vertrauensschutz im Vlkerrecht (Cologne/Berlin, 1971)
Nguyen, Quoc Dinh, Pellet, Alain and Daillier, Patrick, Droit international public,
7th edn (Paris, 2002)
Nolte, Georg (ed), Peace through International Law: the Role of the International Law
Commission, a Colloquium at the Occasion of its Sixtieth Anniversary, Beitrge zum
auslndischen ffentlichen Recht und Vlkerrecht, vol 211 (Berlin, 2009)
Obieta Chalbaud, Jos A, The International Status of the Suez Canal, 2nd edn (The
Hague, 1970)
Oppenheim, Lasa, Jennings, Robert and Watts, Arthur, Oppenheims International
Law, 9th edn (Harlow, 1992) vol 1, Introduction and Part 14

328 Bibliography
Parry, Clive, Grant, John P and Barker, J Craig, Parry and Grant Encyclopaedic
Dictionary of International Law, 2nd edn (Dobbs Ferry, 2004)
Paulus, Andreas L, Die internationale Gemeinschaft im Vlkerrecht The International
Commmunity in Public International Law: English Summary (Mnchen, 2001)
Pellet, Alain, Commentary to Art. 38 ICJ Statute in Andreas Zimmermann,
Christian Tomuschat and Karin Oellers-Frahm (eds), The Statute of the International
Court of Justice: A Commentary (Oxford, 2006) 677792
Pfluger, Franz, Die einseitigen Rechtsgeschfte im Vlkerrecht (Zurich, 1936)
Pufendorf, Samuel, Elementorum Jurisprudentiae Universalis Libri Duo (translation,
Oxford, 1931) vol 2
Quadri, R, Cours gnral de droit international public (1964) (III) 113 Recueil des
Cours de lAcadmie de Droit International de la Haye 245
Quintana, Juan Jos, The Nicaragua Case and the Denunciation of Declarations of
Acceptance of the Compulsory Jurisdiction of the International Court of Justice
(1998) 11 Leiden J International Law 97
Ratner, Steven R, Precommitment Theory and International Law: Starting a
Conversation (20022003) 81 Texas Law Review 2055
Reut-Nicolussi, Eduard, Die sterreichische Neutralittserklrung vom 26.
Oktober 1955 (1956) Internationales Recht und Diplomatie 15
Reuter, Paul, Principes de droit international public (1961) 103 Recueil des Cours
de lAcadmie de Droit International de la Haye 431
Droit international public, 6th edn (Paris, 1983)
Introduction to the Law of Treaties (London/New York, 1995)
Ritter, Jean Pierre, Laffaire des essais nuclaire et la notion de judgement dclaratoire (1975) 21 AFDI 278
Rodley, Nigel, The Treatment of Prisoners under International Law, 2nd edn (Oxford,
1999)
Rosas, Allan Negative Security Assurances and Non-Use of Nuclear Weapons
(1982) 25 GYIL 199
Rousseau, Charles, Droit international public, vol I, Introduction et Sources (Paris, 1970)
Rubin, Alfred P, The International Legal Effects of Unilateral Declarations (1977)
71 American J International Law 1
Ruda, Jos Maria, Recognition of States and Governments in M Bedjaoui (ed),
International Law: Achievements and Prospects (1991) 44965
Ruiz, Jos Juste, Mootness in International Adjudication: the Nuclear Tests Cases
(1977) 20 German YB International Law 358
Schwarzenberger, Georg, International Law as Applied by International Courts and
Tribunals, 3rd edn (London, 1957) vol 1
A Manual of International Law, 5th edn (London, 1967)
Schweisfurth, Theodor, Die Zustimmung der Bundesregierung zur Stationierung
amerikanischer Mittelstreckenraketen in der Bundesrepublik Deutschland (1974)
22 Archiv des Vlkerrechts 195
International Treaties and Third States (1985) 45 Zeitschrift fr auslndisches
ffentliches Recht und Vlkerrecht 653
Shaw, Malcolm N, International Law, 6th edn (Cambridge/New York, 2008)
Sicault, Jean-Didier, Du caractre obligatoire des engagements unilatraux en droit
international public (1979) 83 Revue Gnrale de Droit International Public 633
Sinclair, Ian, The Vienna Convention on the Law of Treaties, 2nd edn (Manchester, 1984)


Bibliography329
Skubiszewski, Krzysztof, Les actes unilatraux des tats in Mohammed Bedjaoui
(ed), Droit international: Bilan et perspectives (Paris, 1991) vol 1, 23151
Unilateral Acts of States in M Bedjaoui (ed), International Law: Achievements
and Prospects (1991) 22140
Strupp, Karl, Neutralisation, Befriedung, Entmilitarisierung (1933)
Sur, Serge, Les affaires des essais nuclaires (1975) 79 Revue Gnrale de Droit
International Public 972
Suy, Eric, Les actes juridiques unilatraux en droit international public (Paris, 1962)
Unilateral Acts as a Source of International Law: Some New Thoughts and
Frustrations in Droit du pouvoir, pouvoir du droit: mlanges offerts Jean Salmon
(Brussels, 2007) 63142
Suy, Eric and Angelet, Nicolas, Rechtsgeschfte, einseitige in Ignaz SeidlHohenveldern (ed), Lexikon des Rechts, Vlkerrecht, 3rd edn (Neuwied, 2001) 31921
Thirlway, HWA, International Customary Law and Codification: an examination of
the continuing role of custom in the present period of codification of international law
(Leiden, 1972)
, Concepts, Principles, Rules and Analogies: International and Municipal
Legal Reasoning (2002) 294 Recueil des Cours de lAcadmie de Droit International
de la Haye 267
Tomuschat, Christian, Commentary to Art. 36 ICJ Statute in Andreas Zimmermann,
Christian Tomuschat and Karin Oellers-Frahm (eds), The Statute of the International
Court of Justice: A Commentary (Oxford, 2006) 589657
The International Law Commission: an Outdated Institution? (2006) 49
German YB International Law 77
Unilateral Acts under International Law in Droits et culture. Mlanges en
lhonneur du Doyen Yadh Ben Achour (Tunis, 2008) 1487
Trebilcock, Anne M, Waiver in Rudolf Bernhardt (ed), Encyclopedia of Public
International Law (Amsterdam, 1984) vol IV 132731
Triepel, Heinrich, Vlkerrecht und Landesrecht (Leipzig, 1899)
United States Department of State, The Suez Canal Problem, 26 July22 September
1956 (Washington, 1956)
Venturini, Gian Carlo, Attitudes et actes unilatraux des tats (1964) (II) 112
Recueil des Cours de lAcadmie de Droit International de la Haye 363
Verdross, Alfred, Le fondement du droit international (1927) 16 Recueil des Cours
de lAcadmie de Droit International de la Haye 247
Vlkerrecht, 3rd edn (Vienna, 1955)
Die immerwhrende Neutralitt sterreichs, 4th edn (Vienna, 1977)
Verdross, Alfred and Simma, Bruno, Universelles Vlkerrecht: Theorie und Praxis, 3rd
edn (Berlin, 1984)
Vicua, Francisco Orrego, The Legal Nature of the Optional Clause and the Right
of a State to Withdraw a Declaration Accepting the Compulsory Jurisdiction of
the International Court of Justice in Liber Amicorum Judge Shigeru Oda (2002) vol
1, 46379
Virally, Michel, La pense juridique (Paris, 1960)
The Sources of International Law in Max Srensen (ed), Manual of Public
International Law (New York/London, 1968) 11674
Unilateral Acts of International Organizations in M Bedjaoui (ed), International
Law: Achievements and Prospects (1991) 24163

330 Bibliography
Visscher, Charles de, Problmes dinterprtation judiciaire en droit international public
(Paris, 1963)
Visscher, Paul de, Remarques sur lvolution de la jurisprudence de la Cour
Internationale de Justice relative au fondement obligatoire des certains actes
unilatraux in Essays in Honour of Manfred Lachs (The Hague/Boston, 1984)
45965
Vitzthum, Wolfgang Graf, Begriff, Geschichte und Rechtsquellen des Vlkerrechts
in Wolfgang Graf Vitzthum (ed), Vlkerrecht, 4th edn (Berlin, 2007) 179
Wengler, Wilhelm, Vlkerrecht (Berlin, 1964) vol I
Whiteman, Marjorie M and Hackworth, Green Haywood, Digest of International
Law (Washington, 1963) vol 1
Zemanek, Karl, Neutral Austria in the United Nations (1961) 15 International
Organization 408
Unilateral Legal Acts Revisited in Karel Wellens (ed), International Law:
Theory and Practice, Essays in Honour of Eric Suy (The Hague, 1998) 20921
Zoller, Elisabeth, La bonne foi en droit international public (Paris, 1977)

ILC Materials
ILC, Conclusions of the Work of the Study Group on the Fragmentation of International
Law: Difficulties arising from the Diversification and Expansion of International Law,
UN Doc A/CN.4/L.682 to be reproduced in (2006) II Yearbook of the International
Law Commission
ILC, Draft Articles on the Law of Treaties with Commentaries, reproduced in
(1966) II Yearbook of the International Law Commission 187
ILC, Guiding Principles applicable to Unilateral Declarations of States Capable of
Creating Legal Obligations, with Commentaries thereto, ILC Report, A/61/10
to be reproduced in (2006) II Yearbook of the International Law Commission
ILC, Replies from Governments to the Questionnaire, Report of the SecretaryGeneral, UN Doc A/CN.4/511 (2000)
ILC, Report of the Chairman of the Working Group, UN Doc A/CN.4/L.646 (2003)
ILC, Report of the Working Group reproduced in (1996) II Yearbook of the International
Law Commission (Part Two) 141, Annex II, Addendum 3
ILC, Report of the Working Group, UN Doc A/CN.4/L.543 (1997)
ILC, Report of the Working Group, UN Doc A/CN.4/L.558 (1998)
ILC, Report of the Working Group, UN Doc A/CN.4/L.588 (1999)
ILC, Report of the Working Group, ILC Report 2000, UN Doc A/55/10 (2000)
ILC, Report on the Work of its Fiftieth Session, General Assembly, Official Records,
Fiftieth Session, Supplement No 10, A/53/10 (1998)
ILC, Report on the Work of its Fifty-second Session, General Assembly, Official
Records, Fifty-fifth Session, Supplement No 10, UN Doc. A/55/10 (2000)
ILC, Report on its Work of the Fifty-fourth Session, General Assembly, Official
Documents, UN Doc A/57/10 (2002)
ILC, Summary Record of the 2524th Meeting, UN Doc A/CN.4/SR.2524 (1998)
ILC, Summary Record of the 2525th Meeting, UN Doc A/CN.4/SR.2525 (1998)
ILC, Summary Record of the 2526th Meeting, UN Doc A/CN.4/SR.2526 (1998)


Bibliography331
ILC, Summary Record of the 2527th Meeting, UN Doc A/CN.4/SR.2527 (1998)
ILC, Summary Record of the 2543rd Meeting, UN Doc A/CN.4/SR.2543 (1998)
ILC, Summary Record of the 2593rd Meeting, UN Doc A/CN.4/SR.2593 (1999)
ILC, Summary Record of the 2594th Meeting, UN Doc A/CN.4/SR.2594 (1999)
ILC, Summary Record of the 2595th Meeting, UN Doc A/CN.4/SR.2595 (1999)
ILC, Summary Record of the 2603rd Meeting, UN Doc A/CN.4/SR.2603 (1999)
ILC, Summary Record of the 2628th Meeting, UN Doc A/CN.4/SR.2628 (2000)
ILC, Summary Record of the 2629th Meeting, UN Doc A/CN.4/SR.2629 (2000)
ILC, Summary Record of the 2695th Meeting, UN Doc A/CN.4/SR.2695 (2001)
ILC, Summary Record of the 2696th Meeting, UN Doc A/CN.4/SR.2696 (2001)

Index
A
acceptance:
as a unilateral act 22, 60, 61
law of treaties 689
no need for 41, 43, 47, 249
presumption of See consent, presumption
of
tacit 78, 228, 302
acquiescence 22, 33
acta sunt servanda 305n
addressees:
of promises 186
reactions effect on promises 18991,
24750
advantages of promises 610, 30711
Advisory Opinion on the Legality of the Threat
or Use of Nuclear Weapons 165
aid, promise of 101
Anglo-Iranian Oil Company case 215
annexes as a means of interpretation 217,
222
arbitrariness as a bar on revocation 123,
1934, 25863, 745
Armed Activities on the Territory of the Congo
case 16972, 186, 192, 2145
Austrian declaration of neutrality See
neutrality
autonomy See unilateral acts of states,
unilateral
B
bilateral 401, 57, 22831
mindset 81, 229, 299
bindingness, legal basis for promises
194207
C
capacity of states 24, 183, 185
Certain German Interests in Upper Polish
Silesia case 867, 101, 231
circumstances:
fundamental change of 193, 246, 254,
25960, 262, 275
importance of 2227
coercion 25, 184, 238
collective state action See unilateral acts of
states
competence to bind a state legally 25,
1701, 1856, 2347
condition, resolutory 257, 263, 275

consent:
presumption of 1967
to revoke 254, 263
consideration 58, 230
corruption 25, 52, 221, 2379
counter measures 38, 63, 78, 296, 305
D
dangers, promises 3007
debts, promise to cancel 10
declaration 202
negotiated 233
of independence See Kosovo
See also interpretation
defective will 2457
drift-net fishing, promise to refrain from 10
E
Eastern Greenland case 93101, 102, 2324,
302
economic zone, declarations establishing
exclusive 63, 65, 77
empowering rule 3007
erga omnes undertakings 122, 131, 158, 172,
214, 227, 274
error 25, 61, 184, 2379
estoppel:
and promisesconceptional differences
2836
concept 1956, 261, 27783
non-promises triggering 2904
promises triggering 28690
F
Filleting of Fish within the Gulf of Saint
Lawrence arbitration 155, 225
Fisheries Jurisdiction case 223
force, declaration resulting from 25,
2379
form 27, 126, 129, 301
formalities, importance of 767, 2267
fraud 25, 61, 184, 2379
Free Zones case 878
Frontier Dispute case 15661, 188, 2134,
2234, 288
G
German approval case 14851
German assurances before WW II 1014,
231

334 Index
good faith:
application in the Nuclear Tests cases 121,
1268
basis of promises bindingness 2017,
250, 2524, 26972, 2958
general principle 1789
greenhouse gases, promise to reduce 1
1
guarantees of non-repetition 167
Guiding Principles 3, 4n, 134, 18394,
2124, 2958, 3212
I
Ihlen declaration See Eastern Greenland case
immunity, promise to grant 15
intention to be legally bound 195, 198201,
20734
internal law 2357
international community as a potential
addressee 8, 145, 186, 214, 308, 310
International Law Commission See Guiding
Principles
international organizations, declarations of
44, 53, 65, 7879, 182
interpretation, restrictive standard of 126,
131, 141, 155, 1601, 172, 191,
21214
interpretative declarations 63, 78
Ireland v. the United Kingdom 1415, 218
Island of Lamu arbitration 82, 96
ius cogens 25, 187, 238, 245, 248, 250, 277,
297
J
judicial proceedings, declarations made in
the course of 63, 758, 224
K
Kosovo, declaration of independence 10, 15
L
LaGrand case 1669, 225
law making:
judicial 1823
treaties 180
law of treaties See Vienna Convention
London Agreement 101
M
manifestation of will to be bound See will
Mavrommatis Jerusalem Concessions case
835, 86, 231
Military and Paramilitary Activities in and
against Nicaragua case 1515, 160,
165, 213, 261, 26773
minorities, declarations on the protection of
8893, 229
multilateral 401, 57

N
neutrality, Austrian declaration of permanent 1048, 229, 231
North Sea Continental Shelf cases 1146,
224, 231, 2801, 284
notice, reasonable 26773, 275
notification 368
Nuclear Tests cases 124, 11638, 1745,
182, 224
O
object:
lawfulness of 2436
possibility of 2436, 254
obligation for third states 187
offer:
law of treaties 68
as a unilateral act 5962
or promise 689, 878, 153, 22834
Optional Clause, declarations under
the 63, 6975, 26872
P
parliamentary debates 223
Petitpierre, assurance to the U.N.
1457
preamble 217, 222
preparatory work 2178
protest 1, 3, 202, 27, 334, 53, 64, 76,
283, 312
pseudo-promise 10910, 145
publicity 20, 121, 123, 125, 146, 147, 187,
23942
Q
Questions relating to the Obligation to
Prosecute or Extradite case 1723,
225
quid pro quo 5, 7, 12, 578, 92, 98, 100,
101, 1201, 125, 135, 197, 2304,
249
R
rareness, false presumption of promises
412
ratification 12, 56, 67, 75, 87, 93, 109, 115,
188, 224, 2467, 3034
recognition 1, 3, 20, 21, 22, 27, 2933, 64,
76, 78, 94 7, 287
registration 109, 112, 130, 140, 226, 2467,
2967
rejection of promises by addressee(s)
1268, 2047, 24750
reliance:
constructive 1268
detrimental 195, 261, 2815
in the Nuclear Tests cases 1268
presumption of 2017, 295


protection of 1268, 1934, 200, 20114,
216, 24950, 253, 264, 282, 284,
289, 292
revocation of promises 193, 25176
S
security assurances, negative 1616
silence 21, 63, 78, 184, 189, 194, 24950
source:
of international law 17683, 2945
of international obligations 17981
South West Africa cases 66, 10910
state responsibility 3940, 63, 78
state sovereignty 131, 195, 198201
stipulation pour autrui 46n
Suez Canal declaration 10814, 174, 226,
231, 301
suspension of a treaty, act declaring 68
Switzerland:
assurance to the U.N. 1457
declaration in relation to UN Sec. Res. 253
13840
T
Temple of Preah Vihear case 286, 293
termination of a treaty, act declaring 68
territorial sea, declarations establishing
width of 65, 77
text as a means of interpretation 21422, 296
time limit 257, 263
torture:
Britsh Pledge before the ECtHR 1415
declarations on 15, 1401
U
unilateral acts of states:
categorization 268
See also recognition, protest, waiver and
notification

Index335
collective state action:
within treaties providing rights 459
joint declarations outside treaties
4954
single subject of international law
414, 545
rules applicable to all 245
strict sense 213, 689
See also unilateral acts of states, autonomy
Unilateral
autonomy:
overall concept 559
from another bi- or multilateral act
637
from another unilateral act 626
lex specialis 6778
definition 401, 207
V
Vienna Convention on the Law of Treaties:
80
analogy to provisions See relevant subject
matter (eg competence, corruption,
fraud, revocation)
application to declarations under Art. 36
(2) ICJ Statute 715
as lex specialis 689
rights for third states 459
W
waiver 1, 3, 21, 27, 346, 77, 151, 304
war:
declaration of 39
material, declaration on the use of
1478
will, manifestation of 4, 3840, 1879,
20811, 2967
See also intention to be legally bound
wilful action, merely 389